Limits...
Feasibility of Computed Tomography-Guided Methods for Spatial Normalization of Dopamine Transporter Positron Emission Tomography Image.

Kim JS, Cho H, Choi JY, Lee SH, Ryu YH, Lyoo CH, Lee MS - PLoS ONE (2015)

Bottom Line: The CT images were modified in two ways, skull-stripping (ssCT) and intensity transformation (itCT).The SUVR values measured with FreeSurfer-generated VOIs (FSVOI) overlaid on original PET images were also used as a gold standard for comparison.Putaminal SUVR values were highly effective for discriminating PD patients from controls.

View Article: PubMed Central - PubMed

Affiliation: Molecular Imaging Research Center, Korea Institute Radiological and Medical Sciences, Seoul, South Korea.

ABSTRACT

Background: Spatial normalization is a prerequisite step for analyzing positron emission tomography (PET) images both by using volume-of-interest (VOI) template and voxel-based analysis. Magnetic resonance (MR) or ligand-specific PET templates are currently used for spatial normalization of PET images. We used computed tomography (CT) images acquired with PET/CT scanner for the spatial normalization for [18F]-N-3-fluoropropyl-2-betacarboxymethoxy-3-beta-(4-iodophenyl) nortropane (FP-CIT) PET images and compared target-to-cerebellar standardized uptake value ratio (SUVR) values with those obtained from MR- or PET-guided spatial normalization method in healthy controls and patients with Parkinson's disease (PD).

Methods: We included 71 healthy controls and 56 patients with PD who underwent [18F]-FP-CIT PET scans with a PET/CT scanner and T1-weighted MR scans. Spatial normalization of MR images was done with a conventional spatial normalization tool (cvMR) and with DARTEL toolbox (dtMR) in statistical parametric mapping software. The CT images were modified in two ways, skull-stripping (ssCT) and intensity transformation (itCT). We normalized PET images with cvMR-, dtMR-, ssCT-, itCT-, and PET-guided methods by using specific templates for each modality and measured striatal SUVR with a VOI template. The SUVR values measured with FreeSurfer-generated VOIs (FSVOI) overlaid on original PET images were also used as a gold standard for comparison.

Results: The SUVR values derived from all four structure-guided spatial normalization methods were highly correlated with those measured with FSVOI (P < 0.0001). Putaminal SUVR values were highly effective for discriminating PD patients from controls. However, the PET-guided method excessively overestimated striatal SUVR values in the PD patients by more than 30% in caudate and putamen, and thereby spoiled the linearity between the striatal SUVR values in all subjects and showed lower disease discrimination ability. Two CT-guided methods showed comparable capability with the MR-guided methods in separating PD patients from controls and showed better correlation between putaminal SUVR values and the parkinsonian motor severity than the PET-guided method.

Conclusion: CT-guided spatial normalization methods provided reliable striatal SUVR values comparable to those obtained with MR-guided methods. CT-guided methods can be useful for analyzing dopamine transporter PET images when MR images are unavailable.

No MeSH data available.


Related in: MedlinePlus

Image processing steps for acquiring skull-stripped CT (ssCT) and intensity-transformed CT (itCT) templates.(a) Inhomogeneity correction and segmentation of MR, (b) creation of whole brain and CSF masks, (c) skull-stripping of inhomogeneity-corrected MR, (d) spatial normalization of skull-stripped MR to MNI template for skull-stripped MR, (e) spatial normalization of masks by applying normalization parameter, (f) creation of probabilistic maps for whole brain and CSF by averaging tissue masks, (g) coregistration of CT to inhomogeneity-corrected MR, (h) creation of skull mask, (i) spatial normalization of skull mask, (j) creation of probabilistic map for skull, (k) spatial normalization of coregistered CT, (l) creation of whole CT template by averaging, (m) creation of template mask for scalp-stripping, (n) spatial normalization of CT to whole CT template, (o) inverse normalization of template mask for scalp-stripping to create individual mask for scalp-stripping, (p) creation of scalp-stripped CT by applying mask, (q) segmentation of scalp-stripped CT into three tissue type by using probabilistic maps, (r) creation of ssCT by applying mask for whole brain segment, (s) coregistration of ssCT to MR by applying parameter coregistering original CT to MR, (t) spatial normalization of coregistered ssCT, (u) creation of ssCT template by averaging, (v) intensity transformation of original CT, (w) coregistration of itCT to MR, (x) spatial normalization of coregistered itCT, (y) creation of itCT template by averaging, and (z) spatial normalization of individual ssCT and itCT to specific CT templates. The processing steps inside the red (ssCT) and blue (itCT) dashed lines represent the image processing steps required for the spatial normalization of CT images with created CT templates.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4492980&req=5

