Machine learning compared with rule-in/rule-out algorithms and logistic regression to predict acute myocardial infarction based on troponin T concentrations.
Machine learning compared with rule-in/rule-out algorithms and logistic regression to predict acute myocardial infarction based on troponin T concentrations.
复制标题
与规则内/排除算法和逻辑回归相比，机器学习可以根据肌钙蛋白T浓度来预测急性心肌梗死。
DOI：
10.1002/emp2.12363
复制
发表时间：
2021-04
影响因子：
2.3
通讯作者：
Björk J
中科院分区：
文献类型：
--
作者：
Björkelund A;Ohlsson M;Lundager Forberg J;Mokhtari A;Olsson de Capretz P;Ekelund U;Björk J
计算机化决策支持工具可以提高急诊科（ED）胸痛患者对急性心肌梗死（AMI）的诊断。主要目的是评估机器学习算法的预测准确性，该算法基于成对的高灵敏度心肌肌钙蛋白T （hs - cTnT）浓度与不同的采样时间、年龄和性别，以排除AMI。在这项基于登记的横断面诊断研究中，我们对2013-2014年瑞典两家医院的5695名胸痛患者进行了回顾性研究，我们使用了200次5倍交叉验证，以比较人工神经网络（ANN）与欧洲指南推荐的hs - cTnT 0/1和0/3小时算法的性能，并与无交互项的逻辑回归进行了比较。主要结局是不能排除AMI的中间风险组的大小，同时保持各模型的敏感性（排除）和特异性（规则）不变。人工神经网络和逻辑回归在受试者工作特征曲线下的面积相似（95%）。在满足采样时间要求（0/1或0/3小时）的患者（n = 4171）中，与推荐的算法相比，使用ANN导致中间组的规模相对减少9.2%（95%置信区间为4.4%至13.8%；从所有测试患者的24.5%至22.2%）。相比之下，使用逻辑回归并没有实质性地减少中间组的规模。机器学习算法允许采样的灵活性，并有可能改善急诊科胸痛患者的风险评估。
Computerized decision‐support tools may improve diagnosis of acute myocardial infarction (AMI) among patients presenting with chest pain at the emergency department (ED). The primary aim was to assess the predictive accuracy of machine learning algorithms based on paired high‐sensitivity cardiac troponin T (hs‐cTnT) concentrations with varying sampling times, age, and sex in order to rule in or out AMI. In this register‐based, cross‐sectional diagnostic study conducted retrospectively based on 5695 chest pain patients at 2 hospitals in Sweden 2013–2014 we used 5‐fold cross‐validation 200 times in order to compare the performance of an artificial neural network (ANN) with European guideline‐recommended 0/1‐ and 0/3‐hour algorithms for hs‐cTnT and with logistic regression without interaction terms. Primary outcome was the size of the intermediate risk group where AMI could not be ruled in or out, while holding the sensitivity (rule‐out) and specificity (rule‐in) constant across models. ANN and logistic regression had similar (95%) areas under the receiver operating characteristics curve. In patients (n = 4171) where the timing requirements (0/1 or 0/3 hour) for the sampling were met, using ANN led to a relative decrease of 9.2% (95% confidence interval 4.4% to 13.8%; from 24.5% to 22.2% of all tested patients) in the size of the intermediate group compared to the recommended algorithms. By contrast, using logistic regression did not substantially decrease the size of the intermediate group. Machine learning algorithms allow for flexibility in sampling and have the potential to improve risk assessment among chest pain patients at the ED.
登录
查看更多内容
DOI：
10.1016/s0140-6736(15)00391-8
发表时间：
2015-12-19
期刊：
Lancet (London, England)
影响因子：
--
作者：
Shah AS;Anand A;Sandoval Y;Lee KK;Smith SW;Adamson PD;Chapman AR;Langdon T;Sandeman D;Vaswani A;Strachan FE;Ferry A;Stirzaker AG;Reid A;Gray AJ;Collinson PO;McAllister DA;Apple FS;Newby DE;Mills NL;High-STEACS investigators
通讯作者：
High-STEACS investigators
影响因子：
4.4
作者：
Mokhtari, Arash;Lindahl, Bertil;Ekelund, Ulf
通讯作者：
Ekelund, Ulf
影响因子：
5.9
作者：
Hansen, Tobias Graversgaard;Pottegard, Anton;Lassen, Annmarie Touborg
通讯作者：
Lassen, Annmarie Touborg
影响因子：
39.3
作者：
Anand, Sonia S.;Islam, Shofiqul;Yusuf, Salim
通讯作者：
Yusuf, Salim
影响因子：
39.3
作者：
Thygesen, Kristian;Mair, Johannes;Jaffe, Allan S.
通讯作者：
Jaffe, Allan S.