Predicting serum phosphate levels in very preterm infants using machine learning - Takeaways - MDSpire

Predicting serum phosphate levels in very preterm infants using machine learning

  • By

  • Åsbjørn S. Westvik

  • Oliver Tomic

  • Charlotte Tscherning

  • Sissel J. Moltu

  • July 15, 2026

Share

  • 1

    Hypophosphatemia is common in very preterm infants and linked to serious complications, necessitating accurate phosphate level monitoring.

  • 2

    Machine learning models were developed to estimate serum phosphate levels using clinical and nutritional data from very preterm infants.

  • 3

    In a study of 120 infants, 33.6% developed hypophosphatemia in the first week, with SGA infants at higher risk.

  • 4

    Six predictors were consistently identified as significant for phosphate level estimation, with elastic net models performing best.

  • 5

    The study suggests that ML models could reduce unnecessary blood sampling in very preterm infants, pending further validation.

Original Source(s)

Predicting serum phosphate levels in very preterm infants using machine learning

  • by Åsbjørn S. Westvik, Oliver Tomic, Charlotte Tscherning, Sissel J. Moltu

  • July 15, 2026

Related Content

Underutilisation of rehabilitation for children with cancer in Gaza: the experience of the non-governmental organisation Children Not Numbers

Temporal trends in clinical and inflammatory features of Kawasaki disease across the COVID-19 pandemic: a single-center experience from Turkey

Exercise dose architecture as a precision-informed developmental framework in autism spectrum disorder: a narrative review of mechanisms, sex- and development-related heterogeneity, and clinical translation