請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/56865
標題: | 動靜脈廔管狹窄偵測之演算法與輔助診斷系統 Monitoring Algorithm and Platform for Arteriovenous Fistula Stenosis in End-Stage Renal Disease Patients |
作者: | Yi-Chun Huang 黃怡君 |
指導教授: | 鍾孝文,林致廷 |
關鍵字: | 動靜脈廔管狹窄,演算法,輔助診斷平台, Arteriovenous Fistula Stenosis,Monitoring Algorithm,Diagnostic Platform, |
出版年 : | 2014 |
學位: | 碩士 |
摘要: | 末期腎臟疾病為世界性的公衛問題,尤其台灣的末期腎臟病在發生率和盛行率都高居世界第一,因此末期腎臟病所附帶的併發症也是亟需解決的醫療問題。大多數末期腎臟病患需仰賴長期血液透析(洗腎)來取代腎臟的功能,定期排除代謝廢物,而動靜脈廔管連接動脈與靜脈,為血液透析常用的血管通路,但長期的使用下,常發生動靜脈廔管狹窄的問題,若無經常性的監測檢查,很可能造成嚴重的併發症,如血栓、出血等等,頻繁的血管疏通手術與治療也造成病患極大的生理與心理壓力。
本動靜脈廔管狹窄輔助診斷系統分為兩個部分:第一部分為判斷廔管附近血管狹窄程度與否的演算法,藉由短時距傅立葉轉換,獲得血管狹窄的聲音訊號特徵,並以分類器建立血管狹窄的模型,以狹窄程度差異較為明顯的手術前後資料為基礎,再加入一般洗腎患者的聲音訊號資料,配合後續追蹤結果,進行演算法調整,以利將來推廣至臨床醫療使用;第二部分為自動化平台的建立,以簡易、自動化的介面與使用方式,自動收取信號、演算法處理、上傳資料庫,期望能讓研究結果不只停留在學術研究階段,而能提供第一線醫護人員便利有效的輔助診斷系統,更加有效的監控病患的血液透析情形,以達到早期診斷、早期治療的目的。 End-stage renal disease (ESRD) is widespread and serious problem around the world, especially in Taiwan. For those suffering from ESRD, regular and long-term hemodialysis is necessary, and arteriovenous fistula (AVF) is a required implant. However, under long-term abnormal blood flow, AVF stenosis is a common and severe problem. Without proper monitoring, frequent treatment and surgeries might have to be conducted, which leads to poor quality of life for the patients. In view of this situation, a reliable and user-friendly computer-aided diagnosis (CAD) system is proposed and realized in this thesis. The system is mainly composed of two parts: one is a reliable diagnosis algorithm, based on blood flow sound analysis. Physiology-based features are extracted by short-time Fourier transform and classified by support vector machine. The other is a wireless cloud platform, realized to apply the diagnosis algorithm to clinical use. An automatic user flow and a user-friendly GUI is designed and developed. The AVF flow signal recorded in the electronic stethoscope would be automatically connected, transmitted and analyzed. The signal and analysis result would be uploaded to the cloud database, which is of high clinical and academic research value. This system aims to assist hospital staff and even patients themselves to effectively monitor the condition of AVF. With this easy-to-use and non-invasive method, the condition of AVF can be frequently monitored and preventive healthcare can be done. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/56865 |
全文授權: | 有償授權 |
顯示於系所單位: | 生醫電子與資訊學研究所 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-103-1.pdf 目前未授權公開取用 | 2.41 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。