Paper Title
Topological Temporal SOM Based Cardiac Arrhythmia Classification

Abstract
Early detection of cardiac arrhythmia is crucial in other to increase the patient’s chance of survival. In this study, a Temporal Self-Organizing Map (TSOM) was used to classify electrocardiogram (ECG) signals. This temporal notion can produce an improvement in learning accuracy for TSOM compared to other clustering types of advance neural approaches, like LVQ and SOM. Interesting results that are attributed mainly to the competitive properties of the TSOM with global consistency was recorded in this study and it unified the informativeness of the instances, which makes it to achieve a global optimum. The proposed method is able to classify the ECG data with high accuracy (97.20 %) and exceeds the results of previous reports of related works, which proves its effectiveness in ECG arrhythmias detection. Keywords - Electrocardiogram ECG, MIT-BIH database, Temporal Self Organizing Map SOM, Premature Ventricular Pathology (PVC).