Release the Krakencoder: A unified brain connectome translation and fusion tool

Brain connectivity can be estimated in many ways, depending on modality and processing strategy. Here we present the Krakencoder, a joint connectome mapping tool that simultaneously, bidirectionally translates between structural (SC) and functional connectivity (FC), and across different atlases and processing choices via a common latent representation. These mappings demonstrate unprecedented accuracy and individual-level identifiability; the mapping between SC and FC has identifiability 42-54% higher than existing models. The Krakencoder combines all connectome flavors via a shared low-dimensional latent space. This "fusion" representation i) better reflects familial relatedness, ii) preserves age- and sex-relevant information and iii) enhances cognition-relevant information. The Krakencoder can be applied without retraining to new, out-of-age-distribution data while still preserving inter-individual differences in the connectome predictions and familial relationships in the latent representations. The Krakencoder is a significant leap forward in capturing the relationship between multi-modal brain connectomes in an individualized, behaviorally- and demographically-relevant way.

Test Retest Reproducibility of Organ Volume Measurements in ADPKD Using 3D Multimodality Deep Learning

RATIONALE AND OBJECTIVES

Following autosomal dominant polycystic kidney disease (ADPKD) progression by measuring organ volumes requires low measurement variability. The objective of this study is to reduce organ volume measurement variability on MRI of ADPKD patients by utilizing all pulse sequences to obtain multiple measurements which allows outlier analysis to find errors and averaging to reduce variability.

MATERIALS AND METHODS

In order to make measurements on multiple pulse sequences practical, a 3D multi-modality multi-class segmentation model based on nnU-net was trained/validated using T1, T2, SSFP, DWI and CT from 413 subjects. Reproducibility was assessed with test-re-test methodology on ADPKD subjects (n = 19) scanned twice within a 3-week interval correcting outliers and averaging the measurements across all sequences. Absolute percent differences in organ volumes were compared to paired students t-test.

RESULTS

Dice similarlity coefficient > 97%, Jaccard Index > 0.94, mean surface distance < 1 mm and mean Hausdorff Distance < 2 cm for all three organs and all five sequences were found on internal (n = 25), external (n = 37) and test-re-test reproducibility assessment (38 scans in 19 subjects). When averaging volumes measured from five MRI sequences, the model automatically segmented kidneys with test-re-test reproducibility (percent absolute difference between exam 1 and exam 2) of 1.3% which was better than all five expert observers. It reliably stratified ADPKD into Mayo Imaging Classification (area under the curve=100%) compared to radiologist.

CONCLUSION

3D deep learning measures organ volumes on five MRI sequences leveraging the power of outlier analysis and averaging to achieve 1.3% total kidney test-re-test reproducibility.

Learning across diverse biomedical data modalities and cohorts: Challenges and opportunities for innovation

In healthcare, machine learning (ML) shows significant potential to augment patient care, improve population health, and streamline healthcare workflows. Realizing its full potential is, however, often hampered by concerns about data privacy, diversity in data sources, and suboptimal utilization of different data modalities. This review studies the utility of cross-cohort cross-category (C4) integration in such contexts: the process of combining information from diverse datasets distributed across distinct, secure sites. We argue that C4 approaches could pave the way for ML models that are both holistic and widely applicable. This paper provides a comprehensive overview of C4 in health care, including its present stage, potential opportunities, and associated challenges.

Remote Associations Between Tau and Cortical Amyloid-β Are Stage-Dependent

Background

Histopathologic studies of Alzheimer's disease (AD) suggest that extracellular amyloid-β (Aβ) plaques promote the spread of neurofibrillary tau tangles. However, these two proteinopathies initiate in spatially distinct brain regions, so how they interact during AD progression is unclear.

Objective

In this study, we utilized Aβ and tau positron emission tomography (PET) scans from 572 older subjects (476 healthy controls (HC), 14 with mild cognitive impairment (MCI), 82 with mild AD), at varying stages of the disease, to investigate to what degree tau is associated with cortical Aβ deposition.

Methods

Using multiple linear regression models and a pseudo-longitudinal ordering technique, we investigated remote tau-Aβ associations in four pathologic phases of AD progression based on tau spread: 1) no-tau, 2) pre-acceleration, 3) acceleration, and 4) post-acceleration.

Results

No significant tau-Aβ association was detected in the no-tau phase. In the pre-acceleration phase, the earliest stage of tau deposition, associations emerged between regional tau in medial temporal lobe (MTL) (i.e., entorhinal cortex, parahippocampal gyrus) and cortical Aβ in lateral temporal lobe regions. The strongest tau-Aβ associations were found in the acceleration phase, in which tau in MTL regions was strongly associated with cortical Aβ (i.e., temporal and frontal lobes regions). Strikingly, in the post-acceleration phase, including 96% of symptomatic subjects, tau-Aβ associations were no longer significant.

Conclusions

The results indicate that associations between tau and Aβ are stage-dependent, which could have important implications for understanding the interplay between these two proteinopathies during the progressive stages of AD.

A robust and interpretable deep learning framework for multi-modal registration via keypoints

We present KeyMorph, a deep learning-based image registration framework that relies on automatically detecting corresponding keypoints. State-of-the-art deep learning methods for registration often are not robust to large misalignments, are not interpretable, and do not incorporate the symmetries of the problem. In addition, most models produce only a single prediction at test-time. Our core insight which addresses these shortcomings is that corresponding keypoints between images can be used to obtain the optimal transformation via a differentiable closed-form expression. We use this observation to drive the end-to-end learning of keypoints tailored for the registration task, and without knowledge of ground-truth keypoints. This framework not only leads to substantially more robust registration but also yields better interpretability, since the keypoints reveal which parts of the image are driving the final alignment. Moreover, KeyMorph can be designed to be equivariant under image translations and/or symmetric with respect to the input image ordering. Finally, we show how multiple deformation fields can be computed efficiently and in closed-form at test time corresponding to different transformation variants. We demonstrate the proposed framework in solving 3D affine and spline-based registration of multi-modal brain MRI scans. In particular, we show registration accuracy that surpasses current state-of-the-art methods, especially in the context of large displacements. Our code is available at https://github.com/alanqrwang/keymorph.

Optic Nerve Diameter on Non-Contrast Computed Tomography and Intracranial Hypertension in Patients With Acute Brain Injury: A Validation Study

Intracranial hypertension is a feared complication of acute brain injury that can cause ischemic stroke, herniation, and death. Identifying those at risk is difficult, and the physical examination is often confounded. Given the widespread availability and use of computed tomography (CT) in patients with acute brain injury, prior work has attempted to use optic nerve diameter measurements to identify those at risk of intracranial hypertension. We aimed to validate the use of optic nerve diameter measurements on CT as a screening tool for intracranial hypertension in a large cohort of brain-injured patients. We performed a retrospective observational cohort study in a single tertiary referral Neuroscience Intensive Care Unit. We identified patients with documented intracranial pressure (ICP) measures as part of their routine clinical care who had non-contrast CT head scans collected within 24 h, and then measured the optic nerve diameters and explored the relationship and test characteristics of these measures to identify those at risk of intracranial hypertension. In a cohort of 314 patients, optic nerve diameter on CT was linearly but weakly associated with ICP. When used to identify those with intracranial hypertension (> 20 mm Hg), the area under the receiver operator curve (AUROC) was 0.68. Using a previously proposed threshold of 0.6 cm, the sensitivity was 81%, specificity 43%, positive likelihood ratio 1.4, and negative likelihood ratio 0.45. CT-derived optic nerve diameter using a threshold of 0.6 cm is sensitive but not specific for intracranial hypertension, and the overall correlation is weak.

