Through the utilization of a machine-learning algorithm to analyze the electronic health records of mothers and their babies, researchers can anticipate the outcomes of at-risk infants during their initial two months of life. This novel approach empowers medical professionals to classify, even prior to delivery, which newborns may experience complications related to premature birth.
The Stanford School of Medicine has recently developed a method that was described in a study published in Science Translational Medicine. According to senior study author Nima Aghaeepour, Ph.D., an associate professor of anesthesiology, perioperative and pain medicine, and pediatrics, the method shifts the focus toward the individual health factors of newborns rather than solely relying on their gestational age. The study’s lead authors include postdoctoral scholar Davide De Francesco, Ph.D., and Jonathan Reiss, MD, an instructor in pediatrics.
Premature birth, traditionally defined as birth occurring at least three weeks early, is associated with various complications that affect babies’ lungs, vision, brain, hearing, and digestive systems. However, the timing of birth only partially predicts an individual infant’s health outcomes, as some babies born quite early may not develop complications while others born at the same stage of pregnancy may become severely ill or even die.
According to Aghaeepour, preterm birth is the leading cause of death in children below five globally, and effective solutions have yet to be found. By prioritizing research efforts toward predicting the health of preterm babies, optimal care can be provided to them.
Most complications arise from premature birth surface days or even weeks after birth, potentially causing significant harm to the newborn’s health during that time. Therefore, identifying babies at risk could facilitate the implementation of preventive measures.
According to David Stevenson, a co-author of the study, although the baby’s condition is the primary focus for treatment decisions in neonatology, the maternal health record can provide valuable insights into the unique impact of the maternal environment on each baby’s development trajectory.