高级检索
数字医学与健康
期刊首页
过刊列表
高级检索
稿件发表
• 论著 •
ENGLISH ABSTRACT
基于机器学习构建心脏外科术后低心排血量综合征风险预测的决策树模型
许欢
洪亮
沈骁
宋晓春
卓荦
章淬
作者及单位信息
·
DOI： 10.3760/cma.j.cn101909-20240313-00039
Structuring a decision tree model based on machine learning for risk prediction of low output syndrome after cardiac surgery
Xu Huan
Hong Liang
Shen Xiao
Song Xiaochun
Zhuo Luo
Zhang Cui
Authors Info & Affiliations
Xu Huan
Department of ICU, Nanjing First Hospital, Nanjing Medical University, Nanjing 210006, China
Hong Liang
Department of ICU, Nanjing First Hospital, Nanjing Medical University, Nanjing 210006, China
Shen Xiao
Department of ICU, Nanjing First Hospital, Nanjing Medical University, Nanjing 210006, China
Song Xiaochun
Department of ICU, Nanjing First Hospital, Nanjing Medical University, Nanjing 210006, China
Zhuo Luo
Department of ICU, Nanjing First Hospital, Nanjing Medical University, Nanjing 210006, China
Zhang Cui
Department of ICU, Nanjing First Hospital, Nanjing Medical University, Nanjing 210006, China
·
DOI： 10.3760/cma.j.cn101909-20240313-00039
0
0
0
0
0
0
PDF下载
APP内阅读
摘要

目的应用机器学习（ML）算法构建预测心脏外科术后低心排血量综合征发病风险的决策树模型，并评价其预测效果。

方法从南京市第一医院重症医学科心脏重症临床数据库中提取1 681例患者的临床资料，纳入基础信息、检验检查信息、手术信息、血流动力学信息等49项临床参数，通过ML建立预测模型，进而应用SHAP方法和部分依赖图筛选主要预测特征及其阈值，最终生成预测心脏术后低心排血量综合征的决策树模型，并通过受试者操作特征（ROC）曲线对模型预测效果进行评价。

结果SHAP算法分析显示术前左室射血分数［SHAP值=0.008 449（0.000 852）］、血乳酸水平［SHAP值=0.007 434（0.000 718）］、术中平均动脉压<65 mmHg（1 mmHg=0.133 kPa）持续时间［SHAP值=0.004 983（0.000 761）］3项参数对决策树模型具有较高的全局重要性，其构建的模型对心脏外科术后低心排血量综合征发病风险具有较高预测价值，ROC曲线下面积为0.791［95% 可信区间（ CI）：0.718~0.864］，预测准确度为0.763（95% CI：0.722~0.800），敏感度为0.736（95% CI：0.597~0.847），特异度为0.766（95% CI：0.723~0.806）。

结论应用ML算法筛选了预测心脏外科术后低心排血量综合征发病风险的特征参数，建立了预测效果良好、易于临床应用的决策树模型。

低心排血量综合征;预测模型;机器学习;决策树
ABSTRACT

ObjectiveTo evaluating the predictive effectiveness of a decision tree model constructed by machine learning(ML) algorithms for predicting the risk of low cardiac output syndrome (LCOS) after cardiac surgery.

MethodsThe clinical data of 49 clinical parameters of 1 681 patients, including basic information, examination and investigation information, surgical information, hemodynamic information were extracted from the clinical database of the cardiac intensive care in Nanjing First Hospital. Then, a prediction model was established through ML. The shapley additive explanations (SHAP) method and partial dependency graph were applied to screen the main prediction features and their thresholds. Finally, a decision tree model for predicting LCOS was generated, and the prediction effect of the model was evaluated through the receiver operating characteristic (ROC) curve.

ResultsAnalysis using the SHAP algorithm showed that three parameters-preoperative left ventricular ejection fraction [SHAP value=0.008 449 (0.000 852)], blood lactate level [SHAP value=0.007 434 (0.000 718)], and the duration of mean arterial pressure<65 mmHg (1 mmHg=0.133 kPa) during surgery [SHAP value=0.004 983 (0.000 761)]——had high global importance in the decision tree model. The model constructed with these parameters had high predictive value for the risk of postoperative LCOS in cardiac surgery patients. The area under the ROC curve was 0.791 [95% Confidence Interval ( CI): 0.718-0.864], with a prediction accuracy of 0.763 (95% CI: 0.722-0.800), sensitivity of 0.736 (95% CI: 0.597-0.847), and specificity of 0.766 (95% CI: 0.723-0.806).

ConclusionsThe study applied ML algorithms to select characteristic parameters for predicting the risk of postoperative LCOS in cardiac surgery patients, and established a decision tree model with good predictive performance and ease of clinical application.

