Align, Reason and Learn: Enhancing Medical Vision-and-Language Pre-training with Knowledge
Open accessThe DOI matches its Crossref record; 15 of the 15 references it cites were found in a registry or an index.
Abstract
Medical vision-and-language pre-training (Med-VLP) has received considerable attention owing to its applicability to extracting generic vision-and-language representations from medical images and texts. Most existing methods mainly contain three elements: uni-modal encoders (i.e., a vision encoder and a language encoder), a multi-modal fusion module, and pretext tasks, with few studies considering the importance of medical domain expert knowledge and explicitly exploiting such knowledge to facilitate Med-VLP. Although there exist knowledge-enhanced vision-and-language pre-training (VLP) methods in the general domain, most require off-the-shelf toolkits (e.g., object detectors and scene graph parsers), which are unavailable in the medical domain. In this paper, we propose a systematic and effective approach to enhance Med-VLP by structured medical knowledge from three perspectives. First, considering knowledge can be regarded as the intermediate medium between vision and language, we align the representations of the vision encoder and the language encoder through knowledge. Second, we inject knowledge into the multi-modal fusion model to enable the model to perform reasoning using knowledge as the supplementation of the input image and text. Third, we guide the model to put emphasis on the most critical information in images and texts by designing knowledge-induced pretext tasks. To perform a comprehensive evaluation and facilitate further research, we construct a medical vision-and-language benchmark including three tasks. Experimental results illustrate the effectiveness of our approach, where state-of-the-art performance is achieved on all downstream tasks. Further analyses explore the effects of different components of our approach and various settings of pre-training.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., et al. (2020) · arXiv · doi:10.48550/arxiv.2010.11929 · Cited by 47,472
Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
Wu, Y., Schuster, M., Chen, Z., Le, Q. V., Norouzi, M., Macherey, W., Krikun, M., Cao, Y., Gao, Q., Macherey, K., et al. (2016) · arXiv · doi:10.48550/arxiv.1609.08144 · Cited by 5,319 · cited 1×
“Numbers are marked for ease of reading. normalization are employed to 𝑯 𝒗𝒔 and 𝑯 𝒍𝒔 and we denote the results as 𝑯 𝒗𝒔 and 𝑯 𝒍𝒔 , respectively, for simplicity.”
Self-citationSciBERT: A Pretrained Language Model for Scientific Text
Beltagy, I., Lo, K., & Cohan, A. (2019) · Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) · doi:10.18653/v1/d19-1371 · Cited by 3,218
LXMERT: Learning Cross-Modality Encoder Representations from Transformers
Tan, H., & Bansal, M. (2019) · Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) · doi:10.18653/v1/d19-1514 · Cited by 2,319 · cited 8×
“In specific, for the single-stream fusion, the models [8, 27, 28, 43] use a single Transformer for early and unconstrained fusion between modalities; for the dual-stream fusion, the models [35, 47, 55] adopt the co-attention mechanism to interact different modalities.”
BackgroundMethodInfluentialViLBERT: Pretraining Task-Agnostic Visiolinguistic Representations for Vision-and-Language Tasks
Lu, J., Batra, D., Parikh, D., & Lee, S. (2019) · arXiv · doi:10.48550/arxiv.1908.02265 · Cited by 2,109
UNITER: UNiversal Image-TExt Representation Learning
Chen, Y.-C., Li, L., Yu, L., El Kholy, A., Ahmed, F., Gan, Z., Cheng, Y., & Liu, J. (2020) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-58577-8_7 · Cited by 2,055 · cited 10×
“For ROCO and MedICaT, we filter non-radiology samples, and for MIMIC-CXR, we only keep images in the frontal view.”
BackgroundMethodInfluentialERNIE: Enhanced Language Representation with Informative Entities
Zhang, Z., Han, X., Liu, Z., Jiang, X., Sun, M., & Liu, Q. (2019) · Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/p19-1139 · Cited by 1,482 · cited 2×
“According to the knowledge injection schemes, existing studies can be classified into four categories: embeddings combination [41, 59], data structure compatibility [16, 32, 45], knowledge supervision [46, 49], and neural-symbolic methods [2].”
