南方医科大学学报

南方医科大学学报 ›› 2025, Vol. 45 ›› Issue (9): 2019-2025.doi: 10.12122/j.issn.1673-4254.2025.09.21

• • 上一篇    

经尿道前列腺钬激光剜除术后低体温风险预测模型：基于逻辑回归、决策树和支持向量机

姜君1, 封硕2, 孙银贵3, 安燕3()   

  1. 1.山东第二医科大学附属医院，手术室，山东 潍坊 261000
    2.山东第二医科大学附属医院，妇科，山东 潍坊 261000
    3.山东第二医科大学附属医院，麻醉科，山东 潍坊 261000
  • 收稿日期:2024-11-25 出版日期:2025-09-20 发布日期:2025-09-28
  • 通讯作者: 安燕 E-mail:anyanfy@sdsmu.edu.cn
  • 作者简介:姜 君，在读硕士研究生，E-mail: fyjiangjun @sdsmu.edu.cn
  • 基金资助:
    山东省中医药科技项目(Q-2023147);潍坊市科学技术发展计划（医学类）(2023YX057);潍坊市卫健委科研项目(WFWSJK-2023-033);潍坊医学院2022年校级教育教学改革与研究课题(2022YB051)

Construction of risk prediction models of hypothermia after transurethral holmium laser enucleation of the prostate based on three machine learning algorithms

Jun JIANG1, Shuo FENG2, Yingui SUN3, Yan AN3()   

  1. 1.Operating Room, Affiliated Hospital of Shandong Second Medical University, Weifang 261000, China
    2.Department of Gynecology, Affiliated Hospital of Shandong Second Medical University, Weifang 261000, China
    3.Department of Anesthesiology, Affiliated Hospital of Shandong Second Medical University, Weifang 261000, China
  • Received:2024-11-25 Online:2025-09-20 Published:2025-09-28
  • Contact: Yan AN E-mail:anyanfy@sdsmu.edu.cn

RichHTML

7

PDF

140

摘要：

目的 运用逻辑回归、决策树和支持向量机构建经尿道前列腺钬激光剜除术后低体温风险预测模型并比较性能，为评估及预防经尿道前列腺钬激光剜除术后低体温提供依据。 方法 回顾性收集本中心403例、另一中心（潍坊市人民医院）120例经尿道前列腺钬激光剜除术后低体温患者的临床资料，采用逻辑回归、决策树和支持向量机3种机器学习方法构建经尿道前列腺钬激光剜除术后低体温风险预测模型，采用准确性、召回率、精确率、F1指数和受试者工作特征（ROC）曲线下面积（AUC）评价模型性能。 结果 纳入患者的手术时长、前列腺重量、术中冲洗量和是否偏瘦，共4个变量。本中心选择70%的数据集（283例）作为训练集，30%的数据集（120例）作为验证集，另一中心数据（120例）作为外部验证集。在训练集、验证集及外部验证集中，支持向量机的精确率及准确率均为最优，在训练集与验证集中支持向量机的ROC均为最优，逻辑回归次之，且二者在2个数据集中的AUC差异不大。对比支持向量机（SVM）模型与逻辑回归和决策树模型发现，SVM在验证集上的精确率、准确率、召回率、F1指数和AUC值方面均超过其他2种模型。SVM在外部验证集上的精确率、准确率均优于逻辑回归和决策树模型，召回率、F1指数、AUC值略低于决策树模型。SVM在训练集上的精确率、准确率、F1指数和AUC值均优于逻辑回归和决策树模型，尽管其召回率略低于决策树模型。 结论 支持向量机在构建经尿道前列腺钬激光剜除术后低体温风险预测模型中具有较高的性能及较好的泛化能力，可为相关临床决策提供指导建议。

关键词: 前列腺, 低体温, 危险因素, 机器学习, 预测模型

Abstract:

Objective To develop risk prediction models for postoperative hypothermia after transurethral holmium laser enucleation of the prostate (HoLEP) using machine learning algorithms. Methods We retrospectively analyzed the clinical data of 403 patients from our center (283 patients in the training set and 120in the internal validation set) and 120 patients from Weifang People's Hospital (as the external validation set). The risk prediction models were built using logistic regression, decision tree and support vector machine (SVM), and model performance was evaluated in terms of accuracy, recall, precision, F1 score and AUC. Results Operation duration, prostate weight, intraoperative irrigation volume, and being underweight were identified as the predictors of postoperative hypothermia following HoLEP. Among the 3 algorithms, SVM showed the best precision rate and accuracy in all the 3 data sets and the best area under the ROC (AUC) in the training set and validation set, followed by logistic regression, which had a similar AUC in the two data sets. SVM outperformed logistic regression and decision tree models in the validation set in precision, accuracy, recall, F1 score, and AUC, and performed well in the external validation set with better precision rate and accuracy than logistic regression and decision tree models but slightly lower recall rate, F1 index, and AUC value than the decision tree model. SVM outperformed logistic regression and decision tree models in precision, accuracy, F1 score, and AUC in the training set, but had slightly lower recall rate than the decision tree. Conclusion Among the 3 models, SVM has the best performance and generalizability for predicting post-HoLEP hypothermia risk to provide support for clinical decisions.

Key words: prostate, hypothermia, risk factors, machine learning, prediction model