MUMBAI, India, May 29 -- Intellectual Property India has published a patent application (202641062165 A) filed by Sjm Institute Of Technology; Prof. Tanuja T; and Dr. Chetan. S, Chitradurga, Karnataka, on May 15, for 'an intelligent ai-based dermoscopic image classification system for skin cancer detection.'
Inventor(s) include Dr. Chetan. S; Dr. Ravi L S; Dr. Naveen K B; Dr. Vishwanath B R; Dr. M Somashekhar; Dr. Sukumar B S; Mr. Ramesha K P; Mr. G H Yogeesh; Dr. Radhakrishna K R; Mr. Shubas S R; and Mr. Hemanth Kumar D T.
The application for the patent was published on May 29, under issue no. 22/2026.
According to the abstract released by the Intellectual Property India: "Skin cancer is among the most life-threatening diseases worldwide, where early and accurate diagnosis is critical for effective treatment and improved patient survival. Traditional diagnostic approaches are often dependent on clinical expertise and manual examination, leading to delayed or inaccurate predictions. The present invention proposes an ECRNet-based hybrid deep learning system for automated skin cancer identification using dermoscopic images. The proposed architecture combines convolutional neural networks, residual learning blocks, and multi-scale feature extraction mechanisms to improve lesion classification accuracy and robustness. The system utilizes a pretrained ResNet18 backbone integrated with enhanced residual learning and attention-based refinement modules for efficient feature extraction. Image preprocessing techniques such as normalization, resizing, augmentation, and noise reduction are incorporated to improve model generalization. The proposed invention enables reliable classification of skin lesions into multiple diagnostic categories with high accuracy and reduced computational complexity, making it suitable for real-time clinical diagnosis, telemedicine platforms, and computer-aided healthcare applications."
Disclaimer: Curated by HT Syndication.
