On April 3, 2020, the Federal Service for Surveillance in Healthcare (Roszdravnadzor) registered the Webiomed Clinical Decision Support System (CDSS) software solution as a medical device.
The Webiomed system is the first software product using artificial intelligence that has successfully passed technical and clinical trials and has been certified for use in medicine in the Russian Federation, said Roman Novitsky, director of K-Lab, a Petrozavodsk company that developed the system.
Webiomed is a component to be embedded in medical information systems and other software products for managing healthcare and medical facilities. It saves the doctor’s time to manually scroll through the patient’s medical records and provides a ready-made assessment of risks such as the overall risk of developing atherosclerosis and its complications, the risk of thromboembolic complications in cardiac arrhythmias, the risk of cardiac arrest in hospitalized patients, the severity of community-acquired pneumonia, and other risks.
Based on these assessments, the system generates recommendations for patient management tactics. Webiomed uses two models created by machine learning: one estimates the individual probability of developing cardiovascular diseases and the other, the individual probability of death from coronary heart disease and stroke.
Before being registered as a medical device, the system went through technical and clinical trials. The latter proved Webiomed’s safety and effectiveness on real medical data obtained in the Russian Federation and not used for machine learning, project development director Alexander Gusev said.
“Webiomed Clinical Decision Support Systemis aimed to help doctors assess the risk of a cardiovascular disease prior to its clinical manifestations as well as the risks of progressing cardiovascular diseases; it does not require any additional examinations and plenty of time to begin information analysis. The doctor only has to push one button which starts the system, which will search for information and provide the doctor with necessary data to make a decision as to what prevention program or treatment should be used. Another important aspect of using AI in medicine involves analytical efforts such as the ability to analyze epidemiological occurrence of specific risk factors. This is essential for increasing efficiency of regional healthcare institutions,” noted Tatyana Kuznetsova, Doctor of Medicine, research supervisor of the medical project, and head of the Faculty of Intermediate Course of Internal Medicine, Phthisiology, Infectious Diseases and Epidemiology at Petrozavodsk State University.
The system approbation was conducted in the Yamal-Nenets Autonomous Area, Kirov Region, and several medical organizations in the Republic of Karelia.
During this period, the system has been used for analyzing over one mio EMRs, including retrospective analysis of standard medical examination data, comparative study of precision of assessing risks by both doctors and AI technology, and collecting and analyzing feedback from doctors who used the Webiomed Clinical Decision Support System for a second opinion.
Pilot projects proved practical relevance of the service: the system is indeed able to lessen doctors’ load as regards routine medical data analysis, and to detect risk factors and high-risk patients with greater accuracy, which helps medical organizations to focus on preventive personalized medicine and thus reduce the burden of cardiovascular decease.
By Andrey Nikitin