生醫電資所教師研究亮點-111年2月份「賴飛羆教授」

研究主題:AI創新研究計畫 臺大醫神-精準醫療人工智慧輔助決策系統

計畫目標為研發一個「精準醫療AI輔助決策系統」,其中包含電子病歷及醫學數據分析處理、基因資訊庫與相關疾病之診斷與治療、醫療期刊書籍文件資訊擷取、精準醫療人工智慧開發。整合電子病歷、醫療影像等資料,發展出各項技術,例如:基因診斷輔助系統Mviewer、AECOPD發作預測系統等。基因診斷輔助系統Mviewer,整合基因、遺傳疾病相關資料庫,藉由文字探勘模組,協助比對基因型與表現型,以輔助次世代定序結果判讀。而AECOPD發作預測系統是針對慢性阻塞性肺病(COPD)可能會導致患者呼吸困難,是一種常見且易危及生命的肺部疾病。為了降低患者急性發作的風險,AECOPD發作預測系統可藉由智慧型穿戴裝置蒐集病患生活習慣、環境,進行即時的偵測,對未來七天是否有可能發生AECOPD進行預測,其準確度可達到92.1%。

The goal of this project is to develop a clinical decision support system for precision medicine with AI technologies. The project contains electronic medical records analysis, genetic diseases diagnosis, information extraction of medical document, and AI technologies development. By integrating electronic medical records, medical images, and other data, various technologies are developed, such as: genetic diagnosis support system Mviewer, chronic obstructive pulmonary disease with acute exacerbation (AECOPD) prediction system, etc.Genetic diagnosis support system, Mviewer, combines relevant databases of genes and genetic diseases, and compares of genotypes and phenotypes by text mining to assist the interpretation of next generation sequencing results. The AECOPD prediction system is for chronic obstructive pulmonary disease (COPD), a common and life-threatening lung disease that causes patients to have difficulty breathing. In order to reduce the risk of acute exacerbation, the AECOPD prediction system can collect the living habits and environment of patients through the wearable device with real-time detection. For prediction of AECOPD within the next 7 days, the accuracy can achieve 92.1% accuracy.