P09004: Eye Movement Tracking Device

Identify Customer Needs

Table of Contents

Step 1. Gather Raw Data From Customers

Interactions with Customers

First Customer Interview
Interviewers: Bryan McCormick
Sponsor: Dr. Amuso
Date 19 March 2008, Senior Design Floor
Paraphrased Interview Transcript
Interviewer: Should the product be portable?
Sponser: Product should be a stand alone, not a part of other equipment.
Interviewer: What is the target market?
Sponser: Faculty/Staff interested in Biomedical and image processing.
Interviewer: Do you have an approximate budget in mind?
Sponser: Its up to the project team to propose a budget for my approval.
Interviewer: Will the device measure actual EOG waves or the physical blinking of the eye?
Sponser: I don't have all the specifics on it but it has to do with the electromagnetic signal created during blinking. See Dr. Phillips(EE) for additional information on EOG waves.
Interviewer: What is the primary motivation behind this project?
Sponser: The #1 goal is for a MSD project, however it must also be practicle. It will also be used to demonstrate to prospective students to get them interested in Biomedical Engineering. It needs to be easy enough to use that a middle school student would be able to operate it.
Interviewer: Do you have any additional background information on EOG waves?
Sponser: Try asking Dr. Phillips and Dr. Borkholder. They aren't currently aware of this project, but they may be able to help with background information.
Interviewer: Is there a preferred method of data output or display?
Sponser: The measurements of the EOG waves need to be time synchronized with the images of the eye, but display and data storage are up to the design team to propose.
Interviewer: Is there an expected duration of each test that will be run with this device?
Sponser: Not really, it should be more of a push when you want to start collecting, push again when you want to stop.
Interviewer: Is there any other additional information you can think of that I should know?
Sponser: Image processing is not within the scope of this project.

Background Interview 1
Interviewers: Bryan McCormick
Interviewee: Dr. Phillips
Date 7 May 2008, 09-3177
Summary of Relevant Information
Eye movements generate electrical signals that can be sensed
There is a charge distribution between the cornea and the retina that causes these signals
Need a biomedical amplifier to detect this
Blinking generates a large amount of noise in the EOG signal
Electrodes are placed on the eye, one each on the top, bottom, left and right
Surface electrodes come in two varieties:
Low impedance-contact in conductive solution on skin, very messy
High impedance-Doesn't need to adhere to skin but is more prone to artifacts
Dr. Saber(EE) may be a useful resource in the field of image processing and motion tracking
Dr. Phillips will be a Faculty Consultant on the project.

Background Interview 2
Interviewers: Bryan McCormick
Interviewee: Jeff Pelz
Date 7 May 2008, 77-2225
Summary of Relevant Information
His eye trackers use a multiplexer to overlay the seen image with the image of the eye
This still has some issues, a CE on the team working towards this issue may give him a reason to help with funding(not really in scope of project)
Eye is a dipole, also need a ground electrode on forehead or back near ear
Dual Purkinje method compares 1st surface and 4th surface reflections (1/60th of a degree accuracy)
Optical system only has 1 degree accuracy
2004 Paper by Babcock and Pelz on how to build apparatus(in references section)
System has issues dealing with blinks
People almost always blink during large eye movements
With binocular system you can use the better signal of the two
His system tracks the center of the pupil and the 1st surface reflection
Eye movements cause a big difference in the two signals
A camera movement will cause the two to move together
This can differentiate eye movements from camera movements
His system uses CMOS cameras and IR diodes from supercircuits.com(in references)
CCD cameras are larger and use more power
With IR illuminator its best to use Black and White camera
Dr. Herbert(amhgss@rit.edu) may be a useful resource with experience in EEG, EOG and evoke potentials
Jeff Pelz will be a Faculty Consultant on this project.

Step 2. Interpret Raw Data in Terms of Customer Needs

In Addition:

Step 3. Organize the Needs into a Hierarchy

  1. Safety
  2. Accuracy
  3. Data Storage
  4. Data Collection
  5. Ease of Use
  6. Durability
  7. Cost
  8. Maintenance
  9. Portability
  10. User Comfort
  11. Aesthetics

Step 4. Establish the Relative Importance of the Needs

Needs Summary
Need Importance
Safety 1
Accuracy 2
Data Storage 2
Data Collection 2
Ease of Use 3
Durability 3
Cost 4
Maintenance 4
Portability 4
User Comfort 5
Aesthetics 5

1=Most Important

5=Least Important

Home | Planning | Concept Development | System Level Design | Detail Design | Testing and Refinement