Rethinking Model Prototyping through the MedMNIST+ Dataset Collection

arXiv:2404.15786v1 Announce Type: cross
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, prioritization of marginal performance improvements on a few, narrowly scoped benchmarks over clinical applicability has slowed down meaningful algorithmic progress. This trend often results in excessive fine-tuning of existing methods to achieve state-of-the-art performance on selected datasets rather than fostering clinically relevant innovations. In response, this work presents a comprehensive benchmark …

abstract arxiv benchmarks challenges clinical collection cs.cv cs.lg dataset datasets deep learning eess.iv improvements integration medical performance practice progress prototyping systems through trend type