Improving the annotation accuracy of medical images in ImageCLEFmed2005 using K-Nearest Neighbor (kNN) classifier

M. M. Abdulrazzaq, Shahrul Azman Mohd Noah

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Content-based image retrieval systems offer solution to store and search the ever increasing amount of digital images currently in existence. These systems retrieve and extract the images based on low level features, such as color, texture and shape. However, these visual features did not enable users to request images based on semantic meanings. Semantic retrieval is of highly importance in various domains and in particular the medical domain that contain images from various medical devices such as MRI and X-ray. Image annotation or classification systems can be considered as a solution for the limitations of existing CBIR systems. In this study, it was proposed a new approach for image classification using multi-level features and machine learning techniques, particularly the K-Nearest Neighbor (kNN) classifier. We experimented the proposed approach on 9000 images available from the ImageCLEFmed2005 dataset. Principle Component Analysis (PCA) was performed to reduce the feature vectors. The accuracy results achieved 89.32% and 92.99% for the respective 80 and 90% of training images. The results show improvement as compared to previous studies for the same dataset.

Original languageEnglish
Pages (from-to)16-26
Number of pages11
JournalAsian Journal of Applied Sciences
Volume8
Issue number1
DOIs
Publication statusPublished - 2015

Fingerprint

Classifiers
Semantics
Image classification
Image retrieval
Magnetic resonance imaging
Learning systems
Textures
Color
X rays

Keywords

  • Feature extraction
  • Image retrieval for medical application
  • ImageCLEFmed2005
  • K-nearest neighbor
  • Machine learning
  • Principal component analysis

ASJC Scopus subject areas

  • General

Cite this

@article{198cf2a7933e42cbbdf272505a91e6ae,
title = "Improving the annotation accuracy of medical images in ImageCLEFmed2005 using K-Nearest Neighbor (kNN) classifier",
abstract = "Content-based image retrieval systems offer solution to store and search the ever increasing amount of digital images currently in existence. These systems retrieve and extract the images based on low level features, such as color, texture and shape. However, these visual features did not enable users to request images based on semantic meanings. Semantic retrieval is of highly importance in various domains and in particular the medical domain that contain images from various medical devices such as MRI and X-ray. Image annotation or classification systems can be considered as a solution for the limitations of existing CBIR systems. In this study, it was proposed a new approach for image classification using multi-level features and machine learning techniques, particularly the K-Nearest Neighbor (kNN) classifier. We experimented the proposed approach on 9000 images available from the ImageCLEFmed2005 dataset. Principle Component Analysis (PCA) was performed to reduce the feature vectors. The accuracy results achieved 89.32{\%} and 92.99{\%} for the respective 80 and 90{\%} of training images. The results show improvement as compared to previous studies for the same dataset.",
keywords = "Feature extraction, Image retrieval for medical application, ImageCLEFmed2005, K-nearest neighbor, Machine learning, Principal component analysis",
author = "Abdulrazzaq, {M. M.} and {Mohd Noah}, {Shahrul Azman}",
year = "2015",
doi = "10.3923/ajaps.2015.16.26",
language = "English",
volume = "8",
pages = "16--26",
journal = "Asian Journal of Applied Sciences",
issn = "1996-3343",
publisher = "Science Alert",
number = "1",

}

TY - JOUR

T1 - Improving the annotation accuracy of medical images in ImageCLEFmed2005 using K-Nearest Neighbor (kNN) classifier

AU - Abdulrazzaq, M. M.

AU - Mohd Noah, Shahrul Azman

PY - 2015

Y1 - 2015

N2 - Content-based image retrieval systems offer solution to store and search the ever increasing amount of digital images currently in existence. These systems retrieve and extract the images based on low level features, such as color, texture and shape. However, these visual features did not enable users to request images based on semantic meanings. Semantic retrieval is of highly importance in various domains and in particular the medical domain that contain images from various medical devices such as MRI and X-ray. Image annotation or classification systems can be considered as a solution for the limitations of existing CBIR systems. In this study, it was proposed a new approach for image classification using multi-level features and machine learning techniques, particularly the K-Nearest Neighbor (kNN) classifier. We experimented the proposed approach on 9000 images available from the ImageCLEFmed2005 dataset. Principle Component Analysis (PCA) was performed to reduce the feature vectors. The accuracy results achieved 89.32% and 92.99% for the respective 80 and 90% of training images. The results show improvement as compared to previous studies for the same dataset.

AB - Content-based image retrieval systems offer solution to store and search the ever increasing amount of digital images currently in existence. These systems retrieve and extract the images based on low level features, such as color, texture and shape. However, these visual features did not enable users to request images based on semantic meanings. Semantic retrieval is of highly importance in various domains and in particular the medical domain that contain images from various medical devices such as MRI and X-ray. Image annotation or classification systems can be considered as a solution for the limitations of existing CBIR systems. In this study, it was proposed a new approach for image classification using multi-level features and machine learning techniques, particularly the K-Nearest Neighbor (kNN) classifier. We experimented the proposed approach on 9000 images available from the ImageCLEFmed2005 dataset. Principle Component Analysis (PCA) was performed to reduce the feature vectors. The accuracy results achieved 89.32% and 92.99% for the respective 80 and 90% of training images. The results show improvement as compared to previous studies for the same dataset.

KW - Feature extraction

KW - Image retrieval for medical application

KW - ImageCLEFmed2005

KW - K-nearest neighbor

KW - Machine learning

KW - Principal component analysis

UR - http://www.scopus.com/inward/record.url?scp=84915746540&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84915746540&partnerID=8YFLogxK

U2 - 10.3923/ajaps.2015.16.26

DO - 10.3923/ajaps.2015.16.26

M3 - Article

VL - 8

SP - 16

EP - 26

JO - Asian Journal of Applied Sciences

JF - Asian Journal of Applied Sciences

SN - 1996-3343

IS - 1

ER -