Sophia I. Thomopoulos

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

This paper investigates why performance plateaus in 3D MRI transformers for brain disease classification and finds that representation collapse, not task difficulty, causes the plateau.

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

This study analyzes fMRI data from OCD patients and controls to identify age-specific differences in brain activation during inhibitory control and error processing.

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

This paper evaluates automated white matter lesion segmentation methods in Parkinson's disease and healthy controls, finding U-Net-pgs to be the most accurate and reliable approach across multiple metrics and datasets.

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

This study evaluates how sample size affects machine learning performance in classifying dementia stages using subcortical brain features from 2,511 ADNI samples.

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

This study examines how the spectral distribution of discriminative signals in MRI transformers affects their scaling behavior with increasing training data, finding that distributed representations improve more with larger datasets.

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

The paper compares three harmonization methods for multisite diffusion MRI data to reduce protocol-related variability while preserving associations with cognitive impairment and PET biomarkers.

Mapping Higher-Order Topology in OCD Brain Networks with Hodge Laplacian

This paper introduces a Hodge-Laplacian framework to analyze higher-order topology in OCD brain networks, identifying loop-level abnormalities not detectable by conventional functional connectivity analysis.

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

Sex-related structural alterations across common epilepsies: a worldwide ENIGMA study

This study examines sex differences in structural brain alterations in common epilepsies using neuroimaging data from a large international cohort, finding some regional differences but no significant sex-by-diagnosis interactions.

Lifespan Normative Modeling of Brain Microstructure

This paper develops a lifespan normative model of brain white matter microstructure using diffusion MRI data from 54,583 individuals, enabling detection of microstructural deviations in diseases like Alzheimer's and 22q11.2 deletion syndrome.

The Regional Vulnerability Index (RVI) as a Neuroimaging-Based Biomarker for Autism: Associations with Likelihood, Cognition, and Longitudinal Social Outcomes

The paper develops and evaluates the Regional Vulnerability Index (RVI-ASD) as a neuroimaging biomarker for autism, showing its stability, heritability, and associations with cognition and social outcomes.

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

The paper uses machine learning to predict autopsy-confirmed neuropathology from clinical, neuroimaging, and CSF biomarkers, finding that different modalities best predict specific pathologies.

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

This paper introduces an asymmetry-aware tractometry framework to quantify left-right hemispheric asymmetries in white matter tracts and applies it to Alzheimer's Disease data to differentiate cognitive states.

SPECTRA: Spatial Inference for Tractometry Toward Precision Mapping of White Matter Microstructure

SPECTRA improves white matter microstructure mapping by incorporating spatial resolution and adaptive error control in tractometry analysis.

Structural Brain Network Alterations in Relation to Treatment and Illness Severity in Bipolar Disorder

This study analyzes structural brain network changes in bipolar disorder, finding widespread connectivity disruptions linked to illness severity and medication use.

BundleWarp: Enhancing White Matter Tractometry and Morphometry with Precise Neuronal Mapping using Streamline-based Nonlinear Registration

BundleWarp improves white matter tractometry by aligning tracts while preserving anatomical features, enhancing sensitivity to disease-related changes in conditions like Alzheimer's and Parkinson's.

Assessing the Influence of Tractography Methods on Detected White Matter Microstructure in Alzheimer's disease

The paper compares deterministic and probabilistic tractography methods in ADNI data, finding bundle-level differences in white matter microstructure metrics between the two approaches.

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

CALM-VLM improves confidence and reliability in brain MRI classification by calibrating vision-language models and enabling selective prediction.

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

MAVeRiC-AD improves Alzheimer's disease classification from MRI by combining multiple models with demographic data, achieving higher accuracy across different sites.

Childhood Maltreatment and Deviations from Normative Brain Structure: Results from 3,711 Individuals from the ENIGMA MDD and ENIGMA PTSD

The paper examines how childhood maltreatment affects brain structure in 3,711 individuals with MDD and PTSD, finding age- and sex-specific brain morphometric differences.

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

The paper examines how increasing training sample size improves the calibration of diffusion tensor imaging normative models using hierarchical Bayesian regression.

Deep Learning to Predict Future Cognitive Decline: A Multimodal Approach Using Brain MRI and Clinical Data

This study compares deep learning methods to predict cognitive decline using brain MRI and clinical data over two years, showing the value of integrating image and tabular data.

Brain Age in Conduct Disorder: A Mega-Analysis of the ENIGMA Antisocial Behavior Working Group

This study investigates brain age differences in youth with conduct disorder compared to controls, finding accelerated brain maturation in those with adolescence-onset CD.

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

This paper introduces SDNorm, a method to reduce variability in diffusion MRI streamline reconstructions, improving reproducibility across different acquisition protocols.

Shared genetic architecture of brain age gap across 30 cohorts worldwide

This study identifies a shared genetic factor underlying brain age gaps across 30 cohorts, finding 19 SNPs and associations with health traits.

Network-based amyloid-beta pathology predicts subsequent cognitive decline in cognitively normal older adults

This study uses machine learning to show that integrating brain connectome data with amyloid-beta PET imaging improves prediction of cognitive decline in older adults.

White Matter Tract Vulnerability to Amyloid Pathology on the Alzheimer's Disease Continuum

This study uses BUAN tractometry to assess how amyloid positivity affects white matter tracts in Alzheimer's disease, revealing localized microstructural changes associated with amyloid pathology.

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

This paper generates synthetic brain MRI data using a Diffusion Transformer to augment neuroimaging datasets and improve dementia classification.

Deterministic versus Probabilistic Tractography: Impact on White Matter Bundle Shape

This study compares deterministic and probabilistic tractography methods for the left arcuate fasciculus across four datasets, finding probabilistic methods yield higher inter-subject shape similarity but may obscure subtle pathological variations.

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

This paper generates synthetic diffusion MRI scalar maps from T1-weighted images using RevGAN, showing they retain microstructural features and perform well in downstream tasks.