MUMBAI, India, May 29 -- Intellectual Property India has published a patent application (202641062220 A) filed by Ms. Puppala Sireesha; Mr. Mallela Sathya Prajith Reddy; and Mr. Avaneesh Thodupunooru, Hyderabad, Telangana, on May 16, for 'a system and method for configurable neuroanatomical analysis using a tri-directional state-space architecture.'
Inventor(s) include Ms. Puppala Sireesha; Mr. Mallela Sathya Prajith Reddy; and Mr. Avaneesh Thodupunooru.
The application for the patent was published on May 29, under issue no. 22/2026.
According to the abstract released by the Intellectual Property India: "This invention is a configurable, hardware-efficient system for analysing three dimensional structural Magnetic Resonance Imaging (MRI) brain scans. The system identifies specific brain regions, measures their volumes, and compares those measurements with an age- and gender-matched reference database. The process begins with an age-sensitive preprocessing pipeline that removes the skull, corrects scanner-related intensity distortions using N4 Bias Field Correction (N4ITK), normalises image intensities using a White Matter-anchored Gaussian Mixture Model (GMM), and aligns the scan to a standard coordinate space. A Tri-Directional 3D Mamba deep learning network then extracts spatial features using a tri-planar projection method that reduces GPU memory usage from quadratic to linear complexity. The system next localises brain regions without atlas-based population maps, applies dedicated segmentation models to measure regional volumes, and computes a Neuro-Structural Deviation Index (NSDI) with supporting correlation reports. The framework is configurable for different neurodevelopmental conditions."
Disclaimer: Curated by HT Syndication.
