Abstract
Purpose
The precise quantification of dopamine transporter (DAT) density on N-(3-[18F]Fluoropropyl)-2β-carbomethoxy-3β-(4-iodophenyl) nortropane positron emission tomography ([18F]FP-CIT PET) imaging is crucial to measure the degree of striatal DAT loss in patients with parkinsonism. The quantitative analysis requires a spatial normalization process based on a template brain. Since the spatial normalization method based on a delayed-phase PET has limited performance, we suggest an early-phase PET-based method and compared its accuracy, referring to the MRI-based approach as a gold standard.
Methods
A total of 39 referred patients from the movement disorder clinic who underwent dual-phase [18F]FP-CIT PET and took MRI within 1 year were retrospectively analyzed. The three spatial normalization methods were applied for quantification of [18F]FP-CIT PET-MRI-based anatomical normalization, PET template-based method based on delayed PET, and that based on early PET. The striatal binding ratios (BRs) were compared, and voxelwise paired t tests were implemented between different methods.
Results
The early image-based normalization showed concordant patterns of putaminal [18F]FP-CIT binding with an MRI-based method. The BRs of the putamen from the MRI-based approach showed higher agreement with early image- than delayed image-based method as presented by Bland-Altman plots and intraclass correlation coefficients (early image-based, 0.980; delayed image-based, 0.895). The voxelwise test exhibited a smaller volume of significantly different counts in putamen between brains processed by early image and MRI compared to that between delayed image and MRI.
Conclusion
The early-phase [18F]FP-CIT PET can be utilized for spatial normalization of delayed PET image when the MRI image is unavailable and presents better performance than the delayed template-based method in quantitation of putaminal binding ratio.
Keywords: Dopamine transporter imaging, [18F]FP-CIT PET, Early-phase image, Spatial normalization, Quantitative analysis
Introduction
Dopamine transporter (DAT) imaging has been widely applied to evaluate the function of presynaptic dopaminergic neurons [1]. The typical patterns of DAT density loss along the rostrocaudal and ventrodorsal axis of a striatum aid in the diagnosis of Parkinson’s disease [2–4]. Among the imaging modalities, positron emission tomography (PET) using N-(3-[18F]Fluoropropyl)-2β-carbomethoxy-3β-(4-iodophenyl) nortropane ([18F]FP-CIT) exhibits highly specific binding to striatum with good spatial resolution [5, 6]. The quantitative evaluation of striatal [18F]FP-CIT binding has been successfully applied for specifying the diagnosis of movement disorder [2]. During the quantitation of [18F]FP-CIT binding, spatial normalization is a prerequisite process to use predefined anatomical regions including striatum on the template space [7, 8]. An anatomical normalization with magnetic resonance image (MRI) is considered as a gold standard [9], but it is not an option in cases where MRI is unavailable. Another method based on the PET-only approach has been used. The averaged [18F]FP-CIT PET image was obtained from normal subjects and used it as a template for normalization [2, 10]. However, the PET template-based normalization overestimates striatal [18F]FP-CIT binding, particularly in Parkinson’s disease patients [11].
Recently, the role of the early-phase [18F]FP-CIT PET image has been investigated extensively [12–15]. Regional cerebral perfusion is coupled to glucose metabolism so that early-phase [18F]FP-CIT shows a similar uptake pattern to 2-deoxy-2-[18F]fluoro-D-glucose ([18F]FDG) PET. Moreover, the early-phase image successfully replaced [18F] FDG PET in the differential diagnosis of atypical parkinsonism [13]. An application of the early-phase image can be extended to spatial normalization. The early-phase and delayed-phase images are paired individual data; thus, normalization parameters for early-phase image can be directly applied to the delayed image.
Here we suggest a method to spatially normalize a delayed-phase [18F]FP-CIT image based on an early-phase image. The delayed-phase image is spatially normalized with predetermined parameters produced during normalization of an early-phase image to [15O]H2O PET template, which is widely used for the spatial normalization process in [18F] FDG PET. The research aims to show the feasibility of early-phase image-based normalization by comparison with the conventional delayed-phase template-based approach, which presents a limited performance in Parkinson’s disease patients.