Low cardiac output syndrome;Prediction model;Machine learning;Decision tree
Zhang Cui, Email: mocdef.3ab6138207615981
Copyright by Chinese Medical Association
No content published by the journals of Chinese Medical Association may be reproduced or abridged without authorization. Please do not use or copy the layout and design of the journals without permission.
All articles published represent the opinions of the authors, and do not reflect the official policy of the Chinese Medical Association or the Editorial Board, unless this is clearly specified.
引用本文

许欢,洪亮,沈骁,等. 基于机器学习构建心脏外科术后低心排血量综合征风险预测的决策树模型[J]. 数字医学与健康,2025,03(01)：27-33.

DOI:10.3760/cma.j.cn101909-20240313-00039

PERMISSIONS

Request permissions for this article from CCC.

评价本文
*以上评分为匿名评价

版权归中华医学会所有。

未经授权，不得转载、摘编本刊文章，不得使用本刊的版式设计。

除非特别声明，本刊刊出的所有文章不代表中华医学会和本刊编委会的观点。

低心排血量综合征（low cardiac output syndrome，LCOS）是一组以心排血量下降、外周脏器灌注不足为特点的临床综合征。目前国内的专家共识将心脏外科术后的LCOS统一定义为：心脏指数（cardiac index，CI）<2.0 L·min -1·m -2并伴有组织灌注不足的临床表现 ［ 1 ］ 。但由于肺动脉漂浮导管或脉搏指示连续心输出量（pulse indicator continuous cardiac output，PiCCO）导管并不是心脏重症患者围术期监测的常规选择，LCOS 的早期临床识别存在困难。
研究者对LCOS的预测模型进行了相关研究 ［ 2 , 3 , 4 ］ ，但此类研究均采用了传统的统计学分析方法，总体样本量偏少，仅能够纳入相对较少的研究指标，而且不同研究者对研究变量的主观选择难以避免相关偏倚。因此，关于心脏外科患者LCOS的风险预测模型研究，目前尚缺乏高质量的循证医学证据支持。
随着信息技术的飞速发展，人工智能（artificial intelligence，AI）技术在病理学 ［ 5 ］ 、影像学 ［ 6 ］ 等医学领域的应用已取得显著的进展，而且在心力衰竭患者的识别、心脏手术预后评估、超声心动图解读等心血管疾病诊疗方面也表现出重要的作用 ［ 7 ］ 。因此，本研究应用机器学习（machine learning，ML）算法将心脏外科患者术前和术中相关临床参数纳入LCOS发病风险的预测模型，进而通过对主要特征预测参数的阈值推断构建预测术后LCOS的决策树模型，以期为LCOS的早期识别提供临床辅助决策工具。
试读结束，您可以通过登录机构账户或个人账户后获取全文阅读权限。
参考文献
[1]
中国医师协会心脏重症专家委员会. 低心排血量综合征中国专家共识[J]. 解放军医学杂志, 2017,42(11):933-944. DOI: 10.11855/j.issn.0577-7402.2017.11.01 .
返回引文位置Google Scholar
百度学术
万方数据
National consensus on low cardiac output syndrome in China[J]. Medical Journal of Chinese People′s Liberation Army, 2017,42(11):933-944. DOI: 10.11855/j.issn.0577-7402.2017.11.01 .
Google Scholar
Baidu Scholar
Wanfang Data
[2]
Ding W , Ji Q , Shi Y ,et al. Predictors of low cardiac output syndrome after isolated coronary artery bypass grafting[J]. Int Heart J, 2015,56(2):144-149. DOI: 10.1536/ihj.14-231 .
返回引文位置Google Scholar
百度学术
万方数据
[3]
Du X , Chen H , Song X ,et al. Risk factors for low cardiac output syndrome in children with congenital heart disease undergoing cardiac surgery: a retrospective cohort study[J]. BMC Pediatr, 2020,20(1):87. DOI: 10.1186/s12887-020-1972-y .
返回引文位置Google Scholar
百度学术
万方数据
[4]
Posada-Martinez EL , Fritche-Salazar JF , Arias-Godinez JA ,et al. Right ventricular longitudinal strain predicts low-cardiac-output syndrome after surgical aortic valve replacement in patients with preserved and mid-range ejection fraction[J]. J Cardiothorac Vasc Anesth, 2021,35(6):1638-1645. DOI: 10.1053/j.jvca.2020.12.008 .
返回引文位置Google Scholar
百度学术
万方数据
[5]
Shafi S , Parwani AV . Artificial intelligence in diagnostic pathology[J]. Diagn Pathol, 2023,18(1):109. DOI: 10.1186/s13000-023-01375-z .
返回引文位置Google Scholar
百度学术
万方数据
[6]
Küstner T , Qin C , Sun C ,et al. The intelligent imaging revolution: artificial intelligence in MRI and MRS acquisition and reconstruction[J]. MAGMA, 2024,37(3):329-333. DOI: 10.1007/s10334-024-01179-2 .
返回引文位置Google Scholar
百度学术
万方数据
[7]
Kilic A . Artificial intelligence and machine learning in cardiovascular health care[J]. Ann Thorac Surg, 2020,109(5):1323-1329. DOI: 10.1016/j.athoracsur.2019.09.042 .
返回引文位置Google Scholar
百度学术
万方数据
[8]
Pérez Vela JL , Martín Benítez JC , Carrasco González M ,et al. Clinical practice guide for the management of low cardiac output syndrome in the postoperative period of heart surgery[J]. Med Intensiva, 2012,36(4):e1-44. DOI: 10.1016/j.medin.2012.02.007 .
返回引文位置Google Scholar
百度学术
万方数据
[9]
Alyass A , Turcotte M , Meyre D . From big data analysis to personalized medicine for all: challenges and opportunities[J]. BMC Med Genomics, 2015,8:33. DOI: 10.1186/s12920-015-0108-y .
返回引文位置Google Scholar
百度学术
万方数据
[10]
Yang J , Li Y , Liu Q ,et al. Brief introduction of medical database and data mining technology in big data era[J]. J Evid Based Med, 2020,13(1):57-69. DOI: 10.1111/jebm.12373 .
