KinderSeg: FastSurfer Database for Age-Specific Brain Volumes in Healthy Children

Code for the paper:

 Ibrahim Zughayyar, Martin Bauer, Christopher Güttler, Ana Luísa Marcelino, Fabienne Kühne, Claudia Buss, Christine Heim, Annette Aigner, 
 Anna Tietze, and Andrea Dell'Orco 
(2024, September 9).
A FastSurfer Database for Age-Specific Brain Volumes in Healthy Children:
A Tool for Quantifying Localized and Global Brain Volume Alterations in Pediatric Patients.

https://doi.org/10.31219/osf.io/dw7p4

https://osf.io/kj7hy/

Run with Docker

git clone https://github.com/ibrazug/kinderseg.git
cd kinderseg/Docker
cp </path/to/your/license.txt> .
docker build -t kinderseg .
docker run --gpus all --rm \
    -v <nifti_data>:/data \
    -v <output_dir>:/output \
    kinderseg \
    --age <age> \
    --threads <threads, optional, default=4>

nifti_data can be either a directory containing NIfTI files or a single NIfTI file. The output will be saved in the output directory in a subdirectory named after the input NIfTI file, without the extension. Example:

docker run --gpus all --rm \
    -v $PWD/sub-001/ses-01/anat/sub-001_ses-01_T1w.nii.gz:/data/sub-001_ses-01_T1w.nii.gz \
    -v ./output:/output \
    kinderseg --age 12

Run with Apptainer/Singularity

git clone https://github.com/ibrazug/kinderseg.git
cd kinderseg/Docker
cp </path/to/your/license.txt> .
docker build -t kinderseg .
apptainer build kinderseg.sif docker-daemon://kinderseg:latest 
apptainer run --nv kinderseg.sif \
    -B <nifti_data>:/data \
    -B <output_dir>:/output \
    --age <age> \
    --threads <threads>

Requirements

• At least 5 GB of RAM
• GPU (optional, but recommended)
• a valid FreeSurfer license: you can get one for free by clicking here.

Run ShinyApp Locally

To run the ShinyApp locally, follow these steps:

1. Ensure you have R Studio installed on your machine. Recommended version: RStudio 2023.06.2 Build 561.
2. Download this repository by clicking here.
3. Navigate to the shinyapp folder within the downloaded repository.
4. Open app.R using R Studio.
5. Install the shiny package by running install.packages("shiny") in the R console if you haven't already.
6. Run the App by clicking on the Run App button in R Studio.

This will start the ShinyApp locally on your machine.

Dataset Statistics

Dataset	N° in total	Ex (QC)	N° after QC	Ex (SC)	N° after SC
HBN	170	25	145	3	142
LOC	125	20	105	7	98
Kids2Health	211	7	204	3	201
Total	506	52	454	13	441

MRI Segmentation and Volumetric Analysis of the dataset subjects

MRI Post-processing

• DICOM data converted to NIfTI format using dcm2niix and Python version 3.9 was used for data manipulation

Segmentation Tools

1. FreeSurfer v6.0: Ran on a Slurm cluster with a 16-core Intel Xeon E5-2650 CPU, Utilized GNU Parallel for parallel processing

2. FastSurfer (V1): Utilized FastSurferCNN pipeline on a workstation with an AMD 3970X CPU and NVIDIA GeForce RTX 3090 GPU

Files:

• 01.generate_segmentation_outputs (freesurfer and fastsurfer).ipynb for MRI data segmentation and volume calculation using Freesurfer and Fastsurfer, generating volumetric statistics tables for both methods
• 02.SumROIs (masks and volumes).ipynb defining and summing volumes for predefined ROIs, and saving the results in CSV files
• Python_environment.yml Python env

Data Analysis

• R version 4.1.2 used for analysis
• Libraries: ggplot2, tidyverse, viridis (R)

Files:

• DSC_RVD.ipynb showing how we calculated DSC and RVD values and how the figures were generated
• HAI.ipynb showing how HAI values were calculated for FastSurfer and FreeSurfer outputs
• sanity_check_plot Comparison to findings in the study by Bethlehem et al. (2022)
• ICC_and_mean_volumes ICC Agreement Calculation and mean Volumes for each ROI.
• Percetilcurves_LR Established using FastSurfer for the entire database
• R_environment R env