arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Computer Science · Computer Vision and Pattern Recognition

Breast cancer diagnosis using level-set statistics and support vector machines

Jianguo Liu, Xiaohui Yuan, Bill P. Buckles

· PubMed

See on the map

Abstract

Breast cancer diagnosis based on microscopic biopsy images and machine learning has demonstrated great promise in the past two decades. Various feature selection (or extraction) and classification algorithms have been attempted with success. However, some feature selection processes are complex and the number of features used can be quite large. We propose a new feature selection method based on level-set statistics. This procedure is simple and, when used with support vector machines (SVM), only a small number of features is needed to achieve satisfactory accuracy that is comparable to those using more sophisticated features. Therefore, the classification can be completed in much shorter time. We use multi-class support vector machines as the classification tool. Numerical results are reported to support the viability of this new procedure.

More like this