Content area
In this paper, we propose a Linux-based operating system, namely, DicomOS, tailored for medical imaging and enhanced interoperability, addressing user-friendly functionality and the main critical needs in radiology workflows. Traditional operating systems in clinical settings face limitations, such as fragmented software ecosystems and platform-specific restrictions, which disrupt collaborative workflows and hinder diagnostic efficiency. Built on Ubuntu 22.04 LTS, DicomOS integrates essential DICOM functionalities directly into the OS, providing a unified, cohesive platform for image visualization, annotation, and sharing. Methods include custom configurations and the development of graphical user interfaces (GUIs) and command-line tools, making them accessible to medical professionals and developers. Key applications such as ITK-SNAP and 3D Slicer are seamlessly integrated alongside specialized GUIs that enhance usability without requiring extensive technical expertise. As preliminary work, DicomOS demonstrates the potential to simplify medical imaging workflows, reduce cognitive load, and promote efficient data sharing across diverse clinical settings. However, further evaluations, including structured clinical tests and broader deployment with a distributable ISO image, must validate its effectiveness and scalability in real-world scenarios. The results indicate that DicomOS provides a versatile and adaptable solution, supporting radiologists in routine tasks while facilitating customization for advanced users. As an open-source platform, DicomOS has the potential to evolve alongside medical imaging needs, positioning it as a valuable resource for enhancing workflow integration and clinical collaboration.
Full text
1. Introduction
Operating systems form the backbone of computational environments, managing resources, executing processes, and serving as the interface between hardware and software applications. In specialized fields such as medical imaging, where precision and reliability are critical, an operating system must support efficient task management and meet the rigorous demands of clinical diagnostics and research. Among the available platforms, Linux [1] stands out for its open-source nature, flexibility, and active developer community, making it a highly adaptable alternative to proprietary systems like Windows and macOS. This adaptability is exemplified by projects such as Debian Med [2], which integrates various medical tools within a Linux-based environment, primarily focusing on ease of installation and software compatibility. While Debian Med excels in providing a pre-configured ecosystem for various medical applications, it relies on third-party tools for imaging workflows, which require additional installation and configuration to support advanced functionalities such as DICOM integration. Similarly, MONAI [3], an open-source framework optimized for deep learning in healthcare, offers powerful tools for training AI models but requires significant technical expertise and does not address the broader challenges of workflow integration in clinical settings. In contrast, DicomOS combines ease of use, pre-configured tools for imaging, and advanced customization tailored to clinical workflows, bridging the gaps identified in these projects. Table 1 summarizes the main differences between DicomOS, Debian Med, and MONAI, highlighting the unique features of each and their suitability for specific use cases in medical imaging. Thanks to its open architecture, Linux enables unrestricted customization of core functionalities, allowing for the creation of environments specifically tailored to clinical needs. Several Linux-based distributions illustrate its adaptability in specialized fields. For example, Neurodebian [4] and Link4neuro [5] provide tools optimized for neuroscience and neuroimaging. Platforms like MITK [6] and OHIF [7] further demonstrate how open-source imaging software enhances accessibility and usability in targeted domains. Neurodebian integrates tools such as FSL [8], AFNI [9], and FreeSurfer [10], creating a unified ecosystem that simplifies setup and ensures software compatibility. These customized environments highlight Linux’s capacity to support field-specific requirements while fostering reproducibility and collaboration, underscoring its potential as a foundation for dedicated radiological systems. The lack of interoperability between specialized software tools in radiology often disrupts workflows. Clinicians frequently rely on platform-specific applications to visualize, analyze, and share diagnostic images, creating barriers to seamless collaboration. For example, a radiologist using a Windows-based tool may encounter compatibility issues with a colleague working on macOS, resulting in delays and inefficiencies. The fragmented nature of medical imaging software exacerbates this problem, with some tools offering only basic visualization while others provide complex, feature-rich interfaces. Recent developments in data-driven graphical user interfaces for DICOM have proposed solutions to dynamically adapt functionalities based on the specific diagnostic scenario, addressing some of these usability challenges [11]. This forces radiologists to switch between applications to complete a single task, such as isolating regions of interest or applying image filters. Consequently, this fragmented approach increases the cognitive load and detracts from the primary objective of efficient diagnosis and reporting. To address these limitations, we present DicomOS, a Linux-based operating system tailored to the needs of radiologists and imaging specialists. DicomOS integrates core Digital Imaging and Communications in Medicine (DICOM) functionalities into a unified platform built on Ubuntu 22.04 LTS. Integrating DICOM functionality natively into the operating system presented several challenges, including ensuring compatibility with diverse DICOM implementations, maintaining performance for large imaging datasets, and creating user-friendly interfaces for accessing complex functionalities. These obstacles were addressed by leveraging widely used libraries such as PyDicom and developing automation scripts to simplify workflows, ensuring seamless integration into the operating system. By consolidating essential imaging operations such as annotation, visualization, and data sharing, DicomOS reduces the dependence on multiple software tools, streamlining workflows and lowering cognitive demand. Radiologists can perform key tasks within a cohesive environment, improving efficiency and collaboration. The open-source nature of Linux also allows DicomOS to be customized for institutional or individual needs, supporting emerging imaging techniques and advanced analytical tools as they become available. DicomOS offers both a graphical user interface and command-line tools, ensuring accessibility for clinicians who require straightforward applications and developers seeking robust customization options. By overcoming the fragmentation of current imaging systems and enabling interoperable, adaptable workflows, DicomOS is designed to enhance workflow integration and facilitate clinical collaboration by consolidating key medical imaging functionalities into a unified platform.
