Fair patient model: Mitigating bias in the patient representation learned from the electronic health records.
Fair patient model: Mitigating bias in the patient representation learned from the electronic health records.
复制标题
公平的患者模型：减少从电子健康记录中了解到的患者代表性的偏差。
DOI：
10.1016/j.jbi.2023.104544
复制
发表时间：
2023
影响因子：
4.5
通讯作者：
Wang,Yanshan
中科院分区：
文献类型：
--
作者：
Sivarajkumar,Sonish;Huang,Yufei;Wang,Yanshan
目的使用一种新的加权损失函数来预训练来自电子健康记录（EHRs）的公平和无偏的患者表征，以减少深度表征学习模型中的偏差并提高公平性。方法在深度表征学习模型中定义了一种新的损失函数，称为加权损失函数，用于平衡不同患者组和特征的重要性。我们将提出的模型，称为公平患者模型（FPM），应用于来自MIMIC-III数据集的34,739名患者样本，并学习了四项临床结果预测任务的患者表征。结果fpm在人口均等、机会差异均等和均等优势比三个公平指标上优于基线模型。FPM也取得了与基线相当的预测性能，平均精度为0.7912。特征分析显示，FPM从临床特征中捕获的信息比基线更多。结论fpm是一种利用加权损失函数从EHR数据中预训练公平、无偏患者表征的新方法。学习到的表示可以用于医疗保健中的各种下游任务，并可以扩展到对公平性很重要的其他领域。
ObjectiveTo pre-train fair and unbiased patient representations from Electronic Health Records (EHRs) using a novel weighted loss function that reduces bias and improves fairness in deep representation learning models.MethodsWe defined a new loss function, called weighted loss function, in the deep representation learning model to balance the importance of different groups of patients and features. We applied the proposed model, called Fair Patient Model (FPM), to a sample of 34,739 patients from the MIMIC-III dataset and learned patient representations for four clinical outcome prediction tasks.ResultsFPM outperformed the baseline models in terms of three fairness metrics: demographic parity, equality of opportunity difference, and equalized odds ratio. FPM also achieved comparable predictive performance with the baselines, with an average accuracy of 0.7912. Feature analysis revealed that FPM captured more information from clinical features than the baselines.ConclusionFPM is a novel method to pre-train fair and unbiased patient representations from the EHR data using a weighted loss function. The learned representations can be used for various downstream tasks in healthcare and can be extended to other domains where fairness is important.