Automatic ROI detection and classification of the Achilles tendon ultrasound images

Benrabha, JS and Meziane, F 2017, Automatic ROI detection and classification of the Achilles tendon ultrasound images , in: International Conference on Internet of Things and Machine Learning (IML2017), 17-18 October 2017, Liverpool, United Kingdom. (In Press)

[img] PDF - Accepted Version
Restricted to Repository staff only

Download (622kB) | Request a copy

Abstract

Ultrasound (US) imaging plays an important role in medical imaging technologies. It is widely used because of its ease of use and low cost compared to other imaging techniques. Specifically, ultrasound imaging is used in the detection of the Achilles Tendon (AT) pathologies as it detects important details. For example, US imaging is used for AT rupture that affects about 1 in 5,000 people worldwide. Decision support systems are important in medical imaging, as they assist radiologist in detecting probable diagnoses and lesions. The work presented in this paper concerns the development of a software application to detect changes in the AT ultrasound images and subsequently classify them into normal or abnormal. We propose an approach that fully automates the detection for the Region of Interest (ROI) in ultrasound AT images. The original image is divided into six blocks with 1 cm size in each direction. The blocks lie inside the vulnerable area considered as our ROI. The proposed system achieved an accuracy of 97.21%.

Item Type: Conference or Workshop Item (Paper)
Schools: Schools > School of Computing, Science and Engineering > Salford Innovation Research Centre (SIRC)
Journal or Publication Title: ACM – International Conference Proceedings Series (ICPS)
Publisher: ACM Digital Library
Related URLs:
Depositing User: Prof Farid Meziane
Date Deposited: 13 Sep 2017 09:55
Last Modified: 13 Sep 2017 13:35
URI: http://usir.salford.ac.uk/id/eprint/43745

Actions (login required)

Edit record (repository staff only) Edit record (repository staff only)

Downloads

Downloads per month over past year