Search

Background When trying to control regional spread of antibiotic-resistant pathogens such as carbapenem-resistant Enterobacteriaceae (CRE), decision makers must choose the highest-yield facilities to target for interventions. The question is, with limited resources, how best to choose these facilities. Methods Using our Regional Healthcare Ecosystem Analyst–generated agent-based model of all Chicago metropolitan area inpatient facilities, we simulated the spread of CRE and different ways of choosing facilities to apply a prevention bundle (screening, chlorhexidine gluconate bathing, hand hygiene, geographic separation, and patient registry) to a resource-limited 1686 inpatient beds. Results Randomly selecting facilities did not impact prevalence, but averted 620 new carriers and 175 infections, saving $6.3 million in total costs compared to no intervention. Selecting facilities by type (eg, long-term acute care hospitals) yielded a 16.1% relative prevalence decrease, preventing 1960 cases and 558 infections, saving $62.4 million more than random selection. Choosing the largest facilities was better than random selection, but not better than by type. Selecting by considering connections to other facilities (ie, highest volume of discharge patients) yielded a 9.5% relative prevalence decrease, preventing 1580 cases and 470 infections, and saving $51.6 million more than random selection. Selecting facilities using a combination of these metrics yielded the greatest reduction (19.0% relative prevalence decrease, preventing 1840 cases and 554 infections, saving $59.6 million compared with random selection). Conclusions While choosing target facilities based on single metrics (eg, most inpatient beds, most connections to other facilities) achieved better control than randomly choosing facilities, more effective targeting occurred when considering how these and other factors (eg, patient length of stay, care for higher-risk patients) interacted as a system.

BACKGROUND: Since special efforts are necessary to vaccinate people living far from fixed vaccination posts, decision makers are interested in knowing the economic value of such efforts. METHODS: Using our immunization geospatial information system platform and a measles compartment model, we quantified the health and economic value of a 2-dose measles immunization outreach strategy for children

Today’s landscape of computational science is evolving rapidly, with a need for new, flexible, and responsive supercomputing platforms for addressing the growing areas of artificial intelligence (AI), data analytics (DA) and convergent collaborative research. To support this community, we designed and deployed the Bridges-2 platform. Building on our highly successful Bridges supercomputer, which was a high-performance computing resource supporting new communities and complex workflows, Bridges-2 supports traditional and nontraditional research communities and applications; integrates new technologies for converged, scalable high-performance computing (HPC), AI, and data analytics; prioritizes researcher productivity and ease of use; and provides an extensible architecture for interoperation with complementary data intensive projects, campuses, and clouds. In this report, we describe Bridges-2’s hardware and configuration, user environments, and systems support and present the results of the successful Early User Program.

Typically, long-term acute care hospitals (LTACHs) have less experience in and incentives to implementing aggressive infection control for drug-resistant organisms such as carbapenem-resistant Enterobacteriaceae (CRE) than acute care hospitals. Decision makers need to understand how implementing control measures in LTACHs can impact CRE spread regionwide. Using our Chicago metropolitan region agent-based model to simulate CRE spread and control, we estimated that a prevention bundle in only LTACHs decreased prevalence by a relative 4.6%–17.1%, averted 1,090–2,795 new carriers, 273–722 infections and 37–87 deaths over 3 years and saved $30.5–$69.1 million, compared with no CRE control measures. When LTACHs and intensive care units intervened, prevalence decreased by a relative 21.2%. Adding LTACHs averted an additional 1,995 carriers, 513 infections, and 62 deaths, and saved $47.6 million beyond implementation in intensive care units alone. Thus, LTACHs may be more important than other acute care settings for controlling CRE, and regional efforts to control drug-resistant organisms should start with LTACHs as a centerpiece.

