FCS
On September 5-6, the annual interdisciplinary forum "Cardiogenetics and Bioinformatics: The Realities of 2024" was held. It was dedicated to the project "Machine Learning Technologies in Predicting Outcomes of Acute Coronary Syndrome", which is being implemented within the framework of a mirror laboratory. The project is a collaboration between the International Laboratory of Bioinformatics and the Scientific and Educational Center of the Medical Institute of Surgut State University, which were the organizers of the forum.
The event was attended by representatives of the mirror laboratory and the "Genetics of Cardiovascular Diseases" consortium, which brings together the efforts of scientific groups and institutes from Moscow, Surgut, St. Petersburg, Novosibirsk, Arkhangelsk, and Tyumen to research the genetics of cardiovascular diseases.
FCS
The project "Machine Learning Technologies in Predicting Outcomes of Acute Coronary Syndrome" aims to create a prognostic system based on artificial intelligence algorithms to assess the risks of developing an adverse event for personalized management of patients with myocardial infarction.
On the first day of the forum, presentations were made by the heads of the mirror laboratory Maria Poptsova (HSE School of Computer Science) and Anton Vorobyov (Surgut State University), as well as representatives from the Academician E.I. Chazov National Medical Research Center of Cardiology, the Hematology Research Center, the Central State Medical Academy, the Russian Medical Academy of Continuous Professional Education, and other medical institutions.
The presentations were focused on topics such as the genetics of pulmonary arterial hypertension, the genomics of dilated cardiomyopathy, predicting outcomes of acute coronary syndrome, multifactorial risk models of cardiovascular diseases, and so on. Ivan Ivanov gave presentations on "Machine Learning Models for Predicting Serious Cardiovascular Events in Patients with Myocardial Infarction and Different VEGFR2 Genotypes" and "Flippases as Regulators of Cellular Programs and Their Relation to Cardiopathologies." The full list of presentations can be found here.
Maria Poptsova
Associate Professor of the Department of Big Data and Information Retrieval, Head of the International Laboratory of Bioinformatics, Academic Supervisor of the "Data Analysis in Biology and Medicine" educational program
This year we are celebrating the fifth anniversary of the cardiogenetics consortium, and I am pleased to note that we have joint projects with five consortium participants and new ones are planned. Thus, thanks to the consortium, a mirror laboratory with Surgut State University on the topic of AI in medicine has emerged, and projects on AI with the "Genotek" company are being implemented. Together with the Academician E.I. Chazov National Medical Research Center of Cardiology, the City Clinical Hospital No. 29, and the "100,000+ Genomes Project" national genetic initiative, we are investigating genetic factors that affect the diagnosis and treatment of the disease.
Bioinformatics and cardiogenetics are developing side by side, albeit not at a rapid pace. They have recently been joined by a third direction—AI—which is expected to add momentum. We are growing so fast that we lack the manpower to implement all our planned projects.
Anton Vorobyev
Deputy Head of the Department of Cardiology, Senior Researcher at the Scientific and Educational Center of the Medical Institute of Surgut State University, Cardiologist at the Surgut Center for Diagnostic and Cardiovascular Surgery
The annual forum on cardiogenetics and bioinformatics is a young but already established scientific and project platform. It is actively developing within the framework of the interregional consortium on the genetics of cardiovascular diseases and attracts a large number of leading Russian specialists in the field of AI, genetics, and personalized cardiology. For me, the value of this event lies in the exchange of expert opinions, the unification of intellectual efforts and biomedical registers for new breakthrough projects in the analysis of omics data.
All the presentations are of high relevance both in Russia and in the world, as the scientific direction of analyzing the genetic data of cardiological patients using machine learning is one of the most current and belongs to the category of cutting-edge research.
Machine learning models for predicting long-term outcomes of myocardial infarction have already been developed by our joint team of HSE—Surgut State University under the leadership of Maria Poptsova. We recently presented an article with the results of this research in one of the leading world medical journals, Frontiers in Medicine.
The future direction of development is the application of large language models for analyzing unstructured information in ten thousand medical records of patients with myocardial infarction in order to search for new risk factors for remote complications of this disease, which is part of a joint project of the mirror laboratory. There is also great interest in a closely related area - deep learning for developing computer vision models that analyze radiographic and tomographic images of the heart and vessels of cardiology patients.
On the second day of the forum, the school "Machine Learning Methods for the Analysis of Medical Data" was held. The school covered the main methods of statistical analysis and preprocessing of patient data, methods for imputing missing medical data, building machine learning models for predicting adverse events, comparing predictive power and methods of interpreting models, and the application of large language models for analyzing unstructured medical records.
The "Cardiogenetics and Bioinformatics: Realities" forum attracts more and more new participants each year who are interested in collaborating with the International Laboratory of Bioinformatics of the Faculty of Computer Science.