MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202621048011 A) filed by Symbiosis International Deemed University on April 15, 2026, for Deep Learning Based Intrusion Detection System For Cyberattack Identification In Internet Of Health Things Networks.
Inventors include Shruti Potey; Sadaf Farooqui; Shreyash Rodge; and Dr. Priya Dasarwar.
The application for the patent was published on June 12, 2026, under issue no. 24/2026.
Abstract: ABSTRACT DEEP LEARNING BASED INTRUSION DETECTION SYSTEM FOR CYBERATTACK IDENTIFICATION IN INTERNET OF HEALTH THINGS NETWORKS The present invention discloses a deep learning based intrusion detection system (100) for identifying cyberattacks in Internet of Health Things (IoHT) networks. The system comprises a Convolutional Neural Network module (110) for binary classification of network traffic into normal and attack categories through spatial feature extraction using convolution layers, activation functions, and pooling layers. A Recurrent Neural Network module (120) performs multi-class classification to identify specific attack types including Denial of Service, Man-in-the-Middle, Reconnaissance, and Brute Force attacks by capturing temporal patterns through hidden state layers. A preprocessing module (130) handles data cleaning, feature extraction, normalization, and label encoding of raw IoHT network traffic data. An anomaly detection module (160) generates real-time alerts and triggers automated protective measures when cyberattacks are detected. The system achieves binary classification accuracy of 98.93% and multi-class classification accuracy of 98.97%, providing robust cybersecurity protection for healthcare environments. [
Disclaimer: Curated by HT Syndication.
