Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/10149
Title: | Study of soft computing techniques for ischemia detection in ECGS |
Researcher: | Murugan S |
Guide(s): | Radhakrishnan, S. |
Keywords: | Ischemia, ECGs, soft computing techniques, component analysis, genetic algorithm |
Upload Date: | 29-Jul-2013 |
University: | Anna University |
Completed Date: | |
Abstract: | Myocardial ischemia is the most common cardiac disease and is characterized by high risk of sudden cardiac death. Myocardial ischemia is caused by a lack of oxygen and nutrients to the contractile cells and may lead to myocardial infarction with its severe consequence of heart failure and arrhythmia. An electrocardiogram (ECG) represents a recording of changes occurring in the electrical potentials between different sites on the skin as a result of the cardiac activity. Myocardial ischemia diagnosis using long duration electrocardiographic recordings is a simple and non invasive method that needs further development before being used in everyday medical practice. There are a few mandatory steps for automated detection of ischemic episodes. After the initial removal of noise, it follows the second stage, when all the important ECG features (J point, isoelectric line and T wave peak) are extracted. Using the above features, in the third stage, each cardiac beat is classified as normal or ischemic. The ST T Complex of the ECG represents the time period from the end of the ventricular depolarization to the end of the corresponding depolarization in the electrical cardiac cycle. In this research, two improved versions of Principal Component Analysis (PCA) and Independent Component Analysis (ICA) has been implemented for feature extraction and selection. The Genetic Algorithm (GA) and Fuzzy logic are combined with PCA and ICA to improve their performance; the algorithms are Genetic PCA (GPCA), Genetic ICA (GICA), Fuzzy-Genetic PCA (FGPCA) and Fuzzy-Genetic ICA (FGICA). In the proposed method, the features are classified using a Genetic based Least Square Support Vector Machine (GLSSVM). The performance is then analyzed with two different classifiers. The ECG beats used in this work are collected from European ST-T database, totally 2040 beats extracted from 17 patients. The results demonstrated that the GLSSVM with Fuzzy-Genetic ICA achieved greater accuracy than other automated diagnostic systems. newline |
Pagination: | xv, 137 |
URI: | http://hdl.handle.net/10603/10149 |
Appears in Departments: | Faculty of Electrical and Electronics Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf | Attached File | 33.65 kB | Adobe PDF | View/Open |
02_certificates.pdf | 643.97 kB | Adobe PDF | View/Open | |
03_abstract.pdf | 16.55 kB | Adobe PDF | View/Open | |
04_acknowledgement.pdf | 13.14 kB | Adobe PDF | View/Open | |
05_contents.pdf | 41.34 kB | Adobe PDF | View/Open | |
06_chapter 1.pdf | 49.7 kB | Adobe PDF | View/Open | |
07_chapter 2.pdf | 93.88 kB | Adobe PDF | View/Open | |
08_chapter 3.pdf | 217.95 kB | Adobe PDF | View/Open | |
09_chapter 4.pdf | 306.4 kB | Adobe PDF | View/Open | |
10_chapter 5.pdf | 1.05 MB | Adobe PDF | View/Open | |
11_chapter 6.pdf | 400.78 kB | Adobe PDF | View/Open | |
12_chapter 7.pdf | 28.96 kB | Adobe PDF | View/Open | |
13_references.pdf | 65.97 kB | Adobe PDF | View/Open | |
14_publications.pdf | 14.07 kB | Adobe PDF | View/Open | |
15_vitae.pdf | 14.27 kB | Adobe PDF | View/Open |
Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
Altmetric Badge: