"我不知道":一种不确定性感知机器学习模型,用于预测急诊室分组的患者倾向
Abubakar Sadiq Bouda Abdulai1, Jean Storm2, Michael Ehrlich3
1Data Strategy Workgroup, Quality Insights Inc., 3001 Chesterfield Ave, Charleston WV 25311, USA.
International journal of medical informatics
|May 3, 2025
概括
对于急诊室分组的符合性预测模型提供了不确定性意识的患者处置预测. 这种方法通过减少过度自信的机器学习输出来提高可靠性和患者安全.
科学领域:
- 医疗信息学 医疗信息学
- 机器学习 机器学习
- 临床决策支持 临床决策支持
背景情况:
- 机器学习 (ML) 模型对于预测急诊室 (ED) 选中的患者态度至关重要.
- 然而,当前的ML模型往往缺乏不确定性量化,导致可能过度自信的预测,这可能会影响临床决策.
研究的目的:
- 开发和评估ED分类的合规预测模型.
- 目标是提供不确定性意识的患者倾向预测,并将高不确定性病例的"不知道"输出纳入.
主要方法:
- 对560,486名成年ED访问进行了回顾性分析.
- 一个极端梯度增强 (XGBoost) 模型被训练,验证,然后被规范化以处理不确定性.
- 该模型的性能在56000个ED病例的样本上进行了测试.
主要成果:
- 标准的XGBoost模型实现了0.9307.7的AUC.
- 符合预测,在60%的信心值,在4.9%的病例中确定了"不知道",提高了对0.74的敏感性和对0.95.9的特异性.
- 更高的信心门 (例如,90%) 增加了"不知道"预测 (34.5%),同时提高了灵敏度 (0.88) 和特异性 (0.99),表明了信心和准确性之间的权衡.
结论:
- 将不确定性意识集成到ML模型中,可以显著提高ED分拣的可靠性.
- 这种方法为临床医生提供了更容易解释的见解,减轻了与过度自信的预测相关的风险.
- 最终,这导致在急诊室环境中提高了患者安全.
相关概念视频
The Availability Heuristic
5.8K
A heuristic is a general problem-solving framework (Tversky & Kahneman, 1974). You can think of these as mental shortcuts that are used to solve problems. Different types of heuristics are used in different types of situations, and the impulse to use a heuristic occurs when one of five conditions is met (Pratkanis, 1989):
5.8K
Methods of Documentation VI: Case Management Model
537
The case management model is a multidisciplinary approach that involves healthcare professionals from diverse disciplines, such as physicians, nurses, therapists, social workers, and pharmacists, working collaboratively to address the various needs of patients. Each healthcare professional brings unique expertise and perspectives, contributing to a more comprehensive understanding of the patient's condition and tailoring treatment plans accordingly.
For example, a patient with a chronic...
For example, a patient with a chronic...
537


