Background and problem statement
Parkinson’s disease (PD) is a neurodegenerative disorder characterized by motor symptoms as tremor (involuntary rhythmic movements of hands, legs, neck or entire body), bradykinesia (movement slowness), rigidity (increased muscle tone) and postural instability and gait dysfunction.
Deep brain stimulation (DBS) is a therapy where a lead is placed deep inside the brain to electrically stimulate a specific area. DBS of the subthalamic nucleus is an established therapy to reduce the motor symptoms in PD. However, it is associated with side effects. One of the most important side effects from a patient’s perspective is deterioration of speech quality as it leads to difficulty in communication and social isolation.
Phonetic data is clinically analyzed using the Multi-Dimensional Voice Program from Kay Pentax. Typical outcome measures indicate the quality of speech but may not be sensitive enough to distinguish subtle changes in speech quality due to different DBS parameter settings.
You will collect phonetic data from healthy volunteers. Data from PD patients will be provided for a range of DBS settings. The goal of this assignment is to establish features in phonetic data that characterize the quality of speech. The measures have to be sensitive enough to (1) distinguish PD patients from healthy subjects and (2) show changes in speech quality for different DBS parameters.
•Perform a literature study on phonetic data analysis
•Collect phonetic data from 10 healthy volunteers
•Establish features that quantify speech characteristics in the collected data
•Develop algorithms in Matlab to analyze phonetic data and extract features
•Compare data from healthy volunteers with data from PD patients
•Compare data from PD patients for different DBS parameters
•Discuss the results
•Report your findings in the format of a scientific journal article
•UT, Biomedical Signals and Systems (BSS)
•Case Western Reserve University, McIntyre Lab, Cleveland, USA
Supervision and info: