The LoDoPaB-CT Dataset: A Benchmark Dataset for Low-Dose CT Reconstruction Methods

TL;DR

The LoDoPaB-CT Dataset provides a large, publicly available benchmark for low-dose CT reconstruction, facilitating fair comparison of deep learning methods in inverse imaging problems.

Contribution

This paper introduces a comprehensive, publicly accessible dataset for low-dose CT reconstruction, including data processing and baseline results, to standardize evaluation.

Findings

Over 40,000 scan slices from 800 patients included

Baseline reconstruction results provided

Dataset enables fair comparison of methods

Abstract

Deep Learning approaches for solving Inverse Problems in imaging have become very effective and are demonstrated to be quite competitive in the field. Comparing these approaches is a challenging task since they highly rely on the data and the setup that is used for training. We provide a public dataset of computed tomography images and simulated low-dose measurements suitable for training this kind of methods. With the LoDoPaB-CT Dataset we aim to create a benchmark that allows for a fair comparison. It contains over 40,000 scan slices from around 800 patients selected from the LIDC/IDRI Database. In this paper we describe how we processed the original slices and how we simulated the measurements. We also include first baseline results.