Stroke is the world’s second leading cause of death and the third leading cause of disability and death combined. Risk factors for stroke are potentially manageable so prevention of this disease is possible. Identification of previously unknown modifiable risk factors for stroke or testing the significance of known factors is an urgent task that should be solved based on a retrospective analysis of the electronic health records of patients with this disease. The paper presents an approach to identifying risk factors for acute cerebrovascular accidents from texts of case histories using natural language processing and machine learning methods. The proposed approach made it possible to identify risk factors for stroke and transient ischemic attack in patients of one of the Moscow clinics. The identified factors are consistent with those found in other studies.
Download PDF from the Proceeding of the Institute for Systems Analysis of the Russian Academy of Science website (in Russian): http://www.isa.ru/proceedings/images/documents/2023-73-2/111-122.pdf
Download PDF from eLibrary (in Russian, registration required): https://elibrary.ru/item.asp?id=54085106
Donitova V. V., Kireev D. A., Kobrinskii B. A., Smirnov I. V., Titova E. V. (2023) Retrieving stroke risk factors based on intellectual analysis of electronic health records // Proceeding of the Institute for Systems Analysis of the Russian Academy of Science, Vol. 73, № 2, pp. 111–122.