UnICLAM:Contrastive Representation Learning with Adversarial Masking for Unified and Interpretable Medical Vision Question Answering
Preprint
Abstract
Medical Visual Question Answering (Medical-VQA) aims to to answer clinical questions regarding radiology images, assisting doctors with decision-making options. Nevertheless, current Medical-VQA models learn cross-modal representations through residing vision and texture encoders in dual separate spaces, which lead to indirect semantic alignment. In this paper, we propose UnICLAM, a Unified and Interpretable Medical-VQA model through Contrastive Representation Learning with Adversarial Masking. Specifically, to learn an aligned image-text representation, we first establish a unified dual-stream pre-training structure with the gradually soft-parameter sharing strategy. Technically, the proposed strategy learns a constraint for the vision and texture encoders to be close in a same space, which is gradually loosened as the higher number of layers. Moreover, for grasping the unified semantic representation, we extend the adversarial masking data augmentation to the contrastive representation learning of vision and text in a unified manner. Concretely, while the encoder training minimizes the distance between original and masking samples, the adversarial masking module keeps adversarial learning to conversely maximize the distance. Furthermore, we also intuitively take a further exploration to the unified adversarial masking augmentation model, which improves the potential ante-hoc interpretability with remarkable performance and efficiency. Experimental results on VQA-RAD and SLAKE public benchmarks demonstrate that UnICLAM outperforms existing 11 state-of-the-art Medical-VQA models. More importantly, we make an additional discussion about the performance of UnICLAM in diagnosing heart failure, verifying that UnICLAM exhibits superior few-shot adaption performance in practical disease diagnosis.Keywords
All Related Versions
This publication has 8 references indexed in Scilit:
- Scaling Language-Image Pre-Training via Masking2023
- Anomaly Matters: An Anomaly-Oriented Model for Medical Visual Question AnsweringIEEE Transactions on Medical Imaging, 2022
- VQAMix: Conditional Triplet Mixup for Medical Visual Question AnsweringIEEE Transactions on Medical Imaging, 2022
- A Question-Centric Model for Visual Question Answering in Medical ImagingIEEE Transactions on Medical Imaging, 2020
- BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding2019
- A dataset of clinically generated visual questions and answers about radiology imagesScientific Data, 2018
- Multi-modal Factorized Bilinear Pooling with Co-attention Learning for Visual Question Answering2017
- Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization2017