The main contributions of this research are highlighted as follows:
DicomOS is designed as a general-purpose platform, making it adaptable for various specialities such as radiology, cardiology, and oncology.
DicomOS provides a user-friendly interface for clinicians and a command-line environment for developers, allowing medical professionals to perform routine tasks while enabling programmers to customize workflows.
Essential DICOM functions, visualization, annotation, and data manipulation, are built directly into the operating system, reducing the need for multiple external applications and simplifying workflows.
Built on Linux, DicomOS allows for continuous improvement and adaptation to meet evolving imaging needs, offering flexibility that proprietary systems cannot match.
2. Development and Customization of DicomOS: Adapting Ubuntu for Medical Imaging
2.1. System Setup and Customization
To develop DicomOS, the base system selected was Ubuntu 22.04 Desktop LTS (Long-Term Support) [12], chosen for its robustness, comprehensive documentation, and compatibility with open-source medical imaging tools [13]. The installation was performed within a VirtualBox 6.1 [14] virtual environment, running on a macOS High Sierra desktop system with a 2.3 GHz Intel Core i5 processor and 8GB 1333 MHz DDR3 memory. This approach allowed for flexibility during the customization process, allowing for safe testing without interfering with existing hardware configurations. Once the base installation was completed, the VirtualBox Guest Additions were installed to enhance the compatibility between the virtual machine and the host system. These additions provided seamless integration features, such as improved graphical performance and shared clipboard functionality, ensuring a better user experience in the virtualized Ubuntu environment.
2.2. Customization of the Visual Environment
Following the installation, significant effort was devoted to customizing Ubuntu’s visual elements to meet the needs of medical professionals while maintaining an intuitive and recognizable user interface. The Canta theme and Vimix icons were selected for their modern and visually appealing design, which aimed to create an inviting environment for users less familiar with Linux systems. The Vimix theme was chosen for its similarity to the macOS interface, a widely adopted medical imaging platform due to its user-friendly design [15]. This customization helped bridge the transition for radiologists accustomed to macOS. The customization process involved manually modifying the GNOME desktop environment by integrating these themes and additional extensions. One notable enhancement was installing the “Dash to Dock” GNOME extension, which replaced the default dock with a more accessible and customizable interface. This addition was designed to provide a familiar experience for users transitioning from other operating systems, ultimately improving the usability [16].
2.3. Branding and Identity Customization
Another key customization focused on creating a unified DicomOS brand identity, starting with the boot splash screen, the first visual element encountered by users during system startup. Plymouth [17], a graphical boot loader used in Linux systems, manages the animations and visuals displayed during the boot sequence. The default Ubuntu splash screen was replaced with a customized version featuring the DicomOS branding and a distinctive logo reflecting its focus on medical imaging. This process required modifications to Plymouth’s themes and configuration files, allowing the default visuals to be replaced with a branded version unique to DicomOS. The login screen was updated to include DicomOS-specific branding and provide a consistent user experience. General user credentials were pre-configured to ensure easy initial access, while individual accounts could be created later to personalize the environment for different medical professionals. Figure 1 illustrates the customized DicomOS interface, showcasing its modern themes, icons, and user-centered design. This tailored visual setup highlights the system’s adaptability and focus on usability.
2.4. Integration of Medical Applications and Usability Enhancements
One of the primary objectives of DicomOS was to incorporate essential medical imaging tools directly into the operating system. Popular applications such as ITK-SNAP [18] and 3D Slicer [19] were installed, as they provide advanced visualization and segmentation tools critical for radiological analysis [20,21]. However, one major limitation of these applications is that they do not natively support direct execution with a double click in Ubuntu. To make these applications more accessible to users with limited technical expertise, additional steps were taken to create executable versions. For 3D Slicer, the process began with modifying file permissions to allow for the installation of necessary extensions. Following this, a custom script was created to simplify the execution of 3D Slicer. This script was stored in
3. Integration of Medical Workflows and Command-Line Tools
The primary objectives of DicomOS were to incorporate essential medical imaging tools directly into the operating system, providing an intuitive platform for medical professionals while supporting deeper customization options for developers and system administrators.
