Biology subjects

Thomopoulos, S. I.

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

At least 37 records · Page 2Linked to original sources

bioRxiv2025

Executive control in Obsessive-Compulsive Disorder: A worldwide mega-analysis of task-based functional neuroimaging data of the ENIGMA-OCD consortium

ObjectiveObsessive-compulsive disorder (OCD) is associated with impaired executive function and altered activity in associated neural circuits, contributing to reduced goal-directed behavior. To investigate neural activation during executive control, we conducted a mega-analysis in the ENIGMA-OCD consortium pooling individual participant data from 475 individuals with OCD and 345 healthy controls across 15 fMRI tasks collected worldwide. MethodsIndividual participant data was uniformly processed using HALFpipe to construct voxelwise statistical images of executive control and task load contrasts. Parameter estimates extracted from regions of interest were entered into multilevel Bayesian models to examine regional and whole-brain effects of diagnosis, and, within OCD, the influence of medication status, symptom severity, and age of onset on task activation. ResultsWe observed a robust task activation pattern across individuals with OCD and control participants in executive control regions across tasks. Relative to controls, individuals with OCD showed moderate to very strong evidence of weaker activation of the dorsolateral prefrontal cortex, precuneus, frontal eye fields, and inferior parietal lobule during executive control (all positive posterior probabilities [P+]<0.1). Individuals with OCD also showed stronger activation in regions of the default mode network during executive function relative to controls. We found little evidence for differential activation during executive control in task- positive regions related to disease onset, severity and medication status. ConclusionIn the first mega-analysis of fMRI studies of executive function in OCD, we found strong evidence of weaker frontoparietal activation during executive control tasks. Our findings also suggest a failure of default mode network regions to appropriately disengage during task performance in OCD.

neuroscience↗

bioRxiv2025

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

Alzheimers disease (AD) is marked by progressive cognitive decline and memory loss, due to the abnormal accumulation of amyloid-beta (A{beta}) plaques, which facilitate the spread of tau pathology, and a gradually spreading pattern of neuronal loss. Understanding how amyloid positivity affects the brains neural pathways is critical for understanding how the brain changes with AD pathology. Tractometry offers a powerful approach for the in vivo, 3D quantitative assessment of white matter tracts, enabling the localization of microstructural abnormalities in diseased populations and those at risk. In this study, we applied BUAN (Bundle Analytics) tractometry to multi-cohort diffusion MRI data from a total of 1,908 participants: 606 participants in ADNI3 (Alzheimers Disease Neuroimaging Initiative Phase 3) and 1,302 participants from the HABS-HD (Health and Aging Brain Study-Health Disparities). Using BUAN and along-tract statistical analysis, we assessed the localized effects of amyloid positivity on white-matter pathways, which may be further influenced by downstream tau accumulation. Amyloid positivity was quantified via amyloid-sensitive positron emission tomography (PET). BUAN enables tract-specific quantification of white matter microstructure and supports statistical testing along the full length of fiber bundles to detect subtle, spatially localized associations. We present 3D visualizations of tractwise amyloid associations, highlighting distinct patterns of white matter degeneration in AD.

neuroscience↗

bioRxiv2025

Deterministic versus Probabilistic Tractography: Impact on White Matter Bundle Shape

In diffusion MRI-based tractography, deterministic and probabilistic algorithms reconstruct white matter using distinct strategies, yet their impact on bundle morphology remains uncertain. Using bundle shape similarity analysis, we compared both methods for the left arcuate fasciculus (AF_L) (The left arcuate fasciculus is a critical white matter tract that connects language comprehension and production areas in the human brain, enabling fluent language processing) across four datasets: Alzheimers Disease Neuroimaging Initiative (ADNI), Human Connectome Project-Aging (HCP-A), National Institute of Mental Health and Neurosciences (NIMHANS), and Pediatric Imaging, Neurocognition, and Genetics (PING). Probabilistic tractography consistently produced higher inter-subject shape similarity, by capturing broader anatomical trajectories and enhancing reproducibility. However, this extensive coverage may obscure subtle pathological variations critical for clinical detection. Bundle shape similarity analysis with atlas corroborated these findings, showing stronger alignment for probabilistic tracking and highlighting its utility in quantitative quality control. These results emphasize the need to balance morphological consistency with sensitivity to neuroanatomical variation when selecting tractography methods for research and clinical applications.

