通过可解释的机器学习和电子健康记录中的错误处理来改善临床决策支持
Mehak Arora1,2, Hassan Mortagy3, Nathan Dwarshuis3
1Department of Electrical and Computer Engineering, Duke University, Durham, NC, 27708, United States.
我们开发了Trust-MAPS,这是一个新的工具,通过结合临床背景来增强机器学习 (ML) 的电子医疗记录 (EMR) 数据处理. 这提高了15%的败血症预测准确度,并增加了模型的解释性.
科学领域:
- 生物医学信息学 生物医学信息学
- 医疗保健中的机器学习
- 临床决策支持系统 临床决策支持系统
背景情况:
- 电子医疗记录 (EMR) 数据经常包含错误和偏见,阻碍了可靠机器学习 (ML) 模型的开发,以支持临床决策.
- 现有的ML算法难以结合医疗数据固有的复杂生理和生物约束,限制了它们的解释性和性能.
研究的目的:
- 开发一种新的数据处理工具,Trust-MAPS,将临床领域的知识集成到EMR数据中,以改善ML模型的错误处理,偏差缓解和可解释性.
- 提高医疗保健应用中使用的ML模型的预测能力和临床相关性.
主要方法:
- 开发了Trust-MAPS算法,该算法将临床知识转化为捕捉生理约束的数学模型.
- 将EMR数据投射到一个受约束的空间中,以识别和量化使用"信任分数"来确定和量化与健康生理学的偏差.
- 将信任评分集成到下游ML任务的功能空间中,证明使用XGBoost和SMOTE的败血症预测模型的实用性.
主要成果:
- 信任-MAPS框架有效处理数据错误并改善预测性能.
- 在发作6小时前预测败血症的接受器操作特征曲线下达到0.91的区域,比基线改善15%.
- 证明了ML模型的偏差降低和增强的解释性.
结论:
- 信任-MAPS预处理显著改善下游分类性能,并减少ML模型中的偏差.
- 信任评分提供了临床上有意义的特征,提高了临床决策支持的预测准确性和可解释性.
- 这种新的方法将临床知识转化为数学约束,以改进高维医学数据分析.
更多相关视频
04:09Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
Published on: October 10, 2018
12:18A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 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...
Errors occurring during blood pressure monitoring
Several factors...
Health Information Technology and Healthcare Information System
Health Information Technology, commonly called HIT, integrates advanced information systems and technology in healthcare settings. Its primary functions include:
Issues And Trends In Healthcare Delivery System
Cost Containment
Payment for healthcare services has historically promoted adoption of costly and often unnecessary or inefficient...
Methods of Documentation VI: Case Management Model
For example, a patient with a chronic...
