基於SVM遺傳疾病輔助篩檢系統 以台大醫院新生兒資訊系統為例

Abstract

本篇論文主要描述國立台灣大學醫學院附設醫院目前新生兒資訊系統開發的架構與狀況，使得樣本蒐集、檢驗、診斷、評估、治療、追蹤病例服務、篩檢數據資料庫管理、臨床研究、院際通訊的工作更為健全。最後以資料探勘的技術來建立一輔助協助診斷系統提升篩檢正確率。以此基於支持向量機的罕見疾病輔助診斷系統，甲基丙二酸血症為例，來分析此目前疑陽性極高的疾病。

The thesis aims to discuss about architecture and status of Newborn Screening Healthcare Information System of National Newborn Screening Center in National Taiwan University Hospital for further integrity and enhancements of sample collections, testing, diagnoses, evaluations, treatments or follow-up services, screening database management, as well as collaboration, communication among hospitals. Finally, we use data mining technology to build the diagnosis decision support system to improve the screening system accuracy. Then we examine the screening result of MMA which is high false-positive rate by the diagnosis decision support system based on support vector machine.