The application of KAZE features to the classification echocardiogram videos

Li, Wei, Qian, Yu, Loomes, Martin J. and Gao, Xiaohong W. ORCID: (2015) The application of KAZE features to the classification echocardiogram videos. Multimodal Retrieval in the Medical Domain: First International Workshop, MRMD 2015, Vienna, Austria, March 29, 2015, Revised Selected Papers. In: First International Workshop Multimodal Retrieval in the Medical Domain (MRMD 2015), 29 Mar 2015, Vienna, Austria. ISBN 9783319244709. ISSN 0302-9743 [Conference or Workshop Item] (doi:10.1007/978-3-319-24471-6_6)

Download (879kB) | Preview


In the computer vision field, both approaches of SIFT and SURF are prevalent in the extraction of scale-invariant points and have demonstrated a number of advantages. However, when they are applied to medical images with relevant low contrast between target structures and surrounding regions, these approaches lack the ability to distinguish salient features. Therefore, this research proposes a different approach by extracting feature points using the emerging method of KAZE. As such, to categorise a collection of video images of echocardiograms, KAZE feature points, coupled with three popular representation methods, are addressed in this paper, which includes the bag of words (BOW), sparse coding, and Fisher vector (FV). In comparison with the SIFT features represented using Sparse coding approach that gives 72% overall performance on the classification of eight viewpoints, KAZE feature integrated with either BOW, sparse coding or FV improves the performance significantly with the accuracy being 81.09%, 78.85% and 80.8% respectively. When it comes to distinguish only three primary view locations, 97.44% accuracy can be achieved when employing the approach of KAZE whereas 90% accuracy is realised while applying SIFT features.

Item Type: Conference or Workshop Item (Paper)
Additional Information: Published as a chapter in: Multimodal Retrieval in the Medical Domain, Volume 9059 of the series Lecture Notes in Computer Science pp 61-72
Research Areas: A. > School of Science and Technology > Computer Science
Item ID: 15373
Notes on copyright: The final publication is available at Springer via
Useful Links:
Depositing User: Xiaohong Gao
Date Deposited: 27 Apr 2015 13:22
Last Modified: 09 Jun 2021 16:35

Actions (login required)

View Item View Item


Activity Overview

Additional statistics are available via IRStats2.