Abstract 10958: Ability of Machine Learning Algorithms to Predict Stage 2 Hypertension Using Health Examination Data from Japanese General Population

Abstract

Introduction: There is a large amount of specific health examination data in Japan. Efforts to utilize these data to predict the future onset of cardiovascular diseases are important from the perspective of improving prognosis and QOL as well as medical economy. We evaluated the prediction ability of machine learning algorithms for newly developed 5-year hypertension. Methods: Among those who underwent examinations from 2008 to 2018, individuals whose baseline (year X), 1 year later (year X+1), and 5 years later (year X+5) data were available were included. Hypertension was defined as > 140/90 mmHg or under treatment with antihypertensive drugs. Those with hypertension in year X or X+1 were excluded. Three different machine learning algorithms, Support vector machine (SVM), Logistic regression (LR), and Multilayer perceptron (MLP) to predict hypertension on year X+5 from single-year data (year X) or consecutive 2 years data (year X and X+1) were assessed, and the predictive ability was compared. Anthropometric measurements, blood pressure and pulse rate, laboratory measurements, and lifestyle related indices were included as explanatory variables. Results: A total of 24,652 participants (age 48.0±10.1 years, male 47.5%) were included and 3,596 individuals (14.6%) had hypertension in year X+5. The indices of predictive ability from single-year data were as follows: SVM, AUC 0.80, accuracy 0.71, precision 0.94; LR, AUC 0.80, accuracy 0.86, precision 0.87; MLP, AUC 0.80, accuracy 0.86, precision 0.86. Those from multi-year data were as follows: SVM, AUC 0.85, accuracy 0.75, precision 0.95; LR, AUC 0.85, accuracy 0.87, precision 0.88; MLP, AUC 0.84, accuracy 0.87, precision 0.89. The variables with the highest importance features were systolic blood pressure (0.49), diastolic blood pressure (0.26), waist circumference (0.13), and body mass index (0.06). Conclusions: The ability to predict new onset of hypertension was better when using data from two consecutive years than when using data from a single year. The AUCs ranged from 0.84 to 0.85 and were generally better than previously reported risk models using statistical methods. Anthropometric measures, along with blood pressure levels, have been identified as major contributors.