MUMBAI, India, July 30 -- Intellectual Property India has published a patent application (202641087197 A) filed by New Horizon College Of Engineering on July 16, 2026, for Multi-Branch Explainable Ecg Intelligence System For Cardiac Disease Prediction And Clinical Decision Support.
Inventors include Chandan Raj B R; Harshith C M; Chandru K; Fredrick George F; Shivani M; Amrutha Varshini D; and Manoj Vrc Swami.
The application for the patent was published on July 24, 2026, under issue no. 30/2026.
Abstract: The present invention relates to a multi-encoder explainable electrocardiogram (ECG) intelligence system for automated cardiac disease identification, retrieval- augmented clinical reasoning, and clinical decision support. The system digitizes ECG data from scanned papers, photographs, PDFs, wearable devices, and digital 10 recordings into standardized ECG signals through preprocessing, waveform extraction, reconstruction, and canonicalization. It generates temporal, morphological, biomarker, spectrogram, wavelet, and Gramian Angular Field (GAF) representations for detecting arrhythmias, atrial fibrillation, myocardial infarction, bundle branch block, ventricular abnormalities, and other cardiac 15 disorders. Multiple encoders for time, morphology, and biomarker features generate complementary representations and are combined into an integrated representation of cardiac diseases called “cardiac embedding.” The embedding enables disease prediction, similarity-based search of similar cases from the past, and explainability of AI through feature attribution, waveform visualization, and attention maps. The 20 system further enables physician-guided continuous learning and generates comprehensive clinical reports containing diagnostic results, retrieved evidence, annotations, and recommendations.
Disclaimer: Curated by HT Syndication.
