The Gonzo Dataset: Human Brain MRI Data of CSF Tracer Evolution Over 72h For Data-Integrated Simulations

We present the Gonzo dataset: brain MRI and derivative data of one healthy-appearing male human volunteer before and during the 72 hours after injection of the contrast agent gadobutrol into the cerebrospinal fluid (CSF) of the spinal canal (intrathecal injection). The data records show the temporal and spatial evolution of the contrast agent in CSF, brain, and adjacent structures. The MRI data includes T1-weighted images, Look-Locker inversion recovery (LL, a technique to determine T1 values), a mixed inversion-recovery spin-echo sequence (Mixed) for all time points (one pre-contrast and four post-contrast acquisitions) and, in addition, T2-weighted, FLAIR, and dynamic DTI data for the pre-contrast session. In addition to raw data, we provide derivatives with the goal of allowing for numerical simulations of the studied tracer transport process. This includes T1 maps (from LL and Mixed) and tracer concentration maps, diffusion tensor maps, as well as unstructured triangulated volume meshes of the brain geometry and associated field data (MRI and derived data mapped onto the computational mesh). We provide brain region markers obtained with a FreeSurfer-based analysis pipeline. An initial regional statistical analysis of the data is presented. The data can be used to study the transport behaviour and the underlying processes of a tracer in the human brain. Tracer transport is both relevant to study water transport as well as new pathways for drug delivery. The composition of the data set allows both reuse by the image processing and the simulation science communities. The dataset is meant to contribute and inspire new studies into the understanding of transport processes in the brain and into method development regarding image analysis and simulation of transport processes.

Full data descriptor is available at: https://doi.org/10.1101/2025.07.23.25331971

Repositories with custom software used in the post-processing and usage example:

• https://github.com/jorgenriseth/gonzo
• https://github.com/jorgenriseth/gMRI2FEM
• https://github.com/timokoch/dumux-braindiffusion-miniapp