Title

A Low-Cost Portable Electroencephalographic System for the Identification and Prevention of Microsleep Episodes

Lead Author Major

Bioengineering and Applied Mathematics

Lead Author Status

Senior

Second Author Major

Bioengineering

Second Author Status

Senior

Third Author Major

Bioengineering

Third Author Status

Senior

Fourth Author Major

Bioengineering

Fourth Author Status

Senior

Fifth Author Major

Bioengineering

Fifth Author Status

Senior

Format

SOECS Senior Project Demonstration

Faculty Mentor Name

Fadi Muheidat

Faculty Mentor Email

fmuheidat@pacific.edu

Faculty Mentor Department

Electrical Engineering

Additional Faculty Mentor Name

Jeff Burmeister

Additional Faculty Mentor Email

jburmeister@pacific.edu

Additional Faculty Mentor Department

Bioengineering

Abstract/Artist Statement

Studies have shown that lapses into microsleep episodes are common in truck drivers–a particularly dangerous event considering the oft-catastrophic consequences of falling asleep at the wheel. In this work, we have undertaken development of a low-profile electroencephalographic (EEG) device that would determine user consciousness levels and alert the user if a microsleep episode is imminent. Fitting within a baseball cap, 9 dry, comb-style electrodes are mounted within an elastic band within the 10-20 EEG placement convention. The transduced, microvolt-level electrical activity is then subjected to a low-pass filter and subsequently amplified by a Delta-Sigma differential amplifier wired in bipolar montage. The device is controlled by an Atmel ATmega microcontroller clocked at 16MHz which interfaces with a Bluetooth low-energy module for wireless transmission of data. The electronics are to be implemented on a printed circuit board, and the device will be rechargeable. To process the acquired data, the device transmits data via Bluetooth 4.1 to an Android app programmed in Java. This smartphone app is based upon an open source Bluetooth app which has been modified to receive, interpret, and send signals to the PCB. The data is separated into discrete time-packets and subjected to a continuous power-spectrum analysis algorithm, which will be trained using labelled EEG data from the Sleep-EDF database. Once the spectral signature of sleep is detected, the Android app will issue an asynchronous interrupt to the device, engaging an alarm system comprised of two vibratory motors and an audible alarm until the user disables the alarm. Currently competitive options that exercise these features are prohibitively expensive, have low battery life, and have only been available in European markets. As the project progresses, the goal is to ensure that brain activity can be accurately detected and interpreted to alert a user before they fall asleep.

Location

School of Engineering & Computer Science

Start Date

4-5-2018 2:30 PM

End Date

4-5-2018 4:00 PM

This document is currently not available here.

Share

COinS
 
May 4th, 2:30 PM May 4th, 4:00 PM

A Low-Cost Portable Electroencephalographic System for the Identification and Prevention of Microsleep Episodes

School of Engineering & Computer Science

Studies have shown that lapses into microsleep episodes are common in truck drivers–a particularly dangerous event considering the oft-catastrophic consequences of falling asleep at the wheel. In this work, we have undertaken development of a low-profile electroencephalographic (EEG) device that would determine user consciousness levels and alert the user if a microsleep episode is imminent. Fitting within a baseball cap, 9 dry, comb-style electrodes are mounted within an elastic band within the 10-20 EEG placement convention. The transduced, microvolt-level electrical activity is then subjected to a low-pass filter and subsequently amplified by a Delta-Sigma differential amplifier wired in bipolar montage. The device is controlled by an Atmel ATmega microcontroller clocked at 16MHz which interfaces with a Bluetooth low-energy module for wireless transmission of data. The electronics are to be implemented on a printed circuit board, and the device will be rechargeable. To process the acquired data, the device transmits data via Bluetooth 4.1 to an Android app programmed in Java. This smartphone app is based upon an open source Bluetooth app which has been modified to receive, interpret, and send signals to the PCB. The data is separated into discrete time-packets and subjected to a continuous power-spectrum analysis algorithm, which will be trained using labelled EEG data from the Sleep-EDF database. Once the spectral signature of sleep is detected, the Android app will issue an asynchronous interrupt to the device, engaging an alarm system comprised of two vibratory motors and an audible alarm until the user disables the alarm. Currently competitive options that exercise these features are prohibitively expensive, have low battery life, and have only been available in European markets. As the project progresses, the goal is to ensure that brain activity can be accurately detected and interpreted to alert a user before they fall asleep.