Biology subjects

Thomopoulos, S. I.

Publications and source records attributed to Thomopoulos, S. I..

At least 19 recordsLinked to original sources

bioRxiv2026

Understanding Performance Plateaus and Representation Collapse in 3D MRI Transformers Trained to Classify Brain Disease

In medical imaging AI, we commonly assume that diagnostic classifiers will increase in accuracy if they are trained on more data, but in reality, their performance often plateaus. To understand this behavior, we ran a spectral analysis of 3D vision transformers, a common supervised deep learning model that can be trained to classify disease based on 3D brain MRI. For simpler classification methods such as linear discriminant analysis, the classifier performance AUC(N), when trained on a sample of size N, is known to depend on the encoder's eigenspectrum. Each class-discriminative direction in the latent space contributes only after the sample size N reaches a threshold, so the spectrum influences how performance scales with training data. Building on this theory, we show that supervised transformers compress almost all of their discriminative signal into one principal component (EVR@1 = 0.90--0.96): after this representation collapse happens, no further directions cross the threshold as training data increases, and accuracy plateaus. In principle, this plateau in performance could be due to either the encoder's learned feature space or the difficulty of the task. To distinguish these cases, we trained three types of supervised 3D transformers (ViT3D, MINiT, NIT) and, as a non-collapsing alternative, a frozen self-supervised encoder (DINO), to classify people with dementia or mild cognitive impairment (MCI) versus cognitively normal (CN) individuals from 3D brain MRI, using data from the Alzheimer's Disease Neuroimaging Initiative (ADNI). We froze each encoder and applied it to a new task, so that only the task difficulty varied. An easy task encoder (Dementia vs CN), transferred to the harder task (MCI vs CN), did not reproduce the plateau (MINiT threshold 9.2), but full retraining did (2758.5). The plateau is therefore due to the learned encoder, rather than task difficulty. Ranking the feature directions by the proportion of the disease signal they contain, we were able to identify a sign of representation collapse, i.e., a single dominant spike in the latent feature eigenspectrum. This spike was 11--26 times larger than predicted from the rest of the spectrum for the disease-trained (supervised) encoders, but was absent for DINO (0.66). To predict how transformer accuracy scales with the number of training scans, a large spike in the encoder spectrum suggests that the representation has collapsed and additional scans will not help; a small spike means that the signal is spread across many encoder latent space directions, and more training data should boost the classifier performance.

neuroscience↗

bioRxiv2026

Machine learning on subcortical brain features: A study of sample size efficiency for neurodegenerative disease classification

Subcortical brain alterations are a key feature of dementia disease progression. Machine learning (ML) has been applied widely to MRI-based brain features in dementia, where performance depends on the model choice, training data size, and input feature characteristics. Most studies compare ML models using a single training sample size. Here, we evaluate the sample-size efficiency of ML models based on subcortical gross volume and vertex-wise shape features for dementia stage classification using 2,511 samples in the Alzheimer's Disease Neuroimaging Initiative (ADNI). Learning curves were generated for dementia vs. cognitively normal controls (CN), dementia vs. mild cognitive impairment (MCI), and MCI vs. CN across increasing training sample sizes. Classification performance improved when increasing sample size for all models, with late- fusion models consistently achieving the highest performance, and Logit-TVL1 outperforming the other shape-based models. Learning curve analysis showed that classification performance was driven by the training sample size and the magnitude of anatomical group differences, and can be used to optimize future model selection tasks in dementia and other brain disorders.

neuroscience↗

bioRxiv2026

When does more data help? Spectral Geometry and Scaling Laws in MRI Transformers

Scaling laws describe how model performance improves as the amount of training data increases, and recent theories such as the zeta law suggest that scaling behavior is influenced by the eigenspectrum of the models latent representation. Here, we evaluated whether the distribution of discriminative signals across spectral modes predicts the future scaling behavior, for MRI transformers trained for disease classification. We trained three supervised 3D vision transformers (ViT3D, MINiT, and NIT) for Alzheimers disease classification using 2,822 training scans from the Alzheimers Disease Neuroimaging Initiative (ADNI); we compared their encoder spectra with that of a frozen self-supervised DINO ViT-B/16 encoder adapted to 3D MRI. The supervised models learned highly concentrated representations, with 90-96% of CLS-token variance captured by a single principal component, whereas DINO distributed signal across many latent directions. Via spectral expansion of the Mahalanobis signal, we found that supervised training concentrated disease information into a single dominant mode, while self-supervised training produced a richer spectral geometry with higher effective rank and discoverability. This led to different scaling behavior: supervised models exhibited flatter AUC(N) curves, yet DINO continued to improve as sample size increased, gaining 11.0 percentage points from N=50 to N=2,822. Overall, the spectral distribution of the discriminative signal, for these different encoder types, influenced how much performance remained discoverable as sample size increased. Distributed representations may retain signal across many latent modes and continue to improve with additional data, whereas concentrated representations tend to exhaust most of the discoverable signal at much lower sample sizes.

neuroscience↗

bioRxiv2026

Comparing Harmonization Approaches for Protocol-Related Variability in Multisite Diffusion MRI Data

Diffusion MRI (dMRI) enables assessment of white matter microstructural abnormalities in Alzheimers disease (AD), and multisite datasets enable more robust modeling of non-biological variation that can confound analyses. The Alzheimers Disease Neuroimaging Initiative (ADNI) includes over 10 dMRI protocols, necessitating robust methods to model protocol-related variability when pooling data. Here, we compared three harmonization approaches: (1) mixed-effects models, (2) ComBat-GAM, and (3) eHarmonize, a reference-based lifespan method. We assessed their ability to reduce protocol-related variability in diffusion tensor imaging fractional anisotropy (FA) and mean diffusivity (MD) while preserving associations with cognitive impairment (CI), and amyloid-beta (A{beta}) and tau PET burden in 1,086 ADNI3/4 participants. All approaches yielded more closely aligned FA/MD distributions across protocols. Associations with clinical indicators of CI were highly consistent across approaches, whereas PET associations were less widespread and more variable. Overall, multiple strategies effectively modeled protocol-related variability while preserving AD-related associations.

neuroscience↗

bioRxiv2026

The ENIGMA-PD-WML Pipeline: A Containerized, User-Friendly Approach for Accurate, Standardized Segmentation of White Matter Lesions in Multi-Site MRI Data

Understanding vascular contributions to disease is a major unmet need. White matter lesions (WML) are an accepted imaging marker of cerebral small vessel disease, giving insights into its related pathologies. A unified approach for WML analyses in large multi-site data is lacking despite the need for pooling of data to overcome the limitations of often small heterogenous MRI studies which make subtyping and identifying patterns within disease groups difficult. Our ENIGMA-PD-WML pipeline is an open-source containerized pipeline containing all the code and packages required for pre-processing, processing and post-processing of T1-weighted and FLAIR data, outputting accurate and reproducible binary WML maps using a UNet approach. The pipeline provides a standardized image analysis approach for WML and outputs data in both native and MNI space to allow for sharing and pooling of data from multiple sites for large-data analysis. In addition to a reliable standardized approach for WML segmentation, key priorities when developing the pipeline included: usability, i.e., requiring minimal manual input and technical expertise to use, and suitability to run on various MRI scanners and acquisition parameters as is common in multi-site data. This paper describes the pipeline in detail, with rationale for each step, providing transparency and facilitating its usage to overcome reproducibility issues in large-scale WML analyses.

neuroscience↗

bioRxiv2026

Comparison of Automated White Matter Lesion Segmentation Approaches for Use in Large, Multi-Site Data Analyses in Parkinson's Disease

BackgroundParkinsons disease (PD) is the second most common neurodegenerative disorder. PD currently lacks effective disease-modifying treatments, likely due to its diverse clinical features and underlying neuropathology. The vascular role in PD is emerging, with vascular mechanisms increasingly implicated, yet the literature remains conflicted, motivating large-data analyses with greater statistical power. White matter lesions (WML) are an accepted imaging marker of small vessel disease. Accurate automated WML segmentation techniques are crucial for large-scale studies in PD due to the impracticality of manual segmentation for extensive datasets and to ensure consistency. Evaluation of the optimum approach in PD for large-scale analysis is lacking. This study aimed to evaluate various automated WML segmentation algorithms to determine the most accurate and reliable method, among those selected, for assessing WML for multi-site large data analysis in PD. MethodsWe assessed whole-brain volumetric T1-weighted and FLAIR images from 201 PD patients (mean age, 66.6 {+/-} 7.86 years) and 64 healthy controls (HC; mean age, 66.3 {+/-} 8.67) across three datasets: the Parkinsons Progression Markers Initiative (PPMI), the University of Pennsylvania (UPenn) and the Montreal Neurological Institute Biobank: Clinical Biological Imaging and Genetic Repository (C-BIG). The sample included different scanners, imaging parameters and lesion loads, as would be expected for multi-site data. WML were manually segmented to provide the gold standard, and four freely available automated algorithms were evaluated: FSLs BIANCA, FreeSurfer, SPMs LST-LPA and U-Net-pgs using the performance metrics: Dice score, Hausdorff distance, recall, precision, F1 score, log absolute volume difference (LOGAVD) and intraclass correlation coefficient (ICC). Subgroup analyses were performed based on lesion load and lobar regions. The associations of data from these automated approaches with age, and with Fazekas and Wahlund visual rating scales, were assessed through partial correlation analysis. ResultsU-Net-pgs performed best overall, with the highest Dice score (PD: 0.46 {+/-} 0.21; HC: 0.39 {+/-} 0.21), recall (PD: 0.76 {+/-} 0.25; HC: 0.62 {+/-} 0.31), precision (PD: 0.49 {+/-} 0.25; HC: 0.63 {+/-} 0.27), F1 score (PD: 0.54 {+/-} 0.22; HC: 0.56 {+/-} 0.22) and ICC (PD: 0.965; HC: 0.967) and lowest Hausdorff distance (PD: 8.89 {+/-} 3.96; HC: 6.33 {+/-} 2.91). U-Net-pgs achieved the lowest LOGAVD in the PD group (0.31 {+/-} 0.31) whereas BIANCA-LOO with a threshold of 0.9 was lowest in HC (0.27 {+/-} 0.30). U-Net also showed superior performances in all lesion loads for PD and overall across various brain regions in both PD and HC. ConclusionOverall, U-Net-pgs emerged as the best performing automated method, of those we evaluated, for WML segmentation in PD and HC within a dataset collected with various scanner and image acquisition parameters. U-Net-pgs consistently outperformed other automated approaches across lesion loads and brain regions, for most metrics. The accuracy and reliability of U-Net-pgs make it a promising tool for large-scale analyses, facilitating future research investigating WML in PD.

neuroscience↗

bioRxiv2026

Predicting Autopsy-Confirmed Neuropathology across Clinical, Neuroimaging, and CSF Biomarkers using Machine Learning

Accurate in vivo prediction of neuropathology is critical for advancing diagnosis and treatment of Alzheimers disease and related dementias (ADRDs). As many individuals with ADRDs have mixed pathologies ({beta}-amyloid, pathologic tau, cerebrovascular disease, vascular brain injury, pathologic TDP-43, hippocampal sclerosis, Lewy bodies), there is interest in determining how accurately we can infer these pathologic changes from clinical data, biofluid assays (e.g., CSF), and neuroimaging. Here we evaluated automated machine learning models trained on data curated by the AD Sequencing Project Phenotype Harmonization Consortium (N=7,894 individuals), to predict 26 autopsy-confirmed neuropathological outcomes. Predictors included in vivo clinical and cognitive composite scores, brain measures from 3D structural MRI and diffusion tensor imaging, image-derived measures of white matter hyperintensities (WMH), and CSF biomarkers. Predictive models were trained using ensemble learning with stratified cross-validation. We assessed performance using Spearmans rank correlation and Matthews correlation coefficient, to accommodate co-occurring pathologic changes. The added value of neuroimaging and CSF versus clinical features alone was quantified. Braak stage was among the most consistently predicted outcomes. CSF biomarkers best predicted {beta}-amyloid and tau pathology, but diffusion MRI metrics best captured vascular brain injury and white matter injury, and outperformed clinical and cognitive measures and anatomical MRI in predicting Lewy body disease. Anatomical measures from structural MRI outperformed standard clinical assessments in assessing neurodegeneration and hippocampal sclerosis, and WMH complemented cognitive measures in predicting TDP-43 pathology. These results establish a baseline for comparing modalities for inferring neuropathology.

neuroscience↗

bioRxiv2026

CALM-VLM: CALIBRATION AND SELECTIVE PREDICTION IN VISION-LANGUAGE MODELS FOR RELIABLE BRAIN MRI CLASSIFICATION

AO_SCPLOWBSTRACTC_SCPLOWRecent advances in vision-language models (VLMs) have demonstrated strong multimodal capabilities for medical image analysis. However, their confidence in diagnostic predictions is often unclear, limiting adoption in clinical settings. We introduce CALM-VLM (CALibration Mechanism for Vision-Language Models), which integrates confidence calibration and selective prediction into a generative 3D VLM. To create CALM-VLM, we fine-tuned the Med3DVLM architecture for Alzheimers disease (AD) and stroke classification as initial test cases. To improve reliability, we incorporated temperature scaling on the VLMs generative outputs. The calibrated model then selectively abstained from predictions when uncertain; this also improved its diagnostic accuracy. Experiments across multi-site MRI datasets, from 10 countries worldwide, show that CALM-VLM improved confidence relative to uncalibrated VLMs. Coverage-adjusted test receiver-operator characteristic curve-area under the curve (ROC-AUC) increased by 5% to 13% for both diagnostic tasks across independent test sets. Our calibrated VLM achieved a test ROC-AUC of 0.951 for AD classification and 0.905 for stroke classification. These findings highlight the importance of calibrated, uncertainty-aware VLMs for trustworthy neuroimaging AI.

neuroscience↗

bioRxiv2026

Distributional and Centile Calibration of Diffusion Tensor Imaging Normative Models as Training Sample Sizes Increase to 40,000 Subjects

Normative modeling (NM) is a powerful framework for quantifying individual deviations in brain structure and function relative to a population reference. However, its clinical utility depends on well-calibrated models trained on heterogeneous datasets such as those found in neuroimaging. Here, we systematically examine the effect of training sample size on the distributional and centile calibration of hierarchical Bayesian regression (HBR)-based NMs. Using multisite 3D diffusion MRI scans of the brain from 54,583 subjects, spanning almost the entire lifespan (age: 4-91 years), we trained NMs of white matter fractional anisotropy, a key microstructural metric, on subsamples ranging from 5,000 to 40,000 subjects. HBR was modeled with a Sinh-Arcsinh likelihood. Model calibration was evaluated using Kernelized Stein Discrepancy (KSD) to assess distributional agreement of Z-scores with the standard normal distribution; we also used Mean Absolute Centile Error (MACE) to quantify centile accuracy. Both metrics showed consistent and substantial improvements as the training sample size increased, indicating reduced posterior uncertainty and improved estimation of distributional parameters, particularly at the centile extremes. These results demonstrate that large training cohorts are essential for well-calibrated NMs derived from heterogeneous neuroimaging data and highlight the importance of large-scale data aggregation for reliable individual-level inference.

bioinformatics↗

bioRxiv2026

Tractometry-Based Quantification of Along-Tract White-Matter Hemispheric Asymmetry in Alzheimer's Disease

White-matter hemispheric asymmetry is a fundamental property of human brain organization and is known to change in aging, neurodevelopment, and neurodegenerative disorders. Tractometry analyzes diffusion-derived microstructural measures along the full length of tracts, localizing changes to specific tract-segments rather than collapsing tracts into a single value. Yet, existing frameworks lack a principled way to quantify left-right hemispheric asymmetries along homologous tracts. Here, we introduce an asymmetry-aware tractometry framework that integrates a symmetric white-matter atlas with BUAN (Bundle Analytics) to enable anatomically consistent, along-tract comparison of homologous pathways. By defining homologous bundles with a shared template and consistent orientation, each left-hemisphere segment is directly matched to its right-hemisphere counterpart, enabling principled, segment-wise comparison and revealing spatially localized asymmetries along-tract. Applying this framework to diffusion MRI data from the Alzheimers Disease Neuroimaging Initiative (ADNI) comprising 1,215 subjects, we demonstrate how this approach reveals systematic left-right asymmetries across major white-matter pathways and show how these patterns differentiate cognitively normal (CN) individuals from those with mild cognitive impairment (MCI) and dementia. This method provides a sensitive and anatomically grounded tool for studying hemispheric specialization and its disruption in aging and disease, and establishes a general approach for asymmetry-aware tractometry in population neuroimaging studies.

neuroscience↗

bioRxiv2025

Assessing the Influence of Tractography Methods on White Matter Microstructure and Tractometry Analysis in Alzheimer's Disease

Tractometry enables detailed mapping of white matter microstructure along individual tracts and is widely used to study disease effects such as those seen in Alzheimers disease (AD). However, how different tractography algorithms influence tractometry outcomes remains unclear. Here, we compared whole-brain deterministic and probabilistic tractography using the BUndle ANalytics (BUAN) framework in the Alzheimers Disease Neuroimaging Initiative (ADNI) dataset, including 118 AD and 728 cognitively normal (CN) participants. Both approaches revealed the expected pattern of lower fractional anisotropy (FA) and higher mean, radial, and axial diffusivity (MD, RD, AxD) in AD, consistent with white matter degeneration. Despite broadly similar global trends, substantial bundle-level differences emerged between the two tractography methods. Probabilistic tracking produced stronger and more spatially extended effects in the fornix, a small and highly curved limbic pathway vulnerable to AD-related degeneration, whereas deterministic tracking showed greater sensitivity in the posterior segments of the right superior longitudinal fasciculus (SLF R). These discrepancies highlight that the choice of tractography algorithm can alter detecting disease effects, emphasizing the need for cross-method validation to ensure the robustness and interpretability of along-tract measures.

neuroscience↗

bioRxiv2025

Inhibitory control and error processing in Obsessive-Compulsive Disorder: A mega-analysis of task-based fMRI data by the ENIGMA-OCD consortium

ObjectiveObsessive-compulsive disorder (OCD) is a chronic condition in which impaired inhibitory control and excessive error monitoring may contribute to the maintenance of obsessions and compulsions. This mega-analysis investigates neural activation during response inhibition and error processing using adult and pediatric data from the ENIGMA-OCD consortium and the ABCD study. MethodsIndividual participant data was uniformly processed using HALFpipe to extract statistical maps for response inhibition and error processing contrasts. Bayesian multilevel models were used to assess regional and whole-brain effects of OCD, with additional analyses examining the association between the OCD clinical profile and task-related activation. ResultsAcross inhibitory control tasks, both individuals with OCD and control participants showed robust activation in regions implicated in response inhibition and error processing. During response inhibition, compared to controls, adults with OCD showed stronger somatomotor cortex activation, while children with OCD showed stronger occipital cortex activation. Children with likely OCD from the ABCD cohort showed reduced activity in the frontoparietal network in the anterior insula/frontal operculum region. During error processing, relative to controls, adults with OCD showed weaker activation in fronto-striatal regions, while children with OCD showed stronger activation in frontoparietal and attention networks. Greater OCD symptom severity was associated with weaker task-related activation in adults and stronger activation in children during response inhibition. ConclusionCase-control differences in brain activation during inhibitory control varied by age group and task contrast. Symptom severity emerged as the main clinical correlate of activation during inhibition, suggesting that inhibitory control deficits in OCD may be both state-dependent and developmentally specific.

