Japan Association for Medical Informatics

The 40th Joint Conference on Medical Informatics / APAMI2020

The 40th Joint Conference on Medical Informatics / APAMI2020

Nov 18, 2020 - Nov 23, 2020
Actcity Hamamatsu

[AP3-E1-1-02] An Application of Heterogeneous Mixture Learning for Mild Cognitive Impairment Subtyping

*Masataka Kikuchi1, Kaori Kobayashi1,2, Kensaku Kasuga3, Akinori Miyashita3, Takeshi Ikeuchi3, Eiji Yumoto4, Yasuto Fushimi2, Toshihiro Takeda5, Shirou Manabe5, Kenichi Kamijo2, Yasushi Matsumura5 (1. Department of Genome Informatics, Graduate School of Medicine, Osaka University, Japan, 2. Medical Solutions Division, NEC Corporation, Japan, 3. Department of Molecular Genetics, Brain Research Institute, Niigata University, Japan, 4. Biometrics Research Laboratories, NEC Corporation, Japan, 5. Department of Medical Informatics, Graduate School of Medicine, Osaka University, Japan)

Alzheimer Disease, Mild Cognitive Impairment, Decision Trees

Mild cognitive impairment (MCI) is known as a group at high risk of conversion to dementia, including Alzheimer's disease (AD). Individuals with MCI show heterogeneity in patterns of pathology, and do not always convert to AD. Detailed subtyping for MCI and accurate prediction of the patients who convert to AD may allow for new trial designs and may enable evaluation of the efficacy of a drug with a small number of patients during clinical trials. In this study, we applied the heterogeneous mixture learning (HML) method to identify subtypes of MCI. As a result, we identified eight subtypes of MCI using the HML approach and categorized into three groups in terms of AD conversion. The identification of these subtypes revealed varying conversion rates to AD, as well as differing levels of biological features.