Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

Gaussian Transformation Enhanced Semi-Supervised Learning for Sleep Stage Classification

Guo, Yifan (2023) Gaussian Transformation Enhanced Semi-Supervised Learning for Sleep Stage Classification. Master's Thesis, University of Pittsburgh. (Unpublished)

[img]
Preview
PDF
Download (1MB) | Preview

Abstract

Sleep disorders are torturing big populations in modern society. To provide efficient help to people with sleep difficulties, accurate sleep monitoring is essential. However, clinical examinations have problems of limited measure durations and biased observations and thus are often insufficient for diagnosis of sleep disorders. Electroencephalogram (EEG) based wearable devices are promising compliments for clinical examinations because of their ability for convenient and reliable long-term monitoring of sleep. This thesis develops both supervised and semi-supervised learning approaches for sleep stage classification, which can be applied on EEG wearable devices. Specifically, we design and implement various EEG based sleep stage classifiers with different feature extraction processes and then compare and analyze these classifiers through experiments. The classifiers are able to obtain satisfactory performance on the test data from a limited number of human subjects, but the learned classifiers usually cannot generalize well on previously unseen human subjects because the EEG signal characteristics vary significantly from person to person. To address this issue, we propose a semi-supervised learning algorithm to mitigate the performance deterioration when dealing with new subjects. We further study and evaluate several Gaussian transformations on EEG band power features to improve the robustness and accuracy of the algorithm.


Share

Citation/Export:
Social Networking:
Share |

Details

Item Type: University of Pittsburgh ETD
Status: Unpublished
Creators/Authors:
CreatorsEmailPitt UsernameORCID
Guo, Yifanyig39@pitt.eduyig39
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Committee ChairMao, Zhi-Hongzhm4@pitt.edu
Committee MemberCan-Cimino, Azimeazime.cancimino@pitt.edu
Committee MemberDallal, Ahmed Hassan SayedAHD12@pitt.edu
Date: 19 January 2023
Date Type: Publication
Defense Date: 4 November 2022
Approval Date: 19 January 2023
Submission Date: 20 October 2022
Access Restriction: No restriction; Release the ETD for access worldwide immediately.
Number of Pages: 59
Institution: University of Pittsburgh
Schools and Programs: Swanson School of Engineering > Electrical and Computer Engineering
Degree: MS - Master of Science
Thesis Type: Master's Thesis
Refereed: Yes
Uncontrolled Keywords: Machine learning, Gaussian transformation, sleep stage classification
Date Deposited: 19 Jan 2023 19:11
Last Modified: 19 Jan 2023 19:11
URI: http://d-scholarship.pitt.edu/id/eprint/43746

Metrics

Monthly Views for the past 3 years

Plum Analytics


Actions (login required)

View Item View Item