Bayesian uncertainty-weighted loss for improved generalisability on polyp segmentation task
This addresses the critical challenge of unfair models in clinical applications by mitigating bias in polyp segmentation for medical imaging.
The paper tackled the problem of poor generalizability in polyp segmentation models across multi-center datasets by adapting a Bayesian uncertainty-weighted loss to focus on underrepresented samples, achieving improved generalizability without sacrificing state-of-the-art performance on the PolypGen dataset.
While several previous studies have devised methods for segmentation of polyps, most of these methods are not rigorously assessed on multi-center datasets. Variability due to appearance of polyps from one center to another, difference in endoscopic instrument grades, and acquisition quality result in methods with good performance on in-distribution test data, and poor performance on out-of-distribution or underrepresented samples. Unfair models have serious implications and pose a critical challenge to clinical applications. We adapt an implicit bias mitigation method which leverages Bayesian predictive uncertainties during training to encourage the model to focus on underrepresented sample regions. We demonstrate the potential of this approach to improve generalisability without sacrificing state-of-the-art performance on a challenging multi-center polyp segmentation dataset (PolypGen) with different centers and image modalities.