Related Experiment Video
Updated: May 16, 2026
13:44
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
Published on: August 30, 2013
A modified undecimated discrete wavelet transform based approach to mammographic image denoising
Eri Matsuyama1, Du-Yih Tsai, Yongbum Lee
1Department of Radiological Technology, Graduate School of Health Sciences, Niigata University, 2-746, Asahimachi-dori, Niigata 951-8518, Japan.
Journal of Digital Imaging
|December 5, 2012
Summary
This study introduces an advanced mammographic image denoising technique using hierarchical wavelet transforms. The novel method significantly reduces noise, improves image quality, and decreases processing time compared to conventional approaches.
Area of Science:
- Medical Imaging
- Signal Processing
- Biomedical Engineering
Background:
- Mammography is crucial for early breast cancer detection.
- Image noise can obscure subtle abnormalities, impacting diagnostic accuracy.
- Existing denoising methods may compromise image quality or efficiency.
Purpose of the Study:
- To develop and validate an effective denoising method for mammographic images.
- To enhance image quality and reduce noise using a novel wavelet transform approach.
- To compare the proposed method's performance against conventional techniques.
Main Methods:
- Utilized hierarchical correlation of discrete stationary wavelet transform coefficients.
- Employed iterative, undecimated, multi-directional wavelet transforms at adjacent scales.
- Applied mutual information for optimal wavelet basis function selection.
Main Results:
- Achieved a significant reduction in computation time (approx. 1/10th of conventional methods).
- Demonstrated statistically significant improvements in image quality via visual assessment.
- Validated the method through computer simulations and application to clinical mammograms.
Conclusions:
- The proposed denoising method is superior and effective for mammographic images.
- Offers substantial improvements in both processing efficiency and diagnostic image quality.
- Represents a valuable advancement in medical image processing for breast cancer screening.
