Human pose estimation for physiotherapy following a car accident using depth-wise separable convolutional neural networks

We all know that car accidents can result in a variety of obvious injuries, but sometimes the effects are not immediately apparent. They appear gradually, and some of the consequences can be avoided or at least mitigated by paying close attention early on. Seeking physical therapy as soon as possible after a car accident can help reduce pain, avoid surgery, and reduce long-term damage. Physical therapy aids in the reduction of pain symptoms, the improvement of flexibility and strength, and the acceleration of recovery. It can also help prevent the long-term effects of a car accident, such as migraines and chronic pain. The sooner one begins physical therapy after a car accident, the more likely (s)he is to avoid future pain and problems. In this paper, we show an interactive computer vision-based application that, with the help of pose estimation, can help any user do physiotherapy exercises at home. The patient can do the exercises anywhere, anytime, and at his or her own pace, without any help from a person. When it comes to dynamic human pose estimation, the process known as "identifying human joints in an image or video and determining their position in space" is used. This is done so that the dynamic position of the human body can be more accurately estimated and evaluated. This goal can be achieved by applying various computer vision strategies used in a number of industries, such as gaming, robotics training, and animation. In this article, we propose a method for dynamic human pose estimation using convolutional neural networks (CNN). This method will soon be used as a form of physical therapy rehabilitation that can be performed in a remote setting. By making an assessment of the patient's posture, such as a victim of a car accident, the physical therapist can determine whether the patient is performing the assigned exercises correctly. With this method, the physiotherapist can make sure that the therapy sessions match how well the patient is getting better. Furthermore, we aim to bring physical therapy to the metaverse by developing a telehealth platform that addresses the shortcomings of traditional videoconferencing-style platforms encountered by therapists in recent years. Physical therapists are currently unable to obtain quantifiable metrics such as range of motion, joint torques, or balance assessment using current 2D platforms. Our goal is to help therapists by providing these metrics in real time, allowing patients to be properly evaluated via remote care. By all means, remote care improves patient accessibility, reduces commuting time, and can help increase the number of touchpoints between therapist and patient.