Translation. Region: Russian Federation –
Source: Center for Diagnostics and Telemedicine
Moscow scientists have assessed the diagnostic accuracy of various artificial intelligence (AI) services for radiology. The recent study evaluated five AI algorithms, including three developed in Russia and two from India and South Korea. The study found that the Russian AI service demonstrated the highest accuracy in detecting pulmonary nodules, outperforming its foreign counterparts. Research results were published in the Chinese Journal of Quantitative Imaging in Medicine and Surgery.
Yuri Vasiliev, CEO of the Diagnostics and Telemedicine Center of the Moscow City Health Department, noted the importance of this achievement. “In a recent study involving Russian and foreign artificial intelligence (AI) services, our own AI solution demonstrated the highest accuracy in detecting lung nodules, ahead of similar solutions from India and South Korea. This achievement is a significant advance in our efforts to improve the quality of medical care.” Currently, radiologists use more than 50 AI services to interpret medical images, and in Moscow, more than 13 million studies have been analyzed using neural networks. The growing AI services market is constantly evolving with solutions that optimize the workload of medical workers while maintaining high standards of work,” Vasiliev said.
“We aim to provide doctors with tools that will not only make their work easier, but also improve the overall quality of medical care. To do this, we have developed a maturity matrix – a comprehensive tool designed to evaluate and compare the effectiveness of various artificial intelligence services. Since the end of 2022, our AI service has consistently occupied a leading position in the field of chest X-ray,” said Yuri Vasiliev.
The AI service, designed for automatic analysis of chest X-rays, currently identifies 14 signs of various pathologies. In addition, it calculates the cardiothoracic ratio and forms a comprehensive X-ray report.
“We have prepared a unique data set consisting of 100 radiographs, including 50 with confirmed pulmonary nodules and 50 without pulmonary nodules. Of these, 25 cases were found to have nodules that were initially questioned by radiologists, but were confirmed using computed tomography. At the same time, the presence of pulmonary nodules in all 50 studies was confirmed by computed tomography,” explained Kirill Arzamasov, head of the Department of Medical Informatics, Radiomics and Radiogenomics at the Center for Diagnostics and Telemedicine.
“The evaluation of AI services was carried out in three stages, at each stage the results were compared with the reference standard confirmed by CT studies. The results showed that the Russian AI service outperforms international analogues in all indicators of diagnostic accuracy.” The dataset is open source access on the website, which allows developers to independently assess the quality of AI services,” said Kirill Arzamasov, head of the Department of Medical Informatics, Radiomics and Radiogenomics at the Center for Diagnostics and Telemedicine of the Moscow Department of Healthcare.
This study is part of a larger experiment to introduce computer vision technologies into the Moscow healthcare system, which will start in 2020. With the support of the Moscow Social Development Complex and the Department of Information Technology, the project aims to promote innovation in the field of artificial intelligence, in particular by supporting developers.
The Diagnostics and Telemedicine Center, established in 1996, plays a leading role in the implementation of AI technologies in medicine in Russia. Its activities are aimed at the development of AI in medicine, the development of diagnostic images, the management of medical units, research and training of medical workers.