Longitudinal imaging highlights preferential basal ganglia circuit atrophy in Huntington's disease.

Huntington's disease is caused by a CAG repeat expansion in the Huntingtin gene (HTT), coding for polyglutamine in the Huntingtin protein, with longer CAG repeats causing earlier age of onset. The variable 'Age' × ('CAG'-L), where 'Age' is the current age of the individual, 'CAG' is the repeat length and L is a constant (reflecting an approximation of the threshold), termed the 'CAG Age Product' (CAP) enables the consideration of many individuals with different CAG repeat expansions at the same time for analysis of any variable and graphing using the CAG Age Product score as the X axis. Structural MRI studies have showed that progressive striatal atrophy begins many years prior to the onset of diagnosable motor Huntington's disease, confirmed by longitudinal multicentre studies on three continents, including PREDICT-HD, TRACK-HD and IMAGE-HD. However, previous studies have not clarified the relationship between striatal atrophy, atrophy of other basal ganglia structures, and atrophy of other brain regions. The present study has analysed all three longitudinal datasets together using a single image segmentation algorithm and combining data from a large number of subjects across a range of CAG Age Product score. In addition, we have used a strategy of normalizing regional atrophy to atrophy of the whole brain, in order to determine which regions may undergo preferential degeneration. This made possible the detailed characterization of regional brain atrophy in relation to CAG Age Product score. There is dramatic selective atrophy of regions involved in the basal ganglia circuit-caudate, putamen, nucleus accumbens, globus pallidus and substantia nigra. Most other regions of the brain appear to have slower but steady degeneration. These results support (but certainly do not prove) the hypothesis of circuit-based spread of pathology in Huntington's disease, possibly due to spread of mutant Htt protein, though other connection-based mechanisms are possible. Therapeutic targets related to prion-like spread of pathology or other mechanisms may be suggested. In addition, they have implications for current neurosurgical therapeutic approaches, since delivery of therapeutic agents solely to the caudate and putamen may miss other structures affected early, such as nucleus accumbens and output nuclei of the striatum, the substantia nigra and the globus pallidus.

Mixed longitudinal and cross-sectional analyses of deep gray matter and white matter using diffusion weighted images in premanifest and manifest Huntington's disease.

Changes in the brain of patients with Huntington's disease (HD) begin years before clinical onset, so it remains critical to identify biomarkers to track these early changes. Metrics derived from tensor modeling of diffusion-weighted MRIs (DTI), that indicate the microscopic brain structure, can add important information to regional volumetric measurements. This study uses two large-scale longitudinal, multicenter datasets, PREDICT-HD and IMAGE-HD, to trace changes in DTI of HD participants with a broad range of CAP scores (a product of CAG repeat expansion and age), including those with pre-manifest disease (i.e., prior to clinical onset). Utilizing a fully automated data-driven approach to study the whole brain divided in regions of interest, we traced changes in DTI metrics (diffusivity and fractional anisotropy) versus CAP scores, using sigmoidal and linear regression models. We identified points of inflection in the sigmoidal regression using change-point analysis. The deep gray matter showed more evident and earlier changes in DTI metrics over CAP scores, compared to the deep white matter. In the deep white matter, these changes were more evident and occurred earlier in superior and posterior areas, compared to anterior and inferior areas. The curves of mean diffusivity vs. age of HD participants within a fixed CAP score were different from those of controls, indicating that the disease has an additional effect to age on the microscopic brain structure. These results show the regional and temporal vulnerability of the white matter and deep gray matter in HD, with potential implications for experimental therapeutics.

A large public dataset of annotated clinical MRIs and metadata of patients with acute stroke.

To extract meaningful and reproducible models of brain function from stroke images, for both clinical and research proposes, is a daunting task severely hindered by the great variability of lesion frequency and patterns. Large datasets are therefore imperative, as well as fully automated image post-processing tools to analyze them. The development of such tools, particularly with artificial intelligence, is highly dependent on the availability of large datasets to model training and testing. We present a public dataset of 2,888 multimodal clinical MRIs of patients with acute and early subacute stroke, with manual lesion segmentation, and metadata. The dataset provides high quality, large scale, human-supervised knowledge to feed artificial intelligence models and enable further development of tools to automate several tasks that currently rely on human labor, such as lesion segmentation, labeling, calculation of disease-relevant scores, and lesion-based studies relating function to frequency lesion maps.

Automatic comprehensive radiological reports for clinical acute stroke MRIs.

