Electromyography (EMG) is a technique that measures the electrical activity of the muscles and it has been used extensively in the field of physiotherapy to assess the muscle function and activity. Grading muscle power is an important aspect of assessing muscle function, as it provides information about the strength and endurance of muscles. Presently, the physiotherapist uses Manual Muscle Testing (MMT) for grading muscle power however it requires the therapist with good expertise. In this work, an Internet of Medical Things (IoMT) based Smart EMG device is designed and developed for monitoring the patients suffering from abnormal musculoskeletal health conditions. Further, the EMG signals are acquired from normal individuals and the patients with abnormal health conditions. Also, the muscle power grading is used to grade the EMG signals and the Convolutional Neural Network (CNN) based deep learning algorithm is utilised to visualize the progress of course of treatment provided to the patients with musculoskeletal problems such as stroke, spinal cord injuries etc. The entire analysis is carried out Google Co-Laboratory based IoT cloud platform and the algorithms are coded using Python programming language. Results demonstrate that the proposed smart IoMT based smart device can predict the different muscle power with an average accuracy of 97.5 % which proves the effectiveness of the device. This work appears to be of high clinical relevance since the proposed device is capable of providing valuable information about muscle function and enable the physiotherapists to design personalised treatment plans for patients with musculoskeletal disorders.