Centre for Health Informatics, Australian Institute of Health Innovation, Macquarie University, Sydney, NSW 2109, Australia.
Faculty of Medicine, Universitas Indonesia, Jakarta, DKI Jakarta 10430, Indonesia.
J Am Med Inform Assoc. 2023 Nov 17;30(12):2050-2063. doi: 10.1093/jamia/ocad180.
This study aims to summarize the research literature evaluating machine learning (ML)-based clinical decision support (CDS) systems in healthcare settings.
We conducted a review in accordance with the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta Analyses extension for Scoping Review). Four databases, including PubMed, Medline, Embase, and Scopus were searched for studies published from January 2016 to April 2021 evaluating the use of ML-based CDS in clinical settings. We extracted the study design, care setting, clinical task, CDS task, and ML method. The level of CDS autonomy was examined using a previously published 3-level classification based on the division of clinical tasks between the clinician and CDS; effects on decision-making, care delivery, and patient outcomes were summarized.
Thirty-two studies evaluating the use of ML-based CDS in clinical settings were identified. All were undertaken in developed countries and largely in secondary and tertiary care settings. The most common clinical tasks supported by ML-based CDS were image recognition and interpretation (n = 12) and risk assessment (n = 9). The majority of studies examined assistive CDS (n = 23) which required clinicians to confirm or approve CDS recommendations for risk assessment in sepsis and for interpreting cancerous lesions in colonoscopy. Effects on decision-making, care delivery, and patient outcomes were mixed.
ML-based CDS are being evaluated in many clinical areas. There remain many opportunities to apply and evaluate effects of ML-based CDS on decision-making, care delivery, and patient outcomes, particularly in resource-constrained settings.
本研究旨在总结评估医疗保健环境中基于机器学习（ML）的临床决策支持（CDS）系统的研究文献。
我们按照 PRISMA-ScR（系统评价和荟萃分析扩展的首选报告项目用于范围综述）进行了综述。检索了包括 PubMed、Medline、Embase 和 Scopus 在内的四个数据库，以评估 2016 年 1 月至 2021 年 4 月期间在临床环境中使用基于 ML 的 CDS 的研究。我们提取了研究设计、护理环境、临床任务、CDS 任务和 ML 方法。使用先前发表的基于临床任务在临床医生和 CDS 之间划分的 3 级分类来检查 CDS 自主性水平；总结了对决策制定、护理提供和患者结局的影响。
确定了 32 项评估医疗保健环境中基于 ML 的 CDS 使用情况的研究。所有研究均在发达国家进行，主要在二级和三级护理环境中进行。基于 ML 的 CDS 支持的最常见临床任务是图像识别和解释（n = 12）和风险评估（n = 9）。大多数研究检查了辅助性 CDS（n = 23），这些研究需要临床医生确认或批准 CDS 对败血症风险评估和结肠镜检查中癌症病变的建议。对决策制定、护理提供和患者结局的影响是混合的。
基于 ML 的 CDS 正在许多临床领域进行评估。在决策制定、护理提供和患者结局方面，特别是在资源有限的环境中，仍然有许多机会应用和评估基于 ML 的 CDS 的效果。
