MUMBAI, India, July 30 -- Intellectual Property India has published a patent application (202641087886 A) filed by Vellore Institute Of Technology on July 17, 2026, for Automated Melanoma Detection System Using Ridge Vessel Enhanced Neural Network Architecture.
Inventors include Sripreethaa; and Yuvaraj.
The application for the patent was published on July 24, 2026, under issue no. 30/2026.
Abstract: ABSTRACT AUTOMATED MELANOMA DETECTION SYSTEM USING RIDGE VESSEL ENHANCED NEURAL NETWORK ARCHITECTURE A computer- implemented method (100) for automated melanoma detection comprises acquiring a skin lesion image (102), suppressing artifacts in the skin lesion image to produce a depilated lesion image (106), and applying Continuous Threshold Ridge Vessel enhancement (108) to the depilated lesion image to produce a CTRV-enhanced lesion image (316). The method (100) processes the CTRV-enhanced lesion image through a Normalized Residual Depthwise Pointwise U-Net (110) comprising an encoder path and a decoder path, wherein the encoder path and the decoder path comprise Normalized Residual Depth Point modules (404, 414). The method (100) generates a dual output (418) comprising a classification result (420) indicating whether the skin lesion image corresponds to melanoma or non-melanoma and a lesion segmentation mask (422) identifying lesion pixels. (Fig. 1)
Disclaimer: Curated by HT Syndication.