BACKGROUND: Although artificial intelligence systems that diagnosis among different conditions from medical images are long term aims, specific goals for automation of human-labor, time-consuming tasks are not only feasible but equally important. Acute conditions that require quantitative metrics, such as acute ischemic strokes, can greatly benefit by the consistency, objectiveness, and accessibility of automated radiological reports. METHODS: We used 1,878 annotated brain MRIs to generate a fully automated system that outputs radiological reports in addition to the infarct volume, 3D digital infarct mask, and the feature vector of anatomical regions affected by the acute infarct. This system is associated to a deep-learning algorithm for segmentation of the ischemic core and to parcellation schemes defining arterial territories and classically-identified anatomical brain structures. RESULTS: Here we show that the performance of our system to generate radiological reports was comparable to that of an expert evaluator. The weight of the components of the feature vectors that supported the prediction of the reports, as well as the prediction probabilities are outputted, making the pre-trained models behind our system interpretable. The system is publicly available, runs in real time, in local computers, with minimal computational requirements, and it is readily useful for non-expert users. It supports large-scale processing of new and legacy data, enabling clinical and translational research. CONCLUSION: The generation of reports indicates that our fully automated system is able to extract quantitative, objective, structured, and personalized information from stroke MRIs.

Artificial intelligence (AI) uses computer software to solve problems that normally require human input. It is likely that AI will take over, or help with, certain tasks in medical imaging, particularly where these tasks are time-consuming and laborious for clinicians. Here, we demonstrate the possibility of using AI to generate radiological reports for brain scans from patients who have had a stroke. These reports provide a summary of what is shown in the scans, and are normally written by clinicians. Our system performs similarly to human experts, is fast, publicly available, and runs on normal computers with minimal computational requirements, meaning that it might be a useful tool for researchers and clinicians to use when assessing and treating patients with stroke.

Automatic comprehensive aspects reports in clinical acute stroke MRIs.

The Alberta Stroke Program Early CT Score (ASPECTS) is a simple visual system to assess the extent and location of ischemic stroke core. The capability of ASPECTS for selecting patients' treatment, however, is affected by the variability in human evaluation. In this study, we developed a fully automatic system to calculate ASPECTS comparable with consensus expert readings. Our system was trained in 400 clinical diffusion weighted images of patients with acute infarcts and evaluated with an external testing set of 100 cases. The models are interpretable, and the results are comprehensive, evidencing the features that lead to the classification. This system adds to our automated pipeline for acute stroke detection, segmentation, and quantification in MRIs (ADS), which outputs digital infarct masks and the proportion of diverse brain regions injured, in addition to the predicted ASPECTS, the prediction probability and the explanatory features. ADS is public, free, accessible to non-experts, has very few computational requirements, and run in real time in local CPUs with a single command line, fulfilling the conditions to perform large-scale, reproducible clinical and translational research.

Digital 3D Brain MRI Arterial Territories Atlas.

The locus and extent of brain damage in the event of vascular insult can be quantitatively established quickly and easily with vascular atlases. Although highly anticipated by clinicians and clinical researchers, no digital MRI arterial atlas is readily available for automated data analyses. We created a digital arterial territory atlas based on lesion distributions in 1,298 patients with acute stroke. The lesions were manually traced in the diffusion-weighted MRIs, binary stroke masks were mapped to a common space, probability maps of lesions were generated and the boundaries for each arterial territory was defined based on the ratio between probabilistic maps. The atlas contains the definition of four major supra- and infra-tentorial arterial territories: Anterior, Middle, Posterior Cerebral Arteries and Vertebro-Basilar, and sub-territories (thalamoperforating, lenticulostriate, basilar and cerebellar arterial territories), in two hierarchical levels. This study provides the first publicly-available, digital, 3D deformable atlas of arterial brain territories, which may serve as a valuable resource for large-scale, reproducible processing and analysis of brain MRIs of patients with stroke and other conditions.

Identifying brain hierarchical structures associated with Alzheimer's disease using a regularized regression method with tree predictors.

Brain segmentation at different levels is generally represented as hierarchical trees. Brain regional atrophy at specific levels was found to be marginally associated with Alzheimer's disease outcomes. In this study, we propose an ℓ1 -type regularization for predictors that follow a hierarchical tree structure. Considering a tree as a directed acyclic graph, we interpret the model parameters from a path analysis perspective. Under this concept, the proposed penalty regulates the total effect of each predictor on the outcome. With regularity conditions, it is shown that under the proposed regularization, the estimator of the model coefficient is consistent in ℓ2 -norm and the model selection is also consistent. When applied to a brain sMRI dataset acquired from the Alzheimer's Disease Neuroimaging Initiative (ADNI), the proposed approach identifies brain regions where atrophy in these regions demonstrates the declination in memory. With regularization on the total effects, the findings suggest that the impact of atrophy on memory deficits is localized from small brain regions, but at various levels of brain segmentation. Data used in preparation of this paper were obtained from the ADNI database.

