Russian doctors, together with specialists from the SberMedII company, part of the Sber ecosystem, trained
Lung cancer is one of the most aggressive and widespread malignant tumors in the world, as well as the most common cause of death among oncological pathologies.
Russian specialists have improvedalgorithms of the “Lung CT” service, which is already operating in 30 clinics in 19 regions of Russia for the diagnosis of viral pneumonia, including COVID-19. Now, using the service, doctors will be able to find minimal nodular tumors in the lungs and determine their quality. If necessary, the algorithm will highlight all potentially dangerous areas associated with the detected tumor.
To do this, the AI takes into account the size of the neoplasms, andif at least one of them exceeds the threshold value of 4 mm, then the service highlights all detected nodular neoplasms, regardless of their size, thereby outlining the doctor's area of distribution of potentially dangerous cells, which must be paid attention to when analyzing and deciding on further actions.
AI service "CT of lungs" was trained on the basis of 12thousands of anonymized images from open sources of medical data. Radiologists manually marked up all the images and, on their basis, trained artificial intelligence algorithms to identify neoplasms and develop a solution that is optimal for a particular situation.
The "CT of the lungs" service has shown its effectiveness in examining the lungs for the consequences of the coronavirus, so its use for the detection of oncology is quite logical and logical.
“Connecting "CT Lung" to searchsigns of cancer can improve the quality of medical care, and timely diagnosis can save the lives of many people,” said Yuri Krestinsky, vice president and head of the healthcare industry at Sberbank.
Source: Sberbank