基于LASSO回归构建针刺治疗缺血性卒中预后模型

Construction and validation of a prognostic model for acupuncture method of ischemic stroke based on LASSO regression model

关键词:  针刺  缺血性卒中  临床预测模型

Key words:  acupuncture  ischemic stroke  clinical prediction model

摘要:

[目的]建立和评估接受针刺治疗缺血性卒中患者预后改善的临床模型。[方法]选取2021年2月—2022年4月就诊于天津中医药大学第一附属医院针灸科住院部的患者,结合患者人口统计学资料、既往病史、入院情况及治疗因素,基于LASSO回归筛选预测因素构建Logistic回归模型,采用列线图对模型进行可视化呈现。[结果]本研究以年龄、高血压病病史、接受康复治疗的情况及发病3个月内针刺次数建立模型预测患者发病3个月改良Rankin量表改善状况。采用Bootstrap法对模型进行内部验证,模型ROC曲线下面积为0.747,Hosmer-Lemeshow拟合优度检验P值为0.52,DOC临床决策曲线的阈值率为29%~88%,说明模型具有较好的区分度、校准性及临床获益度。[结论]基于LASSO回归构建针刺治疗缺血性卒中预后模型具有较好的临床应用价值。

Abstract:

[Objective] To establish and evaluate a clinical model for the prognosis of ischemic stroke patients treated by acupuncture.[Methods] We selected patients from February 2021 to April 2022 in the residential department of First Teaching Hospital of Tianjin University of Traditional Chinese Medicine. Combined with the patient's demographic data,past medical history,admission and treatment factors,the Logistic regression model was constructed based on LASSO regression screening predictors. The nomogram was used to visualize the model. [Results] In this study,the model was established based on age,history of hypertension,rehabilitation treatment,and the number of acupuncture within three months to predict the improvement of the modified Rankin scale at 3 months.The model is internally verified using the Bootstrap method. The area under the ROC curve of the model is 0.747,and the P-value of the Hosmer-Lemeshow goodness-of-fit test is 0.52. DOC clinical decision curve was 29%~88%. It shows that the model has good accuracy,discrimination,calibration and clinical benefit. [Conclusion] The prognostic model for acupuncture method of ischemic stroke based on LASSO regression model has good clinical application value.