After being awarded £347,000, the team will spend 18 months building a prototype system using the Kinect, which will provide real-time feedback to doctors and patients with facial paralysis through an on-screen avatar.
As the patient completes his/her facial exercises, the system will automatically detect and track asymmetries on either side of a patient’s face - such as mouth corner, eyelids and cheeks. This information is then displayed on a screen in real-time to give immediate feedback to the user, and stored to track success levels for each task.
"This tool will recognise the differences between an undesirable one-sided unilateral movement and a desired symmetrical movement, across both sides of the face. This information would then be conveyed via the TV using an animated representation, or avatar, of the patient," Dr. Philip Breedon, lead academic on the project commented.
"Using the avatar as both a visual and oral communicator, the system will take the patient through a series of exercises and indicate the degree of success. It is hoped that the information will also be sent electronically to a clinician to allow them to track the patient’s progress."
This digitalisation of what is still a paper-based exercise for stroke victims provides extra analytical data both for both parties concerned. The patient will be able to see real-time feedback on specific facial asymmetries, allowing them to apply more concentration to said area, and the therapist will be able to give a better diagnosis.
Dr. Breedon added: "The daily recommendation of stroke rehabilitation for facial paralysis is 45 minutes. We're hoping that this technology will not only help to reduce the burden on the NHS but more importantly improve the quality of life of stroke survivors, providing them with encouragement and recognising the accomplishments during their recovery."
Prototype building is due to begin in April 2013. We will be following this closely
Source: Nottingham Trent University