Intellectual Property India Publishes Patent Application for Self-Supervised Learning System For Automated Breast Cancer Diagnosis Using Whole Slide Images

MUMBAI, India, July 30 -- Intellectual Property India has published a patent application (202641088242 A) filed by Sr University on July 20, 2026, for Self-Supervised Learning System For Automated Breast Cancer Diagnosis Using Whole Slide Images.

Inventors include Sangishetti Karnakar; and Dr. P Praveen.

The application for the patent was published on July 24, 2026, under issue no. 30/2026.

Abstract: SELF-SUPERVISED LEARNING SYSTEM FOR AUTOMATED BREAST CANCER DIAGNOSIS USING WHOLE SLIDE IMAGES The invention discloses a self-supervised learning system (100) for automated breast cancer diagnosis using whole slide images (WSIs). The system includes a Self-Supervised Feature Extraction Core (200) configured to perform contrastive learning on unlabelled WSIs to generate a feature embedding map (210), and a Modified U-Net Segmentation Module (300) enhanced with attention mechanisms (310) and residual connections (320) to achieve precise tumour segmentation. A Training Optimization Engine (400) reduces training time by leveraging unlabelled data, while a Performance Evaluation Unit (500) computes diagnostic metrics including Dice coefficient (510) and Intersection over Union (IoU) (520).

Disclaimer: Curated by HT Syndication.