Detailed Annotations of Chest X-Rays via CT Projection for Report Understanding

TL;DR

This paper introduces PAXRay, a dataset with detailed anatomical annotations from CT scans, enabling improved medical report understanding and phrase grounding in X-ray images by leveraging anatomical structures.

Contribution

The work presents a novel pipeline for extracting anatomical structures from CT data and integrates this into the PAXRay dataset, enhancing visual grounding of radiological findings.

Findings

Anatomical information improves grounding accuracy by up to 50%.

The PAXRay dataset facilitates better association between reports and imagery.

Methods leveraging anatomy outperform traditional region proposal techniques.

Abstract

In clinical radiology reports, doctors capture important information about the patient's health status. They convey their observations from raw medical imaging data about the inner structures of a patient. As such, formulating reports requires medical experts to possess wide-ranging knowledge about anatomical regions with their normal, healthy appearance as well as the ability to recognize abnormalities. This explicit grasp on both the patient's anatomy and their appearance is missing in current medical image-processing systems as annotations are especially difficult to gather. This renders the models to be narrow experts e.g. for identifying specific diseases. In this work, we recover this missing link by adding human anatomy into the mix and enable the association of content in medical reports to their occurrence in associated imagery (medical phrase grounding). To exploit anatomical structures in this scenario, we present a sophisticated automatic pipeline to gather and integrate human bodily structures from computed tomography datasets, which we incorporate in our PAXRay: A Projected dataset for the segmentation of Anatomical structures in X-Ray data. Our evaluation shows that methods that take advantage of anatomical information benefit heavily in visually grounding radiologists' findings, as our anatomical segmentations allow for up to absolute 50% better grounding results on the OpenI dataset as compared to commonly used region proposals. The PAXRay dataset is available at https://constantinseibold.github.io/paxray/.