文献速递关节外科创伤骨科

机器学习基于可穿戴传感器性能测试预测全髋关节置换患者的跌倒风险

Machine Learning Predicts the Fall Risk of Total Hip Arthroplasty Patients Based on Wearable Sensor Instrumented Performance Tests.
PubMed 文献 · J Arthroplasty · 2020-08-19

研究纳入72例全髋关节置换患者,在术前、术后2周和术后6周佩戴传感器完成计时起立行走测试,用术前和术后2周数据构建特征集、术后6周总时间作为标签,开发支持向量机和线性判别分析分类器模型。仅用术前传感器指标的支持向量机模型准确率0.87、敏感度0.97、特异度0.46、AUC 0.82;加入术后2周数据后线性判别分析模型准确率0.90、敏感度0.93、特异度0.59、AUC 0.88。作者认为高准确率的跌倒风险预测模型对患者、临床医生和医疗系统有价值。

要点

  • 仅用术前传感器指标的支持向量机模型预测跌倒风险:准确率0.87、敏感度0.97、特异度0.46、AUC 0.82。
  • 加入术后2周数据后线性判别分析模型:准确率0.90、敏感度0.93、特异度0.59、AUC 0.88。
  • 作者认为高风险患者可采取预防措施,低风险患者可导向加速康复 care 项目。

核心结论

  • 跌倒风险预测模型的高准确率对患者、临床医生和医疗系统有价值。
  • 作者认为高风险患者可实施预防措施,低风险患者可导向加速康复 care 项目。

研究设计与样本

  • 研究为监督式机器学习研究,患者n=72,在术前、术后2周和术后6周佩戴传感器完成计时起立行走测试。
  • 术前和术后2周数据用于构建特征集,术后6周总时间用作标签。
  • 开发并测试支持向量机和线性判别分析分类器模型在不同特征集和特征降维方案组合下的表现。
  • 采用10折留部分受试者交叉验证方案评估所有模型的准确率、敏感度、特异度和受试者工作特征曲线下面积(AUC)。

主要发现

  • 仅用术前传感器指标的支持向量机分类器模型:准确率0.87、敏感度0.97、特异度0.46、AUC 0.82。
  • 加入术后2周数据后线性判别分析分类器模型:准确率0.90、敏感度0.93、特异度0.59、AUC 0.88。

关键数值与推荐

  • 仅用术前传感器指标的支持向量机模型:准确率0.87、敏感度0.97、特异度0.46、AUC 0.82。
  • 加入术后2周数据后线性判别分析模型:准确率0.90、敏感度0.93、特异度0.59、AUC 0.88。

临床怎么用

  • 作者认为高风险患者可实施预防措施,低风险患者可导向加速康复 care 项目。

局限性

  • 原文未报告局限性

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

BACKGROUND: The prevalence of falls affects the wellbeing of aging adults and places an economic burden on the healthcare system. Integration of wearable sensors into existing fall risk assessment tools enables objective data collection that describes the functional ability of patients. In this study, supervised machine learning was applied to sensor-derived metrics to predict the fall risk of patients following total hip arthroplasty. METHODS: At preoperative, 2-week, and 6-week postoperative appointments, patients (n = 72) were instrumented with sensors while they performed the timed-up-and-go walking test. Preoperative and 2-week postoperative data were used to form the feature sets and 6-week total times were used as labels. Support vector machine and linear discriminant analysis classifier models were developed and tested on various combinations of feature sets and feature reduction schemes. Using a 10-fold leave-some-subjects-out testing scheme, the accuracy, sensitivity, specificity, and area under the receiver-operator curve (AUC) were evaluated for all models. RESULTS: A high performance model (accuracy = 0.87, sensitivity = 0.97, specificity = 0.46, AUC = 0.82) was obtained with a support vector machine classifier using sensor-derived metrics from only the preoperative appointment. An overall improved performance (accuracy = 0.90, sensitivity = 0.93, specificity = 0.59, AUC = 0.88) was achieved with a linear discriminant analysis classifier when 2-week postoperative data were added to the preoperative data. CONCLUSION: The high accuracy of the fall risk prediction models is valuable for patients, clinicians, and the healthcare system. High-risk patients can implement preventative measures and low-risk patients can be directed to enhanced recovery care programs.

来源:J Arthroplasty,查看原文。版权归原作者所有。

引用

  1. Polus JS, Bloomfield RA, Vasarhelyi EM, et al. Machine Learning Predicts the Fall Risk of Total Hip Arthroplasty Patients Based on Wearable Sensor Instrumented Performance Tests. J Arthroplasty. 2020. PMID: 32928593. doi:10.1016/j.arth.2020.08.034

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。