Automatic magnetic resonance spinal cord segmentation with topology constraints for variable fields of view
Introduction
The human spinal cord is a long thin cylindrical structure of the central nervous system (CNS) extending from the medulla oblongata to the lumbar vertebrae. It is the principal transmission pathway for neural signals between the brain and the rest of the body. The primary function of the spinal cord makes it of great importance in studying diseases that lead to deterioration in CNS function, such as multiple sclerosis (MS). In vivo magnetic resonance imaging (MRI) of the human spinal cord presents a unique diagnostic tool in studying the progression and characteristics of such neurological diseases.
The usefulness of MRI based analysis of the spinal cord can be traced as far back as Losseff et al. (1996) who demonstrated strong association between spinal cord area and disability as measured by Kurtzke's Expanded Disability Status Scale (Kurtzke, 1983) (EDSS) (r = − 0.7, p < 0.001). Since then, significant progress has been made in both the analysis and application of MRI in spinal cord imaging. Kalkers et al. (2002) proposed using MRI derived metrics to evaluate neuroprotective therapies. Studies from Lin et al., 2003, Lin et al., 2004 demonstrated that changes in edge detectability of the spinal cord boundary are related to changes in clinical disability. More recent work (Rocca et al., 2011, Zackowski et al., 2009) has shown the potential of using quantities derived from various imaging modalities as biomarkers to characterize patients with MS. Other studies (Freund et al., 2011) have shown that cross-sectional area of the spinal cord is well correlated with cortical activity.
A common requirement for these studies is a full or partial segmentation of the spinal cord (see Fig. 1) for each subject in the study. Such segmentations are currently performed manually or semi-automatically by human raters, which create two immediate disadvantages. First, human raters are prone to unintended biases and inconsistency in their work. This is particularly common when segmenting small structures such as the spinal cord, and is evident when replicating a segmentation of the same image or comparing between two separate raters. Second, raters require extensive training and time to perform the task. This imposes a strict limitation on the scale of future studies and produces potentially long delays between acquiring the data and completing the analyses.
There have been several attempts (Archip et al., 2002, Burnett et al., 2004, Karangelis and Zimeras, 2002, Nyúl et al., 2005) to automate the segmentation of the spinal cord in computed tomography (CT) imaging. However, such methods are limited by a lack of soft tissue contrast in CT, making it difficult to distinguish between the spinal cord itself and the surrounding cerebrospinal fluid (CSF). Most approaches are restricted to segmenting only the spinal canal, which is insufficient for analysis of spinal cord atrophy. This, in addition to concerns for patient safety, makes MRI a superior choice for imaging the spinal cord. However, MR imaging is not without its own difficulties. Inhomogeneities in receiver coil sensitivity can manifest as spatially distributed intensity biases. Susceptibility may create artifacts in the spinal cord proximal to the posterior fossa region (McGowan and Patel, 2000). Image quality is also degraded by truncation artifacts (Czervionke et al., 1988), ghosting artifacts from the heart and great vessels (Bronskill et al., 1988, Curtin et al., 1989, Hinks and Quencer, 1988, Levy et al., 1988) and contrast (Lycklama et al., 2003). Non-uniformity correction is particularly important for acquisitions from phased-array coils used to assess spinal cord atrophy (Lin et al., 2004).
These drawbacks have delayed the development of fully automated MR spinal segmentation tools. As such, the majority of the methods presented thus far to address this problem have been semi-automated in nature (Coulon et al., 2002, Horsfield et al., 2010, McIntosh and Hamarneh, 2006, Nieniewski and Serneels, 2002, Van Uitert et al., 2005). These approaches vary from the watershed based (Nieniewski and Serneels, 2002) to applications of deformable models (McIntosh and Hamarneh, 2006). To the best of our knowledge there are only three fully automatic methods in the literature for human MRI spinal cord segmentation (Koh et al., 2010, Koh et al., 2011, Mukherjee et al., 2010). Koh et al. (2010) developed a gradient vector flow (Xu and Prince, 1998) magnitude approach as part of a computer-aided diagnosis (CAD) system. Their algorithm estimates the spinal cord using the magnitude of the gradient vector flow edge map, followed by a connected component analysis to remove any holes in the segmentation. In Koh et al. (2011), the same group developed a different approach to the problem using active contour models (Kass et al., 1988) based on saliency maps. Mukherjee et al. (2010) also applied an active contour approach, but instead evolved an image gradient based, open-ended contour using dynamic programming-based energy-minimization. They initialize their method using an estimation of the vertebra bone contour in each 2D slice of the image, which is found using an optimal shortest path directed graph search based on gradient magnitude and gradient orientation. These 2D contours are then evolved under an active contour model that minimizes an energy based on the symmetry of the contour and the smoothness between successive contours (i.e., contours on adjoining 2D slices).
