Related Experiment Videos
Computer-aided detection system for breast masses on digital tomosynthesis mammograms: preliminary experience
Heang-Ping Chan1, Jun Wei, Berkman Sahiner
1Department of Radiology, University of Michigan, 1500 E Medical Center Dr, UHB1F510B, Ann Arbor, MI 48109-0030, USA. chanhp@umich.edu
Radiology
|October 21, 2005
Summary
This study developed a computer-aided detection (CAD) system for digital breast tomosynthesis (DBT) mammograms. The system shows promise for improving breast mass detection accuracy.
Area of Science:
- Medical Imaging
- Computer-Aided Diagnosis
- Radiology
Background:
- Digital Breast Tomosynthesis (DBT) offers improved visualization of breast tissue compared to conventional mammography.
- Accurate detection of breast masses remains a critical challenge in mammography interpretation.
Purpose of the Study:
- To design a computer-aided detection (CAD) system for identifying breast masses on DBT images.
- To conduct a preliminary performance evaluation of the developed CAD system.
Main Methods:
- A CAD system was developed using gradient-field analysis for screening 3D DBT volumes.
- Mass candidates were segmented, and image features were extracted for classification.
- A leave-one-case-out method was employed for training and testing the feature classifier.
Main Results:
- The CAD system achieved a mean area under the receiver operating characteristic curve of 0.91 ± 0.03.
- The system demonstrated a sensitivity of 85% with 2.2 false-positive objects per case.
Conclusions:
- The developed CAD system shows feasibility for breast mass detection in DBT mammography.
- This approach holds potential for enhancing diagnostic accuracy in breast cancer screening.