切换至 "中华医学电子期刊资源库"

中华腔镜泌尿外科杂志(电子版) ›› 2021, Vol. 15 ›› Issue (04) : 280 -284. doi: 10.3877/cma.j.issn.1674-3253.2021.04.003

临床研究

机器学习算法模型预测体外冲击波碎石治疗输尿管结石的疗效
侯祺1, 相洋2, 吴娜珊1, 肖月1, 肖龙1, 李潇1, 王锐1, 孙中义1,()   
  1. 1. 518055 深圳大学总医院泌尿外科
    2. 518000 深圳,鹏城实验室
  • 收稿日期:2021-04-29 出版日期:2021-08-01
  • 通信作者: 孙中义
  • 基金资助:
    国家自然科学基金(82002716); 广东省普通高校特色创新类项目(2019KTSCX146); 深圳市自然科学基金(JCYJ20190808164209301); 深圳市新引进高端人才科研启动基金、深圳大学总医院科研启动基金(SUGH2020QD005)

Machine learning prediction of stone-free rate in patients with ureter stone after treatment of extracorporeal shock wave lithotripsy

Qi Hou1, Yang Xiang2, Nashan Wu1, Yue Xiao1, Long Xiao1, Xiao Li1, Rui Wang1, Zhongyi Sun1,()   

  1. 1. Department of Urology, Shenzhen University General Hospital, Shenzhen University, Shenzhen 518055, China
    2. Pengcheng Laboratory, Shenzhen 518000, China
  • Received:2021-04-29 Published:2021-08-01
  • Corresponding author: Zhongyi Sun
引用本文:

侯祺, 相洋, 吴娜珊, 肖月, 肖龙, 李潇, 王锐, 孙中义. 机器学习算法模型预测体外冲击波碎石治疗输尿管结石的疗效[J]. 中华腔镜泌尿外科杂志(电子版), 2021, 15(04): 280-284.

Qi Hou, Yang Xiang, Nashan Wu, Yue Xiao, Long Xiao, Xiao Li, Rui Wang, Zhongyi Sun. Machine learning prediction of stone-free rate in patients with ureter stone after treatment of extracorporeal shock wave lithotripsy[J]. Chinese Journal of Endourology(Electronic Edition), 2021, 15(04): 280-284.

目的

探讨基于机器学习算法构建体外冲击波碎石术(ESWL)治疗输尿管结石疗效的预测模型,比较各种算法模型的预测效果及优劣。

方法

纳入接受ESWL治疗的输尿管结石患者1 116例。利用患者性别、年龄、身高、体重、病程、临床症状、结石长径和短径等特征因素以及体外冲击波碎石治疗3个月后的清石结局数据,分别构建随机森林(RF)和极致梯度提升树(XGBoost)预测模型,与Logistic回归预测模型相互比较。

结果

Logistic回归预测模型对体外冲击波碎石疗效的预测准确率为84.67%,ROC曲线下面积(AUC)为0.70。随机森林和极致梯度提升树预测模型的预测准确率分别为91.76%、98.75%,AUC分别为0.9904和0.9998。三种预测模型结果提示结石部位与结石负荷影响治疗的成功率。

结论

机器学习算法的随机森林和极致梯度提升树算法可以极大提高ESWL治疗输尿管结石疗效的预测准确率。

Objective

To compare the performance of machine learning prediction model for success rate of extracorporeal shock wave lithotripsy (ESWL) treating ureter stone.

Methods

1 116 patients who underwent ESWL for ureter stone were enrolled. Clinical data including patient’s gender, age, height, weight, disease course, clinical symptom and characteristics of stone and the outcome of treatment were collected. Prediction model was established by random forest, extreme gradient boosting trees and logistic regression in Python 3.7.

Results

Overall predictive accuracy and area under curve (AUC) calculated by Logistic regression prediction model were 84.67% and 0.70. Results of random forest and extreme gradient boosting trees prediction model were 91.76%, 0.9904 and 98.75%, 0.9998. The three prediction models revealed stone-free rate impacted by patients’ stone site and stone burden.

Conclusions

Machine learning models are better than logistic regression for predicting success rate of ESWL treating ureter stone.

