Multi-level of feature extraction and classification for X-Ray medical image

• Main Authors: Abdulrazaq, M, Alshaikhli, Imad Fakhri Taha, Mohd Noah, Shahrul Azman, Fadhil,, Moayad Al Athami
• Format: Article
• Language: English; English
• Published: Institute of Advanced Engineering and Science (IAES) 2018
• Subjects: QA75 Electronic computers. Computer science
• Online Access: http://irep.iium.edu.my/63248/13/63248%20Multi-Level%20of%20Feature%20Extraction%20and%20Classification%20for%20X-Ray%20%20SCOPUS.pdf; http://irep.iium.edu.my/63248/19/63248_Multi-Level%20of%20Feature%20Extraction%20and%20Classification%20for%20X-Ray_article.pdf; http://irep.iium.edu.my/63248/; http://www.iaescore.com/journals/index.php/IJEECS/article/view/11066/8196
• Institution: Universiti Islam Antarabangsa Malaysia

There has been a rise in demand for digitized medical images over the last two decades. Medical images' pivotal role in surgical planning is also an essential source of information for diseases and as medical reference as well as for the purpose of research and training. Therefore, effective techniques for medical image retrieval and classification are required to provide accurate search through substantial amount of images in a timely manner. Given the amount of images that are required to deal with, it is a non-viable practice to manually annotate these medical images. Additionally, retrieving and indexing them with image visual feature cannot capture high level of semantic concepts, which are necessary for accurate retrieval and effective classification of medical images. Therefore, an automatic mechanism is required to address these limitations. Addressing this, this study formulated an effective classification for X-ray medical images using different feature extractions and classification techniques. Specifically, this study proposed pertinent feature extraction algorithm for X-ray medical images and determined machine learning methods for automatic X-ray medical image classification. This study also evaluated different image features (chiefly global, local, and combined) and classifiers. Consequently, the obtained results from this study improved results obtained from previous related studies.