Automated RA platform
Rheumatoid arthritis (RA) is a chronic inflammatory disease affecting around 1% of the population. The cardinal symptom of the disease is affection by swelling and pain from especially the joints of the hands and feet. Fast disease control and monitoring essential as disease can cause damage, disability and decreased quality of life for the patient.
Ultrasound examination of the joints has been shown to be a sensitive and specific method for both detecting signs of early disease and evaluate arthritis activity in established disease. A central problem is that ultrasound scanning is operator dependent with a moderate to good inter observer reliability. This variability in scanning results can potentially affect the monitoring of disease progression. We want to develop a platform for automated ultrasound scanning followed by automated interpretation of disease activity by the system.
The RA platform is built as a table. The platform uses several sensors: A camera and a Leap Motion sensor track the position of the joints; A Force Torque sensor track the pressure against the ultrasound probe. In order to make good ultrasound images, the patient must relax their hand. To facilitate this, the placement of the hand on the table is up to the patient. The location of the joints is then used as a starting point for the robot to start the scan.
A problem with creating a fully automated system is the robot motions seem mechanical. By working with medical experts in ultrasound, a transfer of skills to the robot can happen through demonstrations. This smooth and natural motions in the automated system. It is planned for the ultrasound images to be automatically scored to further assist doctors.