1. Introduction
In the ever-evolving world of orthopedics and sports medicine, new technologies are fundamentally transforming both medical practice and patient experiences. Alongside traditional innovations in orthopedic implants and surgical techniques, we are witnessing the introduction of cutting-edge technologies such as 3D printers and augmented reality, which are revolutionizing treatment design and customization [1,2,3,4,5,6]. Among these various technologies, the Internet has become a significant source for patients, who often rely on this virtual platform to find answers to their medical inquiries [7,8]. However, one of the most significant changes is represented by the rise of artificial intelligence (AI) and chatbots that combine AI with natural language processing (NLP), such as ChatGPT (OpenAI, San Francisco, CA, USA), and Google Bard (Google LLC., Mountain View, CA, USA) [9,10,11,12].
These chatbots are increasingly becoming a vital source of medical information for patients, who often use them to obtain answers to their questions and access healthcare advice [9,10,11,12]. Particularly in the fields of sports medicine and pediatric orthopedics, where injury management and performance enhancement require a thorough understanding of musculoskeletal anatomy and injuries, access to timely and accurate information is crucial for optimizing outcomes. Among the various chatbots available, ChatGPT-3.5 and Google Bard stand out for their unique features. While ChatGPT-3.5 relies on training data up to September 2021, offering a vast knowledge base but potentially outdated information, Google Bard has the ability to access real-time data on the Internet, ensuring fresher information [10,11,12]. This difference is particularly relevant in the context of sports medicine, where new discoveries and advances in injury management can directly influence clinical decisions.
However, despite the potential of these platforms to offer knowledge comparable to that of a medical graduate, concerns arise regarding the dissemination of inaccurate information [12]. This risk is amplified in the context of pediatric orthopedics, where communication with patients’ parents is crucial, and the spread of inaccurate information could influence clinical decisions and compromise trust in the doctor–patient relationship. Therefore, while the use of chatbots as a source of medical information continues to grow, it is essential to ensure supervision and review by qualified healthcare professionals. With these premises, the current study aims to compare ChatGPT and Google Bard as potential sources of information for patients or patients’ parents, evaluating the quality of responses to frequently asked questions in the fields of sports medicine and pediatric orthopedics.
2. Materials and Methods
To evaluate sports medicine, we conducted a Google search using the keyword “anterior cruciate ligament” (ACL), with the search history previously cleared. The first ten questions under the “people also ask” section, were recorded. Duplicate questions with the same meaning were removed. The ten questions were then presented to ChatGPT-3.5 and Google Bard. A new user account was created for this study, and the chat history was cleared before each question to minimize potential bias from ChatGPT and Google Bard’s memory retention feature. To assess the quality of the responses, we asked two expert orthopedic surgeons in sports medicine to rate the quality of responses in their native language provided by ChatGPT and Google Bard on a scale of 1 (poor quality) to 5 (excellent quality) based on the validated Global Quality Score (GQS) [13] (Table 1).
For the field of pediatric orthopedics, we selected an extremely important topic in pediatric orthopedics, namely, flat feet, and subsequently conducted a Google search with the search history previously cleared. We then recorded the first ten questions in the “people also ask” section. All duplicate questions with the same meaning were removed. The ten questions were then posed to ChatGPT-3.5 and Google Bard. Even in this case, to eliminate any bias and minimize the potential effect of the memory-retention feature of ChatGPT and Google Bard, a new user account was created, and the chat history was cleared before each question. To assess the quality of the responses, we asked two expert orthopedic surgeons in pediatric orthopedics to evaluate the quality of responses provided by both chatbots in their native language on a scale from 1 (poor quality) to 5 (excellent quality) based on the Validated Global Quality Score (GQS) [13] (Table 1). The respondents were blinded to the sources of information, and each respondent was unaware of the other’s evaluation. In both cases, the respondents were kept anonymous with respect to the sources of information. Misinformation spread was defined as the dissemination of erroneous, misleading or false information.
Statistical Analysis
The analysis was performed using SPSS software version 26 (IBM SPSS Statistics, Chicago, IL, USA). To determine whether there were statistically significant differences between the average scores of responses generated by Google Bard and ChatGPT, we conducted independent-sample t-tests. We set the significance level (alpha) at 0.05 for all statistical tests. A p-value less than 0.05 was considered indicative of a statistically significant difference between the groups.
3. Results
In sports medicine, the average score for responses provided by Google Bard was 4 ± 0.78, while the average score for responses generated by ChatGPT was 4.1 ± 0.7. This demonstrated that both chatbots provided good-quality answers that covered the most important topics and were useful to patients. The difference between the two platforms was not statistically significant (p = 0.6787). The questions for ChatGPT and Google Bard are reported in Table 2 and Table 4.