pone.0132585.g001: Image processing steps for acquiring skull-stripped CT (ssCT) and intensity-transformed CT (itCT) templates.(a) Inhomogeneity correction and segmentation of MR, (b) creation of whole brain and CSF masks, (c) skull-stripping of inhomogeneity-corrected MR, (d) spatial normalization of skull-stripped MR to MNI template for skull-stripped MR, (e) spatial normalization of masks by applying normalization parameter, (f) creation of probabilistic maps for whole brain and CSF by averaging tissue masks, (g) coregistration of CT to inhomogeneity-corrected MR, (h) creation of skull mask, (i) spatial normalization of skull mask, (j) creation of probabilistic map for skull, (k) spatial normalization of coregistered CT, (l) creation of whole CT template by averaging, (m) creation of template mask for scalp-stripping, (n) spatial normalization of CT to whole CT template, (o) inverse normalization of template mask for scalp-stripping to create individual mask for scalp-stripping, (p) creation of scalp-stripped CT by applying mask, (q) segmentation of scalp-stripped CT into three tissue type by using probabilistic maps, (r) creation of ssCT by applying mask for whole brain segment, (s) coregistration of ssCT to MR by applying parameter coregistering original CT to MR, (t) spatial normalization of coregistered ssCT, (u) creation of ssCT template by averaging, (v) intensity transformation of original CT, (w) coregistration of itCT to MR, (x) spatial normalization of coregistered itCT, (y) creation of itCT template by averaging, and (z) spatial normalization of individual ssCT and itCT to specific CT templates. The processing steps inside the red (ssCT) and blue (itCT) dashed lines represent the image processing steps required for the spatial normalization of CT images with created CT templates.

Mentions: For creation of the ssCT template, the intensity of CT images was first linearly transformed by adding minimum voxel values to eliminate negative voxel values. The skull-stripped MR images were obtained by merging the gray and white matter segments of inhomogeneity-corrected T1-weighted MR images (Fig 1a–1c) and normalized to the skull-stripped Montreal Neurological Institute (MNI) 152 MR template to derive spatial normalization parameters for MR images (Fig 1d). Probabilistic maps for whole brain and cerebrospinal fluid (CSF) were created by averaging the spatially normalized masks of each MR segment (Fig 1e and 1f). Also, the probabilistic map for skull was created with the spatially normalized masks of skull segments of CT images (Fig 1g–1j). The CT images coregistered to individual MR images were spatially normalized with the spatial normalization parameters for MR images (Fig 1g–1k), and we created a temporary whole CT template and a scalp-stripping mask, including skull, whole brain, CSF and soft tissue below the convexity of skull (Fig 1l and 1m). All CT images were spatially normalized to this whole CT template (Fig 1n) and scalp-stripped by applying inversely normalized scalp-stripping mask images (Fig 1o and 1p). Scalp-stripped CT images were segmented into skull, whole brain, and CSF using the probabilistic maps for each tissue type in SPM segmentation tool (Fig 1q), and then ssCT images were obtained by applying a binary whole brain mask to the original CT images (Fig 1r). Finally, the ssCT template was created by averaging all ssCT images normalized to the MNI template space with spatial normalization parameters for MR images (Fig 1s–1u). Detailed image processing steps for acquiring the ssCT template were described in our previous paper [14].


