Elsevier

NeuroImage

Volume 52, Issue 4, 1 October 2010, Pages 1347-1354
NeuroImage

Automated brain tissue segmentation based on fractional signal mapping from inversion recovery Look–Locker acquisition

https://doi.org/10.1016/j.neuroimage.2010.05.001Get rights and content

Abstract

Most current automated segmentation methods are performed on T1- or T2-weighted MR images, relying on relative image intensity that is dependent on other MR parameters and sensitive to B1 magnetic field inhomogeneity. Here, we propose an image segmentation method based on quantitative longitudinal magnetization relaxation time (T1) of brain tissues. Considering the partial volume effect, fractional volume maps of brain tissues (white matter, gray matter, and cerebrospinal fluid) were obtained by fitting the observed signal in an inversion recovery procedure to a linear combination of three exponential functions, which represents the relaxations of each of the tissue types. A Look–Locker acquisition was employed to accelerate the acquisition process. The feasibility and efficacy of this proposed method were evaluated using simulations and experiments. The potential applications of this method in the study of neurological disease as well as normal brain development and aging are discussed.

Introduction

Automated segmentation of brain tissues in magnetic resonance (MR) images has been widely used in the study of brain structure and function (Ashburner and Friston, 2000, Ashburner and Friston, 2005, Thompson et al., 1997). Various segmentation algorithms have been proposed to make binary or non-binary (fractional volume or probability) maps for different brain tissue types, such as white matter (WM), gray matter (GM) and cerebrospinal fluid (CSF). Based on the imaging modalities/contrasts used, the unsupervised segmentation methods can be classified into single-contrast and multi-contrast approaches. For segmentation algorithms based solely on a single contrast, the spin-lattice relaxation time (T1) weighted image has been popularly used due to its capability of acquiring a high-resolution image within feasible time. Since the feasibility of multi-contrast MR imaging for brain segmentation was demonstrated using spin echo and inversion recovery sequences (Vannier et al., 1985), multi-spectral or multi-contrast segmentation methods have demonstrated the advantage of providing a variety of extra information for voxel classification. However, in practice the relatively long acquisition time and additional registration between multi-contrast images have made such methods less popular.

Due to the relatively small size of macroscopic brain structures (in particular cortical foldings) with respect to the conventional voxel size in MR images, the partial volume effect (PVE) has been a main concern for accurate segmentation, particularly for methods relying on a single image intensity/contrast. Non-binary tissue segmentation maps, such as fractional volume or probability of brain tissue components have been proposed to address the PVE. However, it is challenging to generate non-binary tissue maps using single-contrast based algorithms without a priori templates because signal intensity in a voxel contains different tissue types.

Various segmentation methods analyzing the intensity distribution of T1-weighted (T1w) images have been proposed to generate non-binary tissue maps using Gaussian mixture models (Santago and Gage, 1993, Shattuck et al., 2001) and Markov random field models based on Gaussian mixture models (Held et al., 1997, Rajapakse et al., 1997, Zhang et al., 2001). The theoretical description of the abovementioned methods and others have been reviewed in detail (Bezdek et al., 1993, Clarke et al., 1995, Cuadra et al., 2005). The intrinsic weakness of those techniques is their dependence on the T1w image contrast which is related not only to T1 but also to other MR variables such as proton density and spin-spin relaxation time (T2), and the T1w image contrast can be modulated by the transmit and receive radio-frequency (RF) fields, particularly when surface/phase-array coils are used.

Brain segmentation methods using quantitative or semi-quantitative MR parameters have been proposed. Diffusion tensor imaging for example, was utilized for brain segmentation, using the apparent diffusion coefficient and fractional anisotropy to classify the tissue types (Liu et al., 2007). A quantitative T1 map obtained with a fast whole-brain T1 mapping technique was used for brain segmentation (Hetherington et al., 1996). Recently, a high-resolution (1.25 mm3) magnetization transfer (MT) contrast mapping technique with RF inhomogeneity and T1 relaxation corrections was developed (Helms et al., 2008) and employed to classify the subcortical GM, resulting in improved delineation of those brain areas (Helms et al., 2009).

In this study, we present an automated brain tissue segmentation method, FRActional Signal mapping from InvErsion Recovery (FRASIER), which classifies brain tissue based on quantitative T1. FRASIER observes the signal in a dynamic inversion recovery (IR) procedure and fits the data to a linear combination of three different exponential functions, which represent the relaxations of WM, GM and CSF, respectively. Non-binary fractional signal (fs) maps are obtained from the model fitting, and the fs maps are then converted into fractional volume (fv) maps based on the density of those brain tissues. Data collection with the FRASIER method is significantly accelerated using Look–Locker (LL) acquisition (Look and Locker, 1970). In this study, the performance of fv measurement using FRASIER is evaluated by simulations and in vivo experiments.

