Clinica: An Open-Source Software Platform for Reproducible Clinical Neuroscience Studies

Transcranial Blood-Brain Barrier Opening in Alzheimer's Disease Patients Using a Portable Focused Ultrasound System with Real-Time 2-D Cavitation Mapping

Background

Focused ultrasound (FUS) in combination with microbubbles has recently shown great promise in facilitating blood-brain barrier (BBB) opening for drug delivery and immunotherapy in Alzheimer's disease (AD). However, it is currently limited to systems integrated within the MRI suites or requiring post-surgical implants, thus restricting its widespread clinical adoption. In this pilot study, we investigate the clinical safety and feasibility of a portable, non-invasive neuronavigation-guided FUS (NgFUS) system with integrated real-time 2-D microbubble cavitation mapping.

Methods

A phase 1 clinical study with mild to moderate AD patients (N=6) underwent a single session of microbubble-mediated NgFUS to induce transient BBB opening (BBBO). Microbubble activity under FUS was monitored with real-time 2-D cavitation maps and dosing to ensure the efficacy and safety of the NgFUS treatment. Post-operative MRI was used for BBB opening and closure confirmation as well as safety assessment. Changes in AD biomarker levels in both blood serum and extracellular vesicles (EVs) were evaluated, while changes in amyloid-beta (Aβ) load in the brain were assessed through 18F-Florbetapir PET.

Results

BBBO was achieved in 5 out of 6 subjects with an average volume of 983±626 mm3 following FUS at the right frontal lobe both in white and gray matter regions. The outpatient treatment was completed within 34.8±10.7 min. Cavitation dose significantly correlated with the BBBO volume (R2>0.9, N=4), demonstrating the portable NgFUS system's capability of predicting opening volumes. The cavitation maps co-localized closely with the BBBO location, representing the first report of real-time transcranial 2-D cavitation mapping in the human brain. Larger opening volumes correlated with increased levels of AD biomarkers, including Aβ42 (R2=0.74), Tau (R2=0.95), and P-Tau181 (R2=0.86), assayed in serum-derived EVs sampled 3 days after FUS (N=5). From PET scans, subjects showed a lower Aβ load increase in the treated frontal lobe region compared to the contralateral region. Reduction in asymmetry standardized uptake value ratios (SUVR) correlated with the cavitation dose (R2>0.9, N=3). Clinical changes in the mini-mental state examination over 6 months were within the expected range of cognitive decline with no additional changes observed as a result of FUS.

Conclusion

We showed the safety and feasibility of this cost-effective and time-efficient portable NgFUS treatment for BBBO in AD patients with the first demonstration of real-time 2-D cavitation mapping. The cavitation dose correlated with BBBO volume, a slowed increase in pathology, and serum detection of AD proteins. Our study highlights the potential for accessible FUS treatment in AD, with or without drug delivery.

DeepFDR: A Deep Learning-based False Discovery Rate Control Method for Neuroimaging Data

Voxel-based multiple testing is widely used in neuroimaging data analysis. Traditional false discovery rate (FDR) control methods often ignore the spatial dependence among the voxel-based tests and thus suffer from substantial loss of testing power. While recent spatial FDR control methods have emerged, their validity and optimality remain questionable when handling the complex spatial dependencies of the brain. Concurrently, deep learning methods have revolutionized image segmentation, a task closely related to voxel-based multiple testing. In this paper, we propose DeepFDR, a novel spatial FDR control method that leverages unsupervised deep learning-based image segmentation to address the voxel-based multiple testing problem. Numerical studies, including comprehensive simulations and Alzheimer's disease FDG-PET image analysis, demonstrate DeepFDR's superiority over existing methods. DeepFDR not only excels in FDR control and effectively diminishes the false nondiscovery rate, but also boasts exceptional computational efficiency highly suited for tackling large-scale neuroimaging data.

Neurodesk: an accessible, flexible and portable data analysis environment for reproducible neuroimaging

Neuroimaging research requires purpose-built analysis software, which is challenging to install and may produce different results across computing environments. The community-oriented, open-source Neurodesk platform ( https://www.neurodesk.org/ ) harnesses a comprehensive and growing suite of neuroimaging software containers. Neurodesk includes a browser-accessible virtual desktop, command-line interface and computational notebook compatibility, allowing for accessible, flexible, portable and fully reproducible neuroimaging analysis on personal workstations, high-performance computers and the cloud.

Automatic motion artefact detection in brain T1-weighted magnetic resonance images from a clinical data warehouse using synthetic data

Containing the medical data of millions of patients, clinical data warehouses (CDWs) represent a great opportunity to develop computational tools. Magnetic resonance images (MRIs) are particularly sensitive to patient movements during image acquisition, which will result in artefacts (blurring, ghosting and ringing) in the reconstructed image. As a result, a significant number of MRIs in CDWs are corrupted by these artefacts and may be unusable. Since their manual detection is impossible due to the large number of scans, it is necessary to develop tools to automatically exclude (or at least identify) images with motion in order to fully exploit CDWs. In this paper, we propose a novel transfer learning method from research to clinical data for the automatic detection of motion in 3D T1-weighted brain MRI. The method consists of two steps: a pre-training on research data using synthetic motion, followed by a fine-tuning step to generalise our pre-trained model to clinical data, relying on the labelling of 4045 images. The objectives were both (1) to be able to exclude images with severe motion, (2) to detect mild motion artefacts. Our approach achieved excellent accuracy for the first objective with a balanced accuracy nearly similar to that of the annotators (balanced accuracy>80 %). However, for the second objective, the performance was weaker and substantially lower than that of human raters. Overall, our framework will be useful to take advantage of CDWs in medical imaging and highlight the importance of a clinical validation of models trained on research data.

