Cross-Examination for Angle-Closure Glaucoma Feature Detection

Effective feature selection plays a vital role in anterior segment imaging for determining the mechanism involved in angle-closure glaucoma (ACG) diagnosis. This research focuses on the use of redundant features for complex disease diagnosis such as ACG using anterior segment optical coherence tomog...

Full description

Saved in:
Bibliographic Details
Published in:IEEE journal of biomedical and health informatics 2016-01, Vol.20 (1), p.343-354
Main Authors: Issac Niwas, Swamidoss, Weisi Lin, Chee Keong Kwoh, Kuo, C.-C Jay, Sng, Chelvin C., Aquino, Maria Cecilia, Chew, Paul T. K.
Format: Article
Language:English
Subjects:
Accuracy
Algorithms
Angle-closure glaucoma
anterior segment optical coherence tomography
Blocking
Classification
Classification algorithms
Diagnosis
Feature extraction
Glaucoma
Glaucoma, Angle-Closure - diagnosis
Humans
Image Interpretation, Computer-Assisted - methods
Iris
laplacian score
Lenses
Machine Learning
Machine learning algorithms
machine learning classifier
minimum redundancy maximum relevance
Mutual information
Optical Coherence Tomography
Predictive Value of Tests
Redundant
redundant features
Segments
Tomography, Optical Coherence - methods
unsupervised feature selection
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:Effective feature selection plays a vital role in anterior segment imaging for determining the mechanism involved in angle-closure glaucoma (ACG) diagnosis. This research focuses on the use of redundant features for complex disease diagnosis such as ACG using anterior segment optical coherence tomography images. Both supervised [minimum redundancy maximum relevance (MRMR)] and unsupervised [Laplacian score (L-score)] feature selection algorithms have been cross-examined with different ACG mechanisms. An AdaBoost machine learning classifier is then used for classifying the five various classes of ACG mechanism such as iris roll, lens, pupil block, plateau iris, and no mechanism using both feature selection methods. The overall accuracy has shown that the usefulness of redundant features by L-score method in improved ACG diagnosis compared to minimum redundant features by MRMR method.
ISSN:2168-2194
2168-2208
DOI:10.1109/JBHI.2014.2387207