MUMBAI, India, Oct. 5 -- Intellectual Property India has published a patent application (202641115763 A) filed by Srm Institute Of Science And Technology, Ramapuram Campus; and Easwari Engineering College on September 28, 2026, for Non-Invasive Diabetes Prediction Device Using Adaptive Differential Stimulus-Response Analysis.
Inventors include Nishika Sorokhaibam; and Dr. C. Saravanan.
The application for the patent was published on October 02, 2026, under issue no. 40/2026.
Abstract: Abstract: The present invention relates to an EEG-based prognostic prediction system for moderate Traumatic Brain Injury (moTBI) using a hybrid deep learning and boosting framework. The proposed system, termed the NeuroFusion Adaptive Temporal Framework, receives multi-channel EEG signals acquired from patients with moderate Traumatic Brain Injury and performs preprocessing, temporal segmentation, and representation learning. The framework integrates Dual-Channel Residual Bidirectional Long Short-Term Memory (BiLSTM) layers with a Self-Attention Temporal Encoder to capture bidirectional and salient temporal characteristics of EEG signals. The learned representations are subsequently processed using a Gradient-Weighted Adaptive Boosting Meta-Classifier for prognostic prediction. To address class imbalance, the system employs Stratified Bootstrap with Synthetic Minority Oversampling Technique (SMOTE). Adaptive optimization using Adam with cosine annealing scheduling is employed for stable model convergence. The system generates an automated prognostic prediction and associated probability score, thereby providing rapid and objective decision support for neurocritical care. The proposed framework is intended to improve temporal representation, reduce prediction variance, enhance robustness, and support early clinical intervention and rehabilitation planning.
Disclaimer: Curated by HT Syndication.