Human brain responses are modulated when exposed to optimized natural images or synthetically generated images

Understanding how human brains interpret and process information is important. Here, we investigated the selectivity and inter-individual differences in human brain responses to images via functional MRI. In our first experiment, we found that images predicted to achieve maximal activations using a group level encoding model evoke higher responses than images predicted to achieve average activations, and the activation gain is positively associated with the encoding model accuracy. Furthermore, anterior temporal lobe face area (aTLfaces) and fusiform body area 1 had higher activation in response to maximal synthetic images compared to maximal natural images. In our second experiment, we found that synthetic images derived using a personalized encoding model elicited higher responses compared to synthetic images from group-level or other subjects' encoding models. The finding of aTLfaces favoring synthetic images than natural images was also replicated. Our results indicate the possibility of using data-driven and generative approaches to modulate macro-scale brain region responses and probe inter-individual differences in and functional specialization of the human visual system.

Deep learning analysis of blood flow sounds to detect arteriovenous fistula stenosis

For hemodialysis patients, arteriovenous fistula (AVF) patency determines whether adequate hemofiltration can be achieved, and directly influences clinical outcomes. Here, we report the development and performance of a deep learning model for automated AVF stenosis screening based on the sound of AVF blood flow using supervised learning with data validated by ultrasound. We demonstrate the importance of contextualizing the sound with location metadata as the characteristics of the blood flow sound varies significantly along the AVF. We found the best model to be a vision transformer trained on spectrogram images. Our model can screen for stenosis at a performance level comparable to that of a nephrologist performing a physical exam, but with the advantage of being automated and scalable. In a high-volume, resource-limited clinical setting, automated AVF stenosis screening can help ensure patient safety via early detection of at-risk vascular access, streamline the dialysis workflow, and serve as a patient-facing tool to allow for at-home, self-screening.

Modulating human brain responses via optimal natural image selection and synthetic image generation

One of the main goals of neuroscience is to understand how biological brains interpret and process incoming environmental information. Building computational encoding models that map images to neural responses is one way to pursue this goal. Moreover, generating or selecting visual stimuli designed to achieve specific patterns of responses allows exploration and control of neuronal firing rates or regional brain activity responses. Here, we investigated the brain's regional activation selectivity and inter-individual differences in human brain responses to various sets of natural and synthetic (generated) images via two functional MRI (fMRI) studies. For our first fMRI study, we used a pre-trained group-level neural model for selecting or synthesizing images that are predicted to maximally activate targeted brain regions. We then presented these images to subjects while collecting their fMRI data. Our results show that optimized images indeed evoke larger magnitude responses than other images predicted to achieve average levels of activation.Furthermore, the activation gain is positively associated with the encoding model accuracy. While most regions' activations in response to maximal natural images and maximal synthetic images were not different, two regions, namely anterior temporal lobe faces (aTLfaces) and fusiform body area 1 (FBA1), had significantly higher activation in response to maximal synthetic images compared to maximal natural images. On the other hand, three regions; medial temporal lobe face area (mTLfaces), ventral word form area 1 (VWFA1) and ventral word form area 2 (VWFA2), had higher activation in response to maximal natural images compared to maximal synthetic images. In our second fMRI experiment, we focused on probing inter-individual differences in face regions' responses and found that individual-specific synthetic (and not natural) images derived using a personalized encoding model elicited significantly higher responses compared to synthetic images derived from the group-level or other subjects' encoding models. Finally, we replicated the finding showing synthetic images elicited larger activation responses in the aTLfaces region compared to natural image responses in that region. Here, for the first time, we leverage our data-driven and generative modeling framework NeuroGen to probe inter-individual differences in and functional specialization of the human visual system. Our results indicate that NeuroGen can be used to modulate macro-scale brain regions in specific individuals using synthetically generated visual stimuli.

Hyper-convolutions via implicit kernels for medical image analysis

The convolutional neural network (CNN) is one of the most commonly used architectures for computer vision tasks. The key building block of a CNN is the convolutional kernel that aggregates information from the pixel neighborhood and shares weights across all pixels. A standard CNN's capacity, and thus its performance, is directly related to the number of learnable kernel weights, which is determined by the number of channels and the kernel size (support). In this paper, we present the hyper-convolution, a novel building block that implicitly encodes the convolutional kernel using spatial coordinates. Unlike a regular convolutional kernel, whose weights are independently learned, hyper-convolution kernel weights are correlated through an encoder that maps spatial coordinates to their corresponding values. Hyper-convolutions decouple kernel size from the total number of learnable parameters, enabling a more flexible architecture design. We demonstrate in our experiments that replacing regular convolutions with hyper-convolutions can improve performance with less parameters, and increase robustness against noise. We provide our code here: https://github.com/tym002/Hyper-Convolution.

LARO: Learned acquisition and reconstruction optimization to accelerate quantitative susceptibility mapping

Quantitative susceptibility mapping (QSM) involves acquisition and reconstruction of a series of images at multi-echo time points to estimate tissue field, which prolongs scan time and requires specific reconstruction technique. In this paper, we present our new framework, called Learned Acquisition and Reconstruction Optimization (LARO), which aims to accelerate the multi-echo gradient echo (mGRE) pulse sequence for QSM. Our approach involves optimizing a Cartesian multi-echo k-space sampling pattern with a deep reconstruction network. Next, this optimized sampling pattern was implemented in an mGRE sequence using Cartesian fan-beam k-space segmenting and ordering for prospective scans. Furthermore, we propose to insert a recurrent temporal feature fusion module into the reconstruction network to capture signal redundancies along echo time. Our ablation studies show that both the optimized sampling pattern and proposed reconstruction strategy help improve the quality of the multi-echo image reconstructions. Generalization experiments show that LARO is robust on the test data with new pathologies and different sequence parameters. Our code is available at https://github.com/Jinwei1209/LARO-QSM.git.

Machine learning can aid in prediction of IDH mutation from H&E-stained histology slides in infiltrating gliomas

While Machine Learning (ML) models have been increasingly applied to a range of histopathology tasks, there has been little emphasis on characterizing these models and contrasting them with human experts. We present a detailed empirical analysis comparing expert neuropathologists and ML models at predicting IDH mutation status in H&E-stained histology slides of infiltrating gliomas, both independently and synergistically. We find that errors made by neuropathologists and ML models trained using the TCGA dataset are distinct, representing modest agreement between predictions (human-vs.-human κ = 0.656; human-vs.-ML model κ = 0.598). While no ML model surpassed human performance on an independent institutional test dataset (human AUC = 0.901, max ML AUC = 0.881), a hybrid model aggregating human and ML predictions demonstrates predictive performance comparable to the consensus of two expert neuropathologists (hybrid classifier AUC = 0.921 vs. two-neuropathologist consensus AUC = 0.920). We also show that models trained at different levels of magnification exhibit different types of errors, supporting the value of aggregation across spatial scales in the ML approach. Finally, we present a detailed interpretation of our multi-scale ML ensemble model which reveals that predictions are driven by human-identifiable features at the patch-level.

