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

INTRODUCTION: Diffusion 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. METHODS: Here, 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. RESULTS: In 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. DISCUSSION: We show that MINT, by jointly modeling tract shape and microstructure, has potential to disentangle and better interpret the effects of degenerative disease on the brain's neural pathways.

Deep Normative Tractometry for Identifying Joint White Matter Macro- and Micro-structural Abnormalities in Alzheimer's Disease.

This study introduces the Deep Normative Tractometry (DNT) framework, that encodes the joint distribution of both macrostructural and microstructural profiles of the brain white matter tracts through a variational autoencoder (VAE). By training on data from healthy controls, DNT learns the normative distribution of tract data, and can delineate along-tract micro-and macro-structural abnormalities. Leveraging a large sample size via generative pre-training, we assess DNT's generalizability using transfer learning on data from an independent cohort acquired in India. Our findings demonstrate DNT's capacity to detect widespread diffusivity abnormalities along tracts in mild cognitive impairment and Alzheimer's disease, aligning closely with results from the Bundle Analytics (BUAN) tractometry pipeline. By incorporating tract geometry information, DNT may be able to distinguish disease-related abnormalities in anisotropy from tract macrostructure, and shows promise in enhancing fine-scale mapping and detection of white matter alterations in neurodegenerative conditions.

Brain Age Analysis and Dementia Classification using Convolutional Neural Networks trained on Diffusion MRI: Tests in Indian and North American Cohorts.

Deep learning models based on convolutional neural networks (CNNs) have been used to classify Alzheimer's disease or infer dementia severity from T1-weighted brain MRI scans. Here, we examine the value of adding diffusion-weighted MRI (dMRI) as an input to these models. Much research in this area focuses on specific datasets such as the Alzheimer's Disease Neuroimaging Initiative (ADNI), which assesses people of North American, largely European ancestry, so we examine how models trained on ADNI, generalize to a new population dataset from India (the NIMHANS cohort). We first benchmark our models by predicting 'brain age' - the task of predicting a person's chronological age from their MRI scan and proceed to AD classification. We also evaluate the benefit of using a 3D CycleGAN approach to harmonize the imaging datasets before training the CNN models. Our experiments show that classification performance improves after harmonization in most cases, as well as better performance for dMRI as input.

Genotype-specific spinal cord damage in spinocerebellar ataxias: an ENIGMA-Ataxia study.

BACKGROUND: Spinal cord damage is a feature of many spinocerebellar ataxias (SCAs), but well-powered in vivo studies are lacking and links with disease severity and progression remain unclear. Here we characterise cervical spinal cord morphometric abnormalities in SCA1, SCA2, SCA3 and SCA6 using a large multisite MRI dataset. METHODS: Upper spinal cord (vertebrae C1-C4) cross-sectional area (CSA) and eccentricity (flattening) were assessed using MRI data from nine sites within the ENIGMA-Ataxia consortium, including 364 people with ataxic SCA, 56 individuals with preataxic SCA and 394 nonataxic controls. Correlations and subgroup analyses within the SCA cohorts were undertaken based on disease duration and ataxia severity. RESULTS: Individuals in the ataxic stage of SCA1, SCA2 and SCA3, relative to non-ataxic controls, had significantly reduced CSA and increased eccentricity at all examined levels. CSA showed large effect sizes (d>2.0) and correlated with ataxia severity (r<-0.43) and disease duration (r<-0.21). Eccentricity correlated only with ataxia severity in SCA2 (r=0.28). No significant spinal cord differences were evident in SCA6. In preataxic individuals, CSA was significantly reduced in SCA2 (d=1.6) and SCA3 (d=1.7), and the SCA2 group also showed increased eccentricity (d=1.1) relative to nonataxic controls. Subgroup analyses confirmed that CSA and eccentricity are abnormal in early disease stages in SCA1, SCA2 and SCA3. CSA declined with disease progression in all, whereas eccentricity progressed only in SCA2. CONCLUSIONS: Spinal cord abnormalities are an early and progressive feature of SCA1, SCA2 and SCA3, but not SCA6, which can be captured using quantitative MRI.

