Fast method for brain image segmentation: Application to proton magnetic resonance spectroscopic imaging

Bonekamp, David; Horská, Alena; Jacobs, Michael A.; Arslanoğlu, Atilla; Barker, Peter B.

doi:10.1002/mrm.20657

Cited by 12 publications

(12 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although single voxel 1 H-MRS studies often attempt to circumvent this issue by placing the VOI in "mostly" WM or GM (31), partial volume effects quantified in Figure 2, SNR limitations, and misregistration all introduce quantification errors. A remedy in 1 H-MRSI is to apply linear regression to the metabolites' concentrations and either the GM or WM voxel volume fractions (13,(32)(33)(34)(35)(36)(37). In such linear model, the voxels' signals are assumed to be of the form given by Equation [4], while the constraint V jk…”

Section: Discussionmentioning

confidence: 99%

“…The misregistration, SNR and partial volume issues can be substantially reduced by combining absolute 1 H-MRSI metabolic quantification with anatomical high-spatial resolution (~1 mm 3 ) MRI that accompanies it. Using freely available segmentation software, WM/GM/CSF masks can be produced and overlaid on the 1 H-MRSI grid to yield their contents in each voxel (13,14). This information can yield global WM and GM metabolite concentrations by modeling the 1 H-MRSI signal from each voxel as a linear combination of their contributions.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The role of gray and white matter segmentation in quantitative proton MR spectroscopic imaging

et al. 2012

View full text Add to dashboard Cite

Since the brain's gray matter (GM) and white matter (WM) metabolite concentrations differ, their partial volumes can vary the voxel's 1H MR spectroscopy (1H-MRS) signal, reducing the sensitivity to changes. While single-voxel 1H-MRS cannot differentiate the WM from GM signals, partial volume correction is feasible in MR spectroscopic imaging (MRSI), using segmentation of the MRI that is always acquired for VOI placement. To determine the magnitude of this effect on metabolic quantification, we segmented the 1 mm3 resolution MRI into GM, WM and CSF masks that were co-registered with the MRSI grid to yield their partial volumes in every ~1 cm3 spectroscopic voxel. Each voxel then provided one equation with two unknowns – its i- metabolite's GM and WM concentrations: CiGM, CiWM. With the voxels' GM and WM volumes as independent coefficients, that over-determined system of equations can be solved for the global, averaged CiGM and CiWM. Trading off local concentrations differences offers three advantages: (i) higher sensitivity due to combined data from many voxels; (ii) improved specificity to WM versus GM changes; (iii) reduced susceptibility to partial volume effects. These improvements make no additional demands of the protocol, measurement time or hardware. Applying the approach to 18 volunteers' 3D MRSI sets, 480 voxels each, yielded N-acetylaspartate, creatine, choline and myo-inositol CiGM of 8.5±0.7, 6.9±0.6, 1.2±0.2, 5.3±0.6 mM; and CiWM of 7.7±0.6, 4.9±0.5, 1.4±0.1 and 4.4±0.6 mM. We show that unaccounted voxel WM or GM partial volume can vary absolute quantification by 5–10% (more for ratios) that can often as much as double the sample size required to establish statistical significance.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

The role of gray and white matter segmentation in quantitative proton MR spectroscopic imaging

et al. 2012

View full text Add to dashboard Cite

show abstract

“…A plethora of anatomical experiments with different contrast weightings can be conducted to provide useful constraints, and the particular sets that are both practical and useful will vary depending on the context of the experiment. In our experiments, we typically acquire T 1 ‐, T 2 ‐, and proton density‐weighted images, since these can accurately differentiate between the tissues of interest in the brain (27). In experiments involving pathology, companion scans including contrast enhancement or diffusion tensor imaging are often also conducted; these additional datasets can be used to derive even more useful constraints.…”

