常规血液测试的机器学习在几天内预测脊髓损伤结果,有助于早期治疗决策。 Machine learning of routine blood tests predicts spinal cord injury outcomes within days, aiding early treatment decisions.

滑铁卢大学的一项研究发现,通过机器学习分析,常规血液测试可以预测脊髓损伤的严重程度和结果,包括存活情况,甚至在神经检查之前,在录取后几天内进行。 A University of Waterloo study finds that routine blood tests, analyzed with machine learning, can predict spinal cord injury severity and outcomes—including survival—within days of admission, even before neurological exams. 研究人员利用2 600多名美国病人提供的数据,查明了在受伤后头三周收集的电解液和免疫细胞等共同血标记中的模式。 Using data from over 2,600 U.S. patients, researchers identified patterns in common blood markers like electrolytes and immune cells collected in the first three weeks post-injury. 这些模型在早期就准确预测了结果,并随着时间的推移提高了准确性,提供了一个低成本、可广泛使用的工具,用以指导特护治疗和资源使用。 The models accurately forecasted outcomes early on, with accuracy improving over time, offering a low-cost, widely available tool to guide treatment and resource use in intensive care. 这一方法可以加强全球脊髓损伤患者的早期预测和个性化护理。 The approach could enhance early prognosis and personalized care for spinal cord injury patients globally.