MUMBAI, India, Aug. 12 -- Intellectual Property India has published a patent application (202611069757 A) filed by Dr Preeti; Dr. Charu Sharma; Dr Sonia Saroha; Dr. Geeta Gill; Dr. Ram Mehar; Dr. Kapil; Vikrant; and Yashasvi Kaushik on June 03, 2026, for A System And Method For Adaptive Low Back Pain Evaluation And Therapeutic Exercise Optimization Using Machine Learning Techniques.
Inventors include Dr Preeti; Dr. Charu Sharma; Dr Sonia Saroha; Dr. Geeta Gill; Dr. Ram Mehar; Dr. Kapil; Vikrant; and Yashasvi Kaushik.
The application for the patent was published on August 07, 2026, under issue no. 32/2026.
Abstract: The present invention discloses a system and method for adaptive evaluation of low back pain (LBP) and optimization of individualized therapeutic exercise protocols using machine learning techniques. The system (100) comprises a multi-modal data acquisition module (110) aggregating patient demographics, clinical assessment scores (VAS, ODI, ROM), wearable sensor streams (sEMG, IMU), medical imaging metadata, and patient-reported outcomes; a data pre-processing and feature engineering module (120); a machine learning pain classification engine (130) employing ensemble models and convolutional neural networks to determine LBP severity, aetiology, and chronicity; a contraindication rules engine (150) grounded in evidence-based clinical guidelines; a reinforcement learning-based adaptive exercise optimization module (140) employing Proximal Policy Optimization; a continual learning module (160) implementing Elastic Weight Consolidation and ADWIN drift detection; and an explainability module (170) generating SHAP-based feature attributions and natural language rationale. Outputs comprise a personalized pain report, an adaptive exercise protocol, a clinician dashboard, and a FHIR-compliant patient mobile application. The invention addresses the long-felt need for an integrated, safe, personalized, and interpretable LBP management platform.
Disclaimer: Curated by HT Syndication.