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 Alzheimer's disease (AD) effects on the brain's 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.

Mapping gray and white matter volume abnormalities in early-onset psychosis: an ENIGMA multicenter voxel-based morphometry study.

INTRODUCTION: Regional gray matter (GM) alterations have been reported in early-onset psychosis (EOP, onset before age 18), but previous studies have yielded conflicting results, likely due to small sample sizes and the different brain regions examined. In this study, we conducted a whole brain voxel-based morphometry (VBM) analysis in a large sample of individuals with EOP, using the newly developed ENIGMA-VBM tool. METHODS: 15 independent cohorts from the ENIGMA-EOP working group participated in the study. The overall sample comprised T1-weighted MRI data from 482 individuals with EOP and 469 healthy controls. Each site performed the VBM analysis locally using the standardized ENIGMA-VBM tool. Statistical parametric T-maps were generated from each cohort and meta-analyzed to reveal voxel-wise differences between EOP and healthy controls as well as the individual-based association between GM volume and age of onset, chlorpromazine (CPZ) equivalent dose, and other clinical variables. RESULTS: Compared with healthy controls, individuals with EOP showed widespread lower GM volume encompassing most of the cortex, with the most marked effect in the left median cingulate (Hedges' g = 0.55, p = 0.001 corrected), as well as small clusters of lower white matter (WM), whereas no regional GM or WM volumes were higher in EOP. Lower GM volume in the cerebellum, thalamus and left inferior parietal gyrus was associated with older age of onset. Deficits in GM in the left inferior frontal gyrus, right insula, right precentral gyrus and right superior frontal gyrus were also associated with higher CPZ equivalent doses. CONCLUSION: EOP is associated with widespread reductions in cortical GM volume, while WM is affected to a smaller extent. GM volume alterations are associated with age of onset and CPZ equivalent dose but these effects are small compared to case-control differences. Mapping anatomical abnormalities in EOP may lead to a better understanding of the role of psychosis in brain development during childhood and adolescence.

Smaller total and subregional cerebellar volumes in posttraumatic stress disorder: a mega-analysis by the ENIGMA-PGC PTSD workgroup.

Although the cerebellum contributes to higher-order cognitive and emotional functions relevant to posttraumatic stress disorder (PTSD), prior research on cerebellar volume in PTSD is scant, particularly when considering subregions that differentially map on to motor, cognitive, and affective functions. In a sample of 4215 adults (PTSD n = 1642; Control n = 2573) across 40 sites from the ENIGMA-PGC PTSD working group, we employed a new state-of-the-art deep-learning based approach for automatic cerebellar parcellation to obtain volumetric estimates for the total cerebellum and 28 subregions. Linear mixed effects models controlling for age, gender, intracranial volume, and site were used to compare cerebellum volumes in PTSD compared to healthy controls (88% trauma-exposed). PTSD was associated with significant grey and white matter reductions of the cerebellum. Compared to controls, people with PTSD demonstrated smaller total cerebellum volume, as well as reduced volume in subregions primarily within the posterior lobe (lobule VIIB, crus II), vermis (VI, VIII), flocculonodular lobe (lobule X), and corpus medullare (all p-FDR < 0.05). Effects of PTSD on volume were consistent, and generally more robust, when examining symptom severity rather than diagnostic status. These findings implicate regionally specific cerebellar volumetric differences in the pathophysiology of PTSD. The cerebellum appears to play an important role in higher-order cognitive and emotional processes, far beyond its historical association with vestibulomotor function. Further examination of the cerebellum in trauma-related psychopathology will help to clarify how cerebellar structure and function may disrupt cognitive and affective processes at the center of translational models for PTSD.

Examining the association between posttraumatic stress disorder and disruptions in cortical networks identified using data-driven methods.

