Superficial white matter (SWM) microstructural alterations in Parkinson’s disease: a neurite orientation dispersion and density imaging and diffusion tensor imaging study

Introduction

Parkinson’s disease (PD) is a progressive neurodegenerative disorder with an increasing global prevalence, and its debilitating characteristics place a substantial burden on both patients and society (1). PD is a multifaceted clinicopathologic syndrome characterized by the progressive exacerbation of bradykinesia, myotonia, resting tremor, and postural balance impairments, which are intricately linked to the selective degeneration of nigrostriatal dopaminergic neurons and the development of Lewy bodies due to the pathological aggregation of α-synuclein (2). Beyond the pathological alterations described above, impairment of the cortico-basal ganglia-thalamo-cortical circuit, a central hub governing motor and non-motor functions, serves as a critical pathological underpinning of PD (3).

Biochemical studies, analyses of transplanted neurons from PD patients, and cellular and animal model research indicate that aberrant α-synuclein aggregation and its dissemination across the gut, brainstem, and higher brain regions may drive the onset and progression of PD by disrupting the integrity of this circuit (2). Additionally, Shang et al. observed abnormal co-variability in the striatal-cortical and thalamic-cortical circuits through diffusion kurtosis imaging, providing microscopic evidence of the dysfunction of these circuits (4).

Currently, neuroimaging has become a core research direction for elucidating the neural mechanisms of PD. In this field, abnormal changes in deep long-range white matter tracts—such as the corticospinal tract and nigrostriatal pathway—have garnered significant attention. Damage to these long-range tracts, which mediate global signal transmission between deep subcortical regions and the cortex, is closely associated with the motor dysfunction observed in PD (4-8). However, previous research has largely focused on revealing abnormalities in “macro-connectivity” in the cortex-basal ganglia-thalamus-cortex circuit (4), while overlooking the mechanisms of “micro-local integration” in cortical regions—a critical prerequisite for functional integrity. This mechanism primarily relies on short-range fiber connections mediated by superficial white matter (SWM), a key region that remains understudied to date.

Unlike the deep white matter (DWM) tracts that mediate long-range signaling, the SWM is a highly complex and heterogeneous U-shaped fiber composed of axonal bundles, subcortical neurons, glial cells, and blood vessels (9). As the essential pathway for all axonal projections between gray matter and white matter, SWM serves as the core structure mediating local information integration. Compared to DWM, SWM fibers are shorter, more diffusely oriented, exhibit higher synaptic terminal density, and possess thinner myelin sheaths. These unique histological features render it a potential early target for neurodegenerative diseases (9).

Although existing studies employing diffusion tensor imaging (DTI) remain limited, they have confirmed that microstructural abnormalities in SWM constitute the structural basis for PD-related sensorimotor deficits and cognitive decline (10). Beyond PD, microstructural alterations in SWM have also been observed in other neuropsychiatric disorders such as Alzheimer’s disease, schizophrenia, and bipolar disorder (11-13). However, the fractional anisotropy (FA) metric used in conventional DTI has inherent limitations in specificity: while sensitive to overall fiber integrity, it cannot distinguish between loss of neurite density, fiber disorganization, and expansion of the extracellular space—all critical microscopic pathological changes in PD. This lack of specificity severely constrains the precise elucidation of the microscopic pathological mechanisms underlying SWM dysfunction in PD. Therefore, to overcome the limitations of a single model, there is an urgent need to introduce diffusion imaging techniques with higher biological specificity. These should complement classical DTI to achieve multidimensional, synergistic analysis of SWM microstructural damage.

This study innovatively combined traditional DTI with advanced neurite orientation dispersion and density imaging (NODDI) technology to perform multidimensional microstructural analysis of SWM in PD patients. NODDI is a three-compartment biophysical model proposed by Zhang et al. (14), which separates diffusion signals across three tissue compartments (extracellular matrix, intracellular neuronal space, and cerebrospinal fluid) to enable the quantitative analysis of axon-specific morphological features. The neurite density index (NDI) directly reflects the quantity of intact axons (axons and dendrites), while the orientation dispersion index (ODI) characterizes the regularity of axonal spatial arrangement.