Materials and Methods
Patients
Consecutive 154 patients who visited the movement disorder clinic and planned for [18F]FP-CIT PET exam due to parkinsonism from November 2019 to February 2020 were retrospectively enrolled. Then, 52 subjects who underwent dual-phase PET acquisition were included. After excluding 11 patients who did not take MRI within 1 year and 2 patients with structural abnormality identified on MRI, 39 subjects remained for further analysis. These 39 patients have no history of stroke or other brain disorders affecting brain structures according to the MRI findings.
[18F]FP-CIT PET Acquisition
All of the patients were instructed to discontinue drugs affecting DAT density for at least 12 h. 185 MBq (5 mCi) of [18F]FP-CIT was injected intravenously to the patient and an early-phase image was acquired immediately. After 2 h of rest, a delayed-phase image was acquired. For the early-phase, a PET image was obtained followed by a computed tomography (CT) image. PET was acquired for 5 min, matrix size was 400 × 400 pixels, and reconstruction was implemented by ordered-subset expectation maximization (OSEM) method with 6 times of iteration and 21 subsets. The time of flight (TOF) technique was adopted during the reconstruction. Gaussian filter with a full width half maximum (FWHM) 4.0 mm was applied to the image, and attenuation correction was performed using CT image. CT was acquired with tube voltage 120 kV, tube current 150 mA, and a slice thickness of 3.0 mm. For the delayed-phase, CT was obtained followed by 10 min of PET acquisition. A two-dimensional point spread function-based algorithm (True-X) was adopted for PET reconstruction. Other parameters for CT and PET acquisition were identical to the early-phase PET.
Visual Assessment of [18F]FP-CIT PET
The visual assessment of [18F]FP-CIT binding in bilateral putamen and caudate nuclei was performed by two physicians (S.B. and H.C.) with experience in [18F]FP-CIT PET for more than 3 years and 9 years, respectively. The PET images were divided into two groups according to the visual assessment: normal and abnormal scans. The putamen or caudate nucleus with preserved [18F]FP-CIT binding was considered as a normal finding.
Processing and Quantitation of [18F]FP-CIT PET
PET images obtained from both phases were coregistered and resliced to MRI T1- or T2-weighted images. The spatial normalization was performed to move count data of each voxel into Montreal Neurological Institute (MNI) space. First, for MRI-based spatial normalization, the MRI image was segmented with tissue probability map (TPM.nii) into gray matter, white matter, cerebrospinal fluid, bone, soft tissue, and air. The deformation field for segmentation was applied to spatially normalize the delayed PET image. Second, for delayed image-based spatial normalization, we utilized a template generated by averaging delayed images from 75 subjects [10]. Third, for early image-based normalization, [15O]H2O PET template was utilized (PET.nii). Predetermined parameters for early image normalization were applied to normalize delayed image (Fig. 1). A software Statistical Parametric Mapping 12 (SPM12) (Wellcome Centre for Human Neuroimaging, London, UK) was utilized based on MATLAB R2019b (MathWorks, Natick, MA, USA) throughout the whole process except for the PET-based normalization processes which were performed in SPM8. A rex toolbox was applied and divided a brain into 116 regions by automated anatomical labeling (AAL) atlas (https://web.mit.edu/swg/software.htm). Mean counts in each putamen, caudate nucleus, and occipital lobe (superior, middle, and inferior occipital lobe) were measured. The degree of striatal binding was expressed as the putamen and caudate nucleus binding ratios (BRs). A BR of the putamen was defined as [(Mean counts in unilateral putamen)/(Mean counts in occipital lobe)] − 1 and BR of the caudate as [(Mean counts in unilateral caudate)/(Mean counts in occipital lobe)] − 1.
Fig. 1.
A schematic image for processing [18F]FP-CIT PET. [18F]FP-CIT PET was spatially normalized with three different methods. The binding ratios (BRs) of the putamen and caudate nucleus were calculated and compared. Also, voxelwise paired t tests were implemented between groups of processed brains. (1) A delayed phase PET is coregistered to MRI (T1 or T2 image) and resliced. (2) An early-phase PET is coregistered to MRI and resliced. (3) MRI is segmented with tissue probability map (TPM.nii) and the deformation field is calculated. (4) The deformation field for segmentation is applied to spatially normalize coregistered delayed PET. (5) The delayed image is normalized with [18F]FP-CIT PET template. (6) A coregistered early image is normalized to [15O]H2O PET template (PET.nii). (7) The parameters utilized in process 6 are directly applied to normalize the delayed PET. (8) BRs are compared between the MRI-based method and two PET-based approaches. The voxelwise paired t tests are performed to discover voxels with significantly different counts
Statistical Analysis
Frist, BRs of the putamen and caudate from MRI-based spatial normalization were compared to those from early or delayed image-based normalization (Fig. 1). Each side of BRs was respectively evaluated: thus, 78 putamen and caudate BRs were calculated for each method. Scatter plots between two variables were drawn, and Pearson correlation coefficients are calculated. Also, Bland-Altman plots were generated and the mean difference with 95% confidence interval was visualized. To measure agreement between different methods, two-way random, average score intraclass correlation coefficient (ICC) values for BRs were calculated. Subgroup analyses were performed in the putamen and caudate for normal and abnormal DAT density in visual assessment. P values below 0.05 are considered statistically significant. The whole process was performed in R version 4.0.2 (R Foundations for Statistical Computing, Vienna, Austria) with ggplot2 (version 3.3.2) and irr (version 0.84.1) packages.
Second, a voxelwise paired t test was implemented. The voxelwise comparison was performed on brains processed by MRI-based and early or delayed image-based normalization in SPM12 (Fig. 1). A brain mask provided by SPM12 (mask_ICV.nii) was applied to each brain during the analysis. The voxelwise multiple comparison correction was performed with the family-wise error (FWE) rate and the cutoff was 10−3. The extent of the threshold was 50 for all the analyses. For the visualization of t values, MRIcron (https://www.nitrc.org/projects/mricron) was utilized. The striatal (putamen and caudate nucleus) mask produced in WFU PickAtlas (https://www.nitrc.org/projects/wfu_pickatlas/) was applied and t-maps were overlayed on single subject T1 MRI image (single_subj_T1.nii).
Results
Subjects
Among the 39 subjects, 16 subjects were males and 23 subjects were females. The range of age was from 50 to 84 and the median age was 64 years. A median time interval between the PET and the MRI exam was 2.07 months (Table 1). The [18F]FP-CIT binding in putamen was preserved in 37.18% (29/78) while in the caudate nucleus it was maintained in 71.79% (56/78).
Table 1.
