View Proposal #299

If this proposal belongs to you, you are authorized to update it. Use the menu on the right.

ID299
First NameJohn
Last NameBerkowitz
InstitutionCoe College
Speaker Categoryundergraduate student
Title of TalkAn Algorithm for the Detection of Transient Neural Oscillations
AbstractThe analysis of neural activity through measures of electrical potential affords researchers great opportunity to understand in detail the dynamic nature of certain brain processes. Electroencephalography (EEG) and Local Field Potential recording (LFP) are two of the most common methods for measuring this activity. Both are essentially recordings of the electrical potential over time in a highly localized segment of the brain, and produce data sets that can show surprising amounts of structure. Oscillations with very well defined frequencies are the most common examples of structure within these recordings, and a large portion of modern neuroscience research focuses on how different frequency bands of these oscillations relate to different modes of activity for the organism being studied. Such signals can easily be detected and quantified automatically with traditional signal processing tools such as the Fourier transform. However, there also exist very transient oscillations within such recordings that are of interest to researchers. These require more sophisticated techniques to detect, because of their dual localization in both the time and frequency domains. A combination of several classic signal processing tools, namely digital band-pass filters, the Hilbert transform, z-scoring, and numerical derivatives has yielded an efficient and accurate algorithm for the detection of such transient oscillations. This algorithm has been applied to LFP data for sleeping rats and used to detect the well-known phenomena of sleep spindles, which are a hallmark of late stage sleep in mammals.
Subject area(s)
Suitable for undergraduates?Yes
Day Preference
Computer Needed?
Bringing a laptop?
Overhead Needed?
Software requests
Special Needs
Date Submitted10/17/2010
Year2010