Posttraumatic stress disorder (PTSD) is associated with lower cortical thickness (CT) in prefrontal, cingulate, and insular cortices in diverse trauma-affected samples. However, some studies have failed to detect differences between PTSD patients and healthy controls or reported that PTSD is associated with greater CT. Using data-driven dimensionality reduction, we sought to conduct a well-powered study to identify vulnerable networks without regard to neuroanatomic boundaries. Moreover, this approach enabled us to avoid the excessive burden of multiple comparison correction that plagues vertex-wise methods. We derived structural covariance networks (SCNs) by applying non-negative matrix factorization (NMF) to CT data from 961 PTSD patients and 1124 trauma-exposed controls without PTSD. We used regression analyses to investigate associations between CT within SCNs and PTSD diagnosis (with and without accounting for the potential confounding effect of trauma type) and symptom severity in the full sample. We performed additional regression analyses in subsets of the data to examine associations between SCNs and comorbid depression, childhood trauma severity, and alcohol abuse. NMF identified 20 unbiased SCNs, which aligned closely with functionally defined brain networks. PTSD diagnosis was most strongly associated with diminished CT in SCNs that encompassed the bilateral superior frontal cortex, motor cortex, insular cortex, orbitofrontal cortex, medial occipital cortex, anterior cingulate cortex, and posterior cingulate cortex. CT in these networks was significantly negatively correlated with PTSD symptom severity. Collectively, these findings suggest that PTSD diagnosis is associated with widespread reductions in CT, particularly within prefrontal regulatory regions and broader emotion and sensory processing cortical regions.

Childhood trauma moderates schizotypy-related brain morphology: analyses of 1182 healthy individuals from the ENIGMA schizotypy working group.

BACKGROUND: Schizotypy represents an index of psychosis-proneness in the general population, often associated with childhood trauma exposure. Both schizotypy and childhood trauma are linked to structural brain alterations, and it is possible that trauma exposure moderates the extent of brain morphological differences associated with schizotypy. METHODS: We addressed this question using data from a total of 1182 healthy adults (age range: 18-65 years old, 647 females/535 males), pooled from nine sites worldwide, contributing to the Enhancing NeuroImaging Genetics through Meta-Analysis (ENIGMA) Schizotypy working group. All participants completed both the Schizotypal Personality Questionnaire Brief version (SPQ-B), and the Childhood Trauma Questionnaire (CTQ), and underwent a 3D T1-weighted brain MRI scan from which regional indices of subcortical gray matter volume and cortical thickness were determined. RESULTS: A series of multiple linear regressions revealed that differences in cortical thickness in four regions-of-interest were significantly associated with interactions between schizotypy and trauma; subsequent moderation analyses indicated that increasing levels of schizotypy were associated with thicker left caudal anterior cingulate gyrus, right middle temporal gyrus and insula, and thinner left caudal middle frontal gyrus, in people exposed to higher (but not low or average) levels of childhood trauma. This was found in the context of morphological changes directly associated with increasing levels of schizotypy or increasing levels of childhood trauma exposure. CONCLUSIONS: These results suggest that alterations in brain regions critical for higher cognitive and integrative processes that are associated with schizotypy may be enhanced in individuals exposed to high levels of trauma.

The Influence of Brain MRI Defacing Algorithms on Brain-Age Predictions via 3D Convolutional Neural Networks.

In brain imaging research, it is becoming standard practice to remove the face from the individual's 3D structural MRI scan to ensure data privacy standards are met. Face removal - or 'defacing' - is being advocated for large, multi-site studies where data is transferred across geographically diverse sites. Several methods have been developed to limit the loss of important brain data by accurately and precisely removing non-brain facial tissue. At the same time, deep learning methods such as convolutional neural networks (CNNs) are increasingly being used in medical imaging research for diagnostic classification and prognosis in neurological diseases. These neural networks train predictive models based on patterns in large numbers of images. Because of this, defacing scans could remove informative data. Here, we evaluated 4 popular defacing methods to identify the effects of defacing on 'brain age' prediction - a common benchmarking task of predicting a subject's chronological age from their 3D T1-weighted brain MRI. We compared brain-age calculations using defaced MRIs to those that were directly brain extracted, and those with both brain and face. Significant differences were present when comparing average per-subject error rates between algorithms in both the defaced brain data and the extracted facial tissue. Results also indicated brain age accuracy depends on defacing and the choice of algorithm. In a secondary analysis, we also examined how well comparable CNNs could predict chronological age from the facial region only (the extracted portion of the defaced image), as well as visualize areas of importance in facial tissue for predictive tasks using CNNs. We obtained better performance in age prediction when using the extracted face portion alone than images of the brain, suggesting the need for caution when defacing methods are used in medical image analysis.

