1. Introduction
The use of technology to address inefficiencies within the healthcare system and optimize patient safety has an extensive history of development, starting with the documentation and recording of patient care events. The concept of the electronic health record (EHR) emerged in the 1970s with the first official EHR built in 1972 by the Regenstrief Institute at Indiana University and has grown expediently over several decades [1,2]. The EHR provides a historical record of the patient care that was ordered and completed, making the EHR inherently retrospective. While this historical record is necessary for legal, administrative (billing), and diagnostic confirmation (post-test probability), it is cumbersome to use for real-time clinical decision-making like predictions, detection, and prognosis. This limitation leads to the inability to anticipate the healthcare needs of the patient, as well as the disease process that may be occurring; this is otherwise known as pre-test probabilities [3]. An opportunity to obtain more granular, real-time data is to use ambient sensors such as CV. CV mimics human vision that integrates and interprets visual information and could be potentially used to create sophisticated algorithms in real-time.
We recognize that much can be said about CV in terms of development trends and internal causal relationships of the overall and individual application of this technology. However, the purpose of this manuscript is to provide a high-level, comprehensive overview of the application of CV in healthcare settings based on other industries. First, we will briefly review the use of CV in industries outside of healthcare and categorize its application into themes. Following these themes, we review in greater detail how CV has been applied and/or developed for the hospital setting, and then we review outpatient and community settings. To identify future opportunities for the application of CV, we completed journey mapping at the patient, clinician, and system levels. Lastly, we discuss the privacy and safety considerations and ethical implications for the use of CV in the healthcare setting.
2. Overview of Computer Vision
CV is a type of artificial intelligence (AI) that uses digital videos or sequences of images. The goal of CV is to train computers to extract information from the images, essentially enabling computers to “see” and recognize content [4]. The foundations of CV were established during the 1980s; they were marked by the development of algorithms like optical flow and edge detection [5,6]. The advancement of machine learning and statistical techniques in the 1990s empowered computer applications to acquire the ability to understand and process more intricate patterns within visual scenes [7,8]. During the 2000s, the application of CV manifested in more practical domains, including the analysis of medical images and the detection of faces. The use of Convolutional Neural Networks (CNNs) significantly advanced the field of CV in 2012 when CNNS demonstrated high performance in the ImageNet Large Scale Visual Recognition Competition [9], emerging as the predominant learning method in CV. CNNs showed expert-level performance in image classification, object detection, and semantic segmentation across diverse fields, including medicine, surveillance, and autonomous driving [10,11,12,13,14,15,16]. As machine learning models advanced, obtaining a sufficient quantity of labeled data became a challenge. In response to this obstacle, unsupervised techniques like clustering and dimensionality reduction were developed. These approaches delve into the inherent structure of data without relying on explicit guidance to address the scarcity of labeled data [17,18,19]. However, the adoption and therefore application of CV into hospital settings has been slow. Its utilization in healthcare comes with limitations, as ethical and privacy concerns take precedence when involving humans, particularly humans in a vulnerable state (i.e., patients). As CV continues to develop, including the potential to assist in many aspects of patient care (e.g., documentation, recognition of a deteriorating patient, etc.), it is important to revisit how CV could be utilized for the benefit of patients and providers.
CV encompasses a multidisciplinary domain integrating advanced machine learning techniques, pattern recognition, and image processing to empower computers to comprehend the visual content present in images and videos [16,20,21,22]. Typically, CV algorithms start with the acquisition of data through cameras or sensors, followed by preprocessing and image enhancements. CNNs play a pivotal role in automatically learning the representations of visual scene content and contribute significantly to various CV tasks. Due to the CNNs’ robust feature representation capabilities, CNNs have found widespread application as an effective method for extracting meaningful patterns and features [23,24,25]. Despite the complexity of training CNNs due to their numerous layers, the artificial intelligence field addresses these challenges by adopting transfer learning and fine-tuning techniques to enhance model efficiency and representation power [26,27,28]. Object detection, crucial in applications like medical diagnostics and autonomous driving, heavily relies on CNNs, where a CNN backbone network extracts image features, and candidate regions determine target category and location information. CNNs also find applications in object classification across medical imaging, security, and agriculture, among other industries. For instance, in medical imaging, CNNs extract relevant features for categorizing structures such as tumors, aiding in diagnostic processes [29,30,31]. By means of parallel processing, Graphics Processing Units (GPUs) (i.e., computer chips) enhance the efficiency of managing extensive matrix operations essential for the processing of CNNs. This parallel approach substantially diminishes training times, thereby streamlining the implementation of real-time applications utilizing CNNs [32]. The evolution and implementation of CV have undergone notable transformations in the utilization of deep learning models and the expansion of machine learning methods. These deep learning models and methods are outlined further in Table 1.
3. Application of Computer Vision in Industry Outside of Healthcare
The use of CV was identified in 24 major industries including agriculture, engineering and manufacturing, retail, and education, among many others. The application of CV largely fell into four different themes or categories: job enhancement, surveillance and monitoring, automation, and augmented reality. These themes are outlined in Table 2 and provide insight into how CV could be applied in healthcare. Examples of job enhancement include the use of CV to analyze sporting events to inform referee calls [33], the scoring of diving competitions [34], and insurance appraisals to assist with claim reporting [10]. The use of CV in surveillance included the detection of forgery in artwork [35,36,37] as well as other industries, the prevention of cheating in academic and educational settings, and the enforcement of speed limits [14]. The theme of monitoring was the largest, including the use of CV to monitor agricultural crops for disease or insect infestations [15,38,39], detect restocking needs in warehouses or retail stores [40], and identify defective products on assembly lines [41]. The final category of augmented reality included technology such as Apple Vision [42], the ability to try clothes on virtually [43,44,45], and tools to assist people with vision impairment and blindness [46,47,48].
4. Computer Vision Application in Hospital Settings
Current medical applications of CV largely focused on monitoring (detection and measurement) and were mostly in the development and testing phases. The application or use of CV in the hospital included several commercial companies that specialized in patient monitoring for falls (Artisight:
4.1. Radiology
The use of CV in radiology has gained increased attention as it can support timely intervention and enhanced efficiencies within clinical workflows. For example, a recent study demonstrated how CV could support surgeons in diagnosing wrist fractures in pediatric patients [154]. The motivation for this type of CV application is to expedite surgical care in low-resource environments where specialized radiologists are not readily available. Another example of a clinical application is notification software (AIDOC), which is used to detect intracranial hemorrhage [141]. Teleradiology networks can employ this software to expedite stroke workups in critical access hospitals or lower trauma centers. Other CV radiology applications concern detecting anomalies within medical images. A recent study by Lakhani et al. (2017) [99] employed deep models such as 2D-CNN, AlexNet [9], and GoogleNet [227] in their CV approach to detecting tuberculosis in chest radiographs and pulmonary tuberculosis in chest radiographs. Other studies applied ensemble learning for the diagnosis of Alzheimer’s Disease using MRI brain images [139,140,142]. The typical image types utilized for radiology segmentation include X-rays, CT scans (i.e., liver tumor, [138]), MRIs (i.e., brain tumors [149]), and 4D-CT (i.e., brain tissue for stroke workup [148]). While the utilization of deep learning has resulted in precise detection rates in the field of radiology, these approaches require extensive, well-annotated datasets. Without such datasets, deep learning methods may experience overfitting, leading to a reduction in their generalizability [228]. There are a multitude of other examples of CV applications in radiology as shown by recent reviews including emergency radiography, stroke workup, and workflow efficiencies [228,229,230]. The implementation and use of these algorithms have been slower than expected and may be due to a lack of standard user interfaces and differing expectations among clinicians and administrators as reported in a recent study [231].
4.2. Histology
The examination of histological images by pathologists provides diagnostic information crucial for influencing a patient’s clinical outcome. Traditional histological image representation involved extracting texture and color features and employing conventional machine-learning approaches. However, the CV landscape has evolved with the extensive representational power offered by Convolutional Neural Networks (CNNs) [127,134]. For example, Bejonrdi et al. (2017) [112] demonstrated that the application of CNNs for detecting lymph node metastases in breast cancer outperformed eleven pathologists in a simulated time-constrained setting. Tellez et al. (2018) [113] developed a CNN-based CV approach capable of effectively detecting mitosis in Hematoxylin and Eosin whole-slide images . Additionally, Kather et al. (2019) [136] showcased that microsatellite instable tumors in gastrointestinal cancer could be directly predicted from H&E histology using CNNs to classify tumors versus normal tissues. The present challenge involves handling high-resolution histological images, requiring substantial computational resources and extensive training sets. Employing transfer learning and knowledge distillation approaches may partially mitigate this challenge [232].
5. Development and Testing of Computer Vision in the Hospital Setting
Several CV-based tools were identified that had been shown to be effective in the testing and development phases; however, they were not yet put into practice at scale. These include the detection of behaviors and signs related to delirium, pain detection and monitoring, monitoring of sedation depth and signs of patient deterioration, mechanical ventilation, and monitoring of the care setting aimed at improving patient safety and quantifying workload. We provide an overview of these studies in this section.
5.1. Detection and Monitoring of Brain Health
Delirium, a type of acute brain dysfunction, occurs in 50–80% of critically ill patients [233]. An observational pilot study reported that delirious patients had significantly different expressions of facial action units, head pose movement, and other environmental factors such as visitation frequency measured by CV (n = 22) [156]. A different study examining the frequency of caregiver actions reported that delirious patients had more caregiver activity overall, which was most concentrated from 8:00 p.m. to 12:00 a.m. [195]. These observational differences between non-delirious and delirious patients can be automated to aid in the recognition of early warning signs of delirium, the measurement of delirium severity, and aid in subtyping and phenotyping efforts.
5.2. Detection and Quantification of Pain, Agitation, and Level of Sedation
CV has been used to detect pain in a variety of patient populations (infants to aging adults), specific disease states (lung cancer, dementia, chronic back pain, shoulder pain), and after procedures (procedural pain in infants), mostly in community or outpatient settings. A recent scoping review identified one study that tested the feasibility of an automated approach to pain assessments using a deep-learning method in a population of critically ill patients [176,177]. The study tested the accuracy of models to dichotomize pain and rate pain on three levels (0–2), reporting >0.85 accuracy for dichotomized models compared to 0.56 for the three-level model [176].
Automated pain recognition and monitoring have been widely explored in neonatal populations. Several reviews report on CV developed over the past two decades aimed at the recognition of pain, monitoring, and the measurement of intensity [159,163,171,179]. These feasibility studies and developed models have cumulated into a point-of-care mobile health application for procedural pain named PainChek Infant [172]. In a feasibility study of forty infants (age range 2.2–6.9 months), the mobile health application significantly correlated (r = 0.82–0.88, p < 0.0001) with clinician-administered pain scales (Neonatal Facial Coding System and Observer administered Visual Analogue Scale) and demonstrated high interrater reliability (ICC = 0.81–0.97, p < 0.001) and high internal consistency (alpha = 0.82–0.97) [172]. This type of technology could be applied to adult hospitalized patients to improve pain monitoring and assist clinicians with follow-up pain assessments after the administration of analgesics, especially in patient populations that are not able to verbalize their discomfort and pain level. Furthermore, this type of state-of-the-art technology was emphasized by a recent narrative review that discussed the updated definition of pain from the International Association for the Study of Pain and how multidimensional technologies are needed to improve the identification and monitoring of pain [234]. Untreated pain can result in delirium, agitation, hostility, and other adverse consequences such as impaired healing and increased mortality risk [235,236,237]. The use of CV, paired with other artificial intelligence modalities, clinicians, and patients within a model, could improve the proactive recognition and monitoring of pain in hospital environments across populations.
Proof of concept CV models have been developed to recognize and monitor facial and body movements associated with agitation during sedation, such as grimacing. One study used a volunteer to simulate agitation through different levels of grimacing, developing a proof-of-concept algorithm that can be tested in critically ill patients [158]. Another study used a volunteer to simulate limb movement during agitated episodes and then tested this model on five ICU patients. The results of the model were correlated over time with a nurse-administered Riker Sedation Agitation Scale and physiologic signs including heart rate, heart rate variability, systolic blood pressure, and blood pressure variability [157]. Lastly, a study on young children and infants used eye movements to facilitate the measurement of sedation and consciousness levels in young children and infants [168].
5.3. Patient Deterioration
The facial action unit is a comprehensive system that conveys facial movements, subsequently utilized for detecting emotions such as anxiety, stress, fear, or pain [164,166,173]. As presented in reference [175], Giannakakis et al. (2022) illustrated that facial action units are associated with stress levels. This implies that during stressful situations, specific action units, such as cheek raising, lid tightening, and upper lip raising, intensify.
The early warning signs of an impending acute patient deterioration are often subtle and overlooked by busy clinical staff leading to a delay in the escalation of care [238]. To address this limitation, a recent feasibility study examined how subtle changes in facial expressions were associated with a future admission to the intensive care unit. This study used CV to identify facial action units and reported that combinations of the upper face, head position, eye position, lips and jaw, and the lower face were associated with the increased likelihood of admission to the intensive care unit (n = 34 patients) [181]. This algorithm could be used to proactively identify patients at risk of acute deterioration and support early intervention [181]. In a post hoc analysis of these data, a decrease in the number of facial expressions (per time unit) and an increase in the diversity of facial expressions predicted admission to the ICU (AUC = 0.81) [182]. Numerous opportunities exist to expand on this CV algorithm and investigate how other signs, such as the frequency of clinician visits to the patient’s room or the presence of certain respiratory devices [183], could improve early recognition of impending acute deterioration.
5.4. Mechanical Ventilation
CV has been applied to the field of mechanical ventilation to estimate regional lung volumes using light to reconstruct the motion of the lungs and measure the regional pressure distribution [184]. This proof-of-concept model was developed by Zhou et al. using a mannequin that measured and monitored chest expansion with a light projector and cameras. They utilized surface reconstruction of regional chest expansion for their model, which showed good accuracy with an error of 8 mL under 600 mL tidal volume. They compared their methods with other frequently used computational models and reported a 40% reduction in computational costs paired with improved accuracy in their new model. This work needs to be clinically tested and validated.
5.5. Mobility
The use of CV to monitor and document physical activity and early mobilization in critically ill patients was reported in two studies. CV was used in a recent study to develop a tool for real-time patient mobility monitoring in the ICU [187]. Yeung et al. (2019) reported a mean sensitivity of 87.2% and specificity of 89.2% with an AUROC of 0.938 for detecting mobility activities (i.e., get out of bed, get into bed, get out of chair, get into chair). The CV had an accuracy of 68.8% for quantifying the number of healthcare personnel involved in each activity [187]. Another study by Reiter et al. (2016) developed an automated mobility sensor to support the monitoring of patient activity in the surgical ICU. They compared the algorithm’s performance with clinician performance on the identification of physical activity and reported high inter-rater reliability with a weighted Kappa score of 0.86 [186]. These types of models could automate documentation of physical activity in hospital patients, decreasing clinician documentation burden and increasing the accuracy of electronic health records.
5.6. Patient Safety
Several studies have developed models focused on improving patient safety using CV as a monitoring tool. One focus is hand hygiene, an essential component of infection prevention. While compliance is critical for patient safety, monitoring clinician performance is time-consuming. It can be inaccurate as it requires human observation of care procedures both inside and outside the patient room. CV provides an opportunity to automate monitoring. This use case has been demonstrated using depth sensors in a hospital unit and video and depth images in a simulated hospital room [192,220]. Both models achieved sensitivity and specificity greater than 90% in detecting hand hygiene dispenser use and performed better than human observers. In addition to monitoring, future studies could explore how the model could provide real-time feedback to clinicians, or reminders of hand hygiene, leading to further opportunities to improve patient safety.
Surgical procedures and operating rooms (ORs) are the focus of a recent review that highlights how CV could improve patient safety and system efficiencies [198]. A recent study used off-the-shelf camera images to measure the level of situational awareness of surgical teams during timeout procedures in the OR. The model distinguished between teams with good and poor situational awareness, substantiating existing studies in the OR on the use of CV to augment traditional human-based observation assessments [194,198]. Other CV-based models have aided in surgical phase recognition, robot-assisted surgeries, surgical skill assessment, detection of instruments or lesions during surgery, enhanced visual displays in surgeries, and navigation during surgical procedures [194,198,199].
