Stories
Slash Boxes
Comments

SoylentNews is people

posted by Fnord666 on Tuesday February 14 2017, @05:09AM   Printer-friendly
from the just-sleep-it-off dept.

Massachusetts General Hospital (MGH) investigators have developed a novel approach to analyze brainwaves during sleep, which promises to give a more detailed and accurate depiction of neurophysiological changes than provided by a traditional sleep study. In a report published in the January issue of Physiology, the research team describes how applying a technique called multitaper spectral analysis to electroencephalogram (EEG) data provides objective, high-resolution depictions of brainwave activity during sleep that are more informative and easier to characterize than previous approaches. The researchers also present a visual atlas of brain activity during sleep in healthy individuals, highlighting new features of the sleep EEG — including a predictor of REM sleep — that could be of important use to clinicians and researchers.

[...] The approach used by the MGH investigators provides a paradigm shift allowing clinicians to move away from subjective sleep staging and harness the wealth of objective information contained within EEG data. In their report, the team describes how sleep oscillations are far more easily characterized using spectral estimation than by looking at EEG traces. Spectral analysis is a class of approaches that break a waveform signal into its component oscillations — repeating patterns over time — just as a prism breaks white light into its component colors. In the EEG, these oscillations represent the activity of specific brain networks during sleep and wakefulness.

"At a fundamental level, brain activity is truly organized in terms of oscillations and waves," says senior author Patrick Purdon, PhD, MGH Anesthesia. "Spectral analysis is just analyzing the signals in terms of these waves, making it the right tool — and in some ways the perfect tool — for the job." Purdon also points out that traditional sleep scoring is essentially a crude form of spectral analysis, based on recognizing the wave properties by eye.

[...] In their report the researchers use these new vivid images of brain activity to illustrate how the sleep EEG multitaper spectrogram objectively reveals the detailed architecture of an entire night of sleep in a single visualization, rather than 1,000 30-second windows. Repeating patterns of activity — which use color to reflect signal power — become apparent even to the untrained eye, allowing technicians with only a few hours of training to stage with an accuracy comparable to that of traditional sleep scoring.

An abstract of the paper is available.

Michael J. Prerau, Ritchie E. Brown, Matt T. Bianchi, Jeffrey M. Ellenbogen, Patrick L. Purdon. Sleep Neurophysiological Dynamics Through the Lens of Multitaper Spectral Analysis. Physiology, 2016; 32 (1): 60 DOI: 10.1152/physiol.00062.2015


Original Submission

 
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 0) by Anonymous Coward on Tuesday February 14 2017, @07:10AM

    by Anonymous Coward on Tuesday February 14 2017, @07:10AM (#466878)

    Spectral analysis is a class of approaches that break a waveform signal into its component oscillations — repeating patterns over time — just as a prism breaks white light into its component colors.

    You get frequency-space function by applying Fourier Transform to the time-space function - i.e the oscillation.