A common deficiency with current automated algorithms is their limitation to both a single MR-sequence and a particular field of view. The three existing methods mentioned above are designed to be used with T2 and T2*-weighted MRIs because it offers the best soft tissue contrast (Koh et al., 2010, Koh et al., 2011, Mukherjee et al., 2010). None of them provide an intuitive and easily generalized approach for addressing MR-sequences or fields of view outside of the particular dataset they were designed for. This is problematic in spinal cord MR imaging because data is generally only collected across specific subsections of the spinal cord, and the MR-sequence is rarely standardized between datasets. Addressing these limitations is one focus of our work.
There are various reasons why existing image segmentation technologies cannot be readily applied to spinal cord MRIs. For example, the structure of the spinal cord makes typical atlas-based registration highly inaccurate. This happens for two reasons; first, the long thin cylindrical nature of the spinal cord and its small size relative to the neck and torso lead to the spinal cord contributing only minimally to the overall cost function of a registration algorithm. This typically causes the registration algorithm to prioritize the alignment of other structures over the spinal cord during the optimization. Second, the spinal cord is a flexible non-rigid structure, which results in a large degree of variability in both the shape and curvature of the structure in MRIs (see Fig. 2). This variability removes rigid and affine registrations as viable choices for the transformation. Even for free-form deformable registration the accuracy is dramatically limited due to the large deformations often required to properly align the curvatures. This is particularly true for registration with strict regularization constraints on the deformation.
Unsupervised intensity based segmentations encounter a different set of problems. They are prone to misclassification due to the partial voluming of nerve roots and the strong intensity inhomogeneities from the spine coils. The inhomogeneity from the surface coil is primarily a result of the MR signal dropping-off for tissues further away from the coils. However, the problem is exacerbated by the curvature of the spinal cord. Since the cord is not parallel to the coils, the anatomy interacts with the intensity inhomogeneity unevenly. As a result, intensity values along the spinal cord are inconsistent and depend on its curvature during acquisition. This effect is particularly evident for images covering large fields of view, where both the distance from the coil and the curvature of the spinal cord is larger. Fig. 3 shows an example of an axially acquired magnetization transfer-prepared T2*-weighted gradient-echo image and a sagittally acquired T1-weighted MRI of the spinal cord, each from separate subjects. Both images have been segmented with a fuzzy c-means (Bezdek, 1980) approach (Pham, 2001) which includes gain field correction and regularization. We see that the spinal cord could not be properly segmented in any of the cases, including those where the image was manually truncated to include only the spinal cord and CSF. The intensity drop-off seen in the examples could not be handled by the inhomogeneity correction built into the classification algorithm, nor by preprocessing with N3 (Sled et al., 1998), an intensity non-uniformity correction tool commonly used in whole head MRI.
In this work we present a fully automated spinal cord segmentation algorithm that combines deformable registration with topology preserving intensity classification. Our approach is built upon a method by Bazin and Pham (2008), which uses a topological and statistical atlas in the fuzzy c-means model to classify tissues in the brain. We introduce a topological atlas that is appropriate for the spinal cord, and a statistical atlas that is dynamically adjusted to match its variability. In addition, we augment the framework with an intensity atlas that is used with deformable registration to allow the atlases to be properly initialized. Finally, we present a rapid approach for generating the necessary topology and statistical atlases from a single manual segmentation. This provides a quick and automatic procedure for adapting our method to better match datasets with very different fields of views and MR-sequences (e.g., T1-weighted and T2*-weighted). An early version of this work was presented in conference form (Chen et al., 2011), where we demonstrated its capabilities on a limited dataset with a fixed field of view. Our algorithm is freely available as part of the Java Integrated Science Toolkit (JIST) (Lucas et al., 2010), which can be downloaded at http://www.nitrc.org/projects/jist/.