3.1. Development of Graphical User Interfaces for Medical Imaging
To enhance accessibility for different types of users, DicomOS is designed with two primary interaction pathways: one tailored for medical professionals who require straightforward graphical interfaces, and another optimized for programmers and system administrators who prefer command-line access for automation and customization. The usability improvements for medical professionals were developed based on established best practices for medical imaging workflows, and future evaluations will involve real-world testing with clinicians to refine these features further. Figure 3 illustrates the dual workflow in DicomOS, where Python scripts connect with shell commands and .desktop files to create accessible graphical user interface (GUI) applications. This setup enables clinicians to interact with advanced imaging tools through straightforward, executable interfaces without directly using Python scripts. These GUIs are designed to make tasks like image annotation, anonymization, and conversion accessible to users with limited technical expertise, leveraging clear visual menus and double-click execution to lower the technical barrier. Meanwhile, programmers have terminal access to the Python scripts, which they can execute using command-line commands and consult detailed manual pages (man pages) for customization and integration, described further below. This dual approach bridges the gap between usability and flexibility, enabling seamless collaboration between non-technical users and technical experts. Clinicians benefit from an intuitive interface that simplifies complex imaging tasks, while developers can customize and automate operations without interfering with the graphical workflows. To develop these GUIs, DicomOS utilizes Python 3.9 along with libraries such as
As seen in Section 2.4, where executable versions were created for applications like ITK-SNAP and 3D Slicer to improve accessibility, a similar approach was followed for the custom GUIs developed specifically for DicomOS. This process involved creating shell scripts to invoke the underlying Python code, configuring .desktop files to assign application icons and enabling double-click execution. The GUIs were then systematically deployed to a standardized directory structure, with application files organized under /opt, a location commonly reserved for third-party software installations. This ensures clear separation from user-specific files and system-level components while maintaining compatibility with Linux filesystem standards. To facilitate intuitive access, the .desktop configuration was extended to link the GUI applications to system-wide menus, providing a seamless integration into the graphical environment of the operating system. The applications were further enhanced by associating them with high-resolution, scalable icons stored within the operating system’s icon directories, ensuring visual consistency with native applications. The integration process also included setting appropriate permissions for executables and configuration files, ensuring accessibility for all system users without compromising security or functionality. This method was chosen after extensive testing with tools like PyInstaller [27], which, while designed for creating standalone executables from Python scripts, proved insufficient for this project. The GUIs in DicomOS depend on many external libraries and complex dependencies that PyInstaller could not fully resolve, resulting in incomplete or non-functional executables. By contrast, using shell scripts allowed for greater flexibility in managing these dependencies, ensuring that all required libraries could be properly loaded within the Python environment. Additionally, the .desktop configuration bridges the gap between complex software dependencies and user-friendly interfaces. By associating each GUI with a desktop entry, these applications appear alongside native system tools in the graphical environment, with consistent icons and support for double-click execution. This design abstracts technical details, allowing clinical users to interact with advanced imaging tools as if they were standard, pre-installed applications, eliminating the need for familiarity with Linux systems or command-line operations. This approach prioritizes practical implementation and adaptability, addressing the specific requirements of DicomOS without compromising its accessibility for clinical and technical users. Furthermore, it offers a flexible foundation that can be expanded or refined to accommodate future updates and additional features. Creating these executables is illustrated in Figure 2. The figure demonstrates how Python code is linked to an executable using a shell script and a .desktop entry to assign an icon and support double-click functionality. This makes the GUI applications accessible as native applications within the operating system. This approach effectively bridges the gap between the complexity of the development and the user experience, aligning with the broader objectives of DicomOS to improve usability and accessibility.
3.1.1. Dicom-Annotation GUI
One of the primary applications developed is DCMAnnotator, a program designed to facilitate radiologists’ workflow and improve collaboration among specialists. This allows users to load image files in various formats, including DICOM, JPEG, and PNG and provides an interactive interface for the direct annotation of medical images. Users can add lines, rectangles, and textual notes to highlight areas of interest. Annotations are saved in a separate file associated with the original image, ensuring that the original data remain unaltered, which is crucial for maintaining data integrity.
The DCMAnnotator user interface is designed with simplicity and efficiency, adhering to best practices in human–computer interactions. The annotation tools are accessible via on-screen buttons and keyboard shortcuts, which enhance workflow efficiency. Specifically, the application provides buttons to draw lines and rectangles and to add prominently displayed text annotations for easy access. Keyboard shortcuts are implemented to expedite the annotation process; pressing
3.1.2. Image Converter GUI
Another GUI application developed is Image Converter, which facilitates the conversion between DICOM files and common image formats such as JPEG and PNG. This application allows users to select a single DICOM file or a directory containing multiple DICOM files and convert them into image formats. Conversely, it can convert image files into DICOM format. The conversion process involves normalizing the pixel data and properly handling the image metadata. When converting images to DICOM, the application prompts the user to input essential metadata fields such as the patient name, study description, and series description. This ensures that the resulting DICOM files contain the necessary information for clinical use and compliance with DICOM standards. The user interface of Image Converter is designed to be straightforward, with buttons for selecting the conversion direction (“DICOM to Image” or “Image to DICOM”) and dialogues for selecting input files or directories and specifying the output destination. Error handling mechanisms are implemented to inform the user of any issues during the conversion process, enhancing robustness and user experience. The application provides feedback upon successful conversion and includes the normalization of image pixel data to ensure consistent image quality.
3.1.3. DICOM Anonymizer GUI
In addition, the ImgAnonExtract was developed as a user-friendly tool for ensuring the privacy of DICOM files. The application allows users to select individual files or entire directories and specify which metadata fields to redact, such as patient ID, patient name, study date, institution name, and referring physician name. A set of checkboxes enables users to customize the redaction process, with the option to overwrite the original files or save processed copies to a designated location. A backup function is included to prevent data loss, allowing duplicates of the original files to be created before any modifications. ImgAnonExtract also provides advanced features for metadata extraction, enabling users to save selected information in a CSV file. The customizable selection of metadata fields facilitates data analysis and record-keeping. An intuitive interface displays results and operational logs, offering detailed feedback on the anonymization process and highlighting any errors encountered. The application is designed to align with GDPR principles by providing customizable metadata redaction options, logging anonymization actions for traceability, and supporting the exclusion of sensitive data during metadata export. Combined with an intuitive interface, these measures ensure that medical imaging data are handled securely and transparently, facilitating compliance with privacy regulations and institutional policies [28]. For example, the tool provides a backup function to prevent data loss during the anonymization process and logs all operations to create a complete audit trail. It also supports filtering metadata fields during CSV export, enabling users to exclude sensitive information while retaining data essential for research and analysis. By empowering users with these customizable and secure features, ImgAnonExtract serves as a practical solution for maintaining GDPR compliance in medical imaging workflows.