neuroscience↗

bioRxiv2025

MAVeRiC-AD: Mixture-of-experts Agentic Vision-Language Ensemble for Robust MRI Classification of Alzheimer's Disease

Robust classification of Alzheimers disease (AD) from structural T1-weighted MRI (T1w) images remains an unmet clinical need, especially when data is acquired at multiple sites that differ in scanning protocols and population demographics. In this paper, we present MAVeRiC-AD (Mixture-of-experts Agent-guided Vision-Language Ensemble for Robust Imaging-based classification of Alzheimers Disease), an agentic framework that dynamically utilizes the optimal inferencing tool for radiological queries to provide relevant answers to the user. In our framework, we tested three specialized models encoded as callable tools: (1) CNN-AD, a 3D DenseNet trained on T1w intensities only; (2) MOE-VLM, a vision-language model that jointly models the T1w with subject-specific demographics (age, sex, site) via a mixture-of-experts (MoE) projection head; (3) Retrieval engine, a similarity-search module that contextualizes a patient against others from the site and reports % prevalence of AD. A light-weight agent analyzes the user request and then routes the input (image, or image + text) to the appropriate tool, aggregates responses and returns the tool response augmented with its confidence derived from conformal prediction. Experiments were conducted using T1w images from the ADNI (N=4,098) and OASIS-3 (N=600) datasets. Single-site training baselines achieved ROC-AUC = 0.79 (CNN) and 0.82 (VLM) on ADNI. When trained jointly on both sites, MOE-VLM surpassed both image-only and standard vision-language models with ROC-AUC = 0.90 on ADNI and 0.81 on OASIS. MAVeRiC-AD demonstrates that agentic orchestration of complementary expert deep models, coupled with explicit demographic conditioning for multi-site data can improve robustness and interpretability of AD image analysis pipelines and serves as a blueprint for scalable, trustworthy clinical AI assistants.

neuroscience↗

bioRxiv2025

Generating diffusion MRI scalar maps from T1-weighted images using Reversible GANs

Diffusion tensor imaging (DTI) provides valuable insights into brain tissue microstructure, but acquiring high-quality DTI data is time-intensive and not always feasible. To mitigate data scarcity and enhance accessibility, we investigate the generation of synthetic DTI scalar maps--specifically mean diffusivity (MD)--from structural 3D volumetric T1-weighted brain MRI using a reversible generative adversarial network (RevGAN). Unlike conventional pipelines requiring multiple steps, our approach enables a single-step translation from T1 to diffusion-derived measures. We assess the quality and utility of the synthetic maps in two downstream tasks: sex classification and Alzheimers disease classification. Performance comparisons between models trained on real and synthetic DTI maps demonstrate that RevGAN-generated images retain meaningful microstructural features and offer competitive accuracy, underscoring their potential for data augmentation and analysis in neuroimaging workflows. We also examine how well models trained on these data generalize to a new population dataset from India (NIMHANS cohort).

neuroscience↗

bioRxiv2025

Using Diffusion Transformers to Generate Synthetic Diffusion Scalar Maps for Data Augmentation

Generation of high-quality synthetic brain MRI data could be beneficial for advancing neuroimaging research, particularly when access to large-scale, labeled datasets is limited. In this work, we leverage a pretrained Diffusion Transformer (DiT) architecture to synthesize 3D mean diffusivity (MD) scalar maps from the Cam-CAN dataset. To adapt the DiT model--originally trained on 2D natural images--for 3D neuroimaging data, we implemented a preprocessing strategy that tiles 2D slices from 3D volumes into composite 2D images, enabling effective finetuning. The quality of the generated synthetic images was evaluated using Multi-Scale Structural Similarity (MS-SSIM) and Maximum Mean Discrepancy (MMD) metrics, demonstrating high fidelity and anatomical coherence. To assess the utility of synthetic data in downstream tasks, we conducted transfer learning experiments for dementia classification on the ADNI dataset. A sex classification model, trained on both real and synthetic Cam-CAN data, was repurposed for this task, showing that synthetic samples can enhance model performance. These results highlight the potential of diffusion-based generative models for augmenting neuroimaging datasets and supporting clinical applications.

