A Robust Modular Automated Neuroimaging Pipeline for Model Inputs to TheVirtualBrain.

Area of Science:

• Neuroscience
• Computational Biology
• Medical Imaging

Background:

• The Virtual Brain (TVB) is an open-source platform for personalized large-scale brain network modeling.
• Increasing neuroimaging data sharing necessitates robust processing pipelines for heterogeneous datasets.
• Understanding individual differences in brain network dynamics is crucial for brain health research.

Purpose of the Study:

• To present the VirtualBrain-UK Biobank pipeline, an automated, open-source solution for processing UK Biobank neuroimaging data.
• To generate connectome-based modeling inputs compatible with The Virtual Brain platform.
• To enhance the Virtual Brain project's capacity to analyze large and diverse brain imaging datasets.

Main Methods:

• Leveraged the UK Biobank's multimodal MRI processing pipeline.
• Implemented user-defined parcellations for creating whole-brain functional and structural connectomes.
• Integrated quality control reports, streamlined installation, modular packaging, and updated software versions.

Main Results:

• Developed a robust pipeline generating connectome data suitable for large-scale network models.
• Ensured compatibility with various brain imaging modalities and public datasets.
• Validated the pipeline's robustness across healthy and clinical populations, including aging and dementia.

Conclusions:

• The VirtualBrain-UK Biobank pipeline effectively processes neuroimaging data for personalized brain network modeling.
• The pipeline supports large-scale, heterogeneous datasets, advancing the understanding of brain health and individual differences.
• This open-source solution integrates seamlessly into The Virtual Brain ecosystem, facilitating broader research applications.