透過您的圖書館登入
IP:3.133.161.153
  • 學位論文

脈診儀訊號特徵分析及脈波分類

Feature Analysis and Classification of Pulse Measurement System

指導教授 : 林康平

摘要


摘要 本論文主要研究目的為提供一種新的脈波分析及分類方法,以數學函數組合模型對脈波進行擬合,藉由數學函數模型擬合脈波於時域所顯示之特徵,修正脈波時域特徵參數,避免時域分析對部分脈波特徵位置不易判斷之困惱。 本研究與長庚醫院中醫診斷研究室進行合作,以具備可攜式功能之廣用型脈診儀為量測儀器,並記錄醫師診斷判別之脈象分數,收錄36位受測者資料,擷取受測者於平均血壓下之脈波資訊,可用資料數66筆。 以Gamma函數為主配合2組Gaussian函數組成模型,藉由powell最佳化搜尋法,找出與脈波最佳化擬合之參數組合。脈波擬合分析結果,脈波於前50%區域擬合精準度達98%以上,脈波全段擬合準確率達90.05%,將擬合脈波所得函數參數,挑選20筆資料做訓練組資料,初步運用於脈波分類訓練上,對全體訓練組資料分類正確率可達85%,其中弦脈正確率為87.5%,滑脈及澀脈皆為83.33%。

關鍵字

脈波分類 脈波擬合

並列摘要


Abstract The aim of this study is to provide a new pulse wave analysis and classification using the combination of mathematical functions fitting the pulse wave. It could modify the pulse characteristic parameters of time-domain analysis and also avoid the time-domain analysis on part of the pulse position is not easy to judge. We use portable pulse signal measurement system measuring 36 subjects pulse wave data, and record the score of pulse types of TCM diagnostics by the R. TCM. P. And we totally corrected 66 available pulse data measuring under mean blood pressure. We used powell method to search the function parameters of the optimal combination of one Gamma and two Gaussian functions fitting the pulse wave. According to the fitting result, we came up with the accuracy 98% in fitting the first 50% pulse area and 90.05% for whole pulse area. At last, we held a test to classify 20 pulse wave documents by mathematical function parameters and came up with a 85% accuracy comparing to the TCM diagnosis, of which 87.5% correct string pulse, slippery pulse, and astringent pulse are all 83.33% accuracy.

參考文獻


[2]. 林汶正、林康平, “以可程式微控制系統晶片為基礎之生理訊號監測系統”,生物醫學工程研討會論文集,2003
[3]. 陳秉淮, “可攜式脈波量測及分析系統”,中原大學碩士論文, 2008.07
[7]. 汪叔游, “脈波圖及其各同步曲線在時域上與傳統脈學之相互印證”,中醫藥雜誌 4(3), 177-190 1993
[12]. 羅文煬, “廣用型脈診系統設計”,中原大學碩士論文, 2009.07
[9]. Xu, L.et al.,“Pulse images recognition using fuzzy neural network”,Expert Systems with Applications(2008), doi:10.1016/j.eswa.2008.02.028

被引用紀錄


江柏儒(2012)。聲、光、壓三通道脈搏訊號紀錄器〔碩士論文,中原大學〕。華藝線上圖書館。https://doi.org/10.6840/cycu201200819

延伸閱讀