5.7. Quantification of Workload in the ICU
A few studies have demonstrated the feasibility of ambient monitoring of caregiving activities in the ICU using CV. The first study completed a task recognition of caregiving activities over 5.5 h with an accuracy of 70% in a pediatric ICU [226]. The recognized tasks included documentation, observation, and procedures, among others [226]. These tasks were then examined over time for trends. The second study recognized and then categorized patient and caregiver movement (i.e., workload) over the course of 24 h in an adult ICU [195]. The study reported significant differences in patient and caregiver movement throughout the 24 h period, between intubated and non-intubated, delirious and non-delirious, and settings (high dependency unit vs. ICU). Another study developed and validated a Clinical Activity Tracking System (CATS), testing its use in both a simulated and actual ICU room. Like the previous study, more caregiving activity was reported between 7:00 a.m. to 11:00 p.m. compared to 11:00 p.m. to 7:00 a.m. [221]. This system was validated against manual observation with a correlation of r = 0.882 [222]. Improving the quantification and understanding of caregiver workload and function throughout a time period was the focus of these studies as existing monitoring systems are resource intensive and subjective [195,222,226].
6. Computer Vision in Outpatient and Community Settings
CV has been developed in the community and outpatient settings to detect, measure, and monitor patient symptoms, signs of underlying illness or disease, and patient events such as falls. A recent survey identified over thirty different CV models developed to automatically detect underlying symptoms related to medical diagnoses [162]. These include monitoring vascular pulse, pain, facial paralysis, neurologic conditions, neurodevelopmental disorders, psychiatric disorders (i.e., attention deficit hyperactivity disorder (ADHD), autism, depression), and mandibular disorders, among others [162]. This type of computer-assisted diagnosis ranges from the detection of facial shape, facial features, and facial muscular response to voluntary emotion and facial motion analysis. In this section, we briefly overview the subject areas with the highest level of development.
6.1. Pain Detection and Monitoring in Community Settings
The automatic recognition and monitoring of pain in the community and outpatient setting is well-established. A systematic review in 2021 (n = 76 studies) reported on the use of CV to diagnose and treat chronic lower back pain [167]. Also, in 2021, a survey of automated detection of pain summarized studies and CV across populations, providing an in-depth overview of datasets, learning approaches, spatial representations, and machine learning methods used [170]. A narrative review highlighted the state-of-the-art technology published on pain detection and monitoring [234]. Lastly, a scoping review reported on several community and outpatient models to detect and monitor pain using CV [177]. A recent study used recent developments in CV automated segmentation and deep learning along with the updated definition of pain from the International Association for the Study of Pain to develop a sentiment analysis system within a Smart Healthcare System for pain monitoring [180]. This CV model and most models mentioned in the reviews need to be prospectively tested [170].
6.2. Neurologic, Neurodevelopmental, and Psychiatric Disorders
Autism spectrum disorder, a neurodevelopmental disorder, is increasingly prevalent in pediatric populations [239,240]. Time to diagnose and receipt of needed care and resources can be delayed by months, leading to deficiencies in care. To address this gap in clinical care, an interactive, mobile health technology was developed through a series of studies that uses CV in a closed-loop system to automatically code signs and behaviors associated with autism [202]. The intent is for parents to use this technology at home to improve the early recognition of autism and access to needed resources and care. This use case and framework could be expanded to include additional neurodevelopmental disorders with similar impact. A different study developed a CV model that could differentiate between individuals with autism spectrum disorders, ADHD, and healthy controls [200]. Head motion and facial expression were used to distinguish between these disorders [200].
The detection and severity of depression have been automated using CV in a few studies [174,178]. Depression is one of the most common psychiatric disorders and is often underrecognized, leading to delays in patient care and decreased quality of life. The application of CV to detect and measure depression could have widespread implications and lead to the early detection and allocation of resources to improve patient care. For example, such an algorithm could be applied in telehealth outpatient visits where depression may not otherwise be discussed or during a hospitalization where situational depression can increase patient stress and lead to prolonged hospitalization and readmissions. In addition to depression, one study explored the accuracy between clinician-rated and computerized recognition of blunted facial affect [209,241].
To detect facial weakness, Zhuang et al. (2020) developed CV using images and videos of people collected from Google Images and YouTube videos [203]. Six neurologists used a rating scale (likelihood of facial weakness) to label the images. Following model development, the authors concluded that the combination of landmark and intensity features led to the highest accuracy. The ability to detect the shape (i.e., landmark) and texture (i.e., gradient intensity) was contributed by the neurologists who labeled the images [203]. Facial palsy detection has been the focus of several studies. Guarin et al. (2018) retrained a CV facial landmark detection model that was previously trained using healthy individuals with facial palsy patients to develop a more accurate model [201]. Ruiter et al. (2023) studied the use of facial recognition software to identify patterns of facial weakness, and a deep learning model was trained for classification and disease severity in a cohort of myasthenia gravis patients [204]. The images used for training were collected in the outpatient setting. The area under the curve for diagnosis of myasthenia gravis was 0.82 and 0.88 for disease severity [204]. Another recent study assessed the intensity of facial weakness in patients with facial palsy. The intensity was classified into three levels by focusing on specific facial landmarks. The accuracy of detecting palsy patients was 95.61%. The accuracy for class assignment (intensity level) was 95.58% [205].
To improve virtual interactions and patient education efforts, a CV algorithm detected changes in facial expressions indicative of confusion and compared its accuracy to forty medical students [189]. The accuracy of the human raters in identifying confusion was 41% compared to 72% accuracy by the CV algorithm using four different facial action units (lowered brow, raised cheek, stretched lips, and lip corner pulled).
6.3. Falls
With the increasing population of aging adults and their preferences to live at home, falls at home have contributed to a significant increase in the risk of morbidity and mortality in the population [242]. To address the increasing incidence of falls, CV technology has been developed to detect and monitor for risks and signs of falls in the home environment. A literature review completed in 2023 surveyed the use of ambient sensors to detect falls in the home environment. While some studies have used CV to detect falls, other systems are a combination, or hybrid, of wearable and ambient sensor technologies [216]. One example of CV fall detection was developed by Joshi et al. (2017) [215]. A single camera was used to detect four different movements indicative of a fall event, and notifications were sent via email to the designated individual if a fall was detected. The CV model achieved an accuracy of 91.8% [215]. This model, and the majority identified by the recent review, focused on the detection of falls and not on the prediction or identification of early warning signs [216].
7. Journey Mapping and Future Computer Vision Application
To investigate how CV could be applied in the hospital setting, the temporal journeys of the clinician and patient through the healthcare system were mapped and analyzed for opportunities. The overview of the journey map is illustrated in Figure 1. Many identified opportunities to incorporate CV into patient care and the healthcare system overlapped. These identified opportunities are displayed in Figure 2. For example, the use of facial recognition technology to automate patient check-ins in the outpatient and inpatient settings improves the efficiency of the system while also providing a smoother process for the patient. The monitoring of parking lots for available patient parking and the use of interactive displays to provide directions to clinic or hospital appointments benefit the system and the patient. Individuals who were responsible for patient check-in could instead meet, greet, and accompany a patient on their clinic or hospital journey to improve the coordination of care. The integration of CV into clinic and hospital rooms could improve the monitoring of patient conditions, resulting in early detection of acute deterioration or patient discomfort, assist with diagnostic testing, provide real-time feedback on the effectiveness of interventions to ameliorate patient discomfort, and complete auto-documentation of patient care procedures. These examples benefit the clinician, patient, and efficiency of the system. When taking ideas such as these from concept to development, it is crucial to identify who the end-user is, who benefits from the model or service (which may be different than the end-user), what efficiencies are improved, and what unintended consequences may result once the algorithm is in production. Additionally, the privacy, safety, and ethical principles and values must be considered. These are discussed next.
8. Summary and Implications for Computer Vision Use in Healthcare
This rapid review identified a few applications of CV in the hospital setting. Most of the CV in hospitals is still in the feasibility and proof-of-concept stage, lagging behind other healthcare settings and industries. This gap in CV application in hospital settings is likely due to the availability of public datasets to train and develop models, data privacy and security needs, ethical considerations, and barriers inherent within a complex system.. We will discuss these limitations, including ethical and economic considerations. Our journey mapping exercise identified many future opportunities for CV in the hospital and outpatient settings. As future opportunities are considered, it is critical to understand what problem the CV aims to solve, the stakeholders involved in using it, how privacy, safety, and ethical concerns are addressed, and the potential unintended consequences of its use in these settings.
9. Data Privacy and Safety Considerations
Before using a CV model in the hospital setting, it is crucial to consider the data privacy and patient safety requirements. Privacy has multiple meanings that depend on the stakeholder’s perspective [243]. A recently published meta-synthesis highlights perspectives patients and health professionals share on the benefits and risks of artificial intelligence (AI) in healthcare [244]. A theme identified by patients and clinicians involved the importance of data security and use. Both stakeholder groups shared how the storage and protection of these data were essential to prevent records from being hacked and/or leaked. Further, the meta-syntheses reported that the unwarranted use of these data for commercial purposes was a significant concern [244]. These concerns are related to overall AI use in healthcare and are not specific to CV.
Privacy and data management concerns unique to CV center on the nature of ambient intelligence, how it is applied, and what information is captured in the video images [245]. As a recent perspectives article highlighted, it is important to collect the minimum amount of information needed to train and use the model [246]. This could mean using black and white images instead of color or blurring or removing unnecessary pieces that do not contribute needed information. It is also important to consider the inclusion of individuals who are not the focus of the model. For example, the patient may be the focus of data collection, but clinicians and visitors in the hospital room may also be included in the video capture. Everyone and everything within the image field is included in the data collection. It is important that all individuals who may enter the room are either consented (i.e., patient) or informed (i.e., clinician) of the data collection and what privacy protections are in place [246]. If the collected data are being considered for other purposes and the bystanders are reidentified, there should be a process in place to notify and gain consent of those individuals. These concerns emphasize the importance of data management (storage and use) within research studies, production teams, and the healthcare system.
Informed consent or assent of data collection detailing the why, how, and when behind the collection and the privacy protections in place is imperative to complete, particularly when capturing images of patients in a very vulnerable state. Decision points can be built into the consent process, allowing the patient to opt in to use their image data for other purposes. For example, an opt-in for sharing their images with external institutions or scientists can help facilitate the development of public image datasets that may accelerate CV development. It is important to clarify that once a model is in production, images do not need to be retained as it can operate as a closed-loop system. This may limit the transparency, or the ability to review the algorithm to understand the output, but it does improve privacy protections.
Safety considerations for CV in hospitals are multi-factorial. It is essential to consider the end-user of the model. How will the end-user use the information provided by the model in their decision-making? Who is responsible for maintaining the model to ensure its accuracy? CV models can improve patient and clinician safety. For example, a model could recognize the early warning signs of workplace violence and notify clinicians to improve their situational awareness and implement mitigation measures to prevent verbal or physical abuse. Another example that proof-of-concept models have demonstrated in the operating room setting is the detection of missed care, poor situational awareness, or procedure errors. Both examples would improve patient and clinician safety [194,198]. On the other hand, CV models could decrease the safety of patient care. Previous studies have shown how bias is readily introduced into models if the training data are not representative of a diverse population [247]. These biases can lead to embedded stereotypes, discrimination, and exclusion of certain patients [248]. Current deep learning models employed in CV tasks directly derive their knowledge from the training data. Consequently, the performance of the model is heavily influenced by the distribution of the training data. If bias is present in the training set, the model identifies it as a significant context, impacting the generative capabilities of the model for unseen examples. Numerous studies in the literature have explored methods to extract bias-independent feature embeddings, resulting in enhanced performance of neural networks when trained on biased datasets [249,250,251,252]. These methods can be integrated into model development along with representative sampling to minimize the risk of bias.
9.1. Ethical Considerations in Computer Vision
The use of CV in healthcare has broad ethical considerations that need to be addressed as algorithms and models are designed, developed, tested, and deployed. Each stage of the algorithm, including maintenance, should be considered, and continually re-evaluated to ensure the medical ethics of autonomy, beneficence, non-maleficence, and justice are upheld for the end-user. It is also important to define and consider who is the end-user (i.e., patient, a decision-maker for the patient, clinicians, administration, support staff) and proactively address ethical concerns [253]. Depending on the end-user and circumstances concerning the use of the technology, different ethical principles or values may need to be considered. A recent scoping review identified eleven different ethical principles and values on the use of artificial intelligence [254]. These include transparency, justice and fairness, non-maleficence, responsibility, privacy, beneficence, freedom and autonomy, trust, sustainability, and solidarity [254]. Similar themes along with societal implications were summarized in a recent narrative review by Elendu et al. (2023). Inherent within these principles is the importance of placing the patient at the forefront and ensuring that every patient has a fair and equitable opportunity to benefit from the technology [255]. This priority encapsulates the responsibility to ensure the model was built on a representative dataset that can be generalized broadly, i.e., any risk of bias, discrimination, and stereotyping is minimized, and the welfare of the patient is prioritized. To accomplish these goals, it is important to partner with a medical ethicist, sociologist, or patient-community stakeholder group to evaluate the technology from multiple viewpoints within an ethical framework [255]. Questions evaluating the intent of the model, who will use the model and who will benefit from the model, how the model will be implemented and maintained, the acceptability and usability of the model, the transparency of the algorithm and resulting decisions, who holds the ultimate responsibility for performance, and how unintended consequences will be identified, tracked, and evaluated are just a few topics that are essential to work through prior to the inception of the project.
9.2. Economic Considerations
Job replacement and loss are significant concerns regarding the application of artificial intelligence, including CV [256]. To ensure safety and ethical considerations are followed, it is important to build “human in the loop” models that use CV as a tool to inform decisions; however, the human is the critical decision maker on how to use the information provided [257]. CV should enhance processes to improve decision-making, efficiencies within the system, and patient outcomes. It should not replace humans. A framework for evaluating the implications of automation in artificial intelligence was shared in a recent working paper by the National Bureau of Economic Research [256]. This paper discusses the balance between potential displacement and increased demand in non-automated labor tasks that could enhance the human experience. This type of framework is important for healthcare systems to use as the adoption of CV is considered. A recent review studied how artificial intelligence models could result in healthcare cost savings over several years [258]. Although they reported significant cost savings with the use of artificial intelligence for diagnosis and treatment, they highlighted that a major disadvantage to artificial intelligence is the prioritization of accuracy over clinical evaluation and scientific validation.
9.3. Acceptability and Readiness for Computer Vision
The implementation of artificial intelligence in healthcare is impacted by public opinion. In a comprehensive review published by Bekbolatova et al. (2024), the results of Pew Research surveys are highlighted, emphasizing the correlation between familiarity with artificial intelligence and the expressed potential for it to benefit healthcare [259]. While readiness for artificial intelligence is growing, the need to address specific knowledge gaps within the community to increase familiarity with artificial intelligence tools is also growing. Parallel efforts are needed to develop a comprehensive understanding of legislation and guidelines for the responsible use of artificial intelligence in healthcare [259]. A recent 10-question survey focused on the use of CV in healthcare was completed by 233 providers and 50 patients and family members. The potential for the use of CV data in lawsuits (81% clinicians) and privacy breaches (50% patients) were major areas of concern [245]. Future work should focus on further exploring provider, patient, and public perceptions and knowledge needs on CV.
9.4. Data Needs and Considerations
Despite the impressive performance of deep learning models on general datasets, achieving accurate results in the medical domain remains challenging. This difficulty arises mainly from the substantial number of parameters in each layer of CNN models. When a sufficient amount of data is available, as found in large CV datasets like ImageNet [260] (1 million images), the model is better able to generalize and overfitting is mitigated. Acquiring a sufficient sample of labeled data for model training within the healthcare system to produce models that are generalizable and statistically fit can be prohibitively expensive. One potential solution to address overfitting is employing models with fewer parameters [261,262]. However, these compact models often struggle to capture intricate features of the dataset, resulting in reduced detection or classification accuracy. To cope with the scarcity of labeled data, data augmentation is used to generate additional training data [114,263]. While this approach partially resolves the problem, the repetition of images may lead to overfitting. Another strategy involves utilizing transfer learning, where the model is initially trained on a large dataset with available labels and then fine-tuned on the smaller medical datasets [115,264]. This approach aims to leverage pre-existing knowledge from the larger dataset to enhance the performance of the medical data. Each of these solutions is a trade-off in model performance and must be weighed in the development and testing stages. Another option to scale the development of CV models in medicine is to use available deep-learning techniques to classify, segment, and detect specific structures or abnormalities. Detectron2 [265], developed by Facebook AI Research (FAIR), offers a high-quality implementation of state-of-the-art object detection and segmentation models. MMDetection [266], another open-source PyTorch library, facilitates the utilization of pre-trained state-of-the-art models and their training on medical datasets. Torchvision, an official PyTorch library, provides general models that can be tailored for use with medical domain datasets. OpenPose [267] stands out as one of the initial open-source and real-time multi-person models designed to identify human body structures, body key points, as well as facial and hand features in visual footage. These are a sampling of available deep learning techniques, and it is important to consider their development and validation prior to use to develop subsequent CV models. Lastly, the training of CV models demands significant computing resources and expertise, including GPUs and AI specialists, which may not be readily available at every institution due to resource constrained- environments. In light of these challenges, many clinician-scientists opt to use traditional machine learning methods, like logistic regression, that limit model development. Future CV studies may explore how federated learning could expand datasets and computational resources [268].
