A Predictive Model for Severe COVID-19 in the Medicare Population: A Tool for Prioritizing Primary and Booster COVID-19 Vaccination.

Recommendations for prioritizing COVID-19 vaccination have focused on the elderly at higher risk for severe disease. Existing models for identifying higher-risk individuals lack the needed integration of socio-demographic and clinical risk factors. Using multivariate logistic regression and random forest modeling, we developed a predictive model of severe COVID-19 using clinical data from Medicare claims for 16 million Medicare beneficiaries and socio-economic data from the CDC Social Vulnerability Index. Predicted individual probabilities of COVID-19 hospitalization were then calculated for population risk stratification and vaccine prioritization and mapping. The leading COVID-19 hospitalization risk factors were non-white ethnicity, end-stage renal disease, advanced age, prior hospitalization, leukemia, morbid obesity, chronic kidney disease, lung cancer, chronic liver disease, pulmonary fibrosis or pulmonary hypertension, and chemotherapy. However, previously reported risk factors such as chronic obstructive pulmonary disease and diabetes conferred modest hospitalization risk. Among all social vulnerability factors, residence in a low-income zip code was the only risk factor independently predicting hospitalization. This multifactor risk model and its population risk dashboard can be used to optimize COVID-19 vaccine allocation in the higher-risk Medicare population.

Using Deep Learning to Identify High-Risk Patients with Heart Failure with Reduced Ejection Fraction.

Background: Deep Learning (DL) has not been well-established as a method to identify high-risk patients among patients with heart failure (HF). Objectives: This study aimed to use DL models to predict hospitalizations, worsening HF events, and 30-day and 90-day readmissions in patients with heart failure with reduced ejection fraction (HFrEF). Methods: We analyzed the data of adult HFrEF patients from the IBM® MarketScan® Commercial and Medicare Supplement databases between January 1, 2015 and December 31, 2017. A sequential model architecture based on bi-directional long short-term memory (Bi-LSTM) layers was utilized. For DL models to predict HF hospitalizations and worsening HF events, we utilized two study designs: with and without a buffer window. For comparison, we also tested multiple traditional machine learning models including logistic regression, random forest, and eXtreme Gradient Boosting (XGBoost). Model performance was assessed by area under the curve (AUC) values, precision, and recall on an independent testing dataset. Results: A total of 47 498 HFrEF patients were included; 9427 with at least one HF hospitalization. The best AUCs of DL models without a buffer window in predicting HF hospitalizations and worsening HF events in the total patient cohort were 0.977 and 0.972; with a 7-day buffer window the best AUCs were 0.573 and 0.608, respectively. The best AUCs in predicting 30- and 90-day readmissions in all adult patients were 0.597 and 0.614, respectively. An AUC of 0.861 was attained for prediction of 90-day readmission in patients aged 18-64. For all outcomes assessed, the DL approach outperformed traditional machine learning models. Discussion: The DL approach can automate feature engineering during the model learning, which can increase the clinical applicability and lead to comparable or better model performance. However, the lack of granular clinical data, and sample size and imbalance issues may have limited the model's performance. Conclusions: A DL approach using Bi-LSTM was shown to be a feasible and useful tool to predict HF-related outcomes. This study can help inform the future development and deployment of predictive tools to identify high-risk HFrEF patients and ultimately facilitate targeted interventions in clinical practice.

Seeing faces is necessary for face-domain formation.

Here we report that monkeys raised without exposure to faces did not develop face domains, but did develop domains for other categories and did show normal retinotopic organization, indicating that early face deprivation leads to a highly selective cortical processing deficit. Therefore, experience must be necessary for the formation (or maintenance) of face domains. Gaze tracking revealed that control monkeys looked preferentially at faces, even at ages prior to the emergence of face domains, but face-deprived monkeys did not, indicating that face looking is not innate. A retinotopic organization is present throughout the visual system at birth, so selective early viewing behavior could bias category-specific visual responses toward particular retinotopic representations, thereby leading to domain formation in stereotyped locations in inferotemporal cortex, without requiring category-specific templates or biases. Thus, we propose that environmental importance influences viewing behavior, viewing behavior drives neuronal activity, and neuronal activity sculpts domain formation.

Development of the macaque face-patch system.

Face recognition is highly proficient in humans and other social primates; it emerges in infancy, but the development of the neural mechanisms supporting this behaviour is largely unknown. We use blood-volume functional MRI to monitor longitudinally the responsiveness to faces, scrambled faces, and objects in macaque inferotemporal cortex (IT) from 1 month to 2 years of age. During this time selective responsiveness to monkey faces emerges. Some functional organization is present at 1 month; face-selective patches emerge over the first year of development, and are remarkably stable once they emerge. Face selectivity is refined by a decreasing responsiveness to non-face stimuli.

Novel domain formation reveals proto-architecture in inferotemporal cortex.

Primate inferotemporal cortex is subdivided into domains for biologically important categories, such as faces, bodies and scenes, as well as domains for culturally entrained categories, such as text or buildings. These domains are in stereotyped locations in most humans and monkeys. To ask what determines the locations of such domains, we intensively trained seven juvenile monkeys to recognize three distinct sets of shapes. After training, the monkeys developed regions that were selectively responsive to each trained set. The location of each specialization was similar across monkeys, despite differences in training order. This indicates that the location of training effects does not depend on function or expertise, but rather on some kind of proto-organization. We explore the possibility that this proto-organization is retinotopic or shape-based.

Modulation of the brain's functional network architecture in the transition from wake to sleep.

The transition from quiet wakeful rest to sleep represents a period over which attention to the external environment fades. Neuroimaging methodologies have provided much information on the shift in neural activity patterns in sleep, but the dynamic restructuring of human brain networks in the transitional period from wake to sleep remains poorly understood. Analysis of electrophysiological measures and functional network connectivity of these early transitional states shows subtle shifts in network architecture that are consistent with reduced external attentiveness and increased internal and self-referential processing. Further, descent to sleep is accompanied by the loss of connectivity in anterior and posterior portions of the default-mode network and more locally organized global network architecture. These data clarify the complex and dynamic nature of the transitional period between wake and sleep and suggest the need for more studies investigating the dynamics of these processes.

Functional connectivity of the macaque posterior parahippocampal cortex.

Neuroimaging experiments in humans suggest that regions in parietal cortex and along the posterior midline are functionally connected to the medial temporal lobe and are active during memory retrieval. It is unknown whether macaques have a similar network. We examined functional connectivity in isoflurane-anesthetized macaques to identify a network associated with posterior parahippocampal cortex (PPHC). Functional connectivity was observed between the PPHC and retrosplenial, posterior cingulate, superior temporal gyrus, and inferior parietal cortex. PPHC correlations were distinct from regions in parietal and temporal cortex activated by an oculomotor task. Comparison of macaque and human PPHC correlations revealed similarities that suggest the temporal-parietal region identified in the macaque may share a common lineage with human Brodmann area 39, a region thought to be involved in recollection. These results suggest that macaques and humans may have homologous PPHC-parietal pathways. By specifying the location of the putative macaque homologue in parietal cortex, we provide a target for future physiological exploration of this area's role in mnemonic or alternative processes.

Precuneus shares intrinsic functional architecture in humans and monkeys.

Evidence from macaque monkey tracing studies suggests connectivity-based subdivisions within the precuneus, offering predictions for similar subdivisions in the human. Here we present functional connectivity analyses of this region using resting-state functional MRI data collected from both humans and macaque monkeys. Three distinct patterns of functional connectivity were demonstrated within the precuneus of both species, with each subdivision suggesting a discrete functional role: (i) the anterior precuneus, functionally connected with the superior parietal cortex, paracentral lobule, and motor cortex, suggesting a sensorimotor region; (ii) the central precuneus, functionally connected to the dorsolateral prefrontal, dorsomedial prefrontal, and multimodal lateral inferior parietal cortex, suggesting a cognitive/associative region; and (iii) the posterior precuneus, displaying functional connectivity with adjacent visual cortical regions. These functional connectivity patterns were differentiated from the more ventral networks associated with the posterior cingulate, which connected with limbic structures such as the medial temporal cortex, dorsal and ventromedial prefrontal regions, posterior lateral inferior parietal regions, and the lateral temporal cortex. Our findings are consistent with predictions from anatomical tracer studies in the monkey, and provide support that resting-state functional connectivity (RSFC) may in part reflect underlying anatomy. These subdivisions within the precuneus suggest that neuroimaging studies will benefit from treating this region as anatomically (and thus functionally) heterogeneous. Furthermore, the consistency between functional connectivity networks in monkeys and humans provides support for RSFC as a viable tool for addressing cross-species comparisons of functional neuroanatomy.

