對於重症監護病房的病患而言，急性低血壓的發作是一個會導致不可挽回的器官損傷甚至是死亡的關鍵事件。於2009年PhysioNet/Computers in Cardiology的第10屆的挑戰即是針對ICU病患的AHE事件之預測，參賽者可使用PhysioNet所提供的各種生理訊號。
直至目前為止，所有發表的論文多是使用平均動脈壓(MAP)、舒張壓(diastolic ABP)或收縮壓(systolic ABP)等生理訊號的時域特徵來預測。本文相信心血管系統與心臟系統有著一定的關聯性，且透過頻域的分析，我們才更能夠探知訊號的本質。本文透過心電圖訊號與連續血壓訊號，探討有無發生急性低血壓的重症病患在頻域的能量與集中度上所具有的差異。
本文使用Multiscale Entropy計算頻域能量與集中度隨著時間變化的關係，並藉由這些特徵透過統計學的假設檢定獲得具有顯著差異的分類指標。

Acute hypotensive episodes (AHE) is a critical event that can lead to irreversible organ damage and death in intensive care units (ICU). The goal of the 10 th annual PhysioNet/Computers in Cardiology Challenge is to predict which ICU patients will experience AHE within a forecast window of one hour.
In tackling this problem, most of the previous studies extract their features for AHE prediction from the time history of MAP, diastolic ABP and systolic ABP. In contrast, by exploring the interaction within the cardiovascular system, this work employs frequency domain approach. Toward this goal, this work proposes two feature sets: degree of concentration and energy from the spectrogram of the ECG and ABP signals. The mulstiscale entropy of these features have also been studied. The effectiveness of these features is statically investigated by comparing their means between the AHE and non AHI patient groups.

目 錄
論文審定書 i
誌謝 ii
圖　次 vii
表 次 ix
摘要 x
Abstract xi
第一章 序論 1
1.1 前言 1
1.2 文獻回顧 2
1.3 論文架構 3
第二章 急性低血壓事件與心電圖和血壓訊號 4
2.1 急性低血壓事件 4
2.2 心電圖訊號 4
2.3 心電圖波形特徵 5
2.4 血壓量測原理 5
2.5 血壓訊號特徵 6
第三章 實驗流程與架構 7
3.1 實驗流程設計 7
3.2 挑選ECG與ABP訊號的峰值 8
3.3 波形濾波 9
3.3.1 定值濾波 9
3.3.2 法則Ⅰ 10
3.3.3 法則Ⅱ 12
3.3.4 法則Ⅲ 13
3.4 頻域分析 15
3.4.1 訊號擷取 15
3.4.2 移動視窗 16
3.4.3 頻域特徵 16
3.5 Multiscale Entropy 19
3.6 假設檢定 20
第四章 實驗結果與討論 23
4.1 有無發生AHE病患白天8小時的Mode能量與Mode集中度比較 23
4.2 C1與H1族群於預測視窗前1小時的Mode能量與Mode集中度比較 26
4.3 C與H族群於預測視窗前1小時的Mode能量與Mode集中度比較 27
4.4 ECG Mode集中度的MSE特徵變數 29
4.5 門檻值切割分類法 30
第五章 結論與未來展望 33
5.1 結論 33
5.2 未來展望 33
參考文獻 34
附錄Ⅰ C1組與H2組ECG各Mode集中度特徵變數的t* 37
附錄Ⅱ C1組與H2組ECG各Mode能量特徵變數的t* 40
附錄Ⅲ C1組與H2組ABP各Mode集中度特徵變數的t* 43
附錄Ⅵ C1組與H2組ABP各Mode能量特徵變數的t* 46
附錄Ⅴ C1組與H2組SYSTEM各Mode集中度特徵變數的t* 49
附錄Ⅵ C1組與H2組SYSTEM各Mode能量特徵變數的t* 52
附錄Ⅶ C1組與H1組ECG各Mode集中度特徵變數的t* 55
附錄Ⅷ C1組與H1組ECG各Mode能量特徵變數的t* 58
附錄Ⅸ C1組與H1組ABP各Mode集中度特徵變數的t* 61
附錄Ⅹ C1組與H1組ABP各Mode能量特徵變數的t* 64
附錄ⅩⅠ C1組與H1組SYSTEM各Mode集中度特徵變數的t* 67
附錄ⅩⅡ C1組與H1組SYSTEM各Mode能量特徵變數的t* 70
附錄ⅩⅢ C組與H組ECG各Mode集中度特徵變數的t* 73
附錄ⅩⅣ C組與H組ECG各Mode能量特徵變數的t* 76
附錄ⅩⅤ C組與H組ABP各Mode集中度特徵變數的t* 79
附錄ⅩⅥ C組與H組ABP各Mode能量特徵變數的t* 82
附錄ⅩⅦ C組與H組SYSTEM各Mode集中度特徵變數的t* 85
附錄ⅩⅧ C組與H組SYSTEM各Mode能量特徵變數的t* 88

