Department of Anesthesiology, Complexo Hospitalario Universitario de Pontevedra, Pontevedra, PO, Spain.
Einstein (Sao Paulo). 2021 Oct 11;19:eAO6283. doi: 10.31744/einstein_journal/2021AO6283. eCollection 2021.
To explore an artificial intelligence approach based on gradient-boosted decision trees for prediction of all-cause mortality at an intensive care unit, comparing its performance to a recent logistic regression system in the literature, and a logistic regression model built on the same platform.
A gradient-boosted decision trees model and a logistic regression model were trained and tested with the Medical Information Mart for Intensive Care database. The 1-hour resolution physiological measurements of adult patients, collected during 5 hours in the intensive care unit, consisted of eight routine clinical parameters. The study addressed how the models learn to categorize patients to predict intensive care unit mortality or survival within 12 hours. The performance was evaluated with accuracy statistics and the area under the Receiver Operating Characteristic curve.
The gradient-boosted trees yielded an area under the Receiver Operating Characteristic curve of 0.89, compared to 0.806 for the logistic regression. The accuracy was 0.814 for the gradient-boosted trees, compared to 0.782 for the logistic regression. The diagnostic odds ratio was 17.823 for the gradient-boosted trees, compared to 9.254 for the logistic regression. The Cohen's kappa, F-measure, Matthews correlation coefficient, and markedness were higher for the gradient-boosted trees.
The discriminatory power of the gradient-boosted trees was excellent. The gradient-boosted trees outperformed the logistic regression regarding intensive care unit mortality prediction. The high diagnostic odds ratio and markedness values for the gradient-boosted trees are important in the context of the studied unbalanced dataset.
探索一种基于梯度提升决策树的人工智能方法，用于预测重症监护病房的全因死亡率，将其性能与文献中的最近逻辑回归系统以及基于同一平台构建的逻辑回归模型进行比较。
使用医疗信息集市重症监护数据库训练和测试梯度提升决策树模型和逻辑回归模型。成人患者在重症监护病房 5 小时内每 1 小时采集的生理测量值包括 8 个常规临床参数。该研究探讨了模型如何学习对患者进行分类以预测 12 小时内重症监护病房的死亡率或存活率。使用准确性统计数据和接收器操作特征曲线下的面积评估性能。
与逻辑回归相比，梯度提升树的接收器操作特征曲线下面积为 0.89，而逻辑回归为 0.806。梯度提升树的准确性为 0.814，而逻辑回归为 0.782。梯度提升树的诊断优势比为 17.823，而逻辑回归为 9.254。梯度提升树的科恩氏kappa、F 度量、马修斯相关系数和显著性更高。
梯度提升树的区分能力非常出色。与逻辑回归相比，梯度提升树在预测重症监护病房死亡率方面表现更好。在研究的不平衡数据集背景下，梯度提升树的高诊断优势比和显著性值非常重要。
