Reproducibility of Automatic Quantitative Analysis of the Articular Cartilages from MEDIC and weDESS Magnetic Resonance Images of the Knee at 1.5T

Fripp, J., Jurcak, V., Holt, R., Ourselin, S., Engstrom, C. and Crozier, S. (2008). Reproducibility of Automatic Quantitative Analysis of the Articular Cartilages from MEDIC and weDESS Magnetic Resonance Images of the Knee at 1.5T. In: Proceedings of the 16th Annual Meeting of the International Society for Magnetic Resonance in Medicine. The International Society for Magnetic Resonance in Medicine's 16th Scientific Meeting and Exhibition, Toronto, Ontario, Canada, (3657-3657). 3-9 May 2008.


Author Fripp, J.
Jurcak, V.
Holt, R.
Ourselin, S.
Engstrom, C.
Crozier, S.
Title of paper Reproducibility of Automatic Quantitative Analysis of the Articular Cartilages from MEDIC and weDESS Magnetic Resonance Images of the Knee at 1.5T
Conference name The International Society for Magnetic Resonance in Medicine's 16th Scientific Meeting and Exhibition
Conference location Toronto, Ontario, Canada
Conference dates 3-9 May 2008
Convener Butts Pauly, K
Proceedings title Proceedings of the 16th Annual Meeting of the International Society for Magnetic Resonance in Medicine   Check publisher's open access policy
Place of Publication USA
Publisher International Society for Magnetic Resonance in Medicine
Publication Year 2008
Sub-type Fully published paper
Open Access Status Not Open Access
ISSN 1524-6965
Start page 3657
End page 3657
Total pages 1
Language eng
Abstract/Summary Osteoarthritis (OA) of the knee is characterized by degeneration of cartilage tissue. Magnetic Resonance (MR) imaging allows the monitoring of cartilage degeneration by analysing changes in the volume, thickness and surface area of these complex soft tissue structures. To perform this analysis requires the cartilages in the MR images to be accurately segmented. Unfortunately, due to imaging artefacts and low contrast in several regions of the knee joint, like the joint contact areas, tendons and ligaments, fully automated segmentation of the cartilage from MR images is difficult. This has made the clinical use of MR imaging for OA assessment impractical, with clinical studies performed using manual or semi-automatic methods [1]. Usually, high resolution sequences like T1 weighted water excitation FLASH (weFLASH) at 1.5 T have been used to image the cartilages which then appear fairly homogenous and bright compared to other tissue. Recently, the use of water excited Double Echo in the Steady State (weDESS) at 3T has been investigated with results validated against 3T weFLASH [2]. The T2 and T2* weighting in weDESS obtains a rich variation in cartilage appearance that may be important clinically, however, it makes automated segmentation even more difficult. In our previous work [3,4] we developed a segmentation system for cartilages which was trained and tested using a database of fat suppressed SPGR images of healthy volunteers. In this paper we extended the approach to include a tissue estimate for synovial fluid and tested the reproducibility and testretest error for our automatic segmentation algorithm by segmenting each individual cartilage in the knee from weDESS (Te=8.6, Tr=25, freq=63) and MEDIC (Te=22, Tr = 44, freq = 63) images acquired at 1.5 T.
Subjects E1
861502 Medical Instruments
970109 Expanding Knowledge in Engineering
090399 Biomedical Engineering not elsewhere classified
Keyword Automatic Quantative Analysis
articular cartilage
MRI
Q-Index Code E1
Q-Index Status Confirmed Code

 
Versions
Version Filter Type
Citation counts: Google Scholar Search Google Scholar
Created: Mon, 13 Apr 2009, 02:29:55 EST by Ms Kimberley Nunes on behalf of School of Information Technol and Elec Engineering