Development of the predictive model for progression of lifestyle-related diseases using health care record and machine learning

2023 Fiscal Year Final Research Report

• Project/Area Number: 21K16093
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 53020:Cardiology-related
• Research Institution: National Cardiovascular Center Research Institute
• Principal Investigator: Kanaoka Koshiro 国立研究開発法人国立循環器病研究センター, オープンイノベーションセンター, 室長 (70873412)
• Project Period (FY): 2021-04-01 – 2024-03-31
• Keywords: 保健医療データベース / 生活習慣病 / 予後予測

Outline of Final Research Achievements

In this study, we constructed matched medical and care data and performed analyses leading to predictions of future cardiovascular diseases and other conditions. We built an integrated database using medical, care, and health checkup data, creating a large-scale cohort. We focused on patients who were first diagnosed with hypertension during health checkups. Among these patients, visits to medical institutions were limited. However, patient who were treated with antihypertensive drugs were associated with subsequent good blood pressure control compared to those who did not receive medical consultation. A similar trend was observed with dyslipidemia, indicating the need for strategies to facilitate smoother connections to medical consultations in the future.

Free Research Field

循環器内科

Academic Significance and Societal Importance of the Research Achievements

本研究で得られた結果をもとに、健診の受診動向と、その後の医療機関受領状況の可視化及び適切な患者を受診へと結びつけるシステムを確立できれば、潜在的に心血管疾患リスクの高い患者に対して早期のアプローチを行うことができ、循環器病を減少させる一つの方策となる。