Lan Wei

Thesis: 
Development of an EEG based sleep assessment scoring system using an Interpretable Machine Learning Approach

Research has shown that sleep plays an important role in brain development and that extensive sleep is required for rapid brain growth, connectivity, and synaptic plasticity. EEG changes in the first few months after birth exceed any other period of individual development.The aim of this study is to use EEG to examine the quality of sleep by classifying the sleep stages and calculating the percentage of deep sleep (stage 3), and the number of sleep spindles in sleep stages 2 automatically.We will explore two types of sleep EEG data sets from Cork University Hospital. The first set contains EEG data from 180 full-term infants recorded at four months of age while the second set is from 100 infants born prematurely with an EEG recorded pre-discharge at 36 weeks gestational age and again at four months adjusted age.For EEG data from full-term infants, we will examine the quality of sleep by classifying sleep stages. Additionally, we will build an algorithm to identify and count the number of sleep spindles automatically.For EEG data from premature infants, we will examine the quality of sleep by classifying sleep stages, investigate the brain activity during sleep and compare the brain activity between healthy and premature infants.

Supervisior: 
Dr. Catherine Mooney
Email: 
lan.wei@ucdconnect.ie