| Project/Area Number |
17K19930
|
|---|
| Research Category |
Grant-in-Aid for Challenging Research (Exploratory)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Health science and related fields
|
|---|
| Research Institution | Teikyo University |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
佐藤 倫広 東北医科薬科大学, 医学部, 助教 (70717892)
|
|---|
| Project Period (FY) |
2017-06-30 – 2020-03-31
|
| Project Status |
Completed (Fiscal Year 2019)
|
| Budget Amount *help |
¥6,370,000 (Direct Cost: ¥4,900,000、Indirect Cost: ¥1,470,000)
Fiscal Year 2018: ¥390,000 (Direct Cost: ¥300,000、Indirect Cost: ¥90,000)
Fiscal Year 2017: ¥5,980,000 (Direct Cost: ¥4,600,000、Indirect Cost: ¥1,380,000)
|
|---|
| Keywords | 高血圧 / 機械学習 / 成人保健 / 疫学 / 人工知能 |
|---|
| Outline of Final Research Achievements |
For the prediction of hypertension incidence within the next 5 years, the artificial intelligence (AI) was developed based on annual health check-up database from JMDC Inc. We then selected the same number of participants with and without hypertension incidence by the under-sampling method, respectively. We assessed the predictive value of the AI by applying it to data from the Ohasama cohort study. Although the AI developed by the logistic regression method showed better predictive value for incident hypertension than that developed by the neural network method, adequate predictive value was not observed from these two AIs. Categorization of variables, addition of other variables, or adjusting the parameter of neural network model did not significantly enhance the predictive value of AI. We also developed the AI for the prediction of stroke. However, the stroke prediction model from the JMDC database revealed low F value when it was applied to the data from the Ohasama cohort study.
|
| Academic Significance and Societal Importance of the Research Achievements |
JMDCデータで構築した高血圧・脳卒中発症予測の人工知能を大迫研究データに適用することは困難と考えられた。これは学習と検証に用いたデータに含まれる対象者特性の相違が原因と考えられる。傾向スコアマッチングによる両データの特性を一致させる、データのスケール変換などにより大迫研究データと互換性が取れるJMDCデータの再構築をする、といった前処理に関する今後の検討の必要性を明らかにした点で、本研究は萌芽研究として一定の意義を有するものと考える。
|
