Application of least square method for muscular strength estimation in hand motion recognition using surface EMG

The Development of a System for Elbow Joint Range of Motion Measurement Based on Image Recognition and Myoelectric Signals

After a fracture, patients have reduced willingness to bend and extend their elbow joint due to pain, resulting in muscle atrophy, contracture, and stiffness around the elbow. Moreover, this may lead to progressive atrophy of the muscles around the elbow, resulting in permanent functional loss. Currently, a goniometer is used to measure the range of motion, ROM, to evaluate the recovery of the affected limb. However, the measurement process can cause measurement errors ranging from 4 to 5 degrees due to unskilled operation or inaccurate placement, leading to inaccurate judgments of the recovery of the affected limb. In addition, the current measurement methods do not include an assessment of muscle endurance. In this paper, the proposed device combines image recognition and a myoelectric signal sensor to measure the joint movement angle and muscle endurance. The movement angle of the elbow joint is measured using image recognition. Muscle endurance is measured using the myoelectric signal sensor. The measured data are transmitted to a cloud database via an app we have proposed to help medical staff track a patient's recovery status. The average error value of static image recognition and verification is up to 0.84 degrees. The average error value of dynamic image recognition and verification is less than 1%. The average error of total harmonic distortion (THD) verified by the myoelectric signal sensor is less than ±3%. It was proven that our system could be applied to measuring elbow joint range of motion. Since this is pilot research, most of the measurement subjects are healthy people without dysfunction in arm movement, and it is difficult to observe differences in the measurement results. In the future, experiments will be conducted on patients with elbow fractures through the IRB. This is expected to achieve the effect of encouraging patients to be actively rehabilitated at home through their measurement data and images of their actions being displayed in real time using our cheap and compact device and app.