BACK PAIN CLASSIFICATION BASED ON ULTRA-SOUND MUSCLE IMAGE USING MACHINE LEARNING

Abstract

Due to COVID-19 pandemic people who suffer from back pain try avoiding physical contact with physiotherapists. Ultrasound imaging plays very significant role in the clinical diagnosis of central neuraxial blockade which often lead to induce pain in thoracic vertebrae as well as promoting disorders in the normal functions trapezius muscle. Ultrasound image classification could be useful for the identification and estimation of intervertebral levels, important physiological locations such as inter-laminar and middle line spaces, depth between intrathecal and epidural spaces. The main purpose of this research is to propose a deep learning based approach for the classification of ultrasound images for the normal and abnormal trapezius muscle activity for thoracic vertebrae along with a brief review of Artificial Intelligence (AI) based automatic image segmentation. In this research, we propose a method to identify muscle condition based on ultra-sound image. We employ machine learning to classify the severity of the pain and help to descript the medication. The proposed method will help people during the COVID-19 duration to be more safe and secure. Also, with such technology, the therapeutic and diagnostic will be improved in future. In conclusion, the preliminary findings of this study suggest that based on ultrasound images of thoracic vertebrae, the normal and abnormal activity of trapezius muscle can be autonomously classified.