9.5. Computer Vision Datasets
Object detection datasets typically consist of images with annotated bounding boxes and segmented areas depicting objects of interest. The Pascal Visual Object Classes (VOC) [269] dataset stands out as a well-known benchmark, featuring 5000 images across 20 object classes with 12,000 annotations. Another widely used benchmark, the Common Objects in Context (COCO) [270] dataset, offers a substantial dataset of 164,000 images covering 80 object classes, accompanied by 897,000 annotations, encompassing both indoor and outdoor environments. However, in the context of hospital environments, there is currently a lack of sufficient datasets capturing diverse objects under various conditions. For example, the MCIndoor2000 [223] dataset includes 2055 images of three object classes including doors, stairs, and hospital signs. The MYNursingHome [224] dataset focuses on object classification and detection in nursing homes, containing 37,500 images featuring objects commonly found in elderly home care centers, such as toilet seats, tables, and wheelchairs. The Hospital Indoor Object Detection (HIOD) dataset comprises 4417 images covering 56 object categories, including items like surgical lights, IV poles, and bedside monitors, with a total of 51,869 annotations. On average, the images in this dataset contain 10 objects spanning 6.8 object categories [225]. There are several datasets available for medical imaging purposes. The website
This dearth of public datasets is illustrated best by examining the large amount of literature and models developed in neonatal populations. The cumulation of this work over the past two decades has led to a point-of-care mobile application for procedural pain that has passed the feasibility stage [172]. This type of technology could greatly improve pain management not only in neonatal populations but also in adult populations. Several studies aimed at identifying chronic or outpatient pain have used the UNBC-McMaster Pain Archive [160,161,165]. While these images have aided in the development of automated models for pain detection and monitoring in adult outpatient populations, they have not facilitated the expansion of such models into the acute care setting. Public datasets of hospitalized patients across age groups to facilitate this type of modeling are needed [172].
9.6. Limitations of This Review
This review used a broad search strategy. That being said, a systematic review approach was not used, and it is possible that studies involving CV were not included. The CV field is rapidly expanding. Due to that expansion, this review is limited in scope and strove to highlight advances in CV for healthcare clinicians and clinician-scientists (i.e., end-users of technology).
10. Conclusions
This review summarizes the application of CV in healthcare, and we highlight important considerations for the use of CV in healthcare including privacy, safety, and ethical factors. The overall goal is to improve the patient and clinician journey within the industry. There continues to be a paucity of data to train CV and for it to catch up to other industries in its application; substantial work is needed to overcome to barriers of privacy and safety considerations.
This review is based on published research and did not collect any patient identifiers. IRB approval was not needed.
Not applicable. No prospective data collection was completed for this review paper.
Search results and datasheets of extracted data are available upon request to the corresponding author.
All other authors declare no conflict of interest.
Footnotes
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Enlarge this image.Figure 1. The completed journey mapping of the patient and clinician through the healthcare system. This figure depicts a scenario demonstrating the application of computer vision in a hospital setting. Each data point addresses a specific instance where computer vision could effectively enhance the system, optimizing the patient’s experience and the clinical workflow. This system would aim at directing time and resources more efficiently towards patient care management and improvement of patient outcomes.
Enlarge this image.Figure 2. The Venn diagram displays how the main applications of computer vision in healthcare are organized. There are three central domains that we identified through our review: monitoring, job enhancement, and automation. In bold, attention is drawn to applications that are frequently mentioned in the literature. Overlap between categories is shown. The non-bold text highlights potential uses of computer vision that are either in development/testing stages or need proof-of-concept work to be completed.
Display of different types of machine learning models used in computer vision.
| Approach | Supervision | Machine Learning Model | Description |
|---|---|---|---|
| Deep | Supervised | Convolutional Neural Network | Mostly used for classification and segmentation. It includes wide range of model architectures such as ResNet, VGG-net, and AlexNet. |
| Mask-region-based Convolutional Neural Network | CNN type primarily employed for detecting objects in input images. | ||
| YOLO | CNN types primarily employed for image segmentation or classification. | ||
| U-net | Type of CNN mainly used for image segmentation. | ||
| Gated recurrent unit | Type of recurrent neural network tailored for modeling time dependent data to address long-range dependencies in sequential data. | ||
| Long short-term memory (LSTM) | |||
| Vision transformer | Novel category of CNNs. Adopts transformer architecture commonly used in NLP and shows high performance in image classification benchmarks. | ||
| Unsupervised | Convolutional Deep Belief Network (CDBN) | Type of deep generative models that is constructed by stacking max-pooling Convolutional | |
| Autoencoder | A type of neural network that specializes in learning to convert data into a compact and efficient representation, often employed for the purpose of dimensionality reduction. | ||
| Traditional | Supervised | k-nearest neighbors | Assigns class labels or values according to the distance of the input data to the k-nearest neighbors in the training data. |
| Binary Tree | Decision-making algorithm that navigates the tree from root to leaf to make decisions based on specific features or attributes. | ||
| Naïve Bayes | Probabilistic machine learning algorithm that classifies data based on the conditional independence between every pair of features. | ||
| Support vector machine (SVM) | It uses the kernel trick to find a linear decision boundary to separate input data in the transformed space. | ||
| Fuzzy Inference System | A computational model that uses fuzzy logic to perform reasoning on uncertain or imprecise information. | ||
| Fisher’s linear discriminant analysis | Classifies input data based on linear combination of features that represent items in each class. | ||
| Linear Mixed Model | An extension of simple linear models that allow fixed and random effects, useful for complex data | ||
| Logistic/linear regression | This is a statistical model that uses the logistic function to predict the probability of a specific class. | ||
| Supervised | Random forest | Ensemble learning method that comprises multiple trees trained on random subsets of data. The final prediction is aggregated from all trees. | |
| Neural network | It is a conventional machine learning model employed for classification and regression. In comparison to existing deep methods, it exhibits lower accuracy. | ||
| Singular Value Decomposition (SVD) | It decomposes the input feature space into 3 generic and familiar matrices. | ||
| Unsupervised | Fuzzy C-means | A computational model that uses fuzzy logic to perform reasoning on uncertain or imprecise information | |
| Gaussian Mixture Model Segmentation | It uses Gaussian distribution to partition pixels into similar segments |
Uses and categorizations (themes) of computer vision in industry outside healthcare.
| Industry | Themes | |||
|---|---|---|---|---|
| Job Enhancement | Surveillance (S) Monitoring (M) | Automation | Augmented Reality | |
| Agriculture | Monitor crops (M) [ | Weed detection and elimination | ||
| Animal control | Wildlife monitoring (M) [ | |||
| Art security | Forgery detection (S) [ | |||
| Automotive | Parking lot analysis (M) [ | Self-driving cars | ||
| Digital design | Video enhancement [ | |||
| Education | Cheating prevention (S) [ | |||
| Engineering | Importing real life objects into modeling software [ | |||
| Food service | Reduce food waste in restaurants (M) [ | Robotic food delivery | ||
| Gaming | Xbox Kinect [ | |||
| Government | Control traffic lights [ | Detecting natural disasters (M) [ | ||
| Insurance | Insurance appraisals [ | |||
| Law enforcement | Forensic analysis[ | Facial recognition | Speeding enforcement [ | |
| Manufacturing | Defective products on an assembly line [ | |||
| Manufacturing | Workplace inspection [ | |||
| Medical | See | |||
| Military | Terrain Reconnaissance (S) [ | Automate military drones [ | ||
| Movie | Film movie restoration [ | |||
| Retail | Customer behavior analysis (Traffic volume heatmaps) | Detecting defective products (M) [ | Floor-cleaning robots [ | Trying on clothes virtually [ |
| Robotics | Helping robots move around environment | |||
| Social media | Social media recommendation [ | Inappropriate content detection [ | ||
| Space | Tracking asteroids and debris (M) [ | Landing spacecraft [ | ||
| Sports | Finding game highlights in videos [ | Ball tracking in sports [ | ||
| Tech | Facial recognition personal mobile devices (M) [ | Language translation of video and images [ | ||
CV applications in healthcare.
| Areas | Citations | Image Type | Deep Model | Application |
|---|---|---|---|---|
| Medical imaging and diagnosis | [ | CT, F-FDG PET/CT, Chest X-rays | Mask-RCNN, CNN, Transformer, SVM, random forest, k-nearest neighbor | Lung cancer, tuberculosis |
| [ | Iris, cellular retinal, fundus | Binarytree, Random Forest, SVM, neural network, CNN | Changes in vision related to diabetes | |
| [ | HD microscope | Vision transformer, CNN | Cervical cancer | |
| [ | Mammogram, whole slide images, hematoxylin, eosin | YOLO, CNN, random forest, SVM, decision tree, Naïve Bayes, Logistic linear classifier, Linear discriminant classifier, Fischer’s Linear Discriminant analysis, k-nearest neighbor, Autoencoders | Breast cancer, data augmentation | |
| [ | Dermoscopic image | CNN, Gated recurrent unit | Skin cancer detection/segmentation | |
| [ | Endoscopic images, hematoxylin & eosin, whole-slide images | CNN, transformer, U-net | Colorectal, gastrointestinal cancer | |
| [ | Chest X-rays | CNN, transformer, logistic regression | COVID-19 diagnosis | |
| [ | Whole slide images | Vision transformer | Subtyping of papillary renal cell carcinoma | |
| [ | MRI, Histogram, | CNN, Naïve Bayes, Random Forest, Neural | Cancers (brain, bladder, breast, liver, lung, pancreas, prostate, other), | |
| [ | Dual energy X-Ray absorptiometry (DEXA) | SVM | Lumbar spine fractures | |
| Delirium | [ | Surveillance images | CNN, k-nearest neighbors | Delirium monitoring |
| Pain, Agitation, Stress, Level of sedation | [ | Surveillance images, depth image, face images, pain datasets | YOLO, Mask-RCNN, CNN, CDBN, SVM, LSTM, LMM, Neural Network | Activity recognition, detection of pain and discomfort, stress, automated facial analysis for grimacing, agitation, eye localization, depression, anxiety, stress levels, AU |
| Patient deterioration | [ | Color videos | Logistic/linear Regression | Deterioration prediction using AUs |
| Mechanical ventilation | [ | Chest X-rays, ICU videos | U-net, YOLO, TL, Feature descriptor | Need for mechanical ventilation, detect and recognize ventilation objects and positioning, estimate lung volume |
| Mobility | [ | ICU video images | CNN, YOLOv2 | Patient mobilization activities in ICU, NIMS |
| Patient safety | [ | Surgical videos, depth images, video recordings | OpenPose, Yolo, CNN, Mask-RCNN | Surgical team behavioral analysis, patient mobilization activities, hand hygiene, ICU staff monitoring, assessing situational awareness |
| Surgical assistance | [ | Surgical activity images, OR videos | CNN | Robot-assisted surgery, situational awareness in OR |
| Neurological, neurodevelopmental, psychiatric disorders | [ | Whole-body video recording, MRI, PET, patient images | Detectron2, OpenPose, CNN, | Analysis of gain synchrony, balance, Infant neuromotor risk, neurodegenerative disease, behavioral analysis in ASD and ADHD, facial expression in depression, facial weakness |
| Remote monitoring, telemedicine | [ | Surveillance images | Deep reinforcement learning, CNN | In-home elbow rehabilitation |
| Data security and privacy | [ | X-rays, MRI | CNN, Fuzzy CNN | Privacy protections for deep learning algorithms containing medical data |
| Fall detection | [ | Surveillance images | Gaussian Mixture Model, CNN Segmentation, AlphaPose, OpenPose, LSTM | Human fall detection |
| Hospital scene recognition | [ | Indoor images of ICU, hospital, nursing home; pediatric ICU videos | YOLO, CNN, SVM, CATS | ICU and hospital indoor object detection, hand hygiene, ICU activity measurement |
References
1. McDonald, C.J.; Overhage, J.M.; Tierney, W.M.; Dexter, P.R.; Martin, D.K.; Suico, J.G.; Zafar, A.; Schadow, G.; Blevins, L.; Glazener, T. et al. The Regenstrief Medical Record System: A quarter century experience. Int. J. Med. Inform.; 1999; 54, pp. 225-253. [DOI: https://dx.doi.org/10.1016/S1386-5056(99)00009-X]
2. Evans, R.S. Electronic Health Records: Then, Now, and in the Future. Yearb. Med. Inform.; 2016; 25, (Suppl. S1), pp. S48-S61. [DOI: https://dx.doi.org/10.15265/IYS-2016-s006]
3. Tobin, M.J. Why Physiology Is Critical to the Practice of Medicine: A 40-year Personal Perspective. Clin. Chest Med.; 2019; 40, pp. 243-257. [DOI: https://dx.doi.org/10.1016/j.ccm.2019.02.012] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31078207]
4. Prince, S.J. Computer vision: Models, Learning, and Inference; Cambridge University Press: Cambridge, UK, 2012.
5. Burton, A.; Radford, J. Thinking in Perspective: Critical Essays in the Study of Thought Processes; Taylor & Francis: Oxfordshire, UK, 2022.
6. Horn, B.K.; Schunck, B.G. Determining optical flow. Artif. Intell.; 1981; 17, pp. 185-203. [DOI: https://dx.doi.org/10.1016/0004-3702(81)90024-2]
7. Belongie, S.; Carson, C.; Greenspan, H.; Malik, J. Color-and texture-based image segmentation using EM and its application to content-based image retrieval. Proceedings of the Sixth International Conference on Computer Vision; Bombay, India, 7 January 1998; IEEE: Piscataway, NJ, USA, 1998; pp. 675-682.
8. Kirby, M.; Sirovich, L. Application of the Karhunen-Loeve procedure for the characterization of human faces. IEEE Trans. Pattern Anal. Mach. Intell.; 1990; 12, pp. 103-108. [DOI: https://dx.doi.org/10.1109/34.41390]
9. Krizhevsky, A.; Sutskever, I.; Hinton, G.E. Imagenet classification with deep convolutional neural networks. Adv. Neural Inf. Process. Syst.; 2012; 25, pp. 1097-1105. [DOI: https://dx.doi.org/10.1145/3065386]
10. Hansen, U.S. 6 Use Cases for Computer Vision in Insurance. Available online: https://encord.com/blog/computer-vision-use-cases-insurance/ (accessed on 5 January 2024).
11. Dong, S.; Wang, P.; Abbas, K. A survey on deep learning and its applications. Comput. Sci. Rev.; 2021; 40, 100379. [DOI: https://dx.doi.org/10.1016/j.cosrev.2021.100379]
12. Cui, Y.; Chen, R.; Chu, W.; Chen, L.; Tian, D.; Li, Y.; Cao, D. Deep learning for image and point cloud fusion in autonomous driving: A review. IEEE Trans. Intell. Transp. Syst.; 2021; 23, pp. 722-739. [DOI: https://dx.doi.org/10.1109/TITS.2020.3023541]
13. Nawaratne, R.; Alahakoon, D.; De Silva, D.; Yu, X. Spatiotemporal anomaly detection using deep learning for real-time video surveillance. IEEE Trans. Ind. Inform.; 2019; 16, pp. 393-402. [DOI: https://dx.doi.org/10.1109/TII.2019.2938527]
14. Lad, A.; Kanaujia, P.; Soumya, P.; Solanki, Y. Computer Vision enabled Adaptive Speed Limit Control for Vehicle Safety. Proceedings of the 2021 International Conference on Artificial Intelligence and Machine Vision (AIMV); Gandhinagar, India, 24–26 September 2021; pp. 1-5.
15. Sinshaw, N.T.; Assefa, B.G.; Mohapatra, S.K.; Beyene, A.M. Applications of Computer Vision on Automatic Potato Plant Disease Detection: A Systematic Literature Review. Comput. Intell. Neurosci.; 2022; 2022, 7186687. [DOI: https://dx.doi.org/10.1155/2022/7186687]
16. Szeliski, R. Computer Vision: Algorithms and Applications; Springer Nature: Berlin/Heidelberg, Germany, 2022.
17. Le, Q.V. Building high-level features using large scale unsupervised learning. Proceedings of the 2013 IEEE International Conference on Acoustics, Speech and Signal Processing; Vancouver, BC, Canada, 26–31 May 2013; IEEE: Iscataway, NJ, USA, 2013; pp. 8595-8598.
