Medical image classification based on multi-scale non-negative sparse coding

•We propose a multi-scale non-negative sparse coding model to construct visual dictionary thus to overcome the defects of BoVW-based algorithms.•We utilize multi-scale decomposition method to decompose images into multiple scale layers and extract more representative image features.•We introduce fis...

Full description

Saved in:
Bibliographic Details
Published in:Artificial intelligence in medicine Vol. 83; pp. 44 - 51
Main Authors: Zhang, Ruijie, Shen, Jian, Wei, Fushan, Li, Xiong, Sangaiah, Arun Kumar
Format: Journal Article
Language:English
Published: Netherlands Elsevier B.V 01.11.2017
Subjects:
Artificial Intelligence
Diagnostic Imaging - classification
Fisher discriminative analysis
Humans
Image Interpretation, Computer-Assisted - methods
Internal Medicine
Medical image classification
Multi-scale decomposition
Non-negative sparse coding
Other
Pattern Recognition, Automated
Predictive Value of Tests
ROC Curve
The semantic gap
The semantic gap
Multi-scale decomposition
Non-negative sparse coding
Fisher discriminative analysis
Medical image classification
ISSN:0933-3657, 1873-2860, 1873-2860
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•We propose a multi-scale non-negative sparse coding model to construct visual dictionary thus to overcome the defects of BoVW-based algorithms.•We utilize multi-scale decomposition method to decompose images into multiple scale layers and extract more representative image features.•We introduce fisher discriminative analysis algorithm to non-negative sparse coding model thus to exploit more contextual spatial information.•Our model performs superior to other related algorithms in terms of efficiency and classification accuracy. With the rapid development of modern medical imaging technology, medical image classification has become more and more important in medical diagnosis and clinical practice. Conventional medical image classification algorithms usually neglect the semantic gap problem between low-level features and high-level image semantic, which will largely degrade the classification performance. To solve this problem, we propose a multi-scale non-negative sparse coding based medical image classification algorithm. Firstly, Medical images are decomposed into multiple scale layers, thus diverse visual details can be extracted from different scale layers. Secondly, for each scale layer, the non-negative sparse coding model with fisher discriminative analysis is constructed to obtain the discriminative sparse representation of medical images. Then, the obtained multi-scale non-negative sparse coding features are combined to form a multi-scale feature histogram as the final representation for a medical image. Finally, SVM classifier is combined to conduct medical image classification. The experimental results demonstrate that our proposed algorithm can effectively utilize multi-scale and contextual spatial information of medical images, reduce the semantic gap in a large degree and improve medical image classification performance.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0933-3657
1873-2860
1873-2860
DOI:10.1016/j.artmed.2017.05.006