Section: Methodsmentioning

confidence: 99%

Anatomically constrained reconstruction from noisy data

Haldar¹,

Hernando²,

Song³

et al. 2008

Magnetic Resonance in Med

106

124

View full text Add to dashboard Cite

Noise is a major concern in many important imaging applications. To improve data signal-to-noise ratio (SNR), experiments often focus on collecting low-frequency k -space data. This article proposes a new scheme to enable extended k -space sampling in these contexts. It is shown that the degradation in SNR associated with extended sampling can be effectively mitigated by using statistical modeling in concert with anatomical prior information. The method represents a significant departure from most existing anatomically constrained imaging methods, which rely on anatomical information to achieve super-resolution. The method has the advantage that less accurate anatomical information is required relative to super-resolution approaches. Theoretical and experimental results are provided to characterize the performance of the proposed scheme. Magn Reson Med 59:810-818, 2008.

show abstract

“…As a consequence, mapping of V CSF at baseline will benefit from more quantitative interpretation of functional MRI results. Similarly, the knowledge of V CSF in each voxel is also important for determination of brain metabolite concentrations in proton MR spectroscopic imaging (7–11).…”

mentioning

confidence: 99%

A simple approach for three‐dimensional mapping of baseline cerebrospinal fluid volume fraction

Qin

2010

Magnetic Resonance in Med

View full text Add to dashboard Cite

A simple method of measuring baseline cerebrospinal fluid volume fraction (V CSF ) in three-dimensional is proposed that used the characteristic of cerebrospinal fluid with very long T 2 . It is based on the fitting of monoexponential decay of only cerebrospinal fluid signal, using a nonselective T 2 preparation scheme. Three-dimensional gradient-and spin-echo acquisition also improves signal-to-noise ratio efficiency and brain coverage. Both V CSF and T 2,CSF are fitted voxel by voxel and analyzed in different cortical areas across subjects. V CSF is largely regionally dependent (occipital: 8.9 6 1.7%, temporal: 11.4 6 2.4%, and frontal: 21.4 6 6.9%). Measured T 2,CSF was 1573 6 146 msec within cortical lobes as compared with 2062 6 37 msec from ventricle regions. Different parameter set were compared, and the robustness of the new method is demonstrated. Conversely, when comparing with the proposed approach, large overestimation of segmentation based method using T 1 -weighted images is found, and the underlying causes are suggested. Magn Reson Med 65:385-391, 2011. V C 2010 Wiley-Liss, Inc.Key words: CSF volume fraction; T 2 prep; GRASE Long considered as just a partial volume confounder in functional MRI, cerebrospinal fluid (CSF) volume fraction (V CSF ) is known to have large variations between different regions of a subject, especially for cortical gray matter and periventricular areas. As people age, CSF volume starts to increase, whereas brain volume steadily declines (1,2). Recently, a more active role in functional MRI contrast mechanism is being investigated for CSF: namely, redistribution in response to local blood volume changes during activation (3-6). As a consequence, mapping of V CSF at baseline will benefit from more quantitative interpretation of functional MRI results. Similarly, the knowledge of V CSF in each voxel is also important for determination of brain metabolite concentrations in proton MR spectroscopic imaging (7-11).One group of MRI methods to determine tissue composition is based on multicompartment T 2 decay fitting (7,12). This type of techniques need very high signal-tonoise ratio, which requires long time for both acquisition and analysis. A newly proposed promising method (13) acquired T 2 0 decay curves that were fitted with a sophisticated model for oxygen extraction fraction and other parameters including deoxygenated blood volume and CSF volume fraction. This method was recently employed for V CSF measurement (14). Another group of methods use segmentation of high-resolution threedimensional (3D) T 1 -weighted (8,15) or T 2 -weighted (9,16) images. Extracortical CSF (subarachnoid spaces) could not be distinguished from the skull in T 1 -weighted anatomical images because both appear dark (10). Even with high resolution, voxels could still contain CSF partial volume but are segmented as tissue and thus bring in errors for mapping V CSF .Jin and Kim (6) have recently measured both baseline and functional change of V CSF on an animal scanner with a surface receiving coil mo...

show abstract

Fast method for brain image segmentation: Application to proton magnetic resonance spectroscopic imaging

Cited by 12 publications

References 20 publications

The role of gray and white matter segmentation in quantitative proton MR spectroscopic imaging

The role of gray and white matter segmentation in quantitative proton MR spectroscopic imaging

Anatomically constrained reconstruction from noisy data

A simple approach for three‐dimensional mapping of baseline cerebrospinal fluid volume fraction

Contact Info

Product

Resources

About