Cortical microstructural associations with CSF amyloid and pTau.

Diffusion MRI (dMRI) can be used to probe microstructural properties of brain tissue and holds great promise as a means to non-invasively map Alzheimer's disease (AD) pathology. Few studies have evaluated multi-shell dMRI models such as neurite orientation dispersion and density imaging (NODDI) and mean apparent propagator (MAP)-MRI in cortical gray matter where many of the earliest histopathological changes occur in AD. Here, we investigated the relationship between CSF pTau181 and Aß1-42 burden and regional cortical NODDI and MAP-MRI indices in 46 cognitively unimpaired individuals, 18 with mild cognitive impairment, and two with dementia (mean age: 71.8 ± 6.2 years) from the Alzheimer's Disease Neuroimaging Initiative. We compared findings to more conventional cortical thickness measures. Lower CSF Aß1-42 and higher pTau181 were associated with cortical dMRI measures reflecting less hindered or restricted diffusion and greater diffusivity. Cortical dMRI measures, but not cortical thickness measures, were more widely associated with Aß1-42 than pTau181 and better distinguished Aß+ from Aß- participants than pTau+ from pTau- participants. dMRI associations mediated the relationship between CSF markers and delayed logical memory performance, commonly impaired in early AD. dMRI metrics sensitive to early AD pathogenesis and microstructural damage may be better measures of subtle neurodegeneration in comparison to standard cortical thickness and help to elucidate mechanisms underlying cognitive decline.

Predictive Modeling Of Alzheimer's Disease Prognosis Using Anatomical & Diffusion MRI.

Mild cognitive impairment (MCI) is an intermediate stage between healthy aging and Alzheimer's disease (AD), and AD is a progressive neurodegenerative disorder that affects around 50 million people worldwide. As new AD treatments begin to be developed, one key goal of AD research is to predict which individuals with MCI are most likely to progress to AD over a given interval (such as 2 years); if successful, these individuals could be preferentially enrolled in drug trials that aim to slow AD progression. Here we benchmarked a range of MCI-to-AD predictive models including linear regressions, support vector machines, and random forests, using predictors from anatomical and diffusion-weighted brain MRI, age, sex, APOE genotype and standardized clinical scores. In evaluations on 2,448 subjects (1,132 MCI, 883 healthy controls, 433 with dementia) from the ADNI study, models including PCA-compacted features achieved a balanced accuracy of 75.3% (using cortical features) and 89.7% using diffusion MRI measures on test set, suggesting the added prognostic value of microstructural metrics obtainable with diffusion MRI.

Efficiently Training Vision Transformers on Structural MRI Scans for Alzheimer's Disease Detection.

Neuroimaging of large populations is valuable to identify factors that promote or resist brain disease, and to assist diagnosis, subtyping, and prognosis. Data-driven models such as convolutional neural networks (CNNs) have increasingly been applied to brain images to perform diagnostic and prognostic tasks by learning robust features. Vision transformers (ViT) - a new class of deep learning architectures - have emerged in recent years as an alternative to CNNs for several computer vision applications. Here we tested variants of the ViT architecture for a range of desired neuroimaging downstream tasks based on difficulty, in this case for sex and Alzheimer's disease (AD) classification based on 3D brain MRI. In our experiments, two vision transformer architecture variants achieved an AUC of 0.987 for sex and 0.892 for AD classification, respectively. We independently evaluated our models on data from two benchmark AD datasets. We achieved a performance boost of 5% and 9-10% upon fine-tuning vision transformer models pre-trained on synthetic (generated by a latent diffusion model) and real MRI scans, respectively. Our main contributions include testing the effects of different ViT training strategies including pre-training, data augmentation and learning rate warm-ups followed by annealing, as pertaining to the neuroimaging domain. These techniques are essential for training ViT-like models for neuroimaging applications where training data is usually limited. We also analyzed the effect of the amount of training data utilized on the test-time performance of the ViT via data-model scaling curves.Clinical Relevance- The models evaluated in this work could be trained on neuroimaging data to assist in diagnosis, subtyping and prognosis of Alzheimer's disease.

Variational Autoencoders for Generating Synthetic Tractography-Based Bundle Templates in a Low-Data Setting.

White matter tracts generated from whole brain tractography are often processed using automatic segmentation methods with standard atlases. Atlases are generated from hundreds of subjects, which becomes time-consuming to create and difficult to apply to all populations. In this study, we extended our prior work on using a deep generative model - a Convolutional Variational Autoencoder - to map complex and data-intensive streamlines to a low-dimensional latent space given a limited sample size of 50 subjects from the ADNI3 dataset, to generate synthetic population-specific bundle templates using Kernel Density Estimation (KDE) on streamline embeddings. We conducted a quantitative shape analysis by calculating bundle shape metrics, and found that our bundle templates better capture the shape distribution of the bundles than the atlas data used in the original segmentation derived from young healthy adults. We further demonstrated the use of our framework for direct bundle segmentation from whole-brain tractograms.

BundleCleaner: Unsupervised Denoising and Subsampling of Diffusion MRI-Derived Tractography Data.

We present BundleCleaner, an unsupervised multi-step framework that can filter, denoise and subsample bundles derived from diffusion MRI-based whole-brain tractography. Our approach considers both the global bundle structure and local streamline-wise features. We apply BundleCleaner to bundles generated from single-shell diffusion MRI data in an independent clinical sample of older adults from India using probabilistic tractography and the resulting 'cleaned' bundles can better align with the atlas bundles with reduced overreach. In a downstream tractometry analysis, we show that the cleaned bundles, represented with less than 20% of the original set of points, can robustly localize along-tract microstructural differences between 32 healthy controls and 34 participants with Alzheimer's disease ranging in age from 55 to 84 years old. Our approach can help reduce memory burden and improving computational efficiency when working with tractography data, and shows promise for large-scale multi-site tractometry.

Data-driven biomarkers outperform theory-based biomarkers in predicting stroke motor outcomes.

Chronic motor impairments are a leading cause of disability after stroke. Previous studies have predicted motor outcomes based on the degree of damage to predefined structures in the motor system, such as the corticospinal tract. However, such theory-based approaches may not take full advantage of the information contained in clinical imaging data. The present study uses data-driven approaches to predict chronic motor outcomes after stroke and compares the accuracy of these predictions to previously-identified theory-based biomarkers. Using a cross-validation framework, regression models were trained using lesion masks and motor outcomes data from 789 stroke patients (293 female/496 male) from the ENIGMA Stroke Recovery Working Group (age 64.9±18.0 years; time since stroke 12.2±0.2 months; normalised motor score 0.7±0.5 (range [0,1]). The out-of-sample prediction accuracy of two theory-based biomarkers was assessed: lesion load of the corticospinal tract, and lesion load of multiple descending motor tracts. These theory-based prediction accuracies were compared to the prediction accuracy from three data-driven biomarkers: lesion load of lesion-behaviour maps, lesion load of structural networks associated with lesion-behaviour maps, and measures of regional structural disconnection. In general, data-driven biomarkers had better prediction accuracy - as measured by higher explained variance in chronic motor outcomes - than theory-based biomarkers. Data-driven models of regional structural disconnection performed the best of all models tested (R2 = 0.210, p < 0.001), performing significantly better than predictions using the theory-based biomarkers of lesion load of the corticospinal tract (R2 = 0.132, p< 0.001) and of multiple descending motor tracts (R2 = 0.180, p < 0.001). They also performed slightly, but significantly, better than other data-driven biomarkers including lesion load of lesion-behaviour maps (R2 =0.200, p < 0.001) and lesion load of structural networks associated with lesion-behaviour maps (R2 =0.167, p < 0.001). Ensemble models - combining basic demographic variables like age, sex, and time since stroke - improved prediction accuracy for theory-based and data-driven biomarkers. Finally, combining both theory-based and data-driven biomarkers with demographic variables improved predictions, and the best ensemble model achieved R2 = 0.241, p < 0.001. Overall, these results demonstrate that models that predict chronic motor outcomes using data-driven features, particularly when lesion data is represented in terms of structural disconnection, perform better than models that predict chronic motor outcomes using theory-based features from the motor system. However, combining both theory-based and data-driven models provides the best predictions.