Contrast-enhanced to non-contrast-enhanced image translation to exploit a clinical data warehouse of T1-weighted brain MRI

BACKGROUND

Clinical data warehouses provide access to massive amounts of medical images, but these images are often heterogeneous. They can for instance include images acquired both with or without the injection of a gadolinium-based contrast agent. Harmonizing such data sets is thus fundamental to guarantee unbiased results, for example when performing differential diagnosis. Furthermore, classical neuroimaging software tools for feature extraction are typically applied only to images without gadolinium. The objective of this work is to evaluate how image translation can be useful to exploit a highly heterogeneous data set containing both contrast-enhanced and non-contrast-enhanced images from a clinical data warehouse.

METHODS

We propose and compare different 3D U-Net and conditional GAN models to convert contrast-enhanced T1-weighted (T1ce) into non-contrast-enhanced (T1nce) brain MRI. These models were trained using 230 image pairs and tested on 77 image pairs from the clinical data warehouse of the Greater Paris area.

RESULTS

Validation using standard image similarity measures demonstrated that the similarity between real and synthetic T1nce images was higher than between real T1nce and T1ce images for all the models compared. The best performing models were further validated on a segmentation task. We showed that tissue volumes extracted from synthetic T1nce images were closer to those of real T1nce images than volumes extracted from T1ce images.

CONCLUSION

We showed that deep learning models initially developed with research quality data could synthesize T1nce from T1ce images of clinical quality and that reliable features could be extracted from the synthetic images, thus demonstrating the ability of such methods to help exploit a data set coming from a clinical data warehouse.

Subtypes of brain change in aging and their associations with cognition and Alzheimer's disease biomarkers

Structural brain changes underly cognitive changes in older age and contribute to inter-individual variability in cognition. Here, we assessed how changes in cortical thickness, surface area, and subcortical volume, are related to cognitive change in cognitively unimpaired older adults using structural magnetic resonance imaging (MRI) data-driven clustering. Specifically, we tested (1) which brain structural changes over time predict cognitive change in older age (2) whether these are associated with core cerebrospinal fluid (CSF) Alzheimer's disease (AD) biomarkers phosphorylated tau (p-tau) and amyloid-β (Aβ42), and (3) the degree of overlap between clusters derived from different structural features. In total 1899 cognitively healthy older adults (50 - 93 years) were followed up to 16 years with neuropsychological and structural MRI assessments, a subsample of which (n = 612) had CSF p-tau and Aβ42 measurements. We applied Monte-Carlo Reference-based Consensus clustering to identify subgroups of older adults based on structural brain change patterns over time. Four clusters for each brain feature were identified, representing the degree of longitudinal brain decline. Each brain feature provided a unique contribution to brain aging as clusters were largely independent across modalities. Cognitive change and baseline cognition were best predicted by cortical area change, whereas higher levels of p-tau and Aβ42 were associated with changes in subcortical volume. These results provide insights into the link between changes in brain morphology and cognition, which may translate to a better understanding of different aging trajectories.

Coarse-to-fine visual representation learning for medical images via class activation maps

The value of coarsely labeled datasets in learning transferable representations for medical images is investigated in this work. Compared to fine labels which require meticulous effort to annotate, coarse labels can be acquired at a significantly lower cost and can provide useful training signals for data-hungry deep neural networks. We consider coarse labels in the form of binary labels differentiating a normal (healthy) image from an abnormal (diseased) image and propose CAMContrast, a two-stage representation learning framework for medical images. Using class activation maps, CAMContrast makes use of the binary labels to generate heatmaps as positive views for contrastive representation learning. Specifically, the learning objective is optimized to maximize the agreement within fixed crops of image-heatmap pair to learn fine-grained representations that are generalizable to different downstream tasks. We empirically validate the transfer learning performance of CAMContrast on several public datasets, covering classification and segmentation tasks on fundus photographs and chest X-ray images. The experimental results showed that our method outperforms other self-supervised and supervised pretrain methods in terms of data efficiency and downstream performance.

Disentangling Accelerated Cognitive Decline from the Normal Aging Process and Unraveling Its Genetic Components: A Neuroimaging-Based Deep Learning Approach

Background

The progressive cognitive decline, an integral component of Alzheimer's disease (AD), unfolds in tandem with the natural aging process. Neuroimaging features have demonstrated the capacity to distinguish cognitive decline changes stemming from typical brain aging and AD between different chronological points.

Objective

To disentangle the normal aging effect from the AD-related accelerated cognitive decline and unravel its genetic components using a neuroimaging-based deep learning approach.

Methods

We developed a deep-learning framework based on a dual-loss Siamese ResNet network to extract fine-grained information from the longitudinal structural magnetic resonance imaging (MRI) data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) study. We then conducted genome-wide association studies (GWAS) and post-GWAS analyses to reveal the genetic basis of AD-related accelerated cognitive decline.

