A Robust Modular Automated Neuroimaging Pipeline for Model Inputs to TheVirtualBrain

TheVirtualBrain, an open-source platform for large-scale network modeling, can be personalized to an individual using a wide range of neuroimaging modalities. With the growing number and scale of neuroimaging data sharing initiatives of both healthy and clinical populations comes an opportunity to create large and heterogeneous sets of dynamic network models to better understand individual differences in network dynamics and their impact on brain health. Here we present TheVirtualBrain-UK Biobank pipeline, a robust, automated and open-source brain image processing solution to address the expanding scope of TheVirtualBrain project. Our pipeline generates connectome-based modeling inputs compatible for use with TheVirtualBrain. We leverage the existing multimodal MRI processing pipeline from the UK Biobank made for use with a variety of brain imaging modalities. We add various features and changes to the original UK Biobank implementation specifically for informing large-scale network models, including user-defined parcellations for the construction of matching whole-brain functional and structural connectomes. Changes also include detailed reports for quality control of all modalities, a streamlined installation process, modular software packaging, updated software versions, and support for various publicly available datasets. The pipeline has been tested on various datasets from both healthy and clinical populations and is robust to the morphological changes observed in aging and dementia. In this paper, we describe these and other pipeline additions and modifications in detail, as well as how this pipeline fits into the TheVirtualBrain ecosystem.


FC_homotopic_mean_fMRI
Homotopic functional connectivity mean

Category: tvb_IDP_func_TSNR
Temporal signal-to-noise ratios in fMRI fMRI_TSNR Temporal signal-to-noise ratio in the pre-processed fMRI -reciprocal of median (across brain voxels) of voxelwise mean intensity divided by voxelwise timeseries standard deviation fMRI_cleaned_TSNR Temporal signal-to-noise ratio in the artefact-cleaned pre-processed fMRI -reciprocal of median (across brain voxels) of voxelwise mean intensity divided by voxelwise timeseries standard deviation fMRI_num_vol Number of volumes (timepoints) in fMRI scan

Category: tvb_IDP_func_susceptibility_SNR 3
Temporal signal-to-noise ratios in susceptible and non-susceptible brain regions in fMRI 3 fMRI_non-susceptible_TSNR Temporal signal-to-noise ratio in the minimallyprocessed fMRI non-susceptible regions -median (across brain voxels) of voxelwise mean intensity divided by voxelwise timeseries standard deviation fMRI_non-susceptible_cleaned_TSNR Temporal signal-to-noise ratio in the artefact-cleaned pre-processed fMRI non-susceptible regions -median (across brain voxels) of voxelwise mean intensity divided by voxelwise timeseries standard deviation fMRI_susceptible_TSNR Temporal signal-to-noise ratio in the minimallyprocessed fMRI susceptible regions -median (across brain voxels) of voxelwise mean intensity divided by voxelwise timeseries standard deviation fMRI_susceptible_cleaned_TSNR Temporal signal-to-noise ratio in the artefact-cleaned pre-processed fMRI susceptible regions -median (across brain voxels) of voxelwise mean intensity divided by voxelwise timeseries standard deviation

Category: tvb_IDP_all_align_to_T1
Discrepancy between various modalities registered to T1 space and the T1 image

T2_FLAIR_align_to_T1
Discrepancy between the T2_FLAIR brain image (linearly-aligned to the T1) and the T1 brain image dMRI_align_to_T1 Discrepancy between the dMRI brain image (linearlyaligned to the T1) and the T1 brain image fMRI_align_to_T1 Discrepancy between the fMRI brain image (linearlyaligned to the T1) and the T1 brain image fMRI_fieldmap_align_to_T1 Discrepancy between the fMRI gradient echo field map brain image (linearly-aligned to the T1) and the T1 brain image

Category: tvb_IDP_fieldmap_func_align
Discrepancy between the fMRI field map brain image registered to fMRI func space and the fMRI func image fMRI_fieldmap_func_align Discrepancy between the fMRI gradient echo field map brain image and the fMRI image  Figure 2. T1 image analysis from the QC Report. Image shows a coronal, axial, and sagittal view of a subject T1 image (grayscale). Motion artifacts and atrophy are evident (red arrows).

Supplementary
Supplementary Figure 3. T1 brain extraction analysis from the QC Report. Images show a sagittal view of a subject's brain mask (yellow) overlaid on top of their T1w image (grayscale). A slice from a well-processed subject can be seen on the left. A slice from a subject with insufficient brain inclusion in the mask can be seen in the middle (red arrow). A slice from a subject with excess dura mater inclusion in the mask can be seen on the right (light blue arrow).
Supplementary Figure 4. T1 registration analysis from the QC Report. Images show a sagittal view of the MNI standard (red) overlaid on top of a subject's T1w image (grayscale). Two slices from a well-processed subject can be seen on the left and two slices from a poorly-processed subject can be seen on the right. The arrow (blue) points to an instance of poor alignment.
Supplementary Figure 5. T1 White Matter Segmentation from the QC Report. Images show a sagittal view of a subject's white matter mask (yellow) overlaid on top of their T1w image (grayscale). Two slices from a well-processed subject can be seen on the left. Two slices from a subject with misclassified white matter from the dura can be seen on the right (red arrow).
Supplementary Figure 6. T1 ROI segmentation analysis from the QC Report. Images show a view of a subject's labelled grey matter ROIs (coloured) overlaid on top of their T1w image (grayscale). Three slices from a well-processed subject can be seen on top and three slices from a poorlyprocessed subject, where grey matter labelling is sparse, can be seen on bottom.