Automatic determination of bone erosions in Rheumatoid Arthritis using High Resolution peripheral Quantitative CT imaging

Michiel Peters

Rheumatoid Arthritis is a chronic inflammatory auto-immune disease, in which inflammation at the joints is characterized by destruction of peri-articular cortical bone, cartilage and soft tissues. On conventional radiography (CR), which is the gold standard in clinical practice, an erosion is defined as a cortical break and can be seen once the bone loss has formed a large enough cavity to be detectable in the projection. The presence, size and number within a joint and the number of joints with erosions are associated with poor functional outcome and predictors of radiographic progression. However, the quantification of erosions on CR, has its limitations since the sensitivity for the detection of erosions is low with CR.

High Resolution peripheral Quantitative Computed Tomography (HR-pQCT) is a low-dose imaging technique that is able to assess the 3D bone structure at micro-scale (82 µm isotropic resolution) for the peripheral skeleton in vivo. Studies with the HR-pQCT on finger joints have shown that HR-pQCT imaging enables the detection of smaller erosions, resulting in a higher sensitivity of detection of erosions compared to ultrasound, CR and MRI.

The aim in my research is to automatically identify these small erosions, which are specific for RA, using HR-pQCT imaging in patients with RA compared to Healthy Controls. We are developing a (semi-)automatic algorithm for the detection of these small erosions. The advantage is that no subjective and time consuming visual scoring system is needed.

Contact person: Michiel Peters

e-mail address: michiel.peters@maastrichtuniversity.nl