MUMBAI, India, Sept. 28 -- Intellectual Property India has published a patent application (202541026961 A) filed by Bhoopesh S J; Akilan S; Ananda Kumar A; Saswin V; and Rajasekar S S on March 24, 2025, for Colon Cancer Detection Using Mask R-Cnn.
Inventors include Bhoopesh S J; Akilan S; Ananda Kumar A; Saswin V; and Rajasekar S S.
The application for the patent was published on September 25, 2026, under issue no. 39/2026.
Abstract: Early detection of colon polyps is critical for preventing the progression of colorectal cancer. This study proposes a two-stage deep learning approach for accurate colon cancer detection. In the first stage, U-Net, a widely used convolutional neural network (CNN) for medical image segmentation, is employed to precisely identify and segment colon polyps from colonoscopy images. • The segmented regions are then passed to the second stage, where Mask R-CNN, a powerful instance segmentation model, is utilized to classify the segmented polyps and determine whether they are cancerous or benign. The integration ofU-Net for initial polyp detection and Mask R- CNN for classification enhances the overall accuracy and efficiency of the system. • Experimental results demonstrate that this hybrid approach improves the precision of colon cancer detection, enabling timely and effective diagnosis, which is crucial for reducing mortality rates associated with colorcctal cancer.
Disclaimer: Curated by HT Syndication.