表1 ESWL治疗的输尿管结石患者纳入临床数据的基线特征
图1 三种模型在测试模式1条件下的预测ROC曲线
表2 三种算法模型的预测性能比较
图2 随机森林模型中变量的重要性指数
图3 极致梯度提升树模型中变量的重要性指数
[1]
Chaussy C, Brendel W, Schmiedt E. Extracorporeally induced destruction of kidney stones by shock waves[J]. Lancet, 1980, 2(8207): 1265-1268.
[2]
苑海春,薛玉泉,焦杨, 等. 原发性巨输尿管症合并结石行体外碎石疗效[J/CD]. 中华腔镜泌尿外科杂志(电子版), 2020, 14(3): 228-231.
[3]
郭万松,杨波,赵航. 体外冲击波碎石术治疗尿路结石研究进展[J/CD]. 中华腔镜泌尿外科杂志(电子版), 2020, 14(5): 393-396.
[4]
Obermeyer Z, Emanuel EJ. Predicting the future-big data, machine learning, and clinical medicine [J]. N Engl J Med, 2016, 375(13): 1216-1219.
[5]
De Silva D, Ranasinghe W, Bandaragoda T, et al. Machine learning to support social media empowered patients in cancer care and cancer treatment decisions [J]. PLoS One, 2018, 13(10): e0205855.
[6]
Peng Z, Qi H, Huijian Z, et al. Prediction of outcome of extracorpaoreal shock wave lithotripsy for solitary ureteral calcul[J] J Mod Urol, 2013, 18(1): 27-29.
[7]
Sipper M, Moore JH. Conservation machine learning: a case study of random forests [J]. Sci Rep, 2021, 11(1): 3629.
[8]
Yu D, Liu Z, Su C, et al. Copy number variation in plasma as a tool for lung cancer prediction using extreme gradient boosting (xgboost) classifier[J]. Thorac Cancer, 2020, 11(1): 95-102.
[9]
Abraham A, Pedregosa F, Eickenberg M, et al. Machine learning for neuroimaging with scikit-learn[J]. Front Neuroinform, 2014, 8: 14.
[10]
Yang SW, Hyon YK, Na HS, et al. Machine learning prediction of stone-free success in patients with urinary stone after treatment of shock wave lithotripsy [J]. BMC Urol, 2020, 20(1): 88.
[11]
Seckiner I, Seckiner S, Sen H, et al. A neural network - based algorithm for predicting stone - free status after ESWL therapy[J]. Int Braz J Urol, 2017, 43(6): 1110-1114.
[12]
Moorthy K, Krishnan M. Prediction of fragmentation of kidney stones: A statistical approach from NCCT images [J]. Can Urol Assoc J, 2016, 10(7-8): E237-E240.
[13]
Choo MS, Uhmn S, Kim JK, et al. A prediction model using machine learning algorithm for assessing stone-free status after single session shock wave lithotripsy to treat ureteral stones [J]. J Urol, 2018, 200(6): 1371-1377.
[14]
Gomha MA, Sheir KZ, Showky S, et al. Can we improve the prediction of stone-free status after extracorporeal shock wave lithotripsy for ureteral stones? A neural network or a statistical model[J]? J Urol, 2004, 172(1): 175-179.
[15]
Xu ZH, Zhou S, Jia CP, et al. Prediction of proximal ureteral stones clearance after shock wave lithotripsy using an artificial neural network[J]. Urol J, 2021, 18: 6476.
[1] 胡诤贇, 史建伟, 申建伟, 王冰, 蒋春苗, 刘冲. 基于机器学习鉴定早产儿支气管肺发育不良的关键基因[J]. 中华妇幼临床医学杂志(电子版), 2023, 19(04): 446-454.
[2] 范帅华, 郭伟, 郭军. 基于机器学习的决策树算法在血流感染预后预测中应用现状及展望[J]. 中华实验和临床感染病杂志(电子版), 2023, 17(05): 289-293.
[3] 庞亮月, 林焕彩. 人工智能在龋病诊疗中的应用[J]. 中华口腔医学研究杂志(电子版), 2023, 17(03): 162-166.
[4] 徐慧新, 刘波, 唐立钧. 体外冲击波治疗>1 cm输尿管上段结石失败的预测模型建立[J]. 中华腔镜泌尿外科杂志(电子版), 2023, 17(05): 506-511.
[5] 伍学成, 李远伟, 袁武雄, 王建松, 石泳中, 卢强, 李卓, 陈佳, 刘哲, 滕伊漓, 高智勇. 炎症介质谱联合降钙素原在尿源性脓毒血症中的诊断价值[J]. 中华腔镜泌尿外科杂志(电子版), 2023, 17(05): 476-480.
[6] 胡金科, 钟文. 妊娠期输尿管结石的处理与转归[J]. 中华腔镜泌尿外科杂志(电子版), 2023, 17(04): 377-381.
[7] 汪帮琦, 张小明, 肖远松, 聂海波, 吕军, 王尉. 电切镜与输尿管镜在输尿管中下段结石并输尿管开口进镜困难中的应用[J]. 中华腔镜泌尿外科杂志(电子版), 2023, 17(02): 157-159.
[8] 葛云鹏, 崔红元, 宋京海. 人工智能在原发性肝癌诊断、治疗及预后中的应用[J]. 中华肝脏外科手术学电子杂志, 2023, 12(04): 367-371.
[9] 田亚, 吴美龙, 冯晓彬. 人工智能在肝细胞癌诊疗中的应用[J]. 中华肝脏外科手术学电子杂志, 2023, 12(03): 258-262.
[10] 王琦琦, 龚梦元, 冯正源, 韩亮, 王铮, 马清涌, 仵正. 基于术前检测指标构建Logistic回归模型在预测壶腹周围癌病理类型中的价值[J]. 中华肝脏外科手术学电子杂志, 2023, 12(02): 196-200.
[11] 陈显金, 吴芹芹, 何长春, 张庆华. 利用多模态医学数据和机器学习构建脑出血预后预测模型的研究[J]. 中华脑科疾病与康复杂志(电子版), 2023, 13(04): 193-198.
[12] 黄文鹏, 邱永康, 杨琦, 宋乐乐, 陈钊, 范岩, 康磊. PET相关影像组学在肿瘤预后中的研究进展[J]. 中华消化病与影像杂志(电子版), 2023, 13(02): 104-110.
[13] 郭震天, 张宗明, 赵月, 刘立民, 张翀, 刘卓, 齐晖, 田坤. 机器学习算法预测老年急性胆囊炎术后住院时间探索[J]. 中华临床医师杂志(电子版), 2023, 17(9): 955-961.
[14] 熊鑫, 邓勇志. 基于血管内超声的机器学习在冠状动脉病变中的研究进展[J]. 中华诊断学电子杂志, 2023, 11(03): 153-157.
[15] 王俊杰, 尹晓亮, 刘二腾, 陆军, 祁鹏, 胡深, 杨希孟, 陈鲲鹏, 张东, 王大明. 机器学习对预测颈内动脉非急性闭塞患者血管内再通术成功的潜在价值[J]. 中华脑血管病杂志(电子版), 2023, 17(05): 464-470.
阅读次数
全文


摘要