Modelling of sleep behaviors of patients with mood disorders

Loading...
Thumbnail Image
Journal Title
Journal ISSN
Volume Title
Perustieteiden korkeakoulu | Master's thesis
Date
2023-01-23
Department
Major/Subject
Machine Learning, Data Science and Artificial Intelligence
Mcode
SCI3044
Degree programme
Master’s Programme in Computer, Communication and Information Sciences
Language
en
Pages
53+1
Series
Abstract
Sleep is an essential function of the human body. It has a restorative effect on both physical and mental health functions. Short and long-term consequences of sleep disruption include changes to stress response, anxiety, and depression, as well as deficiencies in memory, cognition, and performance. Several methods have been developed to assess sleep. While polysomnography is considered the golden standard of sleep assessment, researchers have focused on alternate ways of tracking sleep using non-intrusive and less costly methods such as actigraphy. Some studies suggested that screen activity from smartphones can be an indicator of the sleep and wake states of an individual as smartphone usage increased drastically in the last decade. Mood disorders are mental health conditions that disrupt the emotional state of individuals. Sudden and extreme mood changes interfere with the patients’ daily rhythm in many ways, including their sleep behavior. Timely diagnosis of the severity of mood disorders plays a critical role in their treatment process. Previous research shows strong links between decreased sleep quality in patients suffering from mood disorders. This thesis uses the data from a digital phenotyping study, Mobile Monitoring of Mood (MoMo-Mood), to analyze the sleep behaviors of patients with mood disorders using some sleep parameters. In addition, a predictive model is built to investigate the severity of depression using the information tracked via actigraph and bed sensor. Lastly, the perceived sleep quality from questionnaires is compared with the data tracked by these sensors to evaluate the differences in the three different groups of patients: bipolar disorder, borderline personality disorder, and major depressive disorder.
Description
Supervisor
Aledavood, Talayeh
Thesis advisor
Aledavood, Talayeh
Keywords
mood disorders, digital phenotyping, sleep assessment, actigraphy, depression
Other note
Citation