Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

deepmriprep: Voxel-based Morphometry (VBM) Preprocessing via Deep Neural Networks

Lukas Fisch, Nils R. Winter, Janik Goltermann, Carlotta Barkhau, Daniel Emden, Jan Ernsting, Maximilian Konowski, Ramona Leenings, Tiana Borgers, Kira Flinkenflügel, Dominik Grotegerd, Anna Kraus, Elisabeth J. Leehr, Susanne Meinert, Frederike Stein, Lea Teutenberg, Florian Thomas-Odenthal, Paula Usemann, Marco Hermesdorf, Hamidreza Jamalabadi, Andreas Jansen, Igor Nenadic, Benjamin Straube, Tilo Kircher, Klaus Berger, Benjamin Risse, Udo Dannlowski, Tim Hahn

TL;DR

Deepmriprep is a neural network-based pipeline that performs all necessary preprocessing steps for VBM analysis of T1-weighted MR images using deep neural networks and matches CAT12 in accuracy for tissue segmentation and image registration across more than 100 datasets and shows strong correlations in VBM results.

Abstract

Voxel-based Morphometry (VBM) has emerged as a powerful approach in neuroimaging research, utilized in over 7,000 studies since the year 2000. Using Magnetic Resonance Imaging (MRI) data, VBM assesses variations in the local density of brain tissue and examines its associations with biological and psychometric variables. Here, we present deepmriprep, a neural network-based pipeline that performs all necessary preprocessing steps for VBM analysis of T1-weighted MR images using deep neural networks. Utilizing the Graphics Processing Unit (GPU), deepmriprep is 37 times faster than CAT12, the leading VBM preprocessing toolbox. The proposed method matches CAT12 in accuracy for tissue segmentation and image registration across more than 100 datasets and shows strong correlations in VBM results. Tissue segmentation maps from deepmriprep have over 95% agreement with ground truth maps, and its non-linear registration, using supervised SYMNet, predicts smooth deformation fields comparable to CAT12. The high processing speed of deepmriprep enables rapid preprocessing of extensive datasets and thereby fosters the application of VBM analysis to large-scale neuroimaging studies and opens the door to real-time applications. Finally, deepmripreps straightforward, modular design enables researchers to easily understand, reuse, and advance the underlying methods, fostering further advancements in neuroimaging research. deepmriprep can be conveniently installed as a Python package and is publicly accessible at https://github.com/wwu-mmll/deepmriprep.

deepmriprep: Voxel-based Morphometry (VBM) Preprocessing via Deep Neural Networks

TL;DR

Deepmriprep is a neural network-based pipeline that performs all necessary preprocessing steps for VBM analysis of T1-weighted MR images using deep neural networks and matches CAT12 in accuracy for tissue segmentation and image registration across more than 100 datasets and shows strong correlations in VBM results.

Abstract

Voxel-based Morphometry (VBM) has emerged as a powerful approach in neuroimaging research, utilized in over 7,000 studies since the year 2000. Using Magnetic Resonance Imaging (MRI) data, VBM assesses variations in the local density of brain tissue and examines its associations with biological and psychometric variables. Here, we present deepmriprep, a neural network-based pipeline that performs all necessary preprocessing steps for VBM analysis of T1-weighted MR images using deep neural networks. Utilizing the Graphics Processing Unit (GPU), deepmriprep is 37 times faster than CAT12, the leading VBM preprocessing toolbox. The proposed method matches CAT12 in accuracy for tissue segmentation and image registration across more than 100 datasets and shows strong correlations in VBM results. Tissue segmentation maps from deepmriprep have over 95% agreement with ground truth maps, and its non-linear registration, using supervised SYMNet, predicts smooth deformation fields comparable to CAT12. The high processing speed of deepmriprep enables rapid preprocessing of extensive datasets and thereby fosters the application of VBM analysis to large-scale neuroimaging studies and opens the door to real-time applications. Finally, deepmripreps straightforward, modular design enables researchers to easily understand, reuse, and advance the underlying methods, fostering further advancements in neuroimaging research. deepmriprep can be conveniently installed as a Python package and is publicly accessible at https://github.com/wwu-mmll/deepmriprep.
Paper Structure (30 sections, 10 equations, 18 figures, 2 tables)

This paper contains 30 sections, 10 equations, 18 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Materials and Methods
  3. Datasets
  4. Training and Validation Datasets
  5. OpenNeuro-Total
  6. OpenNeuro-HD
  7. Test Datasets
  8. Marburg-Münster Affective Disorders Cohort Study (FOR2107/MACS)
  9. Münster Neuroimaging Cohort (MNC)
  10. BiDirect
  11. Input Preprocessing and Ground Truth
  12. Data Augmentation
  13. Tissue Segmentation
  14. Patchwise UNet
  15. Multi-Level Activation Function
  16. ...and 15 more sections

Figures (18)

  • Figure 1: Distribution of age, sex, image resolution, and scanner field strength across 8,279 subjects from OpenNeuro-Total (left) and 685 subjects from OpenNeuro-HD (right).
  • Figure 2: Distribution of age and sex across 1,799 subjects from the Marburg-Münster Affective Disorders Cohort Study (left), 1,194 samples from the Münster Neuroimaging Cohort, and 1,024 subjects from the BiDirect cohort (right).
  • Figure 3: All utilized image augmentations (right) compared to the original image (left).
  • Figure 4: Coronal (left), sagittal (middle), and axial slice (right) of the 128x128x128 voxel patches resulting from the optimized patch positioning. For reference, the T1 template of CAT12, upsampled to the utilized resolution of 0.5mm, is shown in the background.
  • Figure 5: Left: Step activation function $f$ which maps logits to tissue values between 0 and 3. Middle: Activation function $f$ applied to 100,000 values drawn from the normal distribution $N(1,1)$. Right: Tissue value distribution in the MNI template.
  • ...and 13 more figures