Catherine Beck, Ashleigh V. Morrice-West, Peter Muir, Peta L. Hitchens, R. Christopher Whitton
Background:
Bone mineral density (BMD) estimation using computed tomography (CT) imaging is well-established in human medicine through Hounsfield Unit (HU) analysis. This method has potential for veterinary applications, including horses. However, variability in HU values across different CT machines and over time poses challenges. The study aims to evaluate these variations using a standardized phantom to inform opportunistic BMD estimation in horses.
Methods:
A serial measurement study was conducted using 30 scans each on two CT machines—a conventional Siemens Emotion 16-slice CT and a standing Equina by Asto CT. An electron density phantom with eight insert plugs of varying density was scanned. ROI mean HU and standard deviation (noise) were measured. A linear mixed effects model assessed inter-machine differences and temporal variations. Calibration curves for estimating CaHA concentrations (proxy for BMD) were developed for each machine using linear regression.
Results:
HU values were significantly different between machines (P < 0.001), with higher HU recorded on the standing CT for all positive-value ROIs. Noise was also significantly lower on the standing CT. Differences increased with phantom insert density. Over time, HU remained stable for the conventional CT except for one ROI, while the standing CT showed significant HU increases in denser ROIs, attributed to a software upgrade and hardware change. Calibration equations were derived:
Conventional CT: CaHA = -29.58 + 0.98 × ROImean
Standing CT: CaHA = -54.53 + 0.95 × ROImean
Limitations:
The study used a non-ACR certified phantom, and scanning protocols varied between machines. One machine underwent a mid-study software upgrade. These factors may limit the generalizability of findings. Cone-beam CT systems were not evaluated, and results may not extrapolate to these systems.
Conclusions:
HU values differ significantly between CT machines, affecting BMD estimation in horses. Calibration using a bone-equivalent phantom is essential for inter-machine comparison and clinical application. Future work should focus on correlating CT-derived BMD with other validation methods such as DXA and ash fraction.
Calibration curve representing the linear regression line of best fit of mean HU obtained from the ROIs of the four bone-equivalent phantom inserts 5 to 8 against the known CaHA mg/cm.3 concentration of each insert for the conventional (orange) and standing CT (green). The calibration curve equations are Conventional Machine:
Standing Machine:
. Jitter was applied to the data points to prevent overlap (degree 4; N = 60)
How did we do?
Disclaimer: The summary generated in this email was created by an AI large language model. Therefore errors may occur. Reading the article is the best way to understand the scholarly work. The figure presented here remains the property of the publisher or author and subject to the applicable copyright agreement. It is reproduced here as an educational work. If you have any questions or concerns about the work presented here, reply to this email.
