Opportunities and challenges in machine learning‐based newborn screening—A systematic literature review

Abstract

The development and continuous optimization of newborn screening (NBS) programs remains an important and challenging task due to the low prevalence of screened diseases and high sensitivity requirements for screening methods. Recently, different machine learning (ML) methods have been applied to support NBS. However, most studies only focus on single diseases or specific ML techniques making it difficult to draw conclusions on which methods are best to implement. Therefore, we performed a systematic literature review of peer‐reviewed publications on ML‐based NBS methods. Overall, 125 related papers, published in the past two decades, were collected for the study, and 17 met the inclusion criteria. We analyzed the opportunities and challenges of ML methods for NBS including data preprocessing, classification models and pattern recognition methods based on their underlying approaches, data requirements, interpretability on a modular level, and performance. In general, ML methods have the potential to reduce the false positive rate and identify so far unknown metabolic patterns within NBS data. Our analysis revealed, that, among the presented, logistic regression analysis and support vector machines seem to be valuable candidates for NBS. However, due to the variety of diseases and methods, a general recommendation for a single method in NBS is not possible. Instead, these methods should be further investigated and compared to other approaches in comprehensive studies as they show promising results in NBS applications.

Newborn screening for neurodevelopmental diseases: Are we there yet?

“…Recent studies using advanced informatics and data mining tools such as machine learning have demonstrated the potential of such approaches (Baker et al, 2019;James et al, 2020;Robertson et al, 2022;Seo et al, 2022;Tan et al, 2020;Zaunseder et al, 2022).…”

“…While reanalysis of the data can potentially be automated, information technology systems would need to be developed to enable long term tracking of participants. Recent studies using advanced informatics and data mining tools such as machine learning have demonstrated the potential of such approaches (Baker et al, 2019; James et al, 2020; Robertson et al, 2022; Seo et al, 2022; Tan et al, 2020; Zaunseder et al, 2022).…”

How longitudinal observational studies can guide screening strategy for rare diseases

“…Utilizing data from multiple sources, this study highlights the need for accurate and severity‐adjusted case definitions and the importance of well‐characterized longitudinally followed patient cohorts. In the future, multi‐omics data and artificial intelligence‐supported diagnostic pathways might help to overcome the current limitations of case definition and risk stratification 77 …”

“…In the future, multi-omics data and artificial intelligencesupported diagnostic pathways might help to overcome the current limitations of case definition and risk stratification. 77…”

Machine Learning Methods Improve Specificity in Newborn Screening for Isovaleric Aciduria

“…In the context of NBS, a variety of supervised ML methods have been applied to NBS data to predict whether or not a newborn suffers from a condition. The methods and their results were summarized in a recent review [ 15 ]. They enabled a reduction of false positive rates and identification of so far unknown metabolic patterns by relying on complex feature combinations instead of predefined single cut-off values.…”