QA/QC of the unprocessed anatomical, functional, and diffusion MRI data of the Human Connectome PHantom (HCPh) dataset with MRIQC

Motivation: Reliable neuroimaging pipelines require the implementation of robust QA/QC protocols. Goal(s): Demonstrating a comprehensive QA/QC checkpoint on ‘unprocessed’ data of a mid-size dataset with MRIQC. Approach: We employ MRIQC in the visual assessment and training of automatic QC to identify data that must be excluded or flagged within the ‘Human Connectome PHantom’ (HCPh) project. Results: We developed a QA/QC protocol for unprocessed data within the HCPh project with MRIQC, comprehensively describing predefined exclusion criteria. We then demonstrate the application of the protocol to the corresponding data and report the outcomes.Impact: We demonstrate how to streamline QA in a neuroimaging workflow, establishing robust QA/QC protocols with MRIQC. This approach adds to the tooling available to improve neuroimaging analyses, ensuring more accurate and reproducible results.