Data Core
Data Core
批准号:
10729443
负责人:
Aaron James Kaat
金额:
$71.36万
依托单位国家:
美国
项目类别:
财政年份:
2018
资助国家:
美国
项目状态:
未结题
起止时间:
2018-09-15 至 2028-05-31
关键词:
AdherenceAlzheimer&aposs DiseaseAlzheimer&aposs disease related dementiaAutomatic Data ProcessingAwardBehavioralBig DataCellular PhoneClinicalClinical DataCodeCognitiveCommunitiesComputing MethodologiesConsultationsDataData AggregationData AnalysesData AnalyticsData CollectionData ProvenanceData ScientistData SetData Storage and RetrievalDecentralizationDevelopmentEnsureFutureGoalsInformation SystemsInfrastructureInternationalLearningLengthLibrariesLife Cycle StagesLocationLongitudinal StudiesLongitudinal, observational studyMachine LearningMeasurementMeasuresMeta-AnalysisMetadataMethodologyMonitorMovementMultiple SclerosisNoiseOutcome MeasureParkinson DiseaseParticipantPatient Outcomes AssessmentsPatient Self-ReportPatternPerformancePopulationProcessPsoriasisPsychometricsQuality ControlReportingReproducibilityReproducibility of ResultsResearchResearch DesignResearch PersonnelResearch Project GrantsResourcesSamplingScienceScoring MethodServicesSpecific qualifier valueStatistical Data InterpretationStructureSurveysSynapsesSystemUnited States National Institutes of HealthValidationValidity and ReliabilityWorkcardiovascular healthcloud basedcognitive abilitycognitive changedata analysis pipelinedata cleaningdata managementdata pipelinedata repositorydata reusedata sharingdata standardsdensitydesigndirect applicationexperiencehigh standardinformantinnovationinterestlongitudinal analysismHealthmultidisciplinarynovelplatform as a servicerepositoryresearch studyresponsesleep qualitysocialsoftware as a servicesupervised learningsupport toolstoolvalidation studies
中文摘要
数据与统计分析核心摘要
英文摘要
Data and Statistical Analysis Core Summary
The Data and Statistical Analysis Core (Data Core) will provide the data management and analytic support
tools needed throughout the Mobile Toolbox 2 initiative. We will provide a standard data processing pipeline and
centralized data storage center that documents the original data, as well as data provenance for derived scores,
metadata, and paradata regarding measure completion. These processes are necessary to ensure the
reproducibility of results arising from the Mobile Toolbox, as well as the reliability and validity of the measures.
The Data Core will support the Project by providing statistical and psychometric expertise in the development of
new measures and analysis of longitudinal observational studies. Further, the centralized repository developed
and maintained by the Data Core will be a key component of the Extension and Dissemination Core in their
efforts to disseminate data for secondary research purposes. The Mobile Toolbox system and data management
infrastructure was built as an extension of the Sage Synapse Data Platform, a cloud-based data storage and
organization system developed to facilitate sharing data, data analysis, and computational methods distributed
across research teams. The Data Core will be integral in serving the needs of Mobile Toolbox 2. The primary
aims of the core include: Aim 1: Conduct statistical analysis and methodological consultation services for
the Mobile Toolbox; Aim 2: Extend and maintain a central repository for data aggregation, curation, and
dissemination; and Aim 3: Create a learning system to assist in the deployment, design, and execution
of remote studies. The Data Core will manage and conduct analyses needed for psychometric validation and
norming of any new measures identified for inclusion in the Mobile Toolbox assessment library through the
scientific community-engaged approach utilized within the Project. We will develop new normative values and
conduct multi-pronged validation analyses as warranted within diverse and representative samples. This work
will be performed by the team of statisticians and psychometricians responsible for norming and validating the
NIH Toolbox®, Patient Reported Outcome Measurement Information System®, and International Cognitive Ability
Resource. The Data Core will ensure that any new measures can be included within the automated data
processing pipeline, including ensuring that the measures are well-defined using JSON schemas so that the data
are self-describing. This allows the data to be seamlessly integrated into the centralized data repository. Further,
this repository allows data aggregation across studies, supporting novel and innovative “big data” approaches.
The Data Core will enable learning systems to aggregate data and implement unsupervised and semi-supervised
machine learning approaches across studies.
期刊论文(0)
专著(0)
科研奖励(0)
会议论文