Machine learning based multi-modal prediction of future decline toward Alzheimer's disease: An empirical study

Alzheimer's disease (AD) is a neurodegenerative condition that progresses over decades. Early detection of individuals at high risk of future progression toward AD is likely to be of critical significance for the successful treatment and/or prevention of this devastating disease. In this paper, we present an empirical study to characterize how predictable an individual subjects' future AD trajectory is, several years in advance, based on rich multi-modal data, and using modern deep learning methods. Crucially, the machine learning strategy we propose can handle different future time horizons and can be trained with heterogeneous data that exhibit missingness and non-uniform follow-up visit times. Our experiments demonstrate that our strategy yields predictions that are more accurate than a model trained on a single time horizon (e.g. 3 years), which is common practice in prior literature. We also provide a comparison between linear and nonlinear models, verifying the well-established insight that the latter can offer a boost in performance. Our results also confirm that predicting future decline for cognitively normal (CN) individuals is more challenging than for individuals with mild cognitive impairment (MCI). Intriguingly, however, we discover that prediction accuracy decreases with increasing time horizon for CN subjects, but the trend is in the opposite direction for MCI subjects. Additionally, we quantify the contribution of different data types in prediction, which yields novel insights into the utility of different biomarkers. We find that molecular biomarkers are not as helpful for CN individuals as they are for MCI individuals, whereas magnetic resonance imaging biomarkers (hippocampus volume, specifically) offer a significant boost in prediction accuracy for CN individuals. Finally, we show how our model's prediction reveals the evolution of individual-level progression risk over a five-year time horizon. Our code is available at https://github.com/batuhankmkaraman/mlbasedad.

Pulse Sequence Dependence of a Simple and Interpretable Deep Learning Method for Detection of Clinically Significant Prostate Cancer Using Multiparametric MRI

RATIONALE AND OBJECTIVES

Multiparametric magnetic resonance imaging (mpMRI) is increasingly used for risk stratification and localization of prostate cancer (PCa). Thanks to the great success of deep learning models in computer vision, the potential application for early detection of PCa using mpMRI is imminent.

MATERIALS AND METHODS

Deep learning analysis of the PROSTATEx dataset.

RESULTS

In this study, we show a simple convolutional neural network (CNN) with mpMRI can achieve high performance for detection of clinically significant PCa (csPCa), depending on the pulse sequences used. The mpMRI model with T2-ADC-DWI achieved 0.90 AUC score in the held-out test set, not significantly better than the model using K^trans instead of DWI (AUC 0.89). Interestingly, the model incorporating T2-ADC- K^trans better estimates grade. We also describe a saliency "heat" map. Our results show that csPCa detection models with mpMRI may be leveraged to guide clinical management strategies.

CONCLUSION

Convolutional neural networks incorporating multiple pulse sequences show high performance for detection of clinically-significant prostate cancer, and the model including dynamic contrast-enhanced information correlates best with grade.

A transformer-Based neural language model that synthesizes brain activation maps from free-form text queries

Neuroimaging studies are often limited by the number of subjects and cognitive processes that can be feasibly interrogated. However, a rapidly growing number of neuroscientific studies have collectively accumulated an extensive wealth of results. Digesting this growing literature and obtaining novel insights remains to be a major challenge, since existing meta-analytic tools are constrained to keyword queries. In this paper, we present Text2Brain, an easy to use tool for synthesizing brain activation maps from open-ended text queries. Text2Brain was built on a transformer-based neural network language model and a coordinate-based meta-analysis of neuroimaging studies. Text2Brain combines a transformer-based text encoder and a 3D image generator, and was trained on variable-length text snippets and their corresponding activation maps sampled from 13,000 published studies. In our experiments, we demonstrate that Text2Brain can synthesize meaningful neural activation patterns from various free-form textual descriptions. Text2Brain is available at https://braininterpreter.com as a web-based tool for efficiently searching through the vast neuroimaging literature and generating new hypotheses.

Predicting Individual Task Contrasts From Resting-state Functional Connectivity using a Surface-based Convolutional Network

Task-based and resting-state represent the two most common experimental paradigms of functional neuroimaging. While resting-state offers a flexible and scalable approach for characterizing brain function, task-based techniques provide superior localization. In this paper, we build on recent deep learning methods to create a model that predicts task-based contrast maps from resting-state fMRI scans. Specifically, we propose BrainSurfCNN, a surface-based fully-convolutional neural network model that works with a representation of the brain's cortical sheet. BrainSurfCNN achieves exceptional predictive accuracy on independent test data from the Human Connectome Project, which is on par with the repeat reliability of the measured subject-level contrast maps. Conversely, our analyses reveal that a previously published benchmark is no better than group-average contrast maps. Finally, we demonstrate that BrainSurfCNN can generalize remarkably well to novel domains with limited training data.

NeuroGen: Activation optimized image synthesis for discovery neuroscience

Functional MRI (fMRI) is a powerful technique that has allowed us to characterize visual cortex responses to stimuli, yet such experiments are by nature constructed based on a priori hypotheses, limited to the set of images presented to the individual while they are in the scanner, are subject to noise in the observed brain responses, and may vary widely across individuals. In this work, we propose a novel computational strategy, which we call NeuroGen, to overcome these limitations and develop a powerful tool for human vision neuroscience discovery. NeuroGen combines an fMRI-trained neural encoding model of human vision with a deep generative network to synthesize images predicted to achieve a target pattern of macro-scale brain activation. We demonstrate that the reduction of noise that the encoding model provides, coupled with the generative network’s ability to produce images of high fidelity, results in a robust discovery architecture for visual neuroscience. By using only a small number of synthetic images created by NeuroGen, we demonstrate that we can detect and amplify differences in regional and individual human brain response patterns to visual stimuli. We then verify that these discoveries are reflected in the several thousand observed image responses measured with fMRI. We further demonstrate that NeuroGen can create synthetic images predicted to achieve regional response patterns not achievable by the best-matching natural images. The NeuroGen framework extends the utility of brain encoding models and opens up a new avenue for exploring, and possibly precisely controlling, the human visual system.

Magnetic Resonance Imaging Radiomics-Based Machine Learning Prediction of Clinically Significant Prostate Cancer in Equivocal PI-RADS 3 Lesions

BACKGROUND

While Prostate Imaging Reporting and Data System (PI-RADS) 4 and 5 lesions typically warrant prostate biopsy and PI-RADS 1 and 2 lesions may be safely observed, PI-RADS 3 lesions are equivocal.

