MUMBAI, India, July 13 -- Intellectual Property India has published a patent application (202641082199 A) filed by Koneru Lakshmaiah Education Foundation on July 03, 2026, for Continual Personality-Aware Causal Graph Transformer Framework For Multimodal Physiological Emotion Recognition.
Inventors include T. L. Deepika Roy; and Dr. Nulaka Srinivasu.
The application for the patent was published on July 10, 2026, under issue no. 28/2026.
Abstract: Emotion recognition using physiological and behavioral signals has emerged as a critical research area in affective computing, intelligent healthcare, human-computer interaction, adaptive learning systems, and mental health monitoring. Existing multimodal emotion recognition systems often suffer from limited interpretability, inability to model causal relationships among physiological modalities, poor adaptation to individual personality differences, and catastrophic forgetting when exposed to new subjects or sessions.This invention proposes a novel Continual Personality-Aware Causal Graph Transformer Framework that integrates dynamic causal graph learning, Graph Transformer networks, Neural Ordinary Differential Equation (Neural-ODE) memory evolution, personality-aware attention fusion, and counterfactual adaptation mechanisms for lifelong multimodal emotion recognition. The framework utilizes synchronized physiological and behavioral modalities including Electroencephalography (EEG), Electrodermal Activity (EDA), facial activity, eye gaze, pupil dilation, cursor dynamics, and personality traits.The proposed architecture dynamically learns directed causal interactions among modalities, models continuous emotional evolution through Neural-ODE temporal memory, adapts emotional representations according to individual personality characteristics, and continuously acquires new emotional knowledge without catastrophic forgetting. Experimental evaluations demonstrate superior recognition accuracy, robustness to missing modalities, enhanced explainability, and improved cross-subject generalization.The invention provides an interpretable, scalable, and adaptive framework suitable for next-generation intelligent affective systems.
Disclaimer: Curated by HT Syndication.
