可解释的机器学习用于在重症监护室中停止治疗抗生素
Tariq A Dam1, Rogier P Schade2, Jurriaan E M de Steenwinkel3
1Department of Intensive Care Medicine, Center for Critical Care Computational Intelligence (C4I), Amsterdam Medical Data Science (AMDS), Amsterdam Public Health (APH), Amsterdam Cardiovascular Science (ACS), Amsterdam Institute for Infection and Immunity (AII), Amsterdam UMC, Vrije Universiteit, Amsterdam, the Netherlands; Pacmed, Amsterdam, the Netherlands.
在重症监护室 (ICU) 预测抗生素的重新开始是困难的. 较短的抗生素持续时间是重启抗生素的最强有力的预测因素,突出了基于数据的决策,以优化ICU中的抗生素使用.
科学领域:
- 重症监护医疗
- 临床药理学
- 卫生信息学
背景情况:
- 在ICU中确定最佳抗生素持续时间是具有挑战性的,平衡耐药性和感染风险.
- 长期使用抗生素会增加耐药性和副作用;过早停药可能导致感染复发.
- 可解释的机器学习有可能预测ICU患者的抗生素重新开始使用.
研究的目的:
- 开发和评估用于预测ICU患者的抗生素重新开始的机器学习模型.
- 在72小时内确定重启抗生素的关键预测因素.
主要方法:
- 在荷兰两所高等学术医院,从成年ICU患者收集回顾性数据.
- 包括监测数据,实验室结果,消毒策略,药物和文化作为预测因素.
- 训练后勤回归,轻GBM和自动预测模型来预测抗生素的重新启动.
主要成果:
- 分析了2486名患者和3645个抗生素疗程;19. 4%的患者重新开始服用抗生素.
- 最后一个抗生素疗程的较短持续时间是重新开始治疗的最重要的预测因素.
- 后勤回归获得了最好的表现 (AUROC 0. 675);在重新开始治疗组中,90天死亡率更高 (39. 8% vs 25. 0%).
结论:
- 预测ICU中的抗生素重启仍然具有挑战性,但可行.
- 较短的抗生素持续时间是关键预测因素,表明停止治疗的最佳机会.
- 频繁重复使用相同的抗生素和较高的死亡率强调了需要数据驱动的决策支持.
更多相关视频
10:50Design and Use of a Low Cost, Automated Morbidostat for Adaptive Evolution of Bacteria Under Antibiotic Drug Selection
Published on: September 27, 2016
04:09Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
Published on: October 10, 2018
相关概念视频
Antibiotic Selection
Clearance Models: Compartment Models
Steps in Outbreak Investigation
Pulmonary Tuberculosis V
Latent tuberculosis infection occurs when TB bacteria are present in a person's body, but are not causing illness or symptoms. It is not contagious, and preventive treatment is crucial to avoid the...
Healthcare Associated Infections I: Iatrogenic, Exogenic and Endogenic
HAIs significantly increase the cost of health care. Extended stays in healthcare institutions, increased disability, increased costs of medications, including specialized antibiotics, and prolonged recovery times add to the patient's expenses and the healthcare institution and funding bodies.