Characteristics of enrolled subjects
| Characteristics (n = 39) | |
|---|---|
| Age (years) | 64 [50–84]‡ |
| Gender | |
| Male | 16 (41.03%) |
| Female | 23 (58.97%) |
| Time interval between MRI and PET (months) | 2.07 [0–11.27]‡ |
| Visual assessment of the putamen | |
| Bilateral normal | 14 (35.90%) |
| Unilateral abnormal | 1 (2.56%) |
| Bilateral abnormal | 24 (61.54%) |
| Visual assessment of the caudate nucleus | |
| Bilateral normal | 27 (69.23%) |
| Unilateral abnormal | 2 (5.13%) |
| Bilateral abnormal | 10 (25.64%) |
‡The values are reported as median [minimum value–maximum value]
Comparison of Binding Ratios in Different Spatial Normalization Methods
Scatter plots between striatal BRs (putamen and caudate nucleus) by MRI-based normalization and two different PET-based normalization methods were presented (Fig. 2a, b). Pearson correlation coefficient for BRs of the putamen was 0.98 between MRI and early image which was higher than 0.90 between MRI and delayed image. The correlation coefficients for BR of the caudate were 0.90 and 0.75 in early and delayed image-based approaches, respectively. Also, in the subgroup analysis, both normal and abnormal groups showed a higher correlation between MRI and early image than between MRI- and delayed-image-based methods (Fig. 3). Bland-Altman plots revealed higher agreement of BR of the putamen between MRI and early image compared to MRI and delayed image (MRI-Early, 0.24 ± 0.68; MRI-Delay, 0.31 ± 1.49 [mean difference ± 1.96 × standard deviation]) (Fig. 4a). According to the Bland-Altman plots, early image-based normalization did not show a significant trend of BR difference with the MRI-based method in the putamen. On the other hand, the delayed image-based method overestimated BRs of the putamen in lower values while underestimated in higher values. Meanwhile, BRs of the caudate were overall underestimated in the early image and overestimated in the delayed image (MRI-Early, 0.68 ± 1.02; MRI-Delay, − 0.20 ± 1.50) (Fig. 4b). The subgroup analysis showed similar patterns of agreement in both normal and abnormal groups (Fig. 5). The representative cases for overestimation and underestimation of putaminal [18F]FP-CIT binding in delayed image-based approach were exhibited in Fig. 6. Agreement between MRI- and PET-based binding ratios was presented with ICC values. Two-way random, average score ICC values for BR of the putamen were 0.980 between MRI and early image (95% confidence interval [CI] 0.934 to 0.991) while 0.895 between MRI and delayed image (95% CI 0.809 to 0.939). For BR of the caudate, ICC values for MRI early image was 0.840 (95% CI − 0.066 to 0.950) and for MRI delayed image was 0.813 (95% CI 0.703 to 0.882) (Table 2). The subgroup analysis revealed trends of higher ICC values between BRs of the putamen from MRI-early image than MRI-delayed image (Table 3).
Fig. 2.
Scatter plots between striatal binding ratios from MRI and PET-based approaches. a A scatter plot between binding ratios (BRs) of putamen derived from MRI-based and early or delayed PET-based spatial normalization. b A scatter plot between BRs of caudate nucleus derived from MRI-based and early or delayed PET-based spatial normalization. The red dots represent BRs between MRI and delayed PET and green dots for BRs between MRI and early PET-based methods. The cases with normal [18F]FP-CIT binding in the visual assessment are denoted by round dots and abnormal binding by triangular dots. The dashed lines represent points with the same BRs. Pearson’s correlation coefficients (R) and p values are written in the top-left corner of each plot
Fig. 3.
Subgroup analysis: scatter plots between striatal binding ratios from MRI- and PET-based approaches. Scatter plots between binding ratios (BRs) derived from MRI-based and early or delayed PET-based spatial normalization. The cases with a normal or b abnormal putamen [18F]FP-CIT binding in the visual assessment are exhibited in separate plots. Also, the cases with c normal or d abnormal caudate [18F]FP-CIT binding in the visual assessment are exhibited in separate plots. The red dots represent BRs between MRI and delayed PET and green dots for BRs between MRI- and early PET-based methods. The cases with normal [18F]FP-CIT binding in the visual assessment are denoted by round dots and abnormal binding by triangular dots. The dashed lines represent points with the same BRs. Pearson’s correlation coefficients (R) and p values are written in the top-left corner of each plot
Fig. 4.
Bland-Altman plots between striatal binding ratios from MRI- and PET-based methods. a Two Bland-Altman plots between binding ratios (BRs) of putamen derived from MRI-based and early or delayed PET-based spatial normalization are presented in the left and right panel, respectively. b Two Bland-Altman plots between BRs of caudate nucleus derived from MRI-based and early or delayed PET-based spatial normalization are exhibited in the left and right panel, respectively. The blue lines represent the mean difference of BRs and the values are written right above the lines. The red lines represent mean ± 1.96 × standard deviation (SD) for the difference of BRs and the values are written right above the lines
Fig. 5.
