Journal
COMPUTERIZED MEDICAL IMAGING AND GRAPHICS
Volume 26, Issue 5, Pages 303-307Publisher
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/S0895-6111(02)00027-7
Keywords
breast cancer; breast ultrasound; breast lesion discrimination; texture analysis; image analysis
Ask authors/readers for more resources
We investigate the use of Haralick's texture features and posterior acoustic attenuation descriptors (PAAD) for the characterization of ultrasound (US) breast lesions. 71 lesions (24 cyst, 21 benign solid mass and 26 malignant solid masses) were manually segmented on two-dimensional breast US images. 28 Haralick's descriptors and two PAAD were evaluated on these segmented lesions. Mean of Sum Average, Range of Sum Entropy and the second PAAD best discriminated cysts from noncysts. Range of Correlation and the second PAAD best discriminated solid malignant from benign lesions, Computerized analysis of breast US images can increase the specificity of breast sonography by providing a better characterization of solid lesions. (C) 2002 Elsevier Science Ltd. All rights reserved.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available