This study had two primary objectives: (I) to validate the consistency and differences between DTI and NODDI metrics through comparative analysis, thereby assessing specific microstructural damage patterns in the SWM of PD (distinguishing reduced axonal density from increased orientation dispersion within white matter regions); and (II) to explore the associations between these imaging metrics and clinical parameters (e.g., motor symptoms and cognitive function). By employing this combined approach, the study aimed to overcome the limitations of single-model analyses and provide more comprehensive neuroimaging evidence for elucidating the neuropathological mechanisms of PD. We present this article in accordance with the STROBE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-aw-2302/rc).

Methods

Subjects

This study enrolled 35 PD patients hospitalized in the Department of Neurology at Shaoxing People’s Hospital between September 2023 and January 2025. All the patients received standardized clinical evaluations conducted by two neurologists, each with over a decade of experience in diagnosing movement disorders, adhering to the criteria established by the United Kingdom Parkinson’s Disease Association Brain Bank (15). The patients underwent three assessments: the Chinese version of the Mini-Mental State Examination (MMSE) (16), the Unified Parkinson’s Disease Rating Scale—Part III (UPDRS-III), and the Hoehn and Yahr (H-Y) Scale (17). The MMSE was used to evaluate overall cognitive capabilities, including executive function, working memory, attention, and visuospatial abilities. The UPDRS-III and H-Y scale were used to evaluate motor function and disease progression in PD patients.

PD patients were excluded from the study if they met any of the following exclusion criteria: (I) presence of concurrent neurological disorders, including cerebrovascular accidents (strokes), frontotemporal degeneration, progressive supranuclear palsy, multiple system atrophy, and other neurodegenerative diseases, or intracranial space-occupying lesions; and/or (II) inadequate magnetic resonance imaging (MRI) scan data quality due to factors such as changes in body position during the scan. PD patients requiring pharmacotherapy were instructed to discontinue all medications approximately 12 hours prior to MRI scanning under the guidance of a clinical neurologist.

Additionally, 37 healthy individuals were recruited as healthy control (HC) subjects. All HCs completed the MMSE. The inclusion criteria were as follows: (I) no history of neurological or psychiatric disorders; and (II) an MMSE score of 24 or higher. One HC with an arachnoid cyst was excluded, resulting in a final sample size of 36 HCs. Detailed demographic characteristics and clinical data for all subjects are presented in Table 1.

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Shaoxing People’s Hospital (No. 2022-345-Y-01), and informed consent was obtained from all individual subjects.

MRI data acquisition

All MRI images were obtained using a Siemens Magnetom Verio 3.0T MRI scanner (Siemens Healthcare, Erlangen, Germany) at Shaoxing People’s Hospital. The scanning sequences comprised T1-weighted imaging and multi-shell diffusion-weighted imaging. T1-weighted images were obtained using a spoiled gradient echo sequence with the following parameters: repetition time (TR)/echo time (TE) =1,900/2.5 ms, inversion time =900 ms, flip angle =9°, voxel size =1.1 mm × 1.1 mm × 1 mm, matrix =256×256, and field of view (FOV) =280 mm × 280 mm. Multi-shell diffusion data were obtained for each subject using a spin-echo diffusion-weighted echo-planar imaging procedure with the following parameters: TR/TE =9,000/104 ms, flip angle =90°, FOV =256 mm × 256 mm, matrix =128×128, voxel size =2 mm × 2 mm × 4 mm (without slice gap), with 30 diffusion-sensitive directions at b-values of 1,000 and 2,000 s/mm², and one volume of non-diffusion-weighted (b0) images. Additionally, opposed-phase b0 images were acquired and used in conjunction with the “top up” algorithm in FMRIB Software Library (FSL), Oxford Centre for Functional MRI of the Brain, University of Oxford, Oxford, UK for the correction of distortions induced by oscillating gradients in echo-planar imaging.

Data analysis

Structural reconstruction, segmentation, and mask registration

The T1-weighted images were processed for cortical reconstruction and subcortical segmentation using FreeSurfer version 7.3.2 (Martinos Center, Boston, MA, USA), generating gray-white matter boundary masks, diencephalon masks, and white matter masks. All the T1-space masks first underwent 12-parameter affine transformation for linear registration using FMRIB’s Linear Image Registration Tool (FLIRT), part of FSL, Oxford Centre for Functional MRI of the Brain, University of Oxford, Oxford, UK, laying the foundation for spatial alignment for subsequent nonlinear registration. For the DWM mask from the Johns Hopkins University DTI atlas, non-linear registration to diffusion space was achieved via the fiber tract-based spatial statistics (TBSS) pipeline (18,19); specifically, as a core analysis module of the FSL software library, the TBSS registration pipeline invokes the FMRIB Non-linear Image Registration Tool in the same software package to generate a non-linear transformation field, and the DWM mask is accurately non-linearly registered to the diffusion space through the inverse mapping of this transformation field (i.e., the inverted transformation field), ensuring spatial consistency between the masks and diffusion data.

