Fractures strongly influence fluid flow and shear behavior in rock masses. However, direct testing on natural rock fractures is limited by scale, repeatability, and difficulties in controlling fracture geometry, particularly under dislocation. This study examined the effects of fracture dislocation on physical aperture, roughness, and fluid flow in a single induced tensile fracture of Kuru granite. Furthermore, it combined high-resolution photogrammetry and 3D printing to evaluate the feasibility of using fracture replicas for fluid flow testing. A fracture measuring 6 cm × 6 cm was analyzed using high-resolution photogrammetry, and 3D models of well-matched and dislocated surfaces were used to quantify aperture and roughness. Results showed a nonlinear increase in physical aperture with dislocation up to the maximum tested dislocation of 5 mm, accompanied by a decrease in surface roughness. A custom fluid flow setup was developed to test both natural and printed samples under varying hydraulic pressure gradients. Flow tests demonstrated that dislocation enlarged hydraulic apertures and increased flow rates, following a nonlinear relationship with hydraulic pressure gradient. Forchheimer's analysis indicated that hydraulic aperture (e h ) and the non-Darcy coefficient (β) are direction-dependent and highly sensitive to the magnitude of dislocation, resulting in flow anisotropy, jointly controlled by aperture and roughness. Comparisons between natural and printed samples revealed consistent flow trends but overestimated hydraulic apertures compared to the natural rock. The findings highlight the suitability of 3D-printed replicas for controlled fracture flow studies, while underlining current challenges in replicating natural roughness due to limitations in 3D printing and photogrammetry.

To evaluate the correspondence between myelin water fraction (MWF) estimates derived from multi-echo spin echo (MESE) and multi-echo gradient echo (MGRE) imaging in fixed mouse brain tissue, using a panel of myelin basic protein (Mbp) enhancer-edited mouse lines exhibiting graded hypomyelination. Fifteen mouse brains from five genetically modified mouse lines were imaged using ex vivo 7 T MRI with high-resolution 3D MESE and MGRE protocols. MWF maps were computed from the MESE approach using regularized non-negative least squares (NNLS) decomposition, and from the MGRE approach using robust principal component analysis (rPCA). MWF values were then parcellated across major white matter tracts using an atlas-based pipeline and validated against biological markers, including Mbp gene expression and FluoroMyelin fluorescence intensity. Both MESE- and MGRE-derived MWF maps exhibited high sensitivity to myelin content and resolved mouse line-dependent differences across white matter tracts. Region-specific MWF estimates were highly correlated across contrasts (r = 0.96), with MGRE yielding consistently higher MWF values, particularly in smaller tracts. MESE derived MWF values showed subtle underestimation of myelin content in hypomyelinated white matter regions. Both MWF measures showed strong correlations with Mbp messenger ribonucleic acid (mRNA) (r = 0.72-0.97) and FluoroMyelin staining (r = 0.50-0.92), with stronger histological correlations for MGRE-derived values. Both MESE and MGRE sequences provide biologically meaningful estimates of myelin content in fixed tissue but exhibit contrast-specific sensitivities and biases. MGRE combined with rPCA offers time efficient imaging and robustness in fine white matter structures, supporting its utility for high-resolution preclinical myelin mapping.

To characterize the clinical, radiologic, and electrophysiological features of spontaneous autoplugging in superior semicircular canal dehiscence (SSCD) and compare them to untreated and surgically treated cases. This retrospective observational study was conducted in a tertiary referral center in Paris, France. Fifty-nine SSCD ears were categorized into 3 groups: untreated (n=36), autoplugging (n=5), and surgically plugged (n=18). Autoplugging was defined as the absence of a fluid signal in the dehiscent canal on high-resolution 3D T2-weighted MRI in patients with no history of ear surgery. All patients underwent clinical assessment, audiometry, video head impulse test (vHIT), cervical vestibular-evoked myogenic potentials (cVEMP), and high-resolution imaging. Group comparisons used nonparametric tests with FDR correction and power analysis. vHIT gain was significantly reduced in both the autoplugging (0.56±0.09) and surgical groups (0.57±0.13) compared with untreated SSCD (0.77±0.19, P =0.0003). The reduction was confined to the superior canal, whereas lateral and posterior canal gains remained normal. Autophony persisted in the autoplugging group (80%) but was significantly reduced in the surgical group (22.2%, P <0.0001). cVEMP thresholds were abnormally low in untreated (62.0±11.1dB) and autoplugging ears (62.5±9.6dB), and significantly higher post-surgery (83.5±12.3dB, P <0.001). Autoplugging mimics surgical plugging in terms of canal dysfunction but fails to alleviate third-window symptoms. It may represent an incomplete, spontaneous adaptation in SSCD pathophysiology. Level 3.

