A Stanford-led study shows AI can accurately predict which preterm infants will suffer complications based on metabolite patterns in newborn blood samples.
Key Details
- 1AI model used dried blood spot samples from 13,536 premature infants in California, born more than 10 weeks early.
- 2Validated the algorithm on 3,299 preterm infants from Ontario, Canada.
- 3AI identified six key blood measurements, forming a metabolic health index.
- 4Combined with clinical data (gestational age, birth weight, sex, Apgar scores), the index predicted four major complications (intestinal, eye, lung, brain) with over 85% accuracy.
- 5Research published in Science Translational Medicine (DOI: 10.1126/scitranslmed.adv4942).
Why It Matters
This approach enables much more precise risk stratification for preemies, paving the way for personalized neonatal care, early intervention, and potentially reduced morbidity through targeted resource use. It represents a significant advance in applying AI to high-impact clinical prediction using routine biomedical data.
Source
EurekAlert
Related News
•EurekAlert
UVA's $4M NIH Grant Backs AI to Optimize Cardiac MRI Therapies
UVA secures $4 million NIH funding to develop GenStrain, an AI model using MRI data to personalize cardiac pacing therapies.
•EurekAlert
Machine Learning Models Predict Cancer Risk with Routine Health Data
Researchers developed machine learning models using routinely collected health data to predict individual risk for 20 common cancers, outperforming traditional methods for several cancer types.
•EurekAlert
Duke Receives $25M to Lead National Trustworthy AI Hub for Dementia and Aging
Duke University has launched a $25 million national center to make AI for aging and dementia clinically robust and trustworthy.