Improving the Mann–Whitney statistical test for feature selection: An approach in breast cancer diagnosis on mammography

Highlights • An innovative feature selection method (named uFilter) is proposed. • A set of image-based features, from mammography lesions, were explored and successfully ranked. • Classification's performance of four different machine learning algorithms increased in almost all scenarios when...

Full description

Saved in:
Bibliographic Details
Published in:Artificial intelligence in medicine 2015-01, Vol.63 (1), p.19-31
Main Authors: Pérez, Noel Pérez, Guevara López, Miguel A, Silva, Augusto, Ramos, Isabel
Format: Article
Language:English
Subjects:
Algorithms
Bayes Theorem
Breast
Breast cancer CADx
Breast Neoplasms - diagnostic imaging
Cancer
Chi-Square Distribution
Classification
Databases, Factual
Diagnosis
Diagnosis, Computer-Assisted - methods
Discretization
Discriminant Analysis
Feature selection methods
Female
Humans
Internal Medicine
Linear Models
Machine Learning
Machine learning algorithms
Mammography - methods
Mann–Whitney U-test
Models, Statistical
Neural networks
Other
Predictive Value of Tests
Radiographic Image Interpretation, Computer-Assisted - methods
Redundancy analysis
Reproducibility of Results
Statistical tests
uFilter method
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Highlights • An innovative feature selection method (named uFilter) is proposed. • A set of image-based features, from mammography lesions, were explored and successfully ranked. • Classification's performance of four different machine learning algorithms increased in almost all scenarios when using the uFilter method. • The uFilter method statistically improved the breast cancer classification in mammography. • The efficiency of the uFilter method was confirmed by the Wilcoxon statistical test.
ISSN:0933-3657
1873-2860
DOI:10.1016/j.artmed.2014.12.004