Search

Military and law enforcement breachers are exposed to many low-level blasts during their training and occupational experiences in which they detonate explosives to force entry into secured structures. There is a concern that exposure to these repetitive blast events in career breachers could result in cumulative neurological effects. This study aimed to determine concentrations of neurofilament light (NF-L), tau, and amyloid-beta 42 (Aβ42) in serum and in neuronal-derived extracellular vesicles (EVs) in an experienced breacher population, and to examine biomarker associations with neurobehavioral symptoms. Thirty-four participants enrolled in the study: 20 experienced breachers and 14 matched military or civilian law enforcement controls. EV tau concentrations were significantly elevated in experienced breachers (0.3301 ± 0.5225) compared to controls (−0.4279 ± 0.7557; F = 10.43, p = 0.003). No statistically significant changes were observed in EV levels of NF-L or Aβ42 or in serum levels of NF-L, tau, or Aβ42 (p’s > 0.05). Elevated EV tau concentrations correlated with increased Neurobehavioral Symptom Inventory (NSI) score in experienced breachers (r = 0.596, p = 0.015) and predicted higher NSI score (F(1,14) = 7.702, p = 0.015, R2 = 0.355). These findings show that neuronal-derived EV concentrations of tau are significantly elevated and associated with neurobehavioral symptoms in this sample of experienced breachers who have a history of many low-level blast exposures.

Combat military and civilian law enforcement personnel may be exposed to repetitive low-intensity blast events during training and operations. Persons who use explosives to gain entry (i.e., breach) into buildings are known as “breachers” or dynamic entry personnel. Breachers operate under the guidance of established safety protocols, but despite these precautions, breachers who are exposed to low-level blast throughout their careers frequently report performance deficits and symptoms to healthcare providers. Although little is known about the etiology linking blast exposure to clinical symptoms in humans, animal studies demonstrate network-level changes in brain function, alterations in brain morphology, vascular and inflammatory changes, hearing loss, and even alterations in gene expression after repeated blast exposure. To explore whether similar effects occur in humans, we collected a comprehensive data battery from 20 experienced breachers exposed to blast throughout their careers and 14 military and law enforcement controls. This battery included neuropsychological assessments, blood biomarkers, and magnetic resonance imaging measures, including cortical thickness, diffusion tensor imaging of white matter, functional connectivity, and perfusion. To better understand the relationship between repetitive low-level blast exposure and behavioral and imaging differences in humans, we analyzed the data using similarity-driven multi-view linear reconstruction (SiMLR). SiMLR is specifically designed for multiple modality statistical integration using dimensionality-reduction techniques for studies with high-dimensional, yet sparse, data (i.e., low number of subjects and many data per subject). We identify significant group effects in these data spanning brain structure, function, and blood biomarkers.

With the rapid development of computer software and hardware technologies, more and more healthcare data are becoming readily available from clinical institutions, patients, insurance companies and pharmaceutical industries, among others. This access provides an unprecedented opportunity for data science technologies to derive data-driven insights and improve the quality of care delivery. Healthcare data, however, are usually fragmented and private making it difficult to generate robust results across populations. For example, different hospitals own the electronic health records (EHR) of different patient populations and these records are difficult to share across hospitals because of their sensitive nature. This creates a big barrier for developing effective analytical approaches that are generalizable, which need diverse, "big data". Federated learning, a mechanism of training a shared global model with a central server while keeping all the sensitive data in local institutions where the data belong, provides great promise to connect the fragmented healthcare data sources with privacy-preservation. The goal of this survey is to provide a review for federated learning technologies, particularly within the biomedical space. In particular, we summarize the general solutions to the statistical challenges, system challenges and privacy issues in federated learning, and point out the implications and potentials in healthcare., Comment: 18 pages

Privacy concerns on sharing sensitive data across institutions are particularly paramount for the medical domain, which hinders the research and development of many applications, such as cohort construction for cross-institution observational studies and disease surveillance. Not only that, the large volume and heterogeneity of the patient data pose great challenges for retrieval and analysis. To address these challenges, in this paper, we propose a Federated Patient Hashing (FPH) framework, which collaboratively trains a retrieval model stored in a shared memory while keeping all the patient-level information in local institutions. Specifically, the objective function is constructed by minimization of a similarity preserving loss and a heterogeneity digging loss, which preserves both inter-data and intra-data relationships. Then, by leveraging the concept of Bregman divergence, we implement optimization in a federated manner in both centralized and decentralized learning settings, without accessing the raw training data across institutions. In addition to this, we also analyze the convergence rate of the FPH framework. Extensive experiments on real-world clinical data set from critical care are provided to demonstrate the effectiveness of the proposed method on similar patient matching across institutions.

