mIQa: A Highly Scalable and Customizable Platform for Medical Image Quality Assessment

Project Summary NIH is increasing its investment in large mutli center brain MRI studies via projects such as the recently announced BRAIN initiativeThe success of these studies depends on the quality of MRIs and the resulting image measurementsregardless of sample sizeEven though quality control of MRIs and corresponding measurements could be outsourcedmost neuroscience studies rely on in house procedures that combine automatically generated scores with manually guided checkssuch as visual inspectionImplementing these procedures typically requires combining several open source software systemsFor examplethe NIH NIAAA and BD K funded Data Analysis ResourceDARof the National Consortium on Alcohol and Neurodevelopment in AdolescenceNCANDAuses XNAT to consolidate the structuraldiffusionand functional MRIs acquired across five sitesand has also developed their own custom software package to comply with study requirements that called for a multi tierquality controlQCworkflowHoweverthese customone off tools lack support for multi site QC workflows as that would require a unified platformdesign that supports collaboration and sharingand strong cohesion between technologiesTo improve the effectiveness of QC efforts specific to multi center neuroimaging studieswe will develop a widely accessible and broadly compatible software platform that supports simplified creation of custom QC workflows in compliance with study requirementsprovides core functionality for performing QC of medical imagesand automatically generates documentation compliant with the FAIR principlei emaking scientific findings findableaccessibleinteroperableand reusableSpecificallyour multi site open source software platform for Medical Image Quality AssurancemIQawill enable efficient and accurate QC processing by leveraging open sourcestate of the art web interface technologiessuch as a web based dataset caching systemmachine learning to aid in QC processand an interactive electronic notebook platformUsers will be able to configure workflows that not only reflect the specific requirements of medical imaging studies but also minimize the time spent on labor intensive operationssuch as visually reviewing scansIssue tracking technology will enhance communication between geographically distributed team membersas they can easily share image annotations and receive automating notifications of outstanding QC issuesThe system will be easy to deploy as it will be able to interface with various imaging storage backendssuch as local file systems and XNATWhile parts of this functionality have been developed elsewheremIQa is unique as it provides a unifiedstandard interface for efficient QC setupmaintenanceand review for projects analyzing multipleindependently managed data sourcesThe usefulness of this unique QC system will be demonstrated on increasing the efficiency of the diverse QC team of the multi center NCANDA study Narrative The goal of this proposal is to develop multi siteopen source software for Medical Image Quality AssurancemIQato address the QC needs of geographically diverse teams using small and large medical image based studies alikemIQa will enable efficient and accurate QC processing by levering open sourcestate of the art machine learningdata managementand web interface technologiesOur effort will minimize the time spent on labor intensive review and analysis operations by supporting team oriented reviewing that is guided by highly customizable workflows seamlessly interacting with existing data management systems