File(s) under permanent embargo

Reason: The file associated with this record is under a permanent embargo in accordance with the publisher's policy. The full text may be available through the publisher links provided above.

Medical image classification: A comparison of various handcrafted features

journal contribution
posted on 03.04.2020, 10:56 by AD Olayemi, MR Zare, P Muhammad Fermi
This paper compares different feature extraction techniques exploited by various researchers for medical image classification and retrieval. They are categorized into three groups; (i) feature extraction techniques used for shape, (ii) feature extraction techniques used for texture and (iii) local patch based representation such as bag of visual words. The main aim of this work is to determine the capabilities and the challenges of medical images handcrafted feature extraction techniques as well as to see how best to improve the efficiency and accuracy of medical image classification and retrieval. It focused centrally on the analysis of the most commonly used shape and texture feature extraction techniques applied on medical images. Bag of visual words which is a type of local patch based method was also analysed. The limitations of these techniques are discussed as presented in the paper reviewed. We summarized with some conclusions and a recommendation for future exploits.

History

Citation

Int. J. Advance Soft Compu. Appl, Vol. 11, No. 3, November 2019

Version

AM (Accepted Manuscript)

Published in

International Journal of Advances in Soft Computing and its Applications

Volume

11

Issue

3

Pagination

24 - 39

issn

2074-8523

Available date

01/11/2019

Publisher version

http://home.ijasca.com/data/documents/2_p24-39_Medical Image Classification A Comparison of Various Handcrafted Features.pdf

Language

en