MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641094838 A) filed by Dhanalakshmi Srinivasan College Of Engineering And Technology on August 05, 2026, for Oncosight-Ai Powered Web Based Cancer Detection System.
Inventors include K. Sibi Mathavan; Dr. Samuel Chellathurai. A; and Yogaraj. M.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: ABSTRACT OF THE INVENTION ONCOSIGIIT - AI POWERED WEB BASED CANCER DETECTION SYSTEM The present invention relates to an Artificial Intelligence (AI)-powered web-based cancer detection and classification platform designed to assist healthcare professionals in the automated analysis of medical images for cancer detection. The invention integrates deep learning techniques with a secure web-based architecture to provide an end-to-end workflow for patient management, medical image processing, AI-assisted prediction, explainable visualization, and automated report generation.The proposed platform comprises a user authentication module, patient management module, medical image upload module, image preprocessing module, intelligent model selection module, AI-based cancer detection module, explainable Al module, diagnostic report generation module, and database management module. Authorized users can securely access the platform, register patient information, upload medical images, and receive AI-assisted prediction results through a unified web interface.In the current implementation, the platform is configured for the detection and classification of two types of lung cancer using Computed Tomography (CT) scan images and histopathological images. CT scan images are analyzed using a weighted ensemble deep learning framework that combines multiple convolutional neural network (CNN) models to improve prediction reliability. Histopathological images are processed using a dedicated deep learning classification model. The platform automatically identifies the uploaded image modality and routes it to the corresponding Al model.To improve transparency, the invention incorporates an Explainable Artificial Intelligence (XAI) module based on Gradient-weighted Class Activation Mapping (Grad-CAM), which generates visual heatmaps highlighting the image regions contributing to the Al prediction. Prediction results, confidence scores, visual explanations, and patient information are automatically compiled into a structured report that can be securely stored and managed through the web-based platform.The architecture of the proposed system is modular and scalable, enabling future integration of additional cancer-specific Al models and medical imaging modalities without substantial modification of the core platform. While the current implementation demonstrates functionality using two types of lung cancer, the platform is designed for future expansion as additional validated Al models become available.The invention provides a unified AI-assisted platform combining secure web-based access, intelligent image routing, deep learning-based cancer detection and classification, explainable Al, centralized patient record management, and automated report generation, thereby offering an efficient and scalable solution for computer-assisted cancer analysis.
Disclaimer: Curated by HT Syndication.
