MUMBAI, India, Jan. 23 -- Intellectual Property India has published a patent application (202641001511 A) filed by Sr University, Warangal, Telangana, on Jan. 6, for 'a robust deep learning system and method for environmental health impact prediction using fourier feature expansion, gated linear units, and squeeze-and-excitation based residual learning architecture.'
Inventor(s) include Bobbala Rajesh Reddy; and R. Vijaya Prakash.
The application for the patent was published on Jan. 23, under issue no. 04/2026.
According to the abstract released by the Intellectual Property India: "The present invention discloses a robust deep learning system and method for predicting health impact outcomes from environmental air quality data. The invention employs a novel neural architecture that integrates Fourier feature expansion to capture non-linear and periodic relationships, gated linear units to selectively regulate feature interactions, and squeeze-and-excitation mechanisms to adaptively recalibrate feature importance. Residual dense learning, together with normalization and regularization techniques, ensures stable training and enhanced generalization. The system processes multidimensional environmental parameters and generates a continuous health impact score with improved accuracy and robustness compared to conventional models. The disclosed invention is suitable for environmental monitoring, public health risk assessment, and decision-support applications requiring reliable prediction of health impacts arising from environmental conditions."
Disclaimer: Curated by HT Syndication.
