Enhancement of wavelet-based medical image retrieval through feature evaluation using an information gain measure

Authors:
O. H. Karam;A. M. Hamad;S. Ghoniemy;S. Rady
Affiliations:
Ain Shams University, Cairo, Egypt;Ain Shams University, Cairo, Egypt;Ain Shams University, Cairo, Egypt;Ain Shams University, Cairo, Egypt
Venue:
Proceedings of the 2003 ACM symposium on Applied computing
Year:
2003

Citing 8
Cited 2

Wavelet-based image indexing techniques with partial sketch retrieval capability

IEEE ADL '97 Proceedings of the IEEE international forum on Research and technology advances in digital libraries
WALRUS: a similarity retrieval algorithm for image databases

SIGMOD '99 Proceedings of the 1999 ACM SIGMOD international conference on Management of data
ASSERT: a physician-in-the-loop content-based retrieval system for HRCT image databases

Computer Vision and Image Understanding - Special issue on content-based access for image and video libraries
Query by Image and Video Content: The QBIC System

Computer
Windsurf: Region-Based Image Retrieval Using Wavelets

DEXA '99 Proceedings of the 10th International Workshop on Database & Expert Systems Applications
Local versus Global Features for Content-Based Image Retrieval

CBAIVL '98 Proceedings of the IEEE Workshop on Content - Based Access of Image and Video Libraries
NeTra: a toolbox for navigating large image databases

ICIP '97 Proceedings of the 1997 International Conference on Image Processing (ICIP '97) 3-Volume Set-Volume 1 - Volume 1
Fundamentals of Wavelets: Theory, Algorithms, and Applications

Fundamentals of Wavelets: Theory, Algorithms, and Applications

Use of wavelet-based basis functions to extract rotation invariant features for automatic image recognition

WSEAS Transactions on Information Science and Applications
Texture map: an effective representation for image segmentation

C3S2E '09 Proceedings of the 2nd Canadian Conference on Computer Science and Software Engineering

Quantified Score

Hi-index	0.00

Visualization

Abstract

Medical image retrieval covers a major application area, providing significant assistance in medical diagnosis. In this paper, several parameters for a wavelet-based retrieval system for HRCT lung images are analyzed, in order to improve the performance of the system in terms of precision and retrieval time. The information gain measure is used to evaluate extracted features to find the features that have the most discriminating power across image disease classes. A weighted similarity metric based on the evaluation is used for image retrieval. The number of levels for wavelet decomposition is studied for 2 types of wavelets, haar and daubechies-8. A filtering criteria based on the features with the highest information gain is used to enhance the retrieval time. Experiements on HRCT lung images that cover 8 disease classes show that the weighted approach achieves about 10% improvement in the average disease class precision, and 4.75% in the average total system precision, over the unweighted approach. Retrieval time is much enhanced while still maintaining high precision ratios.