Subgroup analysis: Bland-Altman plots between striatal binding ratios from MRI- and PET-based methods. Bland-Altman plots between binding ratios (BRs) derived from MRI-based and early or delayed PET-based spatial normalization are presented in the left and right panel, respectively. The cases with a normal or b abnormal putamen [18F]FP-CIT binding in the visual assessment are exhibited in separate plots. Also, the cases with c normal or d abnormal caudate [18F]FP-CIT binding in the visual assessment are exhibited in separate plots. The blue lines represent the mean difference of BRs, and the values are written right above the lines. The red lines represent mean ± 1.96 × standard deviation (SD) for the difference of BRs, and the values are written right above the lines
Fig. 6.
Representative images showing overestimation or underestimation of putaminal binding ratios in delayed phase PET-based method. a A representative case where a binding ratio (BR) of the putamen is overestimated in delayed PET-based method compared to the MRI-based method. b A representative case where a BR of the putamen is underestimated in delayed PET-based method compared to the MRI-based method. The spatially normalized coronal, sagittal, and axial delayed PET images from delayed, early, and MRI-based approaches, and region of interests (ROIs) defined by automated anatomical labeling (AAL) atlas was placed in top-left, top-right, bottom-left, and bottom-right corners, respectively
Table 2.
Intraclass correlation coefficients (ICCs) between MRI and PET-based spatial normalization approaches
| Groups | ICC for BRs of putamen | ICC for BRs of caudate |
|---|---|---|
| MRI vs. early-phase PET | 0.980 [0.934–0.991] | 0.840 [− 0.066 to 0.950] |
| MRI vs. delayed-phase PET | 0.895 [0.809–0.939] | 0.813 [0.703–0.882] |
All the values are reported as ICC [95% confidence interval of ICC]
BRs binding ratios
Table 3.
Subgroup analysis: intraclass correlation coefficients (ICCs) between MRI and PET-based spatial normalization methods
| FP-CIT binding | Groups | ICC for BRs of putamen | ICC for BRs of caudate |
|---|---|---|---|
| Normal | MRI vs. Early PET | 0.791 [0.350–0.917] | 0.763 [− 0.223 to 0.929] |
| MRI vs. Delayed PET | 0.358 [− 0.291 to 0.736] | 0.693 [0.479–0.820] | |
| Abnormal | MRI vs. Early PET | 0.957 [0.841–0.982] | 0.931 [0.489–0.980] |
| MRI vs. Delayed PET | 0.932 [0.875–0.963] | 0.837 [0.521–0.937] |
ICC values for each subgroup, normal or abnormal [18F]FP-CIT binding in visual assessment, are presented and reported as ICC [95% confidence interval of ICC]
BRs binding ratios
SPM Analysis in Different Spatial Normalization Methods
The voxelwise paired t test was implemented to unveil which voxels contributed to the difference between MRI-based and two PET-based normalization methods. Compared with the MRI-based approach, the early image-based approach showed significantly lower counts in the right caudate nucleus while higher counts in the right ventral putamen (Fig. 7a). Meantime, in the delayed image-based method, it exhibited significantly lower counts in bilateral ventral putamina while higher counts in bilateral dorsal putamina and head of caudate nuclei (Fig. 7b).
Fig. 7.
The voxelwise paired t tests between MRI- and PET-based methods. The SPM analyses were done and voxels with significantly different counts (family-wise error rate < 10−3) between different normalization methods were visualized on single-subject T1 weighted MRI. A striatal mask was applied and only the results for the striatum were presented. a The analysis between brains processed by MRI-based and early image-based methods. b The analysis between brains processed by MRI-based and delayed image-based methods. The t values for voxels having higher counts in the PET-based approach are presented with a red color map and higher counts in the MRI-based approach with a blue color map
Discussion
An assessment of DAT density in the presynaptic neuron can be achieved with [18F]FP-CIT PET. To analyze striatal binding quantitatively, spatial normalization with [18F]FP-CIT template followed by semiautomatic segmentation is one of the commonly applied processes [2, 10, 16]. In this study, we investigated the new role of early-phase [18F]FP-CIT PET image in spatial normalization of the delayed PET image. The early image-based approach showed concordant patterns of putaminal [18F]FP-CIT binding with an MRI-based method in terms of striatal binding ratio and in the voxelwise test. Also, compared with the delayed image-based approach, the early image-based method exhibited higher agreement with the MRI-based method particularly in estimating BR of the putamen.
