Adapting the UK Biobank brain imaging protocol and analysis pipeline for the C-MORE multi-organ study of COVID-19 survivors

Abstract

SARS-CoV-2 infection has been shown to damage multiple organs, including the brain. Multiorgan MRI can provide further insight on the repercussions of COVID-19 on organ health but requires a balance between richness and quality of data acquisition and total scan duration. We adapted the UK Biobank brain MRI protocol to produce high-quality images while being suitable as part of a post-COVID-19 multiorgan MRI exam. The analysis pipeline, also adapted from UK Biobank, includes new imaging-derived phenotypes (IDPs) designed to assess the effects of COVID-19. A first application of the protocol and pipeline was performed in 51 COVID-19 patients post-hospital discharge and 25 controls participating in the Oxford C-MORE study. The protocol acquires high resolution T1, T2-FLAIR, diffusion weighted images, susceptibility weighted images, and arterial spin labelling data in 17 minutes. The automated imaging pipeline derives 1575 IDPs, assessing brain anatomy (including olfactory bulb volume and intensity) and tissue perfusion, hyperintensities, diffusivity, and susceptibility. In the C-MORE data, these quantitative measures were consistent with clinical radiology reports. Our exploratory analysis tentatively revealed that recovered COVID-19 patients had a decrease in frontal grey matter volumes, an increased burden of white matter hyperintensities, and reduced mean diffusivity in the total and normal appearing white matter in the posterior thalamic radiation and sagittal stratum, relative to controls. These differences were generally more prominent in patients who received organ support. Increased T2* in the thalamus was also observed in recovered COVID-19 patients, with a more prominent increase for non-critical patients. This initial evidence of brain changes in COVID-19 survivors prompts the need for further investigations. Follow-up imaging in the C-MORE study is currently ongoing, and this protocol is now being used in large-scale studies. The pipeline is widely applicable and will contribute to new analyses to hopefully clarify the medium to long-term effects of COVID-19.

Reliability of multi-modal MRI-derived brain phenotypes for multi-site assessment of neuropsychiatric complications of SARS-CoV-2 infection

“…For the COVID-CNS project, it provides insights that can guide the design of harmonisation strategies for the project. More broadly, the study is of relevance to the expanding number of studies utilising multi-modal imaging protocols derived from the UK Biobank, including a number of additional studies focused on the neurological impact of COVID-19 (Douaud et al, 2021;Raman et al, 2021) . Overall, our results demonstrate generally good to excellent levels of between-site reliability of imaging derived phenotypes estimated across a wide range of brain MRI modalities in data collected from 4 UK sites participating in a national COVID research consortium.…”

“…Pipeline changes were implemented to accommodate minor differences in imaging parameters between UKB and COVID-CNS protocols, to analyse MRI modalities not included in the UKB MRI protocol, e.g., ASL, and to analyse MRI data acquired using the GE scanner at KCL. Where protocols matched exactly, analysis pipelines were identical to those used in the C-MORE COVID study Raman et al, 2021) . Summaries of pre-processing and IDP estimation are provided below for individual modalities, with further details available in (Douaud et al, 2021;Griffanti et al, 2021) .…”

“…Where protocols matched exactly, analysis pipelines were identical to those used in the C-MORE COVID study Raman et al, 2021) . Summaries of pre-processing and IDP estimation are provided below for individual modalities, with further details available in (Douaud et al, 2021;Griffanti et al, 2021) . For presentation, IDPs reflecting the same phenotypic properties were grouped together into IDP classes (Douaud et al, 2021;Elliott et al, 2018) .…”

“…Sub-cortical volumes were estimated utilizing population priors on shape and intensity variation across subjects (Patenaude et al, 2011) . Using an additional non-linear registration procedure, regional volumes of the olfactory bulbs were estimated using T1-weighted, T2-FLAIR and dMRI data, and a template derived from over 700 UKB individuals (Arthofer et al, 2021;Griffanti et al, 2021;Lange et al, 2020) . T2-FLAIR pre-processing was very similar to the T1w pipeline (with the T1-weighted image used for registration to the MNI standard template).…”

“…Images were segmented using BIANCA to identify white matter (WM) hyperintensities (WMH) (Griffanti et al, 2016) , using the UKB BIANCA training file. Periventricular WMH (pWMH) and deep WMH (dWMH) volumes were defined for complementary subsets of total WM hyperintensities that were, respectively, less than (or more than) 10 mm distant from the lateral ventricles (Griffanti et al, 2021) .…”