Variational Topic Inference for Chest X-Ray Report Generation

TL;DR

This paper introduces a variational topic inference approach for automatic chest X-ray report generation, effectively capturing report diversity and aligning image and language data to produce informative, novel reports with competitive performance.

Contribution

It proposes a novel variational topic inference framework with latent topics and visual attention for improved medical report generation.

Findings

Generates diverse, novel reports rather than copying training data.

Achieves performance comparable to state-of-the-art methods.

Effective in aligning image features with report content.

Abstract

Automating report generation for medical imaging promises to reduce workload and assist diagnosis in clinical practice. Recent work has shown that deep learning models can successfully caption natural images. However, learning from medical data is challenging due to the diversity and uncertainty inherent in the reports written by different radiologists with discrepant expertise and experience. To tackle these challenges, we propose variational topic inference for automatic report generation. Specifically, we introduce a set of topics as latent variables to guide sentence generation by aligning image and language modalities in a latent space. The topics are inferred in a conditional variational inference framework, with each topic governing the generation of a sentence in the report. Further, we adopt a visual attention module that enables the model to attend to different locations in the image and generate more informative descriptions. We conduct extensive experiments on two benchmarks, namely Indiana U. Chest X-rays and MIMIC-CXR. The results demonstrate that our proposed variational topic inference method can generate novel reports rather than mere copies of reports used in training, while still achieving comparable performance to state-of-the-art methods in terms of standard language generation criteria.