DocumentCode
1172078
Title
Learning of Perceptual Similarity From Expert Readers for Mammogram Retrieval
Author
Wei, Liyang ; Yang, Yongyi ; Wernick, Miles N. ; Nishikawa, Robert M.
Author_Institution
Dept. of Biomed. Eng., Illinois Inst. of Technol., Chicago, IL
Volume
3
Issue
1
fYear
2009
Firstpage
53
Lastpage
61
Abstract
Content-based image retrieval relies critically on the use of a computerized measure of the similarity (i.e., relevance) of a query image to other images in a database. In this work, we explore a superivised learning approach for retrieval of mammogram images, of which the goal is to serve as a diagnostic aid for breast cancer. We propose that the most meaningful measure is one that is designed specifically to match that perceived by the radiologists in their interpretation of mammogram lesions. In our approach, we model the notion of similarity as an unknown function of the image features characterizing the lesions, and use modern machine-learning algorithms to learn this function from similarity scores collected from radiologists in reader studies. This approach is evaluated using data collected from an observer study with a set of clinical mammograms. Our results demonstrate that the proposed machine learning approach can be used to model the notion of similarity as judged by expert readers in their interpretation of mammogram images and that it can outperform alternative similarity measures derived from unsupervised learning.
Keywords
biology computing; cancer; image retrieval; mammography; medical image processing; patient diagnosis; breast cancer; content-based image retrieval; diagnostic aid; learning approach; machine-learning algorithms; mammogram lesions; mammogram retrieval; perceptual similarity; radiologist; Biomedical imaging; Breast cancer; Image databases; Image retrieval; Information retrieval; Lesions; Machine learning; Mammography; Medical diagnostic imaging; Pathology; Content-based image retrieval; mammogram; multidimensional scaling; perceptual similarity; similarity measure; supervised learning;
fLanguage
English
Journal_Title
Selected Topics in Signal Processing, IEEE Journal of
Publisher
ieee
ISSN
1932-4553
Type
jour
DOI
10.1109/JSTSP.2008.2011159
Filename
4786551
Link To Document