The photoinitiating system is a decisive factor governing the speed, resolution, and operational robustness of digital light processing (DLP) 3D printing. However, most existing systems rely on inert atmospheres, high irradiation intensities, or prolonged exposure times, severely limiting printing efficiency and practical applicability. Here we report a molecular design paradigm that fundamentally redefines the functional role of multiple-resonance thermally activated delayed fluorescence (MR-TADF) materials, transforming them from efficient light emitters into highly active triplet photocatalysts through selenium-atom engineering. By integrating carbonyl-assisted n-π*/π-π* state coupling with selenium-induced spin-orbit enhancement, the resulting photocatalyst QPSO achieves a near-unity intersystem crossing quantum yield together with an exceptionally large forward-to-reverse intersystem crossing rate constant ratio, thus efficiently channeling exciton flux into long-lived, redox-active triplet states. When combined with a hypervalent iodonium co-initiator, this system enables rapid photopolymerization in ambient air using low-intensity blue light. As a result, single-layer curing is completed within only 1.5-2s across a broad thickness range, affording a printing resolution down to 10µm and a record-high build speed of up to 72cm h-1 at 400µm layer thickness. The platform supports the fabrication of complex hierarchical architectures while exhibiting excellent biocompatibility.

Spatial transcriptomics and in situ hybridisation techniques have become essential for contextualising single-cell RNA sequencing (scRNA-seq) data by mapping cell type- and state-specific mRNA expression within intact tissues. Beyond validating scRNA-seq predictions, these spatial methods provide critical insight into the localisation of specific cell types and their spatial relationships with neighbouring cells. However, current approaches rely on light microscopy (LM), which, despite advances, lacks the resolution needed to resolve detailed cell morphology or subcellular RNA localisation in situ. While correlative imaging methods can enhance spatial context, RNA detection remains limited by the optical constraints of LM. To overcome these challenges, we developed a three-dimensional RNA-based correlated light and electron microscopy (RCLEM) workflow. This method integrates RNA labelling with serial block-face scanning electron microscopy (SBF-SEM), allowing high-resolution 3D visualisation of RNA molecules within their ultrastructural environment. We optimised the RNA labelling protocol for thick, whole-mount tissues by fine-tuning detergent conditions to preserve ultrastructure while maintaining probe accessibility. This RCLEM approach offers several advantages over protein-based correlative methods, including robust detection with nucleotide probes and independence from antibody optimisation or fixation-sensitive epitopes. Applied to choroid plexus (ChP) and liver tissue, this workflow enabled precise 3D correlation between mRNA expression and EM features, supporting detailed morphological characterisation of rare scRNA-seq-defined cell subtypes. This protocol is broadly applicable to other complex tissues and provides a powerful platform for integrative, high-resolution spatial transcriptomic analyses that bridge gene expression with ultrastructure. LAY DESCRIPTION: To overcome the resolution limits of current spatial transcriptomics, we developed a 3D RNA-based correlated light and electron microscopy (RCLEM) workflow. By integrating RNA labelling with scanning electron microscopy, this method enables high-resolution 3D visualisation of mRNA within its ultrastructural environment in thick tissues. Validated in choroid plexus and liver, our antibody-independent RCLEM approach provides a broadly applicable platform bridging gene expression and precise cellular morphology for advanced spatial analyses.

