Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Brain-Shift: Unsupervised Pseudo-Healthy Brain Synthesis for Novel Biomarker Extraction in Chronic Subdural Hematoma

Baris Imre, Elina Thibeau-Sutre, Jorieke Reimer, Kuan Kho, Jelmer M. Wolterink

TL;DR

This work proposes a novel method using anatomy-aware unsupervised diffeomorphic pseudo-healthy synthesis to generate brain deformation fields and indicates that automatically obtained brain deformation fields might contain prognostic value for personalized cSDH treatment.

Abstract

Chronic subdural hematoma (cSDH) is a common neurological condition characterized by the accumulation of blood between the brain and the dura mater. This accumulation of blood can exert pressure on the brain, potentially leading to fatal outcomes. Treatment options for cSDH are limited to invasive surgery or non-invasive management. Traditionally, the midline shift, hand-measured by experts from an ideal sagittal plane, and the hematoma volume have been the primary metrics for quantifying and analyzing cSDH. However, these approaches do not quantify the local 3D brain deformation caused by cSDH. We propose a novel method using anatomy-aware unsupervised diffeomorphic pseudo-healthy synthesis to generate brain deformation fields. The deformation fields derived from this process are utilized to extract biomarkers that quantify the shift in the brain due to cSDH. We use CT scans of 121 patients for training and validation of our method and find that our metrics allow the identification of patients who require surgery. Our results indicate that automatically obtained brain deformation fields might contain prognostic value for personalized cSDH treatment. Our implementation is available on: github.com/Barisimre/brain-morphing

Brain-Shift: Unsupervised Pseudo-Healthy Brain Synthesis for Novel Biomarker Extraction in Chronic Subdural Hematoma

TL;DR

This work proposes a novel method using anatomy-aware unsupervised diffeomorphic pseudo-healthy synthesis to generate brain deformation fields and indicates that automatically obtained brain deformation fields might contain prognostic value for personalized cSDH treatment.

Abstract

Chronic subdural hematoma (cSDH) is a common neurological condition characterized by the accumulation of blood between the brain and the dura mater. This accumulation of blood can exert pressure on the brain, potentially leading to fatal outcomes. Treatment options for cSDH are limited to invasive surgery or non-invasive management. Traditionally, the midline shift, hand-measured by experts from an ideal sagittal plane, and the hematoma volume have been the primary metrics for quantifying and analyzing cSDH. However, these approaches do not quantify the local 3D brain deformation caused by cSDH. We propose a novel method using anatomy-aware unsupervised diffeomorphic pseudo-healthy synthesis to generate brain deformation fields. The deformation fields derived from this process are utilized to extract biomarkers that quantify the shift in the brain due to cSDH. We use CT scans of 121 patients for training and validation of our method and find that our metrics allow the identification of patients who require surgery. Our results indicate that automatically obtained brain deformation fields might contain prognostic value for personalized cSDH treatment. Our implementation is available on: github.com/Barisimre/brain-morphing
Paper Structure (13 sections, 4 equations, 5 figures, 1 table)

This paper contains 13 sections, 4 equations, 5 figures, 1 table.

Table of Contents

  1. Introduction
  2. Materials and Methods
  3. Data
  4. Brain Symmetry Losses
  5. Initial Alignment and Segmentation
  6. Diffeomorphic Pseudo-Healthy Synthesis
  7. cSDH Severity Prediction
  8. Experiments and Results
  9. Initial Alignment and Segmentation
  10. Pseudo-Healthy Synthesis
  11. Surgery Prediction
  12. Discussion and Conclusion
  13. Acknowledgments

Figures (5)

  • Figure 1: CT scans of cSDH patients selected for surgical intervention, showing the ideal midline (blue), shifted midline (pink), cSDH (red), left ventricle (purple), and right ventricle (green). Left: a unilateral hematoma with midline shift. Center: a bilateral hematoma with midline shift. Right: a bilateral hematoma without midline shift.
  • Figure 2: Overview of our process. Images are aligned to a sagittal plane (A) and automatically segmented (B). The CT scan and segmentations are used as input to a deep diffeomorphic registration model (C). Deformation fields provided by this model are used to distinguish between patients requiring surgery by various biomarkers.
  • Figure 3: Axial and coronal slide visualizing registration results and mid-sagittal plane (green). From left to right: the original brain with a unilateral (visible on the right side) cSDH, the magnitude of the deformation field, and the resulting pseudo-healthy brain. The hematoma is mostly removed and ventricular symmetry is restored.
  • Figure 4: Box plots for five biomarkers for three subsets of our data, comparing the distributions of patients that required surgery (orange) compared to the rest (blue). The midline shift (MLS) is hand measured in mm. Hematoma volume (Volume) is inferred from the segmentation results in mm. Maximum, average, and total shift are calculated as explained in subsection \ref{['brain-shift']} in millimeters. Many of the biomarkers fail at distinguishing surgical status of bilateral hematomas.
  • Figure 5: ROC curves of surgical status classifiers in either the full population (left) or subgroups of patients with bilateral (center) or unilateral (right) hematomas.