Section snippets

Automated segmentation

A fast T1 measurement method using IR LL echo-planar imaging at a steady state (IR LL-EPI SS) has been recently proposed (Shin et al., 2009a). In the current study, IR LL-EPI SS method was employed for fast T1 and fv mapping to accelerate scan time. IR LL-EPI SS observes the apparent longitudinal magnetization relaxation (T1) instead of T1 in an IR procedure. Under this condition T1 can be expressed as a function of T1, TR and flip angle, α (1/T1 = 1/T1 ln(cosα)/TR), and the relaxation of

Results

Fig. 1 shows the representative whole-brain T1 and T1 histograms and individual measurements of T1 and T1 in WM and GM. The two lowest Gaussian distributions in the histograms were likely derived from voxels with partial volume between WM and GM (green) and between GM and CSF (yellow). Individual whole-brain T1 and T1 histograms over the 11 subjects are shown in Fig. 2. From the individual T1 histograms, average T1 values in WM and GM were measured to be 925 ± 28 ms and 1531 ± 43 ms, respectively,

Discussion

The PVE is one of the main difficulties related to brain tissue segmentation of MR images. High-resolution MR imaging, such as 1 mm3 isotropic T1w imaging, has been widely used to minimize PVE in segmentation. In this study, we propose a new fv measurement technique by investigating the multi-exponential patterns of the longitudinal magnetization relaxation. FRASIER utilizes the well-understood T1 relaxation in an IR procedure to segment brain tissues. However, the current FRASIER protocol

Conclusion

We have developed a new segmentation approach, FRASIER, which uses quantitative T1s (or T1s) in brain tissues to obtain fractional signal mapping. The FRASIER method can be applied to any IR procedure, and can be accelerated using LL acquisition. In the present study, the FRASIER method provided whole-brain T1 and fv maps within 4.5 min. Experimental data demonstrated that the GM, WM, and CSF of the brain were effectively segmented using FRASIER. This segmentation method would provide an

Acknowledgments

This work was supported by the Intramural Research Program of the National Institute on Drug Abuse (NIDA), National Institutes of Health. The authors would like to thank Dr. Thomas Ross and Dr. Annabelle Belcher of the NIDA for his helpful discussions and suggestions, and Dr. Leon Axel for suggesting the validation of FRASIER using the down-sampling approach.

References (35)

  • M.B. Cuadra et al.

    Comparison and validation of tissue modelization and statistical classification methods in T1-weighted MR brain images

    IEEE Trans. Med. Imaging

    (2005)
  • L.W. de Jong et al.

    Strongly reduced volumes of putamen and thalamus in Alzheimer's disease: an MRI study

    Brain

    (2008)
  • M.J. Donahue et al.

    Theoretical and experimental investigation of the VASO contrast mechanism

    Magn. Reson. Med.

    (2006)
  • T. Ethofer et al.

    Comparison of longitudinal metabolite relaxation times in different regions of the human brain at 1.5 and 3 Tesla

    Magn. Reson. Med.

    (2003)
  • Y. Ge et al.

    Quantitative assessment of iron accumulation in the deep gray matter of multiple sclerosis by magnetic field correlation imaging. AJNR Am

    J. Neuroradiol.

    (2007)
  • K. Held et al.

    Markov random field segmentation of brain MR images

    IEEE Trans. Med. Imaging

    (1997)
  • G. Helms et al.

    High-resolution maps of magnetization transfer with inherent correction for RF inhomogeneity and T1 relaxation obtained from 3D FLASH MRI

    Magn. Reson. Med.

    (2008)
  • Cited by (16)

    • P-ResUnet: Segmentation of brain tissue with Purified Residual Unet

      2022, Computers in Biology and Medicine
      Citation Excerpt :

      Over the decades, magnetic resonance imaging (MRI) technology has been extensively used in the medical field [1–6] and has significantly promoted the development of medicine, neurophysiology, and cognitive neuroscience. The accurate segmentation of brain MRI slices into various tissues, i.e, gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF), is essential in the research of brain development [7], which can assist clinicians to diagnose neurological diseases such as Alzheimer’s disease, Epilepsy, Parkinson’s disease, Multiple sclerosis, and schizophrenia [8–11]. This step has typically been done by a radiologist who spends significant time on manual segmentation.

    • MhURI:A Supervised Segmentation Approach to Leverage Salient Brain Tissues in Magnetic Resonance Images

      2021, Computer Methods and Programs in Biomedicine
      Citation Excerpt :

      Computerized techniques are increasingly being employed for tasks such as image registration, segmentation, and visualization to render a quick medical diagnosis. In brain MRI segmentation, there has been a significant effort to partition the whole of the image into sub-portions like WM, GM, and CSF of the brain [1,2]. In contrast, some works concentrate on the extraction of a specific structure like brain tumors [3], multiple sclerosis lesions [4], or ischemic stroke lesion [5].

    • Investigation of the neurovascular coupling in positive and negative BOLD responses in human brain at 7T

      2014, NeuroImage
      Citation Excerpt :

      Statistical analysis was done using FSL Feat (Version 5.98) (Worsley, 2001). In order to estimate voxel-wise tissue composition independent of slice orientation and image distortions, data from multiple TI experiments were used to generate fractional volume maps of GM, white matter (WM) and cerebrospinal fluid (CSF) (Shin et al., 2010). Assumed T1 values were T1,WM/T1,GM/T1,CSF = 1.4/1.9/4.2 s. Functional ROIs were defined based on the BOLD data as a cluster of voxels having z-values above 2.0 and a significance level of p < 0.05 (automatically corrected for multiple comparisons in FSL based upon GRF theory), after smoothing with a kernel size of 1 mm.

    View all citing articles on Scopus
    View full text