Pancreatic cystic lesions are increasingly detected due to the widespread use of high-resolution cross-sectional imaging, particularly MRI and CT. The reported prevalence of incidental pancreatic cysts is 2% to 13% on CT and 20% to 45% on MRI, reflecting improved imaging sensitivity and an aging population. While most cysts are benign, mucinous lesions such as intraductal papillary mucinous neoplasms (IPMNs) and mucinous cystic neoplasms (MCNs) carry a risk of malignant transformation. MRI with MR cholangiopancreatography (MRCP) is central to evaluating pancreatic cystic lesions, particularly for delineating pancreatic duct anatomy, showing cyst-duct communication, and assessing ductal morphology relevant to malignancy risk. High-resolution 3D MRCP enables detailed visualization of the main and branch ducts and supports differentiation of cyst types, including identification of branch-duct IPMN (BD-IPMN) based on communication with the main pancreatic duct. However, MRCP alone has limitations, such as inability to assess enhancement patterns, vascularity, or subtle mural nodules, which may require contrast-enhanced MRI or endoscopic ultrasound (EUS). Recent advances in MRI technology, including compressed sensing and deep-learning reconstruction, have improved MRCP image quality and enabled abbreviated MRI protocols for surveillance of low-risk cysts. These protocols reduce scan time and cost while keeping diagnostic accuracy but must be applied judiciously to avoid missing subtle high-risk features or concomitant malignancies. This review evaluates the role of MRCP in characterizing ductal communication and high-risk features in pancreatic cystic lesions, discusses technical considerations and diagnostic challenges, and examines the evidence supporting abbreviated MRI protocols. The integration of MRCP into multimodality imaging strategies is emphasized for best risk stratification and clinical management.

During breathing, anatomical structures within the torso move which can result in inaccurate radiation dose delivery. With the introduction of MRI-linear accelerators (MRI-linacs), the ability to simultaneously acquire high-contrast soft tissue images during irradiation allows for compensation of intrafraction motion and deformation by gating or tracking the target volume. The implementation of MRI-guided tracking radiotherapy is not without risks as any lag or organ deformation may not be accounted for. Polymer gel dosimetry has the potential to measure the integral dose delivered to deforming and moving targets as the gels are flexible and provide a high-resolution 3D dose profile. Spherical silicone casts filled with polymer gel and silica beads were compressed to measure the deformation of gel phantoms. The effect of compression on the dose response was studied by irradiating and scanning the spherical phantoms in various states of compression. End-to-end gel dosimetry experiments with MRI-guided tracking radiotherapy were conducted on a prototype MRI-linac (Australian MRI-Linac) using a moving, non-deformable, rectangular-shaped gel dosimeter phantom and an MRI-safe, pneumatically actuated, anthropomorphic, breathing phantom containing a liver-shaped gel dosimeter. Radiochromic film dosimeters within the phantoms were used as secondary validations of the dose profile. The tracking performance of the MRI-linac was assessed by comparing measured dose distributions in the phantoms in static and actuated experiments. There was no significant impact of compression on the dose response in the irradiated spheres. The gel-measured dose profiles in the phantoms matched closely with the film dosimeters for tracked and non-tracked scenarios. The end-to-end gel dosimeter experiments illustrate the improvement in dose conformality with MRI-guided tracking. Deformable 3D gel dosimeters in an anthropomorphic body phantom can be used for assessing 3D geometric accuracy of tracked treatments on MRI-linacs, but care should be taken to account for the oxygen inhibition at the edges of the dosimeter.

Objective.Cone beam computed tomography (CBCT) often has a truncated acquired field of view (FOV) due to the limited detector size, leading to image reconstruction from truncated projection data. CBCT reconstructions using an image volume that just encloses the acquired FOV exhibit reconstruction artifacts due to attenuation in the tissues outside the image volume. On the other hand, extending the high-resolution voxel volume far enough beyond the FOV to fully enclose the imaged body often leads to a significant increase of the computational complexity in model based iterative reconstruction techniques. We propose a multi-resolution reconstruction model that eliminates the out-of-FOV reconstruction artifacts and enables accurate recovery of Hounsfield unit (HU) values within the FOV.Approach.We propose a multi-resolution extended reconstruction volume (MR-ERV) approach that extends the image volume beyond the FOV using separate extension volumes with coarser voxel representation, leading to appropriate modeling of the observed rays outside the FOV without significant increase of the computational complexity. Furthermore, we demonstrate that by augmenting the model with a simple projection extrapolation yields a further reduction of the out-of-FOV artifacts. In this study, the model is evaluated with model based iterative reconstruction minimization using high-resolution 3D CBCT data. The optimization problems considered are non-negativity constrained least-squares estimation, with and without regularization. The optimization is performed using a primal-dual hybrid gradient algorithm.Results.The proposed MR-ERV model effectively removes out-of-FOV reconstruction artifacts and it also achieves accurate HU values within the FOV when the volume extension fully encloses the imaged body in the transaxial direction.Significance.The MR-ERV model provides a platform for computationally efficient and accurate model based iterative reconstruction of CBCT data.