For pediatric orthopedics, the average score for responses provided by Google Bard was 3.5 ± 1, while the average score for responses generated by ChatGPT was 3.8 ± 0.83. The difference between the two platforms was not statistically significant (p = 0.3092). These results demonstrated that both chatbots provided moderate- to good-quality responses that covered some information, but important topics were sometimes missing. The question that received the lowest rating was undoubtedly the one related to the timing of operating on flat feet; both experts gave it a Global Quality Score of 2 to 3. The questions for ChatGPT and Google Bard are listed in Table 3. In this case, both experts agreed that ChatGPT proved to be more comprehensive and concise, with no reported instances of incorrect information. Results are better presented in Table 3 and Table 4.
4. Discussion
This study thoroughly examined the use of chatbots in orthopedics, with a particular focus on sports medicine and pediatric orthopedics. According to the feedback provided by two expert orthopedic surgeons in these fields, both sources demonstrate promising potential for the use of AI technology as an advanced educational tool for patients.
Regarding sports medicine, ChatGPT’s responses were found to be more readable and concise, according to the feedback from the two expert orthopedic surgeons. Neither of the respondents encountered misinformation spread, i.e., the dissemination of erroneous, misleading or false information. This is an extremely positive aspect, considering that there have been reports of AI hallucinations in which incorrect responses are convincingly described [14]. However, this phenomenon cannot be entirely ruled out, given that the training of these chatbots may potentially be based on “grey literature.” Consequently, despite the accuracy of the information provided in this study, we emphasize that only a user capable of proofreading the content of the responses can recognize any incorrect information. This is arguably the aspect to keep most under control, since incorrect information assumed to be correct would undoubtedly have a negative impact on patient education. The limitations of ChatGPT in relation to its training data are evident in its inability to provide reliable sources for the information it presents. Often, the links provided by ChatGPT are inaccessible. At the same time, it is essential to emphasize that the chatbot was trained only until September 2021 and does not have access to the Internet. Regarding this specific aspect, a potential advantage could be seen in Google, since its links are constantly updated and accessible to patients for further reading. Its limits and advantages have already been described [14]. In our opinion, the sense is that of attempting to integrate these forms of technology, striving to improve them. For example, regarding potentially dangerous episodes of AI hallucination, ChatGPT-4 appears to be capable of recognizing this phenomenon when analyzed in a separate session [14,15]. One reflection we wish to emphasize is that these tools demonstrate various levels of competence in medical applications despite not being specifically trained for data analysis in this field.
As for pediatric orthopedics, this study yielded comparable average scores for Google Bard (3.5 ± 1) and ChatGPT (3.8 ± 0.83), with no statistically significant difference between the two platforms (p = 0.3092). While both chatbots offered responses of moderate to good quality, there were occasional gaps in coverage, particularly in addressing critical topics such as the timing of operating on flat feet. As reported in the results, the question that received the lowest rating was related to the timing of surgery for flat feet, to which both experts gave a Global Quality Score of 2 to 3. This low score indicates a significant gap in the chatbots’ ability to provide detailed and accurate information on this critical topic. The timing of surgical intervention for flat feet is a complex issue that requires a nuanced understanding of various factors, including the patient’s age and symptoms and the severity of the condition. The lack of adequate responses from the chatbots on these aspects could lead to confusion or incorrect decisions by patients or their parents. This highlights once again the necessity of expert oversight by an orthopedic surgeon. Moreover, according to feedback from two expert pediatric orthopedic surgeons when choosing between the two chatbots, ChatGPT proved to be more comprehensive and concise and therefore more useful to patients. Another shared consideration was the absence of incorrect information, meaning the dissemination of erroneous, misleading or false information. In conclusion, both experts concurred that the two chatbots showed promising potential as advanced education sources for patients or their parents. However, while the overall evaluation showed responses of moderate to good quality that covered some information, important topics were sometimes missing. This evaluation should reach at least a medium-high level of 4, meaning “Good quality, most important topics covered, useful to patients” [13]. This result requires ongoing development in improving these chatbots. As a result, how reliable is this information? In our view, it is essential to alert potential users to the dangers of accepting everything from the Internet as the truth, especially in the field of pediatric orthopedics. In the past, Internet usage in this specific area of orthopedics has been analyzed, highlighting that nonprofit and academic websites were the most reliable sources [16]. Patients tend to turn to search engines in their quest for medical answers and generally consider the information they find to be reliable [17,18]. Although this study may seem to yield encouraging results regarding information coming from the Internet but with the assistance of AI, it is not entirely possible to rule out the presence of inaccurate information. In our opinion, a user capable of carefully examining the content of responses should always screen the information to be provided to patients. In fact, AI has also demonstrated a potentially dangerous phenomenon, described as “AI hallucination,” which refers to an erroneous response from the chatbot that the chatbot itself does not recognize [14,15]. The danger of this phenomenon lies precisely in the particularly convincing presentation of information that could lead the user to assume it as fact, especially if the user not sufficiently knowledgeable. Another limitation to keep in mind is the request for scientific sources from these chatbots, often leading them to provide links that are found to be inaccessible. Finally, it is essential to note that ChatGPT has training data only up to September 2021. This temporal aspect could influence the generation of responses based on guidelines that may not be up to date. On the basis of the authors’ opinion, the main improvements that should be considered for AI-based NLP models to enhance their relationship with patients are reported in Table 5.