neuroscience↗

bioRxiv2025

A Scalable Toolkit for Modeling 3D Surface-based Brain Geometry

3D surface-based computational mapping is more sensitive to localized brain alterations in neurological, developmental and psychiatric conditions than traditional gross volumetric analysis, providing fine-scale 3D maps of a wide range of surface-based features. Here we introduce a scalable toolkit for large-scale computational surface analysis, with efficient algorithms for multisite data integration, statistical harmonization, accelerated multivariate statistics, and visualization. We showcase the utility of the toolkit by mapping subcortical shape variations and factors that affect them across 21 international samples from the ENIGMA Bipolar Disorder Working Group (N=3,373).

neuroscience↗

bioRxiv2025

A Core Pattern of Cerebellar and Brainstem Degeneration and Reduced Cerebrocerebellar Structural Covariance in Spinocerebellar Ataxia Type 3 (SCA3): MRI Volumetrics from ENIGMA-Ataxia

ObjectiveSpinocerebellar ataxia type 3 (SCA3) is a rare, inherited neurodegenerative disease. Here, we profile the spatial spread of atrophy across the whole brain, determine whether brain degeneration preferentially maps onto specific functional networks, and investigate the relationship between cerebellar and cerebral anatomical changes. MethodsWhole-brain grey and white matter (GM and WM) voxel-based morphometry was performed on 408 individuals with SCA3 (82 pre-ataxic) and 293 controls. The SCA3 cohort was stratified by ataxia severity to investigate disease progression, with cerebellar GM atrophy mapped onto a task-based functional atlas. Volume was correlated with disease duration and intensity. Cerebrocerebellar volumetric covariance was assessed to determine whether atrophy was coupled between infra- and supratentorial regions. ResultsThe pattern of atrophy is spatially consistent but progressive in magnitude across the disease course. The greatest atrophy was found in the pons, cerebellar WM, and cerebellar peduncles; correlations with disease severity and duration were also strongest in these regions. Cerebellar GM atrophy was greatest in functional regions associated with motor execution and planning, attention, and emotional processing. Sparse cerebral cortical atrophy appears only in the most severe disease subgroup, while striatal atrophy begins in the earliest stages. Reduced cerebrocerebellar structural covariance is observed in SCA3 participants versus controls. InterpretationWhile cerebellar and brainstem atrophy become more severe, the pattern of atrophy remains largely consistent as SCA3 progresses. Cerebellar GM degeneration occurs in regions associated with motor, cognitive, and affective control, in line with clinical presentation. Cerebellar atrophy is not directly mirrored by cerebral changes.

neuroscience↗

bioRxiv2025

Lifestyle, Early-life, and Genetic Health Risk Factors Underlying the Brain Age Gap: A Mega-Analysis Across 3,934 Individuals from the ENIGMA MDD Consortium

BackgroundLarge-scale studies show that adults with major depressive disorder (MDD) generally have a higher imaging-predicted age relative to their chronological age (i.e., positive brain age gap) compared to controls, though considerable within-group variation exists. This study examines lifestyle, early-life, and genetic health risk factors contributing to the brain age gap. Identifying risk and resilience factors could help protect brain and mental health. MethodsUsing an established model trained on FreeSurfer-derived brain regions (www.photon-ai.com/enigma_brainage), we generated brain age predictions for 1,846 controls and 2,088 individuals with MDD (aged 18-75) from 12 international cohorts. Polygenic risk scores (PRS) were calculated for major depression, C-reactive protein, and body mass index (BMI) using large-scale GWAS results. Linear mixed models were applied to assess lifestyle (BMI, smoking, education), early-life childhood trauma, and genetic (PRS) health risk associations with the brain age gap. Additionally, we evaluated the link between the brain age gap and peripheral biological age indicators (epigenetic clocks). ResultsHigher brain age gaps were significantly associated with BMI ({beta}=0.01, PFDR=0.02) and smoking ({beta}=0.11, PFDR=0.02), while lower brain age gaps were linked to higher education ({beta}=-0.02, PFDR=0.02). Higher childhood trauma scores predicted a higher brain age gap ({beta}=0.04, P=0.01). Higher brain age gaps were positively associated with all PRS ({beta}s=0.04-0.16, PsFDR=0.02-0.03). There were no significant interactions between diagnosis and assessed factors on the brain age gap. In a multivariable model, only modifiable health factors--BMI, smoking, and education--remained uniquely associated with brain age gaps. ConclusionsGenetic liability for depression and related traits is linked to poorer brain health, but health behaviors potentially offer a key opportunity for intervention. This study underscores the importance of targeting modifiable lifestyle factors to mitigate poor brain health in depressed individuals, an approach perhaps under-recognized in clinical practice.

neuroscience↗

bioRxiv2025

Synthetic Diffusion Tensor Imaging Maps Generated by 2D and 3D Probabilistic Diffusion Models: Evaluation and Applications

Diffusion tensor imaging (DTI) is a key neuroimaging modality for assessing brain tissue microstructure, yet high-quality acquisitions are costly, time-intensive, and prone to artifacts. To address data scarcity and privacy concerns - and to augment the available data for training deep learning methods - synthetic DTI generation has gained interest. Specifically, denoising diffusion probabilistic models (DDPMs) have emerged as a promising approach due to their superior fidelity, diversity, controllability, and stability compared to generative adversarial networks (GANs) and variational autoencoders (VAEs). In this work, we evaluate the quality, fidelity and added value for downstream applications of synthetic DTI mean diffusivity (MD) maps generated by 2D slice-wise and 3D volume-wise DDPMs. We evaluate their computational efficiency and utility for data augmentation in two downstream tasks: sex classification and dementia classification using 2D and 3D convolutional neural networks (CNNs). Our findings show that 3D synthesis outperforms 2D slice-wise generation in downstream tasks. We present a benchmark analysis of synthetic diffusion-weighted imaging approaches, highlighting key trade-offs in image quality, diversity, efficiency, and downstream performance.

neuroscience↗

bioRxiv2025

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

Predicting the trajectory of clinical decline in aging individuals is a pressing challenge, especially for people with mild cognitive impairment, Alzheimers disease, Parkinsons disease, or vascular dementia. Accurate predictions can guide treatment decisions, identify risk factors, and optimize clinical trials. In this study, we compared two deep learning approaches for forecasting changes, over a 2-year interval, in the Clinical Dementia Rating scale sum of boxes score (sobCDR). This is a key metric in dementia research, and scores range from 0 (no impairment) to 18 (severe impairment). To predict decline, we trained a hybrid convolutional neural network that integrates 3D T1-weighted brain MRI scans with tabular clinical and demographic features (including age, sex, body mass index (BMI), and baseline sobCDR). We benchmarked its performance against AutoGluon, an automated multimodal machine learning framework that selects an appropriate neural network architecture. Our results demonstrate the importance of combining image and tabular data in predictive modeling for clinical applications. Deep learning algorithms can fuse image-based brain signatures and tabular clinical data, with potential for personalized prognostics in aging and dementia.

neuroscience↗

bioRxiv2025

Parameter Efficient Fine-tuning of Transformer-based Masked Autoencoder Enhances Resource Constrained Neuroimage Analysis

Recent innovations in artificial intelligence (AI) have increasingly focused on large-scale foundational models that are more general purpose in contrast to conventional models trained to perform specialized tasks. Transformer-based architectures have become the standard backbone in foundation models across data modalities (image, text, audio, video). There has been a keen interest in applying parameter-efficient fine-tuning (PEFT) methods to adapt these models to specialized downstream tasks in language and vision. These methods are particularly essential for medical image analysis where the limited availability of training data could lead to overfitting. In this work, we evaluated different types of PEFT methods on pre-trained vision transformers relative to typical training approaches, such as full fine-tuning and training from scratch. We used a transformer-based masked autoencoder (MAE) framework, to pretrain a vision encoder on T1- weighted (T1-w) brain MRIs. The pretrained vision transformers were then fine-tuned using different PEFT methods that reduced the trainable model parameters to as few as 0.04% of the original model size. Our study shows that: 1. PEFT methods were competitive with or outperformed the reference full fine-tuning approach and outperformed training from scratch, with only a fraction of the trainable parameters; 2. PEFT methods with a 32% reduction in model size boosted Alzheimers disease (AD) classification by 3% relative to full fine-tuning and 11% relative to a 3D CNN, with only 258 training scans; and 3. PEFT methods performed well on diverse neuroimaging tasks including AD and Parkinsons disease (PD) classification, and "brain-age" prediction based on T1-w MRI datasets - a standard benchmark for deep learning models in neuroimaging; 4. smaller model sizes were competitive with larger models in test performance. Our results show the value of adapting foundation models to neuroimaging tasks efficiently and effectively in contrast to training stand- alone special purpose models.

neuroscience↗

bioRxiv2025

Leveraging a Vision-Language Model with Natural Text Supervision for MRI Retrieval, Captioning, Classification, and Visual Question Answering