18. Baldi, P. Autoencoders, unsupervised learning, and deep architectures. Proceedings of the ICML Workshop on Unsupervised and Transfer Learning; Bellevue, WA, USA, 2 July 2011; pp. 37-49.
19. Srivastava, N.; Mansimov, E.; Salakhudinov, R. Unsupervised learning of video representations using lstms. Proceedings of the International Conference on Machine Learning; Lille, France, 6–11 July 2015; PMLR: London, UK, 2015; pp. 843-852.
20. Liu, L.; Ouyang, W.; Wang, X.; Fieguth, P.; Chen, J.; Liu, X.; Pietikäinen, M. Deep learning for generic object detection: A survey. Int. J. Comput. Vis.; 2020; 128, pp. 261-318. [DOI: https://dx.doi.org/10.1007/s11263-019-01247-4]
21. Tan, M.; Pang, R.; Le, Q.V. Efficientdet: Scalable and efficient object detection. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition; Seattle, WA, USA, 13–19 June 2020; pp. 10781-10790.
22. Jaeger, P.F.; Kohl, S.A.; Bickelhaupt, S.; Isensee, F.; Kuder, T.A.; Schlemmer, H.-P.; Maier-Hein, K.H. Retina U-Net: Embarrassingly simple exploitation of segmentation supervision for medical object detection. Proceedings of the Machine Learning for Health Workshop; Virtual, 11 December 2020; PMLR: London, UK, 2020; pp. 171-183.
23. Jogin, M.; Madhulika, M.; Divya, G.; Meghana, R.; Apoorva, S. Feature extraction using convolution neural networks (CNN) and deep learning. Proceedings of the 2018 3rd IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT); Bangalore, India, 18–19 May 2018; IEEE: Piscataway, NJ, USA, pp. 2319-2323.
24. Cai, G.; Wei, X.; Li, Y. Privacy-preserving CNN feature extraction and retrieval over medical images. Int. J. Intell. Syst.; 2022; 37, pp. 9267-9289. [DOI: https://dx.doi.org/10.1002/int.22991]
25. Yang, A.; Yang, X.; Wu, W.; Liu, H.; Zhuansun, Y. Research on feature extraction of tumor image based on convolutional neural network. IEEE Access; 2019; 7, pp. 24204-24213. [DOI: https://dx.doi.org/10.1109/ACCESS.2019.2897131]
26. Tajbakhsh, N.; Shin, J.Y.; Gurudu, S.R.; Todd Hurst, R.; Kendall, C.B.; Gotway, M.B.; Liang, J. On the necessity of fine-tuned convolutional neural networks for medical imaging. Deep Learning and Convolutional Neural Networks for Medical Image Computing: Precision Medicine, High Performance and Large-Scale Datasets; Springer: Berlin/Heidelberg, Germany, 2017; pp. 181-193.
27. Dutta, P.; Upadhyay, P.; De, M.; Khalkar, R. Medical image analysis using deep convolutional neural networks: CNN architectures and transfer learning. Proceedings of the 2020 International Conference on Inventive Computation Technologies (ICICT); Coimbatore, India, 26–28 February 2020; IEEE: Piscataway, NJ, USA, pp. 175-180.
28. Lee, K.-S.; Kim, J.Y.; Jeon, E.-t.; Choi, W.S.; Kim, N.H.; Lee, K.Y. Evaluation of scalability and degree of fine-tuning of deep convolutional neural networks for COVID-19 screening on chest X-ray images using explainable deep-learning algorithm. J. Pers. Med.; 2020; 10, 213. [DOI: https://dx.doi.org/10.3390/jpm10040213] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33171723]
29. Kesav, N.; Jibukumar, M. Efficient and low complex architecture for detection and classification of Brain Tumor using RCNN with Two Channel CNN. J. King Saud Univ.-Comput. Inf. Sci.; 2022; 34, pp. 6229-6242. [DOI: https://dx.doi.org/10.1016/j.jksuci.2021.05.008]
30. Deepak, S.; Ameer, P. Brain tumor classification using deep CNN features via transfer learning. Comput. Biol. Med.; 2019; 111, 103345. [DOI: https://dx.doi.org/10.1016/j.compbiomed.2019.103345] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31279167]
31. Chiao, J.-Y.; Chen, K.-Y.; Liao, K.Y.-K.; Hsieh, P.-H.; Zhang, G.; Huang, T.-C. Detection and classification the breast tumors using mask R-CNN on sonograms. Medicine; 2019; 98, e15200. [DOI: https://dx.doi.org/10.1097/MD.0000000000015200] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31083152]
32. Jorda, M.; Valero-Lara, P.; Pena, A.J. Performance evaluation of cudnn convolution algorithms on nvidia volta gpus. IEEE Access; 2019; 7, pp. 70461-70473. [DOI: https://dx.doi.org/10.1109/ACCESS.2019.2918851]
33. Ahramovich, A. Top Applications for Computer Vision in Sports. Available online: https://builtin.com/articles/computer-vision-sports (accessed on 5 January 2024).
34. Hao, N.; Ruan, S.; Song, Y.; Chen, J.; Tian, L. The Establishment of a precise intelligent evaluation system for sports events: Diving. Heliyon; 2023; 9, e21361. [DOI: https://dx.doi.org/10.1016/j.heliyon.2023.e21361]
35. Rodriguez-Ortega, Y.; Ballesteros, D.M.; Renza, D. Copy-Move Forgery Detection (CMFD) Using Deep Learning for Image and Video Forensics. J. Imaging; 2021; 7, 59. [DOI: https://dx.doi.org/10.3390/jimaging7030059]
36. Tyagi, S.; Yadav, D. ForensicNet: Modern convolutional neural network-based image forgery detection network. J. Forensic Sci.; 2023; 68, pp. 461-469. [DOI: https://dx.doi.org/10.1111/1556-4029.15210] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36719038]
37. Auberson, M.; Baechler, S.; Zasso, M.; Genessay, T.; Patiny, L.; Esseiva, P. Development of a systematic computer vision-based method to analyse and compare images of false identity documents for forensic intelligence purposes-Part I: Acquisition, calibration and validation issues. Forensic Sci. Int.; 2016; 260, pp. 74-84. [DOI: https://dx.doi.org/10.1016/j.forsciint.2016.01.016] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26836242]
38. Story, D.; Kacira, M. Design and implementation of a computer vision-guided greenhouse crop diagnostics system. Mach. Vis. Appl.; 2015; 26, pp. 495-506. [DOI: https://dx.doi.org/10.1007/s00138-015-0670-5]
39. Liu, J.; Wang, X. Plant diseases and pests detection based on deep learning: A review. Plant Methods; 2021; 17, 22. [DOI: https://dx.doi.org/10.1186/s13007-021-00722-9] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33627131]
40. Hussain, M.; Al-Aqrabi, H.; Munawar, M.; Hill, R.; Alsboui, T. Domain Feature Mapping with YOLOv7 for Automated Edge-Based Pallet Racking Inspections. Sensors; 2022; 22, 6927. [DOI: https://dx.doi.org/10.3390/s22186927] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36146273]
41. Panahi, R.; Louis, J.; Podder, A.; Swanson, C.; Pless, S. Bottleneck Detection in Modular Construction Factories Using Computer Vision. Sensors; 2023; 23, 3982. [DOI: https://dx.doi.org/10.3390/s23083982]
42. Masalkhi, M.; Waisberg, E.; Ong, J.; Zaman, N.; Sarker, P.; Lee, A.G.; Tavakkoli, A. Apple Vision Pro for Ophthalmology and Medicine. Ann. Biomed. Eng.; 2023; 51, pp. 2643-2646. [DOI: https://dx.doi.org/10.1007/s10439-023-03283-1] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37332003]
43. Nakamura, R.; Izutsu, M.; Hatakeyama, S. Estimation Method of Clothes Size for Virtual Fitting Room with Kinect Sensor. Proceedings of the 2013 IEEE International Conference on Systems, Man, and Cybernetics; Manchester, UK, 13–16 October 2013; IEEE: Piscataway, NJ, USA, 2013; pp. 3733-3738.
44. Yuan, M.; Khan, I.R.; Farbiz, F.; Yao, S.; Niswar, A.; Foo, M.H. A Mixed Reality Virtual Clothes Try-On System. IEEE Trans. Multimed.; 2013; 15, pp. 1958-1968. [DOI: https://dx.doi.org/10.1109/TMM.2013.2280560]
45. Zhang, W.; Begole, B.; Chu, M.; Liu, J.; Yee, N. Real-time clothes comparison based on multi-view vision. Proceedings of the 2008 Second ACM/IEEE International Conference on Distributed Smart Cameras; Palo Alto, CA, USA, 7–11 September 2008; IEEE: Piscataway, NJ, USA, 2008; pp. 1-10.
46. Budrionis, A.; Plikynas, D.; Daniušis, P.; Indrulionis, A. Smartphone-based computer vision travelling aids for blind and visually impaired individuals: A systematic review. Assist. Technol.; 2022; 34, pp. 178-194. [DOI: https://dx.doi.org/10.1080/10400435.2020.1743381]
47. Tapu, R.; Mocanu, B.; Zaharia, T. A computer vision system that ensure the autonomous navigation of blind people. Proceedings of the 2013 E-Health and Bioengineering Conference (EHB); Iasi, Romania, 21–23 November 2013; IEEE: Piscataway, NJ, USA, 2013; pp. 1-4.
48. Sivan, S.; Darsan, G. Computer vision based assistive technology for blind and visually impaired people. Proceedings of Proceedings of the 7th International Conference on Computing Communication and Networking Technologies; Dallas, TX, USA, 6–8 July 2016; pp. 1-8.
49. Kraus, M. Keeping Track of Animals in the Wild with Computer Vision. Available online: https://www.vantage-ai.com/en/blog/keeping-track-of-animals-in-the-wild-with-computer-vision (accessed on 5 January 2024).
50. Boesch, G. Animal Monitoring with Computer Vision—Case Study. Available online: https://viso.ai/applications/animal-monitoring/ (accessed on 10 January 2024).
51. Spratt, E.L.; Elgammal, A. Computational beauty: Aesthetic judgment at the intersection of art and science. Proceedings of the Computer Vision-ECCV 2014 Workshops; Zurich, Switzerland, 6–7, 12 September 2014; Proceedings, Part I 13 Springer: Berlin/Heidelberg, Germany, 2014; pp. 35-53.
52. Klingler, N. Viso Suite Guide: Develop a Computer Vision Parking Lot Occupancy Application—Viso.ai. Available online: https://viso.ai/product/computer-vision-parking-lot-occupancy-tutorial/ (accessed on 15 July 2023).
53. Vasluianu, F.; Timofte, R. Efficient video enhancement transformer. Proceedings of the 2022 IEEE International Conference on Image Processing (ICIP); Bordeaux France, 16–19 October 2022; IEEE: Piscataway, NJ, USA, 2022; pp. 4068-4072.
54. Huihui, Y.; Daoliang, L.; Yingyi, C. A state-of-the-art review of image motion deblurring techniques in precision agriculture. Heliyon; 2023; 9, e17332. [DOI: https://dx.doi.org/10.1016/j.heliyon.2023.e17332]
55. Alsabhan, W. Student Cheating Detection in Higher Education by Implementing Machine Learning and LSTM Techniques. Sensors; 2023; 23, 4149. [DOI: https://dx.doi.org/10.3390/s23084149]
56. Thompson, B. ActionVFX|How To Transform Real-World Objects Into 3D Assets. Available online: https://www.actionvfx.com/blog/how-to-transform-real-world-objects-into-3d-assets (accessed on 15 December 2023).
57. Combating Food Waste Using AI and Computer Vision—Cogniphi. Available online: https://cogniphi.com/combating-food-waste-using-ai-and-computer-vision/ (accessed on 15 July 2023).
58. An Introduction to the Kinect Sensor|Microsoft Press Store. Available online: https://www.microsoftpressstore.com/articles/article.aspx?p=2201646 (accessed on 10 January 2024).
59. Le, N.V.; Qarmout, M.; Zhang, Y.; Zhou, H.; Yang, C. Hand Gesture Recognition System for Games. Proceedings of the 2021 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE); Brisbane, Australia, 8–10 December 2021; IEEE: Piscataway, NJ, USA, 2021; pp. 1-6.
60. Nodado, J.T.G.; Morales, H.C.P.; Abugan, M.A.P.; Olisea, J.L.; Aralar, A.C.; Loresco, P.J.M. Intelligent traffic light system using computer vision with android monitoring and control. Proceedings of the TENCON 2018—2018 IEEE Region 10 Conference; Jeju Island, Republic of Korea, 28–31 October 2018; IEEE: Piscataway, NJ, USA, 2018; pp. 2461-2466.
61. Rosebrock, A. Detecting Natural Disasters with Keras and Deep Learning—PyImageSearch. Available online: https://pyimagesearch.com/2019/11/11/detecting-natural-disasters-with-keras-and-deep-learning/ (accessed on 5 January 2024).
62. Yilmaz, H. Top Applications of Computer Vision in Insurance (2022 Guide). Available online: https://www.plugger.ai/blog/top-applications-of-computer-vision-in-insurance-2022-guide (accessed on 5 January 2024).
63. Stefanovskyi, O. Computer Vision in Insurance: Vehicle Damage Assessment Case. Available online: https://intelliarts.com/blog/computer-vision-in-insurance-vehicle-damage-assessment/ (accessed on 5 January 2024).
64. Thakkar, P.; Patel, D.; Hirpara, I.; Jagani, J.; Patel, S.; Shah, M.; Kshirsagar, A. A comprehensive review on computer vision and fuzzy logic in forensic science application. Ann. Data Sci.; 2023; 10, pp. 761-785. [DOI: https://dx.doi.org/10.1007/s40745-022-00408-6]
65. Facial Recognition in a Crowd. Available online: https://kintronics.com/solutions/ip-camera-systems/facial-recognition/ (accessed on 10 January 2024).
66. Identity Verification with Deep Learning: ID-Selfie Matching Method. Available online: https://medium.com/coinmonks/identity-verification-with-deep-learning-id-selfie-matching-method-be56d72be632 (accessed on 10 January 2024).
67. Wiggers, K. Ambient’s Computer Vision Detects Dangerous Behaviors, but Raises Privacy Concerns. Available online: https://venturebeat.com/uncategorized/ambients-computer-vision-detects-dangerous-behaviors-but-raises-privacy-concerns/ (accessed on 10 January 2024).
68. Novacura. Quality Inspections of the Production Line Using Computer Vision and Novacura Flow. Available online: https://www.novacura.com/computer-vision-quality-inspections/ (accessed on 10 January 2024).
69. CompScience. CompScience Workplace Safety Analytics. Available online: https://www.compscience.com/about-us/ (accessed on 10 January 2024).
70. Simpson, S. Tactical Multi-Drone Mapping Demonstrated to US Military|Unmanned Systems Technology. Available online: https://www.unmannedsystemstechnology.com/2022/08/tactical-multi-drone-mapping-demonstrated-to-us-military/ (accessed on 15 July 2023).
71. Maltsev, A. Drones at War and Computer Vision. Available online: https://medium.com/@zlodeibaal/drones-at-war-and-computer-vision-a16b8063be7b (accessed on 15 July 2023).
72. Liu, L.; Catelli, E.; Katsaggelos, A.; Sciutto, G.; Mazzeo, R.; Milanic, M.; Stergar, J.; Prati, S.; Walton, M. Digital restoration of colour cinematic films using imaging spectroscopy and machine learning. Sci. Rep.; 2022; 12, 21982. [DOI: https://dx.doi.org/10.1038/s41598-022-25248-5]
73. Tabernik, D.; Šela, S.; Skvarč, J.; Skočaj, D. Deep-learning-based computer vision system for surface-defect detection. Proceedings of the Computer Vision Systems: 12th International Conference, ICVS 2019; Thessaloniki, Greece, 23–25 September 2019; Proceedings 12 Springer: Berlin/Heidelberg, Germany, 2019; pp. 490-500.
74. Canedo, D.; Fonseca, P.; Georgieva, P.; Neves, A.J. A deep learning-based dirt detection computer vision system for floor-cleaning robots with improved data collection. Technologies; 2021; 9, 94. [DOI: https://dx.doi.org/10.3390/technologies9040094]
75. Savit, A.; Damor, A. Revolutionizing Retail Stores with Computer Vision and Edge AI: A Novel Shelf Management System. Proceedings of the 2023 2nd International Conference on Applied Artificial Intelligence and Computing (ICAAIC); Salem, India, 4–6 May 2023; IEEE: Piscataway, NJ, USA, 2023; pp. 69-74.
