Learning to Exploit Temporal Structure for Biomedical Vision-Language Processing
Authors:
Top Cited Papers
- 1 June 2023
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 15016-15027
Abstract
Self-supervised learning in vision-language processing (VLP) exploits semantic alignment between imaging and text modalities. Prior work in biomedical VLP has mostly relied on the alignment of single image and report pairs even though clinical notes commonly refer to prior images. This does not only introduce poor alignment between the modalities but also a missed opportunity to exploit rich self-supervision through existing temporal content in the data. In this work, we explicitly account for prior images and reports when available during both training and fine-tuning. Our approach, named BioViL-T, uses a CNN-Transformer hybrid multi-image encoder trained jointly with a text model. It is designed to be versatile to arising challenges such as pose variations and missing input images across time. The resulting model excels on downstream tasks both in single- and multi-image setups, achieving state-of-the-art (SOTA) performance on (I) progression classification, (II) phrase grounding, and (III) report generation, whilst offering consistent improvements on disease classification and sentence-similarity tasks. We release a novel multi-modal temporal benchmark dataset, MS-CXR-T, to quantify the quality of vision-language representations in terms of temporal semantics. Our experimental results show the advantages of incorporating prior images and reports to make most use of the data.Keywords
All Related Versions
This publication has 38 references indexed in Scilit:
- ChestX-Ray8: Hospital-Scale Chest X-Ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseases2017
- Deep Residual Learning for Image Recognition2016
- Preparing a collection of radiology examinations for distribution and retrievalJournal of the American Medical Informatics Association, 2015
- The Design of SimpleITKFrontiers in Neuroinformatics, 2013
- Toward a Comprehensive Framework for the Spatiotemporal Statistical Analysis of Longitudinal Shape DataInternational Journal of Computer Vision, 2012
- A nonparametric-based rib suppression method for chest radiographsComputers & Mathematics with Applications, 2012
- Symmetric diffeomorphic image registration with cross-correlation: Evaluating automated labeling of elderly and neurodegenerative brainMedical Image Analysis, 2008
- Optimization of mutual information for multiresolution image registrationIEEE Transactions on Image Processing, 2000
- PhysioBank, PhysioToolkit, and PhysioNetCirculation, 2000
- Influence of prior radiologic information on the interpretation of radiographic examinationsAcademic Radiology, 1995