AI预测早至48小时的儿科败血症,从而能够及时采取干预措施。 AI predicts pediatric sepsis up to 48 hours early, enabling timely intervention.

根据JAMA小儿科公布的一项研究,人工智能现在可以预测在出现症状前48小时以下的儿童患败血症的情况。 Artificial intelligence can now predict sepsis in children up to 48 hours before symptoms appear, according to a study published in JAMA Pediatrics. 研究人员分析了从急诊部门对五个保健系统进行头四个小时的访问中得出的例行电子健康记录数据,在器官衰竭发生之前确定高风险儿童。 Researchers analyzed routine electronic health record data from the first four hours of emergency department visits across five health systems, identifying children at high risk before organ failure occurs. 采用新的凤凰塞普西标准验证的AI模型显示,在区分高危病人和低风险病人方面非常准确,减少了不必要治疗的机会。 The AI models, validated using the new Phoenix Sepsis Criteria, showed strong accuracy in distinguishing at-risk patients from low-risk ones, reducing the chance of unnecessary treatments. 该技术由西北大学和Lurie儿童医院开发,是朝着早期拯救生命的干预措施和儿科败血症的精密药物迈出的重要一步,而儿科败血症是全世界儿童死亡率的一个主要原因。 The technology, developed by Northwestern University and Lurie Children's Hospital, represents a major step toward early, life-saving interventions and precision medicine in pediatric sepsis, a leading cause of child mortality worldwide.