MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202641064916 A) filed by Anurag University on May 22, 2026, for A Design And Implementation Of A Real-Time Ecg Analysis System Using Cwt And Cnn.
Inventor includes Dr Akku. Madhusudhan.
The application for the patent was published on June 12, 2026, under issue no. 24/2026.
Abstract: The present invention relates to a design and implementation of a real-time ECG analysis system using CWT and CNN. The proposed system utilizes an AD8232 signal-lead ECG sensor integrated with an Arduino Uno microcontroller to acquire live, analog physiological signals from the human body. To ensure mathematical compatibility with the pre-trained neural network, the microcontroller is programmed to perform precise data acquisition, transmitting digital serial data at strict sampling rate of 360 Hz. A centralized MATLAB architecture acts as the signal processing and inference engine, capturing dynamic data buffers in real-time. The software pipeline employs digital signal processing techniques, including DC offset removal and infinite impulse response(IIR) bandpass filtering (0.5 Hz - 50 Hz), to eliminate environmental noise and baseline wander. The processed ID time-series signal is subsequently transformed into a 2D time-frequency representation using the Continuous Wavelet Transform (CWT). The resulting scalogram images are resized and fed into a transfer-learned AlexNet CNN to perform instantaneous binary classification (Normal vs. Abnormal). The experimental implementation demonstrates the successful integration of budget-friendly, open-source hardware with complex mathematical models. The project identifies and discusses the practical challenges of hardware integration, such as environmental noise sensitivity and domain shift, establishing a robust framework for future portable, AI-driven medical diagnostic edge.
Disclaimer: Curated by HT Syndication.
