The goal in this project was to investigate whether or not, it is possible to detect knee abnormalities using vibration analysis.
Briefly speaking, when you have osteoarthritis, the surface of your joints are not as smooth as they are supposed to be. There will be a lot of friction when the joints move, which will generate a grinding sound. Sometimes the sound is so strong that you can hear it in the room with your ears.
For this project, we built a brace that had 5 accelerometers attached to it. the accelerometers (sensors) would record the vibrational signals from the knee, as the patients walked or moved. the recorded signals were later analyzed using latent variable methods to predict whether or not the patient has osteoarthritis, or meniscus tears.
The brace was designed in such a way that it would not limit the motion of the knee and ensure the sensors are in a relatively fixed position during the test. It may seem that the knee has a simple rotational motion, but in reality the knee has a very complex range of motion and the relative position of the joint changes as the knee swings and extends.
Following two posters were published as the results of the study: