近年人們生活型態的轉變,國人健康狀況普遍不佳,造就許多疾病的產生,其中心血管疾病更是占台灣十大死因的大宗,然而也因為氣候變遷與環境污染越來越嚴重,空氣汙染與氣候異常與心血管疾病如缺血性心臟病、腦血管心臟病等息息相關。本論文研究重大心血管不良事件之影響因素及預測模型,以供日後醫師與學者做為參考。本研究應用Pearson 相關分析討論氣象與環境因子之間互相的關聯性,並利用羅吉斯迴歸找出顯著的變數。另外建立資料採礦方法如隨機森林、支援向量、決策樹、類神經網路等探討不同時間點的環境氣象因子對發生重大心血管不良事件預測模型。實證發現,以羅吉斯迴歸找出影響重大心血管不良事件的顯著變數有二氧化氮、PM2.5、風速、能見度、溫差等。另外建立預測模型發現在1:1 的資料集為最好的預測模式,且最好的預測模型為隨機森林與決策樹模型。
In this study, the data came from National Health Insurance Research Database and selected the person who are 15 years old above and under 100 years old from 2008 to 2013 as the sample of data analyze and modeling. This study also used data mining technology to establish Standard Operation Procedure of National Health Insurance Research Database and built various models such random forest, artificial neural network, decision tree and support vector machine to find out the correlation between environment pollution and major adverse cardiac events. And find influential factors for person suffer from major adverse cardiac events. The results showed that environment pollution factors will cause any affection to major adverse cardiac events and the random forest is the best method to predict person who will suffer from major adverse cardiac events. In conclusion, we hope the result of the study can provide the reference for medical research.