PURPOSE

To construct and cross-validate a machine learning model based on radiomics features from T₂ -weighted imaging (T₂ WI) of PI-RADS 3 lesions to identify clinically significant prostate cancer (csPCa), that is, pathological Grade Group ≥ 2.

STUDY TYPE

Single-center retrospective study.

POPULATION

A total of 240 patients were included (training cohort, n = 188, age range 43-82 years; test cohort, n = 52, age range 41-79 years). Eligibility criteria were 1) magnetic resonance imaging (MRI)-targeted biopsy between 2015 and 2020; 2) PI-RADS 3 index lesion identified on multiparametric MRI; (3) biopsy performed within 1 year of MRI. The percentages of csPCa lesions were 10.6% and 15.4% in the training and test cohorts, respectively.

FIELD STRENGTH/SEQUENCE

A 3 T; T₂ WI turbo-spin echo, diffusion-weighted spin-echo echo planar imaging, dynamic contrast-enhanced MRI with time-resolved T1-weighted imaging.

ASSESSMENT

Multislice volumes-of-interest (VOIs) were drawn in the PI-RADS 3 index lesions on T₂ WI. A total of 107 radiomics features (first-order histogram and second-order texture) were extracted from the segmented lesions.

STATISTICAL TESTS

A random forest classifier using the radiomics features as input was trained and validated for prediction of csPCa. The performance of the machine learning classifier, prostate specific antigen (PSA) density, and prostate volume for csPCa prediction was evaluated using receiver operating characteristic (ROC) analysis.

RESULTS

The trained random forest classifier constructed from the T₂ WI radiomics features good and statistically significant area-under-the-curves (AUCs) of 0.76 (P = 0.022) for prediction of csPCa in the test set. Prostate volume and PSA density showed moderate and nonsignificant performance (AUC 0.62, P = 0.275 and 0.61, P = 0.348, respectively) for csPCa prediction in the test set.

CONCLUSION

The machine learning classifier based on T₂ WI radiomic features demonstrated good performance for prediction of csPCa in PI-RADS 3 lesions.

EVIDENCE LEVEL

4 TECHNICAL EFFICACY: 2.

Cortical response to naturalistic stimuli is largely predictable with deep neural networks

Naturalistic stimuli, such as movies, activate a substantial portion of the human brain, invoking a response shared across individuals. Encoding models that predict neural responses to arbitrary stimuli can be very useful for studying brain function. However, existing models focus on limited aspects of naturalistic stimuli, ignoring the dynamic interactions of modalities in this inherently context-rich paradigm. Using movie-watching data from the Human Connectome Project, we build group-level models of neural activity that incorporate several inductive biases about neural information processing, including hierarchical processing, temporal assimilation, and auditory-visual interactions. We demonstrate how incorporating these biases leads to remarkable prediction performance across large areas of the cortex, beyond the sensory-specific cortices into multisensory sites and frontal cortex. Furthermore, we illustrate that encoding models learn high-level concepts that generalize to task-bound paradigms. Together, our findings underscore the potential of encoding models as powerful tools for studying brain function in ecologically valid conditions.

Heritability of individualized cortical network topography

Human cortex is patterned by a complex and interdigitated web of large-scale functional networks. Recent methodological breakthroughs reveal variation in the size, shape, and spatial topography of cortical networks across individuals. While spatial network organization emerges across development, is stable over time, and is predictive of behavior, it is not yet clear to what extent genetic factors underlie interindividual differences in network topography. Here, leveraging a nonlinear multidimensional estimation of heritability, we provide evidence that individual variability in the size and topographic organization of cortical networks are under genetic control. Using twin and family data from the Human Connectome Project (n = 1,023), we find increased variability and reduced heritability in the size of heteromodal association networks (h2 : M = 0.34, SD = 0.070), relative to unimodal sensory/motor cortex (h2 : M = 0.40, SD = 0.097). We then demonstrate that the spatial layout of cortical networks is influenced by genetics, using our multidimensional estimation of heritability (h2-multi; M = 0.14, SD = 0.015). However, topographic heritability did not differ between heteromodal and unimodal networks. Genetic factors had a regionally variable influence on brain organization, such that the heritability of network topography was greatest in prefrontal, precuneus, and posterior parietal cortex. Taken together, these data are consistent with relaxed genetic control of association cortices relative to primary sensory/motor regions and have implications for understanding population-level variability in brain functioning, guiding both individualized prediction and the interpretation of analyses that integrate genetics and neuroimaging.

Machine Learning Prediction of Stroke Mechanism in Embolic Strokes of Undetermined Source

BACKGROUND AND PURPOSE

One-fifth of ischemic strokes are embolic strokes of undetermined source (ESUS). Their theoretical causes can be classified as cardioembolic versus noncardioembolic. This distinction has important implications, but the categories' proportions are unknown.

METHODS

Using data from the Cornell Acute Stroke Academic Registry, we trained a machine-learning algorithm to distinguish cardioembolic versus non-cardioembolic strokes, then applied the algorithm to ESUS cases to determine the predicted proportion with an occult cardioembolic source. A panel of neurologists adjudicated stroke etiologies using standard criteria. We trained a machine learning classifier using data on demographics, comorbidities, vitals, laboratory results, and echocardiograms. An ensemble predictive method including L1 regularization, gradient-boosted decision tree ensemble (XGBoost), random forests, and multivariate adaptive splines was used. Random search and cross-validation were used to tune hyperparameters. Model performance was assessed using cross-validation among cases of known etiology. We applied the final algorithm to an independent set of ESUS cases to determine the predicted mechanism (cardioembolic or not). To assess our classifier's validity, we correlated the predicted probability of a cardioembolic source with the eventual post-ESUS diagnosis of atrial fibrillation.

RESULTS

Among 1083 strokes with known etiologies, our classifier distinguished cardioembolic versus noncardioembolic cases with excellent accuracy (area under the curve, 0.85). Applied to 580 ESUS cases, the classifier predicted that 44% (95% credibility interval, 39%-49%) resulted from cardiac embolism. Individual ESUS patients' predicted likelihood of cardiac embolism was associated with eventual atrial fibrillation detection (OR per 10% increase, 1.27 [95% CI, 1.03-1.57]; c-statistic, 0.68 [95% CI, 0.58-0.78]). ESUS patients with high predicted probability of cardiac embolism were older and had more coronary and peripheral vascular disease, lower ejection fractions, larger left atria, lower blood pressures, and higher creatinine levels.

CONCLUSIONS

A machine learning estimator that distinguished known cardioembolic versus noncardioembolic strokes indirectly estimated that 44% of ESUS cases were cardioembolic.