76. Smink, A.R.; Frowijn, S.; van Reijmersdal, E.A.; van Noort, G.; Neijens, P.C. Try online before you buy: How does shopping with augmented reality affect brand responses and personal data disclosure. Electron. Commer. Res. Appl.; 2019; 35, 100854. [DOI: https://dx.doi.org/10.1016/j.elerap.2019.100854]
77. Vidal, C. Technology Primer: Social Media Recommendation Algorithms. Available online: https://www.belfercenter.org/publication/technology-primer-social-media-recommendation-algorithms (accessed on 20 July 2023).
78. Yousaf, K.; Nawaz, T. A deep learning-based approach for inappropriate content detection and classification of youtube videos. IEEE Access; 2022; 10, pp. 16283-16298. [DOI: https://dx.doi.org/10.1109/ACCESS.2022.3147519]
79. Yekkehkhany, B.; Shokri, P.; Zadeh, A. A Computer Vision Approach for Detection of Asteroids/comets in Space Satellite Images. Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci.; 2020; 43, pp. 1185-1190. [DOI: https://dx.doi.org/10.5194/isprs-archives-XLIII-B3-2020-1185-2020]
80. Yang, X.; Bobkov, A. Development of a vision system for safe and high-precision soft landing on the Moon. Procedia Comput. Sci.; 2021; 186, pp. 503-511. [DOI: https://dx.doi.org/10.1016/j.procs.2021.04.171]
81. Li, S. Computer vision based autonomous navigation for pin-point landing robotic spacecraft on asteroids. Intelligent Robotics and Applications: First International Conference, ICIRA 2008, Wuhan, China, 15–17 October 2008; Proceedings, Part II 1 Springer: Berlin/Heidelberg, Germany, 2008; pp. 1115-1126.
82. Saskovec, P. AI-Generated Sports Highlights: Different Approaches—KDnuggets. Available online: https://www.kdnuggets.com/ai-generated-sports-highlights-different-approaches (accessed on 5 July 2023).
83. Computer Vision in Sports: Applications and Challenges SuperAnnotate. Available online: https://www.superannotate.com/blog/computer-vision-in-sports (accessed on 5 July 2023).
84. Cioppa, A.; Deliege, A.; Giancola, S.; Ghanem, B.; Van Droogenbroeck, M. Scaling up SoccerNet with multi-view spatial localization and re-identification. Sci. Data; 2022; 9, 355. [DOI: https://dx.doi.org/10.1038/s41597-022-01469-1]
85. About Face ID Advanced Technology—Apple Support. Available online: https://support.apple.com/en-us/102381 (accessed on 20 July 2023).
86. Live Translate on Pixel Phones: Meet Your New Interpreter. Available online: https://store.google.com/intl/en/ideas/articles/meet-your-helpful-interpreter-pixel-6/ (accessed on 20 July 2023).
87. Azure Kinect DK—Develop AI Models|Microsoft Azure. Available online: https://azure.microsoft.com/en-us/products/kinect-dk (accessed on 20 July 2023).
88. CareAI. Sensor Technology. Available online: https://www.care.ai/sensors.html (accessed on 18 January 2024).
89. Artisight. Available online: https://artisight.com/ (accessed on 18 January 2024).
90. Ocuvera. Our Solution. Available online: https://ocuvera.com/our-solution/ (accessed on 18 January 2024).
91. VirtuSense. Available online: https://www.virtusense.ai/ (accessed on 18 January 2024).
92. Inspiren. Available online: https://inspiren.com/solutions/ (accessed on 18 January 2024).
93. Philips. AI Enabled Solutions. Available online: https://www.philips.com/a-w/about/artificial-intelligence/ai-enabled-solutions (accessed on 18 January 2024).
94. AI, S. Philips and Silo AI Develop Computer Vision to Improve the Accuracy of Radiotherapy. Available online: https://www.silo.ai/blog/philips-and-silo-ai-develop-computer-vision-to-improve-the-accuracy-of-radiotherapy (accessed on 18 January 2024).
95. Gao, J.; Yang, Y.; Lin, P.; Park, D.S. Computer Vision in Healthcare Applications. J. Healthc. Eng.; 2018; 2018, 5157020. [DOI: https://dx.doi.org/10.1155/2018/5157020]
96. Olveres, J.; González, G.; Torres, F.; Moreno-Tagle, J.C.; Carbajal-Degante, E.; Valencia-Rodríguez, A.; Méndez-Sánchez, N.; Escalante-Ramírez, B. What is new in computer vision and artificial intelligence in medical image analysis applications. Quant. Imaging Med. Surg.; 2021; 11, pp. 3830-3853. [DOI: https://dx.doi.org/10.21037/qims-20-1151] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34341753]
97. Isikli Esener, I.; Ergin, S.; Yuksel, T. A New Feature Ensemble with a Multistage Classification Scheme for Breast Cancer Diagnosis. J. Healthc. Eng.; 2017; 2017, 3895164. [DOI: https://dx.doi.org/10.1155/2017/3895164] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29065592]
98. Yin, G.; Song, Y.; Li, X.; Zhu, L.; Su, Q.; Dai, D.; Xu, W. Prediction of mediastinal lymph node metastasis based on 18 F-FDG PET/CT imaging using support vector machine in non-small cell lung cancer. Eur. Radiol.; 2021; 31, pp. 3983-3992. [DOI: https://dx.doi.org/10.1007/s00330-020-07466-5] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33201286]
99. Lakhani, P.; Sundaram, B. Deep learning at chest radiography: Automated classification of pulmonary tuberculosis by using convolutional neural networks. Radiology; 2017; 284, pp. 574-582. [DOI: https://dx.doi.org/10.1148/radiol.2017162326] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28436741]
100. Ardila, D.; Kiraly, A.P.; Bharadwaj, S.; Choi, B.; Reicher, J.J.; Peng, L.; Tse, D.; Etemadi, M.; Ye, W.; Corrado, G. End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography. Nat. Med.; 2019; 25, pp. 954-961. [DOI: https://dx.doi.org/10.1038/s41591-019-0447-x] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31110349]
101. Raza, R.; Zulfiqar, F.; Khan, M.O.; Arif, M.; Alvi, A.; Iftikhar, M.A.; Alam, T. Lung-EffNet: Lung cancer classification using EfficientNet from CT-scan images. Eng. Appl. Artif. Intell.; 2023; 126, 106902. [DOI: https://dx.doi.org/10.1016/j.engappai.2023.106902]
102. Sun, R.; Pang, Y.; Li, W. Efficient Lung Cancer Image Classification and Segmentation Algorithm Based on an Improved Swin Transformer. Electronics; 2023; 12, 1024. [DOI: https://dx.doi.org/10.3390/electronics12041024]
103. Said, Y.; Alsheikhy, A.A.; Shawly, T.; Lahza, H. Medical images segmentation for lung cancer diagnosis based on deep learning architectures. Diagnostics; 2023; 13, 546. [DOI: https://dx.doi.org/10.3390/diagnostics13030546] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36766655]
104. Samant, P.; Agarwal, R. Comparative analysis of classification based algorithms for diabetes diagnosis using iris images. J. Med. Eng. Technol.; 2018; 42, pp. 35-42. [DOI: https://dx.doi.org/10.1080/03091902.2017.1412521]
105. Jena, P.K.; Khuntia, B.; Palai, C.; Nayak, M.; Mishra, T.K.; Mohanty, S.N. A novel approach for diabetic retinopathy screening using asymmetric deep learning features. Big Data Cogn. Comput.; 2023; 7, 25. [DOI: https://dx.doi.org/10.3390/bdcc7010025]
106. Kothadiya, D.; Rehman, A.; Abbas, S.; Alamri, F.S.; Saba, T. Attention-based deep learning framework to recognize diabetes disease from cellular retinal images. Biochem. Cell Biol.; 2023; 101, pp. 550-561. [DOI: https://dx.doi.org/10.1139/bcb-2023-0151]
107. Pacal, I. MaxCerVixT: A Novel Lightweight Vision Transformer-Based Approach for Precise Cervical Cancer Detection. Knowl. -Based Syst.; 2024; 289, 111482. [DOI: https://dx.doi.org/10.1016/j.knosys.2024.111482]
108. Abd-Alhalem, S.M.; Marie, H.S.; El-Shafai, W.; Altameem, T.; Rathore, R.S.; Hassan, T.M. Cervical cancer classification based on a bilinear convolutional neural network approach and random projection. Eng. Appl. Artif. Intell.; 2024; 127, 107261. [DOI: https://dx.doi.org/10.1016/j.engappai.2023.107261]
109. Attallah, O. Cervical cancer diagnosis based on multi-domain features using deep learning enhanced by handcrafted descriptors. Appl. Sci.; 2023; 13, 1916. [DOI: https://dx.doi.org/10.3390/app13031916]
110. Payabvash, S.; Aboian, M.; Tihan, T.; Cha, S. Machine learning decision tree models for differentiation of posterior fossa tumors using diffusion histogram analysis and structural MRI findings. Front. Oncol.; 2020; 10, 71. [DOI: https://dx.doi.org/10.3389/fonc.2020.00071]
111. Olberg, S.; Choi, B.S.; Park, I.; Liang, X.; Kim, J.S.; Deng, J.; Yan, Y.; Jiang, S.; Park, J.C. Ensemble learning and personalized training for the improvement of unsupervised deep learning-based synthetic CT reconstruction. Med. Phys.; 2023; 50, pp. 1436-1449. [DOI: https://dx.doi.org/10.1002/mp.16087] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36336718]
112. Bejnordi, B.E.; Veta, M.; Van Diest, P.J.; Van Ginneken, B.; Karssemeijer, N.; Litjens, G.; Van Der Laak, J.A.; Hermsen, M.; Manson, Q.F.; Balkenhol, M. Diagnostic assessment of deep learning algorithms for detection of lymph node metastases in women with breast cancer. JAMA; 2017; 318, pp. 2199-2210. [DOI: https://dx.doi.org/10.1001/jama.2017.14585]
113. Tellez, D.; Balkenhol, M.; Otte-Höller, I.; van de Loo, R.; Vogels, R.; Bult, P.; Wauters, C.; Vreuls, W.; Mol, S.; Karssemeijer, N. Whole-slide mitosis detection in H&E breast histology using PHH3 as a reference to train distilled stain-invariant convolutional networks. IEEE Trans. Med. Imaging; 2018; 37, pp. 2126-2136.
114. Hussain, Z.; Gimenez, F.; Yi, D.; Rubin, D. Differential data augmentation techniques for medical imaging classification tasks. AMIA Annu. Symp. Proc.; 2017; 2017, pp. 979-984.
115. Ravishankar, H.; Sudhakar, P.; Venkataramani, R.; Thiruvenkadam, S.; Annangi, P.; Babu, N.; Vaidya, V. Understanding the mechanisms of deep transfer learning for medical images. Deep Learning and Data Labeling for Medical Applications: First International Workshop, LABELS 2016, and Second International Workshop, DLMIA 2016, Held in Conjunction with MICCAI 2016, Athens, Greece, 21 October 2016; Proceedings 1 Springer: Berlin/Heidelberg, Germany, 2016; pp. 188-196.
116. Prinzi, F.; Insalaco, M.; Orlando, A.; Gaglio, S.; Vitabile, S. A YOLO-based model for breast cancer detection in mammograms. Cogn. Comput.; 2024; 16, pp. 107-120. [DOI: https://dx.doi.org/10.1007/s12559-023-10189-6]
117. Luo, L.; Wang, X.; Lin, Y.; Ma, X.; Tan, A.; Chan, R.; Vardhanabhuti, V.; Chu, W.C.; Cheng, K.-T.; Chen, H. Deep learning in breast cancer imaging: A decade of progress and future directions. IEEE Rev. Biomed. Eng.; 2024; pp. 1-20. [DOI: https://dx.doi.org/10.1109/RBME.2024.3357877]
118. Minh, T.C.; Quoc, N.K.; Cong Vinh, P.; Nhu Phu, D.; Chi, V.X.; Tan, H.M. UGGNet: Bridging U-Net and VGG for Advanced Breast Cancer Diagnosis. arXiv; 2024; [DOI: https://dx.doi.org/10.4108/eetcasa.4681]
119. Xu, W.; He, J.; Shu, Y. DeepHealth: Deep representation learning with autoencoders for healthcare prediction. Proceedings of the 2020 IEEE Symposium Series on Computational Intelligence (SSCI); Canberra, Australia, 1–4 December 2020; IEEE: Piscataway, NJ, USA, 2020; pp. 42-49.
120. Chanda, D.; Onim, M.S.H.; Nyeem, H.; Ovi, T.B.; Naba, S.S. DCENSnet: A new deep convolutional ensemble network for skin cancer classification. Biomed. Signal Process. Control; 2024; 89, 105757. [DOI: https://dx.doi.org/10.1016/j.bspc.2023.105757]
121. Akilandasowmya, G.; Nirmaladevi, G.; Suganthi, S.; Aishwariya, A. Skin cancer diagnosis: Leveraging deep hidden features and ensemble classifiers for early detection and classification. Biomed. Signal Process. Control; 2024; 88, 105306. [DOI: https://dx.doi.org/10.1016/j.bspc.2023.105306]
122. Hu, B.; Zhou, P.; Yu, H.; Dai, Y.; Wang, M.; Tan, S.; Sun, Y. LeaNet: Lightweight U-shaped architecture for high-performance skin cancer image segmentation. Comput. Biol. Med.; 2024; 169, 107919. [DOI: https://dx.doi.org/10.1016/j.compbiomed.2024.107919] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38176212]
123. Zhang, L.; Zhang, J.; Gao, W.; Bai, F.; Li, N.; Ghadimi, N. A deep learning outline aimed at prompt skin cancer detection utilizing gated recurrent unit networks and improved orca predation algorithm. Biomed. Signal Process. Control; 2024; 90, 105858. [DOI: https://dx.doi.org/10.1016/j.bspc.2023.105858]
124. Ponzio, F.; Macii, E.; Ficarra, E.; Di Cataldo, S. Colorectal cancer classification using deep convolutional networks. Proceedings of the 11th International Joint Conference on Biomedical Engineering Systems and Technologies; Funchal, Portugal, 19–21 January 2018; pp. 58-66.
125. Choi, K.; Choi, S.J.; Kim, E.S. Computer-Aided diagonosis for colorectal cancer using deep learning with visual explanations. Proceedings of the 2020 42nd annual international conference of the IEEE engineering in Medicine & Biology Society (EMBC); Montreal, QC, Canada, 20–24 July 2020; IEEE: Piscataway, NJ, USA, 2020; pp. 1156-1159.
126. Khan, A.; Brouwer, N.; Blank, A.; Müller, F.; Soldini, D.; Noske, A.; Gaus, E.; Brandt, S.; Nagtegaal, I.; Dawson, H. Computer-assisted diagnosis of lymph node metastases in colorectal cancers using transfer learning with an ensemble model. Mod. Pathol.; 2023; 36, 100118. [DOI: https://dx.doi.org/10.1016/j.modpat.2023.100118] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36805793]
127. Su, Y.; Bai, Y.; Zhang, B.; Zhang, Z.; Wang, W. Hat-net: A hierarchical transformer graph neural network for grading of colorectal cancer histology images. Proceedings of the British Machine Vision Conference; Online, 22–25 November 2021.
