Related Experiment Video
Updated: Jul 23, 2026
08:30
X-ray Dose Reduction through Adaptive Exposure in Fluoroscopic Imaging
Published on: September 11, 2011
Noise Reduction for a Virtual Grid Using a Generative Adversarial Network in Breast X-ray Images
Sewon Lim1, Hayun Nam2, Hyemin Shin2
1Department of Health Science, General Graduate School of Gachon University, 191, Hambakmoe-ro, Yeonsu-gu, Incheon 21936, Republic of Korea.
Journal of Imaging
|December 22, 2023
Summary
This study introduces a generative adversarial network (GAN) algorithm to reduce noise in breast X-ray images after scatter correction. The novel approach enhances image quality by effectively mitigating noise amplification, improving diagnostic accuracy.
Area of Science:
- Medical Imaging
- Artificial Intelligence in Healthcare
- Radiology
Background:
- Scatter correction in breast X-ray imaging can amplify noise, potentially degrading image quality.
- Virtual grid techniques, while useful, may exacerbate this noise issue.
- Effective noise reduction is crucial for accurate mammography interpretation.
Purpose of the Study:
- To develop and evaluate a noise reduction algorithm for breast X-ray images.
- To address noise amplification issues post-scatter correction.
- To improve the quality of digital breast tomosynthesis images using artificial intelligence.
Main Methods:
- A noise-level estimation algorithm was developed.
- A noise reduction algorithm based on generative adversarial networks (GANs) was created.
- Synthetic scatter breast X-ray images were generated and processed using scatter correction software.
- GANs were trained on 42 noise combinations.
- Image quality was quantitatively assessed using contrast-to-noise ratio (CNR), coefficient of variance (COV), and normalized noise-power spectrum (NNPS).
Main Results:
- The GAN-based noise reduction algorithm effectively reduced noise in breast X-ray images.
- An approximate 2.80% improvement in CNR was observed.
- A 12.50% enhancement in COV was achieved.
- Overall improvement in NNPS was noted across all frequency ranges.
Conclusions:
- The proposed GAN-based noise reduction algorithm successfully mitigates noise amplification after scatter correction.
- The application of this algorithm leads to improved-quality breast X-ray images.
- This AI-driven approach shows promise for enhancing mammography diagnostics.
