The Artificial Intelligence Clinician learns optimal treatment strategies for sepsis in intensive care
The Artificial Intelligence Clinician learns optimal treatment strategies for sepsis in intensive care
复制标题
DOI：
10.1038/s41591-018-0213-5
复制
发表时间：
2018-11-01
期刊：
影响因子：
82.9
通讯作者：
Faisal, A. Aldo
中科院分区：
文献类型：
--
作者：
Komorowski, Matthieu;Celi, Leoa;Faisal, A. Aldo
脓毒症是全球第三大死亡原因，也是医院死亡的主要原因（1-3），但最佳治疗策略仍不确定。特别是，有证据表明，目前静脉输液和血管加压药的给药方法并不理想，可能会对部分患者造成伤害（1，4 -6）。为了解决这个顺序决策问题，我们开发了一个强化学习代理，人工智能（AI）临床医生，它从大量的患者数据中提取隐含的知识，这些数据超过了人类临床医生一生的经验，并通过分析无数（大多数是次优）治疗决策来学习最佳治疗。我们证明，AI临床医生选择的治疗价值平均可靠地高于人类临床医生。在一个独立于训练数据的大型验证队列中，临床医生的实际剂量与AI决策相匹配的患者的死亡率最低。我们的模型为败血症提供了个性化和临床可解释的治疗决策，可以改善患者的预后。
Sepsis is the third leading cause of death worldwide and the main cause of mortality in hospitals(1-3), but the best treatment strategy remains uncertain. In particular, evidence suggests that current practices in the administration of intravenous fluids and vasopressors are suboptimal and likely induce harm in a proportion of patients(1,4-6). To tackle this sequential decision-making problem, we developed a reinforcement learning agent, the Artificial Intelligence (AI) Clinician, which extracted implicit knowledge from an amount of patient data that exceeds by many-fold the life-time experience of human clinicians and learned optimal treatment by analyzing a myriad of (mostly suboptimal) treatment decisions. We demonstrate that the value of the AI Clinician's selected treatment is on average reliably higher than human clinicians. In a large validation cohort independent of the training data, mortality was lowest in patients for whom clinicians' actual doses matched the AI decisions. Our model provides individualized and clinically interpretable treatment decisions for sepsis that could improve patient outcomes.