128. Sabottke, C.F.; Breaux, M.A.; Spieler, B.M. Estimation of age in unidentified patients via chest radiography using convolutional neural network regression. Emerg. Radiol.; 2020; 27, pp. 463-468. [DOI: https://dx.doi.org/10.1007/s10140-020-01782-5]
129. Khalif, K.M.N.K.; Chaw Seng, W.; Gegov, A.; Bakar, A.S.A.; Shahrul, N.A. Integrated Generative Adversarial Networks and Deep Convolutional Neural Networks for Image Data Classification: A Case Study for COVID-19. Information; 2024; 15, 58. [DOI: https://dx.doi.org/10.3390/info15010058]
130. Mezina, A.; Burget, R. Detection of post-COVID-19-related pulmonary diseases in X-ray images using Vision Transformer-based neural network. Biomed. Signal Process. Control; 2024; 87, 105380. [DOI: https://dx.doi.org/10.1016/j.bspc.2023.105380]
131. Varde, A.S.; Karthikeyan, D.; Wang, W. Facilitating COVID recognition from X-rays with computer vision models and transfer learning. Multimed. Tools Appl.; 2024; 83, pp. 807-838. [DOI: https://dx.doi.org/10.1007/s11042-023-15744-9] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37362714]
132. Jin, C.; Chen, W.; Cao, Y.; Xu, Z.; Tan, Z.; Zhang, X.; Deng, L.; Zheng, C.; Zhou, J.; Shi, H. Development and evaluation of an artificial intelligence system for COVID-19 diagnosis. Nat. Commun.; 2020; 11, 5088. [DOI: https://dx.doi.org/10.1038/s41467-020-18685-1] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33037212]
133. Yoo, S.H.; Geng, H.; Chiu, T.L.; Yu, S.K.; Cho, D.C.; Heo, J.; Choi, M.S.; Choi, I.H.; Cung Van, C.; Nhung, N.V. Deep learning-based decision-tree classifier for COVID-19 diagnosis from chest X-ray imaging. Front. Med.; 2020; 7, 427. [DOI: https://dx.doi.org/10.3389/fmed.2020.00427] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32760732]
134. Gao, Z.; Hong, B.; Zhang, X.; Li, Y.; Jia, C.; Wu, J.; Wang, C.; Meng, D.; Li, C. Instance-based vision transformer for subtyping of papillary renal cell carcinoma in histopathological image. Medical Image Computing and Computer Assisted Intervention–MICCAI 2021: 24th International Conference, Strasbourg, France, 27 September–1 October 2021; Proceedings, Part VIII 24 Springer: Berlin/Heidelberg, Germany, 2021; pp. 299-308.
135. Khosravi, P.; Lysandrou, M.; Eljalby, M.; Li, Q.; Kazemi, E.; Zisimopoulos, P.; Sigaras, A.; Brendel, M.; Barnes, J.; Ricketts, C. A deep learning approach to diagnostic classification of prostate cancer using pathology–radiology fusion. J. Magn. Reson. Imaging; 2021; 54, pp. 462-471. [DOI: https://dx.doi.org/10.1002/jmri.27599] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33719168]
136. Kather, J.N.; Pearson, A.T.; Halama, N.; Jäger, D.; Krause, J.; Loosen, S.H.; Marx, A.; Boor, P.; Tacke, F.; Neumann, U.P. Deep learning can predict microsatellite instability directly from histology in gastrointestinal cancer. Nat. Med.; 2019; 25, pp. 1054-1056. [DOI: https://dx.doi.org/10.1038/s41591-019-0462-y]
137. Liu, M.; Zhang, D.; Shen, D. Alzheimer’s Disease Neuroimaging Initiative. Hierarchical fusion of features and classifier decisions for Alzheimer’s disease diagnosis. Hum. Brain Mapp.; 2014; 35, pp. 1305-1319. [DOI: https://dx.doi.org/10.1002/hbm.22254] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/23417832]
138. Li, W.; Jia, F.; Hu, Q. Automatic segmentation of liver tumor in CT images with deep convolutional neural networks. J. Comput. Commun.; 2015; 3, pp. 146-151. [DOI: https://dx.doi.org/10.4236/jcc.2015.311023]
139. Gao, S.; Lima, D. A review of the application of deep learning in the detection of Alzheimer’s disease. Int. J. Cogn. Comput. Eng.; 2022; 3, pp. 1-8. [DOI: https://dx.doi.org/10.1016/j.ijcce.2021.12.002]
140. Helaly, H.A.; Badawy, M.; Haikal, A.Y. Deep Learning Approach for Early Detection of Alzheimer’s Disease. Cogn. Comput.; 2022; 14, pp. 1711-1727. [DOI: https://dx.doi.org/10.1007/s12559-021-09946-2]
141. Kundisch, A.; Hönning, A.; Mutze, S.; Kreissl, L.; Spohn, F.; Lemcke, J.; Sitz, M.; Sparenberg, P.; Goelz, L. Deep learning algorithm in detecting intracranial hemorrhages on emergency computed tomographies. PLoS ONE; 2021; 16, e0260560. [DOI: https://dx.doi.org/10.1371/journal.pone.0260560]
142. Thung, K.-H.; Wee, C.-Y.; Yap, P.-T.; Shen, D.; Initiative, A.s.D.N. Neurodegenerative disease diagnosis using incomplete multi-modality data via matrix shrinkage and completion. NeuroImage; 2014; 91, pp. 386-400. [DOI: https://dx.doi.org/10.1016/j.neuroimage.2014.01.033] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24480301]
143. Jiang, X.; Hu, Z.; Wang, S.; Zhang, Y. Deep Learning for Medical Image-Based Cancer Diagnosis. Cancers; 2023; 15, 3608. [DOI: https://dx.doi.org/10.3390/cancers15143608] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37509272]
144. John, J.; Ravikumar, A.; Abraham, B. Prostate cancer prediction from multiple pretrained computer vision model. Health Technol.; 2021; 11, pp. 1003-1011. [DOI: https://dx.doi.org/10.1007/s12553-021-00586-y]
145. Kott, O.; Linsley, D.; Amin, A.; Karagounis, A.; Jeffers, C.; Golijanin, D.; Serre, T.; Gershman, B. Development of a deep learning algorithm for the histopathologic diagnosis and Gleason grading of prostate cancer biopsies: A pilot study. Eur. Urol. Focus; 2021; 7, pp. 347-351. [DOI: https://dx.doi.org/10.1016/j.euf.2019.11.003] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31767543]
146. Rampun, A.; Chen, Z.; Malcolm, P.; Tiddeman, B.; Zwiggelaar, R. Computer-aided diagnosis: Detection and localization of prostate cancer within the peripheral zone. Int. J. Numer. Methods Biomed. Eng.; 2016; 32, e02745. [DOI: https://dx.doi.org/10.1002/cnm.2745]
147. Brunese, L.; Mercaldo, F.; Reginelli, A.; Santone, A. An ensemble learning approach for brain cancer detection exploiting radiomic features. Comput. Methods Programs Biomed.; 2020; 185, 105134. [DOI: https://dx.doi.org/10.1016/j.cmpb.2019.105134] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31675644]
148. Leemput, S.C.V.D.; Meijs, M.; Patel, A.; Meijer, F.J.A.; Ginneken, B.V.; Manniesing, R. Multiclass Brain Tissue Segmentation in 4D CT Using Convolutional Neural Networks. IEEE Access; 2019; 7, pp. 51557-51569. [DOI: https://dx.doi.org/10.1109/ACCESS.2019.2910348]
149. Kleesiek, J.; Urban, G.; Hubert, A.; Schwarz, D.; Maier-Hein, K.; Bendszus, M.; Biller, A. Deep MRI brain extraction: A 3D convolutional neural network for skull stripping. NeuroImage; 2016; 129, pp. 460-469. [DOI: https://dx.doi.org/10.1016/j.neuroimage.2016.01.024] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26808333]
150. Pei, L.; Vidyaratne, L.; Rahman, M.M.; Iftekharuddin, K.M. Context aware deep learning for brain tumor segmentation, subtype classification, and survival prediction using radiology images. Sci. Rep.; 2020; 10, 19726. [DOI: https://dx.doi.org/10.1038/s41598-020-74419-9]
151. Roth, H.R.; Lu, L.; Farag, A.; Shin, H.-C.; Liu, J.; Turkbey, E.B.; Summers, R.M. Deeporgan: Multi-level deep convolutional networks for automated pancreas segmentation. Medical Image Computing and Computer-Assisted Intervention—MICCAI 2015: 18th International Conference, Munich, Germany, 5–9 October 2015, Proceedings, Part I 18; Springer: Berlin/Heidelberg, Germany, 2015; pp. 556-564.
152. Mehta, S.D.; Sebro, R. Computer-aided detection of incidental lumbar spine fractures from routine dual-energy X-ray absorptiometry (DEXA) studies using a support vector machine (SVM) classifier. J. Digit. Imaging; 2020; 33, pp. 204-210. [DOI: https://dx.doi.org/10.1007/s10278-019-00224-0]
153. Kuo, R.Y.L.; Harrison, C.; Curran, T.A.; Jones, B.; Freethy, A.; Cussons, D.; Stewart, M.; Collins, G.S.; Furniss, D. Artificial Intelligence in Fracture Detection: A Systematic Review and Meta-Analysis. Radiology; 2022; 304, pp. 50-62. [DOI: https://dx.doi.org/10.1148/radiol.211785]
154. Ju, R.-Y.; Cai, W. Fracture detection in pediatric wrist trauma X-ray images using YOLOv8 algorithm. Sci. Rep.; 2023; 13, 20077. [DOI: https://dx.doi.org/10.1038/s41598-023-47460-7] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37973984]
155. Zech, J.R.; Carotenuto, G.; Igbinoba, Z.; Tran, C.V.; Insley, E.; Baccarella, A.; Wong, T.T. Detecting pediatric wrist fractures using deep-learning-based object detection. Pediatr. Radiol.; 2023; 53, pp. 1125-1134. [DOI: https://dx.doi.org/10.1007/s00247-023-05588-8] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36650360]
156. Davoudi, A.; Malhotra, K.R.; Shickel, B.; Siegel, S.; Williams, S.; Ruppert, M.; Bihorac, E.; Ozrazgat-Baslanti, T.; Tighe, P.J.; Bihorac, A. et al. Intelligent ICU for Autonomous Patient Monitoring Using Pervasive Sensing and Deep Learning. Sci. Rep.; 2019; 9, 8020. [DOI: https://dx.doi.org/10.1038/s41598-019-44004-w] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31142754]
157. Chase, J.G.; Agogue, F.; Starfinger, C.; Lam, Z.; Shaw, G.M.; Rudge, A.D.; Sirisena, H. Quantifying agitation in sedated ICU patients using digital imaging. Comput. Methods Programs Biomed.; 2004; 76, pp. 131-141. [DOI: https://dx.doi.org/10.1016/j.cmpb.2004.03.005] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/15451162]
158. Becouze, P.; Hann, C.E.; Chase, J.G.; Shaw, G.M. Measuring facial grimacing for quantifying patient agitation in critical care. Comput. Methods Programs Biomed.; 2007; 87, pp. 138-147. [DOI: https://dx.doi.org/10.1016/j.cmpb.2007.05.005] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/17574701]
159. Brahnam, S.; Nanni, L.; Sexton, R. Introduction to neonatal facial pain detection using common and advanced face classification techniques. Advanced Computational Intelligence Paradigms in Healthcare–1; Springer: Berlin/Heidelberg, Germany, 2007; pp. 225-253.
160. Hammal, Z.; Cohn, J.F. Automatic detection of pain intensity. Proceedings of the 14th ACM International Conference on Multimodal Interaction; Santa Monica, CA, USA, 22–26 October 2012; Volume 2012, pp. 47-52. [DOI: https://dx.doi.org/10.1145/2388676.2388688]
161. Kharghanian, R.; Peiravi, A.; Moradi, F. Pain detection from facial images using unsupervised feature learning approach. Proceedings of the 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC); Orlando, FL, USA, 16–20 August 2016; Volume 2016, pp. 419-422. [DOI: https://dx.doi.org/10.1109/embc.2016.7590729]
162. Thevenot, J.; López, M.B.; Hadid, A. A Survey on Computer Vision for Assistive Medical Diagnosis From Faces. IEEE J. Biomed. Health Inform.; 2018; 22, pp. 1497-1511. [DOI: https://dx.doi.org/10.1109/JBHI.2017.2754861] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28991753]
163. Zamzmi, G.; Kasturi, R.; Goldgof, D.; Zhi, R.; Ashmeade, T.; Sun, Y. A Review of Automated Pain Assessment in Infants: Features, Classification Tasks, and Databases. IEEE Rev. Biomed. Eng.; 2018; 11, pp. 77-96. [DOI: https://dx.doi.org/10.1109/RBME.2017.2777907]
164. Gavrilescu, M.; Vizireanu, N. Predicting depression, anxiety, and stress levels from videos using the facial action coding system. Sensors; 2019; 19, 3693. [DOI: https://dx.doi.org/10.3390/s19173693]
165. Erekat, D.; Hammal, Z.; Siddiqui, M.; Dibeklioğlu, H. Enforcing Multilabel Consistency for Automatic Spatio-Temporal Assessment of Shoulder Pain Intensity. Proceedings of the ACM International Conference on Multimodal Interaction; Virtual, 25–29 October 2020; Volume 2020, pp. 156-164. [DOI: https://dx.doi.org/10.1145/3395035.3425190]
166. Xu, X.; de Sa, V.R. Exploring multidimensional measurements for pain evaluation using facial action units. Proceedings of the 2020 15th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2020); Buenos Aires, Argentina, 16–20 November 2020; IEEE: Piscataway, NJ, USA, 2020; pp. 786-792.