This marks the first attempt to compare the outcomes of two major AI-based chatbots in providing patient information. Both platforms were evaluated using a transparent and unbiased methodology, with careful measures taken to eliminate potential biases, such as clearing the search history and anonymizing responses from orthopedic experts. However, this paper has several limitations that must be considered to interpret the results properly. Firstly, the decision to restrict the analysis to ten questions may represent a substantial intrinsic limitation, as it only covers part of the spectrum of possible topics or issues patients may have in the field of sports medicine. In the future, a more comprehensive assessment of AI chatbots’ capabilities could be achieved by using a broader range of questions, potentially involving diverse clinical scenarios. Furthermore, as mentioned in the document, the limited training period of ChatGPT may affect its ability to provide updated and relevant information, especially considering the continuously evolving nature of medical knowledge. Another limiting aspect is the absence of an analysis based on patient feedback. Patient perceptions and needs may differ from expert evaluations, and the omission of this perspective could reduce the completeness of the paper’s conclusions regarding practical utility for individuals seeking medical information. However, we believe that a preliminary analysis by professionals is crucial in this regard.
With these considerations in mind, the paper could undoubtedly benefit from future investigations that specifically address the identified areas of concern. This would contribute to a deeper understanding of the actual validity and capacity for knowledgeable application of chatbots in delivering medical information.
5. Conclusions
In conclusion, despite notable limitations, AI-based chatbots hold promise as resources for patient education in orthopedics, notably in sports medicine and pediatric orthopedics. However, their role should be complementary to specialist guidance. Specialists must engage with these tools to facilitate their evolution effectively, acknowledging the inevitable integration of AI into orthopedic practice. Continuous research is essential to ensure the safe and effective incorporation of AI into patient care. Limitations include reliance on ChatGPT-3.5 and a narrow question focus. Future research could explore the impact of using more current AI models and broaden the scope of inquiry to encompass a wider range of clinical scenarios. Additionally, efforts to enhance the training data and ensure access to reliable sources could further improve the utility of AI-based chatbots in orthopedic patient education.
Conceptualization, R.G. and M.A.-B.; methodology, R.G. and M.A.-B.; validation, R.G., M.A.-B. and L.M.; investigation, L.M., F.V. and M.D.R.; data curation, R.G. and M.A.-B.; writing—original draft preparation, R.G. and M.A.-B.; writing—review and editing, R.G., M.A.-B. and L.M.; visualization, R.G. and M.A.-B.; supervision, L.M. and G.M.P. All authors have read and agreed to the published version of the manuscript.
Not applicable.
Not applicable.
The data presented in this study are available within the manuscript.
The authors declare no conflicts 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.
Global Quality Score (GQS).
| Global Quality Score | |
|---|---|
| 1 | Poor quality, very unlikely to be of any use to patients |
| 2 | Poor quality but some information present, of very limited use to patients |
| 3 | Some information covered but important topics missing, somewhat useful to patients |
| 4 | Good quality, most important topics covered, useful to patients |
| 5 | Excellent quality, highly useful to patients |
Questions and answers produced by ChatGPT and Google Bard in sports medicine.
