23/06/2026
Congratulations to Dr. Tinku Singh (Assistant Professor, ) for the publication of the paper, “An augmented ECG data-based classification for arrhythmia using optimal feature set” in Health Information Science and Systems, Springer.
This paper presents an Intelligent Arrhythmia Classification System leveraging Electrocardiogram (ECG) data for accurate and timely detection of cardiac abnormalities such as arrhythmia. Recognizing the challenges posed by complex ECG signals and class imbalance in abnormal beats, the proposed framework integrates efficient preprocessing techniques with domain-specific knowledge to enhance feature quality. A Multilayer Perceptron (MLP) is employed for feature learning and classification, while the Synthetic Minority Over-sampling Technique (SMOTE) is used to address data imbalance between normal and abnormal classes. To support scalable and real-time analysis, the system is implemented using a three-node cluster with Apache Kafka and Apache Spark. Evaluated on the benchmark MIT-BIH dataset, the model achieves an overall accuracy of 96.4%, with strong performance on minority classes such as Supra-Ventricular Ectopic Beat (SVEB) and Fusion Beat (F), demonstrating high predictive value, sensitivity, and F1-scores. The results indicate that the proposed approach not only improves diagnostic accuracy but also maintains lower computational complexity compared to existing methods, making it suitable for real-time healthcare monitoring applications.
https://link.springer.com/article/10.1007/s13755-026-00440-3
