A dataset for Computer-Aided Detection of Pulmonary Embolism in CTA images

TL;DR

This paper introduces a publicly available CTA image dataset for pulmonary embolism detection, enabling standardized evaluation and comparison of CAD systems in PE analysis.

Contribution

The authors collected and labeled a comprehensive CTA dataset with 5160 slices from 20 patients, filling a critical gap for benchmarking PE detection methods.

Findings

Dataset includes 66% peripheral PE cases

Contains images with various pulmonary diseases

Facilitates standardized evaluation of CAD systems

Abstract

Todays, researchers in the field of Pulmonary Embolism (PE) analysis need to use a publicly available dataset to assess and compare their methods. Different systems have been designed for the detection of pulmonary embolism (PE), but none of them have used any public datasets. All papers have used their own private dataset. In order to fill this gap, we have collected 5160 slices of computed tomography angiography (CTA) images acquired from 20 patients, and after labeling the image by experts in this field, we provided a reliable dataset which is now publicly available. In some situation, PE detection can be difficult, for example when it occurs in the peripheral branches or when patients have pulmonary diseases (such as parenchymal disease). Therefore, the efficiency of CAD systems highly depends on the dataset. In the given dataset, 66% of PE are located in peripheral branches, and different pulmonary diseases are also included.