Clinical magnetic resonance imaging of the equine foot: An investigation of factors influencing image quality and image interpretation

Byrne, Christian A. (2021) Clinical magnetic resonance imaging of the equine foot: An investigation of factors influencing image quality and image interpretation. MVM(R) thesis, University of Glasgow.

Abstract

Magnetic resonance imaging (MRI) is a fundamental imaging modality for evaluation of the equine foot. Optimising image quality and observer pathology identification is important to maximise the diagnostic value of MRI. There is limited information investigating factors that influence magnetic resonance (MR) image quality in live equine patients in a clinical setting. In addition, agreement between observers assessing pathology on clinical MRI studies of the equine foot has not been investigated. This project aimed to evaluate the influence of patient general anaesthesia (which encompasses the potential effects of motion and weight-bearing) and field strength on clinical MR image quality. In addition, the project aimed to determine the agreement between expert observers for pathology assessment of clinically important anatomical structures of the equine foot.

A total of fifteen routine equine MRI foot studies were acquired from the clinical databases of three different MRI systems: low-field standing, low-field under general anaesthesia and high-field under general anaesthesia. Ten experienced observers (diploma or associate level) assessed entire MRI studies and seven key individual anatomical structures of the equine foot. Observers used an online image assessment platform to grade subjective image quality (briefly, grade 1: textbook quality, grade 2: high diagnostic quality, grade 3: satisfactory diagnostic quality, grade 4: non‐diagnostic), pathology, and their confidence in pathology assessment. Statistical analysis was performed to assess the influence of anaesthesia and field strength on image quality, and to document inter-observer agreement in pathology assessment.

Observers deemed most clinical MRI foot studies to be of diagnostic quality, regardless of acquisition system. There were no significant differences in image quality between low-field standing and low-field under general anaesthesia (for both groups all individual structure image quality median grades= 3). Conversely, high-field under general anaesthesia studies had significantly greater image quality for entire studies and all individual anatomical structures (median grades= 1 for 5/7 structures and 2 for 2/7 structures) compared to low-field under general anaesthesia (all individual structure median grades= 3). There was a general trend of agreement between observers for pathology assessment of anatomical structures of the equine foot. Although absolute agreement for pathology assessment grading was generally low, relative agreement (accounting for the ranking of study pathology grading) was greater. Agreement was lowest for the distal interphalangeal joint (Kendall’s coefficient of concordance= 0.19) and greatest for the navicular bone (Kendall’s coefficient of concordance= 0.70). Importantly there were instances of marked variation in pathology assessment for individual MRI studies. In general, agreement was greater at the extremes of pathology.

The findings indicate that field strength is a more important influencer of image quality than general anaesthesia for MRI of the equine foot in clinical patients. However, the reasons described for reduced image quality appear to differ between MRI systems. There was a general tendency of agreement between observers for pathology assessment. However, there can be notable variation in pathology assessment for individual MRI studies, even when interpretation is performed by experienced observers. Future work is needed to evaluate the influence of image quality factors when imaging other regions of the equine limb and to investigate the processes of lesion identification and subsequent diagnostic decision making by those interpreting MRI images of the equine foot.

• Item Type: Thesis (MVM(R))
• Qualification Level: Masters
• Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Biodiversity, One Health & Veterinary Medicine
• Funder's Name: Horserace Betting Levy Board (HBLB)
• Supervisor's Name: Voute, Dr. Lance C. and Marshall, Dr. John
• Date of Award: 2021
• Depositing User: Theses Team
• Unique ID: glathesis:2021-82590
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 08 Dec 2021 14:51
• Last Modified: 08 Apr 2022 17:08
• Thesis DOI: 10.5525/gla.thesis.82590
• URI: https://theses.gla.ac.uk/id/eprint/82590

Related URLs

• Article DOI
• Enlighten Publications Record