Results

We used our model to process data from 1,313 individuals, training it on 414 cognitively normal people and predicting cognitive assessment for all participants. In our analysis of accelerated cognitive decline GWAS, we identified two genome-wide significant loci: APOE locus (chromosome 19 p13.32) and rs144614292 (chromosome 11 p15.1). Variant rs144614292 (G > T) has not been reported in previous AD GWA studies. It is within the intronic region of NELL1, which is expressed in neurons and plays a role in controlling cell growth and differentiation. The cell-type-specific enrichment analysis and functional enrichment of GWAS signals highlighted the microglia and immune-response pathways.

Conclusions

Our deep learning model effectively extracted relevant neuroimaging features and predicted individual cognitive decline. We reported a novel variant (rs144614292) within the NELL1 gene.

Genomic loci influence patterns of structural covariance in the human brain

Normal and pathologic neurobiological processes influence brain morphology in coordinated ways that give rise to patterns of structural covariance (PSC) across brain regions and individuals during brain aging and diseases. The genetic underpinnings of these patterns remain largely unknown. We apply a stochastic multivariate factorization method to a diverse population of 50,699 individuals (12 studies and 130 sites) and derive data-driven, multi-scale PSCs of regional brain size. PSCs were significantly correlated with 915 genomic loci in the discovery set, 617 of which are newly identified, and 72% were independently replicated. Key pathways influencing PSCs involve reelin signaling, apoptosis, neurogenesis, and appendage development, while pathways of breast cancer indicate potential interplays between brain metastasis and PSCs associated with neurodegeneration and dementia. Using support vector machines, multi-scale PSCs effectively derive imaging signatures of several brain diseases. Our results elucidate genetic and biological underpinnings that influence structural covariance patterns in the human brain.

Hippocampal metabolic subregions and networks: Behavioral, molecular, and pathological aging profiles

INTRODUCTION

Hippocampal local and network dysfunction is the hallmark of Alzheimer's disease (AD).

METHODS

We characterized the spatial patterns of hippocampus differentiation based on brain co-metabolism in healthy elderly participants and demonstrated their relevance to study local metabolic changes and associated dysfunction in pathological aging.

RESULTS

The hippocampus can be differentiated into anterior/posterior and dorsal cornu ammonis (CA)/ventral (subiculum) subregions. While anterior/posterior CA show co-metabolism with different regions of the subcortical limbic networks, the anterior/posterior subiculum are parts of cortical networks supporting object-centered memory and higher cognitive demands, respectively. Both networks show relationships with the spatial patterns of gene expression pertaining to cell energy metabolism and AD's process. Finally, while local metabolism is generally lower in posterior regions, the anterior-posterior imbalance is maximal in late mild cognitive impairment with the anterior subiculum being relatively preserved.

DISCUSSION

Future studies should consider bidimensional hippocampal differentiation and in particular the posterior subicular region to better understand pathological aging.

Associations of neuroinflammatory IL-6 and IL-8 with brain atrophy, memory decline, and core AD biomarkers - in cognitively unimpaired older adults

Concentrations of pro-inflammatory cytokines -interleukin-6 (IL-6) and interleukin-8 (IL-8) - are increased with age and in Alzheimer's disease (AD). It is not clear whether concentrations of IL-6 and IL-8 in the central nervous system predict later brain and cognitive changes over time nor whether this relationship is mediated by core AD biomarkers. Here, 219 cognitively healthy older adults (62-91 years), with baseline cerebrospinal fluid (CSF) measures of IL-6 and IL-8 were followed over time - up to 9 years - with assessments that included cognitive function, structural magnetic resonance imaging, and CSF measurements of phosphorylated tau (p-tau) and amyloid-β (Aβ-42) concentrations (for a subsample). Higher baseline CSF IL-8 was associated with better memory performance over time in the context of lower levels of CSF p-tau and p-tau/Aβ-42 ratio. Higher CSF IL-6 was related to less CSF p-tau changes over time. The results are in line with the hypothesis suggesting that an up-regulation of IL-6 and IL-8 in the brain may play a neuroprotective role in cognitively healthy older adults with lower load of AD pathology.

Regional Glymphatic Abnormality in Behavioral Variant Frontotemporal Dementia

OBJECTIVES

Glymphatic function has not yet been explored in behavioral variant frontotemporal dementia (bvFTD). The spatial correlation between regional glymphatic function and bvFTD remains unknown.

METHOD

A total of 74 patients with bvFTD and 67 age- and sex-matched healthy controls (HCs) were selected from discovery dataset and replication dataset. All participants underwent neuropsychological assessment. Glymphatic measures including choroid plexus (CP) volume, diffusion tensor imaging along the perivascular (DTI-ALPS) index, and coupling between blood-oxygen-level-dependent signals and cerebrospinal fluid signals (BOLD-CSF coupling), were compared between the two groups. Regional glymphatic function was evaluated by dividing DTI-ALPS and BOLD-CSF coupling into anterior, middle, and posterior regions. The bvFTD-related metabolic pattern was identified using spatial covariance analysis based on l8 F-FDG-PET.

RESULTS

Patients with bvFTD showed higher CP volume (p < 0.001); anterior and middle DTI-ALPS (p < 0.001); and weaker anterior BOLD-CSF coupling (p < 0.05) than HCs after controlling for cortical gray matter volume in both datasets. In bvFTD from the discovery dataset, the anterior DTI-ALPS was negatively associated with the expression of the bvFTD-related metabolic pattern (r = -0.52, p = 0.034) and positively related with regional standardized uptake value ratios of l8 F-FDG-PET in bvFTD-related brain regions (r range: 0.49 to 0.62, p range: 0.017 to 0.047). Anterior and middle glymphatic functions were related to global cognition and disease severity.