Multi-shell diffusion data analysis, probabilistic fiber tracking with SWM constraint

The multi-shell diffusion data underwent preprocessing with MRtrix3 (3.0.3), an open-source diffusion MRI analysis software package developed and maintained by an international team of neuroimaging experts, including format conversion, denoising, mask generation, Gibbs loop artifact reduction, motion and distortion correction, and bias field correction (20). The FSL dtifit function was used to fit the diffusion tensor model in each voxel, producing FA and mean diffusivity (MD). This research used Accelerated Microstructure Imaging via Convex Optimization (AMICO) to fit the NODDI model, generating NDI and ODI maps (21). AMICO exemplifies an effective approach that markedly enhances fitting speed relative to current methodologies while preserving accuracy and precision in the estimation of model parameters.

Probabilistic fiber tracking was conducted via bedpostX/probtrackX (FSL toolkit), accommodating up to two crossing fibers per voxel (22). In each hemisphere, fiber tracking began at the ipsilateral gray-white matter boundary as the seed region, whereas subcortical areas, DWM, and contralateral white matter were excluded as masks to guarantee that tracking was restricted to the “SWM” area.

The fiber tracking outcomes were standardized by the total count of successful fiber tracts per subject and filtered with a 0.01% threshold of the waytotal to accurately identify tracts that genuinely represent SWM (23), thus establishing a dependable foundation for creating group-level masks. Individual-level binarized SWM fiber tracts were subjected to non-linear transformation into standard space using the TBSS transformation field (24). Probabilistic SWM masks were constructed using a 50% threshold after group averaging.

TBSS-like analysis

The SWM data were further processed and analyzed using a TBSS-like skeletonization methodology. Initially, all FA pictures were non-linearly registered to standard space and averaged to produce a subject-average FA map. Subsequently, the SWM fiber tracts were subjected to skeletonization: (I) the mean FA map was multiplied by the thresholded SWM mask to produce the SWM average FA map; and (II) a “thinning” procedure (non-maximum suppression perpendicular to local fiber tract structures) was applied to the SWM average FA image to generate the skeletonized SWM average FA (24). This study diminished the probability of false positives in the TBSS skeleton by employing a probabilistic fiber tract-based SWM mask to confine the TBSS analysis. Prior TBSS research has suggested that lower skeletonized mean FA thresholds (>0.15) may capture finer structures that would otherwise disappear at higher thresholds (25). Therefore, by applying a threshold to skeletonized SWM mean FA (FA >0.15), non-white matter elements were removed from the skeleton.

To eradicate residual registration bias and enable intergroup comparisons, local signal maxima were detected along the vertical axis of the skeleton inside each subject’s FA image, limited by the search space and SWM mask. These maxima denote the core sites of SWM, which are remapped to the common skeleton. The tbss_non_FA script, a component of FSL, executed non-linear transformation and skeletonization on the NDI and ODI maps. SWM masks were generated for the frontal, parietal, occipital, and temporal lobes by enlarging the respective cortical regions in the Montreal Neurological Institute (MNI) probability map, a standard template widely used in neuroimaging studies, with a Gaussian kernel to include probable SWM regions. The average NODDI index was subsequently derived from the SWM skeleton for the area of interest analysis.

Statistical analysis

The statistical analysis was performed using the social science statistical software package (SPSS 27.0). The Shapiro-Wilk test was used to verify the normality of the data distribution, and the Levene test was used to assess the homogeneity of variances. Independent samples t-tests or Wilcoxon rank-sum tests were used to evaluate differences in age and MMSE scores between the two groups. The chi-square test was used to analyze gender distribution differences across groups. Correlations between NODDI and DTI metrics and clinical measures—including disease duration, H-Y stage, and UPDRS-III scores—were assessed in the PD patients using Pearson or Spearman correlation analysis as appropriate. Statistical significance was defined as P<0.05.

