Current Trends and Focal Points of Machine Learning in The Domain of Down Syndrome: A Bibliometric Study

Abstract

The purpose of this study is to analyze major areas and trends in down syndrome machine learning (ML). We gathered research literature on machine learning in the field of Down syndrome from the Science Citation Index Expanded in the Web of Science. The evaluation of this data took into account publication years, countries/regions, journals, institutions, citations, and keywords. VOSviewer and CiteSpace on the online analytic platform developed co-occurrence network graphs. We selected 73 relevant research publications from various countries for our analysis. Every year since 2007, relevant articles have increased dramatically. The USA (n = 26), India (n = 9), and China (n = 8) had the most publications, accounting for 76%. IEEE Access (n = 4) and PLOS ONE (n = 3) published most. Universities with the most publications are University of California Davis (10), Vanderbilt (7), and Children's National Medical Center (6). We found that neurodevelopmental and cognitive characteristics, comorbidities, and diagnoses dominate current research by keyword analysis. The study found important areas and trends in using machine learning to diagnose Down syndrome, indicating that this technology could greatly improve early detection. The study identified targeted interventions for Down syndrome patients, which could significantly improve their quality of life.