BackgroundVisualBERT: A Simple and Performant Baseline for Vision and Language
Li, L. H., Yatskar, M., Yin, D., Hsieh, C.-J., & Chang, K.-W. (2019) · arXiv · doi:10.48550/arxiv.1908.03557 · Cited by 1,198 · cited 9×
“In specific, for the single-stream fusion, the models [8, 27, 28, 43] use a single Transformer for early and unconstrained fusion between modalities; for the dual-stream fusion, the models [35, 47, 55] adopt the co-attention mechanism to interact different modalities.”
BackgroundMethodInfluentialK-BERT: Enabling Language Representation with Knowledge Graph
Liu, W., Zhou, P., Zhao, Z., Wang, Z., Ju, Q., Deng, H., & Wang, P. (2020) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v34i03.5681 · Cited by 894
KEPLER: A Unified Model for Knowledge Embedding and Pre-trained Language Representation
Wang, X., Gao, T., Zhu, Z., Zhang, Z., Liu, Z., Li, J., & Tang, J. (2021) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00360 · Cited by 676 · cited 1×
“An overview of the proposed approach is demonstrated in Figure 2, and the details of the general Med-VLP framework, the knowledge extraction process (shown in Figure 1), and the injection of knowledge into the general Med-VLP framework are introduced in the following subsections.”
MethodKnowledge Enhanced Contextual Word Representations
Peters, M. E., Neumann, M., Logan, R., Schwartz, R., Joshi, V., Singh, S., & Smith, N. A. (2019) · Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) · doi:10.18653/v1/d19-1005 · Cited by 661 · cited 2×
“According to the knowledge injection schemes, existing studies can be classified into four categories: embeddings combination [41, 59], data structure compatibility [16, 32, 45], knowledge supervision [46, 49], and neural-symbolic methods [2].”
BackgroundMIMIC-CXR-JPG, a large publicly available database of labeled chest radiographs
Johnson, A. E. W., Pollard, T. J., Greenbaum, N. R., Lungren, M. P., Deng, C.-Y., Peng, Y., Lu, Z., Mark, R. G., Berkowitz, S. J., & Horng, S. (2019) · arXiv · doi:10.48550/arxiv.1901.07042 · Cited by 269
CoLAKE: Contextualized Language and Knowledge Embedding
Sun, T., Shao, Y., Qiu, X., Guo, Q., Hu, Y., Huang, X., & Zhang, Z. (2020) · Proceedings of the 28th International Conference on Computational Linguistics · doi:10.18653/v1/2020.coling-main.327 · Cited by 185
Li, Y., Wang, H., & Luo, Y. (2020) · 2020 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) · doi:10.1109/bibm49941.2020.9313289 · Cited by 81 · cited 2×
“More recently, some studies [15, 22, 29, 37] applied VLP to the medical domain and significantly improved the performance for medical vision-and-language tasks (especially for Med-VQA).”
Background
Cited by
Knowledge-enhanced visual-language pre-training on chest radiology images
Zhang, X., Wu, C., Zhang, Y., Xie, W., & Wang, Y. (2023) · Nature Communications · doi:10.1038/s41467-023-40260-7 · Cited by 207
He, K., Mao, R., Lin, Q., Ruan, Y., Lan, X., Feng, M., & Cambria, E. (2024) · doi:10.2139/ssrn.4809363 · Cited by 69
Li, Y., Kong, C., Zhao, G., & Zhao, Z. (2025) · Artificial Intelligence Review · doi:10.1007/s10462-025-11337-0 · Cited by 7
“KAD (Zhang et al. 2023), (Chen et al. 2022), and MedKLIP (Wu et al. 2023b ) all underscore the significance of knowledge integration during pre-training.”
Metadata
- Authors
- Chen, Z., Li, G., & Wan, X.
- Venue
- Proceedings of the 30th ACM International Conference on Multimedia
- Publisher
- ACM
- Published
- 2022-10-10
- Pages
- 5152-5161
- OpenAlex
- W4304092062
- Cited by
- 82
- References
- 15
- FWCI
- 6.95
- Type
- conference paper
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.