Regularized regression on compositional trees with application to MRI analysis.

A compositional tree refers to a tree structure on a set of random variables where each random variable is a node and composition occurs at each non-leaf node of the tree. As a generalization of compositional data, compositional trees handle more complex relationships among random variables and appear in many disciplines, such as brain imaging, genomics and finance. We consider the problem of sparse regression on data that are associated with a compositional tree and propose a transformation-free tree-based regularized regression method for component selection. The regularization penalty is designed based on the tree structure and encourages a sparse tree representation. We prove that our proposed estimator for regression coefficients is both consistent and model selection consistent. In the simulation study, our method shows higher accuracy than competing methods under different scenarios. By analyzing a brain imaging data set from studies of Alzheimer's disease, our method identifies meaningful associations between memory decline and volume of brain regions that are consistent with current understanding.

3D Attention M-net for Short-axis Left Ventricular Myocardium Segmentation in Mice MR cardiac Images.

Small rodent cardiac magnetic resonance imaging (MRI) plays an important role in preclinical models of cardiac disease. Accurate myocardial boundaries delineation is crucial to most morphological and functional analysis in rodent cardiac MRIs. However, rodent cardiac MRIs, due to animal's small cardiac volume and high heart rate, are usually acquired with sub-optimal resolution and low signal-to-noise ratio (SNR). These rodent cardiac MRIs can also suffer from signal loss due to the intra-voxel dephasing. These factors make automatic myocardial segmentation challenging. Manual contouring could be applied to label myocardial boundaries but it is usually laborious, time consuming, and not systematically objective. In this study, we present a deep learning approach based on 3D attention M-net to perform automatic segmentation of left ventricular myocardium. In the deep learning architecture, we use dual spatial-channel attention gates between encoder and decoder along with multi-scale feature fusion path after decoder. Attention gates enable networks to focus on relevant spatial information and channel features to improve segmentation performance. A distance derived loss term, besides general dice loss and binary cross entropy loss, was also introduced to our hybrid loss functions to refine segmentation contours. The proposed model outperforms other generic models, like U-Net and FCN, in major segmentation metrics including the dice score (0.9072), Jaccard index (0.8307) and Hausdorff distance (3.1754 pixels), which are comparable to the results achieved by state-of-the-art models on human cardiac ACDC17 datasets.Clinical relevance Small rodent cardiac MRI is routinely used to probe the effect of individual genes or groups of genes on the etiology of a large number of cardiovascular diseases. An automatic myocardium segmentation algorithm specifically designed for these data can enhance accuracy and reproducibility of cardiac structure and function analysis.

Deep learning-based detection and segmentation of diffusion abnormalities in acute ischemic stroke.

Background: Accessible tools to efficiently detect and segment diffusion abnormalities in acute strokes are highly anticipated by the clinical and research communities. Methods: We developed a tool with deep learning networks trained and tested on a large dataset of 2,348 clinical diffusion weighted MRIs of patients with acute and sub-acute ischemic strokes, and further tested for generalization on 280 MRIs of an external dataset (STIR). Results: Our proposed model outperforms generic networks and DeepMedic, particularly in small lesions, with lower false positive rate, balanced precision and sensitivity, and robustness to data perturbs (e.g., artefacts, low resolution, technical heterogeneity). The agreement with human delineation rivals the inter-evaluator agreement; the automated lesion quantification of volume and contrast has virtually total agreement with human quantification. Conclusion: Our tool is fast, public, accessible to non-experts, with minimal computational requirements, to detect and segment lesions via a single command line. Therefore, it fulfills the conditions to perform large scale, reliable and reproducible clinical and translational research.

Intracellular location of aquaporin-2 serine 269 phosphorylation and dephosphorylation in kidney collecting duct cells.

Aquaporin-2 (AQP2) is a vasopressin-regulated water channel protein responsible for water reabsorption by the kidney collecting ducts. Under control conditions, most AQP2 resides in the recycling endosomes of principal cells, where it answers to vasopressin with trafficking to the apical plasma membrane to increase water reabsorption. Upon vasopressin withdrawal, apical AQP2 retreats to the early endosomes before joining the recycling endosomes for the next vasopressin stimulation. Prior studies have demonstrated a role of AQP2 S269 phosphorylation in reducing AQP2 endocytosis, thereby prolonging apical AQP2 retention. Here, we studied where in the cells S269 was phosphorylated and dephosphorylated in response to vasopressin versus withdrawal. In mpkCCD collecting cells, vacuolar protein sorting 35 knockdown slowed vasopressin-induced apical AQP2 trafficking, resulting in AQP2 accumulation in the recycling endosomes where S269 was phosphorylated. Rab7 knockdown, which impaired AQP2 trafficking from the early to recycling endosomes, reduced vasopressin-induced S269 phosphorylation. Rab5 knockdown, which impaired AQP2 endocytosis, did not affect vasopressin-induced S269 phosphorylation. Upon vasopressin withdrawal, S269 was not dephosphorylated in Rab5 knockdown cells. In contrast, S269 dephosphorylation upon vasopressin withdrawal was completed in Rab7 or vacuolar protein sorting 35 knockdown cells. We conclude that S269 is dephosphorylated during Rab5-mediated AQP2 endocytosis before AQP2 joins the recycling endosomes upon vasopressin withdrawal. While in the recycling endosomes, AQP2 can be phosphorylated at S269 in response to vasopressin before apical trafficking.