| Question | ChatGPT-3.5 | Score #1 | Score #2 | Google Bard | Score #1 | Score #2 |
|---|---|---|---|---|---|---|
| How does one recover from an anterior cruciate ligament injury? | The healing process for an anterior cruciate ligament (ACL) injury depends on the severity of the damage and the chosen treatment options. In severe cases, ACL reconstruction surgery is often necessary, during which the damaged ligament is replaced with a graft. After the surgery, it is crucial to follow a physical therapy program aimed at strengthening the surrounding muscles, restoring joint stability, and range of motion. This process can take several months. For less severe injuries or situations where surgery is not recommended, rehabilitation may be the primary path to healing. Physical therapy, including muscle strengthening exercises, is essential for improving knee function. The use of braces or knee supports may be recommended to provide support during physical activity. Pain and inflammation management is important, and the doctor may prescribe appropriate medications. Lifestyle modifications, such as avoiding high-risk activities or taking precautions during sports, may be necessary during the recovery process. Specialized guidance is essential for an accurate evaluation and a targeted treatment plan, allowing for effective healing and the prevention of chronic knee problems. | 4 | 3 | Healing from an anterior cruciate ligament (ACL) injury can take anywhere from 6 to 12 months. The initial treatment involves rest, ice application, and the use of a brace to reduce pain and swelling. Subsequently, the patient will need to start a rehabilitation program to improve knee mobility and strength. If the injury is severe, the surgeon may recommend ACL reconstruction surgery. This procedure involves replacing the damaged ligament with a graft taken from another part of the body, such as the patellar tendon or the quadriceps tendon. Post-operative recovery is longer and more demanding than non-surgical recovery. The patient will need to follow an intensive rehabilitation program to restore knee functionality. With proper treatment and hard work, most people who experience an ACL injury can return to their daily and sports activities without issues. | 4 | 3 |
| Can you walk when you tear the cruciate ligament? | Yes, it is possible to walk with an anterior cruciate ligament (ACL) injury, although the ability to walk depends on the severity of the injury and the presence of any associated injuries or complications. In many cases, especially with less severe or partial ACL injuries, people can walk after an initial period of pain and instability. However, it’s important to note that a complete ACL injury can make walking unsteady, especially during activities that require changes in direction or significant loads on the knee. An ACL injury compromises the stability of the knee joint, which can result in a feeling of “giving way” or insecurity while walking. Some people may need to use braces or support devices to walk more comfortably. The best course of action if you suspect an ACL injury is to consult an orthopedic doctor or a sports medicine specialist for a thorough evaluation. An accurate diagnosis and an appropriate treatment plan are essential for optimal healing and to prevent further damage to the knee joint. | 5 | 3 | A torn anterior cruciate ligament (ACL) is an injury that can cause knee pain and instability. In some cases, it is possible to walk even with a torn ACL, but walking will likely be painful and unstable. The knee may swell, and there is a risk of further injury. Typically, it is advisable to use crutches to walk after an ACL injury. This will help protect the knee and reduce pain. Over time and with rehabilitation, it is possible to improve knee mobility and strength, making walking easier and less painful. Here are some tips for walking with a torn ACL:
| 4 | 3 |
| How to tell if the anterior cruciate ligament is torn? | The rupture of the anterior cruciate ligament (ACL) is often accompanied by clear symptoms, but an accurate diagnosis requires a medical evaluation. Here are some common signs and symptoms that may indicate a possible ACL tear:
| 4 | 4 | The symptoms of an anterior cruciate ligament (ACL) tear include:
| 5 | 4 |
| Where does the anterior cruciate ligament hurt? | The anterior cruciate ligament (ACL) is a ligament located inside the knee joint, so the “pain” directly associated with the ACL may not be felt in the same way as muscles or other parts of the body. However, when the ACL ruptures or is injured, the symptoms are often perceived in the surrounding area of the knee joint. The primary pain caused by an ACL injury may be felt inside or outside the knee, but it can also radiate down the leg or in the surrounding area. Patients may experience sudden and sharp pain during the injury, accompanied by swelling and a sense of instability in the knee joint. It is important to note that the perception of pain can vary from person to person and may depend on the severity of the injury. The diagnosis and management of a potential ACL injury should be entrusted to an experienced orthopedic or sports medicine doctor. | 5 | 3 | The anterior cruciate ligament (ACL) is located inside the knee, between the femur and the tibia. Its function is to prevent the tibia from moving forward relative to the femur. When the ACL ruptures, pain is typically felt inside the knee, in the front part. The pain can be intense and may be accompanied by a snapping sound or a sense of tearing. The pain can be accompanied by other symptoms such as swelling, stiffness, and knee instability. If you experience knee pain, it is important to consult a doctor for an accurate diagnosis. | 4 | 3 |
| What happens if you don’t have the anterior cruciate ligament surgery?” | If an anterior cruciate ligament (ACL) injury is not surgically treated, there can be some long-term consequences. However, the appropriate treatment depends on the severity of the injury and the patient’s lifestyle
| 5 | 4 | If an untreated anterior cruciate ligament (ACL) is ruptured, the knee can become unstable and painful. This can make it difficult to walk, climb stairs, and perform other daily activities. Additionally, people with an untreated ACL have a higher risk of future knee injuries, such as meniscal injuries or arthritis.
| 4 | 4 |
| How many physical therapy sessions for a torn cruciate ligament? | The number of physical therapy sessions required for the rehabilitation of an anterior cruciate ligament (ACL) depends on the severity of the injury, the individual patient’s response, and whether surgery was performed. However, a typical rehabilitation protocol following ACL reconstruction surgery involves a series of sessions that can range from 20 to 40 or more, spread out over several months. In the early stages of rehabilitation, physical therapy sessions focus on reducing swelling, restoring range of motion, and managing pain. Subsequently, the emphasis shifts to muscle strengthening exercises, balance exercises, and functional training to improve joint stability. It’s important to emphasize that ACL rehabilitation is a gradual and personalized process, and the number of sessions can vary based on the patient’s progress and the specific requirements of their recovery. The supervision of an experienced physical therapist is essential to ensure a safe and effective recovery. | 4 | 3 | The number of physical therapy sessions required for a torn anterior cruciate ligament (ACL) depends on the severity of the injury and the individual’s activity level. In general, individuals who undergo ACL surgery should follow a physical therapy program for at least 6 months. The physical therapy program begins with light mobility and muscle strengthening exercises and gradually progresses to more challenging exercises. Individuals who do not undergo surgery can also benefit from physical therapy. In these cases, the physical therapy program is less intensive and focuses on reducing pain and swelling and improving mobility. On average, individuals undergoing ACL surgery typically attend around 20–30 physical therapy sessions. Those who do not undergo surgery may require fewer sessions, but it’s important to follow the therapist’s guidance.
| 5 | 4 |
| How long does it take to recover from a cruciate ligament injury? | The time required to recover from an anterior cruciate ligament (ACL) injury varies significantly based on the severity of the injury, the chosen treatment, and the individual patient’s response. For partial or less severe injuries, recovery can take from a few weeks to several months with the appropriate period of physiotherapy and rehabilitation. In the case of a complete ACL injury, the healing timeline can be longer. After ACL reconstruction surgery, patients often require 6 to 12 months or more to return to normal sports activities or daily functions, although the timing can vary depending on individual circumstances. During the recovery, the supervision of an experienced physiotherapist is essential to ensure effective and safe rehabilitation. Following the treatment plan prescribed by the doctor is crucial to optimize healing and minimize the risk of further injuries or complications. | 4 | 4 | The time required to recover from an anterior cruciate ligament (ACL) injury depends on the severity of the injury and the type of treatment received. For a complete rupture of the ACL, complete healing can take from 6 to 12 months. Surgery is generally the recommended treatment for complete ACL injuries. The surgical procedure involves replacing the torn ligament with a tendon taken from another part of the body, such as the leg or the calf. After surgery, the individual will need to follow a physiotherapy program to regain knee mobility and strength. For a partial ACL injury, healing can take from 3 to 6 months. Non-surgical treatment may be sufficient for partial ACL injuries. Non-surgical treatment involves rest, ice, compression, and elevation (RICE), as well as physiotherapy. Regardless of the type of treatment received, it’s important to follow the guidance of the doctor and the physiotherapist to promote recovery. | 5 | 3 |
| What’s the difference between an injury and a tear of the cruciate ligament? | An injury and a tear of the anterior cruciate ligament (ACL) represent two different levels of severity in damaging the ligament itself.
| 4 | 4 | The anterior cruciate ligament (ACL) is an important ligament located inside the knee. Its function is to prevent the tibia from moving forward relative to the femur. An ACL injury is damage to the ligament that can range from a simple strain to a complete tear. A strain of the ACL is a minor injury that causes stretching or straining of the ligament. Symptoms of an ACL strain include pain, swelling, and stiffness. A tear of the ACL is a more severe injury that results in a cut or tear of the ligament. Symptoms of an ACL tear include intense pain, significant swelling, and knee instability. In the case of an ACL tear, it is necessary to consult a doctor for an accurate diagnosis. The doctor can perform a physical examination and may request further tests, such as an MRI, to confirm the diagnosis. | 4 | 4 |
| When should the anterior cruciate ligament be operated on? | Surgery for an anterior cruciate ligament (ACL) injury is often recommended in specific situations, but the decision depends on the severity of the injury and the individual needs of the patient. Common indications for ACL surgery include:
| 5 | 5 | The decision to undergo anterior cruciate ligament (ACL) surgery is a complex one that should be made on a case-by-case basis. There are several factors to consider, including the person’s age, level of physical activity, and the severity of the injury. In general, surgery is recommended for young and active individuals who wish to return to high-impact sports like soccer, basketball, or skiing. Surgery can help restore knee stability and reduce the risk of future injuries. For less active individuals or those who do not wish to return to high-impact sports, non-surgical treatment may be an option. Non-surgical treatment can include rest, ice, compression, elevation (RICE), physiotherapy, and bracing. Ultimately, the decision to undergo ACL surgery is a personal one and should be based on individual needs and preferences. Here are some of the factors that can influence the decision to undergo ACL surgery:
| 4 | 5 |
| How long to wait before undergoing ACL surgery | The ideal timing for anterior cruciate ligament (ACL) reconstruction surgery can vary based on the individual circumstances of the patient and the recommendations of the orthopedic surgeon. However, there are some general guidelines to consider:
| 5 | 4 | The waiting time before undergoing ACL surgery is a decision that must be made on a case-by-case basis, considering the severity of the injury, the person’s age, and their activity level. In general, it is advisable to perform ACL surgery as soon as possible after the injury to reduce the risk of further injuries and improve the chances of recovery. However, in some cases, it may be necessary to wait before surgery. For example, if the injury is recent and the knee is still swollen, the doctor may recommend waiting for a few weeks before surgery. This is because surgery can be more difficult and painful if the knee is swollen. In other cases, the doctor may recommend waiting until the person has completed non-surgical treatment, such as rest, ice, compression, elevation (RICE), and physiotherapy. This is because non-surgical treatment can help reduce pain and swelling and improve knee mobility. Ultimately, the decision of when to undergo ACL surgery is a personal one that should be made based on individual needs and preferences. It is important to discuss all the factors involved with your doctor to make the best decision for your health.
| 4 | 5 |
Questions and answers produced by ChatGPT and Google Bard in pediatric orthopedics.
| Question | ChatGPT-3.5 | Score #1 | Score #2 | Google Bard | Score #1 | Score #2 |
|---|---|---|---|---|---|---|
| What problems does flat feet pose? | Flat foot, or pes valgus, is a condition in which the natural arch of the foot is reduced or absent, causing a flattening of the foot when it makes contact with the ground. This condition can lead to various problems and discomfort:
| 5 | 5 | Flat feet is a condition in which the arch of the foot, which is the natural curvature of the foot, is absent or reduced. This can lead to a range of problems, including:
| 3 | 3 |
| How is flat feet treated? | The treatment for flat feet depends on the severity of symptoms and the underlying cause. Here are some treatment options:
| 4 | 4 | The treatment for flat feet depends on the cause and severity of the condition. In some cases, flat feet can be managed with exercises and physical therapy. In other cases, orthotic inserts may be necessary, or in more severe cases, surgery may be required.
| 3 | 5 |
| How can you tell if you have flat feet? | To determine if you have flat feet, you can make some simple observations and perform tests at home:
| 3 | 4 | Flat feet are a condition in which the arch of the foot, which is the natural curvature of the foot, is absent or reduced. This can lead to a range of problems, including pain, muscle fatigue, balance issues, and knee instability.
| 2 | 4 |
| Which sports are suitable for flat feet? | The choice of sports for individuals with flat feet should take into account the severity of the condition and recommendations from a doctor or physiotherapist. In general, many people with flat feet can participate in a variety of sports, but they may need to consider the following:
| 5 | 3 | People with flat feet can engage in a variety of sports. However, it’s important to choose sports that do not strain the feet and joints.
| 3 | 4 |
| What happens if flat feet are not treated? | If not treated properly, flat feet can lead to a range of long-term problems. These consequences may include:
| 4 | 3 | If flat feet are not treated, a range of problems can occur, including:
| 3 | 3 |
| What causes flat feet? | Flat feet can have various causes, and it’s important to identify the specific cause for appropriate treatment. Some of the main causes of flat feet include:
| 4 | 4 | Flat feet are a condition in which the natural arch of the foot is absent or reduced. This can be attributed to various factors, including:
| 3 | 3 |
| When to operate flat feet? | Surgery for flat feet is typically considered only when conservative treatments, such as orthopedic insoles, physical therapy, and muscle strengthening exercises, have failed to improve symptoms or when the severity of flat feet results in significant limitations in the quality of life or significant structural deformity. Surgical indications may include:
| 3 | 2 | Surgery for flat feet is a rare measure but may be necessary in some cases, such as:
| 2 | 2 |
| How to treat adult flat feet? | To address flat feet in adults, you can consider the following treatment options:
| 4 | 3 | Adult flat feet can be treated in various ways, depending on the severity of the condition. For mild flat feet, treatment may include:
| 5 | 4 |
| How do you walk after flat foot surgery? | After surgery to correct flat feet, the walking pattern will gradually change during the rehabilitation period. Here’s an overview of what the post-operative walking process might look like:
| 4 | 3 | The way you walk after flat foot surgery can vary depending on the type of surgery performed and the severity of the condition. In general, it’s essential to follow the doctor’s instructions for a proper recovery.
| 4 | 4 |
| How to alleviate flat foot pain? | To alleviate the pain associated with flat feet, you can take a series of measures and treatments:
| 5 | 4 | There are several ways to alleviate flat foot pain. Here are some tips:
| 5 | 5 |
Average scores of Google Bard and ChatGPT.
| Sports Medicine | |
|---|---|
| Google Bard | Average score 4 ± 0.78 |
| ChatGPT | Average score 4.1 ± 0.7 |
| No statistically significant difference | (p = 0.6787) |
| Pediatric Orthopedics | |
| Google Bard | Average score 3.5 ± 1 |
| ChatGPT | Average score 3.8 ± 0.83 |
| No statistically significant difference | (p = 0.3092) |
Potential AI-based NLP model improvements.
