Clinical information extraction for Low-resource languages with Few-shot learning using Pre-trained language models and Prompting

arXiv:2403.13369v1 Announce Type: cross
Abstract: Automatic extraction of medical information from clinical documents poses several challenges: high costs of required clinical expertise, limited interpretability of model predictions, restricted computational resources and privacy regulations. Recent advances in domain-adaptation and prompting methods showed promising results with minimal training data using lightweight masked language models, which are suited for well-established interpretability methods. We are first to present a systematic evaluation of these methods in a low-resource setting, by performing multi-class section classification on …

abstract advances arxiv challenges clinical computational costs cs.ai cs.cl cs.lg documents domain expertise extraction few-shot few-shot learning information information extraction interpretability language language models languages low medical predictions privacy prompting regulations resources results type