A resident of “Skolkovo” registered the II-system for the diagnosis of oncological diseases of the skin

A resident of “Skolkovo” registered the II-system for the diagnosis of oncological diseases of the skin

“AIMED” company, a resident of “Skolkovo” (VEB.RF Group), received a registration certificate for the “ProRodynka” medical decision support system. The service based on neural network technologies is intended to help the doctor in making decisions when diagnosing skin neoplasms based on photographs.

At the moment, the “ProRodynky” mobile application contains 980,000 images received from 550,000 users. During the implementation of the project, more than 23,000 malignant neoplasms were detected, including almost 14,000 cases of melanoma.

The software includes several modules of a convolutional neural network, a trained belief system based on diagnostic features of certain pathologies and mathematical calculations. Using the images of skin neoplasms provided, the neural network evaluates and makes a conclusion comparable to the conclusion of a qualified specialist in the field of dermatovenerology and oncology.

Nosologies recognized by the neural network:

Yuliya Shcheglova, senior project manager of the biomedical technology cluster “Skolkovo”: “Our resident “Aimed” received Russia’s first registration certificate of Roszdravnadzor for the support system for medical decisions in dermatology. The service’s neural network allows you to detect all possible types of malignant formations, as well as other types of skin diseases. Before receiving a registration certificate, users of the “ProRodynky” mobile application could only use it to assess the risk of cancer. Obtaining permission from Roszdravnadzor allows medical organizations to purchase software, and doctors to use it during diagnosis. We are glad that with the grant and business support of Skolkovo, the team brought their product to the market faster.”

The service can be used within the framework of the provision of primary health care, including pre-medical and medical, as well as in the conditions of a day hospital and at home, within the framework of the provision of specialized medical care and in scientific activities.

Veniamin Dardyk, general director of AIMED: “Our product is the result of 5 years of great joint work of doctors and programmers. The built model made it possible to timely detect melanoma in thousands of people. We hope that by obtaining the status of a medical product, we will be able to help people detect diseases in time, in the early stages, and thereby get the opportunity for full treatment.”

The main advantages of this system are:

  • Productivity. The system can process several hundred thousand images during a working day;

  • Efficiency. The system works at the level of a good dermatologist-oncologist. In clinical trials conducted at Blokhin National Medical Center, the system showed a sensitivity of 93.5% and a specificity of 84.3%;

  • Accessibility. If the Internet is available, the system services are available from anywhere in Russia.

The project is implemented with the support of the Volga Research Medical University (PIMU). The university’s team of experts not only acted as creators of the service, but also organizes promotion in the regions of Russia.

Related posts