INTERPRETATION

Our findings reveal abnormal glymphatic function, especially in the anterior and middle regions of brain in bvFTD. Regional glymphatic dysfunction may contribute to the pathogenesis of bvFTD. ANN NEUROL 2023;94:442-456.

Multi-modal graph neural network for early diagnosis of Alzheimer's disease from sMRI and PET scans

In recent years, deep learning models have been applied to neuroimaging data for early diagnosis of Alzheimer's disease (AD). Structural magnetic resonance imaging (sMRI) and positron emission tomography (PET) images provide structural and functional information about the brain, respectively. Combining these features leads to improved performance than using a single modality alone in building predictive models for AD diagnosis. However, current multi-modal approaches in deep learning, based on sMRI and PET, are mostly limited to convolutional neural networks, which do not facilitate integration of both image and phenotypic information of subjects. We propose to use graph neural networks (GNN) that are designed to deal with problems in non-Euclidean domains. In this study, we demonstrate how brain networks are created from sMRI or PET images and can be used in a population graph framework that combines phenotypic information with imaging features of the brain networks. Then, we present a multi-modal GNN framework where each modality has its own branch of GNN and a technique that combines the multi-modal data at both the level of node vectors and adjacency matrices. Finally, we perform late fusion to combine the preliminary decisions made in each branch and produce a final prediction. As multi-modality data becomes available, multi-source and multi-modal is the trend of AD diagnosis. We conducted explorative experiments based on multi-modal imaging data combined with non-imaging phenotypic information for AD diagnosis and analyzed the impact of phenotypic information on diagnostic performance. Results from experiments demonstrated that our proposed multi-modal approach improves performance for AD diagnosis. Our study also provides technical reference and support the need for multivariate multi-modal diagnosis methods.

Longitudinal changes in hippocampal texture from healthy aging to Alzheimer's disease

Early detection of Alzheimer's disease is essential to develop preventive treatment strategies. Detectible change in brain volume emerges relatively late in the pathogenic progression of disease, but microstructural changes caused by early neuropathology may cause subtle changes in the MR signal, quantifiable using texture analysis. Texture analysis quantifies spatial patterns in an image, such as smoothness, randomness and heterogeneity. We investigated whether the MRI texture of the hippocampus, an early site of Alzheimer's disease pathology, is sensitive to changes in brain microstructure before the onset of cognitive impairment. We also explored the longitudinal trajectories of hippocampal texture across the Alzheimer's continuum in relation to hippocampal volume and other biomarkers. Finally, we assessed the ability of texture to predict future cognitive decline, over and above hippocampal volume. Data were acquired from the Alzheimer's Disease Neuroimaging Initiative. Texture was calculated for bilateral hippocampi on 3T T1-weighted MRI scans. Two hundred and ninety-three texture features were reduced to five principal components that described 88% of total variance within cognitively unimpaired participants. We assessed cross-sectional differences in these texture components and hippocampal volume between four diagnostic groups: cognitively unimpaired amyloid-β- (n = 406); cognitively unimpaired amyloid-β+ (n = 213); mild cognitive impairment amyloid-β+ (n = 347); and Alzheimer's disease dementia amyloid-β+ (n = 202). To assess longitudinal texture change across the Alzheimer's continuum, we used a multivariate mixed-effects spline model to calculate a 'disease time' for all timepoints based on amyloid PET and cognitive scores. This was used as a scale on which to compare the trajectories of biomarkers, including volume and texture of the hippocampus. The trajectories were modelled in a subset of the data: cognitively unimpaired amyloid-β- (n = 345); cognitively unimpaired amyloid-β+ (n = 173); mild cognitive impairment amyloid-β+ (n = 301); and Alzheimer's disease dementia amyloid-β+ (n = 161). We identified a difference in texture component 4 at the earliest stage of Alzheimer's disease, between cognitively unimpaired amyloid-β- and cognitively unimpaired amyloid-β+ older adults (Cohen's d = 0.23, Padj = 0.014). Differences in additional texture components and hippocampal volume emerged later in the disease continuum alongside the onset of cognitive impairment (d = 0.30-1.22, Padj < 0.002). Longitudinal modelling of the texture trajectories revealed that, while most elements of texture developed over the course of the disease, noise reduced sensitivity for tracking individual textural change over time. Critically, however, texture provided additional information than was provided by volume alone to more accurately predict future cognitive change (d = 0.32-0.63, Padj < 0.0001). Our results support the use of texture as a measure of brain health, sensitive to Alzheimer's disease pathology, at a time when therapeutic intervention may be most effective.

Temporo-basal sulcal connections: a manual annotation protocol and an investigation of sexual dimorphism and heritability

The temporo-basal region of the human brain is composed of the collateral, the occipito-temporal, and the rhinal sulci. We manually rated (using a novel protocol) the connections between rhinal/collateral (RS-CS), collateral/occipito-temporal (CS-OTS) and rhinal/occipito-temporal (RS-OTS) sulci, using the MRI of nearly 3400 individuals including around 1000 twins. We reported both the associations between sulcal polymorphisms as well with a wide range of demographics (e.g. age, sex, handedness). Finally, we also estimated the heritability, and the genetic correlation between sulcal connections. We reported the frequency of the sulcal connections in the general population, which were hemisphere dependent. We found a sexual dimorphism of the connections, especially marked in the right hemisphere, with a CS-OTS connection more frequent in females (approximately 35-40% versus 20-25% in males) and an RS-CS connection more common in males (approximately 40-45% versus 25-30% in females). We confirmed associations between sulcal connections and characteristics of incomplete hippocampal inversion (IHI). We estimated the broad sense heritability to be 0.28-0.45 for RS-CS and CS-OTS connections, with hints of dominant contribution for the RS-CS connection. The connections appeared to share some of their genetic causing factors as indicated by strong genetic correlations. Heritability appeared much smaller for the (rarer) RS-OTS connection.