Sex classification using long-range temporal dependence of resting-state functional MRI time series

A thorough understanding of sex differences that exist in the brains of healthy individuals is crucial for the study of neurological illnesses that exhibit phenotypic differences between males and females. Here we evaluate sex differences in regional temporal dependence of resting-state brain activity in 195 adult male-female pairs strictly matched for total grey matter volume from the Human Connectome Project. We find that males have more persistent temporal dependence in regions within temporal, parietal, and occipital cortices. Machine learning algorithms trained on regional temporal dependence measures achieve sex classification accuracies up to 81%. Regions with the strongest feature importance in the sex classification task included cerebellum, amygdala, and frontal and occipital cortices. Secondarily, we show that even after strict matching of total gray matter volume, significant volumetric sex differences persist; males have larger absolute cerebella, hippocampi, parahippocampi, thalami, caudates, and amygdalae while females have larger absolute cingulates, precunei, and frontal and parietal cortices. Sex classification based on regional volume achieves accuracies up to 85%, highlighting the importance of strict volume-matching when studying brain-based sex differences. Differential patterns in regional temporal dependence between the sexes identifies a potential neurobiological substrate or environmental effect underlying sex differences in functional brain activation patterns.

The Shared Genetic Basis of Educational Attainment and Cerebral Cortical Morphology

Individual differences in educational attainment are linked to differences in intelligence, and predict important social, economic, and health outcomes. Previous studies have found common genetic factors that influence educational achievement, cognitive performance and total brain volume (i.e., brain size). Here, in a large sample of participants from the UK Biobank, we investigate the shared genetic basis between educational attainment and fine-grained cerebral cortical morphological features, and associate this genetic variation with a related aspect of cognitive ability. Importantly, we execute novel statistical methods that enable high-dimensional genetic correlation analysis, and compute high-resolution surface maps for the genetic correlations between educational attainment and vertex-wise morphological measurements. We conduct secondary analyses, using the UK Biobank verbal-numerical reasoning score, to confirm that variation in educational attainment that is genetically correlated with cortical morphology is related to differences in cognitive performance. Our analyses relate the genetic overlap between cognitive ability and cortical thickness measurements to bilateral primary motor cortex as well as predominantly left superior temporal cortex and proximal regions. These findings extend our understanding of the neurobiology that connects genetic variation to individual differences in educational attainment and cognitive performance.

Fidelity imposed network edit (FINE) for solving ill-posed image reconstruction

Deep learning (DL) is increasingly used to solve ill-posed inverse problems in medical imaging, such as reconstruction from noisy and/or incomplete data, as DL offers advantages over conventional methods that rely on explicit image features and hand engineered priors. However, supervised DL-based methods may achieve poor performance when the test data deviates from the training data, for example, when it has pathologies not encountered in the training data. Furthermore, DL-based image reconstructions do not always incorporate the underlying forward physical model, which may improve performance. Therefore, in this work we introduce a novel approach, called fidelity imposed network edit (FINE), which modifies the weights of a pre-trained reconstruction network for each case in the testing dataset. This is achieved by minimizing an unsupervised fidelity loss function that is based on the forward physical model. FINE is applied to two important inverse problems in neuroimaging: quantitative susceptibility mapping (QSM) and under-sampled image reconstruction in MRI. Our experiments demonstrate that FINE can improve reconstruction accuracy.

Machine Learning Methods Predict Individual Upper-Limb Motor Impairment Following Therapy in Chronic Stroke

Background. Accurate prediction of clinical impairment in upper-extremity motor function following therapy in chronic stroke patients is a difficult task for clinicians but is key in prescribing appropriate therapeutic strategies. Machine learning is a highly promising avenue with which to improve prediction accuracy in clinical practice. Objectives. The objective was to evaluate the performance of 5 machine learning methods in predicting postintervention upper-extremity motor impairment in chronic stroke patients using demographic, clinical, neurophysiological, and imaging input variables. Methods. A total of 102 patients (female: 31%, age 61 ± 11 years) were included. The upper-extremity Fugl-Meyer Assessment (UE-FMA) was used to assess motor impairment of the upper limb before and after intervention. Elastic net (EN), support vector machines, artificial neural networks, classification and regression trees, and random forest were used to predict postintervention UE-FMA. The performances of methods were compared using cross-validated R². Results. EN performed significantly better than other methods in predicting postintervention UE-FMA using demographic and baseline clinical data (median REN2=0.91,RRF2=0.88,RANN2=0.83,RSVM2=0.79,RCART2=0.70; P < .05). Preintervention UE-FMA and the difference in motor threshold (MT) between the affected and unaffected hemispheres were the strongest predictors. The difference in MT had greater importance than the absence or presence of a motor-evoked potential (MEP) in the affected hemisphere. Conclusion. Machine learning methods may enable clinicians to accurately predict a chronic stroke patient's postintervention UE-FMA. Interhemispheric difference in the MT is an important predictor of chronic stroke patients' response to therapy and, therefore, could be included in prospective studies.

Deep neural networks and kernel regression achieve comparable accuracies for functional connectivity prediction of behavior and demographics

There is significant interest in the development and application of deep neural networks (DNNs) to neuroimaging data. A growing literature suggests that DNNs outperform their classical counterparts in a variety of neuroimaging applications, yet there are few direct comparisons of relative utility. Here, we compared the performance of three DNN architectures and a classical machine learning algorithm (kernel regression) in predicting individual phenotypes from whole-brain resting-state functional connectivity (RSFC) patterns. One of the DNNs was a generic fully-connected feedforward neural network, while the other two DNNs were recently published approaches specifically designed to exploit the structure of connectome data. By using a combined sample of almost 10,000 participants from the Human Connectome Project (HCP) and UK Biobank, we showed that the three DNNs and kernel regression achieved similar performance across a wide range of behavioral and demographic measures. Furthermore, the generic feedforward neural network exhibited similar performance to the two state-of-the-art connectome-specific DNNs. When predicting fluid intelligence in the UK Biobank, performance of all algorithms dramatically improved when sample size increased from 100 to 1000 subjects. Improvement was smaller, but still significant, when sample size increased from 1000 to 5000 subjects. Importantly, kernel regression was competitive across all sample sizes. Overall, our study suggests that kernel regression is as effective as DNNs for RSFC-based behavioral prediction, while incurring significantly lower computational costs. Therefore, kernel regression might serve as a useful baseline algorithm for future studies.

Machine learning in resting-state fMRI analysis

Machine learning techniques have gained prominence for the analysis of resting-state functional Magnetic Resonance Imaging (rs-fMRI) data. Here, we present an overview of various unsupervised and supervised machine learning applications to rs-fMRI. We offer a methodical taxonomy of machine learning methods in resting-state fMRI. We identify three major divisions of unsupervised learning methods with regard to their applications to rs-fMRI, based on whether they discover principal modes of variation across space, time or population. Next, we survey the algorithms and rs-fMRI feature representations that have driven the success of supervised subject-level predictions. The goal is to provide a high-level overview of the burgeoning field of rs-fMRI from the perspective of machine learning applications.