Cortical morphology in patients with the deficit and non-deficit syndrome of schizophrenia: a worldwide meta- and mega-analyses.

Converging evidence suggests that schizophrenia (SZ) with primary, enduring negative symptoms (i.e., Deficit SZ (DSZ)) represents a distinct entity within the SZ spectrum while the neurobiological underpinnings remain undetermined. In the largest dataset of DSZ and Non-Deficit (NDSZ), we conducted a meta-analysis of data from 1560 individuals (168 DSZ, 373 NDSZ, 1019 Healthy Controls (HC)) and a mega-analysis of a subsampled data from 944 individuals (115 DSZ, 254 NDSZ, 575 HC) collected across 9 worldwide research centers of the ENIGMA SZ Working Group (8 in the mega-analysis), to clarify whether they differ in terms of cortical morphology. In the meta-analysis, sites computed effect sizes for differences in cortical thickness and surface area between SZ and control groups using a harmonized pipeline. In the mega-analysis, cortical values of individuals with schizophrenia and control participants were analyzed across sites using mixed-model ANCOVAs. The meta-analysis of cortical thickness showed a converging pattern of widespread thinner cortex in fronto-parietal regions of the left hemisphere in both DSZ and NDSZ, when compared to HC. However, DSZ have more pronounced thickness abnormalities than NDSZ, mostly involving the right fronto-parietal cortices. As for surface area, NDSZ showed differences in fronto-parietal-temporo-occipital cortices as compared to HC, and in temporo-occipital cortices as compared to DSZ. Although DSZ and NDSZ show widespread overlapping regions of thinner cortex as compared to HC, cortical thinning seems to better typify DSZ, being more extensive and bilateral, while surface area alterations are more evident in NDSZ. Our findings demonstrate for the first time that DSZ and NDSZ are characterized by different neuroimaging phenotypes, supporting a nosological distinction between DSZ and NDSZ and point toward the separate disease hypothesis.

Preliminary validation of a structural magnetic resonance imaging metric for tracking dementia-related neurodegeneration and future decline.

Alzheimer's disease (AD) is a progressive neurodegenerative disease characterized by cognitive decline and atrophy in the medial temporal lobe (MTL) and subsequent brain regions. Structural magnetic resonance imaging (sMRI) has been widely used in research and clinical care for diagnosis and monitoring AD progression. However, atrophy patterns are complex and vary by patient. To address this issue, researchers have made efforts to develop more concise metrics that can summarize AD-specific atrophy. Many of these methods can be difficult to interpret clinically, hampering adoption. In this study, we introduce a novel index which we call an "AD-NeuroScore," that uses a modified Euclidean-inspired distance function to calculate differences between regional brain volumes associated with cognitive decline. The index is adjusted for intracranial volume (ICV), age, sex, and scanner model. We validated AD-NeuroScore using 929 older adults from the Alzheimer's Disease Neuroimaging Initiative (ADNI) study, with a mean age of 72.7 years (SD = 6.3; 55.1-91.5) and cognitively normal (CN), mild cognitive impairment (MCI), or AD diagnoses. Our validation results showed that AD-NeuroScore was significantly associated with diagnosis and disease severity scores (measured by MMSE, CDR-SB, and ADAS-11) at baseline. Furthermore, baseline AD-NeuroScore was associated with both changes in diagnosis and disease severity scores at all time points with available data. The performance of AD-NeuroScore was equivalent or superior to adjusted hippocampal volume (AHV), a widely used metric in AD research. Further, AD-NeuroScore typically performed as well as or sometimes better when compared to other existing sMRI-based metrics. In conclusion, we have introduced a new metric, AD-NeuroScore, which shows promising results in detecting AD, benchmarking disease severity, and predicting disease progression. AD-NeuroScore differentiates itself from other metrics by being clinically practical and interpretable.

