An Intelligent Diagnosis System for Electrocardiogram (ECG) Images Using Artificial Neural Network (ANN)

Electrocardiogram (ECG) is a record of the origin and propagation of electrical potential through cardiac muscles. Electrical activity of the heart is called as electrocardiogram. ECG's provide lots of information about heart abnormalities. ECG trace is an image based on which disease detection is made. It represent signal of cardiac physiology, useful in diagnosing cardiac disorders, where ECG can provide a lot of information regarding the abnormality in the concerned patient [5]. ECGs are analysed by the physicians and the interpretation may vary depending upon their experience [5]. Hence this paper focuses on computer based automated system in the analysis of the ECG signals in which the images are fed into the system and the software extracts the ECG signal from the image and feed it to the ANN (Artificial Neural Network) classifier. ANN has its own database of diseased ECG recordings which will match the inputted ECG to the existing one so that they must be diagnosed and interpreted accurately irrespective of the physicians. Furthermore, it shall display the normality or abnormality of the signal so as to start the early treatment for the problems and many lives could be saved. Converting ECG records into a computer based digitised signal reduces the physical storage space and the retrieval of the requisite information can be made quicker and accurate [3]. The soft computing technique used for carrying out the automation is MATLAB 7.7.