VEMERA

About VEMERA

Rheumatoid arthritis and osteoarthritis are diseases affecting joints. Both cause inflammations in synovial joints, such as hands, wrist and feet, resulting in loss of cartilage and bone erosion.

Patients experience this as pain and stiffness in the affected joints.

Disease progression is currently measured using X-ray images of affected joints. Rheumatologists and radiologists evaluate using a semi quantitative scoring based on visual assessment of radiographs. Joint spaces are measured in order to gain insight in cartilage loss and erosions are (visually) detected.

It is preferable to use objective automatic assessment of disease progression. The methods described in literature describe partially automated measurement of the joint space.

Kauffman developed a semi-automatic joint segmentation and joint space width measurement. This method has been redesigned for foot radiographs.  The segmentation system used connected active appearance models of the bones. The system is able to detect a large variety of hands and feet on radiographs. Also, a method for erosion visualization was developed.

Goal

This research evaluates the visual scoring method to automated assessment of joint damage in RA. The aim is to develop automated methods to measure joint damage with a higher sensitivity to change compared to the currently used scoring method. Furthermore, the aim is to show the relevance of the system for research as well as clinical use.

Current activities

We are currently working on several items of the project:

- Improvements on the automatic segmentation of the foot
- Improvements on the JSW measurements of the foot
- What is the minimal size of an erosion to be detected?
- JSW measurements on series – comparing outcome to currently used clinical values

We are always looking for motivated students (BSc or MSC); just send an email to o.schenk@utwente.nl and we can discuss the possibilities for an assignment!

Click here for a movie about hand-segmentation.

Click here for a movie about foot-segmentation.

Project lead

no picture available
prof.dr.ir. Kees Slump