Personalized Multimorbidity Management for Patients with Type 2 Diabetes Using Reinforcement Learning of Electronic Health Records.
Personalized Multimorbidity Management for Patients with Type 2 Diabetes Using Reinforcement Learning of Electronic Health Records.
复制标题
使用电子健康记录的强化学习对2型糖尿病患者进行个性化多变量管理。
DOI：
10.1007/s40265-020-01435-4
复制
发表时间：
2021-03
期刊：
影响因子：
11.5
通讯作者：
Zhong J
中科院分区：
文献类型：
--
作者：
Zheng H;Ryzhov IO;Xie W;Zhong J
2型糖尿病患者中常见慢性并发症。我们开发了一种基于强化学习（RL）的人工智能算法，用于个性化糖尿病和多发病管理，相对于当前的临床实践，该算法具有改善健康结果的强大潜力。我们使用2009-2017年纽约大学Langone Health门诊医疗电子健康记录的16，665例2型糖尿病患者的回顾性队列，将糖尿病、血压和心血管疾病（CVD）风险建模为健康结局。我们训练了一个RL处方算法，该算法推荐了一种治疗方案，该方案使用患者每次就诊时的个人特征和病史来优化患者的累积健康结果。RL建议在一个独立的患者子集上进行评估。单结果优化RL算法，RL-血压，RL-CVD，建议一致的处方，由临床医生观察到的86.1%，82.9%和98.4%的遭遇，分别。对于RL建议与临床医生处方不同的患者，显示不受控制的药物治疗的患者明显较少。（35%的患者A1 c> 8%），未控制的高血压（16%的患者血压> 140 mmHg）和高CVD风险（25%的病例风险> 20%）与临床医生观察到的结果（分别为43%、27%和31%;所有p < 0.001）相比。2型糖尿病的个性化RL处方框架与临床医生的处方具有高度一致性，并且在糖尿病、血压和CVD风险结果方面有实质性改善。本文的在线版本（10.1007/s40265-020-01435-4）包含补充材料，可供授权用户使用。
Comorbid chronic conditions are common among people with type 2 diabetes. We developed an artificial intelligence algorithm, based on reinforcement learning (RL), for personalized diabetes and multimorbidity management, with strong potential to improve health outcomes relative to current clinical practice. We modeled glycemia, blood pressure, and cardiovascular disease (CVD) risk as health outcomes, using a retrospective cohort of 16,665 patients with type 2 diabetes from New York University Langone Health ambulatory care electronic health records in 2009–2017. We trained an RL prescription algorithm that recommends a treatment regimen optimizing patients’ cumulative health outcomes using their individual characteristics and medical history at each encounter. The RL recommendations were evaluated on an independent subset of patients. The single-outcome optimization RL algorithms, RL–glycemia, RL–blood pressure, and RL–CVD, recommended consistent prescriptions as that observed by clinicians in 86.1%, 82.9%, and 98.4% of the encounters, respectively. For patient encounters in which the RL recommendations differed from the clinician prescriptions, significantly fewer encounters showed uncontrolled glycemia (A1c > 8% in 35% of encounters), uncontrolled hypertension (blood pressure > 140 mmHg in 16% of encounters), and high CVD risk (risk > 20% in 25% of encounters) under RL algorithms compared with those observed under clinicians (43%, 27%, and 31% of encounters, respectively; all p < 0.001). A personalized RL prescriptive framework for type 2 diabetes yielded high concordance with clinicians’ prescriptions, and substantial improvements in glycemia, blood pressure, and CVD risk outcomes. The online version of this article (10.1007/s40265-020-01435-4) contains supplementary material, which is available to authorized users.
登录
查看更多内容
DOI：
10.1136/bmj.e5205
发表时间：
2012-09-03
期刊：
BMJ (Clinical research ed.)
影响因子：
--
作者：
Smith SM;Soubhi H;Fortin M;Hudon C;O'Dowd T
通讯作者：
O'Dowd T
影响因子：
16.2
作者：
Bertsimas, Dimitris;Kallus, Nathan;Zhuo, Ying Daisy
通讯作者：
Zhuo, Ying Daisy
影响因子：
28.1
作者：
Lundberg, Scott M.;Nair, Bala;Lee, Su-In
通讯作者：
Lee, Su-In
影响因子：
9.7
作者：
Druss, BG;Marcus, SC;Pincus, HA
通讯作者：
Pincus, HA
影响因子：
64.8
作者：
Mnih, Volodymyr;Kavukcuoglu, Koray;Hassabis, Demis
通讯作者：
Hassabis, Demis