Intergroup comparisons for SWM microstructure analysis were performed using age, years of education, and gender as non-interest variables. Nonparametric permutation tests in FSL were used for the SWM voxel analysis on the skeletonized NODDI parameter maps, executed 5,000 times. Significant clusters were preliminarily corrected using threshold-free cluster enhancement (TFCE), with the corrected P value <0.05. Clusters with significant intergroup variations in NODDI and DTI metrics throughout the subcortical white matter were identified using the Harvard-Oxford atlas. Correlation analyses were further performed between imaging indicators showing significant differences and clinical scales.

Results

Analysis of general characteristics and scale scores between the PD and HC groups

A total of 35 patients with PD and 36 HCs were enrolled in this study, and their demographic and clinical characteristics are presented in Table 1. The intergroup comparison results showed that there was no statistically significant difference in gender distribution between the two groups (P=0.19). However, the PD group was significantly older than the HC group [median age 70 years, interquartile range (IQR), 66.5–73.5 years vs. 61 years, IQR, 57.25–64.75 years, P<0.001], The PD group exhibited significantly fewer years of education than the HC group (median: 6 years; IQR, 6–6 years vs. 9 years, IQR, 7.5–10.5 years, P=0.008) and significantly lower MMSE scores (median: 25 points; IQR 21–29 points vs. 28 points, IQR, 26.5–29.5 points, P<0.001).

Regional microstructural differences in the SWM of PD patients compared with HCs

The DTI analysis of the SWM revealed significantly reduced FA in the lateral occipital cortex, superior division (LOCs), and precuneous cortex (PcC) of the PD patients, accompanied by increased MD in the LOCs and angular gyrus (AG) (all P_TFCE<0.05).

Extending these DTI findings, the NODDI analysis of SWM showed significantly reduced NDI in the PD patients, not only in the DTI-identified abnormal regions (LOCs, PcC, and AG) but also in the cingulate gyrus, posterior division (CGp) (P_TFCE<0.05). See Table 2 and Figures 1-5 for details.

Figure 1 NDI reduction involves brain regions across the LOCs, PcC, AG and CGp in the SWM of PD patients. AG, angular gyrus; CGp, cingulate gyrus, posterior division; LOCs, lateral occipital cortex, superior division; NDI, neurite density index; PcC, precuneous cortex; PD, Parkinson’s disease; SWM, superficial white matter.

Figure 2 FA reduction involves brain regions across the LOCs and the PcC in the SWM of PD patients. FA, fractional anisotropy; LOCs, lateral occipital cortex, superior division; PcC, precuneous cortex; PD, Parkinson’s disease; SWM, superficial white matter.

Figure 3 MD elevation involves brain regions across the LOCs and the PcC in the SWM of PD patients. LOCs, lateral occipital cortex, superior division; MD, mean diffusivity; PcC, precuneous cortex; PD, Parkinson’s disease; SWM, superficial white matter.

Figure 4 Intergroup comparisons revealed that the PD group exhibited significantly lower NDI and FA values compared to the HC group (all P_TFCE<0.05). FA, fractional anisotropy; HC, healthy control; NDI, neurite density index; PD, Parkinson’s disease; TFCE, threshold-free cluster enhancement.

Figure 5 Intergroup comparisons revealed that the MD value in the PD group was significantly higher than that in the control group. HC, healthy control; MD, mean diffusivity; PD, Parkinson’s disease.

Correlations between SWM microstructural indices and clinical features

To investigate the relationship between SWM microstructural alterations and clinical phenotypes, Spearman’s partial correlation analyses were performed—controlling for confounding factors (age, gender, and years of education)—with bootstrap bias-corrected and accelerated (BCa) confidence intervals (CIs) used to assess the association between clinical and imaging indicators.

The DTI results demonstrated negative correlations between FA in the dorsal visual pathway-default mode network (DMN) and H-Y stage (ρ =−0.405; BCa CI: −0.644, −0.083; P=0.022), and UPDRS-III scores (ρ =−0.366; BCa CI: −0.629, −0.046; P=0.039).

