Advanced Brain-Wave Analysis For Early Diagnosis of Alzheimer’s Disease (AD)Advanced Brain-Wave Analysis For Early Diagnosis of Alzheimer’s Disease (AD)

Jaron MurphyJaron MurphyThe Ohio State University The Ohio State University

Research Alliance in Math and ScienceResearch Alliance in Math and ScienceComputational Sciences and Engineering, Oak Ridge National LaboratoryComputational Sciences and Engineering, Oak Ridge National Laboratory

Dr. Lee Hively & Dr. Nancy MunroDr. Lee Hively & Dr. Nancy Munro

http://www.ccs.ornl.gov/Internships/rams_08/j_murphyhttp://www.ccs.ornl.gov/Internships/rams_08/j_murphy

AbstractAbstractThe goal of this study is improvement in distinguishing various disease states: normal aging, mild cognitive impairment, early Alzheimer’s disease(AD), The goal of this study is improvement in distinguishing various disease states: normal aging, mild cognitive impairment, early Alzheimer’s disease(AD), and Diffuse Lewy Body disease(DLB). With the help of Dr. Robert Sneddon (University of California, Irvine), a Java program has been designed to model and Diffuse Lewy Body disease(DLB). With the help of Dr. Robert Sneddon (University of California, Irvine), a Java program has been designed to model the Sneddon and Shankle qEEG methodology. The program analyzed data received from another collaborator, Dr. Yang Jiang (University of Kentucky the Sneddon and Shankle qEEG methodology. The program analyzed data received from another collaborator, Dr. Yang Jiang (University of Kentucky School of Medicine),School of Medicine), and calculated the ratio of variance of anterior cortical activity to the variance of posterior cortical activity. This measure will be and calculated the ratio of variance of anterior cortical activity to the variance of posterior cortical activity. This measure will be used to identify the optimal cutoff value to discriminate normal from impaired subjects, and thus improve the accuracy for discriminating among disease used to identify the optimal cutoff value to discriminate normal from impaired subjects, and thus improve the accuracy for discriminating among disease states.states.

Future ApplicationsFuture Applications• Anticipating a clinical device in the next several Anticipating a clinical device in the next several

years that could be used by a physician to provide years that could be used by a physician to provide early diagnosis of AD in 5 to 10 years before AD early diagnosis of AD in 5 to 10 years before AD onsetonset

• Ability to provide early diagnosis of neurological Ability to provide early diagnosis of neurological diseases such as:diseases such as:• Parkinson’s disease Parkinson’s disease • Diffuse Lewy Body disease Diffuse Lewy Body disease • Clinical DepressionClinical Depression• Bi-Polar Disorder Bi-Polar Disorder

The Research Alliance in Math and Science program is sponsored by the Office of Advanced Scientific Computing Research, U.S. Department of Energy. The work was performed at the Oak Ridge National Laboratory which is managed by UT-Battelle, LLC under Contract No. De-AC05-00OR22725. This work has been authored by a contractor of the U.S. Government, accordingly, the U.S. Government retains a nonexclusive, royalty-free license to publish or reproduce the published form of this contribution, or allow others to do so, for U.S. Government purposes.

BackgroundBackground• AD is a neurodegenerative disease of the AD is a neurodegenerative disease of the

nervous system that is: nervous system that is: • non-treatablenon-treatable• affects the cognitive abilities of a personaffects the cognitive abilities of a person• renders them functionally useless in renders them functionally useless in

society society • Federal government estimates Federal government estimates

approximately 4 million people in the U.S. approximately 4 million people in the U.S. has ADhas AD

• Number of people with AD will increase in Number of people with AD will increase in future as number of older persons increases future as number of older persons increases

• Advances in our understanding of AD, but Advances in our understanding of AD, but no effective treatment no effective treatment

• Effectiveness of treatments depend on Effectiveness of treatments depend on implementing them at earliest stage of implementing them at earliest stage of disease as possible disease as possible

MethodMethod• Computational analysis of Computational analysis of

electroencephalographyelectroencephalography (EEG) data using the (EEG) data using the nonlinear Tsallis entropy, implemented in Java nonlinear Tsallis entropy, implemented in Java on desktop systemon desktop system

• Analyzed EEG data collected at UK during Analyzed EEG data collected at UK during administration of a delayed visual recall taskadministration of a delayed visual recall task

Fig. 1

Figure 2. Person wearing an Figure 2. Person wearing an electrocap that holds the electrocap that holds the electrodes in placeelectrodes in place

Research ObjectivesResearch Objectives• Implement the qEEG Methodology in JavaImplement the qEEG Methodology in Java• Analyze UK EEG data to see if Shankle and Analyze UK EEG data to see if Shankle and

Sneddon’s results can be confirmedSneddon’s results can be confirmed• Demonstrate early detection of DLB for the first Demonstrate early detection of DLB for the first

time via qEEGtime via qEEG

Figure 3. Diagram of the four brain wave categoriesFigure 3. Diagram of the four brain wave categories

Fig. 3

ResultsResults• Java code to calculate variance ratio Java code to calculate variance ratio

• Sample numerical results Sample numerical results

• Comparison across 40 data setsComparison across 40 data sets

Fig. 2

Fig. 1 Placement diagram of electrodes Fig. 1 Placement diagram of electrodes on the headon the head

Dsfjladfjladfj;dfdfdfsd

Dsfjladfjladfj;dfdfdfsdlklkllklk