event

Ph.D. Proposal Oral Exam - Mark McCurry

Primary tabs

Title:  Automatic Spectral-temporal Modality based EEG Sleep Staging

Committee: 

Dr. Clements, Advisor  

Dr. Barnes, Chair

Dr. Romberg

Abstract: The objective of the proposed research is to investigate a method for sleep state classification using a modality based sparse denoising approach. This method is designed to identify and enhance global structure in dense spectrograms from single channel EEG recordings by viewing the spectrum as a collection of temporal-spectral regions with a roughly block-constant representation organized in an irregular grid. This signal model permits recovering this global structure in the presence of frequent artifacts and high levels of noise. Unlike prior work, this method does not impose a model of fixed frequency bands within the analysis, but it adapts to different global structures in each subject's recording. The goal of this investigation is to further investigate the relative performance of this approach under increased noise and across different datasets when compared to previously published methods.

Status

  • Workflow Status:Published
  • Created By:Daniela Staiculescu
  • Created:04/01/2016
  • Modified By:Fletcher Moore
  • Modified:10/07/2016

Categories

Target Audience