Improve Model Generalization and Robustness to Dataset Bias with Bias-regularized Learning and Domain-guided Augmentation
This addresses the challenge of model generalization across biased medical datasets, such as chest X-rays from different institutions, offering a generally applicable solution for robust biomedical data analysis.
The study tackled the problem of dataset bias in medical imaging by proposing a multi-source domain generalization model that combines a bias-regularized loss function and domain-guided augmentation, significantly improving accuracy and bias measures on unseen domains without retraining.
Deep Learning has thrived on the emergence of biomedical big data. However, medical datasets acquired at different institutions have inherent bias caused by various confounding factors such as operation policies, machine protocols, treatment preference and etc. As the result, models trained on one dataset, regardless of volume, cannot be confidently utilized for the others. In this study, we investigated model robustness to dataset bias using three large-scale Chest X-ray datasets: first, we assessed the dataset bias using vanilla training baseline; second, we proposed a novel multi-source domain generalization model by (a) designing a new bias-regularized loss function; and (b) synthesizing new data for domain augmentation. We showed that our model significantly outperformed the baseline and other approaches on data from unseen domain in terms of accuracy and various bias measures, without retraining or finetuning. Our method is generally applicable to other biomedical data, providing new algorithms for training models robust to bias for big data analysis and applications. Demo training code is publicly available.