Electrooculogram (EOG) Signal Classification Using Moving Average Technique and its Application to Drive Direct Current Motors

Background: Electrooculogram (EOG) signal is one of the bioelectric signals acquired from the human body to study the movements of eyes and also to design and develop assistive devices. These devices can be mobility devices, video gaming devices or any other assistive device. Methods: Assistive devices are especially designed for quadriplegic or spinal cord injured patients. Motors are one of the key components in the design of mobility devices. These motors are to be driven with the help of EOG commands. This paper explains the process of eliminating involuntary eye movements while driving the motors under the control of EOG signals. The system design is carried out in two ways. Initially the system is designed in such a way that the motors are driven even for involuntary eye movements which is a major drawback of the system. Conclusion: This drawback has been overcome successfully by introducing the moving average technique during classification of the EOG signals. The systems overall classification accuracy is also computed by constructing confusion matrix and has produced high sensitivity, specificity with overall average accuracy of 90.91%.