In stimulated emission depletion (STED) nanoscopy, high 3D resolution requires harnessing a 4Pi architecture using two opposing objectives. Here, we provide the step-by-step process for the construction and alignment of a 4Pi-STED nanoscope, commonly referred to as an 'isoSTED nanoscope'. The procedure guides interested researchers through the assembly of the optomechanical components, the configuration of the electronic and control devices, the alignment of the optical beam path and the assessment of the instrument's performance. The protocol offers a detailed roadmap for constructing an isoSTED nanoscope with adaptive optics in about 12 months and is designed for users with expertise in optical instrumentation builds. Once the instrument is finely calibrated, researchers can expect to achieve 3D biological images with isotropic sub-50-nm resolution in thick samples ≤35 µm in depth.

Unsaturated fatty acids (UFAs) play key roles in essential cellular functions such as membrane dynamics, metabolism, and animal development. Disruptions in UFA metabolism are linked to metabolic, cardiovascular, and neurodegenerative disorders. Cellular UFAs composition and quantification are normally determined using methods such as gas chromatography and/or mass spectrometry, which require extraction procedures and prevent analysis of live specimens. Here, we describe a protocol that employs uniform 13C isotope labeling and high-resolution 2D solution-state nuclear magnetic resonance (NMR) spectroscopy to analyze lipid composition and fatty acid unsaturation directly in the model organism Caenorhabditis elegans. The approach enables in vivo assessment of lipid storage compositions with sufficient resolution and sensitivity to distinguish wild-type animals from those with altered fatty acid desaturation. Complementary analysis of total lipid extracts provides information regarding lipid molecules that are not detected in vivo, such as phospholipid molecules organized in biological membranes. Overall, this non-destructive NMR-based method offers a powerful tool for investigating lipid metabolism in C. elegans and other small model systems that can be isotopically enriched. Key features • Solution-state NMR spectroscopy is not destructive and can be used on live cells and multicellular organisms. • 13C isotopic enrichment is required for high-resolution NMR analysis of lipids in live C. elegans. • Lipid signals from live worms arise from the mobile lipid phase in lipid droplets. • NMR provides readouts of lipid compositions in live animals at a highly sensitive rate, enabling precise interpretation of the whole cell lipid metabolism.

Lumbar spine MRI is predominantly performed using 2D FSE sequences. 3D FSE sequences offer potential advantages over 2D, especially with recent advances in deep-learning reconstruction (DLR) and the use of conformal high-density coil arrays to improve the SNR. This study aimed to compare image quality and diagnostically relevant differences between 2D and 3D lumbar spine protocols, both using a prototype conformal coil and DLR. Subjects with lumbar curvature referred for routine lumbar spine MRI were prospectively recruited (n=31). A prototype, conformal spine coil array with 70 total posterior receiver elements was used at 3T. DLR was applied to both the 2D- and 3D-protocols to improve sharpness and reduce noise. For 3D, a prototype DLR ("Sonic DL") technique was applied to provide higher acceleration factors. Three attending musculoskeletal radiologists assessed 2D and 3D T1-weighted (T1), T2-weighted (T2), and STIR sequences for image quality and diagnostic performance. The Gwet agreement coefficient (AC1, AC2) or intraclass correlation (ICC) was used to assess interrater agreement, while the Wilcoxon signed rank test or a sign test with Bonferroni correction assessed differences between 2D and 3D sequences. There was an overall 14% scan time reduction of 3D compared with 2D sequences, and a 38% time reduction for T2 sequences. All 3D sequences exhibited similar image quality to 2D. 3D T2 sequences had less CSF flow artifact than 2D T2 sequences. Interrater agreement was mostly comparable between all 2D and 3D assessments, including excellent agreement for the spinal canal area (ICC = 0.952 [95% CI: 0.907-0.976] for 2D, 0.941 [0.866-0.973] for 3D), good agreement for foraminal stenosis assessments on T2 sequences (AC2 = 0.713 to 0.795 for 2D, 0.706-0.832 for 3D), and good agreement for fracture detection on T1 sequences (AC1 = 0.869 for 2D, 0.815 for 3D). DLR-enabled 3D lumbar spine MRI offers comparable image quality to 2D with reduced overall scan time.

