Multiple-Instance Learning Algorithms for Computer-Aided Detection

Many computer-aided diagnosis (CAD) problems can be best modelled as a multiple-instance learning (MIL) problem with unbalanced data, i.e., the training data typically consists of a few positive bags, and a very large number of negative instances. Existing MIL algorithms are much too computationally...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on biomedical engineering Vol. 55; no. 3; pp. 1015 - 1021
Main Authors: Dundar, M. Murat, Fung, Glenn, Krishnapuram, Balaji, Rao, R. Bharat
Format: Journal Article
Language:English
Published: United States IEEE 01.03.2008
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:
Algorithms
Alternate optimization
Artificial Intelligence
Biomedical imaging
Colonic Neoplasms - diagnostic imaging
Computed tomography
convex hull
Design automation
Fisher discriminant
Humans
Image Enhancement - methods
Image Interpretation, Computer-Assisted - methods
Labeling
Magnetic resonance imaging
Medical diagnostic imaging
multiple-instance learning (MIL)
Pattern Recognition, Automated - methods
Pulmonary Embolism - diagnostic imaging
Radiography
Reproducibility of Results
Sensitivity and Specificity
Studies
Training data
X-ray detection
X-ray detectors
X-ray imaging
fisher discriminant
alternate optimization
convex hull
multiple instance learning
ISSN:0018-9294, 1558-2531
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Many computer-aided diagnosis (CAD) problems can be best modelled as a multiple-instance learning (MIL) problem with unbalanced data, i.e., the training data typically consists of a few positive bags, and a very large number of negative instances. Existing MIL algorithms are much too computationally expensive for these datasets. We describe CH, a framework for learning a convex hull representation of multiple instances that is significantly faster than existing MIL algorithms. Our CH framework applies to any standard hyperplane-based learning algorithm, and for some algorithms, is guaranteed to find the global optimal solution. Experimental studies on two different CAD applications further demonstrate that the proposed algorithm significantly improves diagnostic accuracy when compared to both MIL and traditional classifiers. Although not designed for standard MIL problems (which have both positive and negative bags and relatively balanced datasets), comparisons against other MIL methods on benchmark problems also indicate that the proposed method is competitive with the state-of-the-art.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ObjectType-Article-2
ObjectType-Feature-1
content type line 23
ObjectType-Undefined-1
ObjectType-Feature-3
ISSN:0018-9294
1558-2531
DOI:10.1109/TBME.2007.909544