范围:使用电子健康记录在办公室访问中预测未来的诊断
Pritam Mukherjee1, Marie Humbert-Droz1, Jonathan H Chen1
1Department of Medicine, Stanford Center for Biomedical Informatics, Stanford University, 1265 Welch Rd, Palo Alto, CA, 94305, USA.
我们开发了一个可解释的模型,使用过去的诊断和实验室结果来预测未来的医学诊断. 这种方法有助于医生使用电子健康记录 (EHR),并显示出临床部署的希望.
科学领域:
- 医疗信息学 医疗信息学
- 医疗保健中的机器学习
- 临床决策支持 临床决策支持
背景情况:
- 电子健康记录 (EHR) 包含大量的患者数据.
- 医生需要有效的工具来导航复杂的患者病史和预测诊断.
- 现有的诊断预测模型可能缺乏可解释性或可扩展性.
研究的目的:
- 开发和评估一种可解释和可扩展的模型,用于预测患者接触时可能的诊断.
- 帮助医生更有效地利用EHR数据.
- 将拟议模型的性能与深度学习方法进行比较.
主要方法:
- 从2,701,522名患者的非识别的EHR数据进行了回顾性分析.
- 使用二进制相关性策略开发多标签分类模型.
- 测试后勤回归和随机森林作为数据聚合的各种时间窗口的基础分类器.
- 与基于循环神经网络 (RNN) 的深度学习模型进行比较.
主要成果:
- 最好的模型,利用随机森林分类器与综合的人口,诊断和实验室数据,实现了0.904.90的中位数AUROC.
- 性能与现有方法相当或优于现有方法,包括超越AUROC.中的深度学习模型.
- 模型解释性揭示了有意义的特征关联,突出了临床相关性.
结论:
- 拟议的可解释多标签模型显示了与深度学习方法可比的性能,同时提供了更高的简单性和可解释性.
- 该模型是临床部署的有希望的候选人,有助于诊断预测.
- 需要在多个机构进行进一步的验证.
更多相关视频
07:51Hydra, a Computer-Based Platform for Aiding Clinicians in Cardiovascular Analysis and Diagnosis
Published on: September 26, 2018
06:55Inverse Probability of Treatment Weighting Propensity Score using the Military Health System Data Repository and National Death Index
Published on: January 8, 2020
相关概念视频
Methods of Documentation VII: EMR
Documentation of Nursing Diagnosis
In some settings, data-driven computerized decision support systems are in place, allowing for more accurate nursing diagnoses. The database within one of these systems includes diagnostic labels defining characteristics, activities, and indicators for nursing. A nurse enters...
Steps in Outbreak Investigation
Classification of Illness
An illness is a response to a disease in which the person's level of functioning is changed compared with a previous level. The general classification of illness includes acute and chronic.
Acute illness is severe...
Purpose of Health Records II
Issues And Trends In Healthcare Delivery System
Cost Containment
Payment for healthcare services has historically promoted adoption of costly and often unnecessary or inefficient...
