The paper “Arrhythmia Detection and Classification using Morphological and Dynamic Features of ECG Signals,” submitted to IEEE EMBS, Aug 2010 and written by Can Ye, Miguel Tavares Coimbra and B.V.K. Vijaya Kumar was accepted for publication.

Another VR paper was also submitted, this time to the BTAS conference 2010:

Can Ye, M. Coimbra, B.V.K. Vijaya Kumar, "Investigation of Human Identification using Two-Lead Electrocardiogram (ECG) Signals "

"In this paper, we investigate the applicability of Electrocardiogram (ECG) signals for human identification. Wavelet Transform (WT) and Independent Component Analysis (ICA) methods are applied to extract morphological features that appear to offer excellent discrimination among subjects. The proposed method is aimed at the two-lead ECG configuration that is routinely used in long-term continuous monitoring of heart activity. The information from the two ECG leads is fused to achieve improved subject identification. The proposed method was tested on three public ECG databases, namely, MIT-BIH Arrhythmias Database [1], MIT-BIH Normal Sinus Rhythm Database [2] and Long-Term ST Database [3], in order to evaluate the proposed subject identification method on normal ECG signals as well as ECG signals with arrhythmias. Excellent rank-1 recognition rates (as high as 99.6%) were achieved based on single heartbeats. The proposed method exhibits good identification accuracies not just with the normal ECG signals, but also in the presence of various arrhythmias. This work adds to the growing evidence that ECG signals can be useful
for human identification."

The first VR project paper to be submitted to IEEE EMBS, Aug 2010 was written by Can Ye, Miguel Tavares Coimbra and B.V.K. Vijaya Kumar:

“Arrhythmia Detection and Classification using Morphological and Dynamic Features of ECG Signals,”

Can Ye (in the picture) presented it to all team members during our last meeting.

"In this paper, we propose a new approach for arrhythmia classification based on a combination of morphological and dynamic features computed from Electrocardiogram (ECG) signals. Wavelet Transform (WT) and Independent Component Analysis (ICA) are applied separately to each heartbeat to extract corresponding coefficients providing ‘morphological’ features. In addition, RR interval information is obtained characterizing the ‘rhythm’ around the corresponding heartbeat as ‘dynamic’ features. These two different types of features are then concatenated and Support Vector Machine (SVM) is utilized for the classification of heartbeats into 15 classes. The procedure is applied to the data from two ECG leads independently and the two results are used for the final decision, by rejecting the samples that are inconsistently classified. The proposed method was tested over the entire MIT-BIH Arrhythmias Database and it yields an overall accuracy of 99.91% on 84707 heartbeats, better than any other published results."