Graduate Institute of Biomedical Informatics, College of Medical Science and Technology, Taipei Medical University, Taipei, Taiwan; International Center for Health Information Technology, Taipei Medical University, Taipei, Taiwan.
Department of Biological Sciences, National Sun Yat-sen University, Gushan District, Kaohsiung City, 804, Taiwan.
Comput Methods Programs Biomed. 2019 Mar;170:1-9. doi: 10.1016/j.cmpb.2018.12.027. Epub 2018 Dec 26.
Sepsis is a common and major health crisis in hospitals globally. An innovative and feasible tool for predicting sepsis remains elusive. However, early and accurate prediction of sepsis could help physicians with proper treatments and minimize the diagnostic uncertainty. Machine learning models could help to identify potential clinical variables and provide higher performance than existing traditional low-performance models. We therefore performed a meta-analysis of observational studies to quantify the performance of a machine learning model to predict sepsis.
A comprehensive literature search was conducted through the electronic database (e.g. PubMed, Scopus, Google Scholar, EMBASE, etc.) between January 1, 2000, and March 1, 2018. All the studies published in English and reporting the sepsis prediction using machine learning algorithms were considered in this study. Two authors independently extracted valuable information from the included studies. Inclusion and exclusion of studies were based on the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines.
A total of 7 out of 135 studies met all of our inclusion criteria. For machine learning models, the pooled area under receiving operating curve (SAUROC) for predicting sepsis onset 3 to 4 h before, was 0.89 (95%CI: 0.86-0.92); sensitivity 0.81 (95%CI:0.80-0.81), and specificity 0.72 (95%CI:0.72-0.72) whereas the pooled SAUROC for SIRS, MEWS, and SOFA was 0.70, 0.50, and 0.78. Additionally, diagnostic odd ratio for machine learning, SIRS, MEWS, and SOFA was 15.17 (95%CI: 9.51-24.20), 3.23 (95%CI: 1.52-6.87), 31.99 (95% CI: 1.54-666.74), and 3.75(95%CI: 2.06-6.83).
Our study findings suggest that the machine learning approach had a better performance than the existing sepsis scoring systems in predicting sepsis.
脓毒症是全球医院常见且严重的健康危机。一种创新且可行的脓毒症预测工具仍难以捉摸。然而，早期准确地预测脓毒症可以帮助医生进行适当的治疗，并最大限度地减少诊断的不确定性。机器学习模型可以帮助识别潜在的临床变量，并提供比现有传统低性能模型更高的性能。因此，我们进行了一项观察性研究的荟萃分析，以量化机器学习模型预测脓毒症的性能。
通过电子数据库（如 PubMed、Scopus、Google Scholar、EMBASE 等）全面检索 2000 年 1 月 1 日至 2018 年 3 月 1 日期间发表的文献。本研究考虑了所有使用机器学习算法报告脓毒症预测的英文发表研究。两位作者独立从纳入研究中提取有价值的信息。研究的纳入和排除是基于系统评价和荟萃分析的首选报告项目（PRISMA）指南。
共有 7 项研究符合我们的所有纳入标准。对于机器学习模型，预测 3-4 小时前发生脓毒症的曲线下面积（SAUROC）为 0.89（95%CI：0.86-0.92）；敏感性为 0.81（95%CI：0.80-0.81），特异性为 0.72（95%CI：0.72-0.72），而 SIRS、MEWS 和 SOFA 的 SAUROC 分别为 0.70、0.50 和 0.78。此外，机器学习、SIRS、MEWS 和 SOFA 的诊断比值比为 15.17（95%CI：9.51-24.20）、3.23（95%CI：1.52-6.87）、31.99（95%CI：1.54-666.74）和 3.75（95%CI：2.06-6.83）。
我们的研究结果表明，与现有的脓毒症评分系统相比，机器学习方法在预测脓毒症方面具有更好的性能。
