Probabilistic Robustness in Medical Image Classification
For medical imaging practitioners, it offers a statistically grounded evaluation framework for model trustworthiness, but the contribution is incremental.
The paper investigates probabilistic robustness (PR) as a more practical trustworthiness measure than adversarial robustness for medical image classification, evaluating DL models on MedMNIST v2 under natural corruptions. No concrete performance numbers are provided.
Deep learning (DL) has shown strong performance in medical image classification, but its trustworthy deployment remains challenging in safety-critical clinical settings, where prediction errors under perturbations may lead to severe consequences. Existing studies mainly focus on adversarial robustness (AR) from a worst-case perspective; however, such settings may be less representative of real medical applications. In this work, we investigate probabilistic robustness (PR) as a more practical measure of model trustworthiness. To this end, we construct a set of natural corruption settings for medical image classification and systematically evaluate commonly used DL models on MedMNIST v2 dataset. Our study provides a statistically grounded perspective on assessing the trustworthiness of DL models, thereby supporting their more trustworthy deployment in medical imaging applications.