MUMBAI, India, Nov. 28 -- Intellectual Property India has published a patent application (202531108425 A) filed by National Institute Of Technology, Durgapur, West Bengal, on Nov. 7, for 'system and method for adaptive representation and classification of electrocardiogram (ecg) signals.'
Inventor(s) include Singh, Raghuwansh; Ranjan, Vivek; Ganguly, Anindita; and Halder, Suman.
The application for the patent was published on Nov. 28, under issue no. 48/2025.
According to the abstract released by the Intellectual Property India: "The present disclosure pertains to a system (102) for adaptive representation and classification of the electrocardiogram (ECG) signals. The system (102) receives ECG signals from sensors, identifies fiducial reference points within the ECG signals corresponding to cardiac events and extracts a fixed-duration window aligned to each of the fiducial reference points to generate a segment of interest. The system (102) performs a baseline correction of the generated segment of interest using a stable portion of the signal, resamples the baseline corrected segment of interest to a uniform number of sample points and projects the resampled segment of interest onto orthogonal bases to generate projection coefficients. Further, the system (102) aggregates the generated projection coefficients over a predefined time interval using statistical measures and classifies the aggregated projection coefficients using a machine-learning model to generate an output corresponding to a physiological condition."
Disclaimer: Curated by HT Syndication.