Unsupervised Deep Learning for Bayesian Brain MRI Segmentation

Probabilistic atlas priors have been commonly used to derive adaptive and robust brain MRI segmentation algorithms. Widely-used neuroimage analysis pipelines rely heavily on these techniques, which are often computationally expensive. In contrast, there has been a recent surge of approaches that leverage deep learning to implement segmentation tools that are computationally efficient at test time. However, most of these strategies rely on learning from manually annotated images. These supervised deep learning methods are therefore sensitive to the intensity profiles in the training dataset. To develop a deep learning-based segmentation model for a new image dataset (e.g., of different contrast), one usually needs to create a new labeled training dataset, which can be prohibitively expensive, or rely on suboptimal ad hoc adaptation or augmentation approaches. In this paper, we propose an alternative strategy that combines a conventional probabilistic atlas-based segmentation with deep learning, enabling one to train a segmentation model for new MRI scans without the need for any manually segmented images. Our experiments include thousands of brain MRI scans and demonstrate that the proposed method achieves good accuracy for a brain MRI segmentation task for different MRI contrasts, requiring only approximately 15 seconds at test time on a GPU.

Multi-modal latent factor exploration of atrophy, cognitive and tau heterogeneity in Alzheimer's disease

Individuals with Alzheimer's disease (AD) dementia exhibit significant heterogeneity across clinical symptoms, atrophy patterns, and spatial distribution of Tau deposition. Most previous studies of AD heterogeneity have focused on atypical clinical subtypes, defined subtypes with a single modality, or restricted their analyses to a priori brain regions and cognitive tests. Here, we considered a data-driven hierarchical Bayesian model to identify latent factors from atrophy patterns and cognitive deficits simultaneously, thus exploiting the rich dimensionality within each modality. Unlike most previous studies, our model allows each factor to be expressed to varying degrees within an individual, in order to reflect potential multiple co-existing pathologies. By applying our model to ADNI-GO/2 AD dementia participants, we found three atrophy-cognitive factors. The first factor was associated with medial temporal lobe atrophy, episodic memory deficits and disorientation to time/place ("MTL-Memory"). The second factor was associated with lateral temporal atrophy and language deficits ("Lateral Temporal-Language"). The third factor was associated with atrophy in posterior bilateral cortex, and visuospatial executive function deficits ("Posterior Cortical-Executive"). While the MTL-Memory and Posterior Cortical-Executive factors were discussed in previous literature, the Lateral Temporal-Language factor is novel and emerged only by considering atrophy and cognition jointly. Several analyses were performed to ensure generalizability, replicability and stability of the estimated factors. First, the factors generalized to new participants within a 10-fold cross-validation of ADNI-GO/2 AD dementia participants. Second, the factors were replicated in an independent ADNI-1 AD dementia cohort. Third, factor loadings of ADNI-GO/2 AD dementia participants were longitudinally stable, suggesting that these factors capture heterogeneity across patients, rather than longitudinal disease progression. Fourth, the model outperformed canonical correlation analysis at capturing associations between atrophy patterns and cognitive deficits. To explore the influence of the factors early in the disease process, factor loadings were estimated in ADNI-GO/2 mild cognitively impaired (MCI) participants. Although the associations between the atrophy patterns and cognitive profiles were weak in MCI compared to AD, we found that factor loadings were associated with inter-individual regional variation in Tau uptake. Taken together, these results suggest that distinct atrophy-cognitive patterns exist in typical Alzheimer's disease, and are associated with distinct patterns of Tau depositions before clinical dementia emerges.

Ensemble learning with 3D convolutional neural networks for functional connectome-based prediction

The specificity and sensitivity of resting state functional MRI (rs-fMRI) measurements depend on preprocessing choices, such as the parcellation scheme used to define regions of interest (ROIs). In this study, we critically evaluate the effect of brain parcellations on machine learning models applied to rs-fMRI data. Our experiments reveal an intriguing trend: On average, models with stochastic parcellations consistently perform as well as models with widely used atlases at the same spatial scale. We thus propose an ensemble learning strategy to combine the predictions from models trained on connectivity data extracted using different (e.g., stochastic) parcellations. We further present an implementation of our ensemble learning strategy with a novel 3D Convolutional Neural Network (CNN) approach. The proposed CNN approach takes advantage of the full-resolution 3D spatial structure of rs-fMRI data and fits non-linear predictive models. Our ensemble CNN framework overcomes the limitations of traditional machine learning models for connectomes that often rely on region-based summary statistics and/or linear models. We showcase our approach on a classification (autism patients versus healthy controls) and a regression problem (prediction of subject's age), and report promising results.

Spatial Topography of Individual-Specific Cortical Networks Predicts Human Cognition, Personality, and Emotion

Resting-state functional magnetic resonance imaging (rs-fMRI) offers the opportunity to delineate individual-specific brain networks. A major question is whether individual-specific network topography (i.e., location and spatial arrangement) is behaviorally relevant. Here, we propose a multi-session hierarchical Bayesian model (MS-HBM) for estimating individual-specific cortical networks and investigate whether individual-specific network topography can predict human behavior. The multiple layers of the MS-HBM explicitly differentiate intra-subject (within-subject) from inter-subject (between-subject) network variability. By ignoring intra-subject variability, previous network mappings might confuse intra-subject variability for inter-subject differences. Compared with other approaches, MS-HBM parcellations generalized better to new rs-fMRI and task-fMRI data from the same subjects. More specifically, MS-HBM parcellations estimated from a single rs-fMRI session (10 min) showed comparable generalizability as parcellations estimated by 2 state-of-the-art methods using 5 sessions (50 min). We also showed that behavioral phenotypes across cognition, personality, and emotion could be predicted by individual-specific network topography with modest accuracy, comparable to previous reports predicting phenotypes based on connectivity strength. Network topography estimated by MS-HBM was more effective for behavioral prediction than network size, as well as network topography estimated by other parcellation approaches. Thus, similar to connectivity strength, individual-specific network topography might also serve as a fingerprint of human behavior.

Resting brain dynamics at different timescales capture distinct aspects of human behavior

Linking human behavior to resting-state brain function is a central question in systems neuroscience. In particular, the functional timescales at which different types of behavioral factors are encoded remain largely unexplored. The behavioral counterparts of static functional connectivity (FC), at the resolution of several minutes, have been studied but behavioral correlates of dynamic measures of FC at the resolution of a few seconds remain unclear. Here, using resting-state fMRI and 58 phenotypic measures from the Human Connectome Project, we find that dynamic FC captures task-based phenotypes (e.g., processing speed or fluid intelligence scores), whereas self-reported measures (e.g., loneliness or life satisfaction) are equally well explained by static and dynamic FC. Furthermore, behaviorally relevant dynamic FC emerges from the interconnections across all resting-state networks, rather than within or between pairs of networks. Our findings shed new light on the timescales of cognitive processes involved in distinct facets of behavior.

Joint Analysis of Cortical Area and Thickness as a Replacement for the Analysis of the Volume of the Cerebral Cortex

Cortical surface area is an increasingly used brain morphology metric that is ontogenetically and phylogenetically distinct from cortical thickness and offers a separate index of neurodevelopment and disease. However, the various existing methods for assessment of cortical surface area from magnetic resonance images have never been systematically compared. We show that the surface area method implemented in FreeSurfer corresponds closely to the exact, but computationally more demanding, mass-conservative (pycnophylactic) method, provided that images are smoothed. Thus, the data produced by this method can be interpreted as estimates of cortical surface area, as opposed to areal expansion. In addition, focusing on the joint analysis of thickness and area, we compare an improved, analytic method for measuring cortical volume to a permutation-based nonparametric combination (NPC) method. We use the methods to analyze area, thickness and volume in young adults born preterm with very low birth weight, and show that NPC analysis is a more sensitive option for studying joint effects on area and thickness, giving equal weight to variation in both of these 2 morphological features.

