This cross-sectional observational study was conducted between September and December 2024 at the Department of Radiology of a tertiary academic medical center affiliated with Shiraz University of Medical Sciences (SUMS), Shiraz, Iran. The investigation was part of an ongoing neuroimaging initiative to establish normative morphometric values for subcortical brain structures in adults. The study protocol was approved by the SUMS Medical Ethics Committee (Approval Code: IR.SUMS.MED.REC.1403.733). All procedures were performed in accordance with the principles of the Declaration of Helsinki. Written informed consent was obtained from all participants prior to enrollment, and all imaging data were anonymized before analysis to ensure confidentiality.

A total of 241 neurologically healthy adults, including 107 males and 134 females aged 30–70 years (mean ± SD: 45.2±11.3 years), were prospectively recruited. Participants were consecutively enrolled from individuals undergoing brain MRI for non-neurological clinical indications. Eligibility required age ≥18 years and the availability of high-quality axial and coronal T2-weighted MRI scans that were diagnostically adequate for precise delineation of the CN. Two radiologists independently reviewed all MRI scans to confirm technical adequacy prior to morphometric analysis.

Participants were excluded if they had a history of neurosurgical intervention, congenital brain malformations, intracranial vascular anomalies, neurodegenerative disorders such as Parkinson’s disease or Huntington’s disease, cerebral palsy, or intracranial infection. Individuals with systemic conditions known to affect neurovascular integrity, including diabetes mellitus or uncontrolled hypertension, were also excluded. Eligible participants were identified through electronic health records, and all identifying information was removed prior to data processing.

All MRI examinations were performed using a 1.5-Tesla Philips Gyroscan Intera scanner (Philips Medical Systems, Best, The Netherlands). Participants were positioned supine, and the head was secured in a dedicated head coil and aligned along the canthomeatal line to standardize imaging orientation. The imaging protocol included scout imaging consisting of 8–10 sagittal slices with a slice thickness of 5 mm and an interslice gap of 1 mm. Axial and coronal T2-weighted spin-echo sequences were subsequently acquired using a repetition time of 2800 ms, an echo time of 100 ms, a slice thickness of 2.5 mm, an interslice gap of 0.3 mm, a flip angle of 90°, a field of view of 220 mm, and a matrix size of 352×512. These parameters were selected to optimize gray–white matter contrast and ensure clear visualization of CN morphology. All images were inspected for motion or technical artifacts prior to analysis.

MRI data in DICOM format were transferred to a secure workstation and processed using ImageJ software (Version 1.34; National Institutes of Health, Bethesda, MD, USA). Axial images were reoriented along the anterior commissure–posterior commissure (AC–PC) line to standardize anatomical alignment. Pre-processing involved semi-automated skull stripping to remove non-neural tissue, followed by brightness and contrast adjustments to enhance visualization of CN boundaries.

For morphometric analysis, the CN head was manually delineated bilaterally on axial and coronal T2-weighted images, using standard neuroanatomical atlases as references. In each hemisphere, two linear parameters were measured: CN width (CNW), which is the maximum mediolateral dimension, and CN length (CNL), which is the maximum anteroposterior dimension (Fig. 1). To ensure procedural consistency, only one skilled investigator carried out each measurement. Manual segmentation was selected for its anatomical precision, though it is time-intensive. Future studies may benefit from semi-automated or atlas-based segmentation methods to enhance scalability and reduce inter-rater variability.

The investigator performing the morphometric measurements was blinded to participants’ age and sex. To evaluate measurement reproducibility, 20% of the sample was independently reanalyzed by a second investigator who was also blinded to participant characteristics and followed the same measurement protocol. This procedure minimized observer bias and ensured methodological consistency.

Statistical analyses were conducted using IBM SPSS Statistics (Version 27; IBM Corp., Armonk, NY, USA). The distribution of CN linear measurements was assessed using the Shapiro–Wilk test, which demonstrated significant deviation from normality for all variables (p<0.05). Accordingly, non-parametric statistical methods were applied. Differences between males and females were evaluated using the Mann–Whitney U test, while associations between CN measurements and age were assessed using Spearman’s rank correlation coefficient (ρ). Effect sizes for the Mann–Whitney U tests were calculated, with r values of 0.10, 0.30, and 0.50 interpreted as small, medium, and large effects, respectively. All statistical tests were two-tailed, and statistical significance was defined as α=0.05. Given the exploratory nature of the study, no adjustments for multiple comparisons were made.

The sample included all eligible participants recruited during the study period. Although a priori power analysis was not conducted, the sample size was considered sufficient to detect small-to-moderate effects using non-parametric statistical methods. A post hoc power analysis will be performed to guide sample size estimation for future confirmatory or longitudinal investigations.

Table 1 displays descriptive statistics for bilateral CN line dimensions, including CNL and CNW. Overall, males demonstrated slightly larger measurements than females in both hemispheres. The mean right CNL (RCNL) was 19.88±2.45 mm in males and 19.52±2.24 mm in females. The mean right CNW (RCNW) measured 9.71±1.33 mm in males compared with 9.45±1.15 mm in females. Similar patterns were observed for the left hemisphere, although the differences were less pronounced. The Shapiro–Wilk test revealed significant deviations from normality for all morphometric variables (p<0.05), thereby justifying the use of non-parametric statistical methods (Fig. 2 and 3.).

The Mann–Whitney U test identified a statistically significant sex difference only for RCNW (U=2756.0, p=0.045), with males exhibiting larger values than females. The corresponding effect size (r=0.15) indicates a small but measurable difference (Fig. 4). No significant sex differences were observed for RCNL, LCNL, or LCNW (all p>0.05), suggesting that sexual dimorphism in CN linear dimensions is limited and primarily confined to the right-sided width.

Spearman’s rank correlation analysis demonstrated a weak but statistically significant negative correlation between age and LCNL (ρ=−0.18, p=0.015), indicating a gradual reduction in left CN length with increasing age (Fig. 5). No significant associations were detected between age and the remaining morphometric parameters—RCNL, RCNW, and LCNW (all p>0.05). These findings suggest that age-related structural changes in the CN are subtle and asymmetric, predominantly affecting the longitudinal dimension of the left hemisphere.

This study establishes normative reference values for the linear dimensions of the CN in a neurologically healthy adult population and identifies subtle but statistically significant associations with sex and age. Male participants demonstrated greater RCNW, whereas increasing age was associated with a modest reduction in LCNL. These findings contribute to foundational knowledge of CN morphology and highlight the potential value of linear morphometry as an accessible method for characterizing subcortical anatomy.

The present findings reinforce the practicality of linear MRI measurements as a complementary alternative to volumetric morphometry. Linear metrics can be derived directly from routine clinical MRI sequences, require minimal post-processing, and are less susceptible to segmentation-related errors. These characteristics make linear morphometry particularly suitable for everyday radiological practice and for clinical settings lacking advanced neuroimaging infrastructure. As interest in individualized neuroimaging biomarkers continues to expand, straightforward CN linear measurements may support early identification of structural variation, facilitate longitudinal monitoring, and assist in neurosurgical planning.

The observed sex-related difference in RCNW is consistent with previous volumetric studies reporting larger caudate and putaminal volumes in males.^[1-3] Proposed mechanisms include sex-dependent neurodevelopmental trajectories, androgen-mediated growth influences, and variations in cerebrovascular architecture. Extending these observations to linear morphometric parameters suggests that mediolateral expansion of the CN may serve as a simple anatomical indicator of sexual dimorphism. Given the CN’s involvement in reinforcement learning, motivation, and cognitive flexibility, structural differences may contribute to subtle sex-related variations in behavioral performance.^[4] However, functional correlations were not evaluated in this study and warrant investigation using task-based functional MRI or neuropsychological assessment.

The inverse association between age and LCNL aligns with established evidence of subcortical atrophy accompanying normal aging, particularly within the CN, putamen, and thalamus.^[5,6] The lateralized nature of this finding, predominantly affecting the left hemisphere, suggests possible asymmetric susceptibility to age-related structural change. Potential mechanisms may include hemispheric differences in dopaminergic innervation, microvascular supply, or intrinsic neurobiological resilience. Diffusion-based imaging studies have demonstrated variability in basal ganglia degeneration across hemispheres and disease phenotypes^[15], and ultrastructural investigations have documented neuronal loss within the CN in neurodegenerative disorders.^[16] Although the observed effect size was modest, the reduction in LCNL may represent a structural marker deserving further longitudinal validation. Future studies are needed to determine whether such changes precede or accompany cognitive or motor decline.

The translational relevance of these findings lies primarily in their potential application to stereotactic neurosurgical procedures, including deep brain stimulation, capsulotomy, and tumor resection. Linear morphometry provides rapid, reproducible structural information that may refine anatomical targeting, particularly in individuals with age-related atrophy or anatomical variability. Normative CN benchmarks may also aid in identifying deviations from expected sex- and age-adjusted ranges, which could support early recognition of neurodegenerative or psychiatric conditions. Structural asymmetries of the CN have been reported in schizophrenia, dementia, vascular cognitive impairment, and depressive disorders^[17,18], underscoring their potential diagnostic relevance.

Combining CN linear measurements with advanced imaging modalities, such as diffusion tensor imaging, resting-state functional MRI, or positron emission tomography, may enhance diagnostic and predictive modeling. Linear metrics are straightforward, scalable, and cost-effective, making them well-suited for automated processing pipelines and large cohort analyses. Emerging artificial intelligence applications increasingly rely on structural features to predict brain aging, cognitive decline, and disease susceptibility.^[19] Accordingly, CN linear parameters may serve as informative inputs in AI-based analytical frameworks aimed at improving diagnostic precision.

This study provides one of the relatively few normative datasets for linear CN measurements in healthy adults. The sample size strengthens statistical reliability, and strict inclusion and exclusion criteria enhance internal validity. All MRI scans were obtained using a single 1.5-Tesla system and standardized protocol, thereby minimizing inter-scanner variability. Although manual segmentation is labor-intensive, it ensures high anatomical accuracy. Blinded assessment and reproducibility analysis further support methodological rigor.