VICCA: Visual interpretation and comprehension of chest X-ray anomalies in generated report without human feedback

Abstract

As artificial intelligence (AI) becomes increasingly central to healthcare, the demand for explainable and trustworthy models is paramount. Current report generation systems for chest X-rays (CXR) often lack mechanisms for validating outputs without expert oversight, raising concerns about reliability and interpretability. To address these challenges, we propose a novel multimodal framework designed to enhance the semantic alignment between text and image context and the localization accuracy of pathologies within images and reports for AI-generated medical reports. Our framework integrates two key modules: a Phrase Grounding Model, which identifies and localizes pathologies in CXR images based on textual prompts, and a Text-to-Image Diffusion Module, which generates synthetic CXR images from prompts while preserving anatomical fidelity. By comparing features between the original and generated images, we introduce a dual-scoring system: one score quantifies localization accuracy, while the other evaluates semantic consistency between text and image features. Our approach significantly outperforms existing methods in pathology localization, achieving an 8% improvement in Intersection over Union score. It also surpasses state-of-the-art methods in CXR text-to-image generation, with a 1% gain in similarity metrics. Additionally, the integration of phrase grounding with diffusion models, coupled with the dual-scoring evaluation system, provides a robust mechanism for validating report quality, paving the way for more reliable and transparent AI in medical imaging.