MUMBAI, India, Aug. 31 -- Intellectual Property India has published a patent application (202641099897 A) filed by Keshav Memorial Institute Of Technology on August 19, 2026, for Oral Cancer Detection System Using Deep Learning And Mobilenet Architecture.
Inventors include Mr. E Srinath; Mr. Para Upendar; Dr. Tikkireddi Adithya Kumar; Mr. Rakesh Reddy Gurrala; Ms. Kasagani Lohitha; Ms. G Gayatri Lahari; Ms. Kumkum Gehlot; and Ms. Manne Udayasri.
The application for the patent was published on August 28, 2026, under issue no. 35/2026.
Abstract: Oral cancer, particularly Oral Squamous Cell Carcinoma (OSCC), remains a major global health challenge due to its high mortality rate and the prevalence of late-stage diagnosis. Early detection significantly improves patient outcomes, yet traditional screening methods are often limited by subjectivity and accessibility. With recent progress in medical imaging and artificial intelligence, deep learning–based visual analysis has become a promising solution for supporting early and reliable cancer screening. This project presents a deep learning framework for oral cancer prediction using clinically captured oral cavity images. The workflow involves image preprocessing, augmentation, and feature extraction, followed by classification using a lightweight Convolutional Neural Network (CNN) architecture, MobileNet. The model distinguishes between benign and malignant oral lesions, aiming for high accuracy while maintaining computational efficiency suitable for real-time deployment. Additionally, prediction confidence scores and interpretable outputs enhance transparency and support clinical decision-making. Future extensions of this work may include the integration of multimodal medical data, such as histopathology images, patient clinical history, or fluorescence imaging, to further strengthen diagnostic precision. Incorporating explainable AI (XAI) techniques could provide deeper insights into lesion characteristics and model reasoning. Ultimately, this system has the potential to support early-stage oral cancer identification and assist healthcare professionals in improving patient outcomes through faster and more accessible diagnostic support.
Disclaimer: Curated by HT Syndication.
