基于Logistic曲线模型的颈椎病根性疼痛物理治疗效果预测分析

关键词：颈椎病  物理治疗  疼痛评定  LOGISTIC模型  预测

Keywords:cervical radiculopathy  physical therapy  pain measurement  LOGISTIC model  prediction

摘要：

      摘要 目的：在对治疗过程完成数学表达的基础上，建立能够预测颈椎病根性疼痛物理治疗效果的预测模型。 方法：对一组连续接受10次物理治疗的颈椎病根性疼痛患者在治疗过程中连续进行视觉模拟评分法（VAS）疼痛评定，采用前期建立的LOGISTIC曲线模型，分别以2—9次的测量值完成曲线拟合并预测全疗程的最终效果，将预测结果与实际测量结果进行比较探讨各预测模型的预测能力，确定合理的预测模型。 结果：模型的预测能力随试验治疗次数的增加而增加，其中4次以下测量值完成的预测分析，其预测能力较低。当使用5次以上测量结果完成分析时，预测能力优于4次以下测量值的预测结果，预测效果满意。 结论：采用LOGISTIC回归模型，以4次连续的试验性治疗所获得的5个测量数据能够有效地预测全疗程的治疗效果。

Abstract:

      Abstract Objective: To establish a predictive model to describe the effectiveness of physical therapy on cervical radiculopathy patients′ pain. Method: Multiple physical therapies were administered in 90 cervical radiculopathy patients with pain for 10 consecutive sessions. Visual analogue scale(VAS) was used to measure the pain severity before each session. The first 2 to 9 pain measurements were involved in different predictive models. The predictive results from different models were compared. Result: The more measurements were involved in the predictive model, the more effective prediction will be. When the model only included less than 4 measurements, the ability of predictive model was poor. When more than 5 measurements were used with the model, the predictions were successful. Conclusion: The LOGISTIC model can predict the effectiveness of multiple physical therapies on cervical radiculopathy patients, by 4 consecutive sessions of treatment with 5 measurement results.