Development of an expert system for automatic osteoarthritis diagnosis using numerical characterisations of articular cartilages and wear particles

Tian, Y., Peng, Z., and Liu, X. (2013) Development of an expert system for automatic osteoarthritis diagnosis using numerical characterisations of articular cartilages and wear particles. In: Proceedings of the IEEE/ASME International Conference on Advanced Intelligent Mechatronics. pp. 194-198. From: 2013 IEEE/ASME International Conference on Advanced Intelligent Mechatronics, 9-12 July 2013 , Wollongong, NSW, Australia.

[img] PDF (Published Version) - Published Version
Restricted to Repository staff only

View at Publisher Website: http://dx.doi.org/10.1109/AIM.2013.65840...
 
1
1


Abstract

As a common joint disease often caused by wear and tear and particularly common for aged people, osteoarthritis (OA) occurs with articular cartilage deterioration and wear particle generation. Current clinical OA diagnosis approaches are mainly based on qualitative evaluation of orthopaedists. This not only brings heavy cost to community healthcare, but can also limit the required service to OA patients in regional areas. In this paper, based on our previous work on the numerical analysis of cartilage and wear particles, an expert system has been established for automatic OA diagnosis using both cartilage and wear particle analysis methods. The developed system supported vector machine (SVM) to obtain cartilage and wear particle data and applied a statistical classification method for an OA assessment. This was a first time that wear particle analysis technique was integrated into an OA diagnosis system. Internal evaluations showed that the correct OA degree recognition rates were 80% and 72% based on the cartilage and particle analysis results, respectively. This paper presents the background information, how the system was developed, and the approach used to deal with inconsistent results from cartilage and wear debris analysis. The proposed framework has demonstrated that it is feasible to develop an automatic and objective OA diagnosis system for future clinic applications.

Item ID: 31722
Item Type: Conference Item (Research - E1)
ISBN: 978-1-4673-5319-9
ISSN: 2159-6247
Keywords: bone, geriatrics, medical expert systems, patient diagnosis, pattern classification, statistical analysis, support vector machines
Funders: Australian Research Council (ARC), James Cook University
Projects and Grants: ARC (DP1093975)
Date Deposited: 26 Feb 2014 09:47
FoR Codes: 02 PHYSICAL SCIENCES > 0299 Other Physical Sciences > 029901 Biological Physics @ 50%
02 PHYSICAL SCIENCES > 0299 Other Physical Sciences > 029903 Medical Physics @ 50%
SEO Codes: 92 HEALTH > 9201 Clinical Health (Organs, Diseases and Abnormal Conditions) > 920116 Skeletal System and Disorders (incl. Arthritis) @ 30%
97 EXPANDING KNOWLEDGE > 970102 Expanding Knowledge in the Physical Sciences @ 35%
97 EXPANDING KNOWLEDGE > 970111 Expanding Knowledge in the Medical and Health Sciences @ 35%
Downloads: Total: 1
More Statistics

Actions (Repository Staff Only)

Item Control Page Item Control Page