返回引文位置Google Scholar
百度学术
万方数据
[11]
Obermeyer Z , Emanuel EJ . Predicting the future-big data, machine learning, and clinical medicine[J]. N Engl J Med, 2016,375(13):1216-1219. DOI: 10.1056/NEJMp1606181 .
返回引文位置Google Scholar
百度学术
万方数据
[12]
Goto Y , Maeda T , Nakatsu-Goto Y . Decision tree model for predicting long-term outcomes in children with out-of-hospital cardiac arrest: a nationwide, population-based observational study[J]. Crit Care, 2014,18(3):R133. DOI: 10.1186/cc13951 .
返回引文位置Google Scholar
百度学术
万方数据
[13]
任杰,朱飞奇. 重型颅脑损伤合并多发伤患者早期死亡的决策树模型研究[J]. 中国急救医学, 2022,42(4):343-346. DOI: 10.3969/j.issn.1002-1949.2022.04.013 .
返回引文位置Google Scholar
百度学术
万方数据
Ren J , Zhu FQ . A decision tree model for early death in the patients with severe traumatic brain injury and polytrauma[J]. Chinese Journal of Critical Care Medicine, 2022,42(4):343-346. DOI: 10.3969/j.issn.1002-1949.2022.04.013 .
Goto CitationGoogle Scholar
Baidu Scholar
Wanfang Data
[14]
Elhazmi A , Al-Omari A , Sallam H ,et al. Machine learning decision tree algorithm role for predicting mortality in critically ill adult COVID-19 patients admitted to the ICU[J]. J Infect Public Health, 2022,15(7):826-834. DOI: 10.1016/j.jiph.2022.06.008 .
返回引文位置Google Scholar
百度学术
万方数据
[15]
Akman B , Kaya AT . Effects of coronary arter y calcified plaque and stent on severity and survival of COVID-19 patients: a decision tree model study [J]. Kardiologiia, 2023,63(7):54-61. DOI: 10.18087/cardio.2023.7.n2251 .
返回引文位置Google Scholar
百度学术
万方数据
[16]
Donal E , L′official G , Kosmala W . New guidelines for managing chronic heart failure patients and new needs in echocardiography[J]. Int J Cardiol, 2022,353:71-72. DOI: 10.1016/j.ijcard.2022.01.035 .
返回引文位置Google Scholar
百度学术
万方数据
[17]
Hagendorff A , Stöbe S , Kandels J ,et al. Diagnostic role of echocardiography for patients with heart failure symptoms and preserved left ventricular ejection fraction[J]. Herz, 2022,47(4):293-300. DOI: 10.1007/s00059-022-05118-6 .
返回引文位置Google Scholar
百度学术
万方数据
[18]
Amabili P , Benbouchta S , Roediger L ,et al. Low cardiac output syndrome after adult cardiac surgery: predictive value of peak systolic global longitudinal strain[J]. Anesth Analg, 2018,126(5):1476-1483. DOI: 10.1213/ANE.0000000000002605 .
返回引文位置Google Scholar
百度学术
万方数据
[19]
Becher H , Alhumaid W , Windram J ,et al. Contrast echocardiography in heart failure: update 2023[J]. Curr Heart Fail Rep, 2024,21(2):63-72. DOI: 10.1007/s11897-024-00647-z .
返回引文位置Google Scholar
百度学术
万方数据
[20]
Bari V , Vaini E , Pistuddi V ,et al. Short-term multiscale complexity analysis of cardiovascular variability improves low cardiac output syndrome risk stratification after coronary artery bypass grafting[J]. Physiol Meas, 2019,40(4):044001. DOI: 10.1088/1361-6579/ab12f0 .
返回引文位置Google Scholar
百度学术
万方数据
[21]
Cluntun AA , Badolia R , Lettlova S ,et al. The pyruvate-lactate axis modulates cardiac hypertrophy and heart failure[J]. Cell Metab, 2021,33(3):629-648.e10. DOI: 10.1016/j.cmet.2020.12.003 .
返回引文位置Google Scholar
百度学术
万方数据
[22]
Yi L , Tang D , Xiang X ,et al. New mechanisms: from lactate to lactylation to rescue heart failure[J]. Biosci Trends, 2024,18(1):105-107. DOI: 10.5582/bst.2024.01000 .
返回引文位置Google Scholar
百度学术
万方数据
[23]
Zeger SL , Irizarry R , Peng RD . On time series analysis of public health and biomedical data[J]. Annu Rev Public Health, 2006,27:57-79. DOI: 10.1146/annurev.publhealth.26.021304.144517 .
返回引文位置Google Scholar
百度学术
万方数据
[24]
Tseng PY , Chen YT , Wang CH ,et al. Prediction of the development of acute kidney injury following cardiac surgery by machine learning[J]. Critical Care, 2020,24(1):478. DOI: 10.1186/s13054-020-03179-9 .
返回引文位置Google Scholar
百度学术
万方数据
[25]
Bihorac A , Ozrazgat-Baslanti T , Ebadi A ,et al. Mysurgeryrisk: development and validation of a machine-learning risk algorithm for major complications and death after surgery[J]. Ann Surg, 2019,269(4):652-662. DOI: 10.1097/SLA.0000000000002706 .
返回引文位置Google Scholar
百度学术
万方数据
备注信息
A
章淬，Email： mocdef.3ab6138207615981
B