Global signal regression strengthens association between resting-state functional connectivity and behavior

Global signal regression (GSR) is one of the most debated preprocessing strategies for resting-state functional MRI. GSR effectively removes global artifacts driven by motion and respiration, but also discards globally distributed neural information and introduces negative correlations between certain brain regions. The vast majority of previous studies have focused on the effectiveness of GSR in removing imaging artifacts, as well as its potential biases. Given the growing interest in functional connectivity fingerprinting, here we considered the utilitarian question of whether GSR strengthens or weakens associations between resting-state functional connectivity (RSFC) and multiple behavioral measures across cognition, personality and emotion. By applying the variance component model to the Brain Genomics Superstruct Project (GSP), we found that behavioral variance explained by whole-brain RSFC increased by an average of 47% across 23 behavioral measures after GSR. In the Human Connectome Project (HCP), we found that behavioral variance explained by whole-brain RSFC increased by an average of 40% across 58 behavioral measures, when GSR was applied after ICA-FIX de-noising. To ensure generalizability, we repeated our analyses using kernel regression. GSR improved behavioral prediction accuracies by an average of 64% and 12% in the GSP and HCP datasets respectively. Importantly, the results were consistent across methods. A behavioral measure with greater RSFC-explained variance (using the variance component model) also exhibited greater prediction accuracy (using kernel regression). A behavioral measure with greater improvement in behavioral variance explained after GSR (using the variance component model) also enjoyed greater improvement in prediction accuracy after GSR (using kernel regression). Furthermore, GSR appeared to benefit task performance measures more than self-reported measures. Since GSR was more effective at removing motion-related and respiratory-related artifacts, GSR-related increases in variance explained and prediction accuracies were unlikely the result of motion-related or respiratory-related artifacts. However, it is worth emphasizing that the current study focused on whole-brain RSFC, so it remains unclear whether GSR improves RSFC-behavioral associations for specific connections or networks. Overall, our results suggest that at least in the case for young healthy adults, GSR strengthens the associations between RSFC and most (although not all) behavioral measures. Code for the variance component model and ridge regression can be found here: https://github.com/ThomasYeoLab/CBIG/tree/master/stable_projects/preprocessing/Li2019_GSR.

A CONVOLUTIONAL AUTOENCODER APPROACH TO LEARN VOLUMETRIC SHAPE REPRESENTATIONS FOR BRAIN STRUCTURES

We propose a novel machine learning strategy for studying neuroanatomical shape variation. Our model works with volumetric binary segmentation images, and requires no preprocessing such as the extraction of surface points or a mesh. The learned shape descriptor is invariant to affine transformations, including shifts, rotations and scaling. Thanks to the adopted autoencoder framework, inter-subject differences are automatically enhanced in the learned representation, while intra-subject variances are minimized. Our experimental results on a shape retrieval task showed that the proposed representation outperforms a state-of-the-art benchmark for brain structures extracted from MRI scans.

Deep convolutional neural networks for segmenting 3D in vivo multiphoton images of vasculature in Alzheimer disease mouse models

The health and function of tissue rely on its vasculature network to provide reliable blood perfusion. Volumetric imaging approaches, such as multiphoton microscopy, are able to generate detailed 3D images of blood vessels that could contribute to our understanding of the role of vascular structure in normal physiology and in disease mechanisms. The segmentation of vessels, a core image analysis problem, is a bottleneck that has prevented the systematic comparison of 3D vascular architecture across experimental populations. We explored the use of convolutional neural networks to segment 3D vessels within volumetric in vivo images acquired by multiphoton microscopy. We evaluated different network architectures and machine learning techniques in the context of this segmentation problem. We show that our optimized convolutional neural network architecture with a customized loss function, which we call DeepVess, yielded a segmentation accuracy that was better than state-of-the-art methods, while also being orders of magnitude faster than the manual annotation. To explore the effects of aging and Alzheimer's disease on capillaries, we applied DeepVess to 3D images of cortical blood vessels in young and old mouse models of Alzheimer's disease and wild type littermates. We found little difference in the distribution of capillary diameter or tortuosity between these groups, but did note a decrease in the number of longer capillary segments (>75μm) in aged animals as compared to young, in both wild type and Alzheimer's disease mouse models.

VoxelMorph: A Learning Framework for Deformable Medical Image Registration

We present VoxelMorph, a fast learning-based framework for deformable, pairwise medical image registration. Traditional registration methods optimize an objective function for each pair of images, which can be time-consuming for large datasets or rich deformation models. In contrast to this approach, and building on recent learning-based methods, we formulate registration as a function that maps an input image pair to a deformation field that aligns these images. We parameterize the function via a convolutional neural network (CNN), and optimize the parameters of the neural network on a set of images. Given a new pair of scans, VoxelMorph rapidly computes a deformation field by directly evaluating the function. In this work, we explore two different training strategies. In the first (unsupervised) setting, we train the model to maximize standard image matching objective functions that are based on the image intensities. In the second setting, we leverage auxiliary segmentations available in the training data. We demonstrate that the unsupervised model's accuracy is comparable to state-of-the-art methods, while operating orders of magnitude faster. We also show that VoxelMorph trained with auxiliary data improves registration accuracy at test time, and evaluate the effect of training set size on registration. Our method promises to speed up medical image analysis and processing pipelines, while facilitating novel directions in learning-based registration and its applications. Our code is freely available at https://github.com/voxelmorph/voxelmorph.

A Bayesian Disease Progression Model for Clinical Trajectories

In this work, we consider the problem of predicting the course of a progressive disease, such as cancer or Alzheimer's. Progressive diseases often start with mild symptoms that might precede a diagnosis, and each patient follows their own trajectory. Patient trajectories exhibit wild variability, which can be associated with many factors such as geno-type, age, or sex. An additional layer of complexity is that, in real life, the amount and type of data available for each patient can differ significantly. For example, for one patient we might have no prior history, whereas for another patient we might have detailed clinical assessments obtained at multiple prior time-points. This paper presents a probabilistic model that can handle multiple modalities (including images and clinical assessments) and variable patient histories with irregular timings and missing entries, to predict clinical scores at future time-points. We use a sigmoidal function to model latent disease progression, which gives rise to clinical observations in our generative model. We implemented an approximate Bayesian inference strategy on the proposed model to estimate the parameters on data from a large population of subjects. Furthermore, the Bayesian framework enables the model to automatically fine-tune its predictions based on historical observations that might be available on the test subject. We applied our method to a longitudinal Alzheimer's disease dataset with more than 3,000 subjects [1] with comparisons against several benchmarks.