Section snippets
Methods
The topology-preserving, anatomy-driven segmentation (TOADS) algorithm (Bazin and Pham, 2008) provides the main model for our segmentation approach. It is a fuzzy c-means (FCM) (Bezdek, 1980) based intensity classification algorithm that is capable of preserving the digital topology of the anatomy being segmented. The principal idea behind the model is to use prior knowledge about the target anatomy and its surrounding structures to constrain the topology of the final segmentation. This
Materials
Our algorithm was applied to two datasets having different population characteristics, MR protocols and scanners, and fields of view.
Experimental results
We perform several experiments to demonstrate the performance and applications of our spinal cord segmentation tool. Our first experiment evaluates the accuracy of our algorithm relative to human raters. It also considers the effect of our registration choice for initializing our atlases. We then evaluate the impact of our statistical atlas construction approach, and the size of the Gaussian smoothing kernel used in its construction. Lastly, we perform a large scale evaluation using the MT
Discussion
In this work we have presented a fully automatic approach for segmenting the spinal cord and cerebrospinal fluid from MRIs. Unlike existing methods, our approach is designed to be highly generalizable to spinal cord images of any field of view or MR contrast. The only criterion necessary for our algorithm is a reasonable registration between an atlas and the image being segmented. For data where such a registration is unreliable (i.e., if the images differs too greatly from the provided atlas),
Conclusion
MRI of the spinal cord presents many challenges, such as noise and artifacts, which make automatic segmentation of the spinal cord and CSF a difficult task. We have presented a topology preserving approach for addressing this problem, and have shown its effectiveness for both accuracy and robustness. In addition, as spinal cord imaging rarely has a standard field of view or MR contrast, we assume that the atlases provided with our algorithm are not optimal for every type of spinal cord MRI
Acknowledgments
Funding for this work was supported in part by NIH/NINDS grant R01-NS070906, the Intramural Research Program of NINDS, and the National MS Society (NMSS).
Conflict of interest
The authors have no conflict of interest to report regarding this work.
References (48)
- et al.
A reproducible evaluation of ANTs similarity metric performance in brain image registration
NeuroImage
(2011) - et al.
Homeomorphic brain image segmentation with topological and statistical atlases
Med. Image Anal.
(2008) - et al.
Simple paradigm for extra-cerebral tissue removal: algorithm and analysis
NeuroImage
(2011) - et al.
Motion artifacts in brain and spine MR
Radiol. Clin. N. Am.
(1988) - et al.
Rapid semi-automatic segmentation of the spinal cord from magnetic resonance images: application in multiple sclerosis
NeuroImage
(2010) - et al.
Evaluation of volume-based and surface-based brain image registration methods
NeuroImage
(2010) - et al.
Spinal-cord MRI in multiple sclerosis
Lancet Neurol.
(2003) - et al.
Technical issues for MRI examination of the posterior fossa
J. Neurol. Sci.
(2000) Spatial models for fuzzy clustering
Comp. Vision Image Underst.
(2001)- et al.
An overlap invariant entropy measure of 3D medical image alignment
Pattern Recogn.
(1999)
Semi-automatic spinal cord segmentation and quantification
Multimodal volume registration by maximation of mutual information
Med. Image Anal.
A knowledge-based approach to automatic detection of the spinal cord in CT images
IEEE Trans. Med. Imag.
Digital homeomorphisms in deformable registration
A convergence theorem for the fuzzy ISODATA clustering algorithms
IEEE Trans. Pattern Anal. Mach. Intell.
Syrinx-like artifacts on MR images of the spinal cord
Radiology
A deformable-model approach to semi-automatic segmentation of CT images demonstrated by application to the spinal canal
Med. Phys.
