MUMBAI, India, Sept. 22 -- Intellectual Property India has published a patent application (202641109777 A) filed by Rathla Roop Singh; and Dr. D. Vasumathi on September 12, 2026, for A Computer-Implemented Radiomic-Guided Adaptive Feature Modulation System And Method For Automated Bone Tumor Grading Using A Compact Deep Learning Architecture.
Inventors include Rathla Roop Singh; and Dr. D. Vasumathi.
The application for the patent was published on September 18, 2026, under issue no. 38/2026.
Abstract: The present invention relates to a computer-implemented system and method for automated grading of bone tumors from digital medical images using adaptive integration of quantitative radiomic analysis and deep learning. The system comprises one or more processors configured to preprocess a digital medical image, identify a tumor region, extract quantitative radiomic descriptors, generate deep feature representations, derive adaptive modulation parameters from the extracted quantitative radiomic descriptors, apply the adaptive modulation parameters to the generated deep feature representations to obtain an adaptively modulated feature representation, and classify the adaptively modulated feature representation to generate an automated tumor grading output, wherein the quantitative radiomic descriptors are utilized to regulate the learned deep feature representations prior to classification, thereby establishing computational interaction between handcrafted radiomic descriptors and learned deep neural representations. The invention is applicable to computer-implemented medical image analysis and artificial intelligence-based clinical decision- support systems.
Disclaimer: Curated by HT Syndication.