Impact of sex and APOE-ε4 genotype on patterns of regional brain atrophy in Alzheimer's disease and healthy aging

Alzheimer's Disease (AD) is a heterogeneous disease that disproportionately affects women and people with the APOE-ε4 susceptibility gene. We aim to describe the not-well-understood influence of both risk factors on the dynamics of brain atrophy in AD and healthy aging. Regional cortical thinning and brain atrophy were modeled over time using non-linear mixed-effect models and the FreeSurfer software with t1-MRI scans from the Alzheimer's Disease Neuroimaging Initiative (N = 1,502 subjects, 6,728 images in total). Covariance analysis was used to disentangle the effect of sex and APOE genotype on the regional onset age and pace of atrophy, while correcting for educational level. A map of the regions mostly affected by neurodegeneration is provided. Results were confirmed on gray matter density data from the SPM software. Women experience faster atrophic rates in the temporal, frontal, parietal lobes and limbic system and earlier onset in the amygdalas, but slightly later onset in the postcentral and cingulate gyri as well as all regions of the basal ganglia and thalamus. APOE-ε4 genotypes leads to earlier and faster atrophy in the temporal, frontal, parietal lobes, and limbic system in AD patients, but not in healthy patients. Higher education was found to slightly delay atrophy in healthy patients, but not for AD patients. A cohort of amyloid positive patients with MCI showed a similar impact of sex as in the healthy cohort, while APOE-ε4 showed similar associations as in the AD cohort. Female sex is as strong a risk factor for AD as APOE-ε4 genotype regarding neurodegeneration. Women experience a sharper atrophy in the later stages of the disease, although not a significantly earlier onset. These findings may have important implications for the development of targeted intervention.

QuNex-An integrative platform for reproducible neuroimaging analytics

Introduction

Neuroimaging technology has experienced explosive growth and transformed the study of neural mechanisms across health and disease. However, given the diversity of sophisticated tools for handling neuroimaging data, the field faces challenges in method integration, particularly across multiple modalities and species. Specifically, researchers often have to rely on siloed approaches which limit reproducibility, with idiosyncratic data organization and limited software interoperability.

Methods

To address these challenges, we have developed Quantitative Neuroimaging Environment & Toolbox (QuNex), a platform for consistent end-to-end processing and analytics. QuNex provides several novel functionalities for neuroimaging analyses, including a "turnkey" command for the reproducible deployment of custom workflows, from onboarding raw data to generating analytic features.

Results

The platform enables interoperable integration of multi-modal, community-developed neuroimaging software through an extension framework with a software development kit (SDK) for seamless integration of community tools. Critically, it supports high-throughput, parallel processing in high-performance compute environments, either locally or in the cloud. Notably, QuNex has successfully processed over 10,000 scans across neuroimaging consortia, including multiple clinical datasets. Moreover, QuNex enables integration of human and non-human workflows via a cohesive translational platform.

Discussion

Collectively, this effort stands to significantly impact neuroimaging method integration across acquisition approaches, pipelines, datasets, computational environments, and species. Building on this platform will enable more rapid, scalable, and reproducible impact of neuroimaging technology across health and disease.

Interaction between spatial perception and temporal perception enables preservation of cause-effect relationship: Visual psychophysics and neuronal dynamics

INTRODUCTION

Visual perception of moving objects is integral to our day-to-day life, integrating visual spatial and temporal perception. Most research studies have focused on finding the brain regions activated during motion perception. However, an empirically validated general mathematical model is required to understand the modulation of the motion perception. Here, we develop a mathematical formulation of the modulation of the perception of a moving object due to a change in speed, under the formulation of the invariance of causality.

METHODS

We formulated the perception of a moving object as the coordinate transformation from a retinotopic space onto perceptual space and derived a quantitative relationship between spatiotemporal coordinates. To validate our model, we undertook the analysis of two experiments: (i) the perceived length of the moving arc, and (ii) the perceived time while observing moving stimuli. We performed a magnetic resonance imaging (MRI) tractography investigation of subjects to demarcate the anatomical correlation of the modulation of the perception of moving objects.

RESULTS

Our theoretical model shows that the interaction between visual-spatial and temporal perception, during the perception of moving object is described by coupled linear equations; and experimental observations validate our model. We observed that cerebral area V5 may be an anatomical correlate for this interaction. The physiological basis of interaction is shown by a Lotka-Volterra system delineating interplay between acetylcholine and dopamine neurotransmitters, whose concentrations vary periodically with the orthogonal phase shift between them, occurring at the axodendritic synapse of complex cells at area V5.

CONCLUSION

Under the invariance of causality in the representation of events in retinotopic space and perceptual space, the speed modulates the perception of a moving object. This modulation may be due to variations of the tuning properties of complex cells at area V5 due to the dynamic interaction between acetylcholine and dopamine. Our analysis is the first significant study, to our knowledge, that establishes a mathematical linkage between motion perception and causality invariance.