Abstract. The Levantine Basin is an ultra-oligotrophic region and the formation site of Levantine Intermediate Water. A high-resolution 3D coupled hydrodynamic-biogeochemical model (SYMPHONIE-Eco3MS) was used to investigate the seasonal and interannual variability of dissolved oxygen (O2) in the Levantine Basin and to estimate its basin-wide budget over the period 2013–2020. The model results show a pronounced seasonal cycle of air–sea exchanges. During winter, cooling and vertical mixing induce an undersaturation in oxygen of the surface layer by up to 2 % across the entire basin, leading to atmospheric oxygen absorption. In contrast, during the stratified period, primary production and warming induce a slight oversaturation and subsequent oxygen release to the atmosphere. The annual budget over the 7-year period shows that the basin acts as a net sink for atmospheric oxygen. The oxygen budget analyses further indicate that the surface layer (0–150 m) acts as a source of dissolved oxygen for intermediate depths through winter vertical export, whose amplitude is significantly governed by the magnitude of heat fluxes. At the basin and annual scale, we estimate a net lateral oxygen input into the basin from the Ionian Sea and a net export towards the Aegean Sea, with this lateral export at both surface and intermediate layers enhanced when winter heat loss is intense. Biogeochemically, the Levantine Basin alternates between autotrophic and heterotrophic states on an annual basis, depending on the intensity of winter surface heat loss. Spatially, the Rhodes Gyre, a quasi-permanent cyclonic structure and major site of intermediate water formation, emerges as a significant oxygen pump in winter, with annual uptake rates twice as high as the rest of the Levantine Basin, and shows enhanced biological production during the productive season, contributing to 41 % of the net annual oxygen production in the surface layer in the basin. This study highlights the need for further modeling studies on pluri-annual and multi-decadal scales to explore interannual variability and evolution of the annual oxygen budget across the entire Eastern Basin, particularly in the context of climate change.

Learning mesh representation is important for many 3D tasks. Conventional convolution for regular data (i.e., images) cannot directly be applied to meshes since each vertex's neighbors are unordered. Previous methods use isotropic filters or predefined local coordinate systems or learning weighting matrices for each template vertex to overcome the irregularity. Learning weighting matrices to resample the vertex's neighbors into an implicit canonical order is the most effective way to capture the local structure of each vertex. However, learning weighting matrices for each vertex increases the model size linearly with the vertex number. Thus, large parameters are required for high-resolution 3D shapes, which is not favorable for many applications. In this paper, we learn spectral dictionary (i.e., bases) for the weighting matrices such that the model size is independent of the resolution of 3D shapes. The coefficients of the weighting matrix bases are learned from the spectral features of the template and its hierarchical levels in a weight-sharing manner. Furthermore, we introduce an adaptive sampling method that learns the hierarchical mapping matrices directly to improve the performance without increasing the model size at the inference stage. Comprehensive experiments demonstrate that our model produces state-of-the-art results with a much smaller model size.

Animal morphology reflects both evolutionary history and present-day adaptation. Male mammal copulatory structures such as the baculum (penile bone) are ideal for studying these processes because of their complexity and high interspecific variability. In primates, however, research has focused mostly on baculum length. Here we investigate the evolution of primate baculum anatomy and morphology using the largest dataset assembled to date. We combined high-resolution 3D micro-CT reconstructions with advanced non-landmark methods to quantify key traits, including baculum position, discrete shape categories, and a continuous descriptor of overall baculum complexity derived from alpha-shapes analysis. Using stochastic character mapping on a primate phylogeny, we inferred ancestral states and evolutionary transitions for baculum position, shape type, and complexity. Reconstructions indicate that a proximally positioned baculum extending beyond the penile mid-shaft represents the ancestral condition, retained in Strepsirrhini and shifted distally in Haplorrhini. A stick-shaped baculum is inferred as ancestral, with subsequent transitions to Y-shaped and pear-shaped morphologies. Overall, this study provides the first phylogenetically explicit reconstruction of primate baculum evolution, revealing repeated shifts in anatomy and morphology complexity. These results offer a framework for testing functional and selective hypotheses on copulatory structures and highlight the value of 3D morphometrics for understanding morphological diversity across species.

