Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Segmentation of the cortical plate in fetal brain MRI with a topological loss

Priscille de Dumast, Hamza Kebiri, Chirine Atat, Vincent Dunet, Mériam Koob, Meritxell Bach Cuadra

TL;DR

This paper proposes for the first time the integration of a topological constraint, as an additional loss function, to enhance the morphological consistency of a deep learning-based segmentation of the fetal cortical plate.

Abstract

The fetal cortical plate undergoes drastic morphological changes throughout early in utero development that can be observed using magnetic resonance (MR) imaging. An accurate MR image segmentation, and more importantly a topologically correct delineation of the cortical gray matter, is a key baseline to perform further quantitative analysis of brain development. In this paper, we propose for the first time the integration of a topological constraint, as an additional loss function, to enhance the morphological consistency of a deep learning-based segmentation of the fetal cortical plate. We quantitatively evaluate our method on 18 fetal brain atlases ranging from 21 to 38 weeks of gestation, showing the significant benefits of our method through all gestational ages as compared to a baseline method. Furthermore, qualitative evaluation by three different experts on 130 randomly selected slices from 26 clinical MRIs evidences the out-performance of our method independently of the MR reconstruction quality.

Segmentation of the cortical plate in fetal brain MRI with a topological loss

TL;DR

This paper proposes for the first time the integration of a topological constraint, as an additional loss function, to enhance the morphological consistency of a deep learning-based segmentation of the fetal cortical plate.

Abstract

The fetal cortical plate undergoes drastic morphological changes throughout early in utero development that can be observed using magnetic resonance (MR) imaging. An accurate MR image segmentation, and more importantly a topologically correct delineation of the cortical gray matter, is a key baseline to perform further quantitative analysis of brain development. In this paper, we propose for the first time the integration of a topological constraint, as an additional loss function, to enhance the morphological consistency of a deep learning-based segmentation of the fetal cortical plate. We quantitatively evaluate our method on 18 fetal brain atlases ranging from 21 to 38 weeks of gestation, showing the significant benefits of our method through all gestational ages as compared to a baseline method. Furthermore, qualitative evaluation by three different experts on 130 randomly selected slices from 26 clinical MRIs evidences the out-performance of our method independently of the MR reconstruction quality.

Paper Structure

This paper contains 12 sections, 4 figures, 3 tables.

Table of Contents

  1. INTRODUCTION
  2. METHODOLOGY
  3. Topological loss
  4. Network architecture
  5. Training strategy
  6. EVALUATION
  7. Quantitative Evaluation
  8. Qualitative Evaluation
  9. RESULTS
  10. CONCLUSION
  11. ACKNOWLEDGMENTS
  12. Compliance with Ethical Standards

Figures (4)

  • Figure 1: Figure inspired and adapted from TopologyPreserving2019. Our method, TopoCP, integrates a topological loss based on persistent homology to a 2D U-Net segmentation of cortical plate fetal MRI.
  • Figure 2: Segmentation results on 35 GA atlas. (a) T2w (left) and GT segmentation overlaid (right). (b) Baseline U-Net and (c) TopoCP: predicted likelihood (left) and estimated segmentation (right). Likelihood probabilities: 0 1. Case 1 illustrates a net improvement in the segmentation of the midsagittal area and frontal cortical foldings. Case 2 shows a more accurate detection of the deep sulci with TopoCP.
  • Figure 3: Evolution of the performance metrics for atlas images from 21 to 38 weeks of gestation.
  • Figure 4: Segmentation results on 23 (top) and 32 (bottom) gestational weeks fetuses: (a) T2w SR image; (b) Baseline U-Net (c) and TopoCP.