National Institute for Health Research Health Protection Research Unit in Healthcare Associated Infections and Antimicrobial Resistance, Imperial College London, London, UK; Université de Paris, IAME, INSERM, F-75018 Paris, France.
Université de Paris, Laboratoire de Parasitologie-Mycologie, Groupe Hospitalier Saint-Louis-Lariboisière-Fernand-Widal, Assistance Publique-Hôpitaux de Paris (AP-HP), Paris, France.
Clin Microbiol Infect. 2020 Oct;26(10):1300-1309. doi: 10.1016/j.cmi.2020.02.006. Epub 2020 Feb 12.
Machine learning (ML) allows the analysis of complex and large data sets and has the potential to improve health care. The clinical microbiology laboratory, at the interface of clinical practice and diagnostics, is of special interest for the development of ML systems.
This narrative review aims to explore the current use of ML In clinical microbiology.
References for this review were identified through searches of MEDLINE/PubMed, EMBASE, Google Scholar, biorXiv, arXiV, ACM Digital Library and IEEE Xplore Digital Library up to November 2019.
We found 97 ML systems aiming to assist clinical microbiologists. Overall, 82 ML systems (85%) targeted bacterial infections, 11 (11%) parasitic infections, nine (9%) viral infections and three (3%) fungal infections. Forty ML systems (41%) focused on microorganism detection, identification and quantification, 36 (37%) evaluated antimicrobial susceptibility, and 21 (22%) targeted the diagnosis, disease classification and prediction of clinical outcomes. The ML systems used very diverse data sources: 21 (22%) used genomic data of microorganisms, 19 (20%) microbiota data obtained by metagenomic sequencing, 19 (20%) analysed microscopic images, 17 (18%) spectroscopy data, eight (8%) targeted gene sequencing, six (6%) volatile organic compounds, four (4%) photographs of bacterial colonies, four (4%) transcriptome data, three (3%) protein structure, and three (3%) clinical data. Most systems used data from high-income countries (n = 71, 73%) but a significant number used data from low- and middle-income countries (n = 36, 37%). Performance measures were reported for the 97 ML systems, but no article described their use in clinical practice or reported impact on processes or clinical outcomes.
In clinical microbiology, ML has been used with various data sources and diverse practical applications. The evaluation and implementation processes represent the main gap in existing ML systems, requiring a focus on their interpretability and potential integration into real-world settings.
机器学习（ML）允许对复杂和大型数据集进行分析，并且有可能改善医疗保健。临床微生物学实验室处于临床实践和诊断的交界处，是开发 ML 系统的特别关注点。
本叙述性评论旨在探讨 ML 在临床微生物学中的当前应用。
通过搜索 MEDLINE/PubMed、EMBASE、Google Scholar、biorXiv、arXiV、ACM 数字图书馆和 IEEE Xplore 数字图书馆，截至 2019 年 11 月，确定了这篇综述的参考文献。
我们发现了 97 个旨在协助临床微生物学家的 ML 系统。总体而言，82 个 ML 系统（85%）针对细菌感染，11 个（11%）针对寄生虫感染，9 个（9%）针对病毒感染，3 个（3%）针对真菌感染。40 个 ML 系统（41%）专注于微生物检测、鉴定和定量，36 个（37%）评估抗菌药物敏感性，21 个（22%）针对诊断、疾病分类和临床结果预测。ML 系统使用了非常多样化的数据来源：21 个（22%）使用微生物基因组数据，19 个（20%）使用宏基因组测序获得的微生物群落数据，19 个（20%）分析显微镜图像，17 个（18%）光谱数据，8 个（8%）针对基因测序，6 个（6%）挥发性有机化合物，4 个（4%）细菌菌落照片，4 个（4%）转录组数据，3 个（3%）蛋白质结构，3 个（3%）临床数据。大多数系统使用来自高收入国家的数据（n=71，73%），但相当数量的系统使用来自低收入和中等收入国家的数据（n=36，37%）。报道了 97 个 ML 系统的性能指标，但没有一篇文章描述了它们在临床实践中的使用情况，也没有报道对流程或临床结果的影响。
在临床微生物学中，已经使用了各种数据来源和不同的实际应用。评估和实施过程是现有 ML 系统的主要差距，需要关注其可解释性和潜在的实际应用。
