Multi-modal Understanding And Generation For Medical Images And Text Via Vision-language Pre-training

Moon Jong Hak, Lee Hyungyung, Shin Woncheol, Kim Young-hak, Choi Edward. IEEE Journal of Biomedical and Health Informatics 2021

[Paper] [Code]    
Applications Attention Mechanism BERT Has Code Model Architecture Multimodal Models Training Techniques

Recently a number of studies demonstrated impressive performance on diverse vision-language multi-modal tasks such as image captioning and visual question answering by extending the BERT architecture with multi-modal pre-training objectives. In this work we explore a broad set of multi-modal representation learning tasks in the medical domain, specifically using radiology images and the unstructured report. We propose Medical Vision Language Learner (MedViLL), which adopts a BERT-based architecture combined with a novel multi-modal attention masking scheme to maximize generalization performance for both vision-language understanding tasks (diagnosis classification, medical image-report retrieval, medical visual question answering) and vision-language generation task (radiology report generation). By statistically and rigorously evaluating the proposed model on four downstream tasks with three radiographic image-report datasets (MIMIC-CXR, Open-I, and VQA-RAD), we empirically demonstrate the superior downstream task performance of MedViLL against various baselines, including task-specific architectures. The source code is publicly available at: https://github.com/SuperSupermoon/MedViLL

Similar Work