Please use this identifier to cite or link to this item:
Title: Machine learning algorithm for sleep study
Authors: Guo, Shuli
Keywords: DRNTU::Engineering::Electrical and electronic engineering
Issue Date: 2017
Abstract: This project involves implementation of machine learning algorithm for sleep study. It aims to diagnose Obstructive Sleep Apnea (OSA) by implementing a machine learning algorithm. The standard and conventional diagnosis of sleep disorder is Polysomnography (PSG), also known as sleep study. During the process of PSG, various bio-signals were collected as parameters to diagnose sleep disorder syndromes. However, using PSG involves the analysis of huge amount of data, which is time consuming. Therefore, the objective of this project is to study and develop a machine learning based algorithm that is able to analyze data automatically to perform sleep disorder diagnosis. As snore parameters are essential factors to predict sleep disorder, hence snoring sound recorded at National University Hospital (NUH) PSG laboratory is used as the data for this project. With this objective, the machine learning algorithm was developed in three stages including feature extraction, feature selection and classification. Features such as formants frequency, Mel-frequency cepstral coefficients (MFCCs), energy were extracted, and then, fisher’s ratio coefficients were calculated to select the features, lastly, classification was done by using support vector machine(SVM). 78.3% accuracy was obtained from the classification learner in the result. MATLAB Scripts were programmed for implementing the whole project. In conclusion, the features used are discriminating and the performance of the classification learner is satisfying.
Schools: School of Electrical and Electronic Engineering 
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:EEE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
  Restricted Access
FYP report2.86 MBAdobe PDFView/Open

Page view(s) 50

Updated on Jul 12, 2024

Download(s) 50

Updated on Jul 12, 2024

Google ScholarTM


Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.