The NODDI analysis revealed that NDI in the parieto-occipital junction and posterior cingulate cortex was negatively correlated with disease duration (ρ =−0.425; BCa CI: −0.648, −0.216; P=0.015), UPDRS-III scores (ρ =−0.375; BCa CI: −0.647, −0.043; P=0.034), total UPDRS scores (ρ =−0.38; BCa CI: −0.626, −0.091; P=0.032), and H-Y stage (ρ =−0.357; BCa CI: −0.577, −0.142; P=0.045).

All the P values above are unadjusted prior to false discovery rate correction for multiple comparisons; none of the corresponding 95% BCa CIs included zero. See Figure 6 for details.

Figure 6 Correlation heatmap between clinical and imaging indicators. Cluster 1 FA was negatively correlated with H-Y stage and UPDRS-III scores. The NDI was negatively correlated with the disease duration, UPDRS-III scores, total UPDRS scores, and H-Y stage. FA, fractional anisotropy; H-Y, Hoehn and Yahr; MD, mean diffusivity; MMSE, Mini-Mental State Examination; NDI, neurite density index; UPDRS-III, Unified Parkinson’s Disease Rating Scale—Part III.

Discussion

This study, combining DTI with NODDI techniques, revealed that SWM lesions in the parieto-occipital junction of PD patients exhibit a distinct spatial hierarchy and pathological specificity.

The NODDI analysis revealed significantly reduced NDI in the LOCs, PcC, AG, and CGp of the PD patients, indicating decreased axonal density in these regions. These areas not only include DTI-abnormal zones but also extend to the CGp. This demonstrates the high sensitivity and specificity of the NDI metrics.

Notably, no significant changes in ODI were observed in this study. The ODI reflects the dispersion of nerve fiber orientation, and its stability has significant pathophysiological implications. Normal ODI indicates that damage at this stage primarily manifests as quantitative loss of nerve terminals rather than abnormal fiber bundle alignment (14). This aligns with synapse loss and axonal degeneration caused by abnormal α-synuclein aggregation in PD (26). Conversely, the CGp in the SWM exhibited a marked reduction in NDI only, with no alterations in FA, MD, and ODI. This NDI-specific decline suggests that PD pathology may initially disrupt subcortical connections of the CGp through localized neurofibrillary atrophy, preceding the disintegration of macroscopic fiber tracts. Thus, the region may represent an early target for PD pathology affecting subcortical microarchitecture, with the spatial distribution highly consistent with the α-synuclein propagation pathways described in Braak staging theory (27).

Conversely, the combined abnormalities of NDI/FA/MD observed in the LOCs indicate that this region has progressed to an advanced stage of structural degeneration, highly consistent with the pathological features of Braak stage V (27). Further, as a core node of the ventral visual (“what”) pathway, the LOCs primarily governs shape representation and object recognition in visual stimuli (28). Multidimensional microstructural damage in this region may compromise the integrity of the “visual processing area-basal ganglia-motor cortex” pathway. This finding offers a microstructural imaging perspective for understanding the structural disruption of visual-motor pathways in advanced PD.

The PcC, as a key part of the DMN, mainly works through cognitive regulation, but it also plays important roles in memory processing and visual-spatial integration (29). The concurrent reduction in NDI and FA in this region suggests composite pathological alterations in SWM, potentially involving decreased axonal density and disrupted myelin integrity, leading to impaired local information transmission in the DMN. This reveals the pathological progression of PD cognitive networks from isolated synaptic damage to widespread structural disintegration of fiber tracts.

Notably, this structural target has been cross-validated by multiple methodologies: Resting-state functional MRI studies identified selective functional deficits in the ventral DMN among PD patients with cognitive impairment (30). Fluorodeoxyglucose-positron emission tomography also revealed a significant positive correlation between PcC metabolism and cognitive scores (31). The authoritative pathological assessment guidelines in Lancet Neurology confirmed high-frequency Lewy body deposition in the PcC of PD patients with dementia, with the pathological burden significantly correlated with cognitive impairment severity (32). This multidimensional evidence collectively confirms that the PcC is a key core region for cognitive impairment in PD, and its dual structural and functional deterioration represents a critical driver of pathological progression in the cognitive network.

Notably, this study found no significant correlation in the SWM between the imaging metrics and MMSE scores. As a global cognitive screening tool, the MMSE has limited sensitivity for detecting early-stage cognitive impairment in PD (33). Further, neurodegenerative changes in PD exhibit insidious progression, with microstructural brain damage typically occurring covertly and preceding clinical symptoms.