The overestimation and underestimation of putaminal [18F]FP-CIT binding in the delayed image was mainly observed in low and high binding ratios, respectively. The finding is consistent with the previous study where BRs of the putamen were exaggerated in Parkinson’s disease patients [11]. The SPM analysis revealed that the overestimation was caused by voxels in bilateral dorsal putamina while underestimation was by voxels in ventral putamina (Fig. 7b). The volume of significantly different voxels between brains processed by delayed PET and MRI was larger than that between early PET and MRI (Fig. 7a, b). Meanwhile, BRs of the caudate presented deviation from the gold standard in both PET-based normalization approaches. The negative deviation in the early image was originated from the right caudate nucleus and the positive deviation in the delayed image from the head of the caudate nucleus (Fig. 7a, b).
An early-phase [18F]FP-CIT PET image shows cerebral perfusion and radioactivity is observed throughout the whole gray matter; thus, it provides abundant morphological information of the brain. On the contrary, the delayed image offers morphological information only for striatum. Besides, the normalization to delayed PET template exaggerates counts, particularly in regions of DAT density loss [11]. As a result, early image facilitates more accurate spatial normalization compared to delayed image-based normalization. However, the agreement of BR of the caudate with the MRI-based approach was similar in two PET-based methods, possibly due to the relatively smaller volume of a caudate nucleus compared to a putamen. It may have affected the spatial normalization process and cut down the advantages of the early PET-based approach.
The subgroup analyses in the putamen revealed that the performance of an early image-based method depends on the degree of DAT density loss (Fig. 3a, b and Table 3). The correlation coefficients and ICC values between early image and MRI-based approaches were higher in the abnormal putamen binding group compared to the normal group. The early image-based method has comparable performance to the MRI-based method in cases of apparent DAT density loss. However, it showed a tendency of underestimation of DAT density in patients with near-normal [18F]FP-CIT binding. Even though spatial normalization using early-phase images showed better results than delay-phase images, it still had a bias in a subgroup with near-normal [18F]FP-CIT PET. Therefore, the interpretation should be cautious when early-phase image-based normalization was used in quantitative analyses of [18F]FP-CIT PET.
One of the limitations in this study is that the MRI was not obtained simultaneously with PET and the time interval between the two exams ranged from 0 to as long as 11.27 months. The anatomical changes were visually not significant between MRI and CT images obtained along with PET; thus, it may not have influenced the spatial normalization greatly. Another potential issue is that the MRI exams in each patient were implemented in different protocols and may have influenced anatomical normalization processes. However, the segmentation process could have reduced the impact of varying MRI acquisition methods on the normalization process [7].
One of the previous studies acquired dynamic [18F]FP-CIT PET at an early time point and generated a template image for spatial normalization of PET [17]. They obtained a mean template image from the enrolled subjects, and the image may be optimized to the included patient population. The greatest advantage of our method is that additional 5-min acquisition of early-phase [18F]FP-CIT PET enables accurate quantitation of putaminal DAT density loss and shows better performance than the conventional PET template-based approach. Although these results are predictable, there has been no comparison study that quantified striatal [18F]FP-CIT binding in different normalization methods. Considering that an MNI-template provided by SPM has been widely used, the approach using the early image and the SPM PET template is easy-to-use and reproducible in multiple centers. Furthermore, as early-phase images of several brain imaging tracers are similar to perfusion images [18–21], spatial normalization using an early-phase image can be widely applied to other types of brain PET imaging analyses.