A lack of software reproducibility has become increasingly apparent in the last several years, calling into question the validity of scientific findings affected by published tools. Reproducibility issues may have numerous sources of error, including the underlying numerical stability of algorithms and implementations employed. Various forms of instability have been observed in neuroimaging, including across operating system versions, minor noise injections, and implementation of theoretically equivalent algorithms. In this paper we explore the effect of various perturbation methods on a typical neuroimaging pipeline through the use of i) targeted noise injections, ii) Monte Carlo Arithmetic, and iii) varying operating systems to identify the quality and severity of their impact. The work presented here demonstrates that even low order computational models such as the connectome estimation pipeline that we used are susceptible to noise. This suggests that stability is a relevant axis upon which tools should be compared, developed, or improved, alongside more commonly considered axes such as accuracy/biological feasibility or performance. The heterogeneity observed across participants clearly illustrates that stability is a property of not just the data or tools independently, but their interaction. Characterization of stability should therefore be evaluated for specific analyses and performed on a representative set of subjects for consideration in subsequent statistical testing. Additionally, identifying how this relationship scales to higher-order models is an exciting next step which will be explored. Finally, the joint application of perturbation methods with post-processing approaches such as bagging or signal normalization may lead to the development of more numerically stable analyses while maintaining sensitivity to meaningful variation., Comment: 9 pages, 5 figures, 1 table, paper published at IJHPCA

Background The choice of preprocessing pipeline introduces variability in neuroimaging analyses that affects the reproducibility of scientific findings. Features derived from structural and functional MRI data are sensitive to the algorithmic or parametric differences of preprocessing tasks, such as image normalization, registration, and segmentation to name a few. Therefore it is critical to understand and potentially mitigate the cumulative biases of pipelines in order to distinguish biological effects from methodological variance. Methods Here we use an open structural MRI dataset (ABIDE), supplemented with the Human Connectome Project, to highlight the impact of pipeline selection on cortical thickness measures. Specifically, we investigate the effect of (i) software tool (e.g., ANTS, CIVET, FreeSurfer), (ii) cortical parcellation (Desikan-Killiany-Tourville, Destrieux, Glasser), and (iii) quality control procedure (manual, automatic). We divide our statistical analyses by (i) method type, i.e., task-free (unsupervised) versus task-driven (supervised); and (ii) inference objective, i.e., neurobiological group differences versus individual prediction. Results Results show that software, parcellation, and quality control significantly affect task-driven neurobiological inference. Additionally, software selection strongly affects neurobiological (i.e. group) and individual task-free analyses, and quality control alters the performance for the individual-centric prediction tasks. Conclusions This comparative performance evaluation partially explains the source of inconsistencies in neuroimaging findings. Furthermore, it underscores the need for more rigorous scientific workflows and accessible informatics resources to replicate and compare preprocessing pipelines to address the compounding problem of reproducibility in the age of large-scale, data-driven computational neuroscience.

Open-access neuroimaging datasets have reached petabyte scale and continue to grow. The ability to leverage the entirety of these datasets is limited to a restricted number of research laboratories with both the capacity and infrastructure to process the data. Whereas Big Data engines have significantly reduced application performance penalties with respect to data movement, their applied strategies (e.g. data locality, in-memory computing and lazy evaluation) are not necessarily practical within neuroimaging workflows where intermediary results may need to be materialized to shared storage for post-processing analysis. In this paper we evaluate the performance advantage brought by Intel® Optane™ DC persistent memory for the processing of large neuroimaging datasets using the two available configurations modes: Memory mode and App Direct mode. We employ a synthetic algorithm on the 76 GiB and 603 GiB BigBrain, as well as apply a standard neuroimaging application on the Consortium for Reliability and Reproducibility (CoRR) dataset using 25 and 96 parallel processes in both cases. Our results show that the performance of applications leveraging persistent memory is superior to that of other storage devices, with the exception of DRAM. This is the case in both Memory and App Direct mode, irrespective of the amount of data and parallelism. Furthermore, persistent memory in App Direct mode is believed to benefit from the use of DRAM as a cache for writing when output data is significantly smaller than available memory. We believe the use of persistent memory will be beneficial to both neuroimaging applications running on HPC or visualization of large, high-resolution images.