Neurodesk: An accessible, flexible, and portable data analysis environment for reproducible neuroimaging

Neuroimaging data analysis often requires purpose-built software, which can be challenging to install and may produce different results across computing environments. Beyond being a roadblock to neuroscientists, these issues of accessibility and portability can hamper the reproducibility of neuroimaging data analysis pipelines. Here, we introduce the Neurodesk platform, which harnesses software containers to support a comprehensive and growing suite of neuroimaging software (https://www.neurodesk.org/). Neurodesk includes a browser-accessible virtual desktop environment and a command line interface, mediating access to containerized neuroimaging software libraries on various computing platforms, including personal and high-performance computers, cloud computing and Jupyter Notebooks. This community-oriented, open-source platform enables a paradigm shift for neuroimaging data analysis, allowing for accessible, flexible, fully reproducible, and portable data analysis pipelines.

Automatic segmentation of the choroid plexuses: Method and validation in controls and patients with multiple sclerosis

Choroid Plexuses (ChP) are structures located in the ventricles that produce the cerebrospinal fluid (CSF) in the central nervous system. They are also a key component of the blood-CSF barrier. Recent studies have described clinically relevant ChP volumetric changes in several neurological diseases including Alzheimer's, Parkinson's disease, and multiple sclerosis (MS). Therefore, a reliable and automated tool for ChP segmentation on images derived from magnetic resonance imaging (MRI) is a crucial need for large studies attempting to elucidate their role in neurological disorders. Here, we propose a novel automatic method for ChP segmentation in large imaging datasets. The approach is based on a 2-step 3D U-Net to keep preprocessing steps to a minimum for ease of use and to lower memory requirements. The models are trained and validated on a first research cohort including people with MS and healthy subjects. A second validation is also performed on a cohort of pre-symptomatic MS patients having acquired MRIs in routine clinical practice. Our method reaches an average Dice coefficient of 0.72 ± 0.01 with the ground truth and a volume correlation of 0.86 on the first cohort while outperforming FreeSurfer and FastSurfer-based ChP segmentations. On the dataset originating from clinical practice, the method reaches a Dice coefficient of 0.67 ± 0.01 (being close to the inter-rater agreement of 0.64 ± 0.02) and a volume correlation of 0.84. These results demonstrate that this is a suitable and robust method for the segmentation of the ChP both on research and clinical datasets.

Data Augmentation in High Dimensional Low Sample Size Setting Using a Geometry-Based Variational Autoencoder

In this paper, we propose a new method to perform data augmentation in a reliable way in the High Dimensional Low Sample Size (HDLSS) setting using a geometry-based variational autoencoder (VAE). Our approach combines the proposal of 1) a new VAE model, the latent space of which is modeled as a Riemannian manifold and which combines both Riemannian metric learning and normalizing flows and 2) a new generation scheme which produces more meaningful samples especially in the context of small data sets. The method is tested through a wide experimental study where its robustness to data sets, classifiers and training samples size is stressed. It is also validated on a medical imaging classification task on the challenging ADNI database where a small number of 3D brain magnetic resonance images (MRIs) are considered and augmented using the proposed VAE framework. In each case, the proposed method allows for a significant and reliable gain in the classification metrics. For instance, balanced accuracy jumps from 66.3% to 74.3% for a state-of-the-art convolutional neural network classifier trained with 50 MRIs of cognitively normal (CN) and 50 Alzheimer disease (AD) patients and from 77.7% to 86.3% when trained with 243 CN and 210 AD while improving greatly sensitivity and specificity metrics.

Forecasting individual progression trajectories in Alzheimer's disease

The anticipation of progression of Alzheimer's disease (AD) is crucial for evaluations of secondary prevention measures thought to modify the disease trajectory. However, it is difficult to forecast the natural progression of AD, notably because several functions decline at different ages and different rates in different patients. We evaluate here AD Course Map, a statistical model predicting the progression of neuropsychological assessments and imaging biomarkers for a patient from current medical and radiological data at early disease stages. We tested the method on more than 96,000 cases, with a pool of more than 4,600 patients from four continents. We measured the accuracy of the method for selecting participants displaying a progression of clinical endpoints during a hypothetical trial. We show that enriching the population with the predicted progressors decreases the required sample size by 38% to 50%, depending on trial duration, outcome, and targeted disease stage, from asymptomatic individuals at risk of AD to subjects with early and mild AD. We show that the method introduces no biases regarding sex or geographic locations and is robust to missing data. It performs best at the earliest stages of disease and is therefore highly suitable for use in prevention trials.

Generalizable deep learning model for early Alzheimer's disease detection from structural MRIs

Early diagnosis of Alzheimer's disease plays a pivotal role in patient care and clinical trials. In this study, we have developed a new approach based on 3D deep convolutional neural networks to accurately differentiate mild Alzheimer's disease dementia from mild cognitive impairment and cognitively normal individuals using structural MRIs. For comparison, we have built a reference model based on the volumes and thickness of previously reported brain regions that are known to be implicated in disease progression. We validate both models on an internal held-out cohort from The Alzheimer's Disease Neuroimaging Initiative (ADNI) and on an external independent cohort from The National Alzheimer's Coordinating Center (NACC). The deep-learning model is accurate, achieved an area-under-the-curve (AUC) of 85.12 when distinguishing between cognitive normal subjects and subjects with either MCI or mild Alzheimer's dementia. In the more challenging task of detecting MCI, it achieves an AUC of 62.45. It is also significantly faster than the volume/thickness model in which the volumes and thickness need to be extracted beforehand. The model can also be used to forecast progression: subjects with mild cognitive impairment misclassified as having mild Alzheimer's disease dementia by the model were faster to progress to dementia over time. An analysis of the features learned by the proposed model shows that it relies on a wide range of regions associated with Alzheimer's disease. These findings suggest that deep neural networks can automatically learn to identify imaging biomarkers that are predictive of Alzheimer's disease, and leverage them to achieve accurate early detection of the disease.

