MUMBAI, India, Oct. 26 -- Intellectual Property India has published a patent application (202641119653 A) filed by Vardhaman College Of Engineering on October 06, 2026, for System And Method For Real-Time Anomaly Detection In Biomedical Signals Using Wavelet Transform And Machine Learning.
Inventors include G. A. E. Satish Kumar; Vangeepuram Srinivasa Raghavan; Malela Sriram; Bhanothu Manvika; Bandreddy Chandra Sai Durgaa Bhavani; Kummari Dheekshitha; and Goli Tanishka Reddy.
The application for the patent was published on October 23, 2026, under issue no. 43/2026.
Abstract: System and Method for Real-Time Anomaly Detection in Biomedical Signals Using Wavelet Transform and Machine Learning is the proposed invention. The suggested invention collects physiological signals such as electrocardiogram (ECG), electroencephalogram (EEG), electromyogram (EMG), photo plethysmogram (PPG), and respiration signals from one or several sensing devices. The obtained signals undergo preprocessing to diminish noise, baseline fluctuations, and motion artefacts, and are then divided into temporal segments. A multi-tiered wavelet transformation disaggregates each signal into time-frequency elements from which statistical, energy-derived, entropy-related, morphological, and temporal characteristics are obtained. The derived representations are analysed by a specialised state- space machine-learning model designed to recognise temporal physiological trends and adaptively modify latent states in reaction to notable signal fluctuations. An adaptive anomaly scoring system identifies divergences from established normal patterns by integrating wavelet-domain and temporal-state data. When beyond a dynamic threshold, an irregularity is detected and categorised based on its nature and confidence level, prompting an alarm to be issued via a local or remote monitoring system. The suggested innovation facilitates rapid, uninterrupted, and resource-efficient identification of anomalies in biological signals, making it appropriate for wearable, edge, bedside, and distant healthcare monitoring uses.
Disclaimer: Curated by HT Syndication.
