Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Realistic Collimated X-Ray Image Simulation Pipeline

Benjamin El-Zein, Dominik Eckert, Thomas Weber, Maximilian Rohleder, Ludwig Ritschl, Steffen Kappler, Andreas Maier

Abstract

Collimator detection remains a challenging task in X-ray systems with unreliable or non-available information about the detectors position relative to the source. This paper presents a physically motivated image processing pipeline for simulating the characteristics of collimator shadows in X-ray images. By generating randomized labels for collimator shapes and locations, incorporating scattered radiation simulation, and including Poisson noise, the pipeline enables the expansion of limited datasets for training deep neural networks. We validate the proposed pipeline by a qualitative and quantitative comparison against real collimator shadows. Furthermore, it is demonstrated that utilizing simulated data within our deep learning framework not only serves as a suitable substitute for actual collimators but also enhances the generalization performance when applied to real-world data.

A Realistic Collimated X-Ray Image Simulation Pipeline

Abstract

Collimator detection remains a challenging task in X-ray systems with unreliable or non-available information about the detectors position relative to the source. This paper presents a physically motivated image processing pipeline for simulating the characteristics of collimator shadows in X-ray images. By generating randomized labels for collimator shapes and locations, incorporating scattered radiation simulation, and including Poisson noise, the pipeline enables the expansion of limited datasets for training deep neural networks. We validate the proposed pipeline by a qualitative and quantitative comparison against real collimator shadows. Furthermore, it is demonstrated that utilizing simulated data within our deep learning framework not only serves as a suitable substitute for actual collimators but also enhances the generalization performance when applied to real-world data.

Paper Structure

This paper contains 16 sections, 8 equations, 6 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Methods
  3. Randomized Collimator Simulation Pipeline
  4. Binary Mask Sampling Strategy:
  5. Collimator Physics:
  6. Scattered Radation Simulation:
  7. Scatter Estimation:
  8. Scatter Correction:
  9. Poisson Noise Simulation:
  10. Experiments
  11. Real vs. Simulated:
  12. Application Case DNN:
  13. Results
  14. Framework Validation
  15. Network Evaluation
  16. ...and 1 more sections

Figures (6)

  • Figure 1: Illustrative case for collimator detection depicted in two contrast settings. (a) Contrast adjusted to full image. (b) Contrast adjusted to the orange box. The collimated area (c) is shown as a binary mask. In (d), the intensity profile along the dashed line is compared to the collimated area to visualize the complexity of image-based collimator detection.
  • Figure 2: Example of a rectangular binary mask being transformed afterwards by rotation and shape distortion to cover the range of essential deviations in clinical practice.
  • Figure 3: Physical properties to be considered when modeling collimators in a basic X-ray system. Edge-blurring introduced by the focal spot characteristic not being an ideal point, as well as increasing intensities within the collimated region due to photons that get scattered by the Compton effect.
  • Figure 4: Comparison of a real collimated image with the ouput of the pipeline based on an open field image acquired by the same setup. Both of the images are shown in two different contrast ranges. Besides being in the complete value range, intensities are limited to the indicated box regions.
  • Figure 5: Line plots comparing real collimator damping on a real X-ray image as well as presenting the differences between the real and simulated collimator.
  • ...and 1 more figures