3.1.4. DICOM File Search GUI
To aid in efficiently locating relevant images within large datasets, the DicomSearch application was developed. This GUI allows users to search and preview DICOM files within a selected directory based on specific metadata fields. Users can specify search criteria such as PatientID, StudyDate, or Modality. The application scans the directory, reads the DICOM files’ metadata, and displays a list of files matching the search criteria. The interface includes options to preview the DICOM files, showing essential metadata for quick reference. Users can also open and view individual DICOM images directly from the application, facilitating the immediate inspection of images of interest.
3.1.5. DICOM File Organizer GUI
Managing large volumes of DICOM files can be challenging without proper organization. The DicomOrg application streamlines the management of DICOM files by organizing them into subfolders based on selected metadata criteria. Users can choose a criterion such as PatientID, StudyDate, or Modality, and the application automatically sorts and moves the DICOM files into corresponding subfolders within the selected directory. This organizational tool helps maintain an orderly file structure, making navigating and managing medical imaging data easier.
3.1.6. DICOM Command Shell GUI
To provide a more interactive experience for users who prefer graphical interfaces over command-line tools, the DicomShell GUI was developed. This application allows users to execute various commands related to DICOM file management through a graphical interface. The available commands include the following:
list : lists all DICOM files in the selected directory.view : displays extended information about a specified DICOM file.analyze : allows for the selection of a criterion for histogram display to analyze the distribution of metadata fields.extract : extracts advanced image features from a DICOM file.extract_all : extracts features from all DICOM files in the directory.compare : compares two DICOM files using the Structural Similarity Index (SSIM) and shows the difference map [29].annotate : adds annotations to a specified DICOM file.
The application includes a help option that provides descriptions of available commands and their usage, enhancing user support and accessibility. By integrating these functionalities into a GUI, the application makes it easier for users without command-line experience to perform complex tasks.
3.1.7. DICOM Server Navigator GUI
With the increasing need for remote access to medical imaging data, the DicomServer was developed to facilitate interaction with remote DICOM servers. This application enables users to navigate the contents of a DICOM server, browse directories and files, and download selected items seamlessly. Users can select files or folders from the server and preview images or download them to a local directory. The application dynamically retrieves server content, decodes file paths to ensure the proper handling of special characters and spaces, and organizes files and folders into a dropdown menu for intuitive navigation. To enhance usability, the interface provides buttons for navigation, image viewing, and downloading, complemented by a responsive logging area that displays feedback on operations and logs any errors encountered during network communication. When users select a DICOM file, the application fetches the data from the server, processes the pixel array, and applies appropriate windowing parameters to render a visually accurate grayscale preview. Window centre and width values, extracted from the DICOM metadata, are used to optimize visualization. If these values are unavailable, the application calculates them dynamically based on pixel intensity distribution. The DicomServer also supports downloading single files or entire folders, with recursive downloading for nested directories. The downloaded files are stored in a user-specified local directory, and subfolders are automatically created to mirror the server structure. For DICOM files, metadata such as Patient ID and acquisition parameters are extracted and displayed in the log, providing additional context for clinical use. Robust error handling ensures reliability, including timeouts for server requests and safeguards against invalid DICOM files. By integrating these advanced functionalities, the DicomServer GUI provides a user-friendly and efficient platform for remote medical imaging data management, ensuring secure and streamlined access to critical resources for clinical and research purposes.
3.1.8. DWI Longitudinal Analysis GUI
For clinicians involved in longitudinal studies, the DWIAnalyze application provides advanced tools for analyzing diffusion-weighted imaging (DWI) data across different time points [30]. This application facilitates the comparison of acute and chronic DWI scans for the same patient by automating critical processes such as image registration, difference computation, and report generation. Users can input directories containing acute and chronic DWI scans, with the application automatically matching patient images based on standardized directory structures. Preprocessing steps, including Gaussian smoothing and intensity normalization, are applied to ensure the consistency and comparability of the data. The workflow begins with the automated alignment of acute and chronic scans through image registration. If images differ in size, the application resamples them to ensure compatibility. Once aligned, different images are computed, highlighting structural changes between the two scans. The application calculates statistical metrics to quantify these changes, including the mean difference, standard deviation, and maximum values. In addition, it calculates the Structural Similarity Index (SSIM), a widely used metric to assess the degree of structural similarity between images. These quantitative analyses are complemented by a visual representation of the central slice of the difference map, color-coded for clarity. The PDF report generated provides a detailed summary of the findings. It includes the file path of the difference image, the computed metrics (mean, standard deviation, and maximum differences), and the SSIM score. The report also offers an interpretation of the results based on these metrics. For instance, high mean difference values may indicate significant structural alterations, while low variability (standard deviation) suggests uniform changes across the brain. The SSIM score is particularly informative, with lower values pointing to substantial structural changes and higher scores indicating stability between scans. The tool’s reliability is further supported by its integration of validated preprocessing techniques, such as Gaussian smoothing and intensity normalization, which minimize artefacts and enhance data consistency. By leveraging established metrics like SSIM and generating comprehensive PDF reports, DWIAnalyze ensures that its outputs accurately reflect structural changes between scans, providing clinicians with robust, interpretable insights into patient progression or recovery. This approach aids clinicians in evaluating disease progression or recovery. For example, in a longitudinal study of a traumatic brain injury patient, the report may reveal regions with significant differences that warrant further investigation or confirm minimal changes indicative of recovery. By combining automated analysis, robust metrics, and visual outputs, DWIAnalyze streamlines longitudinal imaging studies, offering clinicians a powerful tool for informed decision-making and enhanced patient care.
