MUMBAI, India, July 7 -- Intellectual Property India has published a patent application (202641078559 A) filed by Sr University on June 25, 2026, for Deep Learning–based Analysis Of Melanoma Skin Cancer From Dermoscopic Images.
Inventors include Dasari Chaithanya; and Suresh Kumar Mandala.
The application for the patent was published on July 03, 2026, under issue no. 27/2026.
Abstract: DEEP LEARNING–BASED ANALYSIS OF MELANOMA SKIN CANCER FROM DERMOSCOPIC IMAGES A deep learning– based analysis system for melanoma skin cancer from dermoscopic images is disclosed. The system receives dermoscopic images and processes them through a structured multi-stage pipeline. The images undergo pre-processing including noise removal, illumination correction, color normalization, hair and artifact elimination, and resolution standardization to reduce image variability and improve lesion visibility. A deep neural network architecture is employed to perform automated lesion analysis, feature extraction, and melanoma classification. The network learns clinically relevant visual characteristics including asymmetry, border irregularity, color heterogeneity, and texture variations associated with melanoma. The system may further perform lesion segmentation to focus analysis on the lesion region and generate prediction outputs with confidence information to assist clinical interpretation. The invention enables automated, consistent, and accurate analysis of dermoscopic images, thereby supporting early melanoma detection, reducing manual interpretation, and improving clinical screening and diagnostic decision-making.
Disclaimer: Curated by HT Syndication.
