Experimental and numerical diagnosis of fatigue foot using convolutional neural network

Fatigue is an essential criterion for physiotherapy in injured athletes. Muscle fatigue mechanism also is a crucial matter in designing a workout program. It is mainly related to physical injury, cerebrovascular accident, spinal cord injury, and rheumatologic disease. The leg is one of the organs in the body where fatigue is visible, and usually, the first fatigue traces in the human body are shown. The main objective of the article is to diagnosis tired and untired feet base on digital footprint images. Therefore, the foot images of students in the age group of 20-30 were examined. The device is a digital footprint scanner. This device includes a plate screen equipped with pressure sensors and footprints in the image. A treadmill is used for 8 min to tire our test individuals. Therefore, six methods of k-nearest-neighbor classifier, multilayer perceptron, support vector machine, naïve Bayesian learning, decision tree, and convolutional neural network (CNN) architecture are presented to achieve the goal. First, the images are grayscale and divide into four regions, and the mean and variance of pressure in each of the four areas are extracted as features. Finally, the classification is accomplished using machine learning methods. Then, the results are compared with a proposed CNN architecture. The presented CNN method is outperforming other approaches and can be used for future fatigue diagnosis systems.