Regional depth-specific subchondral bone density measures in osteoarthritic and normal distal femora: in vivo precision and preliminary comparisons
Date
2023-03-07
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
ORCID
0009-0008-5662-0766
Type
Thesis
Degree Level
Masters
Abstract
Knee osteoarthritis (OA) is a degenerative disease which affects the cartilage and underlying
bone of afflicted joints. OA can be diagnosed by clinical symptoms such as joint pain, stiffness
and swelling or radiographically according to the presence of key structural changes associated
with the disease. The investigation of subchondral bone is a growing topic of interest in the study
of OA, though its exact relationship with disease initiation and progression is currently not well
understood. Previous studies have investigated subchondral bone’s role in OA at the proximal
tibia and patella; however there is a paucity of research at the distal femur. Additionally, many of
the existing strategies to study subchondral bone in OA have relied on time-consuming manual
techniques, limiting their applicability in studies with large sample sizes. This highlights a need
for the development of automated techniques which assess subchondral bone at the distal femur.
The primary goal of this research was to develop an automated workflow capable of
precisely assessing regional density measures at the distal femur to differentiate between OA and
normal subjects. This resulted in the following objectives: 1) Develop an automated workflow
for assessing regional density measures at the distal femur; 2) determine precision errors of the
acquired density measures; and 3) compare distal femoral density measures between individuals
with and without knee OA.
Following the development of an automated workflow, in vivo precision (assessed via
root mean square coefficients of variation, CV%RMS) at the distal femur was assessed and ranged
from 1.6% to 3.6%. These results are consistent with the in vivo precision errors found at the
proximal tibia and patella, indicating they are suitable for investigating differences between OA
and normal distal femora.
For the comparison analysis, mean regional density measures were not found to differ
significantly between normal and OA distal femora; however, peak density clusters at depths of
2.5-5mm were found to be ~16% lower in OA distal femora.
Thesis results indicate that suitable precision can be obtained at the distal femur with
automated methods. Results also suggest that subchondral density patterns may differ in OA and
normal distal femora in complex ways, not necessarily detectable via regional analyses alone.
Description
Keywords
osteoarthritis, imaging
Citation
Degree
Master of Science (M.Sc.)
Department
Mechanical Engineering
Program
Mechanical Engineering