Neural Network Applications in Sleep Stage Processing

Acknowledgement

We take this opportunity to express our sincere deference to Dr. Vaishali

Kulkarni, Head of Department, EXTC for including this project in the curricu-

We would like to thank our project co-ordinators Prof. Vipul Gohil and Prof.

Sumita Nainan, for guiding us throughout the project.

We are very grateful to our project mentors Nieves Crasto and Richa Upad-

hyay, whose profound encouragement, co-operation, guidance and keen supervi-

sion at every stage of the project, inspired us in pursuing and completing the

project within schedule. Without their guidance and support, this project would

not have been possible.

We convey our immense gratitude to Dr. S.Y.Mahiskar, Dean, MPSTME for

the continuous help and encouragement and the friendly atmosphere of education

provided by him.

Payal Paranjapaye (D030)

Richa Sharma (D044)

Shruti Singh (D049)

B.Tech EXTC

Abstract

The basic idea of this project is to classify EEG signals recorded during sleep into

the various stages, like REM (Rapid Eye Movement) stage, NREM (Non-REM)

stage, wake stage and movement stage. Firstly, temporal and spectral features

are extracted from the signals. Once the features are extracted, they are classified

into various stages using logistic regression. Finally, we compare the accuracy of

the temporal and spectral techniques.

Contents

1 Introduction 1

1.1 EEG Signal Characteristics . . . . . . . . . . . . . . . . . . . . . 2

1.1.1 Delta Waves (0.5 to 3 Hz) . . . . . . . . . . . . . . . . . . 2

1.1.2 Theta Waves (3 to 8 Hz) . . . . . . . . . . . . . . . . . . . 3

1.1.3 Alpha Waves (8 to 12 Hz) . . . . . . . . . . . . . . . . . . 3

1.1.4 Beta Waves (12 to 38 Hz) . . . . . . . . . . . . . . . . . . 3

1.1.5 Gamma Waves (38 to 42 Hz) . . . . . . . . . . . . . . . . 4

1.2 Types of Sleep Stages . . . . . . . . . . . . . . . . . . . . . . . . . 5

1.2.1 Non-rapid eye movement sleep (non-REM) . . . . ....