3.1.9. Medical Image Editor GUI
The DicomViewer is an advanced tool that allows users to perform various image processing operations on medical images. It supports loading images in multiple formats, including DICOM, JPEG, and PNG. The application provides a range of functionalities, such as the following:
Image enhancement operations like adjusting contrast, brightness, and saturation.
Applying filters such as sharpening, smoothing, Gaussian blurring, and histogram equalization.
Performing geometric transformations like flipping images horizontally or vertically.
Edge detection using Sobel and Canny operators.
Noise addition and image denoising techniques.
Image cropping and resizing.
The interface includes menus and toolbar icons for easy access to these functions, with visual feedback provided through real-time updates of the displayed image. The application supports batch processing, allowing users to apply operations to all images within a selected folder. Users can save the modified images and maintain a history of modifications to undo changes or restore the original image. Technical considerations in developing the Medical Image Editor included handling different image formats, managing color spaces, and ensuring efficient processing for large images. The application leverages OpenCV for image processing operations and provides an intuitive interface for non-technical users. These GUI applications were designed to be accessible to medical professionals without requiring programming skills or familiarity with command-line interfaces. By providing intuitive interfaces and guiding users through processes such as conversion, anonymization, searching, organizing, and editing, these tools aim to streamline workflows and reduce the potential for errors. The applications can be launched directly by double-clicking their icons, making them readily available within the DicomOS environment.
3.2. Development of Command-Line Tools for Programmers
Recognizing the needs of programmers and system administrators, DicomOS includes a suite of command-line tools as part of the Dicom-to package. These tools enhance the operating system’s versatility by enabling the automation and customization of medical imaging tasks. The primary commands developed are
The
This command converts the DICOM image
This command extracts metadata from all DICOM files within the directory named
The
This command converts DICOM images inside the
4. Results
The development and customization of DicomOS yielded a specialized Linux-based operating system tailored to the needs of medical imaging professionals. The primary outcome was successfully integrating essential medical imaging applications into a user-friendly environment, enhancing usability and workflow efficiency. By customizing the Ubuntu desktop environment with themes and icons resembling macOS, DicomOS provided an interface to reduce the learning curve for users transitioning from macOS. This visual consistency minimizes the learning curve of transitioning to a Linux-based operating system, enhancing user satisfaction and adoption rates [15]. Critical medical imaging applications like ITK-SNAP and 3D Slicer were seamlessly integrated into DicomOS. Custom execution scripts and desktop launchers were developed, enabling these applications to be executed directly from the desktop environment without needing command-line interaction. This integration streamlined workflows by providing quick access to advanced imaging tools, improving the efficiency of medical image analysis. Furthermore, a specialized GUI application suite was developed to address specific needs in medical imaging tasks. Dicom-Annotation, ConvertGUI, and DICOMAnonymizerGUI provided user-friendly interfaces for image annotation, conversion between DICOM and common image formats, and the anonymization of sensitive patient information, respectively. Each GUI tool, as summarized in Table 2, provides intuitive access to essential imaging functions, improving the overall user experience of DicomOS. These applications were designed following best practices in human–computer interaction, emphasizing simplicity and efficiency, including intuitive menus and keyboard shortcuts that facilitated a more efficient workflow, reducing the time required for routine tasks and minimizing potential errors. For programmers and system administrators, command-line tools such as
5. Discussion and Conclusions
5.1. Limitations
While the customization and development of DicomOS demonstrate the potential of adapting open-source operating systems for medical imaging, several challenges and limitations were encountered. A significant limitation of the current implementation is the absence of a distributable ISO (International Organization for Standardization) image for DicomOS. Currently, DicomOS is distributed as a VirtualBox disk image file (
5.2. Future Directions
Future work will address the challenges revealed by the preceding analysis and explore opportunities for improvement. In particular, the next phase will involve conducting in-depth evaluations with a representative sample of users to validate the effectiveness of DicomOS in real-world clinical settings. Detailed data will be collected through structured tests with medical professionals and IT administrators, including metrics, such as task completion time, error rates, and user satisfaction scores. These analyses will provide essential feedback to drive iterative improvements, ensuring that the system continues to meet the evolving needs of its users. Additionally, future development will include comprehensive performance evaluations comparing DicomOS to other medical imaging platforms. These evaluations will measure processor speed, memory usage, and image rendering times to ensure competitive performance while maintaining its focus on usability and modularity. Moreover, expanding the range of integrated applications and tools, including support for emerging imaging modalities and advanced data analysis techniques [36], can enhance the utility and relevance of DicomOS in medical imaging. In particular, integrating AI-based tools represents a critical direction for future development. Recent studies, such as [37], illustrate how machine learning can improve the accuracy and efficiency of diagnostic workflows in radiology. In our work, planned AI tools include automatic segmentation, diagnostic prediction, and anomaly detection, all designed to improve workflow efficiency and diagnostic accuracy. These functionalities will leverage advanced frameworks such as TensorFlow and PyTorch, ensuring compatibility with existing clinical workflows and allowing for future expansions. Automatic segmentation modules offer clinicians an intuitive interface for processing DICOM images, applying segmentation algorithms, and exporting results in formats suitable for clinical reporting. Diagnostic prediction tools will generate interpretable outputs, such as probability scores and visual overlays, highlighting regions of interest to assist clinicians in decision-making. Additionally, anomaly detection algorithms could identify unexpected patterns in imaging data, providing early warnings for potential clinical issues. A potential enhancement for DicomOS includes integrating a GUI for melanoma classification using pre-trained weights from models like the Vision Transformer (ViT) described in [38]. This interface would allow clinicians to classify dermoscopic images by leveraging AI capabilities directly within the DicomOS environment, providing a streamlined and user-friendly workflow for diagnostic support. DicomOS will integrate these AI-based enhancements into its graphical environment to ensure practical usability. These tools will leverage the existing modular architecture of DicomOS, which supports the addition of Python-based plugins and .desktop executables. This architecture enables AI functionalities to be embedded as standalone modules, ensuring compatibility with the current GUI while maintaining consistent workflows for clinicians. This integration will prioritise user-friendliness, minimizing the learning curve for clinicians while maintaining advanced customization options for developers. Furthermore, including interactive tutorials and example workflows will facilitate the adoption and encourage the widespread use of these AI-powered features in diverse clinical settings.