BACKGROUND: Microneedle patch (MNP) technology is designed to simplify the process of vaccine administration; however, depending on its characteristics, MNP technology may provide additional benefits beyond the point-of-use, particularly for vaccine supply chains. METHODS: Using the HERMES modeling software, we examined replacing four routine vaccines – Measles-containing vaccine (MCV), Tetanus toxoid (TT), Rotavirus (Rota) and Pentavalent (Penta) – with MNP versions in the routine vaccine supply chains of Benin, Bihar (India), and Mozambique. RESULTS: Replacing MCV with an MNP (5 cm(3)-per-dose, 2-month thermostability, current single-dose price-per-dose) improved MCV availability by 13%, 1% and 6% in Benin, Bihar and Mozambique, respectively, and total vaccine availability by 1% in Benin and Mozambique, while increasing the total cost per dose administered by $0.07 in Benin, $0.56 in Bihar and $0.11 in Mozambique. Replacing TT with an MNP improved TT and total vaccine availability (3% and

Background Annual influenza vaccination is a key to preventing widespread influenza infections. Recent reports of influenza vaccine effectiveness (VE) indicate that vaccination in prior years may reduce VE in the current season, suggesting vaccine interference. The purpose of this study is to evaluate the potential effect of repeat influenza vaccinations in the presence of vaccine interference. Methods Using literature-based parameters, an age-structured influenza equation-based transmission model was used to determine the optimal vaccination strategy, while considering the effect of varying levels of interference. Results The model shows that, even in the presence of vaccine interference, revaccination reduces the influenza attack rate and provides individual benefits. Specifically, annual vaccination is a favored strategy over vaccination in alternate years, as long as the level of residual protection is less than 58% or vaccine interference effect is minimal. Furthermore, the negative impact of vaccine interference may be offset by increased vaccine coverage levels. Conclusions Even in the presence of potential vaccine interference, our work provides a population-level perspective on the potential merits of repeated influenza vaccination. This is because repeat vaccination groups had lower attack rates than groups that omitted the second vaccination unless vaccine interference was at very high, perhaps implausible, levels.

Objective The objective of this study was to quantify the impact of crime on physical activity location accessibility, leisure-time physical activity (LTPA), and obesity among African American women. Methods An agent-based model was developed in 2016 to represent resource-limited Washington, DC, communities and their populations to simulate the impact of crime on LTPA and obesity among African American women under different circumstances. Results Data analysis conducted between 2016 and 2017 found that in the baseline scenario, African American women had a 25% probability of exercising. Reducing crime so more physical activity locations were accessible (increasing from 10% to 50%) decreased the annual rise in obesity prevalence by 2.69%. Increasing the probability of African American women to exercise to 37.5% further increased the impact of reducing crime on obesity (2.91% annual decrease in obesity prevalence). Conclusions These simulations showed that crime may serve as a barrier to LTPA. Reducing crime and increasing propensity to exercise through multilevel interventions (i.e., economic development initiatives to increase time available for physical activity and subsidized health care) may promote greater than linear declines in obesity prevalence. Crime prevention strategies alone can help prevent obesity, but combining such efforts with other ways to encourage physical activity can yield even greater benefits.

Introduction Decreased live attenuated influenza vaccine (LAIV) effectiveness in the U.S. prompted the Advisory Committee on Immunization Practices in August 2016 to recommend against this vaccine’s use. However, overall influenza uptake increases when LAIV is available and, unlike the U.S., LAIV has retained its effectiveness in other countries. These opposing countercurrents create a dilemma. Methods To examine the potential consequences of the decision to not recommend LAIV, which may result in decreased influenza vaccination coverage in the U.S. population, a Markov decision analysis model was used to examine influenza vaccination options in U.S. children aged 2–8 years. Data were compiled and analyzed in 2016. Results Using recently observed low LAIV effectiveness values, fewer influenza cases will occur if LAIV is not used compared with having LAIV as a vaccine option. However, having the option to use LAIV may be favored if LAIV effectiveness returns to prior levels or if the absence of vaccine choice substantially decreases overall vaccine uptake. Conclusions Continued surveillance of LAIV effectiveness and influenza vaccine uptake is warranted, given their importance in influenza vaccination policy decisions.

