Electroacupuncture therapy significantly improves muscle strength in patients with peripheral facial palsy.
In the study, a surface nerve electromyogram (EMG) recognition model was created that utilized a multiview convolutional neural network to investigate the value of electroacupuncture as a treatment for peripheral facial palsy. The model was compared with traditional single-view convolutional neural networks, and the influence of multiple view aggregation methods on facial recognition accuracy was analyzed.
As for the data, they were obtained from 150 patients randomly divided into two groups: a control group receiving only basic treatment and a treatment group given both basic treatment and electroacupuncture. The therapeutic effect of electroacupuncture was evaluated after a four-week treatment period based on surface EMG parameters and the Horsfall-Barratt scale.
The results indicated that the multiview convolutional neural networks had superior facial recognition accuracy compared to all single-view networks. The treatment group showed a higher overall treatment effectiveness rate compared to the control group, demonstrating that electroacupuncture can significantly improve muscle strength in patients with peripheral facial palsy. Furthermore, the proposed multiview aggregation network demonstrated higher face recognition accuracy than other multiview aggregation methods.