Style transfer generative adversarial networks to harmonize multisite MRI to a single reference image to avoid overcorrection.

Recent work within neuroimaging consortia have aimed to identify reproducible, and often subtle, brain signatures of psychiatric or neurological conditions. To allow for high-powered brain imaging analyses, it is often necessary to pool MR images that were acquired with different protocols across multiple scanners. Current retrospective harmonization techniques have shown promise in removing site-related image variation. However, most statistical approaches may over-correct for technical, scanning-related, variation as they cannot distinguish between confounded image-acquisition based variability and site-related population variability. Such statistical methods often require that datasets contain subjects or patient groups with similar clinical or demographic information to isolate the acquisition-based variability. To overcome this limitation, we consider site-related magnetic resonance (MR) imaging harmonization as a style transfer problem rather than a domain transfer problem. Using a fully unsupervised deep-learning framework based on a generative adversarial network (GAN), we show that MR images can be harmonized by inserting the style information encoded from a single reference image, without knowing their site/scanner labels a priori. We trained our model using data from five large-scale multisite datasets with varied demographics. Results demonstrated that our style-encoding model can harmonize MR images, and match intensity profiles, without relying on traveling subjects. This model also avoids the need to control for clinical, diagnostic, or demographic information. We highlight the effectiveness of our method for clinical research by comparing extracted cortical and subcortical features, brain-age estimates, and case-control effect sizes before and after the harmonization. We showed that our harmonization removed the site-related variances, while preserving the anatomical information and clinical meaningful patterns. We further demonstrated that with a diverse training set, our method successfully harmonized MR images collected from unseen scanners and protocols, suggesting a promising tool for ongoing collaborative studies. Source code is released in USC-IGC/style_transfer_harmonization (github.com).

The Influence of Brain MRI Defacing Algorithms on Brain-Age Predictions via 3D Convolutional Neural Networks.

In brain imaging research, it is becoming standard practice to remove the face from the individual's 3D structural MRI scan to ensure data privacy standards are met. Face removal - or 'defacing' - is being advocated for large, multi-site studies where data is transferred across geographically diverse sites. Several methods have been developed to limit the loss of important brain data by accurately and precisely removing non-brain facial tissue. At the same time, deep learning methods such as convolutional neural networks (CNNs) are increasingly being used in medical imaging research for diagnostic classification and prognosis in neurological diseases. These neural networks train predictive models based on patterns in large numbers of images. Because of this, defacing scans could remove informative data. Here, we evaluated 4 popular defacing methods to identify the effects of defacing on 'brain age' prediction - a common benchmarking task of predicting a subject's chronological age from their 3D T1-weighted brain MRI. We compared brain-age calculations using defaced MRIs to those that were directly brain extracted, and those with both brain and face. Significant differences were present when comparing average per-subject error rates between algorithms in both the defaced brain data and the extracted facial tissue. Results also indicated brain age accuracy depends on defacing and the choice of algorithm. In a secondary analysis, we also examined how well comparable CNNs could predict chronological age from the facial region only (the extracted portion of the defaced image), as well as visualize areas of importance in facial tissue for predictive tasks using CNNs. We obtained better performance in age prediction when using the extracted face portion alone than images of the brain, suggesting the need for caution when defacing methods are used in medical image analysis.