Medical Image Imputation from Image Collections

We present an algorithm for creating high resolution anatomically plausible images consistent with acquired clinical brain MRI scans with large inter-slice spacing. Although large data sets of clinical images contain a wealth of information, time constraints during acquisition result in sparse scans that fail to capture much of the anatomy. These characteristics often render computational analysis impractical as many image analysis algorithms tend to fail when applied to such images. Highly specialized algorithms that explicitly handle sparse slice spacing do not generalize well across problem domains. In contrast, we aim to enable application of existing algorithms that were originally developed for high resolution research scans to significantly undersampled scans. We introduce a generative model that captures fine-scale anatomical structure across subjects in clinical image collections and derive an algorithm for filling in the missing data in scans with large inter-slice spacing. Our experimental results demonstrate that the resulting method outperforms state-of-the-art upsampling super-resolution techniques, and promises to facilitate subsequent analysis not previously possible with scans of this quality. Our implementation is freely available at https://github.com/adalca/papago.

Dissociable influences of APOE ε4 and polygenic risk of AD dementia on amyloid and cognition

OBJECTIVE

To investigate the effects of genetic risk of Alzheimer disease (AD) dementia in the context of β-amyloid (Aβ) accumulation.

METHODS

We analyzed data from 702 participants (221 clinically normal, 367 with mild cognitive impairment, and 114 with AD dementia) with genetic data and florbetapir PET available. A subset of 669 participants additionally had longitudinal MRI scans to assess hippocampal volume. Polygenic risk scores (PRSs) were estimated with summary statistics from previous large-scale genome-wide association studies of AD dementia. We examined relationships between APOE ε4 status and PRS with longitudinal Aβ and cognitive and hippocampal volume measurements.

RESULTS

APOE ε4 was strongly related to baseline Aβ, whereas only weak associations between PRS and baseline Aβ were present. APOE ε4 was additionally related to greater memory decline and hippocampal atrophy in Aβ+ participants. When APOE ε4 was controlled for, PRS was related to cognitive decline in Aβ+ participants. Finally, PRSs were associated with hippocampal atrophy in Aβ- participants and weakly associated with baseline hippocampal volume in Aβ+ participants.

CONCLUSIONS

Genetic risk factors of AD dementia demonstrate effects related to Aβ, as well as synergistic interactions with Aβ. The specific effect of faster cognitive decline in Aβ+ individuals with higher genetic risk may explain the large degree of heterogeneity in cognitive trajectories among Aβ+ individuals. Consideration of genetic variants in conjunction with baseline Aβ may improve enrichment strategies for clinical trials targeting Aβ+ individuals most at risk for imminent cognitive decline.

The human cortex possesses a reconfigurable dynamic network architecture that is disrupted in psychosis

Higher-order cognition emerges through the flexible interactions of large-scale brain networks, an aspect of temporal coordination that may be impaired in psychosis. Here, we map the dynamic functional architecture of the cerebral cortex in healthy young adults, leveraging this atlas of transient network configurations (states), to identify state- and network-specific disruptions in patients with schizophrenia and psychotic bipolar disorder. We demonstrate that dynamic connectivity profiles are reliable within participants, and can act as a fingerprint, identifying specific individuals within a larger group. Patients with psychotic illness exhibit intermittent disruptions within cortical networks previously associated with the disease, and the individual connectivity profiles within specific brain states predict the presence of active psychotic symptoms. Taken together, these results provide evidence for a reconfigurable dynamic architecture in the general population and suggest that prior reports of network disruptions in psychosis may reflect symptom-relevant transient abnormalities, rather than a time-invariant global deficit.

Temporal changes in brain dynamics are linked with cognitive abilities, but neither their stability nor relationship to psychosis is clear. Here, authors describe the dynamic neural architecture in healthy controls and patients with psychosis and find that they are stable over time and can predict psychotic symptoms.

Polygenic Risk of Spasmodic Dysphonia is Associated With Vulnerable Sensorimotor Connectivity

Spasmodic dysphonia (SD), or laryngeal dystonia, is an isolated task-specific dystonia of unknown causes and pathophysiology that selectively affects speech production. Using next-generation whole-exome sequencing in SD patients, we computed polygenic risk score from 1804 genetic markers based on a genome-wide association study in another form of similar task-specific focal dystonia, musician's dystonia. We further examined the associations between the polygenic risk score, resting-state functional connectivity abnormalities within the sensorimotor network, and SD clinical characteristics. We found that the polygenic risk of dystonia was significantly associated with decreased functional connectivity in the left premotor/primary sensorimotor and inferior parietal cortices in SD patients. Reduced connectivity of the inferior parietal cortex was correlated with the age of SD onset. The polygenic risk score contained a significant number of genetic variants lying near genes related to synaptic transmission and neural development. Our study identified a polygenic contribution to the overall genetic risk of dystonia in the cohort of SD patients. Associations between the polygenic risk and reduced functional connectivity of the sensorimotor and inferior parietal cortices likely represent an endophenotypic imaging marker of SD, while genes involved in synaptic transmission and neuron development may be linked to the molecular pathophysiology of this disorder.

Population Based Image Imputation

We present an algorithm for creating high resolution anatomically plausible images consistent with acquired clinical brain MRI scans with large inter-slice spacing. Although large databases of clinical images contain a wealth of information, medical acquisition constraints result in sparse scans that miss much of the anatomy. These characteristics often render computational analysis impractical as standard processing algorithms tend to fail when applied to such images. Highly specialized or application-specific algorithms that explicitly handle sparse slice spacing do not generalize well across problem domains. In contrast, our goal is to enable application of existing algorithms that were originally developed for high resolution research scans to significantly undersampled scans. We introduce a model that captures fine-scale anatomical similarity across subjects in clinical image collections and use it to fill in the missing data in scans with large slice spacing. Our experimental results demonstrate that the proposed method outperforms current upsampling methods and promises to facilitate subsequent analysis not previously possible with scans of this quality.

Diffeomorphic functional brain surface alignment: Functional demons

Aligning brain structures across individuals is a central prerequisite for comparative neuroimaging studies. Typically, registration approaches assume a strong association between the features used for alignment, such as macro-anatomy, and the variable observed, such as functional activation or connectivity. Here, we propose to use the structure of intrinsic resting state fMRI signal correlation patterns as a basis for alignment of the cortex in functional studies. Rather than assuming the spatial correspondence of functional structures between subjects, we have identified locations with similar connectivity profiles across subjects. We mapped functional connectivity relationships within the brain into an embedding space, and aligned the resulting maps of multiple subjects. We then performed a diffeomorphic alignment of the cortical surfaces, driven by the corresponding features in the joint embedding space. Results show that functional alignment based on resting state fMRI identifies functionally homologous regions across individuals with higher accuracy than alignment based on the spatial correspondence of anatomy. Further, functional alignment enables measurement of the strength of the anatomo-functional link across the cortex, and reveals the uneven distribution of this link. Stronger anatomo-functional dissociation was found in higher association areas compared to primary sensory- and motor areas. Functional alignment based on resting state features improves group analysis of task based functional MRI data, increasing statistical power and improving the delineation of task-specific core regions. Finally, a comparison of the anatomo-functional dissociation between cohorts is demonstrated with a group of left and right handed subjects.