许欢：酝酿和设计试验、实施研究、采集数据、分析/解释数据、起草文章、统计分析；洪亮：酝酿和设计试验、采集数据、分析/解释数据、对文章的知识性内容作批评性审阅、统计分析；沈骁：实施研究、采集数据、对文章的知识性内容作批评性审阅、技术和材料支持；宋晓春：实施研究、采集数据、对文章的知识性内容作批评性审阅、技术和材料支持；卓荦：分析/解释数据、起草文章、技术和材料支持；章淬：酝酿和设计试验、对文章的知识性内容作批评性审阅、获取研究经费、行政和材料支持

C
许欢, 洪亮, 沈骁, 等. 基于机器学习构建心脏外科术后低心排血量综合征风险预测的决策树模型[J]. 数字医学与健康, 2025, 3(1): 27-33. DOI: 10.3760/cma.j.cn101909-20240313-00039.
D
所有作者声明无利益冲突
E
南京市卫生科技发展专项资金 (ZKX19021)
评论 （0条）
注册
登录
时间排序
暂无评论，发表第一条评论抢沙发
最新推荐
更多
尘肺病小阴影密集度的CT分级方法及参考片探讨
翟荣存 等
中华劳动卫生职业病杂志 2021,39(06)
自身免疫性甲状腺病防治的挑战和曙光
单忠艳 等
国际内分泌代谢杂志 2021,41(01)
生物可降解支架：曙光就在前方
韩雅玲
中华心血管病杂志 2019,47(04)
发热伴肺部阴影鉴别诊断专家共识
发热伴肺部阴影鉴别诊断共识专家组
中华结核和呼吸杂志 2016,39(03)
MedAI助手（体验版）
文档即答
智问智答
机器翻译
回答内容由人工智能生成，我社无法保证其准确性和完整性，该生成内容不代表我们的态度或观点，仅供参考。
生成快照
文献快照

你好，我可以帮助您更好的了解本文，请向我提问您关注的问题。

0/2000

《中华医学会杂志社用户协议》 | 《隐私政策》

《SparkDesk 用户协议》 | 《SparkDesk 隐私政策》

网信算备340104764864601230055号 | 网信算备340104726288401230013号

技术支持：

历史对话
本文全部
还没有聊天记录
设置
模式
纯净模式沉浸模式
字号
小中大