Conclusion
The spatial normalization of delayed phase [18F]FP-CIT PET with early-phase image enables more precise quantitation of putaminal DAT density loss compared to the template-based approach, and it may replace the MRI-based approach when MRI is not available.
Funding
This work was supported by the National Research Foundation of Korea (NRF) grants funded by the Korea government (MSIT) (NRF-2019R1F1A1061412 and NRF-2019K1A3A1A14065446).
Compliance with Ethical Standards
Conflict of Interest
Sungwoo Bae, Hongyoon Choi, Wonseok Whi, Jin Chul Paeng, Gi Jeong Cheon, Keon Wook Kang, and Dong Soo Lee declare that they have no conflict of interest.
Ethical Approval
All procedures performed in the study regarding human participants were in accordance with the ethical standards of Helsinki declaration in 1964 and its later amendments. The study was approved by an institutional review board (IRB) of Seoul National University Hospital (IRB number: 2006-184-1135).
Informed Consent
The need for informed consent was waived by the IRB.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Palermo G, Ceravolo R. Molecular imaging of the dopamine transporter. Cells. 2019;8:872. doi: 10.3390/cells8080872. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Oh M, Kim JS, Kim JY, Shin KH, Park SH, Kim HO, et al. Subregional patterns of preferential striatal dopamine transporter loss differ in Parkinson disease, progressive supranuclear palsy, and multiple-system atrophy. J Nucl Med. 2012;53:399–406. doi: 10.2967/jnumed.111.095224. [DOI] [PubMed] [Google Scholar]
- 3.Ba F, Martin WW. Dopamine transporter imaging as a diagnostic tool for parkinsonism and related disorders in clinical practice. Parkinsonism Relat Disord. 2015;21:87–94. doi: 10.1016/j.parkreldis.2014.11.007. [DOI] [PubMed] [Google Scholar]
- 4.Hong CM, Ryu HS, Ahn BC. Early perfusion and dopamine transporter imaging using 18F-FP-CIT PET/CT in patients with parkinsonism. Am J Nucl Med Mol Imaging. 2018;8:360–372. [PMC free article] [PubMed] [Google Scholar]
- 5.Kazumata K, Dhawan V, Chaly T, Antonini A, Margouleff C, Belakhlef A, et al. Dopamine transporter imaging with fluorine-18-FPCIT and PET. J Nucl Med. 1998;39:1521–1530. [PubMed] [Google Scholar]
- 6.Lee I, Kim JS, Park JY, Byun BH, Park SY, Choi JH, et al. Head-to-head comparison of 18F-FP-CIT and 123I-FP-CIT for dopamine transporter imaging in patients with Parkinson's disease: a preliminary study. Synapse. 2018;72:e22032. doi: 10.1002/syn.22032. [DOI] [PubMed] [Google Scholar]
- 7.Ashburner J, Friston KJ. Voxel-based morphometry--the methods. Neuroimage. 2000;11:805–821. doi: 10.1006/nimg.2000.0582. [DOI] [PubMed] [Google Scholar]
- 8.Crinion J, Ashburner J, Leff A, Brett M, Price C, Friston K. Spatial normalization of lesioned brains: performance evaluation and impact on fMRI analyses. Neuroimage. 2007;37:866–875. doi: 10.1016/j.neuroimage.2007.04.065. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Gispert JD, Pascau J, Reig S, Martinez-Lazaro R, Molina V, Garcia-Barreno P, et al. Influence of the normalization template on the outcome of statistical parametric mapping of PET scans. Neuroimage. 2003;19:601–612. doi: 10.1016/S1053-8119(03)00072-7. [DOI] [PubMed] [Google Scholar]