In summary, this study confirmed that subcortical lesions in PD are not isolated localized alterations but rather a systemic process extending from the core regions of the DMN (PcC, manifested as reduced NDI and decreased FA) to the visual association cortex (extralateral occipital lobe, manifested as diffuse DTI/NDI abnormalities with a normal ODI). This comprehensive assessment method provides a reliable in vivo research tool for analyzing and quantifying pathological alterations in PD, including their spatial hierarchy and pathological specificity.

This study had a number of limitations, First, due to its cross-sectional design, a definitive dynamic causal relationship between SWM lesions and PD disease progression cannot be established. Second, the correlation analysis results were not adjusted for the false discovery rate, indicating that the robustness of these conclusions necessitates validation in larger, independent cohorts. Third, the study did not fully control for potential confounding factors such as vascular risk factors (e.g., hypertension, diabetes, and microvascular disease burden) (34,35). Finally, a potential limitation of this study is that the NODDI model fixes the intrinsic diffusion coefficient. While PD pathology may alter brain tissue diffusion rates, existing studies using multi-subject and parametric simulations have demonstrated that the NDI and ODI exhibit good robustness to moderate variations in diffusion coefficients (36). Therefore, it is hypothesized that such moderate changes in diffusion rates have a negligible impact on core findings, and future studies with higher b-values may provide further validation.

Future research should focus on conducting multicenter longitudinal follow-up studies, expanding sample sizes, and performing detailed stratified analyses (e.g., based on cognitive subtypes and disease stages) to clarify the dynamic evolution patterns of SWM microstructural metrics and their predictive value for PD clinical progression. Further, integrating molecular imaging techniques (e.g., positron emission tomography) or pathological biopsies to explore the correspondence between SWM damage and pathological alterations such as α-synuclein deposition and neuroinflammation will aid in elucidating its underlying pathogenic mechanisms.

Conclusions

This study systematically revealed multidimensional microlesion characteristics in the occipito-parietal junction region of PD patients, highlighting the sensitivity and specificity advantages of NODDI technology in capturing such microscopic early alterations. It confirmed the potential association between these lesions and disease severity (duration and motor function). These findings enrich the evidence base for SWM damage in PD, offering a novel SWM-level perspective for understanding the pathological mechanisms underlying the “visual-motor-cognitive” synergistic impairment in PD.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-aw-2302/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-aw-2302/dss

