Diagnostic Accuracy of Machine Learning Models to Identify Congenital Heart Disease: A Meta-Analysis

Abstract

Background: 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.Methods: 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.Results: 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 (n=11, 69%) of studies had a high risk of patient selection bias, unclear bias on index test (n=9, 56%) and flow and timing (n=12, 75%) while low risk of bias was reported for the reference standard (n=10, 62%).Conclusion: 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.

[Retracted] Machine Learning Algorithms Are Applied in Nanomaterial Properties for Nanosecurity

“…Therefore, as a result of the preceding conclusion, it is far necessary that simulation platforms that are equipped to creating large volumes of information for academic tool analysis styles have a connection to nanoma-terials' in vivo affects. I discovered that the combination of in vitro data and nanomaterial descriptors provides a more accurate representation of in vivo reactions than descriptors alone [2,20]. According to some researchers, it is possible that expected in vitro responses can be used to predict in vivo features in this manner, as an extension of this notion.…”

“…There have been various critiques of the application of ML in nanotoxicology during the previous decade [3,15,20], and we include a few remarkable examples of research that have been reported below. These were selected in order to demonstrate the breadth of tool expertise gained by techniques of modelling nanomaterial habitats in some software packages, among other activities.…”

Using Innovative Machine Learning Methods to Screen and Identify Predictors of Congenital Heart Diseases

“…ML approaches have also been utilized to identify and predict CHDs. Most of these studies tried to identify CHDs based on heart sounds, images, ECG, and genetic makeup (7,8). Three previous studies predicted the occurrence of CHDs based on a limited number of self-reported factors from questionnaires (9-11).…”

A review on the impact of deep learning in the identification of atrial septal defect and a comparative study on the algorithms employed in the imaging modalities

“…The diverseness of the diagnostic techniques used to teach those models and the diversification of the CHD reputation blanketed among the research is a chief constraint. [20]. Different machine learning algorithms and deep learning and have been carried out to examine the outcomes and evaluation of the UCI Machine Learning Heart Disease dataset by Rohit Bharti et al [22].…”