Large multimodal models are now extensively used worldwide, with the most powerful ones trained on massive, general-purpose datasets. Despite their rapid deployment, concerns persist regarding the quality and domain relevance of the training data, especially in radiology, medical research, and neuroscience. Additionally, healthcare data privacy is paramount when querying models trained on medical data, as is transparency regarding service hosting and data storage. So far, most deep learning algorithms in radiologic research are designed to perform a specific task (e.g., diagnostic classification) and cannot be prompted to perform multiple tasks using natural language. In this work, we introduce a framework based on vector retrieval and contrastive learning to efficiently learn visual brain MRI concepts via natural language supervision. We show how the method learns to identify factors that affect the brain in Alzheimers disease (AD) via joint embedding and natural language supervision. First, we pre-train separate text and image encoders using self-supervised learning, and jointly fine-tune these encoders to develop a shared embedding space. We train our model to perform multiple tasks, including MRI retrieval, MRI captioning, and MRI classification. We show its versatility by developing a retrieval and re-ranking mechanism along with a transformer decoder for visual question answering. Clinical RelevanceBy learning a cross-modal embedding of radiologic features and text, our approach can learn to perform diagnostic and prognostic assessments in AD research as well as to assist practicing clinicians. Integrating medical imaging with clinical descriptions and text prompts, we aim to provide a general, versatile tool for detecting radiologic features described by text, offering a new approach to radiologic research.

neuroscience↗

bioRxiv2024

Lifespan Normative Modeling of Brain Microstructure

Normative models of brain metrics based on large populations could be extremely valuable for detecting brain abnormalities in patients with a variety of disorders, including degenerative, psychiatric and neurodevelopmental conditions, but no such models exist for the brains white matter (WM) microstructure. Here we present a large-scale normative model of brain WM microstructure - based on 19 international diffusion MRI datasets covering almost the entire lifespan (totaling N=54,583 individuals; age: 4-91 years). We extracted regional diffusion tensor imaging (DTI) metrics using a standardized analysis and quality control protocol and used hierarchical Bayesian regression (HBR) to model the statistical distribution of derived WM metrics as a function of age and sex. We extracted the average lifespan trajectories and corresponding centile curves for each WM region. We illustrate the utility of the method by applying it to detect and visualize profiles of WM microstructural deviations in a variety of contexts: in mild cognitive impairment, Alzheimers disease, and 22q11.2 deletion syndrome - a neurogenetic condition that markedly increases risk for schizophrenia. The resulting large-scale model provides a common reference to identify disease effects on the brains microstructure in individuals or groups, and to compare disorders, and discover factors affecting WM abnormalities. The derived normative models are a valuable resource publicly available to the community, adaptable and extendable to future datasets as the available data expands.

neuroscience↗

bioRxiv2024

Evaluating Synthetic Diffusion MRI Maps created with Diffusion Denoising Probabilistic Models

Generative AI models, such as Stable Diffusion, DALL-E, and MidJourney, have recently gained widespread attention as they can generate high-quality synthetic images by learning the distribution of complex, high-dimensional image data. These models are now being adapted for medical and neuroimaging data, where AI-based tasks such as diagnostic classification and predictive modeling typically use deep learning methods, such as convolutional neural networks (CNNs) and vision transformers (ViTs), with interpretability enhancements. In our study, we trained latent diffusion models (LDM) and denoising diffusion probabilistic models (DDPM) specifically to generate synthetic diffusion tensor imaging (DTI) maps. We developed models that generate synthetic DTI maps of mean diffusivity by training on real 3D DTI scans, and evaluating realism and diversity of the synthetic data using maximum mean discrepancy (MMD) and multi-scale structural similarity index (MS-SSIM). We also assess the performance of a 3D CNN-based sex classifier, by training on combinations of real and synthetic DTIs, to check if performance improved when adding the synthetic scans during training, as a form of data augmentation. Our approach efficiently produces realistic and diverse synthetic data, potentially helping to create interpretable AI-driven maps for neuroscience research and clinical diagnostics.

