Abstract
Yoga improves flexibility and strength in the body and mind. This technique may reduce tension and discomfort. Medical or non-medical yoga and physiotherapy may alter the musculoskeletal, visceral, neurological, immunological, and endocrine systems. By merging early well-being practices with modern technology, clinicians may now study the biochemical effects of physiotherapy and yoga and develop more effective retrieval tactics. Deep learning systems, designed to examine and interpret vast amounts of composite data, have helped healthcare professionals identify complicated biochemical reactions that affect recovery. Thus, this research provides an EYPLM-BPERW approach to evaluate yoga and physiotherapy. Deep learning is used to examine the biochemical effects of therapeutic workouts on patient recovery in the EYPLM-BPERW model. The process includes feature extraction, classification, and parameter optimization. EYPLM-BPERW is used with VGG16 model for feature extraction at the primary stage to gather essential spatial characteristics from imaging and motion analysis of physiotherapy and yoga postures. Classification uses the BiLSTM classifier. Finally, the Adam optimizer method optimises the BiLSTM model parameters to improve recovery prediction accuracy. The EYPLM-BPERW model is extensively tested on the benchmark database to prove its classification accuracy. The wide comparing findings showed that EYPLM-BPERW outperformed current techniques.
Keywords: Adam Optimizer, Bidirectional Long Short-Term Memory, Biochemical Perspective, Feature Extraction, Physiotherapy, Yoga.