Abstract
The pathogenesis of osteoarthritis (OA) includes complex events in the whole joint. In this project, we combined machine-learning techniques in a texture analysis framework and evaluated it in a longitudinal study, where magnetic resonance images of knees were used to quantify the tibial trabecular bone in both a marker for OA diagnosis and another marker for prediction of tibial cartilage loss.
By multiple-instance learning, we also investigated which region of the tibia provided the best prognosis for cartilage loss. The inferior part of the tibial bone was classified as the most relevant region and a preliminary radiological reading of the knees with high and low risks of cartilage loss suggested the prognosis marker captured aspects of the tibia vertical trabecularization to define the prognosis.
Besides presenting a bone marker able to predict disease progression and diagnostic marker superior to other OA biomarkers, our findings underlined the importance of the trabecular bone to the understanding of the OA pathology.
By multiple-instance learning, we also investigated which region of the tibia provided the best prognosis for cartilage loss. The inferior part of the tibial bone was classified as the most relevant region and a preliminary radiological reading of the knees with high and low risks of cartilage loss suggested the prognosis marker captured aspects of the tibia vertical trabecularization to define the prognosis.
Besides presenting a bone marker able to predict disease progression and diagnostic marker superior to other OA biomarkers, our findings underlined the importance of the trabecular bone to the understanding of the OA pathology.
| Original language | English |
|---|
| Publisher | Department of Computer Science, Faculty of Science, University of Copenhagen |
|---|---|
| Number of pages | 112 |
| Publication status | Published - 2012 |