MUMBAI, India, Aug. 24 -- Intellectual Property India has published a patent application (202641094607 A) filed by Dayananda Sagar University on August 05, 2026, for Automated Diagnostic Platform For Differentiating Mechanical And Neurological Back Pain.
Inventors include Sushma D. S; Mamatha R M; Vidya Shree R; Shravankumar Dhangi; Soram Wanglen; Madhuri H D; Nayana H J; Ayush Yadav; Divyanshu Verma; Amartya Das; and Dr. Santosh Kumar J.
The application for the patent was published on August 21, 2026, under issue no. 34/2026.
Abstract: The invention relates to an automated diagnostic platform for differentiating mechanical and neurological back pain using a multi-modal machine learning framework to improve diagnostic accuracy, reduce clinical assessment time, and support evidence-based decision-making. The method includes acquiring patient-specific data comprising clinical history records, radiological imaging, electromyography signals, and wearable sensor-based movement data, and preprocessing the acquired data into structured, modality-specific feature representations. The framework fuses the feature representations into a unified diagnostic representation using trainable weighting coefficients that adaptively emphasize the most diagnostically relevant modality for a given patient. The platform further incorporates a trained classification model that differentiates between mechanical and neurological origins of back pain, flags ambiguous or overlapping presentations for clinician review, and supports severity grading of identified symptoms. Model optimization techniques, including weighted cross-entropy training, stratified validation, quantization, and parameter pruning, are employed to improve diagnostic reliability while enabling deployment on portable point-of-care devices with reduced computational resource requirements. The invention provides improved diagnostic accuracy, interpretability, and deployment flexibility for clinical decision support in hospital, clinic, and telehealth environments. The proposed method is applicable to orthopedic diagnostics, physiotherapy assessment, remote patient monitoring, and resource-constrained healthcare settings requiring rapid, objective differentiation between mechanical and neurological back pain.
Disclaimer: Curated by HT Syndication.
