Related Experiment Video
Updated: Aug 8, 2026
13:44
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
Published on: August 30, 2013
Detection of microcalcifications in digital mammograms using wavelet filter and Markov random field model
Sung-Nien Yu1, Kuan-Yuei Li, Yu-Kun Huang
1Department of Electrical Engineering, National Chung Cheng University, Chia-Yi, Taiwan, ROC. yusn@ee.ccu.edu.tw
Summary
This study developed a computer-aided diagnosis system to detect clustered microcalcifications (MCs) in mammograms. The system achieved 92% sensitivity and removed 98.9% of false positives, aiding early breast cancer detection.
Area of Science:
- Medical Imaging
- Computer-Aided Diagnosis
- Breast Cancer Screening
Background:
- Clustered microcalcifications (MCs) are early indicators of breast cancer in mammograms.
- Accurate detection of MCs is crucial for timely diagnosis and treatment.
Purpose of the Study:
- To develop a computer-aided diagnosis (CAD) system for detecting clustered microcalcifications (MCs) in digital mammograms.
- To improve the accuracy of MCs detection by reducing false positives.
Main Methods:
- A two-stage approach was employed: initial detection of suspicious MCs using wavelet filtering and mean pixel value (MPV) thresholding.
- Texture features, including Markov random field parameters and auxiliary quantities, were extracted for MCs recognition.
- Bayes classifier and back-propagation neural network were utilized for classification.
Main Results:
- The system demonstrated high performance on 20 mammograms with 25 radiologist-marked MC clusters.
- Achieved 92% sensitivity (true positive rate).
- Successfully removed 98.9% of false positives (1341 out of 1356), with an average of 0.75 false positives per image.
Conclusions:
- The developed CAD system effectively detects clustered microcalcifications in digital mammograms.
- Texture features derived from Markov random fields, combined with auxiliary features, significantly enhance MCs recognition accuracy.
- The system shows promise for improving early breast cancer detection rates.