neuroscience↗

bioRxiv2025

Evaluation of Deep Learning Algorithms to Predict Multiple Dementia-Related Neuropathologies from Brain MRI, Clinical and Genetic Data

Alzheimers disease and related dementias (ADRD) involve overlapping neurodegenerative and vascular pathologies--such as amyloid-{beta} (A{beta}), tau, cerebral amyloid angiopathy (CAA), TDP-43, and alpha-synuclein--that complicate diagnosis and treatment. While PET and CSF biomarkers are useful for detecting A{beta} and tau, they are invasive, expensive, and not widely available. In contrast, magnetic resonance imaging (MRI) is non-invasive and widely accessible, offering an opportunity for pathology prediction when combined with deep learning. Most prior studies have focused on single-pathology detection, but there remains a need for models that can jointly predict multiple co-occurring pathologies. In this work, we evaluate deep learning models that integrate structural MRI with demographic, clinical, and genetic data to classify six autopsy-confirmed neuropathologies: A{beta}, tau, CAA, TDP-43, hippocampal sclerosis, and dementia with Lewy bodies. We compare our hybrid deep learning model to AutoGluon, an automated machine learning framework. Our findings support the potential of multimodal AI to enable non-invasive, comprehensive neuropathological profiling in ADRD.

neuroscience↗

bioRxiv2025

DTI versus NODDI White Matter Microstructural Biomarkers of Alzheimer's Disease

Diffusion MRI (dMRI) is a powerful tool to assess white matter (WM) microstructural abnormalities in Alzheimers disease (AD). The fourth phase of the Alzheimers Disease Neuroimaging Initiative (ADNI) now includes multiple multishell dMRI protocols, enabling both traditional and advanced dMRI model analyses. There is a need to evaluate whether multishell data offer deeper insights into WM pathology in AD than more widely available single-shell data by overcoming single-shell model limitations. Here, we fit single-shell DTI and multishell NODDI to dMRI data from 533 ADNI3/4 participants to assess their sensitivity to key clinical indicators of AD such as cognitive impairment, amyloid-beta and tau PET burden. Overall, we found that NODDI offered no major advantages in detecting cognitive impairment and tau pathology, but NODDI was marginally more sensitive to amyloid pathology.

neuroscience↗

bioRxiv2025

Streamline Density Normalization: A Robust Approach to Mitigate Bundle Variability in Multi-Site Diffusion MRI

Tractometry enables quantitative analysis of tissue microstructure is sensitive to variability introduced during tractography and bundle segmentation. Differences in processing parameters and bundle geometry can lead to inconsistent streamline reconstructions and sampling, ultimately affecting the reproducibility of tractometry analysis. In this study, we introduce Streamline Density Normalization (SDNorm), a supervised two-step method designed to reduce variability in bundle reconstructions. SDNorm first computes streamline weights using linear regression to match a subjects bundle to a template streamline density map, then iteratively prunes streamlines to achieve a target density using a novel metric called effective Streamline Point Density (eSPD). We evaluate SDNorm across multiple bundles and acquisition protocols in dMRI data from a subset of subjects from Alzheimers Disease Neuroimaging Initiative and demonstrate that it can significantly reduce variability in streamline density, improve consistency in along-tract microstructure profiles, and provide useful metrics for automated bundle quality control. These results suggest that SDNorm can help enhance the reproducibility and robustness of bundle reconstruction across heterogeneous image acquisition protocols and tractography settings, making it well-suited for large-scale and multi-site neuroimaging studies.

neuroscience↗