Saved in:
Bibliographic Details
Main Authors: Luo, Yuanjiang, Li, Hongxiang, Wu, Xuan, Cao, Meng, Huang, Xiaoshuang, Zhu, Zhihong, Liao, Peixi, Chen, Hu, Zhang, Yi
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2405.20607
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929375934414848
author Luo, Yuanjiang
Li, Hongxiang
Wu, Xuan
Cao, Meng
Huang, Xiaoshuang
Zhu, Zhihong
Liao, Peixi
Chen, Hu
Zhang, Yi
author_facet Luo, Yuanjiang
Li, Hongxiang
Wu, Xuan
Cao, Meng
Huang, Xiaoshuang
Zhu, Zhihong
Liao, Peixi
Chen, Hu
Zhang, Yi
contents Existing mainstream approaches follow the encoder-decoder paradigm for generating radiology reports. They focus on improving the network structure of encoders and decoders, which leads to two shortcomings: overlooking the modality gap and ignoring report content constraints. In this paper, we proposed Textual Inversion and Self-supervised Refinement (TISR) to address the above two issues. Specifically, textual inversion can project text and image into the same space by representing images as pseudo words to eliminate the cross-modeling gap. Subsequently, self-supervised refinement refines these pseudo words through contrastive loss computation between images and texts, enhancing the fidelity of generated reports to images. Notably, TISR is orthogonal to most existing methods, plug-and-play. We conduct experiments on two widely-used public datasets and achieve significant improvements on various baselines, which demonstrates the effectiveness and generalization of TISR. The code will be available soon.
format Preprint
id arxiv_https___arxiv_org_abs_2405_20607
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Textual Inversion and Self-supervised Refinement for Radiology Report Generation
Luo, Yuanjiang
Li, Hongxiang
Wu, Xuan
Cao, Meng
Huang, Xiaoshuang
Zhu, Zhihong
Liao, Peixi
Chen, Hu
Zhang, Yi
Computer Vision and Pattern Recognition
Existing mainstream approaches follow the encoder-decoder paradigm for generating radiology reports. They focus on improving the network structure of encoders and decoders, which leads to two shortcomings: overlooking the modality gap and ignoring report content constraints. In this paper, we proposed Textual Inversion and Self-supervised Refinement (TISR) to address the above two issues. Specifically, textual inversion can project text and image into the same space by representing images as pseudo words to eliminate the cross-modeling gap. Subsequently, self-supervised refinement refines these pseudo words through contrastive loss computation between images and texts, enhancing the fidelity of generated reports to images. Notably, TISR is orthogonal to most existing methods, plug-and-play. We conduct experiments on two widely-used public datasets and achieve significant improvements on various baselines, which demonstrates the effectiveness and generalization of TISR. The code will be available soon.
title Textual Inversion and Self-supervised Refinement for Radiology Report Generation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2405.20607