Lumbar spine segmentation in MR images: a dataset and a public benchmark

TL;DR

This paper introduces a large, multi-center lumbar spine MRI dataset with reference segmentations, along with a benchmark for evaluating segmentation algorithms to foster collaboration and improve diagnostic accuracy.

Contribution

It provides a publicly available dataset, a semi-automatic annotation method, and a benchmark for lumbar spine MRI segmentation algorithms.

Findings

nnU-Net performed comparably to the baseline algorithm.

The dataset includes 447 MRI series from 218 patients.

A continuous challenge setup enables fair algorithm comparison.

Abstract

This paper presents a large publicly available multi-center lumbar spine magnetic resonance imaging (MRI) dataset with reference segmentations of vertebrae, intervertebral discs (IVDs), and spinal canal. The dataset includes 447 sagittal T1 and T2 MRI series from 218 patients with a history of low back pain and was collected from four different hospitals. An iterative data annotation approach was used by training a segmentation algorithm on a small part of the dataset, enabling semi-automatic segmentation of the remaining images. The algorithm provided an initial segmentation, which was subsequently reviewed, manually corrected, and added to the training data. We provide reference performance values for this baseline algorithm and nnU-Net, which performed comparably. Performance values were computed on a sequestered set of 39 studies with 97 series, which were additionally used to set up a continuous segmentation challenge that allows for a fair comparison of different segmentation algorithms. This study may encourage wider collaboration in the field of spine segmentation and improve the diagnostic value of lumbar spine MRI.