MUMBAI, India, Sept. 22 -- Intellectual Property India has published a patent application (202641108301 A) filed by Psg College Of Technology And Applied Research on September 09, 2026, for Ai Based Diagnostic Stethoscope.
Inventors include Dr. Ramesh H; Dr. Julius Fusic S; and Dr. P. R Jain Vinith.
The application for the patent was published on September 18, 2026, under issue no. 38/2026.
Abstract: Title: "AI BASED DIAGNOSTIC STETHOSCOPE" Disclosed herein is an invention typically relates to Al based diagnostic stethoscope integrates advanced artificial intelligence with traditional auscultation methods to diagnose the heart diseases accurately. By utilizing Mel-frequency cepstral coefficients (MFCCs) for feature 10 extraction and employing a hybrid deep learning architecture that integrates Convolution Neural Networks (CNNs) with Long Short-Term Memory (LSTM) networks, Steth-AI enhances the diagnostic accuracy of heart sound analysis. The system extracts 52 MFCC features from each heart sound recording, capturing critical spectral characteristics. The 15 extracted features are then analyzed using various deep learning algorithms, with the LSTM-CNN model demonstrating superior performance in classifying cardiac conditions compared to conventional models. The trained LSTM-CNN model is implemented into an ESP32 microcontroller, and an AI-capable IoT-based mobile application is built 20 for smart and real-time cardiac health monitoring. This method not only overcomes the drawbacks of conventional stethoscopes but also provides a substantial improvement in accessibility and diagnostic accuracy
Disclaimer: Curated by HT Syndication.