Malaria-endemic countries have to decide how much of their limited resources for vector control to allocate toward implementing long-lasting insecticidal nets (LLINs) versus indoor residual spraying (IRS). To help the Mozambique Ministry of Health use an evidence-based approach to determine funding allocation toward various malaria control strategies, the Global Fund convened the Mozambique Modeling Working Group which then used JANUS, a software platform that includes integrated computational economic, operational, and clinical outcome models that can link with different transmission models (in this case, OpenMalaria) to determine the economic value of vector control strategies. Any increase in LLINs (from 80% baseline coverage) or IRS (from 80% baseline coverage) would be cost-effective (incremental cost-effectiveness ratios ≤ $114/disability-adjusted life year averted). However, LLIN coverage increases tend to be more cost-effective than similar IRS coverage increases, except where both pyrethroid resistance is high and LLIN usage is low. In high-transmission northern regions, increasing LLIN coverage would be more cost-effective than increasing IRS coverage. In medium-transmission central regions, changing from LLINs to IRS would be more costly and less effective. In low-transmission southern regions, LLINs were more costly and less effective than IRS, due to low LLIN usage. In regions where LLINs are more cost-effective than IRS, it is worth considering prioritizing LLIN coverage and use. However, IRS may have an important role in insecticide resistance management and epidemic control. Malaria intervention campaigns are not a one-size-fits-all solution, and tailored approaches are necessary to account for the heterogeneity of malaria epidemiology.

Background Gavi recommends solar refrigerators for vaccine storage in areas with less than eight hours of electricity per day, and WHO guidelines are more conservative. The question remains: Can solar refrigerators provide value where electrical outages are less frequent? Methods Using a HERMES-generated computational model of the Mozambique routine immunization supply chain, we simulated the use of solar versus electric mains-powered refrigerators (hereafter referred to as “electric refrigerators”) at different locations in the supply chain under various circumstances. Results At their current price premium, the annual cost of each solar refrigerator is 132% more than each electric refrigerator at the district level and 241% more at health facilities. Solar refrigerators provided savings over electric refrigerators when one-day electrical outages occurred more than five times per year at either the district level or the health facilities, even when the electric refrigerator holdover time exceeded the duration of the outage. Two-day outages occurring more than three times per year at the district level or more than twice per year at the health facilities also caused solar refrigerators to be cost saving. Lowering the annual cost of a solar refrigerator to 75% more than an electric refrigerator allowed solar refrigerators to be cost saving at either level when one-day outages occurred more than once per year, or when two-day outages occurred more than once per year at the district level or even once per year at the health facilities. Conclusion Our study supports WHO and Gavi guidelines. In fact, solar refrigerators may provide savings in total cost per dose administered over electrical refrigerators when electrical outages are less frequent. Our study identified the frequency and duration at which electrical outages need to occur for solar refrigerators to provide savings in total cost per dose administered over electric refrigerators at different solar refrigerator prices.

Introduction Evidence suggests that immunization supply chains are becoming outdated and unable to deliver needed vaccines due to growing populations and new vaccine introductions. Redesigning a supply chain could result in meeting current demands. Methods The Ministries of Health in Benin in Mozambique recognized known barriers to the immunization supply chain and undertook a system redesign to address those barriers. Changes were made to introduce an informed push system while consolidating storage points, introducing transport loops, and increasing human resource capacity for distribution. Evaluations were completed in each country. Results Evaluation in each country indicated improved performance of the supply chain. The Effective Vaccine Management (EVM) assessment in Benin documented notable improvements in the distribution criteria of the tool, increasing from 40% to 100% at the district level. In Mozambique, results showed reduced stockouts at health facility level from 79% at baseline to less than 1% at endline. Coverage rates of DTP3 also increased from 68.9% to 92.8%. Discussion Benin and Mozambique are undertaking system redesign in order to respond to constraints identified in the vaccine supply chain. Results and learnings show improvements in supply chain performance and make a strong case for system redesign. These countries demonstrate the feasibility of system redesign for other countries considering how to address outdated supply chains.