參考文獻
Auroy Y﹐Narchi P﹐Messiah A﹐Litt L﹐Rouvier B﹐Samii K﹐1997﹐“Serious complications related to regional anesthesia: results of a prospective survey in France﹐”Anesthesiology﹐vol. 87﹐pp 479-486.
Chen X﹐Xu D﹐Zhang G﹐Mukkamala R﹐2009﹐“Forecasting Acute Hypotensive Episodes in Intensive Care Patients Based on a Peripheral Arterial Blood Pressure Waveform﹐”Computers in Cardiology﹐pp 545-548.
Henriques J﹐Rocha T﹐2009﹐“Prediction of Acute Hypotensive Episodes Using Neural Network Multi-models﹐”Computers in Cardiology﹐pp 54-552.
Langley P﹐King ST﹐Zheng D﹐Bowers EJ﹐Wang K﹐Allen J﹐Murra A﹐2009﹐“Predicting Acute Hypotensive Episodes from Mean Arterial Pressure﹐”Computers in Cardiology﹐pp 553-556.
Mneimneh MA﹐Povinelli RJ﹐2009﹐“A Rule-Based Approach For the Prediction ofAcute Hypotensive Episodes﹐”Computers in Cardiology﹐pp 557-560.
Chiarugi1 F﹐Karatzanis I﹐Sakkalis1 V﹐Tsamardinos1 I﹐Dermitzaki TH﹐Foukarakis M﹐Vrouchos G﹐2009﹐“Predicting the Occurrence of Acute Hypotensive Episodes: The PhysioNet Challenge﹐”Computers in Cardiology﹐pp 621-624.
Fournier PA﹐Roy JF﹐2009﹐“Acute Hypotension Episode Prediction Using Information Divergence for eature Selection, and Non-Parametric Methods for Classification﹐”Computers in Cardiology﹐pp 625-628.
Hayn D﹐Jammerbund B﹐Kollmann A﹐Schreier G﹐2009﹐“A Biosignal Analysis System Applied for Developing an Algorithm Predicting Critical Situations of High Risk Cardiac Patients by Hemodynamic Monitoring﹐”Computers in Cardiology﹐pp 629-632.
Jin K﹐Stockbridge N﹐2009﹐“Smoothing and Discriminating MAP Data﹐”Computers in Cardiology﹐pp 633-636.
Jousset F﹐Lemay M﹐Vesin JM﹐2009﹐“Computers in Cardiology / Physionet Challenge 2009: Predicting Acute Hypotensive Episodes﹐”Computers in Cardiology﹐pp 637-640.
Ho TCT﹐Chen X﹐2009﹐“Utilizing Histogram to Identify Patients Using Pressors for Acute ypotension﹐”Computers in Cardiology﹐pp 797-800.
Zhigang Wang﹐Lijuan Lai﹐Dongsheng Xiong﹐Xiaoming Wu﹐2010﹐“Study on Predicting Method for Acute Hypotensive Episodes Based on Wavelet Transform and Support Vector Machine﹐”IEEE Biomedical Engineering and Informatics﹐pp 1041-1045.
T. Rocha﹐S. Paredes﹐P. Carvalho﹐J. Henriques﹐M. Harris﹐2010﹐“Wavelet based Time Series Forecast with Application to Acute Hypotensive Episodes Prediction﹐”IEEE EMBS Buenos Aires﹐pp 2403-2406.
Teresa Rocha﹐Simao Paredes﹐Paulo de Carvalho﹐Jorge Henriques﹐2011﹐“Prediction of acute hypotensive episodes by means of neural network multi-models﹐”Computers in Biology and Medicine 41﹐pp 881-890.
劉勝義，2011，“睡眠醫學實務”，合記圖書出版社，pp 157-158.
B.P. Lathi﹐1998﹐“Signal Processing & Linear System﹐”OXFORD﹐P194.
Madalena Costa﹐Ary L﹐Goldberger﹐C.-K. Pen﹐2002﹐“Multiscale Entropy Analysis of Complex Physiologic Time Series﹐” PHYSICAL REVIEW LETTERS﹐vol 89﹐068102.
Madalena Costa﹐Ary L. Goldberger﹐C.-K. Pen﹐2005﹐“Multiscale entropy analysis of biological signal﹐”PHYSICAL REVIEW﹐021906

Martina Udovičić﹐Ksenija Baždarić﹐Lidija Bilić-Zulle﹐Mladen Petrovečki﹐2007﹐“What we need to know when calculating the coefficient of correlation﹐” Biochemia Medica﹐17(1):10-5.
Rangaraj M. Rangayyan﹐2002“Biomedical Signal Analysis : A Case-Study Approach﹐”IEEE Press Series in Biomedical Engineering﹐
李宜容﹐2010﹐“使用不同組合的腦電圖、眼動圖及肌動圖訊號自動偵測快速動眼脆眠期﹐”碩士論文﹐國立中山大學機械與機電工程學系