MUMBAI, India, Oct. 12 -- Intellectual Property India has published a patent application (202641118507 A) filed by Dr R Gomalavalli; K. Neela; Sinthu Murugan; Karthik Nainar; Ganga Murugan; Mrs. P. Rajyalakshmi; and Mrs. P. N. Vijayasree on October 03, 2026, for A Field Programmable Gate Array Based Accelerator For Streaming Measurement Of Lung Lesions With Uncertainty-Triggered Partial Reconfiguration, And Method Of Operating The Same.
Inventors include Dr R Gomalavalli; K. Neela; Sinthu Murugan; Karthik Nainar; Ganga Murugan; Mrs. P. Rajyalakshmi; Mrs. P. N. Vijayasree; and Dr. S. Kannappan.
The application for the patent was published on October 09, 2026, under issue no. 41/2026.
Abstract: An FPGA-based accelerator (100) measures lung lesions in CT image data in a single streaming pass. A single-pass connected-component labelling unit (118) with a line buffer (116), merge table (120) and feature accumulator memory (122) accumulates, for each lesion, voxel counts, partial-volume fractions, boundary gradients and coordinate projections on fixed directions, without storing image frames. A measurement and uncertainty unit (124) computes the lesion volume, greatest dimension and its standard uncertainty from voxel spacing, slice thickness, boundary contrast and noise. A parallel comparator bank (126) generates a trigger when the uncertainty interval contains a programmable size threshold. Only for triggered lesions does a reconfiguration controller (130) load a fine-analysis module (136) into a reconfigurable partition (134) by dynamic partial reconfiguration, while coarse measurement continues. Fine-analysis hardware is thus used only where the size category is ambiguous, reducing area and energy.
Disclaimer: Curated by HT Syndication.