Background Research has shown that the distance to the nearest immunization location can ultimately prevent someone from getting immunized. With the introduction of human papillomavirus (HPV) vaccine throughout the world, a major question is whether the target populations can readily access immunization. Methods In anticipation of HPV vaccine introduction in Mozambique, a country with a 2015 population of 25,727,911, our team developed Strategic Integrated Geo-temporal Mapping Application) to determine the potential economic impact of HPV immunization. We quantified how many people in the target population are reachable by the 1377 existing immunization locations, how many cannot access these locations, and the potential costs and disease burden averted by immunization. Results If the entire 2015 cohort of 10-year-old girls goes without HPV immunization, approximately 125 (111-139) new cases of HPV 16,18-related cervical cancer are expected in the future. If each health center covers a catchment area with a 5-km radius (ie, if people travel up to 5 km to obtain vaccines), then 40% of the target population could be reached to prevent 50 (44-55) cases, 178 (159-198) disability-adjusted life years, and US $202,854 (US $140,758-323,693) in health care costs and lost productivity. At higher catchment area radii, additional increases in catchment area radius raise population coverage with diminishing returns. Conclusions Much of the population in Mozambique is unable to reach any existing immunization location, thereby reducing the potential impact of HPV vaccine. The geospatial information system analysis can assist in planning vaccine introduction strategies to maximize access and help the population reap the maximum benefits from an immunization program.

Shared high-performance computing (HPC) platforms, such as those provided by XSEDE and Compute Canada, enable researchers to carry out large-scale computational experiments at a fraction of the cost of the cloud. Most systems require the use of distributed filesystems (e.g. Lustre) for providing a highly multi-user, large capacity storage environment. These suffer performance penalties as the number of files increases due to network contention and metadata performance. We demonstrate how a combination of two technologies, Singularity and SquashFS, can help developers, integrators, architects, and scientists deploy large datasets (O(10M) files) on these shared systems with minimal performance limitations. The proposed integration enables more efficient access and indexing than normal file-based dataset installations, while providing transparent file access to users and processes. Furthermore, the approach does not require administrative privileges on the target system. While the examples studied here have been taken from the field of neuroimaging, the technologies adopted are not specific to that field. Currently, this solution is limited to read-only datasets. We propose the adoption of this technology for the consumption and dissemination of community datasets across shared computing resources., Comment: 5 pages, 2 figures, 2 tables. Submitted to PEARC 2020 conference

This is a presentation meant to orient researchers new to the NeuroHub Platform project about our goals and high-level vision for the project.

