Intellectual Property India Publishes Patent Application for Iot Based Smart Device To Predict Acute Heart And Cardiac Disease Using Strong Machine Learning Algorithms

MUMBAI, India, July 13 -- Intellectual Property India has published a patent application (202641081790 A) filed by Sathyabama Institute Of Science And Technology on July 02, 2026, for Iot Based Smart Device To Predict Acute Heart And Cardiac Disease Using Strong Machine Learning Algorithms.

Inventors include Dr. S. Bangaru Kamatchi; Dr. Aiswarya S; Dr. J. Jeslin Shanthamalar; Dr. Sheema. D; Dr M. P Vaishnnave; Madhushri K; and Dr. P. M. Lavanya.

The application for the patent was published on July 10, 2026, under issue no. 28/2026.

Abstract: IoT based smart device to predict acute heart and cardiac disease using strong machine learning algorithms ABSTRACT: The Internet of Things facilitates seamless connectivity between individuals and items, while its integration with the Cloud enhances our lives. A multimodal deep learning approach integrating convolutional neural networks (CNNs) and long short-term memory (LSTM) networks for the early identification and prediction of heart disease utilizing data gathered from wearable devices. This integrated multi-model deep learning approach is employed to ascertain precise accuracy and precision values. Then, the features from ECG and PPG are recovered using CNN and the accelerometer characteristics are extracted using the LSTM model. The integrated features are subsequently categorized utilizing a hybrid CNN-LSTM network architecture. The algorithm is assessed utilizing a publicly accessible benchmark dataset.The data from the initial stage will be conveyed to the succeeding stage for a corresponding number of heart disease categories. In the subsequent phase, a TS fuzzy logic system refined by the Giza Pyramids Construction (GPC) methodology (GPC-TS-Fuzzy) is employed to categorize each signal. The MIT-BIH arrhythmia dataset is utilized to assess the suggested strategy. A thorough assessment of the proposed method revealed average performance metrics of 98.58% for accuracy, 98.13% for sensitivity, and 96.47% for specificity. The findings illustrate the efficacy of a multimodal strategy for the early identification and forecasting of cardiac disease with wearable technology and deep learning algorithms.

Disclaimer: Curated by HT Syndication.