Feasibility of Computed Tomography-Guided Methods for Spatial Normalization of Dopamine Transporter Positron Emission Tomography Image.

Kim JS, Cho H, Choi JY, Lee SH, Ryu YH, Lyoo CH, Lee MS - PLoS ONE (2015)

Image processing steps for acquiring skull-stripped CT (ssCT) and intensity-transformed CT (itCT) templates.(a) Inhomogeneity correction and segmentation of MR, (b) creation of whole brain and CSF masks, (c) skull-stripping of inhomogeneity-corrected MR, (d) spatial normalization of skull-stripped MR to MNI template for skull-stripped MR, (e) spatial normalization of masks by applying normalization parameter, (f) creation of probabilistic maps for whole brain and CSF by averaging tissue masks, (g) coregistration of CT to inhomogeneity-corrected MR, (h) creation of skull mask, (i) spatial normalization of skull mask, (j) creation of probabilistic map for skull, (k) spatial normalization of coregistered CT, (l) creation of whole CT template by averaging, (m) creation of template mask for scalp-stripping, (n) spatial normalization of CT to whole CT template, (o) inverse normalization of template mask for scalp-stripping to create individual mask for scalp-stripping, (p) creation of scalp-stripped CT by applying mask, (q) segmentation of scalp-stripped CT into three tissue type by using probabilistic maps, (r) creation of ssCT by applying mask for whole brain segment, (s) coregistration of ssCT to MR by applying parameter coregistering original CT to MR, (t) spatial normalization of coregistered ssCT, (u) creation of ssCT template by averaging, (v) intensity transformation of original CT, (w) coregistration of itCT to MR, (x) spatial normalization of coregistered itCT, (y) creation of itCT template by averaging, and (z) spatial normalization of individual ssCT and itCT to specific CT templates. The processing steps inside the red (ssCT) and blue (itCT) dashed lines represent the image processing steps required for the spatial normalization of CT images with created CT templates.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4492980&req=5

pone.0132585.g001: Image processing steps for acquiring skull-stripped CT (ssCT) and intensity-transformed CT (itCT) templates.(a) Inhomogeneity correction and segmentation of MR, (b) creation of whole brain and CSF masks, (c) skull-stripping of inhomogeneity-corrected MR, (d) spatial normalization of skull-stripped MR to MNI template for skull-stripped MR, (e) spatial normalization of masks by applying normalization parameter, (f) creation of probabilistic maps for whole brain and CSF by averaging tissue masks, (g) coregistration of CT to inhomogeneity-corrected MR, (h) creation of skull mask, (i) spatial normalization of skull mask, (j) creation of probabilistic map for skull, (k) spatial normalization of coregistered CT, (l) creation of whole CT template by averaging, (m) creation of template mask for scalp-stripping, (n) spatial normalization of CT to whole CT template, (o) inverse normalization of template mask for scalp-stripping to create individual mask for scalp-stripping, (p) creation of scalp-stripped CT by applying mask, (q) segmentation of scalp-stripped CT into three tissue type by using probabilistic maps, (r) creation of ssCT by applying mask for whole brain segment, (s) coregistration of ssCT to MR by applying parameter coregistering original CT to MR, (t) spatial normalization of coregistered ssCT, (u) creation of ssCT template by averaging, (v) intensity transformation of original CT, (w) coregistration of itCT to MR, (x) spatial normalization of coregistered itCT, (y) creation of itCT template by averaging, and (z) spatial normalization of individual ssCT and itCT to specific CT templates. The processing steps inside the red (ssCT) and blue (itCT) dashed lines represent the image processing steps required for the spatial normalization of CT images with created CT templates.
Mentions: For creation of the ssCT template, the intensity of CT images was first linearly transformed by adding minimum voxel values to eliminate negative voxel values. The skull-stripped MR images were obtained by merging the gray and white matter segments of inhomogeneity-corrected T1-weighted MR images (Fig 1a–1c) and normalized to the skull-stripped Montreal Neurological Institute (MNI) 152 MR template to derive spatial normalization parameters for MR images (Fig 1d). Probabilistic maps for whole brain and cerebrospinal fluid (CSF) were created by averaging the spatially normalized masks of each MR segment (Fig 1e and 1f). Also, the probabilistic map for skull was created with the spatially normalized masks of skull segments of CT images (Fig 1g–1j). The CT images coregistered to individual MR images were spatially normalized with the spatial normalization parameters for MR images (Fig 1g–1k), and we created a temporary whole CT template and a scalp-stripping mask, including skull, whole brain, CSF and soft tissue below the convexity of skull (Fig 1l and 1m). All CT images were spatially normalized to this whole CT template (Fig 1n) and scalp-stripped by applying inversely normalized scalp-stripping mask images (Fig 1o and 1p). Scalp-stripped CT images were segmented into skull, whole brain, and CSF using the probabilistic maps for each tissue type in SPM segmentation tool (Fig 1q), and then ssCT images were obtained by applying a binary whole brain mask to the original CT images (Fig 1r). Finally, the ssCT template was created by averaging all ssCT images normalized to the MNI template space with spatial normalization parameters for MR images (Fig 1s–1u). Detailed image processing steps for acquiring the ssCT template were described in our previous paper [14].

