An ANN Approach to EEG Scoring Anand Lakshmanan ECE 539 Dec 12 , 2001 Frequently, presenters must deliver material of a technical nature to an audience unfamiliar with the topic or vocabulary. The material may be complex or heavy with detail. To present technical material effectively, use the following guidelines from Dale Carnegie Training®.   Consider the amount of time available and prepare to organize your material. Narrow your topic. Divide your presentation into clear segments. Follow a logical progression. Maintain your focus throughout. Close the presentation with a summary, repetition of the key steps, or a logical conclusion. Keep your audience in mind at all times. For example, be sure data is clear and information is relevant. Keep the level of detail and vocabulary appropriate for the audience. Use visuals to support key points or steps. Keep alert to the needs of your listeners, and you will have a more receptive audience.

Introduction EEG – Electroencephalogram Measures brain activity. Used to diagnose and cure brain related disorders. EEG is often contaminated with eye artefacts..eg blinks. These “extra” signals are currently removed manually during data analysis. Involves time and money. I have used ANN modeled on Startle blink under noise. In your opening, establish the relevancy of the topic to the audience. Give a brief preview of the presentation and establish value for the listeners. Take into account your audience’s interest and expertise in the topic when choosing your vocabulary, examples, and illustrations. Focus on the importance of the topic to your audience, and you will have more attentive listeners.

Data Collection & Preprocessing Source : Personal Involvement in experiment set up for data collection at The Psychology Department UW Madison. 43 subjects * 21sessions * 10000 = data points for classification. DOS Snapshot Stream – Software by the company HEM Data Corp. This program collects data from channels as per the specified sampling rate and gain settings of an attached Bio-Electric Amplifier. Hardware Contour following Integration of Raw Startle Data. The data is streamed to a stimulus file. Matlab used to read the stimulus file. Scaling of data. 60 Hz digital software notch filtering using Matlab Signal Processing ToolBox

Approach Pattern Classification problem that distinguishes EEG from Eye Artefact MLP with back propagation. Program written on the lines of the all powerful “bp.m” written by Prof .Hu

Results and Future I tried out several combinations and trained the network extensively. Attractive classification rate of 90% so far using learn rate = 0.01 and momentum = 0.4 on a 3 layer MLP with 2 neurons in hidden layer. Yet to do cross validation and hopefully Crate will improve.