neuroscience↗

bioRxiv2024

A Surface-based deep learning approach for cortical shape analysis

Advances in deep learning hold promise for predicting clinical factors from human brain images. In this study, we applied a spherical harmonics-based convolutional neural network approach (SPHARM-Net) to MRI-derived brain shape metrics to predict age, sex, and Alzheimers disease (AD) diagnosis. MRI-derived brain features included vertex-wise cortical curvature, convexity, thickness, and surface area. SPHARM-Net performs convolutions using the spherical harmonic transforms, eliminating the need to explicitly define neighborhood size, and achieving rotational equivariance. Sex classification and age regression were carried out in a large sample of healthy adults (UK Biobank; N=32,979), and AD classification performance was tested in a large, publicly available sample (ADNI; N=1,213). SPHARM-Net showed strong performance for sex classification (accuracy=0.91; balanced accuracy= 0.91; AUC=0.97), and age regression (average absolute error=2.97 years; R-squared=0.77; Pearsons coefficient=0.9). AD classification also performed well (accuracy=0.86; balanced accuracy=0.83; AUC=0.9). Our experiments demonstrate promising preliminary performance using the SPHARM-Net for two widely studied benchmarking tasks and for AD classification. Future work will include comparisons of shape-based methods and extending these analysis to more challenging tasks such as mood disorder classification.

neuroscience↗

bioRxiv2024

The Pattern and Staging of Brain Atrophy in Spinocerebellar Ataxia Type 2 (SCA2): MRI Volumetrics from ENIGMA-Ataxia

ObjectiveSpinocerebellar ataxia type 2 (SCA2) is a rare, inherited neurodegenerative disease characterised by progressive deterioration in both motor coordination and cognitive function. Atrophy of the cerebellum, brainstem, and spinal cord are core features of SCA2, however the evolution and pattern of whole-brain atrophy in SCA2 remain unclear. We undertook a multi-site, structural magnetic resonance imaging (MRI) study to comprehensively characterize the neurodegeneration profile of SCA2. MethodsVoxel-based morphometry analyses of 110 participants with SCA2 and 128 controls were undertaken to assess groupwise differences in whole-brain volume. Correlations with clinical severity and genotype, and cross-sectional profiling of atrophy patterns at different disease stages, were also performed. ResultsAtrophy in SCA2 relative to controls was greatest (Cohens d>2.5) in the cerebellar white matter (WM), middle cerebellar peduncle, pons, and corticospinal tract. Very large effects (d>1.5) were also evident in the superior cerebellar, inferior cerebellar, and cerebral peduncles. In cerebellar grey matter (GM), large effects (d>0.8) mapped to areas related to both motor coordination and cognitive tasks. Strong correlations (|r|>0.4) between volume and disease severity largely mirrored these groupwise outcomes. Stratification by disease severity showed a degeneration pattern beginning in cerebellar and pontine WM in pre-clinical subjects; spreading to the cerebellar GM and cerebro-cerebellar/corticospinal WM tracts; then finally involving the thalamus, striatum, and cortex in severe stages. InterpretationThe magnitude and pattern of brain atrophy evolves over the course of SCA2, with widespread, non-uniform involvement across the brainstem, cerebellar tracts, and cerebellar cortex; and late involvement of the cerebral cortex and striatum.

neuroscience↗

bioRxiv2024

Comparison of Explainable AI Models for MRI-based Alzheimer's Disease Classification

Deep learning models based on convolutional neural networks (CNNs) have been used to classify Alzheimers disease or infer dementia severity from 3D T1-weighted brain MRI scans. Here, we examine the value of adding occlusion sensitivity analysis (OSA) and gradient-weighted class activation mapping (Grad-CAM) to these models to make the results more interpretable. Much research in this area focuses on specific datasets such as the Alzheimers Disease Neuroimaging Initiative (ADNI) or National Alzheimers Coordinating Center (NACC), which assess people of North American, predominantly European ancestry, so we examine how well models trained on these data generalize to a new population dataset from India (NIMHANS cohort). We also evaluate the benefit of using a combined dataset to train the CNN models. Our experiments show feature localization consistent with knowledge of AD from other methods. OSA and Grad-CAM resolve features at different scales to help interpret diagnostic inferences made by CNNs.

