Classification and visualization based on derived image features: application to genetic syndromes

Data transformations prior to analysis may be beneficial in classification tasks. In this article we investigate a set of such transformations on 2D graph-data derived from facial images and their effect on classification accuracy in a high-dimensional setting. These transformations are low-variance...

Full description

Saved in:
Bibliographic Details
Published inPloS one Vol. 9; no. 11; p. e109033
Main Authors Balliu, Brunilda, Würtz, Rolf P, Horsthemke, Bernhard, Wieczorek, Dagmar, Böhringer, Stefan
Format Journal Article
LanguageEnglish
Published United States Public Library of Science 18.11.2014
Public Library of Science (PLoS)
Subjects
Accuracy
Algorithms
Asymmetry
Bioinformatics
Biology and Life Sciences
Biometric Identification - methods
Classification
Classifiers
Craniofacial Abnormalities - classification
Data processing
Datasets
Generalized linear models
Humans
Image classification
Image Processing, Computer-Assisted - methods
Medical statistics
Medicine and Health Sciences
Performance evaluation
Physical Sciences
Principal components analysis
Regression analysis
Transformations
Online AccessGet full text

Cover

More Information
Summary:Data transformations prior to analysis may be beneficial in classification tasks. In this article we investigate a set of such transformations on 2D graph-data derived from facial images and their effect on classification accuracy in a high-dimensional setting. These transformations are low-variance in the sense that each involves only a fixed small number of input features. We show that classification accuracy can be improved when penalized regression techniques are employed, as compared to a principal component analysis (PCA) pre-processing step. In our data example classification accuracy improves from 47% to 62% when switching from PCA to penalized regression. A second goal is to visualize the resulting classifiers. We develop importance plots highlighting the influence of coordinates in the original 2D space. Features used for classification are mapped to coordinates in the original images and combined into an importance measure for each pixel. These plots assist in assessing plausibility of classifiers, interpretation of classifiers, and determination of the relative importance of different features.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
Conceived and designed the experiments: DW BH SB. Performed the experiments: BB DW SB. Analyzed the data: BB RPW. Contributed reagents/materials/analysis tools: DW SB. Wrote the paper: BB RPW DW SB.
Competing Interests: The authors have declared that no competing interests exist.
ISSN:1932-6203
1932-6203
DOI:10.1371/journal.pone.0109033