167. D’Antoni, F.; Russo, F.; Ambrosio, L.; Vollero, L.; Vadalà, G.; Merone, M.; Papalia, R.; Denaro, V. Artificial Intelligence and Computer Vision in Low Back Pain: A Systematic Review. Int. J. Environ. Res. Public Health; 2021; 18, 10909. [DOI: https://dx.doi.org/10.3390/ijerph182010909]
168. Prinsen, V.; Jouvet, P.; Al Omar, S.; Masson, G.; Bridier, A.; Noumeir, R. Automatic eye localization for hospitalized infants and children using convolutional neural networks. Int. J. Med. Inform.; 2021; 146, 104344. [DOI: https://dx.doi.org/10.1016/j.ijmedinf.2020.104344] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33360844]
169. Versluijs, Y.; Moore, M.G.; Ring, D.; Jayakumar, P. Clinician Facial Expression of Emotion Corresponds with Patient Mindset. Clin. Orthop. Relat. Res.; 2021; 479, pp. 1914-1923. [DOI: https://dx.doi.org/10.1097/CORR.0000000000001727] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33835095]
170. Hassan, T.; Seus, D.; Wollenberg, J.; Weitz, K.; Kunz, M.; Lautenbacher, S.; Garbas, J.U.; Schmid, U. Automatic Detection of Pain from Facial Expressions: A Survey. IEEE Trans. Pattern Anal. Mach. Intell.; 2021; 43, pp. 1815-1831. [DOI: https://dx.doi.org/10.1109/TPAMI.2019.2958341] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31825861]
171. Oster, H. Automated facial analysis of infant pain expressions: Progress and future directions. Lancet Digit. Health; 2021; 3, pp. e613-e614. [DOI: https://dx.doi.org/10.1016/S2589-7500(21)00207-7] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34481767]
172. Hoti, K.; Chivers, P.T.; Hughes, J.D. Assessing procedural pain in infants: A feasibility study evaluating a point-of-care mobile solution based on automated facial analysis. Lancet Digit. Health; 2021; 3, pp. e623-e634. [DOI: https://dx.doi.org/10.1016/S2589-7500(21)00129-1] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34481769]
173. Bringuier, S.; Macioce, V.; Boulhais, M.; Dadure, C.; Capdevila, X. Facial expressions of pain in daily clinical practice to assess postoperative pain in children: Reliability and validity of the facial action summary score. Eur. J. Pain; 2021; 25, pp. 1081-1090. [DOI: https://dx.doi.org/10.1002/ejp.1729]
174. Liu, D.; Liu, B.; Lin, T.; Liu, G.; Yang, G.; Qi, D.; Qiu, Y.; Lu, Y.; Yuan, Q.; Shuai, S.C. et al. Measuring depression severity based on facial expression and body movement using deep convolutional neural network. Front. Psychiatry; 2022; 13, 1017064. [DOI: https://dx.doi.org/10.3389/fpsyt.2022.1017064] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36620657]
175. Giannakakis, G.; Koujan, M.R.; Roussos, A.; Marias, K. Automatic stress analysis from facial videos based on deep facial action units recognition. Pattern Anal. Appl.; 2022; 25, pp. 521-535. [DOI: https://dx.doi.org/10.1007/s10044-021-01012-9]
176. Wu, C.L.; Liu, S.F.; Yu, T.L.; Shih, S.J.; Chang, C.H.; Yang Mao, S.F.; Li, Y.S.; Chen, H.J.; Chen, C.C.; Chao, W.C. Deep Learning-Based Pain Classifier Based on the Facial Expression in Critically Ill Patients. Front. Med.; 2022; 9, 851690. [DOI: https://dx.doi.org/10.3389/fmed.2022.851690]
177. Zhang, M.; Zhu, L.; Lin, S.Y.; Herr, K.; Chi, C.L.; Demir, I.; Dunn Lopez, K.; Chi, N.C. Using artificial intelligence to improve pain assessment and pain management: A scoping review. J. Am. Med. Inform. Assoc.; 2023; 30, pp. 570-587. [DOI: https://dx.doi.org/10.1093/jamia/ocac231]
178. Ma, Y.; Shen, J.; Zhao, Z.; Liang, H.; Tan, Y.; Liu, Z.; Qian, K.; Yang, M.; Hu, B. What Can Facial Movements Reveal? Depression Recognition and Analysis Based on Optical Flow Using Bayesian Networks. IEEE Trans. Neural Syst. Rehabil. Eng.; 2023; 31, pp. 3459-3468. [DOI: https://dx.doi.org/10.1109/TNSRE.2023.3305351] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37581961]
179. Heiderich, T.M.; Carlini, L.P.; Buzuti, L.F.; Balda, R.C.X.; Barros, M.C.M.; Guinsburg, R.; Thomaz, C.E. Face-based automatic pain assessment: Challenges and perspectives in neonatal intensive care units. J. Pediatr.; 2023; 99, pp. 546-560. [DOI: https://dx.doi.org/10.1016/j.jped.2023.05.005] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37331703]
180. Ghosh, A.; Umer, S.; Khan, M.K.; Rout, R.K.; Dhara, B.C. Smart sentiment analysis system for pain detection using cutting edge techniques in a smart healthcare framework. Clust. Comput.; 2023; 26, pp. 119-135. [DOI: https://dx.doi.org/10.1007/s10586-022-03552-z] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35125934]
181. Madrigal-Garcia, M.I.; Rodrigues, M.; Shenfield, A.; Singer, M.; Moreno-Cuesta, J. What Faces Reveal: A Novel Method to Identify Patients at Risk of Deterioration Using Facial Expressions. Crit. Care Med.; 2018; 46, pp. 1057-1062. [DOI: https://dx.doi.org/10.1097/CCM.0000000000003128] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29578879]
182. Madrigal-Garcia, M.I.; Archer, D.; Singer, M.; Rodrigues, M.; Shenfield, A.; Moreno-Cuesta, J. Do Temporal Changes in Facial Expressions Help Identify Patients at Risk of Deterioration in Hospital Wards? A Post Hoc Analysis of the Visual Early Warning Score Study. Crit. Care Exp.; 2020; 2, e0115. [DOI: https://dx.doi.org/10.1097/CCE.0000000000000115] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32671346]
183. Do, Q.T.; Chaudri, J. Creating Computer Vision Models for Respiratory Status Detection. Proceedings of the 2022 44th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC); Glasgow, UK, 11–15 July 2022; Volume 2022, pp. 1350-1353. [DOI: https://dx.doi.org/10.1109/embc48229.2022.9871978]
184. Zhou, C.; Chase, J.G. Low-cost structured light imaging of regional volume changes for use in assessing mechanical ventilation. Comput. Methods Programs Biomed.; 2022; 226, 107176. [DOI: https://dx.doi.org/10.1016/j.cmpb.2022.107176]
185. Ebrahimian, S.; Homayounieh, F.; Rockenbach, M.A.; Putha, P.; Raj, T.; Dayan, I.; Bizzo, B.C.; Buch, V.; Wu, D.; Kim, K. Artificial intelligence matches subjective severity assessment of pneumonia for prediction of patient outcome and need for mechanical ventilation: A cohort study. Sci. Rep.; 2021; 11, 858. [DOI: https://dx.doi.org/10.1038/s41598-020-79470-0]
186. Reiter, A.; Ma, A.; Rawat, N.; Shrock, C.; Saria, S. Process Monitoring in the Intensive Care Unit: Assessing Patient Mobility Through Activity Analysis with a Non-Invasive Mobility Sensor. Medical Image Computing and Computer-Assisted Intervention–MICCAI 2016: 19th International Conference, Athens, Greece, 17–21 October 2016; Springer: Berlin/Heidelberg, Germany, 2016; Volume 9900, pp. 482-490. [DOI: https://dx.doi.org/10.1007/978-3-319-46720-7_56]
187. Yeung, S.; Rinaldo, F.; Jopling, J.; Liu, B.; Mehra, R.; Downing, N.L.; Guo, M.; Bianconi, G.M.; Alahi, A.; Lee, J. et al. A computer vision system for deep learning-based detection of patient mobilization activities in the ICU. Npj Digit. Med.; 2019; 2, 11. [DOI: https://dx.doi.org/10.1038/s41746-019-0087-z] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31304360]
188. Kumar, S.; Singhal, P.; Krovi, V.N. Computer-vision-based decision support in surgical robotics. IEEE Des. Test; 2015; 32, pp. 89-97. [DOI: https://dx.doi.org/10.1109/MDAT.2015.2465135]
189. Marie Postma-Nilsenová, E.P.; Tates, K. Automatic Detection of Confusion in Elderly Users of a Web-Based Health Instruction Video. Telemed. e-Health; 2015; 21, pp. 514-519. [DOI: https://dx.doi.org/10.1089/tmj.2014.0061]
190. Kennedy-Metz, L.R.; Mascagni, P.; Torralba, A.; Dias, R.D.; Perona, P.; Shah, J.A.; Padoy, N.; Zenati, M.A. Computer vision in the operating room: Opportunities and caveats. IEEE Trans. Med. Robot. Bionics; 2020; 3, pp. 2-10. [DOI: https://dx.doi.org/10.1109/TMRB.2020.3040002] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33644703]
191. Gul, M.A.; Yousaf, M.H.; Nawaz, S.; Ur Rehman, Z.; Kim, H. Patient monitoring by abnormal human activity recognition based on CNN architecture. Electronics; 2020; 9, 1993. [DOI: https://dx.doi.org/10.3390/electronics9121993]
192. Singh, A.; Haque, A.; Alahi, A.; Yeung, S.; Guo, M.; Glassman, J.R.; Beninati, W.; Platchek, T.; Fei-Fei, L.; Milstein, A. Automatic detection of hand hygiene using computer vision technology. J. Am. Med. Inform. Assoc.; 2020; 27, pp. 1316-1320. [DOI: https://dx.doi.org/10.1093/jamia/ocaa115] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32712656]
193. Ahmed, I.; Jeon, G.; Piccialli, F. A deep-learning-based smart healthcare system for patient’s discomfort detection at the edge of internet of things. IEEE Internet Things J.; 2021; 8, pp. 10318-10326. [DOI: https://dx.doi.org/10.1109/JIOT.2021.3052067]
194. Dias, R.D.; Kennedy-Metz, L.R.; Yule, S.J.; Gombolay, M.; Zenati, M.A. Assessing Team Situational Awareness in the Operating Room via Computer Vision. Proceedings of the 2022 IEEE Conference on Cognitive and Computational Aspects of Situation Management (CogSIMA); Salerno, Italy, 6–10 June 2022; Volume 2022, pp. 94-96. [DOI: https://dx.doi.org/10.1109/cogsima54611.2022.9830664]
195. Chan, P.Y.; Tay, A.; Chen, D.; De Freitas, M.; Millet, C.; Nguyen-Duc, T.; Duke, G.; Lyall, J.; Nguyen, J.T.; McNeil, J. et al. Ambient intelligence-based monitoring of staff and patient activity in the intensive care unit. Aust. Crit. Care; 2023; 36, pp. 92-98. [DOI: https://dx.doi.org/10.1016/j.aucc.2022.08.011] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36244918]
196. Bai, L.; Wang, G.; Wang, J.; Yang, X.; Gao, H.; Liang, X.; Wang, A.; Islam, M.; Ren, H. OSSAR: Towards Open-Set Surgical Activity Recognition in Robot-assisted Surgery. arXiv; 2024; [DOI: https://dx.doi.org/10.48550/arXiv.2402.06985]
197. Tseng, L.-A.; Lin, H.-C.; Bai, M.-Y.; Li, M.-F.; Lee, Y.-L.; Chiang, K.-J.; Wang, Y.-C.; Guo, J.-M. DeepVinci: A Semantic Segmentation Model with Edge Super-vision and Densely Multi-scale Pyramid Module for DaVinci Gynecological Surgery. 2024; [DOI: https://dx.doi.org/10.21203/rs.3.rs-3828259/v1]
198. Chadebecq, F.; Vasconcelos, F.; Mazomenos, E.; Stoyanov, D. Computer Vision in the Surgical Operating Room. Visc. Med.; 2020; 36, pp. 456-462. [DOI: https://dx.doi.org/10.1159/000511934] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33447601]
199. Chen, Z.; Zhang, Y.; Yan, Z.; Dong, J.; Cai, W.; Ma, Y.; Jiang, J.; Dai, K.; Liang, H.; He, J. Artificial intelligence assisted display in thoracic surgery: Development and possibilities. J. Thorac. Dis.; 2021; 13, pp. 6994-7005. [DOI: https://dx.doi.org/10.21037/jtd-21-1240] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35070382]
200. Jaiswal, S.; Valstar, M.F.; Gillott, A.; Daley, D. Automatic Detection of ADHD and ASD from Expressive Behaviour in RGBD Data. Proceedings of the 2017 12th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2017); Washington, DC, USA, 30 May–3 June 2017; pp. 762-769.
201. Guarin, D.L.; Dusseldorp, J.; Hadlock, T.A.; Jowett, N. A Machine Learning Approach for Automated Facial Measurements in Facial Palsy. JAMA Facial Plast. Surg.; 2018; 20, pp. 335-337. [DOI: https://dx.doi.org/10.1001/jamafacial.2018.0030]
202. Hashemi, J.; Dawson, G.; Carpenter, K.L.H.; Campbell, K.; Qiu, Q.; Espinosa, S.; Marsan, S.; Baker, J.P.; Egger, H.L.; Sapiro, G. Computer Vision Analysis for Quantification of Autism Risk Behaviors. IEEE Trans. Affect. Comput.; 2021; 12, pp. 215-226. [DOI: https://dx.doi.org/10.1109/TAFFC.2018.2868196]
203. Zhuang, Y.; McDonald, M.; Uribe, O.; Yin, X.; Parikh, D.; Southerland, A.M.; Rohde, G.K. Facial Weakness Analysis and Quantification of Static Images. IEEE J. Biomed. Health Inform.; 2020; 24, pp. 2260-2267. [DOI: https://dx.doi.org/10.1109/JBHI.2020.2964520]
204. Ruiter, A.M.; Wang, Z.; Yin, Z.; Naber, W.C.; Simons, J.; Blom, J.T.; van Gemert, J.C.; Verschuuren, J.; Tannemaat, M.R. Assessing facial weakness in myasthenia gravis with facial recognition software and deep learning. Ann. Clin. Transl. Neurol.; 2023; 10, pp. 1314-1325. [DOI: https://dx.doi.org/10.1002/acn3.51823]
205. Parra-Dominguez, G.S.; Garcia-Capulin, C.H.; Sanchez-Yanez, R.E. Automatic Facial Palsy Diagnosis as a Classification Problem Using Regional Information Extracted from a Photograph. Diagnostics; 2022; 12, 1528. [DOI: https://dx.doi.org/10.3390/diagnostics12071528]
206. Ardalan, A.; Yamane, N.; Rao, A.K.; Montes, J.; Goldman, S. Analysis of gait synchrony and balance in neurodevelopmental disorders using computer vision techniques. Health Inform. J.; 2021; 27, 14604582211055650. [DOI: https://dx.doi.org/10.1177/14604582211055650]
207. Chambers, C.; Seethapathi, N.; Saluja, R.; Loeb, H.; Pierce, S.R.; Bogen, D.K.; Prosser, L.; Johnson, M.J.; Kording, K.P. Computer vision to automatically assess infant neuromotor risk. IEEE Trans. Neural Syst. Rehabil. Eng.; 2020; 28, pp. 2431-2442. [DOI: https://dx.doi.org/10.1109/TNSRE.2020.3029121]
208. He, L.; Li, H.; Wang, J.; Chen, M.; Gozdas, E.; Dillman, J.R.; Parikh, N.A. A multi-task, multi-stage deep transfer learning model for early prediction of neurodevelopment in very preterm infants. Sci. Rep.; 2020; 10, 15072. [DOI: https://dx.doi.org/10.1038/s41598-020-71914-x]
209. Cowan, T.; Masucci, M.D.; Gupta, T.; Haase, C.M.; Strauss, G.P.; Cohen, A.S. Computerized analysis of facial expressions in serious mental illness. Schizophr Res; 2022; 241, pp. 44-51. [DOI: https://dx.doi.org/10.1016/j.schres.2021.12.026]
210. Alhazmi, A.K.; Alanazi, M.A.; Alshehry, A.H.; Alshahry, S.M.; Jaszek, J.; Djukic, C.; Brown, A.; Jackson, K.; Chodavarapu, V.P. Intelligent Millimeter-Wave System for Human Activity Monitoring for Telemedicine. Sensors; 2024; 24, 268. [DOI: https://dx.doi.org/10.3390/s24010268]
211. Li, R.; St George, R.J.; Wang, X.; Lawler, K.; Hill, E.; Garg, S.; Williams, S.; Relton, S.; Hogg, D.; Bai, Q. Moving towards intelligent telemedicine: Computer vision measurement of human movement. Comput. Biol. Med.; 2022; 147, 105776. [DOI: https://dx.doi.org/10.1016/j.compbiomed.2022.105776] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35780600]
212. Zhang, X.; Ding, J.; Wu, M.; Wong, S.T.; Van Nguyen, H.; Pan, M. Adaptive privacy preserving deep learning algorithms for medical data. Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision; Virtual, 5–9 January 2021; pp. 1169-1178.
213. Deepika, J.; Rajan, C.; Senthil, T. Security and privacy of cloud-and IoT-based medical image diagnosis using fuzzy convolutional neural network. Comput. Intell. Neurosci.; 2021; 2021, 6615411. [DOI: https://dx.doi.org/10.1155/2021/6615411] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33790958]
214. Ren, L.; Zhang, D. A New Data Model for the Privacy Protection of Medical Images. Comput. Intell. Neurosci.; 2022; 2022, 5867215. [DOI: https://dx.doi.org/10.1155/2022/5867215] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35855794]
215. Joshi, N.B.; Nalbalwar, S.L. A fall detection and alert system for an elderly using computer vision and Internet of Things. Proceedings of the 2017 2nd IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT); Bengaluru, India, 19–20 May 2017; pp. 1276-1281.
216. Orejel Bustos, A.S.; Tramontano, M.; Morone, G.; Ciancarelli, I.; Panza, G.; Minnetti, A.; Picelli, A.; Smania, N.; Iosa, M.; Vannozzi, G. Ambient assisted living systems for falls monitoring at home. Expert Rev. Med. Devices; 2023; 20, pp. 821-828. [DOI: https://dx.doi.org/10.1080/17434440.2023.2245320]
217. Al Nahian, M.J.; Ghosh, T.; Uddin, M.N.; Islam, M.M.; Mahmud, M.; Kaiser, M.S. Towards artificial intelligence driven emotion aware fall monitoring framework suitable for elderly people with neurological disorder. Proceedings of the International Conference on Brain Informatics; Padua, Italy, 19 September 2020; Springer: Berlin/Heidelberg, Germany, 2020; pp. 275-286.
218. Chen, W.; Jiang, Z.; Guo, H.; Ni, X. Fall detection based on key points of human-skeleton using openpose. Symmetry; 2020; 12, 744. [DOI: https://dx.doi.org/10.3390/sym12050744]
219. Ramirez, H.; Velastin, S.A.; Meza, I.; Fabregas, E.; Makris, D.; Farias, G. Fall detection and activity recognition using human skeleton features. IEEE Access; 2021; 9, pp. 33532-33542. [DOI: https://dx.doi.org/10.1109/ACCESS.2021.3061626]
220. Awwad, S.; Tarvade, S.; Piccardi, M.; Gattas, D.J. The use of privacy-protected computer vision to measure the quality of healthcare worker hand hygiene. Int. J. Qual. Health Care; 2019; 31, pp. 36-42. [DOI: https://dx.doi.org/10.1093/intqhc/mzy099]
221. Guo, P.; Chiew, Y.S.; Shaw, G.M.; Shao, L.; Green, R.; Clark, A.; Chase, J.G. Clinical Activity Monitoring System (CATS): An automatic system to quantify bedside clinical activities in the intensive care unit. Intensive Crit. Care Nurs.; 2016; 37, pp. 52-61. [DOI: https://dx.doi.org/10.1016/j.iccn.2016.05.003]
222. Peng, G.; Yeong Shiong, C.; Shaw, G.; Chase, G. Validation of clinical activity tracking system in Intensive Care Unit to assess nurse workload distribution. Proceedings of the 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC); Milan, Italy, 25–29 August 2015; Volume 2015, pp. 458-461. [DOI: https://dx.doi.org/10.1109/embc.2015.7318398]
223. Bashiri, F.S.; LaRose, E.; Peissig, P.; Tafti, A.P. MCIndoor20000: A fully-labeled image dataset to advance indoor objects detection. Data Brief; 2018; 17, pp. 71-75. [DOI: https://dx.doi.org/10.1016/j.dib.2017.12.047]
224. Ismail, A.; Ahmad, S.A.; Soh, A.C.; Hassan, M.K.; Harith, H.H. MYNursingHome: A fully-labelled image dataset for indoor object classification. Data Brief; 2020; 32, 106268. [DOI: https://dx.doi.org/10.1016/j.dib.2020.106268]
225. Hu, D.; Li, S.; Wang, M. Object detection in hospital facilities: A comprehensive dataset and performance evaluation. Eng. Appl. Artif. Intell.; 2023; 123, 106223. [DOI: https://dx.doi.org/10.1016/j.engappai.2023.106223]
226. Lea, C.; Facker, J.; Hager, G.; Taylor, R.; Saria, S. 3D Sensing Algorithms Towards Building an Intelligent Intensive Care Unit. AMIA Jt. Summits Transl. Sci. Proc.; 2013; 2013, pp. 136-140. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24303253]
227. Szegedy, C.; Liu, W.; Jia, Y.; Sermanet, P.; Reed, S.; Anguelov, D.; Erhan, D.; Vanhoucke, V.; Rabinovich, A. Going deeper with convolutions. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition; Boston, MA, USA, 8–10 June 2015; pp. 1-9.
