Related Experiment Video
Updated: Jun 5, 2025
15:48
Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging
Published on: December 15, 2014
22.4K
Scatter correction for contrast-enhanced digital breast tomosynthesis with a dual-layer detector
Xiangyi Wu1, Xiaoyu Duan1, Hailiang Huang1
1Stony Brook Medicine, Department of Radiology, Stony Brook, New York, United States.
Journal of Medical Imaging (Bellingham, Wash.)
|December 6, 2024
Summary
A new scatter correction method improves dual-layer contrast-enhanced digital breast tomosynthesis (DL-CEDBT) imaging. This technique enhances image quality and lesion detection by accurately removing X-ray scatter, crucial for early cancer diagnosis.
Area of Science:
- Medical Imaging
- Radiology
- Biomedical Engineering
Background:
- Contrast-enhanced digital breast tomosynthesis (CEDBT) uses neo-angiogenesis to highlight breast tumors.
- Dual-layer (DL) detectors enable simultaneous high-energy (HE) and low-energy (LE) image acquisition, reducing scan time and motion artifacts.
- X-ray scatter significantly degrades image quality and lesion detectability in CEDBT.
Purpose of the Study:
- To develop a practical, accurate, and robust scatter correction (SC) method for DL-CEDBT.
- To address the unique scatter characteristics of DL-CEDBT systems.
Main Methods:
- A hybrid SC method combining iterative convolution and empirical interpolation was proposed.
- Monte Carlo simulations generated scatter point spread functions for various breast parameters.
- Performance was evaluated using digital breast phantoms and compared against existing SC methods, with Mean Absolute Relative Error (MARE) as the accuracy metric.
Main Results:
- The proposed hybrid SC method demonstrated superior accuracy and robustness across different phantom types and views (CC, MLO).
- Achieved MARE below for both LE and HE images.
- Post-SC, cupping artifacts were eliminated, and signal difference-to-noise ratio improved by 82.0% for 8 mm iodine objects.
Conclusions:
- A practical SC method for DL-CEDBT was successfully developed.
- The method provides accurate and robust scatter estimates, significantly improving image quality.
- Enhanced image quality is expected to improve lesion detectability in DL-CEDBT examinations.
