Skip to content

Read-only version: For security reasons, the Research Information System (FIS) has been temporarily replaced by this read-only version. All public content and full texts remain available; login, editing, submission and statistics are not available. Questions: fis@uni-bamberg.de

Home › Publications › MedMNIST-C : Comprehensive benchmark and improved classifier robustness by simulating realistic image corruptions

Publication

MedMNIST-C : Comprehensive benchmark and improved classifier robustness by simulating realistic image corruptions

Di Salvo, Francesco; Doerrich, Sebastian; Ledig, Christian (2025): MedMNIST-C : Comprehensive benchmark and improved classifier robustness by simulating realistic image corruptions, in: Bamberg: Otto-Friedrich-Universität, S. 1–11.

Details

Faculty/Chair
Faculty/Professorship
Author
By
... ; Doerrich, Sebastian; ...
Publisher Information
Otto-Friedrich-Universität : Bamberg
Year of publication
2025
Source/Other editions
ADSMI : MICCAI Workshop on Advancing Data Solutions in Medical Imaging AI, October 10, 2024 ; Marrakesh, Morocco, arXiv, 2024, S. 1–11
Year of first publication
2024
Language
English
Licence
URN
Abstract

The integration of neural-network-based systems into clinical practice is limited by challenges related to domain generalization and robustness. The computer vision community established benchmarks such as ImageNet-C as a fundamental prerequisite to measure progress towards those challenges. Similar datasets are largely absent in the medical imaging community which lacks a comprehensive benchmark that spans across imaging modalities and applications. To address this gap, we create and open-source MedMNIST-C, a benchmark dataset based on the MedMNIST+ collection covering 12 datasets and 9 imaging modalities. We simulate task and modality-specific image corruptions of varying severity to comprehensively evaluate the robustness of established algorithms against real-world artifacts and distribution shifts. We further provide quantitative evidence that our simple-to-use artificial corruptions allow for highly performant, lightweight data augmentation to enhance model robustness. Unlike traditional, generic augmentation strategies, our approach leverages domain knowledge, exhibiting significantly higher robustness when compared to widely adopted methods. By introducing MedMNIST-C and open-sourcing the corresponding library allowing for targeted data augmentations, we contribute to the development of increasingly robust methods tailored to the challenges of medical imaging.

GND Keywords
Data Augmentation
Bildanalyse
Medizin
Keywords
Robustness
Corruptions
Data augmentation
DDC Classification
004 Computer science
RVK Classification
ST 300
Peer Reviewed
yes
Type
Conferenceobject
URI
Activation date
2025-10-27

Permalink: https://fis.uni-bamberg.de/handle/uniba/109532 · Data as JSON · Cite: BibTeX · RIS

Data as of: 2026-11-16

Restricted read-only access: login, editing and submission are temporarily disabled.