Title: Breast cancer diagnosis: a statistical analysis-based approach

Authors: Duo Zhou; Dinesh P. Mital; Shankar Srinivasan

Addresses: Pfizer Inc. Office: 219-7-71, 235 East 42nd Street, New York, NY 10017-5755, USA ' Department of Health Informatics, School of Health Related Professions, UMDNJ, 65 Bergen Street, Newark, NJ-07101, USA ' Department of Health Informatics, School of Health Related Professions, UMDNJ, 65 Bergen Street, Newark, NJ-07101, USA

Abstract: The capability of automatic recognition of data patterns has made machine learning very popular in modern science. Well, in clinical settings, the goal is not to find a function that most closely fits the data, but to find one that will most accurately predict outcome from future input. A new field, statistical learning, a framework for machine learning by combining statistics and functional analysis, has become more promising. In this paper, a field study on breast cancer diagnosis was performed to evaluate a variety of statistical learning techniques.

Keywords: breast cancer; cancer diagnosis; statistical learning; benign tumours; malignant tunours; clump; epithelial; scatter; machine learning.

DOI: 10.1504/IJMEI.2013.057194

International Journal of Medical Engineering and Informatics, 2013 Vol.5 No.4, pp.321 - 333

Published online: 16 Oct 2013 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article