Fine-Grained Image-Text Alignment in Medical Imaging Enables Explainable Cyclic Image-Report Generation

TL;DR

This paper introduces AdaMatch, a novel model that aligns image regions with report words in chest X-ray analysis, enhancing explainability and improving report generation accuracy through adaptive patch extraction and bidirectional language modeling.

Contribution

The paper presents AdaMatch with adaptive patch extraction and bidirectional language modeling for explainable and accurate CXR report generation, a novel approach in medical image-text alignment.

Findings

Outperforms existing methods on two CXR datasets

Provides explicit explanations for image regions and report words

Enhances interpretability of CXR report generation

Abstract

To address these issues, we propose a novel Adaptive patch-word Matching (AdaMatch) model to correlate chest X-ray (CXR) image regions with words in medical reports and apply it to CXR-report generation to provide explainability for the generation process. AdaMatch exploits the fine-grained relation between adaptive patches and words to provide explanations of specific image regions with corresponding words. To capture the abnormal regions of varying sizes and positions, we introduce the Adaptive Patch extraction (AdaPatch) module to acquire the adaptive patches for these regions adaptively. In order to provide explicit explainability for CXR-report generation task, we propose an AdaMatch-based bidirectional large language model for Cyclic CXR-report generation (AdaMatch-Cyclic). It employs the AdaMatch to obtain the keywords for CXR images and `keypatches' for medical reports as hints to guide CXR-report generation. Extensive experiments on two publicly available CXR datasets prove the effectiveness of our method and its superior performance to existing methods.