Rethinking model prototyping through the MedMNIST+ dataset collection.

ABSTRACT: The integration of deep learning based systems in clinical practice is often impeded by challenges rooted in limited and heterogeneous medical datasets. In addition, the field has increasingly prioritized marginal performance gains on a few, narrowly scoped benchmarks over clinical applicability, slowing down meaningful algorithmic progress. This trend often results in excessive fine-tuning of existing methods on selected datasets rather than fostering clinically relevant innovations. In response, this work introduces a comprehensive benchmark for the MedMNIST+ dataset collection, designed to diversify the evaluation landscape across several imaging modalities, anatomical regions, classification tasks and sample sizes. We systematically reassess commonly used Convolutional Neural Networks

SUBMITTER: Doerrich S 

PROVIDER: S-EPMC11883007 | biostudies-literature | 2025 Mar

REPOSITORIES: biostudies-literature