Promising Outcomes of Original Grant:
In our original study, the aim was to develop computational tools based on signal processing and machine learning for the diagnosis and prediction of Parkinson's Disease (PD) and related diseases based on electroencephalogram (measures electrical activity in the brain; EEG) data in pre-symptomatic individuals. We successfully classified individuals with REM behavior disorder (RBD) based on their subsequent diagnosis using data recorded on average of eight years before their diagnosis was given. Although these results tentatively confirm our hypothesis of an EEG biomarker for PD, we did see performance variability that may be linked to recently proposed relationships between RBD, PD and other synucleinopathies.
Objectives for Supplemental Investigation:
In our supplemental study, we will first attempt to validate our original results using a larger data set collected from several new sites. Once validated, we will develop several aspects of a computational platform by exploring new EEG features and classification techniques. We are especially interested in understanding the evolution of these diseases and how they affect the performance of our system, which in turn will provide new data for understanding the neurophysiological basis of this evolution and possibly better stratification of diagnosis.
Importance of This Research for the Development of a New PD Therapy:
If we are ultimately successful in developing a biomarker for PD and related diseases, such as Lewy body dementia, with real predictive power in pre-symptomatic individual, it will open the door to new treatments in several ways. First, early detection enables early prescription of existing treatments that are known to have a greater effect if administered early, thereby reducing the number of years a person with PD spends in the later, more debilitating phases of the disease. Second, it allows for development of new types of treatments that may be administered in the earlier phases of the disease. Finally, it provides a tool for screening research participants that will greatly improve the chances of success by increasing the power of the studies.