Related Experiment Video
Updated: Jul 13, 2026
05:28
Clinical Imaging of Microwave Mammography
Published on: November 14, 2025
Computing mammographic density from a multiple regression model constructed with image-acquisition parameters from a
Lee-Jane W Lu1, Thomas K Nishino, Tuenchit Khamapirad
1Department of Preventive Medicine and Community Health, The University of Texas Medical Branch, Galveston, TX 77555-1109, USA.
Physics in Medicine and Biology
|August 3, 2007
Summary
This study adapted a histogram segmentation method (HSM) for digital mammograms to measure breast density, a breast cancer risk factor. Mathematical models accurately predict breast density, potentially automating this process for future research.
Area of Science:
- Radiology
- Medical Imaging
- Oncology
Background:
- Breast density, the proportion of fibroglandular tissue, is a key indicator of breast cancer risk.
- Conventional measurement uses labor-intensive histogram segmentation methods (HSM) on film mammograms.
- Digital mammography offers potential for improved breast density assessment.
Purpose of the Study:
- To adapt and modify the histogram segmentation method (HSM) for measuring breast density from digital mammograms.
- To develop mathematical models for objective and automated breast density computation.
- To assess the predictive power of mammogram acquisition parameters and image statistics on breast density measurements.
Main Methods:
- Adaptation and modification of the histogram segmentation method (HSM) for full-field digital mammography.
- Multiple regression model analyses incorporating instrument parameters (e.g., compression thickness, radiation dose) and image pixel intensity statistics.
- Calculation of intra-class correlation coefficients for %-density reproducibility.
Main Results:
- Mammogram acquisition parameters and image pixel intensity statistics were strong predictors of threshold values (R(2) = 0.93) and %-density (R(2) = 0.84).
- High intra-class correlation coefficients (0.80 to 0.94) indicate good reproducibility of %-density measurements.
- Developed mathematical models demonstrate potential for objective and automated breast density calculation.
Conclusions:
- Adapted mathematical models can objectively and automatically compute breast density from digital mammograms, bypassing manual HSM.
- These models show high predictive accuracy and reproducibility, facilitating breast cancer risk assessment.
- Further research can refine these models to significantly aid breast cancer research studies.
