Our goal of the project was to develop a simple and reliable method to detect insomnia, given a dataset of a patient’s alpha waves obtained by an EEG apparatus. We tested digital signal processing principles we learned in classes, including filtering principles and Windowing.
It is known that larger deviations from a population-wide standard deviation of about 1 Hz have been associated with pathological processes, such as deteriorating working memory performance, schizophrenia, Alzheimer’s disease, and addiction. Thus we specifically looked into changes in alpha frequency during both wake before sleep-onset periods and wake after sleep-onset periods.
Still our method needs to be double checked to see whether other pathological processes mentioned above also interfere with our method. From the very initial design of our project, we struggled to think of implementing a target program for diagnosing insomnia, and we are much satisfied with the results.