Open AccessPrediction model of physical activity level and hypertension based on artificial neural network
Author(s) -
Liang Hua,
Yu Wang,
Jinying Li,
Wanjun Zhang,
Yi-Ting Wang
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1848/1/012096
Subject(s) - artificial neural network , physical activity , body mass index , waist , medicine , physical activity level , physical examination , population , predictive modelling , computer science , physical therapy , artificial intelligence , machine learning , environmental health
Purpose: Taking the population of Henan Provincial People’s Hospital Chronic Disease Control Center from July 1, 2020 to December 31, 2020 who came to the hospital for physical examination as the research object, analyze and explore the role and contribution of physical activity level in the prediction of hypertension by using the LSTM network model, which can provide references for the clinical diagnosis of hypertension. Methods: Randomly select 2000 physical examination data, remove missing and invalid data, and preprocess them. Finally, select factors such as gender, age, body mass index, weight grade, waist-to-hip ratio, physical activity level, body fat percentage and other factors to establish a neural network prediction model. Then test and study the model, focusing on exploring the contribution of physical activity level to prediction. Result: The level of physical activity has certain advantages in predicting the prevalence of hypertension, but the predictive ability in the later stage is insufficient