Bottom Line: The CT images were modified in two ways, skull-stripping (ssCT) and intensity transformation (itCT).The SUVR values measured with FreeSurfer-generated VOIs (FSVOI) overlaid on original PET images were also used as a gold standard for comparison.Putaminal SUVR values were highly effective for discriminating PD patients from controls.

View Article: PubMed Central - PubMed

Affiliation: Molecular Imaging Research Center, Korea Institute Radiological and Medical Sciences, Seoul, South Korea.

ABSTRACT

Background: Spatial normalization is a prerequisite step for analyzing positron emission tomography (PET) images both by using volume-of-interest (VOI) template and voxel-based analysis. Magnetic resonance (MR) or ligand-specific PET templates are currently used for spatial normalization of PET images. We used computed tomography (CT) images acquired with PET/CT scanner for the spatial normalization for [18F]-N-3-fluoropropyl-2-betacarboxymethoxy-3-beta-(4-iodophenyl) nortropane (FP-CIT) PET images and compared target-to-cerebellar standardized uptake value ratio (SUVR) values with those obtained from MR- or PET-guided spatial normalization method in healthy controls and patients with Parkinson's disease (PD).

Methods: We included 71 healthy controls and 56 patients with PD who underwent [18F]-FP-CIT PET scans with a PET/CT scanner and T1-weighted MR scans. Spatial normalization of MR images was done with a conventional spatial normalization tool (cvMR) and with DARTEL toolbox (dtMR) in statistical parametric mapping software. The CT images were modified in two ways, skull-stripping (ssCT) and intensity transformation (itCT). We normalized PET images with cvMR-, dtMR-, ssCT-, itCT-, and PET-guided methods by using specific templates for each modality and measured striatal SUVR with a VOI template. The SUVR values measured with FreeSurfer-generated VOIs (FSVOI) overlaid on original PET images were also used as a gold standard for comparison.

Results: The SUVR values derived from all four structure-guided spatial normalization methods were highly correlated with those measured with FSVOI (P < 0.0001). Putaminal SUVR values were highly effective for discriminating PD patients from controls. However, the PET-guided method excessively overestimated striatal SUVR values in the PD patients by more than 30% in caudate and putamen, and thereby spoiled the linearity between the striatal SUVR values in all subjects and showed lower disease discrimination ability. Two CT-guided methods showed comparable capability with the MR-guided methods in separating PD patients from controls and showed better correlation between putaminal SUVR values and the parkinsonian motor severity than the PET-guided method.

Conclusion: CT-guided spatial normalization methods provided reliable striatal SUVR values comparable to those obtained with MR-guided methods. CT-guided methods can be useful for analyzing dopamine transporter PET images when MR images are unavailable.

No MeSH data available.


Related in: MedlinePlus