Geologically deep learning for high-resolution sparse 3D oil reservoir modeling

Three-dimensional reconstruction of ionospheric disturbance structures during the pre-storm phase of the 10–11 May 2024 geomagnetic storm using a compressed sensing-based tomographic technique

Debris flows represent significant landslide hazards and assessing their destructive potential requires understanding and predicting key parameters such as volume and peak discharge. However, a lack of detailed field data limits our understanding of the spatio-temporal evolution of these quantities. This study addresses these unknowns by utilizing a new suite of high-resolution measurements from 3D LiDAR scanners, which measure instantaneous velocity and flow depth fields. We analyze four debris flows at three monitoring stations along the Illgraben fan. Discharge and volume estimates derived from traditional methods are compared against a novel column-wise (CW) method, which calculates discharge by integrating area and velocity for discrete vertical columns, allowing the exclusion of deposited material (where v = 0 ). We classify the analyzed events into three types: a) fast front, b) depositional and c) wavy. For type a and b events, traditional methods overestimate event volume by >2× compared to the CW approach. This overestimation is attributed to assuming a constant or depth-dependent velocity that does not account for the decrease in velocity after the passage of the front or waves. The overestimation further results from the inclusion of deposited material (e.g. levees) in the cross-sectional area. For type c events, traditional methods underestimate peak discharge associated with wave arrivals, which reached up to 5× the front discharge. Our results align well with empirical relations between volume and peak discharge and are an important step towards obtaining more accurate values of these quantities, which are crucial for the design of check dams and retention basins. • Four debris flows were recorded at three monitoring stations along the Illgraben fan. • High-resolution 3D LiDARs enable a column-wise method to get discharge and volume. • Traditional methods overestimate volume >2× for fast front and depositional flows. • For events with surge waves, traditional approaches underestimate peak discharge. • The peak discharge of these waves can be up to 5× the front discharge. • Traditional methods fail to exclude deposited material and overestimate velocities.

Integrating 3D Point Cloud analysis for potentially unstable rock blocks characterization: a method for assessing size and shape distribution

This study aims to quantify and to correct for partial volume effect (PVE) in lymph node metastases of prostate cancer (PC) on Lu-177 SPECT images using 3D-printed phantoms of these structures to establish a ground truth. Ten individuals with clearly delineated, SPECT-positive lymph node metastases were retrospectively selected from a cohort of PC patients undergoing radioligand therapy (RLT). Manual segmentation of the metastases was performed on the CT component. The segmented lymph nodes were 3D-printed as patient-specific lymph node phantoms with volumes ranging from 0.18 to 23.7ml using a high-resolution 3D printer. These phantoms were filled with Lu-177 and analyzed in a SPECT/CT system to quantify PVE. An exponential curve fit of the recovery coefficient (RC) versus the surface-area to volume ratio (SA:V) was derived from the spheres of a NEMA ICE body phantom and used to correct for PVE in the lymph node phantoms. This method was compared with other post-reconstruction partial volume corrections (PVC). For a tumor to background ratio (TBR) of 10:1 the RCs varied widely, from 82% to 10.4% depending on phantom size. Using the NEMA IEC body phantom, an exponential correlation was established between SA:V and RC, with R2 values exceeding 0.97 across measurements at four different TBRs. Using this curve for PVE correction, the RCs of the lymph node phantoms had an average deviation from the ground truth of 1.15 ± 10.15% (average ± standard error of the mean). This method had a higher accuracy than the other PVCs studied. The low RCs obtained for lymph node metastases suggest that Lu-177-dosimetry is in these structures grossly inaccurate without PVE correction. Applying the SA:V/RC curve established using the NEMA IEC body phantom for this purpose reduces PVE-related errors considerably.

The cerebellum shapes distributed motor and association networks through precisely organized pathways linking the cerebellar cortex to the deep cerebellar nuclei (DCN), its principal output structures. Whether these cortical and nuclear compartments scale proportionally across primates, and how their relative expansion relates to cerebellar organization, remains unresolved. Here, we generate multimodal, high-resolution cross-species 3D cerebellar atlases in marmoset, macaque, and human by integrating iron-sensitive and diffusion MRI (MAP-MRI) with histological markers, enabling direct delineation of cerebellar lobules and DCN subdivisions within a unified framework. Comparative volumetric analyses reveal strikingly nonuniform scaling of cerebellar cortical-nuclear architecture: cortical expansion markedly outpaces DCN enlargement, indicating disproportionate growth of input relative to output systems. This divergence is accompanied by progressive reorganization of the DCN, with increasing dominance of the dentate nucleus. In contrast, cortical expansion is driven by posterior hemispheric territories, with lobules VI-IX and Crus I/II, linked to higher-order functions, showing the strongest scaling and largest absolute gains, whereas DCN subdivisions show selective rather than uniform scaling. Together, these findings establish nonuniform cortical-nuclear scaling as a systems-level organizational principle that reshapes cerebellar input-output architecture across primates.