{
  "abstract": "Recent technological advancements have enabled clinicians to integrate data into predictive models, potentially transforming early diagnosis in neonatology. Using predictive models to detect neonatal conditions such as sepsis, necrotising enterocolitis and mortality could lead to earlier interventions and improved outcomes.1 For the diagnosis of hypoxic ischaemic encephalopathy (HIE), early detection is critical as therapeutic hypothermia (TH) is typically initiated within the first 6 hours after birth. However, delays in transferring infants born at outlying hospitals to facilities equipped for TH can postpone the initiation of treatment.2 Using machine learning (ML) to predict HIE sooner could streamline the transfer process and enable timely intervention.",
  "authors": [
    {
      "affiliations": [
        "Neonatology, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA"
      ],
      "name": "Kristyn S Beam"
    }
  ],
  "title": "Validation of a machine learning algorithm for predicting hypoxic ischaemic encephalopathy",
  "uid": "a75260a9-de26-5b4a-894d-e579032b0dd8"
}