MUMBAI, India, Aug. 17 -- Intellectual Property India has published a patent application (202641094821 A) filed by Dr. Tamilarasan. S; Mr. Vinay G; Dr. Avinash N; Mr. Sandeep K H; Mr. Chethan P J; Dr. Subrahmanya H M; Mr. Shravan H S; Mr. Lohitesh Kumar N; and Mrs. Karthiga G on August 05, 2026, for A Computer-Implemented System And Method For Early Cancer Detection And Multi-Class Classification Using A Hybrid Cnn–bilstm Network.
Inventors include Dr. Tamilarasan. S; Mr. Vinay G; Dr. Avinash N; Mr. Sandeep K H; Mr. Chethan P J; Dr. Subrahmanya H M; Mr. Shravan H S; Mr. Lohitesh Kumar N; and Mrs. Karthiga G.
The application for the patent was published on August 14, 2026, under issue no. 33/2026.
Abstract: ABSTRACT OF THE INVENTION The present invention relates to a computer-implemented system and method for early cancer detection and multi-class cancer classification using a hybrid Convolutional Neural Network (CNN) and Bidirectional Long Short-Term Memory (BiLSTM) network. The system receives medical images from one or more imaging modalities, preprocesses the images through resizing, normalization, and noise removal, and extracts discriminative features using the CNN. The extracted features are further processed by the BiLSTM network to learn contextual relationships and improve classification performance. The refined features are classified into one of multiple cancer categories or a normal class using a classification module. The system generates a predicted cancer class, confidence score, and diagnostic report to assist healthcare professionals in clinical decision-making. The proposed invention provides improved diagnostic accuracy, reduced processing time, and automated multi-class cancer classification, making it suitable for deployment in hospitals, diagnostic laboratories, telemedicine platforms, and other healthcare environments.
Disclaimer: Curated by HT Syndication.