- 10.Kim YI, Im HJ, Paeng JC, Lee JS, Eo JS, Kim DH, et al. Validation of simple quantification methods for 18F-FP-CIT PET using automatic delineation of volumes of interest based on statistical probabilistic anatomical mapping and isocontour margin setting. Nucl Med Mol Imaging. 2012;46:254–260. doi: 10.1007/s13139-012-0159-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kim JS, Cho H, Choi JY, Lee SH, Ryu YH, Lyoo CH, et al. Feasibility of computed tomography-guided methods for spatial normalization of dopamine transporter positron emission tomography image. PLoS One. 2015;10:e0132585. doi: 10.1371/journal.pone.0132585. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Jin S, Oh M, Oh SJ, Oh JS, Lee SJ, Chung SJ, et al. Differential diagnosis of parkinsonism using dual-phase F-18 FP-CIT PET imaging. Nucl Med Mol Imaging. 2013;47:44–51. doi: 10.1007/s13139-012-0182-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Jin S, Oh M, Oh SJ, Oh JS, Lee SJ, Chung SJ, et al. Additional value of early-phase 18F-FP-CIT PET image for differential diagnosis of atypical parkinsonism. Clin Nucl Med. 2017;42:e80–ee7. doi: 10.1097/RLU.0000000000001474. [DOI] [PubMed] [Google Scholar]
- 14.Min JH, Park DG, Yoon JH, An YS. Dual-phase 18F-FP-CIT PET in corticobasal syndrome. Clin Nucl Med. 2019;44:e49–e50. doi: 10.1097/RLU.0000000000002355. [DOI] [PubMed] [Google Scholar]
- 15.Yoon JH, Ahn YS. Dual-phase 18F-FP-CIT PET in corticobasal syndrome underlying AD pathology. Eur J Nucl Med Mol Imaging. 2019;46:2208–2209. doi: 10.1007/s00259-019-04376-7. [DOI] [PubMed] [Google Scholar]
- 16.Jeong E, Oh SY, Pahk K, Lee CN, Park KW, Lee JS, et al. Feasibility of PET template-based analysis on F-18 FP-CIT PET in patients with De novo Parkinson’s disease. Nucl Med Mol Imaging. 2013;47:73–80. doi: 10.1007/s13139-013-0196-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Ma Y, Dhawan V, Mentis M, Chaly T, Spetsieris PG, Eidelberg D. Parametric mapping of [18F] FPCIT binding in early stage Parkinson’s disease: a PET study. Synapse. 2002;45:125–133. doi: 10.1002/syn.10090. [DOI] [PubMed] [Google Scholar]
- 18.Hsiao IT, Huang CC, Hsieh CJ, Hsu WC, Wey SP, Yen TC, et al. Correlation of early-phase 18F-florbetapir (AV-45/Amyvid) PET images to FDG images: preliminary studies. Eur J Nucl Med Mol Imaging. 2012;39:613–620. doi: 10.1007/s00259-011-2051-2. [DOI] [PubMed] [Google Scholar]
- 19.Hsiao T, Huang CC, Hsieh CJ, Wey SP, Kung MP, Yen TC, et al. Perfusion-like template and standardized normalization-based brain image analysis using 18F-florbetapir (AV-45/Amyvid) PET. Eur J Nucl Med Mol Imaging. 2013;40:908–920. doi: 10.1007/s00259-013-2350-x. [DOI] [PubMed] [Google Scholar]
- 20.Rodriguez-Vieitez E, Carter SF, Chiotis K, Saint-Aubert L, Leuzy A, Scholl M, et al. Comparison of early-phase 11C-deuterium-l-deprenyl and 11C-Pittsburgh compound B PET for assessing brain perfusion in Alzheimer disease. J Nucl Med. 2016;57:1071–1077. doi: 10.2967/jnumed.115.168732. [DOI] [PubMed] [Google Scholar]
- 21.Daerr S, Brendel M, Zach C, Mille E, Schilling D, Zacherl MJ, et al. Evaluation of early-phase [18F]-florbetaben PET acquisition in clinical routine cases. Neuroimage Clin. 2017;14:77–86. doi: 10.1016/j.nicl.2016.10.005. [DOI] [PMC free article] [PubMed] [Google Scholar]