SPIDER Baseline nnU-Net

About

Editors:

jaspervdgraaf

Contact email:

jasper.vandergraaf@radboudumc.nl

Image Version:

18195469-6b3c-4daa-95b2-f9f6fbbb6d8b — July 3, 2023

Summary

This algorithm is one of the two baseline algorithms of the SPIDER-Challenge.

The SPIDER Challenge focuses on the segmentation of three anatomical structures in lumbar spine MRI: vertebrae, intervertebral discs (IVDs), and spinal canal. The segmentation task requires similar algorithms to produce separate masks for each vertebra, IVD, and the spinal canal in the lumbar spine MRI volume. This is an nnU-Net based algorithm trained on the complete SPIDER training and validation set using five-fold cross-validation. The labeling sequence of the SPIDER dataset was altered before training this algorithm. The spinal canal is labeled as 1. All IVDs and vertebrae are labeled between 2 and 49 from caudal to cranial (e.g. 2 = IVD L5/S1; 3 = vertebra L5; 4 = IVD L4/L5; 5 = vertebra L4 etc.).

Mechanism

The mechanism of this algorithm is explained in this paper.

Interfaces

This algorithm implements all of the following input-output combinations:

	Inputs	Outputs
1	Sagittal Spine MRI Description Sagittal MRI of the spine Kind Image Read from `/input/images/sagittal-spine-mri/<uuid>.mha` or `/input/images/sagittal-spine-mri/<uuid>.tif` Sagittal Spine MRI	Sagittal spine MR segmentation Description 0: background, 1-25: different vertebrae numbered from the bottom (i.e. L5 = 1) 100: spinal canal 101-125 for partially visible vertebrae 201-225: different intervertebral discs (also numbered from the bottom, i.e. L5/S1 = 201) Kind Segmentation Write to `/output/images/sagittal-spine-mr-segmentation/<uuid>.mha` or `/output/images/sagittal-spine-mr-segmentation/<uuid>.tif` Sagittal spine MR segmentation

Validation and Performance

Left empty by the Algorithm Editors

Challenge Performance

Date	Challenge	Phase	Rank
July 3, 2023	spider	Validation and tuning phase	2
July 3, 2023	spider	Testing phase	1

Uses and Directions

This algorithm was developed for research purposes only.

Warnings

Left empty by the Algorithm Editors

Common Error Messages

Left empty by the Algorithm Editors

Information on this algorithm has been provided by the Algorithm Editors, following the Model Facts labels guidelines from Sendak, M.P., Gao, M., Brajer, N. et al. Presenting machine learning model information to clinical end users with model facts labels. npj Digit. Med. 3, 41 (2020). 10.1038/s41746-020-0253-3