MUMBAI, India, Jan. 23 -- Intellectual Property India has published a patent application (202541116026 A) filed by G. Puthillbai; J. A. Adlin Layola; V. Gokul & V. Devatarika; N. Vinitha; G. Vairamanikandan; S. Padmavathy; and S. Vasudhevan, Chennai, Tamil Nadu, on Nov. 24, 2025, for 'deep learning framework for early diagnosis of cancer using multimodal medical imaging.'
Inventor(s) include G. Puthillbai; J. A. Adlin Layola; V. Gokul & V. Devatarika; N. Vinitha; G. Vairamanikandan; S. Padmavathy; and S. Vasudhevan.
The application for the patent was published on Jan. 23, under issue no. 04/2026.
According to the abstract released by the Intellectual Property India: "The invention provides a deep learning-based framework for early diagnosis of cancer using multimodal medical imaging, including MRI, CT, PET, ultrasound, and histopathology data. The system comprises a pre-processing module for noise reduction, normalization, and multimodal registration, followed by a hybrid architecture integrating convolutional neural networks for spatial feature extraction and transformer encoders for global contextual modeling. A multimodal fusion module correlates structural and'metabolic features, enabling accurate detection and segmentation of tumors through attention-guided U-Net or Mask R-CNN models. A classification module predicts tumor type, stage, and malignancy probability, while a visualization interface generates heatmaps, segmentation masks, and attention maps for clinical interpretation.. The invention improves early cancer detection accuracy, enhances explainability, and supports deployment across cloud, edge, and hospital environments."
Disclaimer: Curated by HT Syndication.