Introduction Since 2009, CBRAIN, a collaborative, web-based research platform, has served a broad international community of researchers in performing large-scale data and computational neuroscience (Sherif, 2014). With over 800 users at 193 sites in 32 different countries and hosting over 60 software pipelines, CBRAIN has provided over 35 million CPU hours on computing resources around the world, including Stampede2 at TACC, the world's largest supercomputer dedicated to academic research. CBRAIN is a central component in the Healthy Brains for Healthy Lives infrastructure (HBHL, 2018) and the Canadian Open Neuroscience Platform (CONP, 2018), which require new features connecting existing platforms (e.g. LORIS (Das, 2016), OpenNeuro (Gorgolewski, 2017), and BrainCode (Vaccarino, 2018), utilizing community-based standards (e.g. BIDS (Gorgolewski, 2016), Boutiques (Glatard, 2018), and CARMIN (Glatard, 2015)), and offering new usage modalities and interfaces. Methods CBRAIN's primary purpose is to provide an ecosystem that abstracts away the low-level details of data movement and computational execution on advanced research computing resources. CBRAIN provides an orchestration system consisting of a central control instrument, termed a Portal, which communicates and submit tasks to remote compute servers, called Bourreaux. Portals and Bourreaux access remote data resources through passive DataProviders.The CBRAIN platform provides a unifying service layer for access to remote computing resources around the world (e.g. Compute Canada, XSEDE, and the CCC-Axis). CBRAIN is a Ruby on Rails application, is completely open source (https://github.com/aces/cbrain) and provided as a service free of charge (https://portal.cbrain.mcgill.ca). Requirements from new national and international initiatives lead to developments increasing interoperability, functionality and usability supporting a wider community and integrating with a broader set of community-driven tools and standards. Results New features developed in the CBRAIN platform are: RESTful API: A fully documented and functional CBRAIN RESTful API is now available at https://app.swaggerhub.com/apis/prioux/CBRAIN/5.1.0 and allows projects to utilize CBRAIN as a backend technology. To promote a community standard, CBRAIN will support CARMIN, a common web API for remote pipeline execution, such that any CARMIN compliant tool can use CBRAIN as a backend without rewriting their package. Datalad and S3 Integration: A Datalad DataProvider provides an interface to move data from Datalad (Datalad, 2018) versioned resources into the CBRAIN ecosystem. A new S3 DataProvider provides data movement from cloud-based resources. BIDS Compatibility: Capitalizing on the BIDS standard, an automatic parallelization capability to ensure BIDSApps run efficiently. Provenance and error-handling are also available, and the user only needs to specify the BIDS-formatted input dataset and the pipeline to execute. Boutiques Integration: To enable computational pipelines to be discoverable and shareable, and ease the burden of integration and deployment, we have adopted the Boutiques JSON standard to define our computational pipelines and to pull new pipelines from the Boutiques Repository. New Interactive Interface: A user-focused, interactive and dynamic CBRAIN user-interface has been designed, built with React.js and GraphQL, interacting through the CBRAIN API to launch tasks and manage data, The UI is highly modular and can quickly be adapted to new user and visualization requirements over time. New modular visualizations have also been developed using React.js. Conclusions CBRAIN has for nearly a decade served as a platform for accomplishing large-scale, large-data neuroinformatics. The new implementations provide a connection to a larger community of neuroinformatics and scientific research and promote FAIR standards in computational science., {"references":["CONP (2018), Canadian Open Neuroscience Platform, https://conp.ca","Das, S. (2016), 'The MNI data-sharing and processing ecosystem', NeuroImage, vol. 124, pp. 1188–1195.","Datalad (2018), available at http://www.datalad.org.","Glatard, T. (2018). Boutiques: a flexible framework to integrate command-line applications in computing platforms. GigaScience, 7, giy016.","Glatard, T. (2015). CARMIN: a common web API for remote pipeline execution. Frontiers in Neuroscience, vol. 9, no. 00053.","Gorgolewski, K.J. (2016). The brain imaging data structure, a format for organizing and describing outputs of neuroimaging experiments. Scientific Data, vol. 3, 160044.","Gorgolewski, K.J. (2017). OpenNeuro—a free online platform for sharing and analysis of neuroimaging data. Organization for Human Brain Mapping. Vancouver, Canada, 1677.","HBHL (2018), Healthy Brains for Healthy Lives, http://www.mcgill.ca/hbhl","Sherif, T. (2014), 'CBRAIN: a web-based, distributed computing platform for collaborative neuroimaging research', Front Neuroinform, vol. 8, pp. 54.","Vaccarino, A. L. (2018). Brain-CODE: A Secure Neuroinformatics Platform for Management, Federation, Sharing and Analysis of Multi-Dimensional Neuroscience Data, vol. 12, pp. 28."]}

Introduction Prior studies showed that live attenuated influenza vaccine (LAIV) is more effective than inactivated influenza vaccine (IIV) in children aged 2–8 years, supporting the Centers for Disease Control and Prevention (CDC) recommendations in 2014 for preferential LAIV use in this age group. However, 2014–2015 U.S. effectiveness data indicated relatively poor effectiveness of both vaccines, leading CDC in 2015 to no longer prefer LAIV. Methods An age-structured model of influenza transmission and vaccination was developed, which incorporated both direct and indirect protection induced by vaccination. Based on this model, the cost effectiveness of influenza vaccination strategies in children aged 2–8 years in the U.S. was estimated. The base case assumed a mixed vaccination strategy where 33.3% and 66.7% of vaccinated children aged 2–8 years receive LAIV and IIV, respectively. Analyses were performed in 2014–2015. Results Using published meta-analysis vaccine effectiveness data (83% LAIV and 64% IIV), exclusive LAIV use would be a cost-effective strategy when vaccinating children aged 2–8 years, whereas IIV would not be preferred. However, when 2014–2015 U.S. effectiveness data (0% LAIV and 15% IIV) were used, IIV was likely to be preferred. Conclusions The cost effectiveness of influenza vaccination in children aged 2–8 years is highly dependent on vaccine effectiveness; the vaccine type with higher effectiveness is preferred. In general, exclusive IIV use is preferred over LAIV use, as long as vaccine effectiveness is higher for IIV than for LAIV.