MUMBAI, India, June 30 -- Intellectual Property India has published a patent application (202641074840 A) filed by Mlr Institute Of Technology on June 17, 2026, for Frequency-Aware Vision Transformer Framework For Automated Glaucoma Detection Using Empirical Wavelet Transform-Based Multi-Sub Band Analysis.
Inventors include Dr. Vijaya Kumar Velpula; Dr. Kiran Chand Ravi; Ms. Varshini; and Mr. P. Rahul.
The application for the patent was published on June 26, 2026, under issue no. 26/2026.
Abstract: This invention presents a Frequency Aware Vision Transformer model for automatic detection of glaucoma from retinal fundus images. The invention combines the frequency domain representation obtained through Empirical Wavelet Transform with a Vision Transformer architecture based on deep learning. This combination makes it possible to exploit both spatial and frequency domain features during training. Initially, the input image is divided into various sub bands corresponding to different frequency bands. Next, each subband is transformed into a patch embedding format and analyzed with a Vision Transformer with cross-frequency attention. This innovation can detect structural variations in the optic nerve head effectively.
Disclaimer: Curated by HT Syndication.
