Comparison of Supervised Machine Learning Algorithms for Classifying of Home Discharge Possibility in Convalescent Stroke Patients: A Secondary Analysis
Comparison of Supervised Machine Learning Algorithms for Classifying of Home Discharge Possibility in Convalescent Stroke Patients: A Secondary Analysis
复制标题
DOI：
10.1016/j.jstrokecerebrovasdis.2021.106011
复制
发表时间：
2021-07-26
影响因子：
2.5
通讯作者：
Araki, Osamu
中科院分区：
文献类型：
--
作者：
Imura, Takeshi;Toda, Haruki;Araki, Osamu
目的：在脑卒中康复过程中，对出院回家的可能性进行分类，以支持决策是很重要的。已经有几项关于监督机器学习算法的研究，但只有少数研究比较了基于相同数据集的不同算法对家庭放电可能性分类的性能。因此，我们的目的是评估五种监督机器学习算法对中风患者出院可能性的分类。材料和方法：这是一个二次分析的基础上，从我们的机构数据库中的481例脑卒中患者的数据。通过构建基于同一数据集的分类系统，比较了决策树（DT）、线性判别分析（LDA）、k-近邻（k-NN）、支持向量机（SVM）和随机森林（RF）等5种有监督机器学习算法的分类效果。包括分类准确性、曲线下面积（AUC）和F1评分（精确度和召回率的加权平均值）在内的几个参数用于模型评估。结果：k-NN模型的分类准确率最高（84.0%），AUC（0.88）和F1评分（87.8）适中。SVM模型还显示出高的分类准确性（82.6%）沿着，具有最高的AUC（0.91）、灵敏度（94.4）、阴性预测值（87.5）和阴性似然比（0.088）。DT、LDA和RF模型具有较高的分类准确度（> 79.9%），AUC（> 0.84）和F1评分（> 83.8）适中。结论：在模型性能方面，k-NN和SVM似乎是分类中风患者出院可能性的最佳候选算法。
Objectives: Classifying the possibility of home discharge is important during stroke rehabilitation to support decision-making. There have been several studies on supervised machine learning algorithms, but only a few have compared the performance of different algorithms based on the same dataset for the classification of home discharge possibility. Therefore, we aimed to evaluate five supervised machine learning algorithms for the classification of home discharge possibility in stroke patients. Materials and Methods: This was a secondary analysis based on the data of 481 stroke patients from the database of our institution. Five models developed by supervised machine learning algorithms, including deci-sion tree (DT), linear discriminant analysis (LDA), k-nearest neighbors (k-NN), support vector machine (SVM), and ran-dom forest (RF) were compared by constructing a classifica-tion system based on the same dataset. Several parameters including classification accuracy, area under the curve (AUC), and F1 score (a weighted average of precision and recall) were used for model evaluation. Results: The k-NN model had the best classification accuracy (84.0%) with a moderate AUC (0.88) and F1 score (87.8). The SVM model also showed high classification accuracy (82.6%) along with the highest AUC (0.91), sensitivity (94.4), negative predictive value (87.5), and negative likelihood ratio (0.088). The DT, LDA, and RF models had high classification accuracies (> 79.9%) with moderate AUCs (> 0.84) and F1 scores (> 83.8). Conclusions: Regarding model performance, the k -NN and SVM seemed the best candidate algorithms for clas-sifying the possibility of home discharge in stroke patients.