PRIOR: Prototype Representation Joint Learning from Medical Images and Reports

Contrastive learning based vision-language joint pre-training has emerged as a successful representation learning strategy. In this paper, we present a prototype representation learning framework incorporating both global and local alignment between medical images and reports. In contrast to standard global multi-modality alignment methods, we employ a <PRE_TAG>local alignment module</POST_TAG> for fine-grained representation. Furthermore, a cross-modality conditional reconstruction module is designed to interchange information across modalities in the training phase by reconstructing masked images and reports. For reconstructing long reports, a sentence-wise prototype <PRE_TAG>memory bank</POST_TAG> is constructed, enabling the network to focus on low-level localized visual and high-level clinical linguistic features. Additionally, a non-auto-regressive generation paradigm is proposed for reconstructing non-sequential reports. Experimental results on five downstream tasks, including supervised classification, zero-shot classification, image-to-text retrieval, semantic segmentation, and object detection, show the proposed method outperforms other state-of-the-art methods across multiple datasets and under different dataset size settings. The code is available at https://github.com/QtacierP/PRIOR.