Automated Recognition of Dyskinesia and Motor Fluctuation in Patients with Parkinson's Disease Using Surface Electromyography

Grant Winners

  • Ming-Shun Cheng, Sc.D. – College of Health Care Sciences
  • Cheryl Hill, Ph.D. – College of Health Care Sciences
  • William Kelleher, Ph.D. – Center for Psychological Studies
  • Serge Roy, Sc.D. – Boston University

Deans

  • Richard Davis – College of Health Care Sciences
  • Karen Grosby – Center for Psychological Studies

Abstract

Award Winners

Clinically, the hallmarks of Parkinson's Disease (PD) include bradykinesia, rigidity and tremor. Levadopa, together with a decarboxylase inhibitor, continues to be the standard of therapy for PD. However, studies have shown that although Levadopa appears to be effective, patients who have taken Levadopa for several years begin to develop other symptoms such as Levadopa-induced dyskinesia (LID) along with an on-off phenomenon. These motor abnormalities can negatively impact quality of life by interfering with the ability to carry out daily activities, work, and social interaction. Assessments that accurately discriminate between the various kinds of abnormalities, their magnitude, and time history corresponding to dosing or intervention schedules may have a significant influence on therapeutic interventions aimed at reducing motor abnormalities. Current methods of accomplishing this objective are severely inadequate primarily because of their reliance on self-reports by the patient, home diaries, or a variety of clinical rating scales. These methods have been known for their poor reliability, poor sensitivity or time requirement to complete. The aim of this project is to develop a new method for identifying motor abnormalities based on surface electromyography (EMG) technique. Specifically, the pilot study will identify the EMG characteristics of motor abnormalities over time in patients with LID and use these signal features as inputs to an artificial neural network system to automatically and accurately recognize the presence and severity of LID. Future studies under separate funding will address the possibility of utilizing such methods that will automatically monitor motor fluctuations in a patient's home during the course of carrying out their routine activities of daily living. The accomplishment of this goal could greatly improve the limited capabilities of subjective reports that clinicians and researchers currently rely on to monitor the effectiveness of conservative (exercise), pharmacological (levodopa/carbidopa) and surgical (stem cell, deep brain stimulation) interventions in patients with PD.