• Like


Flash Player 9 (or above) is needed to view presentations.
We have detected that you do not have it on your computer. To install it, go here.

Microsoft PowerPoint - ourfinal-1.ppt [Compatibility Mode]

Uploaded on


  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Be the first to comment
    Be the first to like this
No Downloads


Total Views
On Slideshare
From Embeds
Number of Embeds



Embeds 0

No embeds

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

    No notes for slide


  • 1. Group EEG (Electroencephalogram) (El t h l ) Anthony Hampton, Tony Nuth, Miral Patel Nuth, (Portions credited to Jack Shelley-Tremblay and E. Keogh) Shelley- 05/09/2008 5/9/2008 1
  • 2. Outline Introduction Goal Methodology M th d l Results Discussion Conclusion 5/9/2008 2
  • 3. Goals The goal of this project is to evaluate EEG data from 19 subjects using various techniques of mathematics mathematics. 5/9/2008 3
  • 4. EEG A recording of the electrical waves that sweeps over the brain’s surface. It is measured by the electrodes that is placed on top of the scalp. 5/9/2008 4
  • 5. Parts of the brain We are interested in the Primary Motor Cortex, and Pre-motor Cortex Pre- 5/9/2008 5
  • 6. Understand the Waves In neurophysiology, an action potential (also neurophysiology, known as a nerve impulse or spike) is a pulse- spike) pulse- like wave of voltage that travels along several types of cell membranes. membranes. 5/9/2008 6
  • 7. Placement International 10- 10-20 system, Place an Pl Electrode on Each point. 5/9/2008 7
  • 8. Event related Potentials An event-related potential (ERP) is any stereotyped event- p ( ) y yp electrophysiological response to an internal or external stimulus. More simply, it is any measured brain response that is directly the result of a thought or perception. perception. By collecting multiple trials of the same type of stimuli, we can enhance the signal and reduce the noise using simple math(typically neuroscientist use this) this). 5/9/2008 8
  • 9. Defined an Epoch A series of time points locked in a significant point time. The button press is our significant point in time time. 5/9/2008 9
  • 10. A subject’s data subject s Example of an Epoch E h 5/9/2008 10
  • 11. Same data Same data At a different Epoch E h 5/9/2008 11
  • 12. Time Series A time series is a sequence of data points, points, measured typically at successive times, times spaced at (often uniform) time intervals. 5/9/2008 12
  • 13. Time Series What is a time series? 5/9/2008 13
  • 14. What EM clustering does How do we classify points and estimate parameters of the models in a mixture at the same time? Adaptive ft l t i Ad ti soft clustering: EM. D t EM. Data points are assigned to each group with a probability equal t a lik lih d of th t b bilit l to likelihood f that point belonging to that group. 5/9/2008 14
  • 15. What is EM - Expectation Maximization A statistical model that makes use of the finite Gaussian mixture models. A set of parameters are recomputed until a desired value is reached Initial I iti l variables are randomly i iti li d i bl d l initialized 5/9/2008 15
  • 16. The methods of EM Initialization: Pick start values for parameters (for us it was making random models and setting a sigma) g g ) Iteratively process until parameters converge g Expectation (E) step: Calculate weights for every data point and update the weights to affect further steps ff t f th t Maximization (M) step: Maximize a log likelihood function with the weights given by E 5/9/2008 step to update the parameters 16
  • 17. Evaluate Initialized 2 models of data using the mean of the EEG g entered using the first half of data over time for the first model and the second half of over time for the second model Co pa ed each ode and created Compared eac model a d c eated a weight matrix e g t at Normalized the data 5/9/2008 17
  • 18. OSB algorithm OSB : Optimal Subsequence Bijection It is an algorithm that determines the optimal subsequence bijection between two sequences of real numbers. We were given a code ‘tsDAGjump4’ that worked for only one channels and we modified so it can work for more than 1 channels.( works for 40 channels) Modification: Created difference matrix with each entry containing differences of corresponding elements. 5/9/2008 18
  • 19. Why OSB algorithm? The OSB is efficient because we use DAG(Directed Acyclic Graph), cheapest path to find the solution. By using DAG in OSB, we get p y g , g perfect and correct results on Time Series dataset. DAG helps us to get rid of outlier elements and get one-to-one or one-to- onto bijection of a sequences. Comparing OSB with DTW using warping window OSB shows window, that by skipping elements improves results. 5/9/2008 19
  • 20. Directed Acyclic Graph This is a simple example of DAG. By skipping over outlier elements we get perfect result. 5/9/2008 20
  • 21. OSB algorithm This program is used to find: Ts - Time Series DAG – Directed Acyclic Graph OSB between two sequences of real numbers D is to find distance between two elements. C is for the jump cost.(penalty for skipping an element) W- weight of edges 5/9/2008 21
  • 22. OSB algorithm g Find b Fi d subsequences of tf two elements. l t Create dissimilarity matrix. Use shortest path algorithm on Directed Acyclic Graph. Find jump cost. Fi d j t Nodes are index pairs of matrix. The main thing in the algorithm is to find edge weights of DAG. 5/9/2008 22
  • 23. Wavelet “Wavelets are mathematical functions that cut up Wavelets data into different frequency components, and then study each component with a resolution matched to its scale.” (IEEE Computational Science and Engineering, Summer 1995, vol. 2, num. 2, published b the IEEE Computer Society) bli h d by h C S i ) 5/9/2008 23
  • 24. Discussion of Results We used OSB to obtain our results. Our results consist of two sequences a and b then find subsequences a’ of a and b’ of b so that a’ matches best a b a with b’. Results are divided on two parts: Cluster Precision - means x% of time that you will cluster an epoch correctly. y Cluster Recall - means y% of time that you will cluster a known left or right button 5/9/2008 24
  • 25. Discussion of Results Example: typeout1: [75 76 27 21 22] = a rtypeout1: [77 72 24 22 24] = b ltypeout2: [25 24 73 79 78] = a’ rtypeout2: [23 28 76 78 76] = b’ t t2 For Cluster Precision: left button cluster: 76, right button cluster: 72 (from a and b) Formula : 76/ (76+72) = 0 51351351 = 51% 0.51351351 For Cluster Recall: left button cluster: 76, right button cluster: 28 (from a and b’) Now, Now we applied the formula so: 76/ (76+28) = 0 73076923 = 73% 0.73076923 Note: Apply same formula to find both right cluster precision and recall. 5/9/2008 25
  • 26. Conclusion We’re given total 19 subjects (EEG Datasets) but we We re derived correct result for only 10 subjects. Other 9 subjects gave us all zeros as a result. ltypeout1: [0 0 0 0 0] rtypeout1: [0 0 0 0 0] ltypeout2: [0 0 0 0 0] rtypeout2: [0 0 0 0 0] …..error…… 5/9/2008 26
  • 27. Extra References Yang Ran, Expectation Maximization : An Ran, Approach to Parameter Estimation”, www.umiacs.umd.edu/ shaohua/enee698 www umiacs umd edu/~shaohua/enee698 a_f03/em.ppt Andrew Blake Bill Freeman “Learning Blake, Freeman, Learning and Vision: Generative Methods”-ppt ICCV Methods”- 2003 October 12 2003 12, 5/9/2008 27
  • 28. Thank You 5/9/2008 28