Rab7 involves Vps35 to mediate AQP2 sorting and apical trafficking in collecting duct cells.

Aquaporin-2 (AQP2) is a vasopressin-regulated water channel protein responsible for osmotic water reabsorption by kidney collecting ducts. In response to vasopressin, AQP2 traffics from intracellular vesicles to the apical plasma membrane of collecting duct principal cells, where it increases water permeability and, hence, water reabsorption. Despite continuing efforts, gaps remain in our knowledge of vasopressin-regulated AQP2 trafficking. Here, we studied the functions of two retromer complex proteins, small GTPase Rab7 and vacuolar protein sorting 35 (Vps35), in vasopressin-induced AQP2 trafficking in a collecting duct cell model (mpkCCD cells). We showed that upon vasopressin removal, apical AQP2 returned to Rab5-positive early endosomes before joining Rab11-positive recycling endosomes. In response to vasopressin, Rab11-associated AQP2 trafficked to the apical plasma membrane before Rab5-associated AQP2 did so. Rab7 knockdown resulted in AQP2 accumulation in early endosomes and impaired vasopressin-induced apical AQP2 trafficking. In response to vasopressin, Rab7 transiently colocalized with Rab5, indicative of a role of Rab7 in AQP2 sorting in early endosomes before trafficking to the apical membrane. Rab7-mediated apical AQP2 trafficking in response to vasopressin required GTPase activity. When Vps35 was knocked down, AQP2 accumulated in recycling endosomes under vehicle conditions and did not traffic to the apical plasma membrane in response to vasopressin. We conclude that Rab7 and Vps35 participate in AQP2 sorting in early endosomes under vehicle conditions and apical membrane trafficking in response to vasopressin.

Using deep Siamese neural networks for detection of brain asymmetries associated with Alzheimer's Disease and Mild Cognitive Impairment.

In recent studies, neuroanatomical volume and shape asymmetries have been seen during the course of Alzheimer's Disease (AD) and could potentially be used as preclinical imaging biomarkers for the prediction of Mild Cognitive Impairment (MCI) and AD dementia. In this study, a deep learning framework utilizing Siamese neural networks trained on paired lateral inter-hemispheric regions is used to harness the discriminative power of whole-brain volumetric asymmetry. The method uses the MRICloud pipeline to yield low-dimensional volumetric features of pre-defined atlas brain structures, and a novel non-linear kernel trick to normalize these features to reduce batch effects across datasets and populations. By working with the low-dimensional features, Siamese networks were shown to yield comparable performance to studies that utilize whole-brain MR images, with the advantage of reduced complexity and computational time, while preserving the biological information density. Experimental results also show that Siamese networks perform better in certain metrics by explicitly encoding the asymmetry in brain volumes, compared to traditional prediction methods that do not use the asymmetry, on the ADNI and BIOCARD datasets.

Cortical thickness atrophy in the transentorhinal cortex in mild cognitive impairment.

This study examines the atrophy rates of subjects with mild cognitive impairment (MCI) compared to controls in four regions within the medial temporal lobe: the transentorhinal cortex (TEC), entorhinal cortex (ERC), hippocampus, and amygdala. These regions were manually segmented and then corrected for undesirable longitudinal variability via Large Deformation Diffeomorphic Metric Mapping (LDDMM) based longitudinal diffeomorphometry. Diffeomorphometry techniques were used to compare thickness measurements in the TEC with the ERC. There were more significant changes in thickness atrophy rate in the TEC than medial regions of the entorhinal cortex. Volume measures were also calculated for all four regions. Classifiers were constructed using linear discriminant analysis to demonstrate that average thickness and atrophy rate of TEC together was the most discriminating measure compared to the thickness and volume measures in the areas examined, in differentiating MCI from controls. These findings are consistent with autopsy findings demonstrating that initial neuronal changes are found in TEC before spreading more medially in the ERC and to other regions in the medial temporal lobe. These findings suggest that the TEC thickness could serve as a biomarker for Alzheimer's disease in the prodromal phase of the disease.