5.3. Conclusions
In conclusion, the development of DicomOS showcases the potential for creating specialized operating systems that directly address the needs of medical imaging professionals. By integrating essential tools, enhancing usability, and providing both GUI and command-line options, DicomOS stands as a valuable resource that can improve efficiency and effectiveness in medical imaging workflows. Addressing current limitations, such as creating a distributable ISO image, and engaging with the user community are recommended to refine further and expand its capabilities, ensuring that it remains a relevant and useful tool in the rapidly evolving field of medical imaging. DicomOS exemplifies how open-source platforms can address specialized needs in professional domains, contributing to the discourse on customizable and cost-effective technological solutions in healthcare, and further positioning it as a model for innovation in clinical and research settings. By addressing current limitations and incorporating cutting-edge innovations, DicomOS has the potential to become an indispensable tool for enhancing efficiency, collaboration, and innovation in medical imaging workflows.
Conceptualization, T.C. and O.G.; methodology, T.C.; software, T.C.; validation, T.C., O.G. and R.P.; formal analysis, T.C. and O.G.; investigation, T.C., O.G. and R.P.; resources, S.V.; writing—original draft preparation, T.C.; writing—review and editing, O.G., R.P. and S.V.; supervision, O.G., R.P. and S.V.; funding acquisition, S.V. All authors have read and agreed to the published version of the manuscript.
Not applicable.
Not applicable.
Data is contained within the article.
The authors declare no conflicts of interest.
The following abbreviations are used in this manuscript:
| AI | Artificial intelligence |
| DICOM | Digital Imaging and Communications in Medicine |
| GNOME | GNU Network Object Model Environment |
| GPU | Graphics Processing Unit |
| GUI | Graphical user interface |
| ISO | International Organization for Standardization |
| SSIM | Structural Similarity Index Measure |
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.
Figure 1. Screenshot of the DicomOS interface, showing the customized theme, icons, and new graphical user interface applications tailored for clinical use.
Figure 2. Example process for creating GUI executables in DicomOS, showing Python code execution through a shell script and desktop entry to facilitate easy user access.
Figure 3. Workflow integration in DicomOS, demonstrating the development of GUI executables for medical professionals and command-line tools for programmers. The two sections are connected by a shared Python script layer, which supports GUI and command-line functionalities. This structure enables DicomOS to cater to the needs of both medical and technical users, providing an intuitive GUI for clinicians while offering direct, customizable access for programmers.
Comparison of DicomOS with other open-source systems for medical imaging. The table highlights the key features of each system, focusing on clinical functionalities and customization capabilities.
| Feature | DicomOS | Debian Med | MONAI |
|---|---|---|---|
| DICOM Support | Native integration with visualization, annotation, and data management capabilities | Limited support through third-party applications | No direct DICOM integration, focused on AI-based medical imaging workflows |
| Ease of Use | User-friendly graphical interface for clinicians and command-line tools for developers | Primarily oriented toward Linux experts | Requires advanced technical expertise for implementation and use |
| Integrated Tools | ITK-SNAP, 3D Slicer, and custom GUIs | Extensive ecosystem, but lacks automated integration | Powerful tools for AI training, but without GUI functionalities |
| Accessibility | Under development for ISO image distribution on physical hardware | Available as Debian packages | Open-source framework for Python |
| Customization | Highly customizable due to its open-source nature and Linux foundation | Limited to the compatibility of available packages | Extendable via Python modules, focusing on AI workflows rather than clinical needs |
Summary of GUIs integrated into DicomOS for medical imaging tasks.
