School of Computing


Using Accelerometer-Based Activity Recognition to Improve Motor Performance in Parkinson's Disease

In this project we aim to develop a method that can be used to assess one of the properties of motor skill: the efficiency of motion. This will have application in degenerative conditions such as Parkinson's Disease and Dementia which have a significant impact on motor abilities.

Through sensors worn on the body or embedded into objects of daily use we can infer the activities performed by a subject. However, so far relatively little work has been invested into a further, detailed analysis of these segmented activities, although extracting their characteristics, i.e. how these activities were performed, would be beneficial to a variety of applications. Applications of exploring this include rehabilitation, pain therapy, sports and professional training in tool usage, e.g. for mechanics among others. Information about the development of these motor performances, i.e. if there is an increase or decline over time, can be very beneficial, particularly in medicine and specifically in degenerative conditions such as Parkinson's Disease where an assessment of a decline in motor ability is a common diagnostic tool.

Our method for measuring motor efficiency will be based on the energy distribution in Principal Component Analysis (PCA) and we will use it to infer a single, normalised metric that is intimately linked to signal complexity and allows comparison of (subject-specific) time-series. We will evaluate the approach on artificially distorted signals and apply it to a simple kitchen task to show its applicability to real-life data streams.