BRAIN project: OpenNeuroPET: An Archive for PET data

National Institute Of Mental Health

Investigators

Linked publications & trials

Abstract

IIn recent years, the importance of data sharing has increasingly been recognized by the neuroimaging community because of the poor replicability of findings, the need for appropriate quality control, the greater statistical power provided by larger samples, and the higher scientific impact of multilateral collaborations. Our funding bodies and scientific journals also increasingly encourage or require that the data be shared. However, data sharing can be clumsy and cumbersome in the absence of well-articulated guidelines, clear data formatting standards, and well-managed repositories. Building on the recommendations of leaders in the PET field, and in collaboration with other highly successful repositories of human neuroimaging data, an application was submitted to the NIH BRAIN initiative for funding to establish a useful and freely-shared repository for human brain PET data. That application was favorably reviewed and awarded funding in 2021 and our work commenced. As a first step, I collaborated with Dr. Gitte Knudsen and a large international group of leaders in PET brain imaging to publish a paper (Knudsen et al 2020) that specified the content and format for publications and archives; these guidelines have been adopted worldwide, bringing much-needed cohesion to the field. Building on these guidelines, the goals of the BRAIN grant are threefold: 1) to establish the OpenNeuroPET Archive; 2) to support the use and adoption of the OpenNeuroPET Archive by reaching out to the community, maintaining a helpdesk, and ensuring compliance with international regulations; and 3) to establish and provide human molecular imaging brain atlases (MIBAs) that provide the mean three-dimensional distribution in brain of molecular imaging targets in aggregated samples of people. Four institutions are collaborating to accomplish these goals: the NIMH Intramural Research Program, Copenhagen University Hospital (in collaboration with Dr. Gitte Knudsen), Massachusetts General Hospital (in collaboration with Dr. Douglas Greve), and Stanford University (in collaboration with Dr. Russell Poldrack). The next step was to complete a highly detailed specification of all the data that needed to be entered. The 17-page document was reviewed by the field, then approved by the BIDS Oversight Group as the accepted PET BIDS standard. We initially considered whether it would be preferable to work with individual centers to help them properly format their existing data or to create a software system that would allow them to easily and usefully acquire future data. We chose the latter path and, through the newly established OpenNeuroPET Archive, are establishing a comprehensive data analysis pipeline for PET data. This pipeline will automatically require that all data be entered and stored in the proper format that can be seamlessly uploaded to the Archive. The data analysis pipeline is user friendly, validated by leaders in the field, and functional enough that researchers will choose to use it, regardless of whether they deposit data to the Archive. Furthermore, the current data analysis pipeline is a major source of variability and errors in brain imaging data, a problem well known in functional MRI (fMRI) that also occurs in PET. Our software system, built to circumvent such errors, is thus likely to be useful for both current and future data acquisition, ensuring that the field will be strongly motivated to maintain it. The data analysis pipeline for the OpenNeuroPET Archive is comprehensive, aims to be applicable to all PET imaging centers, is opensource, and includes the following features: 1) Conversion software and associated documentation to convert all primary data (PET, MRI, and plasma) into BIDS format, ensuring that data can be easily shared between centers. 2) Software for pharmacokinetic modeling of brain and plasma data using PETSurfer. 3) Accessible quality control datasets that can be used to confirm the reproducibility of the analytic pipeline, given that existing pipelines rely on less accessible datasets. 4) Molecular Imaging Brain Atlases (MIBAs). Most non-PET researchers do not have the expertise necessary to analyze primary PET data themselves. Thus, another key facet of the OpenNeuroPET Archive is the analysis and cross-subject averaging of the archived PET data to create atlases for each molecular targetthat is, MIBAs for the human brain. The proposed MIBAs would be analogous to the Allen Human Brain Atlas, which reports the regional distribution in postmortem human brain of the densities of gene transcripts. The MIBAs can be displayed and used for quantitation and correlation with other imaging modalities, including MRI and fMRI. This software package is currently being tested at several PET centers to ensure that it robustly interfaces with the large variety of software systems and cameras. My collaborators and I expect a complete software package to be available by September 2023. After that, we will continue to make improvements as requested by the field, have educational sessions at international meetings, maintain a HelpDesk, and curate the deposited datasets. Most of this curation will be performed by computer that is, using a validator to ensure that each submission conforms to the PET BIDS Standard. Because some archives have closed after funding expires (e.g., the fMRI Data Center at UCLA), another goal of the OpenNeuroPET Archive is to make the Archive compatible with the NIMH Data Archive, which could then assume maintenance of the system after the Brain grant's termination in 2026.