MUMBAI, India, May 1 -- Intellectual Property India has published a patent application (202641026333 A) filed by Dr. T. Vigneswari, Kovilvenni, Tamil Nadu, on March 6, for '3d neural structure analysis with pointnet++ using point cloud for alzheimer's diagnosis.'
Inventor(s) include Dr. T. Vigneswari; J. Harini; K. Iswarya; K. Kaviya; and S. A. Lakshaya.
The application for the patent was published on May 1, under issue no. 18/2026.
According to the abstract released by the Intellectual Property India: "The present invention relates to a system and computer-implemented method for early detection of Alzheimer's disease using three-dimensional analysis of magnetic resonance imaging (MRI) brain data. A plurality of two-dimensional MRI slices are acquired and subjected to preprocessing operations including noise reduction, normalization, and segmentation to eflhance structural clarity. The processed slices are spatially aligned and reconstructed into a threedimensional point cloud by mapping voxel intensity values into spatial coordinates, thereby preserving anatomical continuity and depth information.The reconstructed point cloud is analysed using a deep learning model based on a PointNet++ architecture configured to extract hierarchical spatial features representing structural variations in brain tissues. The model generates classification outputs and confidence scores indicative of neurodegenerative patterns associated with Alzheimer's disease. An interactive visualization module renders the reconstructed three-dimensional brain structure and presents analytical results to the user.By enabling volumetric analysis of MRl data instead of conventional slice-wise interpretation, the invention improves diagnostic accuracy and supports early clinical decision-making for neurological disorder detection."
Disclaimer: Curated by HT Syndication.
