Advancing Newborn Genetic Screening with Machine Learning
A groundbreaking study demonstrates how machine learning can standardize gene selection in newborn genetic screening, improving accuracy and public health outcomes.
Over the past decade, efforts to enhance newborn genetic screening have expanded significantly, driven by advancements in genomic technologies. The BabySeq Project, initiated more than ten years ago, pioneered the integration of genomic sequencing results into newborn care, providing valuable insights into genetic conditions from birth. Today, more than 30 international programs are exploring the use of genomic sequencing (NBSeq) to identify genetic disorders early in life. However, these programs often vary widely in the genes they select for screening, leading to inconsistencies that can impact clinical outcomes.
A recent study by researchers at Mass General Brigham introduces a novel, data-driven approach to refining gene selection in NBSeq programs using machine learning. Published in Genetics in Medicine, the research emphasizes the importance of choosing relevant genes based on scientific evidence, natural history of conditions, and treatment effectiveness. Co-senior author Nina Gold, MD, highlighted that thoughtful gene selection is crucial for maximizing the benefits of genomic screening in newborns.
The study analyzed 4,390 genes across 27 NBSeq initiatives, revealing that only about 1.7%—74 genes—were included in over 80% of these programs. Key factors influencing gene inclusion were their presence on the U.S. Recommended Uniform Screening Panel, robust data on natural disease progression, and strong evidence supporting treatment options. To help standardize gene selection, the team developed a machine learning model incorporating 13 predictive factors. This model accurately forecasts gene inclusion decisions among various programs and provides a ranked list that can adapt to new scientific evidence and regional health priorities.
This innovative approach aims to harmonize NBSeq programs worldwide, ensuring more consistent and informed decision-making. As Green from Mass General Brigham noted, such tools can improve the impact of genetic screening by aligning practices with the latest research and public health goals, ultimately benefiting early diagnosis and intervention in newborns.
For more details, see the full study by Thomas Minten et al. in Genetics in Medicine (2025). Source: https://medicalxpress.com/news/2025-05-machine-newborn-genetic-screening.html
Stay Updated with Mia's Feed
Get the latest health & wellness insights delivered straight to your inbox.
Related Articles
Majority of Children with Pediatric Asthma Meet Diagnostic Guidelines
A new study reveals that around 60% of children diagnosed with asthma meet international guideline criteria, emphasizing the importance of objective testing for accurate diagnosis and treatment.
Hidden Risks of Bowel Cancer in Common Colorectal Polyps
New research reveals that individuals with both adenomas and serrated polyps have a significantly higher risk of developing bowel cancer. Early detection and regular colonoscopies are crucial for prevention.
Advancements in Bird Flu Vaccine and Insights from Organoid Research Reveal Impact of H5N1 on Airways
Researchers develop a promising bird flu vaccine and use organoid models to understand how H5N1 damages airway tissues, opening new avenues for treatment and prevention.
