MMM – Machine Learning Analysis of Movement Measurements
Not too long ago, wearable devices to track the physical activity of individuals such as fitness trackers, running watches, etc., became popular and conquered the market.
Obviously, such devices are a help for many people to motivate themselves to be physical active. Moreover, it is a way to evaluate the physical fitness and to track personal progress.
By means of kinematic recordings of different exercises for upper limb strengthening, this thesis shows how the basic idea of tracking physical activity can be extended to the area of physiotherapy.
Therefore, three classification procedures are elaborated in order to approach the problem from different angles. Classical machine learning algorithms (Logistic Regression, Linear Discriminant Analysis, Random Forest Classifier, Support Vector Classifier) as well as Neural Networks are taken into account. Individual strengths and weaknesses of the different approaches are identified.
It is shown that automatic detection and classification of physiotherapeutic exercises is possible with high accuracy. Even single exercise repetitions can be detected with two of the three elaborated approaches.
FeeL – Sensory Feedback for Leg Prosthesis
The execution of complex movements, as they occur in everyday life in a variety of ways, is based on a sophisticated control loop, with a feed-forward initialization of the movements, and a feedback-based control of the movement. In the case of persons with amputations, the feedback of the movement apparatus is interrupted, and the movement is therefore less precise. Hence, many lower-limb amputees have problems walking naturally and efficiently on ramps, stairways or uneven terrain such as meadows or gravels naturally and efficiently. The result is a markedly increased fall incidence in people with amputations of the lower extremities, as well as movement disorders, which often lead to chronic pain.
Within FeeL, the aim is to investigate how the closure of this control loop with a technical replacement of the pressure feedback from the sole of the foot has the effect on the stability and dynamics of movements of amputees as well as on pain related to the amputation. For this purpose, a vibro-tactile feedback system has to be developed that measures the pressure between the prosthetic foot and the ground and stimulates the stump of the amputee with vibrators.In order to make the feeling of feedback more natural, the suitability of Targeted Sensory Reinnervation on the lower limb will be investigated. Thereby an afferent sensory nerve that originally innervated the one part of the sole of the foot is dissected, and connected to a skin nerve from the stump. This newly innervated skin area can then be used to stimulate the sural nerve and thus can pass information from the prosthetic foot onto the foot area in the somatosensory cortex. Thus, the feedback is more authentic for the user. To Acive the goals of the Project we collaborate with the Medical University Vienna, the Orthopedic Hospital Speising and Otto Bock Healthcare Products.
Our Prothestics Group focusses on pain problems prostheses wearers are experiencing, and on prosthetic devices for the upper extremities.
Diagnostic Devices for Sensory Systems
Corneal Surface of the Eye
To correct for visual deficits, it is possible to reshape the cornea through the use of lasers (LASIK). This often leads to spectacular improvements of visual accuracy. However, in some patient this effect decreases over time. To better understand these developments, we use image processing techniques to automatically determine the thickness of the corneal epithelium.
These tests not only show which information can be obtained from modern recording technologies. They may also help us to understand which factors are relevant for the visual assessment of other people, and how our brain reaches it impressions.