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

Charley Gros 1 Benjamin de Leener 1 Atef Badji 1 Josefina Maranzano 2 Dominique Eden 1 Sara Dupont 1, 3 Jason Talbott 3 Ren Zhuoquiong 4 Yaou Liu 4 Tobias Granberg 5 Russell Ouellette 6 Yasuhiko Tachibana 7 Masaaki Hori 8 Kouhei Kamiya 8 Lydia Chougar 8, 9 Leszek Stawiarz 10 Jan Hillert 10 Elise Bannier 11, 12 Anne Kerbrat 11, 12 Gilles Edan 13, 12 Pierre Labauge 14 Virginie Callot 15 Jean Pelletier 15 Bertrand Audoin 15 Henitsoa Rasoanandrianina 15 Jean-Christophe Brisset 16 Paola Valsasina 17 Maria Rocca 17 Massimo Filippi 17 Rohit Bakshi 18 Shahamat Tauhid 19 Ferran Prados 20 Marios Yiannakas 21 Hugh Kearney 21 Olga Ciccarelli 21 Seth Smith 22 Constantina Andrada Treaba 23 Caterina Mainero 23 Jennifer Lefeuvre 24 Daniel Reich 24 Govind Nair 24 Vincent Auclair 25 Donald Mclaren 25 Allan Martin 26 Michael Fehlings 26 Shahabeddin Vahdat 27, 28 Ali Khatibi 27, 29 Julien Doyon 29, 27 Timothy Shepherd 30 Erik Charlson 30 Sridar Narayanan 29 Julien Cohen-Adad 1
12 VisAGeS - Vision, Action et Gestion d'informations en Santé
INSERM - Institut National de la Santé et de la Recherche Médicale : U1228, Inria Rennes – Bretagne Atlantique , IRISA-D5 - SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE
20 NMR Research Unit [London]
Institute of Neurology [London]
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.
Document type :
Journal articles
Complete list of metadatas
Contributor : Elise Bannier <>
Submitted on : Monday, November 26, 2018 - 9:28:23 AM
Last modification on : Friday, September 13, 2019 - 9:50:02 AM

Links full text



Charley Gros, Benjamin de Leener, Atef Badji, Josefina Maranzano, Dominique Eden, et al.. Automatic segmentation of the spinal cord and intramedullary multiple sclerosis lesions with convolutional neural networks. NeuroImage, Elsevier, 2019, 184, pp.901 - 915. ⟨10.1016/j.neuroimage.2018.09.081⟩. ⟨hal-01934566⟩



Record views