TR-GAN: Multi-Session Future MRI Prediction With Temporal Recurrent Generative Adversarial Network

Magnetic Resonance Imaging (MRI) has been proven to be an efficient way to diagnose Alzheimer's disease (AD). Recent dramatic progress on deep learning greatly promotes the MRI analysis based on data-driven CNN methods using a large-scale longitudinal MRI dataset. However, most of the existing MRI datasets are fragmented due to unexpected quits of volunteers. To tackle this problem, we propose a novel Temporal Recurrent Generative Adversarial Network (TR-GAN) to complete missing sessions of MRI datasets. Unlike existing GAN-based methods, which either fail to generate future sessions or only generate fixed-length sessions, TR-GAN takes all past sessions to recurrently and smoothly generate future ones with variant length. Specifically, TR-GAN adopts recurrent connection to deal with variant input sequence length and flexibly generate future variant sessions. Besides, we also design a multiple scale & location (MSL) module and a SWAP module to encourage the model to better focus on detailed information, which helps to generate high-quality MRI data. Compared with other popular GAN architectures, TR-GAN achieved the best performance in all evaluation metrics of two datasets. After expanding the Whole MRI dataset, the balanced accuracy of AD vs. cognitively normal (CN) vs. mild cognitive impairment (MCI) and stable MCI vs. progressive MCI classification can be increased by 3.61% and 4.00%, respectively.

Multi-scale discriminative regions analysis in FDG-PET imaging for early diagnosis of Alzheimer's disease

OBJECTIVE

Alzheimer's disease (AD) is a degenerative brain disorder, one of the main causes of death in elderly people, so early diagnosis of AD is vital to prompt access to medication and medical care. Fluorodeoxyglucose positron emission tomography (FDG-PET) proves to be effective to help understand neurological changes via measuring glucose uptake. Our aim is to explore information-rich regions of FDG-PET imaging, which enhance the accuracy and interpretability of AD-related diagnosis.

APPROACH

We develop a novel method for early diagnosis of AD based on multi-scale discriminative regions in FDG-PET imaging, which considers the diagnosis interpretability. Specifically, a multi-scale region localization (MSRL) module is discussed to automatically identify disease-related discriminative regions in full-volume FDG-PET images in an unsupervised manner, upon which a confidence score is designed to evaluate the prioritization of regions according to the density distribution of anomalies. Then, the proposed multi-scale region classification (MSRC) module adaptively fuses multi-scale region representations and makes decision fusion, which not only reduces useless information but also offers complementary information. Most of previous methods concentrate on discriminating AD from cognitively normal (CN), while mild cognitive impairment (MCI), a transitional state, facilitates early diagnosis. Therefore, our method is further applied to multiple AD-related diagnosis tasks, not limited to AD vs. CN.

MAIN RESULTS

Experimental results on the ADNI dataset show that the proposed method achieves superior performance over state-of-the-art FDG-PET-based approaches. Besides, some cerebral cortices highlighted by extracted regions cohere with medical research, further demonstrating the superiority.

SIGNIFICANCE

This work offers an effective method to achieve AD diagnosis and detect disease-affected regions in FDG-PET imaging. Our results could be beneficial for providing an additional opinion on the clinical diagnosis.

A Reproducible Deep-Learning-Based Computer-Aided Diagnosis Tool for Frontotemporal Dementia Using MONAI and Clinica Frameworks

Despite Artificial Intelligence (AI) being a leading technology in biomedical research, real-life implementation of AI-based Computer-Aided Diagnosis (CAD) tools into the clinical setting is still remote due to unstandardized practices during development. However, few or no attempts have been made to propose a reproducible CAD development workflow for 3D MRI data. In this paper, we present the development of an easily reproducible and reliable CAD tool using the Clinica and MONAI frameworks that were developed to introduce standardized practices in medical imaging. A Deep Learning (DL) algorithm was trained to detect frontotemporal dementia (FTD) on data from the NIFD database to ensure reproducibility. The DL model yielded 0.80 accuracy (95% confidence intervals: 0.64, 0.91), 1 sensitivity, 0.6 specificity, 0.83 F1-score, and 0.86 AUC, achieving a comparable performance with other FTD classification approaches. Explainable AI methods were applied to understand AI behavior and to identify regions of the images where the DL model misbehaves. Attention maps highlighted that its decision was driven by hallmarking brain areas for FTD and helped us to understand how to improve FTD detection. The proposed standardized methodology could be useful for benchmark comparison in FTD classification. AI-based CAD tools should be developed with the goal of standardizing pipelines, as varying pre-processing and training methods, along with the absence of model behavior explanations, negatively impact regulators' attitudes towards CAD. The adoption of common best practices for neuroimaging data analysis is a step toward fast evaluation of efficacy and safety of CAD and may accelerate the adoption of AI products in the healthcare system.

