MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202641047822 A) filed by Prasad V. Potluri Siddhartha Institute Of Technology on April 15, 2026, for Detection Of Cardiac Arrhythmia Using Multi Perspective Convolutional Neutral Network For Ecg Heartbeat Classification.
Inventors include Dr. V. Sita Mahalakshmi; K. Krishna Sai Ujwal; and Dr. G. Venkata Ramana Reddy.
The application for the patent was published on June 12, 2026, under issue no. 24/2026.
Abstract: The present invention relates to a system and method for· the detection of cardiac arrhythmia using a Multi-Perspective Convolutional Neural Network (MPCNN) for Electrocardiogram (ECG) heartbeat classification. The invention addresses limitations of conventional arrhythmia detection -techniques by providing an automated, accurate, and robust framework for analyzing ECG signals. The proposed system utilizes multiple perspectives of ECG data, including temporal, morphological, and frequency-domain -representations, to enhance feature extraction and classification performance. In one embodiment, the ECG signals are pre-processed to remove noise and artifacts, followed by segmentation into individual heartbeats. The segmented signals are then transformed into multiple . representations, which are simultaneously processed through parallel convolutional neural network branches. ·These branches extract complementary features, which are subsequently fused to generate a comprehensive feature set for classification. The MPCNN model is trained to classify heartbeats into predefined categories of normal and abnormal rhythms, enabling early detection of cardia ; arrhythmias. The system demonstrates improved accuracy, generalization, and robustness compared to single-perspective and traditional machine learning approaches. Additionally, the invention is capable of handling signal variability and imbalanced datasets commonly encountered in clinical scenarios. The proposed method can be implemented in real-time monitoring systems, wearable healthcare devices, and clinical decision support systems, thereby facilitating timely diagnosis and improving patient outcomes.
Disclaimer: Curated by HT Syndication.
