基于四元数与ViT模型的颈肩疾病康复患者疼痛表情分类研究

Research on Classification of Pain Expression in Rehabilitation Patients with Neck or Shoulder Diseases Using Quaternion-ViT Model

与日俱增的社会生活压力增大了人们患有颈椎病、肩周炎等疾病的风险。此类颈肩疾病患者以疼痛症状为主要表现,患者进行康复治疗的评定时,患者会同步呈现不同程度的面部疼痛表情,对这些疼痛表情进行特征信息分析,可以给临床评价提供重要的参考。提出了一种基于深度学习算法的疼痛表情分类模型QA-ViT,可用于患者康复治疗中的辅助诊断。QA-ViT模型可有效地对患者表情疼痛程度进行分类,对3种不同程度疼痛的分类准确率高达92.18%。

The increasing pressure of social life increases the risk of people suffering from cervical spondylosis,periarthritis of shoulder and other diseases.The patients with such neck and shoulder diseases mainly show pain symptoms.When the patients evaluate the rehabilitation treatment,they will show different levels of facial pain expressions synchronously.The analysis of the characteristics of these pain expressions can provide important reference for clinical evaluation.A pain expression classification model QA ViT based on deep learning algorithm is proposed,which can be used for auxiliary diagnosis in patients'rehabilitation treatment.QA ViT model can effectively classify the degree of facial pain of patients,and the classification accuracy of three different degrees of pain is as high as 92.18%.