Specialists from the I.M. Sechenov First Moscow State Medical University have created an innovative web application capable of automatically recognizing brain tumors on MRI scans. Using computer vision technologies, the program not only finds neoplasms but also determines their type and exact location.
The web application not only finds the tumor and determines its location but also classifies the neoplasm, that is, makes a preliminary conclusion about its nature. The model determines that the detected formation is, for example, a glioma, meningioma, or pituitary adenoma.
The application code is available on GitHub, a platform for collaborative development of IT projects. The advanced open model YOLO v11, trained on more than 5000 standardized images from open sources, was used in the process. Thanks to modern methods of augmentation and experimentation, it was possible to achieve tumor detection accuracy of over 97%.
To date, a working prototype of a web application and a server application with a web interface have been created, allowing demonstrating the operation of the model in real time. In the future, it is planned to improve the accuracy of the model, integrate data from Russian radiological laboratories, and optimize the server application for the convenience of doctors. It is also possible to expand the capabilities of the model so that it can detect not only tumors but also other brain pathologies on MRI scans.
Subscribe to the news of "First Technical" in Google News to stay up to date with the latest events from the world of science and technology.
Read materials on the topic:
The first domestic knee module for prostheses was created in Russia
A test for rapid detection of fungal infection in the lungs was created in Russia