英国DeepMind公司Shakir Mohamed和Joseph R. Ledsam研究组宣布他们研制了一个临床应用的方法，以持续预测急性肾损伤。相关论文于2019年8月1日发表于国际顶尖学术期刊《自然》上。

研究组开发了一种深度学习方法，用于对患者未来病情恶化的持续风险预测。该方法建立在近期工作的基础上，这些工作从电子健康记录中建模不良事件，并以急性肾损伤这种常见且可能危及生命的情况为例。研究模型是在一个涵盖不同临床环境的大型纵向电子健康记录数据集上建立的，包括172个住院病人和1062个门诊地点的703,782名成年病人。模型预测了55.8%的住院病人急性肾损伤，以及90.2%的急性肾损伤需要后续透析治疗，前置时间高达48小时，每一个真正的警报对应2个错误警报。除了预测未来急性肾损伤，该模型还提供了信心评估和对每个预测最显著的临床特征列表，以及预测未来临床相关血液测试的轨迹。虽然对急性肾损伤的识别和及时治疗是一项挑战，但该方法可能为在能够早期治疗的时间窗内识别处于危险中的患者提供机会。

对病情恶化的早期预测可能在支持卫生保健专业人员方面发挥重要作用，因为据估计，11%的医院死亡是由于未能及时识别和治疗病情恶化的患者。要实现这一目标，需要不断更新和准确地预测患者的风险，并在个人层面提供足够的上下文和足够的时间来采取行动。

附：英文原文

Title: A clinically applicable approach to continuous prediction of future acute kidney injury

Author: Nenad Tomaev, Xavier Glorot, Jack W. Rae, Michal Zielinski, Harry Askham, Andre Saraiva, Anne Mottram, Clemens Meyer, Suman Ravuri, Ivan Protsyuk, Alistair Connell, Can O. Hughes, Alan Karthikesalingam, Julien Cornebise, Hugh Montgomery, Geraint Rees, Chris Laing, Clifton R. Baker, Kelly Peterson, Ruth Reeves, Demis Hassabis, Dominic King, Mustafa Suleyman, Trevor Back, Christopher Nielson, Joseph R. Ledsam, Shakir Mohamed

Issue&Volume: Volume 572 Issue 7767

Abstract: The early prediction of deterioration could have an important role in supporting healthcare professionals, as an estimated 11% of deaths in hospital follow a failure to promptly recognize and treat deteriorating patients. To achieve this goal requires predictions of patient risk that are continuously updated and accurate, and delivered at an individual level with sufficient context and enough time to act. Here we develop a deep learning approach for the continuous risk prediction of future deterioration in patients, building on recent work that models adverse events from electronic health records and using acute kidney injurya common and potentially life-threatening conditionas an exemplar. Our model was developed on a large, longitudinal dataset of electronic health records that cover diverse clinical environments, comprising 703,782 adult patients across 172 inpatient and 1,062 outpatient sites. Our model predicts 55.8% of all inpatient episodes of acute kidney injury, and 90.2% of all acute kidney injuries that required subsequent administration of dialysis, with a lead time of up to 48 h and a ratio of 2 false alerts for every true alert. In addition to predicting future acute kidney injury, our model provides confidence assessments and a list of the clinical features that are most salient to each prediction, alongside predicted future trajectories for clinically relevant blood tests. Although the recognition and prompt treatment of acute kidney injury is known to be challenging, our approach may offer opportunities for identifying patients at risk within a time window that enables early treatment.

DOI: 10.1038/s41586-019-1390-1

Source: https://www.nature.com/articles/s41586-019-1390-1