RadLing: Towards Efficient Radiology Report Understanding
This addresses the need for efficient radiology report understanding, though it appears incremental as it builds on existing Electra architecture with domain-specific adaptations.
The authors tackled the problem of limited pretrained language models for radiology by developing RadLing, a model trained on 500K radiology reports that achieves state-of-the-art results in fine-tuning tasks for this domain.
Most natural language tasks in the radiology domain use language models pre-trained on biomedical corpus. There are few pretrained language models trained specifically for radiology, and fewer still that have been trained in a low data setting and gone on to produce comparable results in fine-tuning tasks. We present RadLing, a continuously pretrained language model using Electra-small (Clark et al., 2020) architecture, trained using over 500K radiology reports, that can compete with state-of-the-art results for fine tuning tasks in radiology domain. Our main contribution in this paper is knowledge-aware masking which is a taxonomic knowledge-assisted pretraining task that dynamically masks tokens to inject knowledge during pretraining. In addition, we also introduce an knowledge base-aided vocabulary extension to adapt the general tokenization vocabulary to radiology domain.