Funding: This research was partially supported by the Zhejiang Provincial Medical and Health Science and Technology Program (Nos. 2023KY1251 and 2023KY1239), the Shaoxing Municipal Medical and Health Science and Technology Program (No. 2022SY06), the Shaoxing Municipal Health Science and Technology Program (No. 2023SKY044), and the Shaoxing Municipal Basic Public Welfare Program (No. 2023A14024).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-aw-2302/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Shaoxing People’s Hospital (No. 2022-345-Y-01) and informed consent was obtained from all individual participants.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Zhu J, Cui Y, Zhang J, Yan R, Su D, Zhao D, Wang A, Feng T. Temporal trends in the prevalence of Parkinson's disease from 1980 to 2023: a systematic review and meta-analysis. Lancet Healthy Longev 2024;5:e464-79. [Crossref] [PubMed]
2. Morris HR, Spillantini MG, Sue CM, Williams-Gray CH. The pathogenesis of Parkinson's disease. Lancet 2024;403:293-304. [Crossref] [PubMed]
3. Baev KV, Greene KA, Marciano FF, Samanta JE, Shetter AG, Smith KA, Stacy MA, Spetzler RF. Physiology and pathophysiology of cortico-basal ganglia-thalamocortical loops: theoretical and practical aspects. Prog Neuropsychopharmacol Biol Psychiatry 2002;26:771-804. [Crossref] [PubMed]
4. Shang S, Wang L, Yao J, Lv X, Xu Y, Dou W, Zhang H, Ye J, Chen YC. Characterizing microstructural patterns within the cortico-striato-thalamo-cortical circuit in Parkinson's disease. Prog Neuropsychopharmacol Biol Psychiatry 2024;135:111116. [Crossref] [PubMed]
5. Shih YC, Tseng WI, Montaser-Kouhsari L. Recent advances in using diffusion tensor imaging to study white matter alterations in Parkinson's disease: A mini review. Front Aging Neurosci 2022;14:1018017. [Crossref] [PubMed]
6. Zhang Y, Wu IW, Buckley S, Coffey CS, Foster E, Mendick S, Seibyl J, Schuff N. Diffusion tensor imaging of the nigrostriatal fibers in Parkinson's disease. Mov Disord 2015;30:1229-36. [Crossref] [PubMed]
7. Luo C, Song W, Chen Q, Yang J, Gong Q, Shang HF. White matter microstructure damage in tremor-dominant Parkinson's disease patients. Neuroradiology 2017;59:691-8. [Crossref] [PubMed]
8. Zhu N, He S, Huang Z, Zhao J, Zhang C, Chu J, Ling Y, Zhao C, Wang J, Huang Y, Yang Z, Jiang N, Chen L, Chu J. Cerebral microstructural alteration patterns in motor subtypes of Parkinson's disease: a neurite orientation dispersion and density imaging study. Quant Imaging Med Surg 2025;15:12454-66. [Crossref] [PubMed]
9. Van Dyken PC, Khan AR, Palaniyappan L. Imaging of the superficial white matter in health and disease. Imaging Neurosci (Camb) 2024.
10. Zhang Y, Huang B, Chen Q, Wang L, Zhang L, Nie K, Huang Q, Huang R. Altered microstructural properties of superficial white matter in patients with Parkinson's disease. Brain Imaging Behav 2022;16:476-91. [Crossref] [PubMed]
11. Phillips OR, Joshi SH, Piras F, Orfei MD, Iorio M, Narr KL, Shattuck DW, Caltagirone C, Spalletta G, Di Paola M. The superficial white matter in Alzheimer's disease. Hum Brain Mapp 2016;37:1321-34.
12. Ji E, Guevara P, Guevara M, Grigis A, Labra N, Sarrazin S, Hamdani N, Bellivier F, Delavest M, Leboyer M, Tamouza R, Poupon C, Mangin JF, Houenou J. Increased and Decreased Superficial White Matter Structural Connectivity in Schizophrenia and Bipolar Disorder. Schizophr Bull 2019;45:1367-78. [Crossref] [PubMed]
13. Zhang S, Wang Y, Deng F, Zhong S, Chen L, Luo X, Qiu S, Chen P, Chen G, Hu H, Lai S, Huang H, Jia Y, Huang L, Huang R. Disruption of superficial white matter in the emotion regulation network in bipolar disorder. Neuroimage Clin 2018;20:875-82. [Crossref] [PubMed]
14. Zhang H, Schneider T, Wheeler-Kingshott CA, Alexander DC. NODDI: practical in vivo neurite orientation dispersion and density imaging of the human brain. Neuroimage 2012;61:1000-16. [Crossref] [PubMed]
15. Hughes AJ, Daniel SE, Kilford L, Lees AJ. Accuracy of clinical diagnosis of idiopathic Parkinson's disease: a clinico-pathological study of 100 cases. J Neurol Neurosurg Psychiatry 1992;55:181-4. [Crossref] [PubMed]
16. Le Bihan D. Microcomputer simulation of nuclear magnetic resonance imaging contrasts. J Radiol 1985;66:303-8.