| GUI Application | Function |
|---|---|
| ITK-SNAP | Advanced medical image segmentation and visualization tool, often used for annotating structures in 3D medical images. |
| 3D Slicer | Provides powerful tools for visualization and analysis, including segmentation, registration, and quantitative imaging. |
| DCMAnnotator | Allows users to annotate medical images with lines, rectangles, and text, saving annotations separately to preserve original image data. |
| Image Converter | Converts DICOM files to JPEG, PNG, and other formats, and vice versa, supporting metadata input for DICOM conversions. |
| ImgAnonExtract | Anonymizes selected metadata fields in DICOM files, with options to overwrite or save anonymized copies and create backups. |
| DicomSearch | Enables the search and preview of DICOM files based on metadata fields, facilitating the quick location of relevant images. |
| DicomOrg | Organizes DICOM files into subfolders by criteria such as PatientID or StudyDate, streamlining data management. |
| DicomShell | Provides a GUI for command-line operations like listing, viewing, analyzing, extracting features, and comparing DICOM files. |
| DicomServer | Allows browsing and downloading files from a remote DICOM server, including the preview and secure transfer of selected files. |
| DWIAnalyze | Compares acute and chronic DWI scans for a patient, performing image registration, difference computation, and report generation. |
| DicomViewer | Offers a range of image processing functions (e.g., contrast adjustment, filtering, and edge detection) for various medical image formats. |
References
1. Torvalds, L. Just for Fun: The Story of an Accidental Revolutionary; Harper Audio: New York, NY, USA, 2001.
2. Möller, S.; Krabbenhöft, H.N.; Tille, A.; Paleino, D.; Williams, A.; Wolstencroft, K.; Goble, C.; Holland, R.; Belhachemi, D.; Plessy, C. Community-driven computational biology with Debian Linux. BMC Bioinform.; 2010; 11, S5. [DOI: https://dx.doi.org/10.1186/1471-2105-11-S12-S5]
3. Cardoso, M.J.; Li, W.; Brown, R.; Ma, N.; Kerfoot, E.; Wang, Y.; Murrey, B.; Myronenko, A.; Zhao, C.; Yang, D. et al. Monai: An open-source framework for deep learning in healthcare. arXiv; 2022; arXiv: 2211.02701
4. Halchenko, Y.O.; Hanke, M. Open is not enough. Let’s take the next step: An integrated, community-driven computing platform for neuroscience. Front. Neuroinform.; 2012; 6, 22. [DOI: https://dx.doi.org/10.3389/fninf.2012.00022]
5. Nemoto, K.; Dan, I.; Rorden, C.; Ohnishi, T.; Tsuzuki, D.; Okamoto, M.; Yamashita, F.; Asada, T. Lin4Neuro: A customized Linux distribution ready for neuroimaging analysis. BMC Med. Imaging; 2011; 11, 3. [DOI: https://dx.doi.org/10.1186/1471-2342-11-3]
6. Wolf, I.; Vetter, M.; Wegner, I.; Nolden, M.; Bottger, T.; Hastenteufel, M.; Schobinger, M.; Kunert, T.; Meinzer, H.P. The medical imaging interaction toolkit (MITK): A toolkit facilitating the creation of interactive software by extending VTK and ITK. Proceedings of the Medical imaging 2004: Visualization, Image-Guided Procedures, and Display, SPIE; San Diego, CA, USA, 14–19 February 2004; Volume 5367, pp. 16-27.
7. Ziegler, E.; Urban, T.; Brown, D.; Petts, J.; Pieper, S.D.; Lewis, R.; Hafey, C.; Harris, G.J. Open health imaging foundation viewer: An extensible open-source framework for building web-based imaging applications to support cancer research. JCO Clin. Cancer Inform.; 2020; 4, pp. 336-345. [DOI: https://dx.doi.org/10.1200/CCI.19.00131]
8. Jenkinson, M.; Beckmann, C.F.; Behrens, T.E.; Woolrich, M.W.; Smith, S.M. Fsl. Neuroimage; 2012; 62, pp. 782-790. [DOI: https://dx.doi.org/10.1016/j.neuroimage.2011.09.015]
9. Cox, R.W. AFNI: What a long strange trip it’s been. Neuroimage; 2012; 62, pp. 743-747. [DOI: https://dx.doi.org/10.1016/j.neuroimage.2011.08.056]
10. Fischl, B. FreeSurfer. Neuroimage; 2012; 62, pp. 774-781. [DOI: https://dx.doi.org/10.1016/j.neuroimage.2012.01.021]
11. Gambino, O.; Rundo, L.; Cannella, V.; Vitabile, S.; Pirrone, R. A framework for data-driven adaptive GUI generation based on DICOM. J. Biomed. Inform.; 2018; 88, pp. 37-52. [DOI: https://dx.doi.org/10.1016/j.jbi.2018.10.009]
12. Ubuntu: The Official Ubuntu Documentation. 2023; Available online: https://help.ubuntu.com/ (accessed on 10 October 2023).
13. Topol, E.J. High-performance medicine: The convergence of human and artificial intelligence. Nat. Med.; 2019; 25, pp. 44-56. [DOI: https://dx.doi.org/10.1038/s41591-018-0300-7]
14. Oracle Corporation. Oracle VM VirtualBox Version 6.1. 2023; Available online: https://www.virtualbox.org/ (accessed on 10 October 2023).
15. Ventola, C.L. Mobile devices and apps for health care professionals: Uses and benefits. Pharm. Ther.; 2014; 39, 356.
16. Johnson, C.M.; Johnson, T.R.; Zhang, J. A user-centered framework for redesigning health care interfaces. J. Biomed. Inform.; 2005; 38, pp. 75-87. [DOI: https://dx.doi.org/10.1016/j.jbi.2004.11.005]
17. Plymouth Theming Guide. 2023; Available online: https://www.freedesktop.org/wiki/Software/Plymouth/ (accessed on 10 October 2023).
