P1‐270: Machine Learning Prediction of Future Incidence of Alzheimer's Disease Using Population‐wide Electronic Health Records
Year: 2019
Type: article
Study design: Observational Study
Abstract: Prediction of future incidence of Alzheimer's disease may help better intervention strategy to delay the onset of diseases. Here we tested a machine learning model to predict future incidence of AD using administrative EHR derived from one million individuals. [Data] We obtained de-identified EHR from one million Korean individuals (representativeness-validated) from 2005 to 2012 (for 8 years) from Korean National Health Insurance Service database. EHR contained diagnoses (ICD-9/10), medications, laboratory values, annual health check-up database, and socio-demographics. [Event of interest] We used EHR from elders age above 65 years old: N=430,133. Our event of interest was the month of incidence of AD. We defined the incidence of AD based on diagnostic codes and prescription of anti-dementia (N=672). [Control samples] Total 100 sets of random bootstrap samples of controls with an equal number to AD cases were drawn repeatedly. [Modeling] We trained and validated random forest, support vector machine, and logistic regression to predict incidence of AD in 1,2,3, and 4 subsequent years. Data was split into training (60%), validation (20%), and test set (20%). In predicting future incidence of AD, random forest showed the best performance in 1 year prediction with accuracy of 0.76±0.03; in 2 year, 0.73±0.02; in 3 year, 0.67±0.03; in 4 year, 0.66±0.04. Average duration of EHR was 1936 days in AD and 2694 days in controls. Features selected in logistic regression included well-known AD risk factors (age, smoking, essential hypertension), as well as relatively new, potential factors (waist, LDL and HDL-cholesterol, hemoglobin, urine glucose, amoxicillin.
Source: Alzheimer s & Dementia
Authors: Ji-Hwan Park, Han-Eol Cho, Jun Min Cha, Jong Hun Kim, Shinjae Yoo, Hyoung-Seop Kim, Jiook Cha
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W2981080702