Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Synthetic Data Generation for 3D Myocardium Deformation Analysis

Shahar Zuler, Dan Raviv

TL;DR

This work tackles the scarcity of high-resolution CT datasets with ground-truth 3D deformation annotations for myocardium analysis by proposing a synthetic data generation pipeline. It builds a LV-oriented deformation framework using a coordinate system and torsion-based warping to create paired systolic and diastolic frames with 3D optical-flow ground-truth. Leveraging a combination of 4D and 3D CT scans, manual LV segmentations, and carefully chosen deformation parameters, the approach yields a diverse synthetic dataset with comprehensive GT annotations. The method and accompanying code aim to enable the development of accurate myocardium deformation analysis algorithms for clinical imaging and diagnostics.

Abstract

Accurate analysis of 3D myocardium deformation using high-resolution computerized tomography (CT) datasets with ground truth (GT) annotations is crucial for advancing cardiovascular imaging research. However, the scarcity of such datasets poses a significant challenge for developing robust myocardium deformation analysis models. To address this, we propose a novel approach to synthetic data generation for enriching cardiovascular imaging datasets. We introduce a synthetic data generation method, enriched with crucial GT 3D optical flow annotations. We outline the data preparation from a cardiac four-dimensional (4D) CT scan, selection of parameters, and the subsequent creation of synthetic data from the same or other sources of 3D cardiac CT data for training. Our work contributes to overcoming the limitations imposed by the scarcity of high-resolution CT datasets with precise annotations, thereby facilitating the development of accurate and reliable myocardium deformation analysis algorithms for clinical applications and diagnostics. Our code is available at: http://www.github.com/shaharzuler/cardio_volume_skewer

Synthetic Data Generation for 3D Myocardium Deformation Analysis

TL;DR

This work tackles the scarcity of high-resolution CT datasets with ground-truth 3D deformation annotations for myocardium analysis by proposing a synthetic data generation pipeline. It builds a LV-oriented deformation framework using a coordinate system and torsion-based warping to create paired systolic and diastolic frames with 3D optical-flow ground-truth. Leveraging a combination of 4D and 3D CT scans, manual LV segmentations, and carefully chosen deformation parameters, the approach yields a diverse synthetic dataset with comprehensive GT annotations. The method and accompanying code aim to enable the development of accurate myocardium deformation analysis algorithms for clinical imaging and diagnostics.

Abstract

Accurate analysis of 3D myocardium deformation using high-resolution computerized tomography (CT) datasets with ground truth (GT) annotations is crucial for advancing cardiovascular imaging research. However, the scarcity of such datasets poses a significant challenge for developing robust myocardium deformation analysis models. To address this, we propose a novel approach to synthetic data generation for enriching cardiovascular imaging datasets. We introduce a synthetic data generation method, enriched with crucial GT 3D optical flow annotations. We outline the data preparation from a cardiac four-dimensional (4D) CT scan, selection of parameters, and the subsequent creation of synthetic data from the same or other sources of 3D cardiac CT data for training. Our work contributes to overcoming the limitations imposed by the scarcity of high-resolution CT datasets with precise annotations, thereby facilitating the development of accurate and reliable myocardium deformation analysis algorithms for clinical applications and diagnostics. Our code is available at: http://www.github.com/shaharzuler/cardio_volume_skewer
Paper Structure (12 sections, 3 figures)

This paper contains 12 sections, 3 figures.

Table of Contents

  1. Introduction
  2. Related Work
  3. Speckle Tracking
  4. Magnetic Resonance (MR) Tagging
  5. Coordinate System
  6. Method
  7. Data Preparation
  8. Initial Parameters
  9. Torsion selection and Warping
  10. Storing GT annotations
  11. Implementation Details
  12. Conclusion

Figures (3)

  • Figure 1: Example of Synthetic Sample: Visualization of a synthetic sample before (top) and after (bottom) deformation, along with arrows representing selected ground truth annotations (middle).
  • Figure 2: Visualization of coordinate system. The longitudinal direction is established as the principal component of the LV, intersecting the center of mass of the LV. Radial direction lines are perpendicular to the principal component, while circumferential direction lines are normal to both the intersecting radial line and the principal component itself.
  • Figure 3: A visualization of five different torsion centers evenly distributed for a constant torsion.