Machine learning analysis and future risk prediction of weather-sensitive cardiovascular disease mortality during summer in Tokyo, Japan

Abstract

Climate-sensitive diseases developing from heat or cold stress threaten human health. Therefore, the future health risk induced by climate change and aging societies worldwide should be assessed. In this study, we developed the prediction model for mortality of cardiovascular diseases such as myocardial infarction and cerebral infarction, which are known weather- or climate-sensitive diseases, using machine learning techniques. We targeted daily mortality of ischaemic heart disease (IHD) and cerebrovascular disease in the 23 wards of Tokyo and in Osaka City, Japan during summer. The significance of delayed effects of daily maximum temperature and other weather elements on mortality was previously demonstrated using a distributed lag nonlinear model. We conducted machine learning (ML) including specified lag days, with important features of several temperature-related elements and air pressure-related elements for the mortality risk of IHD and cerebrovascular disease during the previous summers, respectively. These models, learned the past data, were used to evaluate the future risk of IHD mortality in Tokyo’s 23 wards owing to climate change by applying transfer learning architecture (TL). The ML incorporating TL predicted that the daily IHD mortality risk in Tokyo was averagely increased 29% and 35% at the 95th and 99th percentiles using a high-level warming climate scenario in 2045–2055, compared to the risk simulated using ML in 2009–2019.