17. Hoehn MM, Yahr MD. Parkinsonism: onset, progression and mortality. Neurology 1967;17:427-42. [Crossref] [PubMed]
18. Jenkinson M, Smith S. A global optimisation method for robust affine registration of brain images. Med Image Anal 2001;5:143-56.
19. Hua K, Zhang J, Wakana S, Jiang H, Li X, Reich DS, Calabresi PA, Pekar JJ, van Zijl PC, Mori S. Tract probability maps in stereotaxic spaces: analyses of white matter anatomy and tract-specific quantification. Neuroimage 2008;39:336-47.
20. Tournier JD, Smith R, Raffelt D, Tabbara R, Dhollander T, Pietsch M, Christiaens D, Jeurissen B, Yeh CH, Connelly A. MRtrix3: A fast, flexible and open software framework for medical image processing and visualisation. Neuroimage 2019;202:116137.
21. Daducci A, Canales-Rodríguez EJ, Zhang H, Dyrby TB, Alexander DC, Thiran JP. Accelerated Microstructure Imaging via Convex Optimization (AMICO) from diffusion MRI data. Neuroimage 2015;105:32-44.
22. Grill-Spector K. The neural basis of object perception. Curr Opin Neurobiol 2003;13:159-66.
23. Nazeri A, Chakravarty MM, Felsky D, Lobaugh NJ, Rajji TK, Mulsant BH, Voineskos AN. Alterations of superficial white matter in schizophrenia and relationship to cognitive performance. Neuropsychopharmacology 2013;38:1954-62.
24. Smith SM, Jenkinson M, Johansen-Berg H, Rueckert D, Nichols TE, Mackay CE, Watkins KE, Ciccarelli O, Cader MZ, Matthews PM, Behrens TE. Tract-based spatial statistics: voxelwise analysis of multi-subject diffusion data. Neuroimage 2006;31:1487-505.
25. Bach M, Laun FB, Leemans A, Tax CM, Biessels GJ, Stieltjes B, Maier-Hein KH. Methodological considerations on tract-based spatial statistics (TBSS). Neuroimage 2014;100:358-69.
26. Calabresi P, Mechelli A, Natale G, Volpicelli-Daley L, Di Lazzaro G, Ghiglieri V. Alpha-synuclein in Parkinson's disease and other synucleinopathies: from overt neurodegeneration back to early synaptic dysfunction. Cell Death Dis 2023;14:176.
27. Braak H, Del Tredici K, Rüb U, de Vos RA, Jansen Steur EN, Braak E. Staging of brain pathology related to sporadic Parkinson's disease. Neurobiol Aging 2003;24:197-211.
28. Grill-Spector K, Kourtzi Z, Kanwisher N. The lateral occipital complex and its role in object recognition. Vision Res 2001;41:1409-22. [Crossref] [PubMed]
29. Cavanna AE, Trimble MR. The precuneus: a review of its functional anatomy and behavioural correlates. Brain 2006;129:564-83. [Crossref] [PubMed]
30. Schindlbeck KA, Vo A, Mattis PJ, Villringer K, Marzinzik F, Fiebach JB, Eidelberg D. Cognition-Related Functional Topographies in Parkinson's Disease: Localized Loss of the Ventral Default Mode Network. Cereb Cortex 2021;31:5139-50. [Crossref] [PubMed]
31. Zhihui S, Yinghua L, Hongguang Z, Yuyin D, Xiaoxiao D, Lulu G, Yi L, Kangli F, Ying Z. Correlation analysis between 18F-fluorodeoxyglucose positron emission tomography and cognitive function in first diagnosed Parkinson's disease patients. Front Neurol 2023;14:1195576. [Crossref] [PubMed]
32. Dickson DW, Braak H, Duda JE, Duyckaerts C, Gasser T, Halliday GM, Hardy J, Leverenz JB, Del Tredici K, Wszolek ZK, Litvan I. Neuropathological assessment of Parkinson's disease: refining the diagnostic criteria. Lancet Neurol 2009;8:1150-7. [Crossref] [PubMed]
33. Snyder A, Gruber-Baldini AL, Rainer von Coelln F, Savitt JM, Reich SG, Armstrong MJ, Shulman LM. Comparison of Mini-Mental State Examination and Montreal Cognitive Assessment Ratings Across Levels of Parkinson's Disease Severity. J Parkinsons Dis 2021;11:1995-2003.
34. Uchida Y, Onda K, Nishimaki K, Kucharska-Newton A, Windham BG, Wasserman BA, Oishi K. Longitudinal Changes in Brain Diffusion Characteristics Associated With Cognition and Vascular Risk Factors: The ARIC-NCS Study. Neurology 2025;105:e213867. [Crossref] [PubMed]
35. Uchida Y, Nishimaki K, Soldan A, Pettigrew C, Ho SG, Moghekar A, Wang MC, Miller MI, Albert M, Oishi K. Change points for dynamic biomarkers in the Alzheimer's disease pathological cascade: A 30-year cohort study. Alzheimers Dement 2025;21:e70544. [Crossref] [PubMed]
36. Guerrero JM, Adluru N, Bendlin BB, Goldsmith HH, Schaefer SM, Davidson RJ, Kecskemeti SR, Zhang H, Alexander AL. Optimizing the intrinsic parallel diffusivity in NODDI: An extensive empirical evaluation. PLoS One 2019;14:e0217118. [Crossref] [PubMed]