Multi-channel enhancement of the adaptive bases algorithm
Topology preserving automatic segmentation of the spinal cord in magnetic resonance images
Quantification of spinal cord atrophy from magnetic resonance images via a B-spline active surface model
Magn. Res. Med.
MR imaging artifacts of the axial internal anatomy of the cervical spinal cord
Am. J. Roentgenol.
Characteristic features of MR truncation artifacts
Am. J. Roentgenol.
Measure of the amount of ecologic association between species
Ecology
Disability, atrophy and cortical reorganization following spinal cord injury
Brain
Cited by (67)
Lumbar Spinal Canal Segmentation in Cases with Lumbar Stenosis Using Deep-U-Net Ensembles
2023, World NeurosurgeryA pipeline to quantify spinal cord atrophy with deep learning: Application to differentiation of MS and NMOSD patients
2021, Physica MedicaCitation Excerpt :El Mendili et al. [15] offered a double threshold-based segmentation method wherein the first threshold approximately segmented SC using Otsu’s thresholding, and the second was the same proposed by Losseff [8]. Chen et al. [16] proposed an atlas-based method in which a non-linear transformation was computed between the dataset and an intensity atlas of SC. Recently, advancing machine learning algorithms such as deep learning methods have revolutionized the field of data analysis [17], which enable computer vision researchers to substantially enhance the performance of biomedical image processing [17].
Automatic spinal cord segmentation from axial-view MRI slices using CNN with grayscale regularized active contour propagation
2021, Computers in Biology and MedicineMapping the rest of the human connectome: Atlasing the spinal cord and peripheral nervous system
2021, NeuroImageCitation Excerpt :Understanding the anatomic and connectomic variability of the SC and spinal nerves across patients is crucial in many clinical settings. A review of the recent clinical literature on PNS disorders suggests that a sophisticated 3D atlas of the PNS could substantially enhance the ability to diagnose a wide range of conditions, including peripheral neuropathy or plexopathy (Takahara and Kwee, 2010; Healy et al., 2020), peripheral and optic neuritis (Dailey et al., 1997; Maravilla and Bowen, 1998; Moore et al., 2001; Meltzer et al., 2018), peripheral nerve sheath tumors (Weber et al., 2000; Lee et al., 2020), traumatic injuries, neuralgias and nerve compressions (DeSouza et al., 2014; Konieczny et al., 2020), axonotmesis and neurotmesis (Chen et al., 2011; Chen et al., 2013), nerve root irregularities and loss of unidirectional nerve course, nerve entrapment due to carpal tunnel syndrome (Kabakci et al., 2007; Khalil et al., 2008; Stein et al., 2009; Hiltunen et al., 2012; Negm et al., 2017), etc. For many other conditions, preliminary studies indicate that the availability of a stereotactic, deformable, population-based and connectomic atlas of the SC and PNS would be useful for predicting the course of disease, for monitoring disease progression, and for monitoring the effects of therapeutic interventions (Valsasina et al., 2005; Ohgiya et al., 2007; Cruz et al., 2009; Bosma and Stroman, 2012; Tarawneh et al., 2020).
Voxelwise analysis of diffusion MRI of cervical spinal cord using tract-based spatial statistics
2020, Magnetic Resonance ImagingCorrelation between EDSS scores and cervical spinal cord atrophy at 3T MRI in multiple sclerosis: A systematic review and meta-analysis
2020, Multiple Sclerosis and Related DisordersCitation Excerpt :The methods adopted in more than two studies were the Losseff method (Losseff et al., 1996) (n = 4) and the Horsfield method (Horsfield et al., 2010) (n = 10). The remaining studies used other semiautomated or automated segmentation methods including a threshold-based method (Cohen et al., 2012), the Chen method (Chen et al., 2013), the Lukas method (Lukas et al., 2004), the Canny method (Canny, 1986) and PropSeg (Yiannakas et al., 2016). We separated the studies into the Losseff method group, the Horsfield method group and the “other methods” group according to spinal cord segmentation method.