MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202621042290 A) filed by Prof. Dhiraj Jadhav; Kabir Khake; Jay Paunikar; Vaidehi Jalwadi; and Sahil Kakad on April 02, 2026, for A Deep Learning-Based Medical Imaging System For Reconstruction Of Damaged Mri Slices.
Inventors include Prof. Dhiraj Jadhav; Kabir Khake; Jay Paunikar; Vaidehi Jalwadi; and Sahil Kakad.
The application for the patent was published on June 12, 2026, under issue no. 24/2026.
Abstract: This invention presents a computer-implemented system for reconstructing damaged or missing slices in three-dimensional T 1 -weighted MRI volumes. The framework integrates preprocessing techniques such as background cropping and intensity normalization with a volumetric deep learning model to restore missing anatomical information. By learning spatial continuity across adjacent slices, the system generates accurate and structurally consistent reconstructions. Experimental results demonstrate approximately 95% reconstruction accuracy, significantly reducing the need for rescanning procedures. The proposed approach enhances diagnostic reliability and provides an efficient solution for medical imaging restoration.
Disclaimer: Curated by HT Syndication.