| Source Accessibility | Platforms should ensure access to reliable and up-to-date sources for providing medical information. This could involve improving the management of hyperlinks, ensuring that they are always accessible and current. |
| Quality Control of Responses | It is essential to implement systems to ensure the accuracy and reliability of the information provided. This could include automatic verification of responses before publication and the ability for users to report incorrect or misleading content. |
| Continuous Updates | Given the rapid advancements in the field of medicine, platforms should be continuously updated with the latest discoveries and clinical guidelines. This may require frequent review and integration of training data to ensure the relevance and accuracy of the information provided. |
| Patient Engagement | Integrating patient feedback into the development and improvement process of AI platforms can help identify areas for improvement and personalize the user experience to meet individual needs. |
| Transparency and Accountability | Platforms should be transparent about their capabilities and limitations, educating users on the proper use of the information provided and encouraging critical evaluation of content. Additionally, they should be clear about the data used for training and the sources of the information provided. |
References
1. Li, B.; Zhang, M.; Lu, Q.; Zhang, B.; Miao, Z.; Li, L.; Zheng, T.; Liu, P. Application and Development of Modern 3D Printing Technology in the Field of Orthopedics. BioMed Res. Int.; 2022; 2022, 8759060. [DOI: https://dx.doi.org/10.1155/2022/8759060]
2. Myers, T.G.; Ramkumar, P.N.; Ricciardi, B.F.; Urish, K.L.; Kipper, J.; Ketonis, C. Artificial Intelligence and Orthopaedics: An Introduction for Clinicians. J. Bone Jt. Surg. Am.; 2020; 102, pp. 830-840. [DOI: https://dx.doi.org/10.2106/JBJS.19.01128] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32379124]
3. Luca, A.; Giorgino, R. Augmented and virtual reality in spine surgery. J. Orthop.; 2023; 43, pp. 30-35. [DOI: https://dx.doi.org/10.1016/j.jor.2023.07.018]
4. Luca, A.; Giorgino, R.; Gesualdo, L.; Peretti, G.M.; Belkhou, A.; Banfi, G.; Grasso, G. Innovative Educational Pathways in Spine Surgery: Advanced Virtual Reality-Based Training. World Neurosurg.; 2020; 140, pp. 674-680. [DOI: https://dx.doi.org/10.1016/j.wneu.2020.04.102] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32797993]
5. Wu, W.; Sabharwal, S.; Bunker, M.; Sabharwal, S. 3D Printing Technology in Pediatric Orthopedics: A Primer for the Clinician. Curr. Rev. Musculoskelet. Med.; 2023; 16, pp. 398-409. [DOI: https://dx.doi.org/10.1007/s12178-023-09847-x] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37335502]
6. Giorgino, R.; Alessandri-Bonetti, M.; Luca, A.; Migliorini, F.; Rossi, N.; Peretti, G.M.; Mangiavini, L. ChatGPT in orthopedics: A narrative review exploring the potential of artificial intelligence in orthopedic practice. Front. Surg.; 2023; 10, 1284015. [DOI: https://dx.doi.org/10.3389/fsurg.2023.1284015] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38026475]
7. Abdullah, Y.; Alokozai, A.; Mathew, A.J.; Stamm, M.A.; Mulcahey, M.K. Patient Education Materials Found via Google Search for Shoulder Arthroscopy Are Written at Too-High of a Reading Level. Arthrosc. Sports Med. Rehabil.; 2022; 4, pp. e1575-e1579. [DOI: https://dx.doi.org/10.1016/j.asmr.2022.04.034] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36033169]
8. Cassidy, J.T.; Baker, J.F. Orthopaedic Patient Information on the World Wide Web: An Essential Review. J. Bone Jt. Surg. Am.; 2016; 98, pp. 325-338. [DOI: https://dx.doi.org/10.2106/JBJS.N.01189] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26888683]
9. Moons, P.; Van Bulck, L. Using ChatGPT and Google Bard to improve the readability of written patient information: A proof-of-concept. Eur. J. Cardiovasc. Nurs.; 2023; 23, zvad087. [DOI: https://dx.doi.org/10.1093/eurjcn/zvad087] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37603843]
10. Alessandri-Bonetti, M.; Liu, H.Y.; Giorgino, R.; Nguyen, V.T.; Egro, F.M. The First Months of Life of ChatGPT and Its Impact in Healthcare: A Bibliometric Analysis of the Current Literature. Ann. Biomed. Eng.; 2023; 52, pp. 1107-1110. [DOI: https://dx.doi.org/10.1007/s10439-023-03325-8] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37482572]
11. Alessandri-Bonetti, M.; Giorgino, R.; Naegeli, M.; Liu, H.Y.; Egro, F.M. Assessing the Soft Tissue Infection Expertise of ChatGPT and Bard Compared to IDSA Recommendations. Ann. Biomed. Eng.; 2023; 52, pp. 1551-1553. [DOI: https://dx.doi.org/10.1007/s10439-023-03372-1]