228. Yamashita, R.; Nishio, M.; Do, R.K.G.; Togashi, K. Convolutional neural networks: An overview and application in radiology. Insights Into Imaging; 2018; 9, pp. 611-629. [DOI: https://dx.doi.org/10.1007/s13244-018-0639-9]
229. Rezazade Mehrizi, M.H.; van Ooijen, P.; Homan, M. Applications of artificial intelligence (AI) in diagnostic radiology: A technography study. Eur. Radiol.; 2021; 31, pp. 1805-1811. [DOI: https://dx.doi.org/10.1007/s00330-020-07230-9]
230. Katzman, B.D.; van der Pol, C.B.; Soyer, P.; Patlas, M.N. Artificial intelligence in emergency radiology: A review of applications and possibilities. Diagn. Interv. Imaging; 2023; 104, pp. 6-10. [DOI: https://dx.doi.org/10.1016/j.diii.2022.07.005] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35933269]
231. Kim, B.; Romeijn, S.; van Buchem, M.; Mehrizi, M.H.R.; Grootjans, W. A holistic approach to implementing artificial intelligence in radiology. Insights Into Imaging; 2024; 15, 22. [DOI: https://dx.doi.org/10.1186/s13244-023-01586-4]
232. Komura, D.; Ishikawa, S. Machine learning methods for histopathological image analysis. Comput. Struct. Biotechnol. J.; 2018; 16, pp. 34-42. [DOI: https://dx.doi.org/10.1016/j.csbj.2018.01.001]
233. Stollings, J.L.; Kotfis, K.; Chanques, G.; Pun, B.T.; Pandharipande, P.P.; Ely, E.W. Delirium in critical illness: Clinical manifestations, outcomes, and management. Intensive Care Med.; 2021; 47, pp. 1089-1103. [DOI: https://dx.doi.org/10.1007/s00134-021-06503-1] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34401939]
234. Berger, S.E.; Baria, A.T. Assessing Pain Research: A Narrative Review of Emerging Pain Methods, Their Technosocial Implications, and Opportunities for Multidisciplinary Approaches. Front. Pain Res.; 2022; 3, 896276. [DOI: https://dx.doi.org/10.3389/fpain.2022.896276]
235. Nordness, M.F.; Hayhurst, C.J.; Pandharipande, P. Current Perspectives on the Assessment and Management of Pain in the Intensive Care Unit. J. Pain Res.; 2021; 14, pp. 1733-1744. [DOI: https://dx.doi.org/10.2147/JPR.S256406] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34163231]
236. Pisani, M.A.; Devlin, J.W.; Skrobik, Y. Pain and Delirium in Critical Illness: An Exploration of Key 2018 SCCM PADIS Guideline Evidence Gaps. Semin. Respir. Crit. Care Med.; 2019; 40, pp. 604-613. [DOI: https://dx.doi.org/10.1055/s-0039-1698809]
237. Devlin, J.W.; Skrobik, Y.; Gélinas, C.; Needham, D.M.; Slooter, A.J.C.; Pandharipande, P.P.; Watson, P.L.; Weinhouse, G.L.; Nunnally, M.E.; Rochwerg, B. et al. Clinical practice guidelines for the prevention and management of pain, agitation/sedation, delirium, immobility, and sleep disruption in adult patients in the ICU. Crit. Care Med.; 2018; 46, pp. e825-e873. [DOI: https://dx.doi.org/10.1097/CCM.0000000000003299]
238. Herasevich, S.; Lipatov, K.; Pinevich, Y.; Lindroth, H.; Tekin, A.; Herasevich, V.; Pickering, B.W.; Barwise, A.K. The Impact of Health Information Technology for Early Detection of Patient Deterioration on Mortality and Length of Stay in the Hospital Acute Care Setting: Systematic Review and Meta-Analysis. Crit. Care Med.; 2022; 50, pp. 1198-1209. [DOI: https://dx.doi.org/10.1097/CCM.0000000000005554] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35412476]
239. Davidovitch, M.; Slobodin, O.; Weisskopf, M.G.; Rotem, R.S. Age-Specific Time Trends in Incidence Rates of Autism Spectrum Disorder Following Adaptation of DSM-5 and Other ASD-Related Regulatory Changes in Israel. Autism Res.; 2020; 13, pp. 1893-1901. [DOI: https://dx.doi.org/10.1002/aur.2420] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33098226]
240. Christensen, D.L.; Maenner, M.J.; Bilder, D.; Constantino, J.N.; Daniels, J.; Durkin, M.S.; Fitzgerald, R.T.; Kurzius-Spencer, M.; Pettygrove, S.D.; Robinson, C. et al. Prevalence and Characteristics of Autism Spectrum Disorder Among Children Aged 4 Years—Early Autism and Developmental Disabilities Monitoring Network, Seven Sites, United States, 2010, 2012, and 2014. MMWR Surveill. Summ.; 2019; 68, pp. 1-19. [DOI: https://dx.doi.org/10.15585/mmwr.ss6802a1]
241. Cowan, S.L.; Preller, J.; Goudie, R.J.B. Evaluation of the E-PRE-DELIRIC prediction model for ICU delirium: A retrospective validation in a UK general ICU. Crit. Care; 2020; 24, 123. [DOI: https://dx.doi.org/10.1186/s13054-020-2838-2] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32228666]
242. Guirguis-Blake, J.M.; Michael, Y.L.; Perdue, L.A.; Coppola, E.L.; Beil, T.L.; Thompson, J.H. U.S. Preventive Services Task Force Evidence Syntheses, formerly Systematic Evidence Reviews. Interventions to Prevent Falls in Community-Dwelling Older Adults: A Systematic Review for the U.S. Preventive Services Task Force; Agency for Healthcare Research and Quality (US): Rockville, MD, USA, 2018.
243. Petersen, C. Through Patients’ Eyes: Regulation, Technology, Privacy, and the Future. Yearb. Med. Inform.; 2018; 27, pp. 10-15. [DOI: https://dx.doi.org/10.1055/s-0038-1641193]
244. Vo, V.; Chen, G.; Aquino, Y.S.J.; Carter, S.M.; Do, Q.N.; Woode, M.E. Multi-stakeholder preferences for the use of artificial intelligence in healthcare: A systematic review and thematic analysis. Soc. Sci. Med.; 2023; 338, 116357. [DOI: https://dx.doi.org/10.1016/j.socscimed.2023.116357] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37949020]
245. Glancova, A.; Do, Q.T.; Sanghavi, D.K.; Franco, P.M.; Gopal, N.; Lehman, L.M.; Dong, Y.; Pickering, B.W.; Herasevich, V. Are We Ready for Video Recognition and Computer Vision in the Intensive Care Unit? A Survey. Appl. Clin. Inform.; 2021; 12, pp. 120-132. [DOI: https://dx.doi.org/10.1055/s-0040-1722614]
246. Gerke, S.; Yeung, S.; Cohen, I.G. Ethical and Legal Aspects of Ambient Intelligence in Hospitals. JAMA; 2020; 323, pp. 601-602. [DOI: https://dx.doi.org/10.1001/jama.2019.21699] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31977033]
247. Bender, E.M.; Gebru, T.; McMillan-Major, A.; Shmitchell, S. On the Dangers of Stochastic Parrots: Can Language Models Be Too Big?. Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency; Virtual Event, 3–10 March 2021; pp. 610-623. [DOI: https://dx.doi.org/10.1145/3442188.3445922]
248. Fournier-Tombs, E.; McHardy, J. A Medical Ethics Framework for Conversational Artificial Intelligence. J. Med. Internet Res.; 2023; 25, e43068. [DOI: https://dx.doi.org/10.2196/43068]
249. Kim, B.; Kim, H.; Kim, K.; Kim, S.; Kim, J. Learning not to learn: Training deep neural networks with biased data. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition; Long Beach, CA, USA, 15–20 June 2019; pp. 9012-9020.
250. Karkkainen, K.; Joo, J. Fairface: Face attribute dataset for balanced race, gender, and age for bias measurement and mitigation. Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision; Virtual, 5–9 January 2021; pp. 1548-1558.
251. Bahng, H.; Chun, S.; Yun, S.; Choo, J.; Oh, S.J. Learning de-biased representations with biased representations. Proceedings of the International Conference on Machine Learning; Virtual, 13–18 July 2020; PMLR: London, UK, 2020; pp. 528-539.
252. Nam, J.; Cha, H.; Ahn, S.; Lee, J.; Shin, J. Learning from failure: De-biasing classifier from biased classifier. Adv. Neural Inf. Process. Syst.; 2020; 33, pp. 20673-20684.
253. Morley, J.; Floridi, L. An ethically mindful approach to AI for health care. Lancet; 2020; 395, pp. 254-255. [DOI: https://dx.doi.org/10.1016/S0140-6736(19)32975-7] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31982053]
254. Jobin, A.; Ienca, M.; Vayena, E. The global landscape of AI ethics guidelines. Nat. Mach. Intell.; 2019; 1, pp. 389-399. [DOI: https://dx.doi.org/10.1038/s42256-019-0088-2]
255. Elendu, C.; Amaechi, D.C.; Elendu, T.C.; Jingwa, K.A.; Okoye, O.K.; John Okah, M.; Ladele, J.A.; Farah, A.H.; Alimi, H.A. Ethical implications of AI and robotics in healthcare: A review. Medicine; 2023; 102, e36671. [DOI: https://dx.doi.org/10.1097/MD.0000000000036671] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38115340]
256. Acemoglu, D.; Restrepo, P. Artificial intelligence, automation, and work. The Economics of Artificial Intelligence: An Agenda; University of Chicago Press: Chicago, IL, USA, 2018; pp. 197-236.
257. Braun, M.; Hummel, P.; Beck, S.; Dabrock, P. Primer on an ethics of AI-based decision support systems in the clinic. J. Med. Ethics; 2020; 47, e3. [DOI: https://dx.doi.org/10.1136/medethics-2019-105860]
258. Khanna, N.N.; Maindarkar, M.A.; Viswanathan, V.; Fernandes, J.F.E.; Paul, S.; Bhagawati, M.; Ahluwalia, P.; Ruzsa, Z.; Sharma, A.; Kolluri, R. et al. Economics of Artificial Intelligence in Healthcare: Diagnosis vs. Treatment. Healthcare; 2022; 10, 2493. [DOI: https://dx.doi.org/10.3390/healthcare10122493] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36554017]
259. Bekbolatova, M.; Mayer, J.; Ong, C.W.; Toma, M. Transformative Potential of AI in Healthcare: Definitions, Applications, and Navigating the Ethical Landscape and Public Perspectives. Healthcare; 2024; 12, 125. [DOI: https://dx.doi.org/10.3390/healthcare12020125] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38255014]
260. Deng, J.; Dong, W.; Socher, R.; Li, L.-J.; Li, K.; Fei-Fei, L. Imagenet: A large-scale hierarchical image database. Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition; Miami, FL, USA, 20–25 June 2009; IEEE: Piscataway, NJ, USA, 2009; pp. 248-255.
261. Ying, X. An overview of overfitting and its solutions. Proc. J. Phys. Conf. Ser.; 2019; 1168, 022022. [DOI: https://dx.doi.org/10.1088/1742-6596/1168/2/022022]
262. Montesinos López, O.A.; Montesinos López, A.; Crossa, J. Overfitting, model tuning, and evaluation of prediction performance. Multivariate Statistical Machine Learning Methods for Genomic Prediction; Springer: Berlin/Heidelberg, Germany, 2022; pp. 109-139.
263. Garcea, F.; Serra, A.; Lamberti, F.; Morra, L. Data augmentation for medical imaging: A systematic literature review. Comput. Biol. Med.; 2023; 152, 106391. [DOI: https://dx.doi.org/10.1016/j.compbiomed.2022.106391]
264. Salehi, A.W.; Khan, S.; Gupta, G.; Alabduallah, B.I.; Almjally, A.; Alsolai, H.; Siddiqui, T.; Mellit, A. A Study of CNN and Transfer Learning in Medical Imaging: Advantages, Challenges, Future Scope. Sustainability; 2023; 15, 5930. [DOI: https://dx.doi.org/10.3390/su15075930]
265. Girshick YWaAKaFMaW-YLaR. Detectron2. 2019; Available online: https://detectron2.readthedocs.io/en/latest/ (accessed on 15 July 2023).
266. Chen, K.; Wang, J.; Pang, J.; Cao, Y.; Xiong, Y.; Li, X.; Sun, S.; Feng, W.; Liu, Z.; Xu, J. MMDetection: Open mmlab detection toolbox and benchmark. arXiv; 2019; [DOI: https://dx.doi.org/10.48550/arXiv.1906.07155]
267. Cao, Z.; Hidalgo, G.; Simon, T.; Wei, S.-E.; Sheikh, Y. Openpose: Realtime multi-person 2d pose estimation using part affinity fields. arXiv; 2018; [DOI: https://dx.doi.org/10.48550/arXiv.1812.08008]
268. Chang, K.; Balachandar, N.; Lam, C.; Yi, D.; Brown, J.; Beers, A.; Rosen, B.; Rubin, D.L.; Kalpathy-Cramer, J. Distributed deep learning networks among institutions for medical imaging. J. Am. Med. Inform. Assoc.; 2018; 25, pp. 945-954. [DOI: https://dx.doi.org/10.1093/jamia/ocy017] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29617797]
269. Everingham, M.; Van Gool, L.; Williams, C.K.; Winn, J.; Zisserman, A. The pascal visual object classes (voc) challenge. Int. J. Comput. Vis.; 2010; 88, pp. 303-338. [DOI: https://dx.doi.org/10.1007/s11263-009-0275-4]
270. Lin, T.-Y.; Maire, M.; Belongie, S.; Hays, J.; Perona, P.; Ramanan, D.; Dollár, P.; Zitnick, C.L. Microsoft coco: Common objects in context. Proceedings of the Computer Vision–ECCV 2014: 13th European Conference; Zurich, Switzerland, 6–12 September 2014; Proceedings, Part V 13 Springer: Berlin/Heidelberg, Germany, 2014; pp. 740-755.
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Details
; Nalaie, Keivan 2
; Raghu, Roshini 3
; Ayala, Ivan N 3
; Busch, Charles 4 ; Bhattacharyya, Anirban 5 ; Pablo Moreno Franco 6
; Diedrich, Daniel A 7 ; Pickering, Brian W 7 ; Herasevich, Vitaly 7
1 Division of Nursing Research, Department of Nursing, Mayo Clinic, Rochester, MN 55905, USA;
2 Division of Nursing Research, Department of Nursing, Mayo Clinic, Rochester, MN 55905, USA;
3 Division of Nursing Research, Department of Nursing, Mayo Clinic, Rochester, MN 55905, USA;
4 Division of Nursing Research, Department of Nursing, Mayo Clinic, Rochester, MN 55905, USA;
5 Department of Critical Care Medicine, Mayo Clinic, Jacksonville, FL 32224, USA;
6 Department of Transplantation Medicine, Mayo Clinic, Jacksonville, FL 32224, USA
7 Department of Anesthesiology and Perioperative Medicine, Mayo Clinic, Rochester, MN 55905, USA;