|Title:||ECG pattern recognition|
Heart -- Diseases -- Diagnosis
Hong Kong Polytechnic University -- Dissertations
|Pages:||vi, 76 leaves : ill. ; 30 cm|
|Abstract:||Holter ECG has become a common diagnostic tool for monitoring patients who have cardiac diseases. The long term recorded ECG allows cardiologists and physicians to analyze a patient's heart function up to 24 hours continuously. These ECG signals provide information that can be used for detecting the transient arrhythmias, which may not present during the regular or exercise ECG tests in a hospital. Many useful parameters, such as the heart rate variations, atrial and ventricular arrhythmias, are commonly used to evaluate the symptomatic patients and the patients who have had myocardial infarction. During the past years, many computer aided ECG monitoring systems have been reported . Although such systems can detect ECG patterns in real time, they are not suitable for high speed analysis of the long term recorded ECG signals. In this project, an efficient and fast algorithm has been developed for detection of QRS complexes and classification of normal and abnormal beats including premature atrial contraction (PAC), premature ventricular contraction (PVC) and right bundle branch block (RBBB). The algorithm uses slope, amplitude, duration and polarity information to detect the QRS complexes. After locating the positions of the QRS complexes, features such as amplitude, duration of the QRS complex are extracted for classification of the beats. The algorithm has been implemented on 486-DX33 and 486-DX66 personal computers with Turbo C++ Language (version 4.5 for Windows) and has been evaluated using records from MIT/BIH arrhythmia database. The experimental results show that the algorithm correctly detects 99.697% of the QRS complexes and identifies 99.058% of the detected beats. On average, processing 30 minutes ECG data (including QRS detection and ECG pattern recognition) takes 56 seconds and 24 seconds when the algorithm is implemented on 486-DX33 personal computer and 486-DX66 personal computer respectively. In addition, a software written in Turbo C++ Language (version 4.5 for Windows) has been developed to aid in the detection of the QRS complexes and classification of the normal beats, PACs, PVCs and RBBBs. The functions of this software include displaying ECG signals with annotations, detection of QRS complexes, recognition of ECG patterns, and beat-by-beat comparison of annotation files.|
Files in This Item:
|b1403346x.pdf||For PolyU Staff & Students||2.39 MB||Adobe PDF||View/Open|
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item: