Neural Network-Based Exercise Training and Limb Function Evaluation System for Traditional Chinese Medicine Guiding Technique

Abstract

Exercise training plays a pivotal role in enhancing limb function and overall physical performance. Through targeted and progressive exercise regimes, individuals can improve strength, flexibility, coordination, and endurance in their limbs. This paper presents a novel Neural Network-Based Exercise Training and Limb Function Evaluation System tailored for Traditional Chinese Medicine (TCM) guiding techniques. This paper constructed a novel Multi-Layer Fuzzy Pattern Neural Network (MLFPNN) for the estimation of limbs for exercise training. The proposed MLFPNN model acquires information about the limb muscles through the acquired information features are normalized. With the normalized features, TCM is evaluated for the computation of the feature for the exercise training in MLFPNN. The proposed model uses the multilayer fuzzy for the estimation of the limb features associated with the limb function. The estimated features of the limb are applied over the pattern network for the classification of limb function based on TCM with MLFPNN. The proposed MLFPNN model evaluates the 10 features in the limb muscle estimation for TCM-based exercise training. Experimental analysis is conducted for the proposed MLFPNN to achieve a higher prediction based on the actual values. The comparative analysis demonstrated that the proposed MLFPNN model achieves an accuracy of 92.5% while conventional SVM, RF, and k-NN achieve a classification accuracy of 88.3%, 90.7%, and 87.6% respectively. The findings stated that the proposed MLFPNN model is significant for the limb function estimation for the TCM-based training.