Developed a software module for automatic image comparison based on classical and advanced distance metrics to enhance the precision of biomedical image analysis. The study addresses the growing demand for intelligent tools capable of quantitative comparison of complex image structures, overcoming the limitations of existing systems such as ImageJ and Axio Vision. The research focuses on integrating Frechet, Hausdorff, Gromov – Frechet, Gromov – Hausdorff, and fuzzy Frechet and Hausdorff metrics within a unified modular architecture designed for distributed data environments. The methods of the research rely on the use of modern programming technologies (Java, PHP, Vue.js, Laravel, MySQL, OpenCV) and the principles of software modularity and service-oriented design. Investigated the design and implementation of a web-based interface that enables users to upload, preview, and compare images interactively. The architecture integrates a RESTful API, microservice-based communication, and visualization components to represent metric results numerically and graphically. Established a relational database schema including entities for users, studies, comparators, and results, ensuring scalability and data integrity. Developed the core module consisting of the RestClient, MainComparator, and a set of specialized metric classes that implement the Comparator interface, allowing easy extension by new algorithms. The developed system demonstrates a substantial improvement over existing software by combining traditional and modern metric-based methods and enabling remote interaction and integration with cloud services. Comparative analysis with ImageJ, Axio Vision, and HIAMS systems revealed that the proposed module uniquely supports both Gromov-based and fuzzy metrics while maintaining a user-friendly web interface and REST API access. The results confirmed the efficiency of the designed module for segmentation and clustering of biomedical images and its integration into the “BRECCAD” system for automatic breast cancer diagnosis. The proposed approach enhances reproducibility, modularity, and analytical accuracy, providing a foundation for further development of intelligent diagnostic software.