Elsevier

NeuroImage

Volume 54, Supplement 1, January 2011, Pages S180-S188
NeuroImage

Automatic MRI segmentation and morphoanatomy analysis of the vestibular system in adolescent idiopathic scoliosis

https://doi.org/10.1016/j.neuroimage.2010.04.002Get rights and content

Abstract

The vestibular system is the sensory organ responsible for perceiving head rotational movements and maintaining postural balance of human body. The objectives of this study are to propose an innovative computational technique capable of automatically segmenting the vestibular system and to analyze its geometrical features from high resolution T2-weighted MR images. In this study, the proposed technique was used to test the hypothesis that the morphoanatomy of vestibular system in adolescent idiopathic scoliosis (AIS) patients is different from healthy control subjects. The findings could contribute significantly to the understanding of the etiopathogenesis of AIS. The segmentation pipeline consisted of extraction of region of interest, image pre-processing, K-means clustering, and surface smoothing. The geometry of this high-genus labyrinth structure was analyzed through automatic partition into genus-0 units and approximation using the best-fit circle and plane for each unit. The metrics of the best-fit planes and circles were taken as shape measures. The proposed technique was applied on a cohort of 20 right-thoracic AIS patients (mean age 14.7 years old) and 20 age-matched healthy girls. The intermediate results were validated by subjective scoring. The result showed that the distance between centers of lateral and superior canals and the angle with vertex at the center of posterior canal were significantly smaller in AIS than in healthy controls in the left-side vestibular system with p = 0.0264 and p = 0.0200 respectively, but not in the right-side counterparts. The detected morphoanatomical changes are likely to be associated with subclinical postural, vestibular and proprioceptive dysfunctions reported frequently in AIS. This study has demonstrated that the proposed method could be applied in MRI-based morphoanatomy studies of vestibular system clinically.

Introduction

The vestibular system plays an important role in maintaining the body equilibrium by sensing body movements and sending signals to the brain. Located in the inner ear compartment of the temporal bone, the vestibular system contains both osseous and membranous components. The membranous labyrinth is filled with endolymph and surrounded by perilymph within the osseous labyrinth. The vestibular system itself consists of three semicircular canals, namely lateral, posterior and superior canals, which are approximately orthogonal to each other, with cross-sectional diameter of about 0.8 mm each (Gray, 1985). Mathematically, the vestibular labyrinth is of a special genus-3 topology.

Anatomical and functional research on vestibular system has been carried out for decades in the fields of otorhinolaryngology and neuroscience (Truex and Carpenter, 1969). Thanks to the state-of-the-art imaging technologies, the 3-D structure of the vestibular system can be imaged non-invasively. However, current work on automatic segmentation and shape analysis of this complex organ is still very limited. In the study by Seemann et al. (2005), the auditory and vestibular systems were segmented in an interactive way to fuse the CT and MRI data for visualization purpose. In Vinchon's study (2007), the orientations of the vestibular system were used to determine the internal coordinate system for cranio-facial surgery. The 3-D view of the vestibular system has also been reconstructed based on the contours labeled in serial celloidin sections (Dai et al., 2004), while volume rendering has been applied to view the 3-D structure reconstructed from transmission electron microscopy sections of rats (Ross et al., 1989).

MRI is an ideal non-invasive modality for imaging in vivo semicircular canals without ionizing radiation (Gunny and Yousry, 2007). Due to the presence of the fluid (i.e., endolymph and perilymph), the vestibular system can be visualized in T2-weighted magnetic resonance images (T2W MRI). T2W fast spin echo sequence is capable of imaging the endolymph and perilymph with high signals, and the osseous labyrinth walls with low signals. Therefore, this study was performed based on the T2W MRI scans.

Adolescent idiopathic scoliosis (AIS) is defined as a three-dimensional structural deformity of the spine that occurs in otherwise healthy children, predominantly adolescent girls. AIS affects about 4% of the adolescent population worldwide. Etiopathogenesis of AIS is still not completely certain (Miller, 1999). To enable evidence-based treatment and prognosis of curve progression, research efforts to reveal the etiology of AIS have remained active for decades (Cowell et al., 1972, Fidler and Jowett, 1976, Harrington, 1977, Raso, 2000, Burwell, 2003).

Poor postural balance control has been reported to be associated with AIS patients in multiple studies. The postural balance control in human requires proper collaboration of somatosensory, visual, and vestibular systems. Byl and Gray (1993) reported that in an experiment involving 50 AIS patients and 20 age-matched normal controls, the AIS group had a similar response as healthy participants in stable static balance conditions. However, once their visual and somatosensory systems were simultaneously challenged, the AIS patients had a significantly higher mean body sway than the controls. AIS patients with moderate to severe curves had poorer performance compared with mild cases in maintaining their balance when the visual and somatosensory systems were challenged simultaneously. With a different set of measures, AIS patients were found to rely much more on ankle proprioception to regulate the amplitude of the balance control when compared with healthy participants (Simoneau et al., 2006). Guo et al. (2006) reported the association of abnormal somatosensory evoked potentials (SSEPs) with AIS, suggesting the abnormal somatosensory function in AIS. Herman et al. (1985) proposed that the altered processing of vestibular information contributes to the etiology of AIS.

In the current study, as motivated by the above findings and our hypothesis that AIS is associated with morphoanatomical changes in the vestibular system, we propose a computational approach to quantitatively evaluate and compare the morphoanatomy of the vestibular system in T2W MRI, with the comparison of AIS and healthy controls being one clinical application.

Section snippets

Participants and imaging protocol

As AIS has significantly higher prevalence in girls than boys with moderate-to-severe scoliotic curves (Winiarski et al., 2005, Raggio, 2006), only girls were recruited in this study. 20 right-thoracic AIS girls (7 severe; 8 moderate; 5 mild) and 20 age-matched healthy girls were recruited from the Scoliosis Clinic of Prince of Wales Hospital, Hong Kong. The detailed demographic information of the two groups is shown in Table 1. All subjects were right-handed Chinese with normal neurological

Results

In order to validate the segmentation results, we performed the quality control over all the segmentation results by visually comparing the segmented surface with the volume data in the ROI visualized using volume rendering. The bottom row of Fig. 8 shows an example of the bilateral vestibular segmentation results, which exhibited good similarity with the volume rendering of the original data (top row of Fig. 8). For each subject involved in our data analysis, the quality of all segmented

Discussions

Recently, studies have been performed to analyze the morphological brain changes in AIS, such as the splenium of corpus callosum (Wang et al., 2009), local white matter attenuation (Shi et al., 2009), and regional brain volumes (Liu et al., 2008). However, little research has been carried out to study the detailed morphology of vestibular system in AIS. This paper demonstrated an innovative vestibular segmentation pipeline using the following major components, namely (1) automatic ROI

Conflict of interest statement

The authors declare that there are no conflicts of interest.

Acknowledgments

The work described in this paper was supported by a grant from the Research Grants Council of the Hong Kong Special Administrative Region, China (Project ID: CUHK 4453/06M).

References (32)

  • M. Fidler et al.

    Muscle imbalance in the aetiology of scoliosis

    J. Bone Joint Surg. Br.

    (1976)
  • W. Gander et al.

    Least-Squares Fitting of Circles and Ellipses

    (1994)
  • H. Gray

    Anatomy of the Human Body: Gray's Anatomy

    (1985)
  • R. Gunny et al.

    Imaging anatomy of the vestibular and visual systems

    Curr. Opin. Neurol.

    (2007)
  • X. Guo et al.

    Balance control in adolescents with idiopathic scoliosis and disturbed somatosensory function

    Spine

    (2006)
  • P. Harrington

    The etiology of idiopathic scoliosis

    Clin. Orthop.

    (1977)
  • Cited by (54)

    • Landmark constrained registration of high-genus surfaces applied to vestibular system morphometry

      2015, Computerized Medical Imaging and Graphics
      Citation Excerpt :

      Length and thickness of each canal were measured and the difference between two groups of subjects was shown statistically significant. Partition approaches have also been proposed to analyze and register VS surfaces by partitioning them into simple open surface patches [28]. However, partition approaches greatly rely on the consistency of surface cuts, which are usually difficult to obtain.

    View all citing articles on Scopus
    1

    For technical issues.

    2

    For clinical issues.

    View full text