Hoodbhoy Zahra, Jiwani Uswa, Sattar Saima, Salam Rehana, Hasan Babar, Das Jai K
Department of Pediatrics and Child Health at the Aga Khan University, Karachi, Pakistan.
Front Artif Intell. 2021 Jul 8;4:708365. doi: 10.3389/frai.2021.708365. eCollection 2021.
With the dearth of trained care providers to diagnose congenital heart disease (CHD) and a surge in machine learning (ML) models, this review aims to estimate the diagnostic accuracy of such models for detecting CHD. A comprehensive literature search in the PubMed, CINAHL, Wiley Cochrane Library, and Web of Science databases was performed. Studies that reported the diagnostic ability of ML for the detection of CHD compared to the reference standard were included. Risk of bias assessment was performed using Quality Assessment for Diagnostic Accuracy Studies-2 tool. The sensitivity and specificity results from the studies were used to generate the hierarchical Summary ROC (HSROC) curve. We included 16 studies (1217 participants) that used ML algorithm to diagnose CHD. Neural networks were used in seven studies with overall sensitivity of 90.9% (95% CI 85.2-94.5%) and specificity was 92.7% (95% CI 86.4-96.2%). Other ML models included ensemble methods, deep learning and clustering techniques but did not have sufficient number of studies for a meta-analysis. Majority (=11, 69%) of studies had a high risk of patient selection bias, unclear bias on index test (=9, 56%) and flow and timing (=12, 75%) while low risk of bias was reported for the reference standard (=10, 62%). ML models such as neural networks have the potential to diagnose CHD accurately without the need for trained personnel. The heterogeneity of the diagnostic modalities used to train these models and the heterogeneity of the CHD diagnoses included between the studies is a major limitation.
由于缺乏经过培训的护理人员来诊断先天性心脏病（CHD），且机器学习（ML）模型激增，本综述旨在评估此类模型检测CHD的诊断准确性。我们在PubMed、CINAHL、Wiley Cochrane图书馆和Web of Science数据库中进行了全面的文献检索。纳入了那些报告了与参考标准相比ML检测CHD诊断能力的研究。使用诊断准确性研究质量评估-2工具进行偏倚风险评估。研究中的敏感性和特异性结果用于生成分层汇总ROC（HSROC）曲线。我们纳入了16项研究（1217名参与者），这些研究使用ML算法诊断CHD。七项研究使用了神经网络，总体敏感性为90.9%（95%CI 85.2-94.5%），特异性为92.7%（95%CI 86.4-96.2%）。其他ML模型包括集成方法、深度学习和聚类技术，但没有足够数量的研究进行荟萃分析。大多数研究（=11，69%）存在患者选择偏倚的高风险，索引测试（=9，56%）以及流程和时间方面（=12，75%）的偏倚不明确，而参考标准的偏倚风险较低（=10，62%）。神经网络等ML模型有潜力在无需训练有素人员的情况下准确诊断CHD。用于训练这些模型的诊断方式的异质性以及研究中纳入的CHD诊断的异质性是一个主要限制。