neuroscience↗

bioRxiv2024

Amyloid, Tau, and APOE in Alzheimer's Disease: Impact on White Matter Tracts

Alzheimers disease (AD) is characterized by cognitive decline and memory loss due to the abnormal accumulation of amyloid-beta (A{beta}) plaques and tau tangles in the brain; its onset and progression also depend on genetic factors such as the apolipoprotein E (APOE) genotype. Understanding how these factors affect the brains neural pathways is important for early diagnostics and interventions. Tractometry is an advanced technique for 3D quantitative assessment of white matter tracts, localizing microstructural abnormalities in diseased populations in vivo. In this work, we applied BUAN (Bundle Analytics) tractometry to 3D diffusion MRI data from 730 participants in ADNI3 (phase 3 of the Alzheimers Disease Neuroimaging Initiative; age range: 55-95 years, 349M/381F, 214 with mild cognitive impairment, 69 with AD, and 447 cognitively healthy controls). Using along-tract statistical analysis, we assessed the localized impact of amyloid, tau, and APOE genetic variants on the brains neural pathways. BUAN quantifies microstructural properties of white matter tracts, supporting along-tract statistical analyses that identify factors associated with brain microstructure. We visualize the 3D profile of white matter tract associations with tau and amyloid burden in Alzheimers disease; strong associations near the cortex may support models of disease propagation along neural pathways. Relative to the neutral genotype, APOE{varepsilon} 3/{varepsilon}3, carriers of the AD-risk conferring APOE{varepsilon} 4 genotype show microstructural abnormalities, while carriers of the protective{varepsilon} 2 genotype also show subtle differences. Of all the microstructural metrics, mean diffusivity (MD) generally shows the strongest associations with AD pathology, followed by axial diffusivity (AxD) and radial diffusivity (RD), while fractional anisotropy (FA) is typically the least sensitive metric. Along-tract microstructural metrics are sensitive to tau and amyloid accumulation, showing the potential of diffusion MRI to track AD pathology and map its impact on neural pathways.

neuroscience↗

bioRxiv2024

Microstructural Mapping of Neural Pathways in Alzheimer's Disease using Macrostructure-Informed Normative Tractometry

IntroductionDiffusion MRI is sensitive to the microstructural properties of brain tissues, and shows great promise in detecting the effects of degenerative diseases. However, many approaches analyze single measures averaged over regions of interest, without considering the underlying fiber geometry. MethodsHere, we propose a novel Macrostructure-Informed Normative Tractometry (MINT) framework, to investigate how white matter microstructure and macrostructure are jointly altered in mild cognitive impairment (MCI) and dementia. We compare MINT-derived metrics with univariate metrics from diffusion tensor imaging (DTI), to examine how fiber geometry may impact interpretation of microstructure. ResultsIn two multi-site cohorts from North America and India, we find consistent patterns of microstructural and macrostructural anomalies implicated in MCI and dementia; we also rank diffusion metrics sensitivity to dementia. DiscussionWe show that MINT, by jointly modeling tract shape and microstructure, has potential to disentangle and better interpret the effects of degenerative disease on the brains neural pathways.

neuroscience↗

bioRxiv2024

Counterfactual MRI Generation with Denoising Diffusion Models for Interpretable Alzheimer's Disease Effect Detection

Generative AI models have recently achieved mainstream attention with the advent of powerful approaches such as stable diffusion, DALL-E and MidJourney. The underlying breakthrough generative mechanism of denoising diffusion modeling can generate high quality synthetic images and can learn the underlying distribution of complex, high-dimensional data. Recent research has begun to extend these models to medical and specifically neuroimaging data. Typical neuroimaging tasks such as diagnostic classification and predictive modeling often rely on deep learning approaches based on convolutional neural networks (CNNs) and vision transformers (ViTs), with additional steps to help in interpreting the results. In our paper, we train conditional latent diffusion models (LDM) and denoising diffusion probabilistic models (DDPM) to provide insight into Alzheimers disease (AD) effects on the brains anatomy at the individual level. We first created diffusion models that could generate synthetic MRIs, by training them on real 3D T1-weighted MRI scans, and conditioning the generative process on the clinical diagnosis as a context variable. We conducted experiments to overcome limitations in training dataset size, compute time and memory resources, testing different model sizes, effects of pretraining, training duration, and latent diffusion models. We tested the sampling quality of the disease-conditioned diffusion using metrics to assess realism and diversity of the generated synthetic MRIs. We also evaluated the ability of diffusion models to conditionally sample MRI brains using a 3D CNN-based disease classifier relative to real MRIs. In our experiments, the diffusion models generated synthetic data that helped to train an AD classifier (using only 500 real training scans) -and boosted its performance by over 3% when tested on real MRI scans. Further, we used implicit classifier-free guidance to alter the conditioning of an encoded individual scan to its counterfactual (representing a healthy subject of the same age and sex) while preserving subject-specific image details. From this counterfactual image (where the same person appears healthy), a personalized disease map was generated to identify possible disease effects on the brain. Our approach efficiently generates realistic and diverse synthetic data, and may create interpretable AI-based maps for neuroscience research and clinical diagnostic applications.

bioinformatics↗