Kotter Adam, Abdelrahman Samir, Wan Yi-Ki Jacob, Madaras-Kelly Karl, Morgan Keaton L, Kan Chin Fung Kelvin, Del Fiol Guilherme
Department of Biomedical Informatics, University of Utah, Salt Lake City, UT 84108, USA.
College of Pharmacy, Idaho State University, Meridian, ID 83209, USA.
Diagnostics (Basel). 2025 Jan 28;15(3):307. doi: 10.3390/diagnostics15030307.
: Sepsis is a life-threatening response to infection and a major cause of hospital mortality. Machine learning (ML) models have demonstrated better sepsis prediction performance than integer risk scores but are less widely used in clinical settings, in part due to lower interpretability. This study aimed to improve the interpretability of an ML-based model without reducing its performance in non-ICU sepsis prediction. : A logistic regression model was trained to predict sepsis onset and then converted into a more interpretable integer point system, STEWS, using its regression coefficients. We compared STEWS with the logistic regression model using PPV at 90% sensitivity. : STEWS was significantly equivalent to logistic regression using the two one-sided tests procedure (0.051 vs. 0.051; = 0.004). : STEWS demonstrated equivalent performance to a comparable logistic regression model for non-ICU sepsis prediction, suggesting that converting ML models into more interpretable forms does not necessarily reduce predictive power.
脓毒症是对感染的一种危及生命的反应，也是医院死亡率的主要原因。机器学习（ML）模型在脓毒症预测方面表现出比整数风险评分更好的性能，但在临床环境中的应用并不广泛，部分原因是其可解释性较低。本研究旨在提高基于ML的模型的可解释性，同时不降低其在非ICU脓毒症预测中的性能。：训练一个逻辑回归模型来预测脓毒症的发作，然后利用其回归系数将其转换为一个更具可解释性的整数评分系统，即STEWS。我们使用90%灵敏度下的阳性预测值（PPV）将STEWS与逻辑回归模型进行比较。：使用双侧检验程序，STEWS与逻辑回归模型显著等效（0.051对0.051；Z = 0.004）。：对于非ICU脓毒症预测，STEWS表现出与类似逻辑回归模型等效的性能，这表明将ML模型转换为更具可解释性的形式不一定会降低预测能力。