ClinicaDL: An open-source deep learning software for reproducible neuroimaging processing

BACKGROUND AND OBJECTIVE

As deep learning faces a reproducibility crisis and studies on deep learning applied to neuroimaging are contaminated by methodological flaws, there is an urgent need to provide a safe environment for deep learning users to help them avoid common pitfalls that will bias and discredit their results. Several tools have been proposed to help deep learning users design their framework for neuroimaging data sets. Software overview: We present here ClinicaDL, one of these software tools. ClinicaDL interacts with BIDS, a standard format in the neuroimaging field, and its derivatives, so it can be used with a large variety of data sets. Moreover, it checks the absence of data leakage when inferring the results of new data with trained networks, and saves all necessary information to guarantee the reproducibility of results. The combination of ClinicaDL and its companion project Clinica allows performing an end-to-end neuroimaging analysis, from the download of raw data sets to the interpretation of trained networks, including neuroimaging preprocessing, quality check, label definition, architecture search, and network training and evaluation.

CONCLUSIONS

We implemented ClinicaDL to bring answers to three common issues encountered by deep learning users who are not always familiar with neuroimaging data: (1) the format and preprocessing of neuroimaging data sets, (2) the contamination of the evaluation procedure by data leakage and (3) a lack of reproducibility. We hope that its use by researchers will allow producing more reliable and thus valuable scientific studies in our field.

Combining Neuroimaging and Omics Datasets for Disease Classification Using Graph Neural Networks

Both neuroimaging and genomics datasets are often gathered for the detection of neurodegenerative diseases. Huge dimensionalities of neuroimaging data as well as omics data pose tremendous challenge for methods integrating multiple modalities. There are few existing solutions that can combine both multi-modal imaging and multi-omics datasets to derive neurological insights. We propose a deep neural network architecture that combines both structural and functional connectome data with multi-omics data for disease classification. A graph convolution layer is used to model functional magnetic resonance imaging (fMRI) and diffusion tensor imaging (DTI) data simultaneously to learn compact representations of the connectome. A separate set of graph convolution layers are then used to model multi-omics datasets, expressed in the form of population graphs, and combine them with latent representations of the connectome. An attention mechanism is used to fuse these outputs and provide insights on which omics data contributed most to the model's classification decision. We demonstrate our methods for Parkinson's disease (PD) classification by using datasets from the Parkinson's Progression Markers Initiative (PPMI). PD has been shown to be associated with changes in the human connectome and it is also known to be influenced by genetic factors. We combine DTI and fMRI data with multi-omics data from RNA Expression, Single Nucleotide Polymorphism (SNP), DNA Methylation and non-coding RNA experiments. A Matthew Correlation Coefficient of greater than 0.8 over many combinations of multi-modal imaging data and multi-omics data was achieved with our proposed architecture. To address the paucity of paired multi-modal imaging data and the problem of imbalanced data in the PPMI dataset, we compared the use of oversampling against using CycleGAN on structural and functional connectomes to generate missing imaging modalities. Furthermore, we performed ablation studies that offer insights into the importance of each imaging and omics modality for the prediction of PD. Analysis of the generated attention matrices revealed that DNA Methylation and SNP data were the most important omics modalities out of all the omics datasets considered. Our work motivates further research into imaging genetics and the creation of more multi-modal imaging and multi-omics datasets to study PD and other complex neurodegenerative diseases.

Unravelling the impact of frontal lobe impairment for social dysfunction in myotonic dystrophy type 1

Myotonic dystrophy type 1 is an autosomal dominant multisystemic disorder affecting muscular and extra muscular systems, including the central nervous system. Cerebral involvement in myotonic dystrophy type 1 is associated with subtle cognitive and behavioural disorders, of major impact on socio-professional adaptation. The social dysfunction and its potential relation to frontal lobe neuropsychology remain under-evaluated in this pathology. The neuroanatomical network underpinning that disorder is yet to disentangle. Twenty-eight myotonic dystrophy type 1 adult patients (mean age: 46 years old) and 18 age and sex-matched healthy controls were included in the study. All patients performed an exhaustive neuropsychological assessment with a specific focus on frontal lobe neuropsychology (motivation, social cognition and executive functions). Among them, 18 myotonic dystrophy type 1 patients and 18 healthy controls had a brain MRI with T1 and T2 Flair sequences. Grey matter segmentation, Voxel-based morphometry and cortical thickness estimation were performed with Statistical Parametric Mapping Software SPM12 and Freesurfer software. Furthermore, T2 white matter lesions and subcortical structures were segmented with Automated Volumetry Software. Most patients showed significant impairment in executive frontal functions (auditory working memory, inhibition, contextualization and mental flexibility). Patients showed only minor difficulties in social cognition tests mostly in cognitive Theory of Mind, but with relative sparing of affective Theory of Mind and emotion recognition. Neuroimaging analysis revealed atrophy mostly in the parahippocampal and hippocampal regions and to a lesser extent in basal ganglia, regions involved in social navigation and mental flexibility, respectively. Social cognition scores were correlated with right parahippocampal gyrus atrophy. Social dysfunction in myotonic dystrophy type 1 might be a consequence of cognitive impairment regarding mental flexibility and social contextualization rather than a specific social cognition deficit such as emotion recognition. We suggest that both white matter lesions and grey matter disease could account for this social dysfunction, involving, in particular, the frontal-subcortical network and the hippocampal/arahippocampal regions, brain regions known, respectively, to integrate contextualization and social navigation.