当前位置: X-MOL 学术Artif. Intell. Med. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
The Physiological Deep Learner: First application of multitask deep learning to predict hypotension in critically ill patients
Artificial Intelligence in Medicine ( IF 6.1 ) Pub Date : 2021-05-29 , DOI: 10.1016/j.artmed.2021.102118
Ményssa Cherifa 1 , Yannet Interian 2 , Alice Blet 3 , Matthieu Resche-Rigon 4 , Romain Pirracchio 5
Affiliation  

Critical care clinicians are trained to analyze simultaneously multiple physiological parameters to predict critical conditions such as hemodynamic instability. We developed the Multi-task Learning Physiological Deep Learner (MTL-PDL), a deep learning algorithm that predicts simultaneously the mean arterial pressure (MAP) and the heart rate (HR).

In an external validation dataset, our model exhibited very good calibration: R2 of 0.747 (95% confidence interval, 0.692 to 0.794) and 0.850 (0.815 to 0.879) for respectively, MAP and HR prediction 60-minutes ahead of time. For acute hypotensive episodes defined as a MAP below 65 mmHg for 5 min, our MTL-PDL reached a predictive value of 90% for patients at very high risk (predicted MAP ≤ 60 mmHg) and 2‰ for patients at low risk (predicted MAP >70 mmHg).

Based on its excellent prediction performance, the Physiological Deep Learner has the potential to help the clinician proactively adjust the treatment in order to avoid hypotensive episodes and end-organ hypoperfusion.



中文翻译:

生理深度学习器:首次应用多任务深度学习来预测危重患者的低血压

重症监护临床医生经过培训,可以同时分析多个生理参数,以预测诸如血流动力学不稳定等危急情况。我们开发了多任务学习生理深度学习器 (MTL-PDL),这是一种深度学习算法,可同时预测平均动脉压 (MAP) 和心率 (HR)。

在外部验证数据集中,我们的模型表现出非常好的校准:对于提前 60 分钟的 MAP 和 HR 预测,R 2 分别为 0.747(95% 置信区间,0.692 至 0.794)和 0.850(0.815 至 0.879)。对于定义为 MAP 低于 65 mmHg 持续 5 分钟的急性低血压发作,我们的 MTL-PDL 对极高风险(预测 MAP ≤ 60 mmHg)患者的预测值达到 90%,对低风险患者(预测 MAP)达到 2‰ >70 毫米汞柱)。

基于其出色的预测性能,生理深度学习器有可能帮助临床医生主动调整治疗,以避免低血压发作和终末器官灌注不足。

更新日期:2021-06-11
down
wechat
bug