Available masters and doctoral projects

An overview of opportunities for students to work with scientists at Reykjavik University on research projects. Please contact the person listed within each project for further information.

Undergraduate level projects are on the Icelandic portion of RU's website: Available undergraduate level projects

  • Snore detection

Snore detection

MSc level project, School of Science and Engineering

Abstract

Develop an algorithm to automatically classify snores. Snores can originate from different sites in the upper airway and snores originating from different places may have specific properties such as relative power distribution in certain bands. During sleep a range of breathing noises can be detected such as mouth breathing, loud breathing, and snoring. Snoring can occur as an individual event or a series of events, snore trains. Snoring is not used in routine diagnosis of sleep disorders, snore characteristics have been linked to specific sites of snore origin. The method will be developed by analyzing a data set containing manually scored snore events of subjects sleeping in a controlled environment, and of subjects sleeping in their homes.

Description

Snoring is not just an annoyance to those who have to listen to it, but can also be an indicator of an underlying respiratory problem during sleep and an independent risk factor for disease. Snore sounds can be recorded during a standard sleep study conducted either in the hospital or at home. Respiratory sounds range from loud breathing to snoring and wheezing sounds. The common cause of the respiratory sounds is an obstruction in the airway. The site of the obstruction is an important component in the diagnosis and the decision to choose treatment. Common treatments to snoring involve surgery on the upper airway, depending on location of snore origin specific parts of the upper airway need to be operated on.

Goal

In this project data from sleep studies will be analyzed and clustered depending on the respiratory sound properties. The clustering will be used to stratify the study subjects and identifying common factors in subject groups.

Funded positions

No

Other

This project is a collaboration between Reykjavik University's Language and Voice Lab and Nox Medical.

Contact

Jón Guðnason, jg@ru.is