Simple Summary Whether it is for COVID-19 primary vaccination or the administration of booster vaccines, prioritization criteria need to be established to optimize COVID-19 vaccination programs accounting for both clinical and social vulnerability risks for severe COVID-19 disease. We developed a dual socio-clinical risk model for severe COVID-19 disease in the Medicare population, which is comprised mostly of individuals aged 65 and over. Our model generated risk levels correlated with regionalized COVID-19 case hospitalization rates and mapped them at the county and zip code levels. The model and map can be used by health jurisdictions to reach out to unvaccinated individuals. Our model approach can also be applied to identify Medicare beneficiaries who were in early vaccination groups to be vaccinated to identify those who might maximally benefit from an additional dose of COVID-19 vaccine if and when vaccine immunity wanes. Abstract 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.

Most applications of machine intelligence have focused on demonstrating crystallized intelligence. Crystallized intelligence relies on accessing problem-specific knowledge, skills and experience stored in long term memory. In this paper, we challenge the AI community to design AIs to completely take tests of fluid intelligence which assess the ability to solve novel problems using problem-independent solving skills. Tests of fluid intelligence such as the NNAT are used extensively by schools to determine entry into gifted education programs. We explain the differences between crystallized and fluid intelligence, the importance and capabilities of machines demonstrating fluid intelligence and pose several challenges to the AI community, including that a machine taking such a test would be considered gifted by school districts in the state of California. Importantly, we show existing work on seemingly related fields such as transfer, zero-shot, life-long and meta learning (in their current form) are not directly capable of demonstrating fluid intelligence but instead are task-transductive mechanisms.

To evaluate how blast exposure impacts peripheral biomarkers.in military personnel enrolled in 10-day blast training.On day 7, 21 military personnel experienced peak overpressure2 pounds per square inch (psi); while 29 military personnel experienced peak overpressure ≥5 psi. Blood samples were collected each day to measure changes in amyloid beta (Aβ), neurofilament light chain (NFL), and tau concentrations.Within 24 hours following exposure ≥5 psi, the ≥5 psi group had lower Aβ42 (These findings provide an initial report of acute alterations in biomarker concentrations following blast exposure.

Background: Blast exposure is a potential hazard in modern military operations and training, especially for some military occupations. Helmets, peripheral armor, hearing protection, and eye protection worn by military personnel provide some acute protection from blast effects but may not fully protect personnel against cumulative effects of repeated blast overpressure waves experienced over a career. The current study aimed to characterize the long-term outcomes of repeated exposure to primary blast overpressure in experienced career operators with an emphasis on the assessment of hearing and vestibular outcomes.Methods: Participants included experienced “breachers” (military and law enforcement explosives professionals who gain entry into structures through controlled detonation of charges) and similarly aged and experienced “non-breachers” (non-breaching military and law enforcement personnel). Responses to a clinical interview and performance on audiological and vestibular testing were compared.Results: Hearing loss, ringing in the ears, irritability, and sensitivity to light or noise were more common among breachers than non-breachers. Breachers reported more combat exposure than non-breachers, and subsequently, memory loss and difficulty concentrating were associated with both breaching and combat exposure. Vestibular and ocular motor outcomes were not different between breachers and non-breachers.Conclusion: Hearing-related, irritability, and sensitivity outcomes are associated with a career in breaching. Future studies examining long-term effects of blast exposure should take measures to control for combat exposure.

At present, there are no set guidelines establishing cumulative limits for blast exposure numbers and intensities in military personnel, in combat or training operations. The objective of the current study was to define lung injury, pathology, and associated behavioral changes from primary repeated blast lung injury under appropriate exposure conditions and combinations (i.e. blast overpressure (BOP) intensity and exposure frequency) using an advanced blast simulator. Male Sprague Dawley rats were exposed to BOP frontally and laterally at a pressure range of ~ 8.5–19 psi, for up to 30 daily exposures. The extent of lung injury was identified at 24 h following BOP by assessing the extent of surface hemorrhage/contusion, Hematoxylin and Eosin staining, and behavioral deficits with open field activity. Lung injury was mathematically modeled to define the military standard 1% lung injury threshold. Significant levels of histiocytosis and inflammation were observed in pressures ≥ 10 psi and orientation effects were observed at pressures ≥ 13 psi. Experimental data demonstrated ~ 8.5 psi is the threshold for hemorrhage/contusion, up to 30 exposures. Modeling the data predicted injury risk up to 50 exposures with intensity thresholds at 8 psi for front exposure and 6psi for side exposures, which needs to be validated further.

A unique cohort of military personnel exposed to isolated blast was studied to explore acute peripheral cytokine levels, with the aim of identifying blast-specific biomarkers. Several cytokines, including interleukin (IL) 6, IL-10 and tumor necrosis factor alpha (TNFα) have been linked to pre-clinical blast exposure, but remained unstudied in clinical blast exposure. To address this gap, blood samples from 62 military personnel were obtained at baseline, and daily, during a 10-day blast-related training program; changes in the peripheral concentrations of IL-6, IL-10 and TNFα were evaluated using an ultrasensitive assay. Two groups of trainees were matched on age, duration of military service, and previous history of blast exposure(s), resulting in moderate blast cases and no/low blast controls. Blast exposures were measured using helmet sensors that determined the average peak pressure in pounds per square inch (psi). Moderate blast cases had significantly elevated concentrations of IL-6 (F1,60 = 18.81, p

Personalized medicine is a ubiquitous term that has come to be used to describe a medical model that proposes the customization of healthcare, such that decisions and/or treatments are tailored to each individual patient. Under this type of clinical practice model, diagnostic and prognostic decisions are often based upon selecting the most appropriate therapy based on a patient’s genetic, demographic, and/or other pertinent information. The primary aim of this paper is to use a personalized medicine framework to better understand the relationship between neuropsychological testing and the progression of symptoms in a blast-induced mild Traumatic Brain Injury (mTBI) patient population. In this paper, we extended our earlier work on Constrained Spectral Partitioning (CSP), a graph-based approach that incorporates additional information from separate graphs to help improve the clustering quality on both graphs simultaneously. While our previous work demonstrated the effectiveness of this algorithm in its ability to accurately classify whether symptoms improved or declined over time, that work did not provide any insights into the progression of symptoms. Therefore, this paper sought to identify, from a clinical perspective, whether symptoms increased/decreased over time. To accomplish this, we developed a decision tree classifier to classify symptom progression based on the outputs from our CSP algorithm. We present results from four separate decision tree classifiers that illustrate the adaptability of these algorithms for utilization as decision rules for the treatment of patients following blast-induced mTBI. Decision tree classifier models are useful in the healthcare setting because patient health data (e.g., diagnosis of a condition or a type of treatment) can be imput into the model and, based on the health data variables, a resulting outcome can be suggested, and providers can use this outcome as information to direct their clinical treatment.

White matter hyperintensities (WMHs) are foci of abnormal signal intensity in white matter regions seen with magnetic resonance imaging (MRI). WMHs are associated with normal ageing and have shown prognostic value in neurological conditions such as traumatic brain injury (TBI). The impracticality of manually quantifying these lesions limits their clinical utility and motivates the utilization of machine learning techniques for automated segmentation workflows.This study develops a concatenated random forest framework with image features for segmenting WMHs in a TBI cohort. The framework is built upon the Advanced Normalization Tools (ANTs) and ANTsR toolkits. MR (3D FLAIR, T2- and T1-weighted) images from 24 service members and veterans scanned in the Chronic Effects of Neurotrauma Consortium's (CENC) observational study were acquired. Manual annotations were employed for both training and evaluation using a leave-one-out strategy. Performance measures include sensitivity, positive predictive value, [Formula: see text] score and relative volume difference.Final average results were: sensitivity = 0.68 ± 0.38, positive predictive value = 0.51 ± 0.40, [Formula: see text] = 0.52 ± 0.36, relative volume difference = 43 ± 26%. In addition, three lesion size ranges are selected to illustrate the variation in performance with lesion size.Paired with correlative outcome data, supervised learning methods may allow for identification of imaging features predictive of diagnosis and prognosis in individual TBI patients.

Background Hypoxia is a critical secondary injury mechanism in traumatic brain injury (TBI), and early intervention to alleviate post-TBI hypoxia may be beneficial. NVX-108, a dodecafluoropentane perfluorocarbon, was screened for its ability to increase brain tissue oxygen tension (PbtO 2 ) when administered soon after TBI. Methods Ketamine-acepromazine anesthetized rats ventilated with 40% oxygen underwent moderate controlled cortical impact (CCI)-TBI at time 0 (T0). Rats received either no treatment (NON, n =8) or 0.5 ml/kg intravenous (IV) NVX-108 (NVX, n =9) at T15 (15 min after TBI) and T75. Results Baseline cortical PbtO 2 was 28±3 mm Hg and CCI-TBI resulted in a 46±6% reduction in PbtO 2 at T15 ( P P =0.013) were found when comparing either absolute or percentage change of PbtO 2 to post-injury (mixed-model ANOVA) suggesting that administration of NVX-108 increased PbtO 2 above injury levels while it remained depressed in the NON group. Specifically in the NVX group, PbtO 2 increased to a peak 143% of T15 ( P =0.02) 60 min after completion of NVX-108 injection (T135). Systemic blood pressure was not different between the groups. Conclusion NVX-108 caused an increase in PbtO 2 following CCI-TBI in rats and should be evaluated further as a possible immediate treatment for TBI.

OBJECTIVE The aim of this study was to assess, in two experiments, the safety and efficacy of the PFC emulsion Oxycyte as an oxygen therapeutic for TBI to test the hypothesis that early administration of this oxygen-carrying fluid post-TBI would improve brain tissue oxygenation (Pbt O2 ). METHODS The first experiment assessed the effects of Oxycyte on cerebral vasoactivity in healthy, uninjured rats using intravital microscopy. The second experiment investigated the effect of Oxycyte on cerebral Pbt O2 using the PQM in TBI model. Animals in the Oxycyte group received a single injection of Oxycyte (6 mL/kg) shortly after TBI, while NON animals received no treatment. RESULTS Oxycyte did not cause vasoconstriction in small- (

A common problem with spatiotemporal data is how to simplify the data to discover an underlying network that consists of cohesive spatial regions (nodes) and relationships between those regions (edges). This network discovery problem naturally exists in a multitude of domains including climate data (dipoles), astronomical data (gravitational lensing) and the focus of this paper, fMRI scans of human subjects. Whereas previous work requires strong supervision, we propose an unsupervised matrix tri-factorization formulation with complex constraints and spatial regularization. We show that this formulation works well in controlled experiments with synthetic networks and is able to recover the underlying ground-truth network. We then show that for real fMRI data our approach can reproduce well known results in neurology regarding the default mode network in resting-state healthy and Alzheimer affected individuals.

Objective The purpose of this study is to utilize a natural history approach to describe and understand symptom recovery in personnel diagnosed with a blast-related mild traumatic brain injury (mTBI) resulting from an improvised explosive device blast. Participants and design The population included military personnel who experienced a blast mTBI while mounted (vehicle; n = 176) or dismounted (on foot; n = 37) (N = 213). Patients had no co-morbid psychiatric or muscle-skeletal issues and were treated within 72 h of injury. Prevalence and duration of self-reported symptoms were separately analyzed by injury context (mounted vs dismounted). Results Headache was prominently reported in both mounted (85%) and dismounted (75%) populations. The mean time from injury to return to full duty was between 7.8 d (mounted) and 8.5 d (dismounted). The dismounted population reported visual changes that lasted 0.74 d longer. Conclusion Our analysis implicates that headache is a common and acutely persistent symptom in mTBI regardless of injury context. Additionally, patients in mounted vs dismounted injury did not report significant differences in symptom prevalence. Although knowing the injury context (i.e., dismounted vs mounted) may be beneficial for providers to understand symptom presentations and deliver accurate anticipatory guidance for patients with blast-related mTBI, no significant differences were observed in this population. This may be due to the population characteristic as the trajectory of recovery may vary for patients who were not able to return to full duty within 30 d or required higher levels of care.

Objective:To explore gene expression after moderate blast exposure (vs baseline) and proteomic changes after moderate- (vs low-) blast exposure.Methods:Military personnel (N = 69) donated blood for quantification of protein level, and peak pressure exposures were detected by helmet sensors before and during a blast training program (10 days total). On day 7, some participants (n = 29) sustained a moderate blast (mean peak pressure = 7.9 psi) and were matched to participants with no/low-blast exposure during the training (n = 40). PAXgene tubes were collected from one training site at baseline and day 10; RNA-sequencing day 10 expression was compared with each participant's own baseline samples to identify genes and pathways differentially expressed in moderate blast-exposed participants. Changes in amyloid precursor protein (APP) from baseline to the day of blast and following 2 days were evaluated. Symptoms were assessed using a self-reported form.Results:We identified 1,803 differentially expressed genes after moderate blast exposure; the most altered network was APP. Significantly reduced levels of peripheral APP were detected the day after the moderate blast exposure and the following day. Protein concentrations correlated with the magnitude of the moderate blast exposure on days 8 and 9. APP concentrations returned to baseline levels 3 days following the blast, likely due to increases in the genetic expression of APP. Onset of concentration problems and headaches occurred after moderate blast.Conclusions:Moderate blast exposure results in a signature biological profile that includes acute APP reductions, followed by genetic expression increases and normalization of APP levels; these changes likely influence neuronal recovery.

Mild traumatic brain injury is a common outcome of blast exposure, and current literature indicates high rates of comorbid posttraumatic stress disorder (PTSD) in military personnel. Blast-exposed rats display PTSD-like behavior, suggesting relationships may exist between PTSD and blast exposure. Other studies demonstrate the roles of stathmin and corticosterone associated with fear- and anxiety-like behaviors in rodent models. Furthermore, studies have observed ranges of responses to both physical and psychological trauma in animal populations (Elder 2012, Ritov 2016). This study exposed rodents to repeated blast overpressure (BOP) and analyzed behavioral responses and molecular variables at 3 weeks and 6 months after exposure. We applied a modified version of a previously reported behavioral profiling approach that separates "affected" and "unaffected" rats based on the presence of anxiety-like behaviors (Ritov, 2016). We report that "affected" 3 week animals showed higher plasma corticosterone and amygdalar stathmin levels, while "affected" 6 month animals had lower prefrontal cortex stathmin. Higher corticosterone also paralleled anxiety behavior in "affected" 3 week animals, which was not observed in 6 month animals, indicating possible negative feedback loop mechanisms. Elevated levels of amygdalar stathmin correlated with anxiety behaviors in "affected" 3 week and 6 month animals, indicating sustained molecular changes. We conclude that this unique analysis may provide more information about response to blast. This type of analysis should also be considered when treating clinical populations, since individual differences may affect behavioral and long-term outcomes. Future studies should elucidate relationships of stress and fear responses in the context of BOP.

Traumatic brain injury (TBI) is one of the most common forms of neurotrauma that has affected more than 250,000 military service members over the last decade alone. While in battle, service members who experience TBI are at significant risk for the development of normal TBI symptoms, as well as risk for the development of psychological disorders such as Post-Traumatic Stress Disorder (PTSD). As such, these service members often require intense bouts of medication and therapy in order to resume full return-to-duty status. The primary aim of this study is to identify the relationship between the administration of specific medications and reductions in symptomology such as headaches, dizziness, or light-headedness. Service members diagnosed with mTBI and seen at the Concussion Restoration Care Center (CRCC) in Afghanistan were analyzed according to prescribed medications and symptomology. Here, we demonstrate that in such situations with sparse labels and small feature sets, classic analytic techniques such as logistic regression, support vector machines, naive Bayes, random forest, decision trees, and k-nearest neighbor are not well suited for the prediction of outcomes. We attribute our findings to several issues inherent to this problem setting and discuss several advantages of spectral graph methods.

Social science is broadly defined as the analysis of human behavior whether it be at an individual or a group level. In this work, we explore the analysis of human behavior encoded as a trail of their events over time and space, which we refer to as behavioral event data. We show that such data offers challenges to data mining algorithm designers as the data to analyze is naturally multi-way, involves complex patterns that form/reform over time, and has complex interactions between groups in the population. Though the data naturally lends itself to be represented as graphs and tensors we show how existing techniques are limited in their usefulness and outline our own algorithms to overcome these challenges. In this paper, using the adversarial event behavior of blue and red forces, we show three core problems and solutions in event behavior analysis: (1) Decomposing behavior to identify areas of intense activity, (2) Predicting what groups of events are likely to occur, and (3) Analysis to identify interacting behavior given a known template.

The Department of Defense continues to invest in the development and validation of several accident investigation and reporting methodologies geared toward the accurate analysis of human factors issues. Recently, a joint memorandum of agreement was signed by each of the military services to use the DoD Human Factors Analysis and Classification System (HFACS). While a great deal of research has been devoted to assessing the validity of HFACS as applied to DoD mishaps, less research has focused on the utility of the model as both an accident investigation and reporting tool. The primary aim of this research was to apply lifted rule probabilities at the nanocode level within HFACS to identify common linkages within the DoD version of HFACS.

Core proteins of mitochondrial protein import are found in all mitochondria, suggesting a common origin of this import machinery. Despite the presence of a universal core import mechanism, there are specific proteins found only in a few groups of organisms. One of these proteins is the translocase of outer membrane 70 (Tom70), a protein that is essential for the import of preproteins with internal targeting sequences into the mitochondrion. Until now, Tom70 has only been found in animals and Fungi. We have identified a tom70 gene in the human parasitic anaerobic stramenopile Blastocystis sp. that is neither an animal nor a fungus. Using a combination of bioinformatics, genetic complementation, and immunofluorescence microscopy analyses, we demonstrate that this protein functions as a typical Tom70 in Blastocystis mitochondrion-related organelles. Additionally, we identified putative tom70 genes in the genomes of other stramenopiles and a haptophyte, that, in phylogenies, form a monophyletic group distinct from the animal and the fungal homologues. The presence of Tom70 in these lineages significantly expands the evolutionary spectrum of eukaryotes that contain this protein and suggests that it may have been part of the core mitochondrial protein import apparatus of the last common ancestral eukaryote.

A unique cohort of military personnel exposed to isolated blast was studied to explore acute peripheral cytokine levels, with the aim of identifying blast-specific biomarkers. Several cytokines, including interleukin (IL) 6, IL-10 and tumor necrosis factor alpha (TNFα) have been linked to pre-clinical blast exposure, but remained unstudied in clinical blast exposure. To address this gap, blood samples from 62 military personnel were obtained at baseline, and daily, during a 10-day blast-related training program; changes in the peripheral concentrations of IL-6, IL-10 and TNFα were evaluated using an ultrasensitive assay. Two groups of trainees were matched on age, duration of military service, and previous history of blast exposure(s), resulting in moderate blast cases and no/low blast controls. Blast exposures were measured using helmet sensors that determined the average peak pressure in pounds per square inch (psi). Moderate blast cases had significantly elevated concentrations of IL-6 (F1,60 =18.81, p < 0.01) and TNFα (F1,60 =12.03, p < 0.01) compared to no/low blast controls; levels rebounded to baseline levels the day after blast. On the day of the moderate blast exposure, the extent of the overpressure (psi) in those exposed correlated with IL-6 (r = 0.46, p < 0.05) concentrations. These findings indicate that moderate primary blast exposure results in changes, specifically acute and transient increases in peripheral inflammatory markers which may have implications for neuronal health.

The analysis of data represented as graphs is common having wide scale applications from social networks to medical imaging. A popular analysis is to cut the graph so that the disjoint subgraphs can represent communities (for social network) or background and foreground cognitive activity (for medical imaging). An emerging setting is when multiple data sets (graphs) exist which opens up the opportunity for many new questions. In this paper we study two such questions: i) For a collection of graphs find a single cut that is good for all the graphs and ii) For two collections of graphs find a single cut that is good for one collection but poor for the other. We show that existing formulations of multiview, consensus and alternative clustering cannot address these questions and instead we provide novel formulations in the spectral clustering framework. We evaluate our approaches on functional magnetic resonance imaging (fMRI) data to address questions such as: "What common cognitive network does this group of individuals have?" and "What are the differences in the cognitive networks for these two groups?" We obtain useful results without the need for strong domain knowledge.