Abstract
Shape features are usually considered for discriminating benign and malignant breast tumors in ultrasound image. A contour extraction method is presented for the purpose of discrimination. A half contouring is devised for reducing the influence of shadow effect resulting from the ultrasonic energy decay. Based on the cue of contour variation, degree standard variation and texture information are developed as useful features for the discrimination. Above 70% accuracy could be obtained by using the adopted half contour information. Experiments confirm the feasibility of the proposed method.
Original language | English |
---|---|
Title of host publication | Proceedings of 2014 International Conference on Machine Learning and Cybernetics, ICMLC 2014 |
Publisher | IEEE Computer Society |
Pages | 533-537 |
Number of pages | 5 |
ISBN (Electronic) | 9781479942169 |
DOIs | |
State | Published - 13 01 2014 |
Event | 13th International Conference on Machine Learning and Cybernetics, ICMLC 2014 - Lanzhou, China Duration: 13 07 2014 → 16 07 2014 |
Publication series
Name | Proceedings - International Conference on Machine Learning and Cybernetics |
---|---|
Volume | 2 |
ISSN (Print) | 2160-133X |
ISSN (Electronic) | 2160-1348 |
Conference
Conference | 13th International Conference on Machine Learning and Cybernetics, ICMLC 2014 |
---|---|
Country/Territory | China |
City | Lanzhou |
Period | 13/07/14 → 16/07/14 |
Bibliographical note
Publisher Copyright:© 2014 IEEE.
Keywords
- Adaptive thresholding
- Breast tumor
- Contour extraction
- Disk expansion
- Ultrasound image