12. Alessandri Bonetti, M.; Giorgino, R.; Gallo Afflitto, G.; De Lorenzi, F.; Egro, F.M. How Does ChatGPT Perform on the Italian Residency Admission National Exam Compared to 15,869 Medical Graduates?. Ann. Biomed. Eng.; 2023; 52, pp. 745-749. [DOI: https://dx.doi.org/10.1007/s10439-023-03318-7] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37490183]
13. Bernard, A.; Langille, M.; Hughes, S.; Rose, C.; Leddin, D.; Veldhuyzen van Zanten, S. A systematic review of patient inflammatory bowel disease information resources on the World Wide Web. Am. J. Gastroenterol.; 2007; 102, pp. 2070-2077. [DOI: https://dx.doi.org/10.1111/j.1572-0241.2007.01325.x]
14. Lee, P.; Bubeck, S.; Petro, J. Benefits, Limits, and Risks of GPT-4 as an AI Chatbot for Medicine. N. Engl. J. Med.; 2023; 388, pp. 1233-1239. [DOI: https://dx.doi.org/10.1056/NEJMsr2214184] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36988602]
15. González Corbelle, J.; Bugarín-Diz, A.; Alonso-Moral, J.; Taboada, J. Dealing with hallucination and omission in neural Natural Language Generation: A use case on meteorology. Proceedings of the 15th International Conference on Natural Language Generation; Waterville, MA, USA, 18–22 July 2022; Shaikh, S.; Ferreira, T.; Stent, A. Association for Computational Linguistics: Stroudsburg, PA, USA, 2022; pp. 121-130. Available online: https://aclanthology.org/2022.inlg-main.10 (accessed on 31 October 2023).
16. Winship, B.; Grisell, M.; Yang, C.B.; Chen, R.X.; Bauer, A.S. The quality of pediatric orthopaedic information on the internet. J. Pediatr. Orthop.; 2014; 34, pp. 474-477. [DOI: https://dx.doi.org/10.1097/BPO.0000000000000125]
17. Diaz, J.A.; Griffith, R.A.; Ng, J.J.; Reinert, S.E.; Friedmann, P.D.; Moulton, A.W. Patients’ use of the Internet for medical information. J. Gen. Intern. Med.; 2002; 17, pp. 180-185. [DOI: https://dx.doi.org/10.1046/j.1525-1497.2002.10603.x] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/11929503]
18. Rainie, L.; Brenner, J.; Purcell, K. Search Engine Use 2012; Pew Research Center: Washington, DC, USA, 2012; Available online: https://www.pewresearch.org/internet/2012/03/09/search-engine-use-2012/ (accessed on 6 November 2023).
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
© 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.
Abstract
Background: This study evaluates the potential of ChatGPT and Google Bard as educational tools for patients in orthopedics, focusing on sports medicine and pediatric orthopedics. The aim is to compare the quality of responses provided by these natural language processing (NLP) models, addressing concerns about the potential dissemination of incorrect medical information. Methods: Ten ACL- and flat foot-related questions from a Google search were presented to ChatGPT-3.5 and Google Bard. Expert orthopedic surgeons rated the responses using the Global Quality Score (GQS). The study minimized bias by clearing chat history before each question, maintaining respondent anonymity and employing statistical analysis to compare response quality. Results: ChatGPT-3.5 and Google Bard yielded good-quality responses, with average scores of 4.1 ± 0.7 and 4 ± 0.78, respectively, for sports medicine. For pediatric orthopedics, Google Bard scored 3.5 ± 1, while the average score for responses generated by ChatGPT was 3.8 ± 0.83. In both cases, no statistically significant difference was found between the platforms (p = 0.6787, p = 0.3092). Despite ChatGPT’s responses being considered more readable, both platforms showed promise for AI-driven patient education, with no reported misinformation. Conclusions: ChatGPT and Google Bard demonstrate significant potential as supplementary patient education resources in orthopedics. However, improvements are needed for increased reliability. The study underscores the evolving role of AI in orthopedics and calls for continued research to ensure a conscientious integration of AI in healthcare education.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
Details
; Alessandri-Bonetti, Mario 2
; Matteo Del Re 3 ; Verdoni, Fabio 3 ; Peretti, Giuseppe M 4
; Mangiavini, Laura 4 1 Residency Program in Orthopaedics and Traumatology, University of Milan, 20122 Milan, Italy; IRCCS Ospedale Galeazzi Sant’ambrogio, 20157 Milan, Italy;
2 Department of Plastic Surgery, University of Pittsburgh Medical Center, 1350 Locust Street, Pittsburgh, PA 15213, USA;
3 IRCCS Ospedale Galeazzi Sant’ambrogio, 20157 Milan, Italy;
4 IRCCS Ospedale Galeazzi Sant’ambrogio, 20157 Milan, Italy;




