Related Experiment Video
Updated: May 3, 2026
13:44
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
Published on: August 30, 2013
42.6K
Assessment of a novel mass detection algorithm in mammograms
Ehsan Kozegar1, Mohsen Soryani, Behrouz Minaei
1Department of Computer Engineering, Iran University of Science and Technology (IUST), Tehran, Iran.
Journal of Cancer Research and Therapeutics
|February 13, 2014
Summary
This study presents an efficient method for detecting breast masses in mammograms using adaptive thresholding and machine learning. The developed algorithm outperforms existing methods in mass detection accuracy.
Area of Science:
- Medical Imaging
- Computer-Aided Diagnosis
- Machine Learning in Healthcare
Background:
- Mammography is crucial for early breast abnormality detection.
- Visually detecting masses on mammograms presents significant challenges.
Purpose of the Study:
- To implement an efficient method for mass detection in mammograms.
- To improve the accuracy of identifying breast abnormalities.
Main Methods:
- A two-step approach: adaptive thresholding for suspicious region extraction.
- Machine learning for reducing false positives from the initial detection.
- Validation on mini-MIAS and INBreast databases.
Main Results:
- The proposed mass detection algorithm demonstrates superior performance.
- FROC analysis confirms the algorithm's effectiveness compared to competing methods.
Conclusions:
- Balancing sensitivity and false positive rates is critical for effective mass detection.
- Mass detection should be treated as a cost-sensitive problem due to unequal misclassification costs.