18. Yushkevich, P.A.; Gao, Y.; Gerig, G. ITK-SNAP: An interactive tool for semi-automatic segmentation of multi-modality biomedical images. Proceedings of the 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC); Orlando, FL, USA, 6–20 August 2016; pp. 3342-3345.
19. Fedorov, A.; Beichel, R.; Kalpathy-Cramer, J.; Finet, J.; Fillion-Robin, J.C.; Pujol, S.; Bauer, C.; Jennings, D.; Fennessy, F.; Sonka, M. et al. 3D Slicer as an image computing platform for the Quantitative Imaging Network. Magn. Reson. Imaging; 2012; 30, pp. 1323-1341. [DOI: https://dx.doi.org/10.1016/j.mri.2012.05.001]
20. Kikinis, R.; Pieper, S.D.; Vosburgh, K.G. 3D Slicer: A Platform for Subject-Specific Image Analysis, Visualization, and Clinical Support. Intraoperative Imaging and Image-Guided Therapy; Springer: New York, NY, USA, 2014.
21. Yushkevich, P.A.; Pashchinskiy, A.; Oguz, I.; Mohan, S.; Schmitt, J.E.; Stein, J.M.; Zukić, D.; Vicory, J.; McCormick, M.; Yushkevich, N. et al. User-guided segmentation of multi-modality medical imaging datasets with ITK-SNAP. Neuroinformatics; 2019; 17, pp. 83-102. [DOI: https://dx.doi.org/10.1007/s12021-018-9385-x]
22. Kushniruk, A.W.; Patel, V.L. Cognitive and usability engineering methods for the evaluation of clinical information systems. J. Biomed. Inform.; 2004; 37, pp. 56-76. [DOI: https://dx.doi.org/10.1016/j.jbi.2004.01.003] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/15016386]
23. Lundh, F. An Introduction to Tkinter. 1999; Voulme 539, 540.Available online: www.pythonware.com/library/tkinter/introduction/index.htm (accessed on 21 April 2024).
24. Bisong, E.; Bisong, E. Matplotlib and seaborn. Building Machine Learning and Deep Learning Models on Google Cloud Platform: A Comprehensive Guide for Beginners; Apress: Berkeley, CA, USA, 2019; pp. 151-165.
25. Pydicom Community. pydicom: DICOM handling in Python. Available online: https://pydicom.github.io (accessed on 21 February 2023).
26. Pianykh, O.S. Digital Imaging and Communications in Medicine (DICOM): A Practical Introduction and Survival Guide; Springer: Berlin/Heidelberg, Germany, 2012; Volume 10.
27. PyInstaller Development Team. PyInstaller: Package Python Applications. Version 5.13.; 2024; Available online: https://pyinstaller.org/en/stable/ (accessed on 6 December 2024).
28. Wilkinson, M.D.; Dumontier, M.; Aalbersberg, I.J.; Appleton, G.; Axton, M.; Baak, A.; Blomberg, N.; Boiten, J.W.; da Silva Santos, L.B.; Bourne, P.E. et al. The FAIR Guiding Principles for scientific data management and stewardship. Sci. Data; 2016; 3, pp. 1-9. [DOI: https://dx.doi.org/10.1038/sdata.2016.18]
29. Brunet, D.; Vrscay, E.R.; Wang, Z. On the mathematical properties of the structural similarity index. IEEE Trans. Image Process.; 2011; 21, pp. 1488-1499. [DOI: https://dx.doi.org/10.1109/TIP.2011.2173206] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/22042163]
30. Bammer, R. Basic principles of diffusion-weighted imaging. Eur. J. Radiol.; 2003; 45, pp. 169-184. [DOI: https://dx.doi.org/10.1016/S0720-048X(02)00303-0] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/12595101]
31. Kerrisk, M. The Linux Programming Interface: A Linux and UNIX System Programming Handbook; No Starch Press: San Francisco, CA, USA, 2010.
32. The Fedora Project. Fedora: The Fedora Project. Available online: https://getfedora.org (accessed on 15 January 2023).
33. Arch Linux Contributors. Arch Linux: A Lightweight and Flexible Linux Distribution. Available online: https://archlinux.org (accessed on 15 January 2023).
34. Docker Inc. Docker: Empowering App Development for Scalability. Available online: https://www.docker.com (accessed on 15 January 2023).
35. Red Hat Inc. Ansible: Simple Automation Across IT Environments. Available online: https://www.ansible.com (accessed on 24 July 2024).
36. Esteva, A.; Robicquet, A.; Ramsundar, B.; Kuleshov, V.; DePristo, M.A.; Chou, K.; Cui, C.; Corrado, G.S.; Thrun, S.; Dean, J. A guide to deep learning in healthcare. Nat. Med.; 2019; 25, pp. 24-29. [DOI: https://dx.doi.org/10.1038/s41591-018-0316-z]
37. Onakpojeruo, E.P.; Mustapha, M.T.; Ozsahin, D.U.; Ozsahin, I. Enhanced MRI-based brain tumour classification with a novel Pix2pix generative adversarial network augmentation framework. Brain Commun.; 2024; 6, fcae372. [DOI: https://dx.doi.org/10.1093/braincomms/fcae372]
38. Cirrincione, G.; Cannata, S.; Cicceri, G.; Prinzi, F.; Currieri, T.; Lovino, M.; Militello, C.; Pasero, E.; Vitabile, S. Transformer-based approach to melanoma detection. Sensors; 2023; 23, 5677. [DOI: https://dx.doi.org/10.3390/s23125677] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37420843]
© 2025 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.