類神經網路在醫學檢驗的應用──以肝病為例

Abstract

隨著經濟發展，國人平均壽命逐漸提高，醫療保健益形重要。定期做健康檢查的人也愈來愈多。據統計指出，約有十分之一的受檢者，可透過健康檢查篩選出疾病。換言之，判讀正常人的資料便佔據了專業醫師大部份的時間，造成無謂的浪費。唯目前台灣利用健康檢查資料進行疾病診斷以提昇檢查效率的研究仍相當罕見，本論文利用健康檢查受檢人員之檢驗數據，藉由倒傳遞類神經網路的學習能力，學習檢驗數據和疾病間的相互關係，建構一疾病診斷模式，一方面輔助醫療人員進行可能疾病症候的篩選，以減輕醫療人員工作負擔；另一方面亦可做為遠距醫療之前端作業。本研究以肝病為例，利用91個無肝病和79個有肝病的混合資料，進行訓練、測試和回想。結果分類的正確率高達88%~94%，可做為專業醫療人員之參考。

Recent economic advances have greatly contributed to the increasing length of an individual average life span, thus making quality medical care a crucial issue. An increasing number of individuals are taking periodic medical examinations, during which, approximately 10% are found to be ill. In other words, doctors spend most of their time diagnosing healthy individuals and recording their information, reflecting a waste of resources. The feasibility of using information from a medical examination to process diagnosed illnesses for increasing inspection efficiency has received scarce attention. By using inspection data from individuals' health examinations, this study presents a neural network-based procedure for diagnosing diseases. The proposed procedure can assist medical care professionals in effectively diagnosing corresponding diseases. The proposed model can be also used as a pre-positive system for telemedicine. A case study involving the diagnosis of liver diseases in ninety one healthy liver patients and seventy nine diseased liver patient’s data demonstrates the effectiveness of the proposed approach. Analysis results show that the proposed model has a diagnosis accuracy of 88~94%. 英文摘要……………………………………………………………………..II 誌謝………………………………………………………………………… III 目錄………………………………………………………………………… IV 圖目錄……………………………………………………………………….VI 表目錄………………………………………………………………………VII 第一章 緒論……………………………………………………………….. 1 1.1 研究背景…………………………………………………………….. 1 1.2 研究動機…………………………………………………………….. 5 1.3 研究目的…………………………………………………………….. 5 1.4 研究限制…………………………………………………………….. 6 1.5 研究架構…………………………………………………………….. 7 第二章 文獻探討………………………………………………………….. 9 2.1 類神經網路簡介…………………………………………………….. 9 2.2 類神經網路於產業之應用………………………………………….15 2.3 類神經網路於疾病診斷與分類之應用…………………………… 18 第三章 研究方法…………………………………………………………. 26 第四章 實例說明…………………………………………………………. 30 4.1 案例說明……………………………………………………………. 30 4.2 疾病自動分類診斷系統模式建構…………………………………. 32 第五章 結論………………………………………………………………. 42 5.1 結論…………………………………………………………………. 42 5.2 未來展望……………………………………………………………. 42 參考文獻……………………………………………………………………. 44 附錄A 健康檢查檢驗項目及意義..………………………………………. 47 附錄B 世界B型肝炎帶原者比率分佈概況.……………………………. 48 附錄C 人類肝炎病毒……………………………………………………... 49