Learning and memory: while you rest, your brain keeps working.

A recent study shows that brain activity recorded while the human subject is at 'rest' is significantly affected by a prior learning episode. These results suggest that understanding resting brain activity may be critical to understanding how humans learn from experience.

Evidence for a frontoparietal control system revealed by intrinsic functional connectivity.

Two functionally distinct, and potentially competing, brain networks have been recently identified that can be broadly distinguished by their contrasting roles in attention to the external world versus internally directed mentation involving long-term memory. At the core of these two networks are the dorsal attention system and the hippocampal-cortical memory system, a component of the brain's default network. Here spontaneous blood-oxygenation-level-dependent (BOLD) signal correlations were used in three separate functional magnetic resonance imaging data sets (n = 105) to define a third system, the frontoparietal control system, which is spatially interposed between these two previously defined systems. The frontoparietal control system includes many regions identified as supporting cognitive control and decision-making processes including lateral prefrontal cortex, anterior cingulate cortex, and inferior parietal lobule. Detailed analysis of frontal and parietal cortex, including use of high-resolution data, revealed clear evidence for contiguous but distinct regions: in general, the regions associated with the frontoparietal control system are situated between components of the dorsal attention and hippocampal-cortical memory systems. The frontoparietal control system is therefore anatomically positioned to integrate information from these two opposing brain systems.

Distinct cortical anatomy linked to subregions of the medial temporal lobe revealed by intrinsic functional connectivity.

The hippocampus and adjacent cortical structures in the medial temporal lobe (MTL) contribute to memory through interactions with distributed brain areas. Studies of monkey and rodent anatomy suggest that parallel pathways converge on distinct subregions of the MTL. To explore the cortical areas linked to subregions of the MTL in humans, we examined cortico-cortical and hippocampal-cortical correlations using high-resolution, functional connectivity analysis in 100 individuals. MTL seed regions extended along the anterior to posterior axis and included hippocampus and adjacent structures. Results revealed two separate brain pathways that correlated with distinct subregions within the MTL. The body of the hippocampus and posterior parahippocampal cortex correlated with lateral parietal cortex, regions along the posterior midline including posterior cingulate and retrosplenial cortex, and ventral medial prefrontal cortex. By contrast, anterior hippocampus and the perirhinal/entorhinal cortices correlated with distinct regions in the lateral temporal cortex extending into the temporal pole. The present results are largely consistent with known connectivity in the monkey and provide a novel task-independent dissociation of the parallel pathways supporting the MTL memory system in humans. The cortical pathways include regions that have undergone considerable areal expansion in humans, providing insight into how the MTL memory system has evolved to support a diverse array of cognitive domains.

Disruption of large-scale brain systems in advanced aging.

Cognitive decline is commonly observed in advanced aging even in the absence of disease. Here we explore the possibility that normal aging is accompanied by disruptive alterations in the coordination of large-scale brain systems that support high-level cognition. In 93 adults aged 18 to 93, we demonstrate that aging is characterized by marked reductions in normally present functional correlations within two higher-order brain systems. Anterior to posterior components within the default network were most severely disrupted with age. Furthermore, correlation reductions were severe in older adults free from Alzheimer's disease (AD) pathology as determined by amyloid imaging, suggesting that functional disruptions were not the result of AD. Instead, reduced correlations were associated with disruptions in white matter integrity and poor cognitive performance across a range of domains. These results suggest that cognitive decline in normal aging arises from functional disruption in the coordination of large-scale brain systems that support cognition.

Intrinsic fluctuations within cortical systems account for intertrial variability in human behavior.

The resting brain is not silent, but exhibits organized fluctuations in neuronal activity even in the absence of tasks or stimuli. This intrinsic brain activity persists during task performance and contributes to variability in evoked brain responses. What is unknown is if this intrinsic activity also contributes to variability in behavior. In the current fMRI study, we identify a relationship between human brain activity in the left somatomotor cortex and spontaneous trial-to-trial variability in button press force. We then demonstrate that 74% of this brain-behavior relationship is attributable to ongoing fluctuations in intrinsic activity similar to those observed during resting fixation. In addition to establishing a functional and behavioral significance of intrinsic brain activity, these results lend new insight into the origins of variability in human behavior.

Distinct brain networks for adaptive and stable task control in humans.

Control regions in the brain are thought to provide signals that configure the brain's moment-to-moment information processing. Previously, we identified regions that carried signals related to task-control initiation, maintenance, and adjustment. Here we characterize the interactions of these regions by applying graph theory to resting state functional connectivity MRI data. In contrast to previous, more unitary models of control, this approach suggests the presence of two distinct task-control networks. A frontoparietal network included the dorsolateral prefrontal cortex and intraparietal sulcus. This network emphasized start-cue and error-related activity and may initiate and adapt control on a trial-by-trial basis. The second network included dorsal anterior cingulate/medial superior frontal cortex, anterior insula/frontal operculum, and anterior prefrontal cortex. Among other signals, these regions showed activity sustained across the entire task epoch, suggesting that this network may control goal-directed behavior through the stable maintenance of task sets. These two independent networks appear to operate on different time scales and affect downstream processing via dissociable mechanisms.

Breakdown of functional connectivity in frontoparietal networks underlies behavioral deficits in spatial neglect.

Spatial neglect is a syndrome following stroke manifesting attentional deficits in perceiving and responding to stimuli in the contralesional field. We examined brain network integrity in patients with neglect by measuring coherent fluctuations of fMRI signals (functional connectivity). Connectivity in two largely separate attention networks located in dorsal and ventral frontoparietal areas was assessed at both acute and chronic stages of recovery. Connectivity in the ventral network, part of which directly lesioned, was diffusely disrupted and showed no recovery. In the structurally intact dorsal network, interhemispheric connectivity in posterior parietal cortex was acutely disrupted but fully recovered. This acute disruption, and disrupted connectivity in specific pathways in the ventral network, strongly correlated with impaired attentional processing across subjects. Lastly, disconnection of the white matter tracts connecting frontal and parietal cortices was associated with more severe neglect and more disrupted functional connectivity. These findings support a network view in understanding neglect.

Unrest at rest: default activity and spontaneous network correlations.

A series of recent empirical observations demonstrate structured activity patterns that exist during passive task states. One observation is that a network of regions, referred to as the default network, shows preferentially greater activity during passive task states as compared to a wide range of active tasks. The second observation is that distributed regions spontaneously increase and decrease their activity together within functional-anatomic networks, even under anesthesia. We believe these rest activity patterns may reflect neural functions that consolidate the past, stabilize brain ensembles, and prepare us for the future. Accumulating data further suggest that differences in rest activity may be relevant to understanding clinical conditions such as Alzheimer's disease and autism. Maps of spontaneous network correlations also provide tools for functional localization and study of comparative anatomy between primate species. For all of these reasons, we advocate the systematic exploration of rest activity.

Moving GLM ballistocardiogram artifact reduction for EEG acquired simultaneously with fMRI.

OBJECTIVE: Simultaneous acquisition of electroencephalogram (EEG) and functional magnetic resonance imaging (fMRI) enables studies of brain activity at both high temporal and high spatial resolution. However, EEG acquired in a magnetic field is contaminated by ballistocardiogram (BKG) artifact. The most commonly used method of BKG artifact reduction, averaged artifact subtraction (AAS), was not designed to account for overlapping BKG waveforms generated by adjacent beats. We describe a new method based on a moving general linear model (mGLM) that accounts for overlapping BKG waveforms. METHODS: Simultaneous EEG-fMRI at 3 Tesla was performed in nine normal human subjects (8-11 runs/subject, 5.52 min/run). Gradient switching artifact was effectively reduced using commercially supplied procedures. Cardiac beats were detected using a novel correlation detector algorithm applied to the EKG trace. BKG artifact was reduced using both mGLM and AAS. RESULTS: mGLM recovered BKG waveforms outlasting the median inter-beat interval. mGLM more effectively than AAS removed variance in the EEG attributable to BKG artifact. CONCLUSIONS: mGLM offers advantages over AAS especially in the presence of variable heart rate. SIGNIFICANCE: The BKG artifact reduction procedure described herein improves the technique of simultaneous EEG-fMRI. Potential applications include basic investigations of the relationship between scalp potentials and functional imaging signals as well as clinical localization of epileptic foci.

Coherent spontaneous activity identifies a hippocampal-parietal memory network.

Despite traditional theories emphasizing parietal contributions to spatial attention and sensory-motor integration, functional MRI (fMRI) experiments in normal subjects suggest that specific regions within parietal cortex may also participate in episodic memory. Here we examined correlations in spontaneous blood-oxygenation-level-dependent (BOLD) signal fluctuations in a resting state to identify the network associated with the hippocampal formation (HF) and determine whether parietal regions were elements of that network. In the absence of task, stimuli, or explicit mnemonic demands, robust correlations were observed between activity in the HF and several parietal regions (including precuneus, posterior cingulate, retrosplenial cortex, and bilateral inferior parietal lobule). These HF-correlated regions in parietal cortex were spatially distinct from those correlated with the motion-sensitive MT+ complex. Reanalysis of event-related fMRI studies of recognition memory showed that the regions spontaneously correlated with the HF (but not MT+) were also modulated during directed recollection. These regions showed greater activity to successfully recollected items as compared with other trial types. Together, these results associate specific regions of parietal cortex that are sensitive to successful recollection with the HF.

Spontaneous neuronal activity distinguishes human dorsal and ventral attention systems.

On the basis of task-related imaging studies in normal human subjects, it has been suggested that two attention systems exist in the human brain: a bilateral dorsal attention system involved in top-down orienting of attention and a right-lateralized ventral attention system involved in reorienting attention in response to salient sensory stimuli. An important question is whether this functional organization emerges only in response to external attentional demands or is represented more fundamentally in the internal dynamics of brain activity. To address this question, we examine correlations in spontaneous fluctuations of the functional MRI blood oxygen level-dependent signal in the absence of task, stimuli, or explicit attentional demands. We identify a bilateral dorsal attention system and a right-lateralized ventral attention system solely on the basis of spontaneous activity. Further, we observe regions in the prefrontal cortex correlated with both systems, a potential mechanism for mediating the functional interaction between systems. These findings demonstrate that the neuroanatomical substrates of human attention persist in the absence of external events, reflected in the correlation structure of spontaneous activity.

The human brain is intrinsically organized into dynamic, anticorrelated functional networks.

During performance of attention-demanding cognitive tasks, certain regions of the brain routinely increase activity, whereas others routinely decrease activity. In this study, we investigate the extent to which this task-related dichotomy is represented intrinsically in the resting human brain through examination of spontaneous fluctuations in the functional MRI blood oxygen level-dependent signal. We identify two diametrically opposed, widely distributed brain networks on the basis of both spontaneous correlations within each network and anticorrelations between networks. One network consists of regions routinely exhibiting task-related activations and the other of regions routinely exhibiting task-related deactivations. This intrinsic organization, featuring the presence of anticorrelated networks in the absence of overt task performance, provides a critical context in which to understand brain function. We suggest that both task-driven neuronal responses and behavior are reflections of this dynamic, ongoing, functional organization of the brain.