Format

Send to

Choose Destination
Neuroimage. 2019 Jan 1;184:901-915. doi: 10.1016/j.neuroimage.2018.09.081. Epub 2018 Oct 6.

Automatic segmentation of the spinal cord and intramedullary multiple sclerosis lesions with convolutional neural networks.

Author information

1
NeuroPoly Lab, Institute of Biomedical Engineering, Polytechnique Montreal, Montreal, QC, Canada.
2
NeuroPoly Lab, Institute of Biomedical Engineering, Polytechnique Montreal, Montreal, QC, Canada; Department of Neuroscience, Faculty of Medicine, University of Montreal, Montreal, QC, Canada.
3
McConnell Brain Imaging Centre, Montreal Neurological Institute, Montreal, Canada.
4
NeuroPoly Lab, Institute of Biomedical Engineering, Polytechnique Montreal, Montreal, QC, Canada; Department of Radiology and Biomedical Imaging, Zuckerberg San Francisco General Hospital, University of California, San Francisco, CA, USA.
5
Department of Radiology and Biomedical Imaging, Zuckerberg San Francisco General Hospital, University of California, San Francisco, CA, USA.
6
Department of Radiology, Xuanwu Hospital, Capital Medical University, Beijing, 100053, PR China.
7
Department of Radiology, Xuanwu Hospital, Capital Medical University, Beijing, 100053, PR China; Department of Radiology, Beijing Tiantan Hospital, Capital Medical University, Beijing, 100050, PR China.
8
Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden; Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Boston, USA.
9
National Institute of Radiological Sciences, Chiba, Chiba, Japan.
10
Juntendo University Hospital, Tokyo, Japan.
11
Juntendo University Hospital, Tokyo, Japan; Hospital Cochin, Paris, France.
12
Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden.
13
CHU Rennes, Radiology Department, France; Univ Rennes, Inria, CNRS, Inserm, IRISA UMR, 6074, Visages U1128, France.
14
Univ Rennes, Inria, CNRS, Inserm, IRISA UMR, 6074, Visages U1128, France; CHU Rennes, Neurology Department, France.
15
MS Unit, DPT of Neurology, University Hospital of Montpellier, France.
16
Aix Marseille Univ, CNRS, CRMBM, Marseille, France; APHM, CHU Timone, CEMEREM, Marseille, France.
17
APHM, CHU Timone, CEMEREM, Marseille, France; APHM, Department of Neurology, CHU Timone, APHM, Marseille, France.
18
Observatoire Français de la Sclérose en Plaques (OFSEP), Univ Lyon, Université Claude Bernard Lyon 1, Hospices Civils de Lyon, CREATIS-LRMN, UMR 5220 CNRS & U 1044 INSERM, Lyon, France.
19
Neuroimaging Research Unit, INSPE, Division of Neuroscience, San Raffaele Scientific Institute, Vita-Salute San Raffaele University, Milan, Italy.
20
Brigham and Women's Hospital, Harvard Medical School, Boston, USA.
21
Queen Square MS Centre, UCL Institute of Neurology, Faculty of Brain Sciences, University College London, London, UK; Center for Medical Image Computing (CMIC), Department of Medical Physics and Biomedical Engineering, University College London, London, United Kingdom.
22
Queen Square MS Centre, UCL Institute of Neurology, Faculty of Brain Sciences, University College London, London, UK.
23
Vanderbilt University, Tennessee, USA.
24
Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Boston, USA.
25
National Institute of Neurological Disorders and Stroke, National Institutes of Health, Maryland, USA.
26
Biospective Inc., Montreal, QC, Canada.
27
Division of Neurosurgery, Department of Surgery, University of Toronto, Toronto, ON, Canada.
28
Neurology Department, Stanford University, USA; Functional Neuroimaging Unit, CRIUGM, Université de Montréal, Montreal, QC, Canada.
29
McConnell Brain Imaging Centre, Montreal Neurological Institute, Montreal, Canada; Functional Neuroimaging Unit, CRIUGM, Université de Montréal, Montreal, QC, Canada.
30
NYU Langone Medical Center, New York, USA.
31
NeuroPoly Lab, Institute of Biomedical Engineering, Polytechnique Montreal, Montreal, QC, Canada; Functional Neuroimaging Unit, CRIUGM, Université de Montréal, Montreal, QC, Canada. Electronic address: https://skype.com/jcohenadad.

Abstract

The spinal cord is frequently affected by atrophy and/or lesions in multiple sclerosis (MS) patients. Segmentation of the spinal cord and lesions from MRI data provides measures of damage, which are key criteria for the diagnosis, prognosis, and longitudinal monitoring in MS. Automating this operation eliminates inter-rater variability and increases the efficiency of large-throughput analysis pipelines. Robust and reliable segmentation across multi-site spinal cord data is challenging because of the large variability related to acquisition parameters and image artifacts. In particular, a precise delineation of lesions is hindered by a broad heterogeneity of lesion contrast, size, location, and shape. The goal of this study was to develop a fully-automatic framework - robust to variability in both image parameters and clinical condition - for segmentation of the spinal cord and intramedullary MS lesions from conventional MRI data of MS and non-MS cases. Scans of 1042 subjects (459 healthy controls, 471 MS patients, and 112 with other spinal pathologies) were included in this multi-site study (n = 30). Data spanned three contrasts (T1-, T2-, and T2-weighted) for a total of 1943 vol and featured large heterogeneity in terms of resolution, orientation, coverage, and clinical conditions. The proposed cord and lesion automatic segmentation approach is based on a sequence of two Convolutional Neural Networks (CNNs). To deal with the very small proportion of spinal cord and/or lesion voxels compared to the rest of the volume, a first CNN with 2D dilated convolutions detects the spinal cord centerline, followed by a second CNN with 3D convolutions that segments the spinal cord and/or lesions. CNNs were trained independently with the Dice loss. When compared against manual segmentation, our CNN-based approach showed a median Dice of 95% vs. 88% for PropSeg (p ≤ 0.05), a state-of-the-art spinal cord segmentation method. Regarding lesion segmentation on MS data, our framework provided a Dice of 60%, a relative volume difference of -15%, and a lesion-wise detection sensitivity and precision of 83% and 77%, respectively. In this study, we introduce a robust method to segment the spinal cord and intramedullary MS lesions on a variety of MRI contrasts. The proposed framework is open-source and readily available in the Spinal Cord Toolbox.

KEYWORDS:

Convolutional neural networks; MRI; Multiple sclerosis; Segmentation; Spinal cord

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center