Reducing self-supervised learning complexity improves weakly-supervised classification performance in computational pathology

arXiv:2403.04558v1 Announce Type: new
Abstract: Deep Learning models have been successfully utilized to extract clinically actionable insights from routinely available histology data. Generally, these models require annotations performed by clinicians, which are scarce and costly to generate. The emergence of self-supervised learning (SSL) methods remove this barrier, allowing for large-scale analyses on non-annotated data. However, recent SSL approaches apply increasingly expansive model architectures and larger datasets, causing the rapid escalation of data volumes, hardware prerequisites, and overall expenses, limiting access …

abstract annotations arxiv classification clinicians complexity computational cs.ai cs.cv cs.lg data deep learning emergence extract generate insights pathology performance self-supervised learning ssl supervised learning type weakly-supervised