Personalized fingerprinting of sleep brain waves provides powerful new tools for understanding disease

An EEG measures brain waves on the scalp, and scientists often use an electroencephalogram (EEG) to study brain activity
during sleep.
The brief oscillation waveforms in the sleep EEG represent short-term coordinated brain activity, and a particularly important set of sleep brain wave events is called sleep spindle waves, which are short oscillation waveforms that usually last less than 1-2 seconds
.
Known to be a short-lived oscillating event associated with memory consolidation (i.
e.
, the conversion of short-term memory into long-term memory), changes in the spindle wave are associated with many neurodegenerative diseases, such as schizophrenia, autism, and Alzheimer's disease, as well as natural aging
.
In fact, since the mid-30s of the 20th century, people have begun to study sleep EEG
by observing brain wave graphics drawn by machines on paper tape.
However, even with the latest machine learning and signal processing algorithms that detect sleep waveforms, many of the important features of sleep EEGs are still based on waveforms that were most easily observed nearly a century ago
.
In this study, the researchers asked the question: What could be found if the concept of sleep brain waves could be extended beyond the scope of what the eye could recognize?

A team of researchers from Massachusetts General Hospital devised a new EEG-type typing method that extracts tens of thousands of electronic events from the brain waves of a sleeping person—the key being that instead of looking at waveforms based on fixed sleep stages (such as wakefulness, REM, and non-REM phases 1-3) as in standard sleep studies, they describe a complete continuous graph of the progressive changes that occur in the brain during sleep—using continuous dynamics to describe a generalized transient time-frequency event over a wide frequency range
。 Information from these waveforms is used to create images of brain activity — the team used all this data to create a graphical representation called a slow oscillation power and phase histogram, which provides a powerful visualization of activity across all waveforms as a function of
continuous sleep depth and cortical synchronization activity.

When the team looked at a group of healthy participants, each with two nights' sleep records, the observed pattern seemed like a fingerprint — each person was highly specific and had strong consistency
from night to night.
These results reveal new ways in which brain activity varies from person to person, even
in healthy people selected as controls.
The researchers then compared activity between healthy subjects and people with schizophrenia, a disease
that reduces spindle activity.
Using their method, the team saw not only known differences in people with schizophrenia, but also differences
in other spindle-like waveforms that appear at other frequencies in the brain.
This suggests that EEG biomarkers for schizophrenia may contribute to a better understanding of the mechanisms of the disease and contribute to the development of targeted treatments
.

The findings were published in the
journal Sleep.
"This approach is really exciting," said
co-author Dr.
Dara Manoach, a co-author in the Department of Psychiatry at Massachusetts General Hospital.
"We look forward to seeing how we advance our understanding of schizophrenia, as well as other neurodevelopmental disorders characterized by sleep differences, such as autism and childhood epilepsy
.
" Senior author Dr Michael Prerau said: "We are only beginning to understand the range of
neurodiversity that exists in the general population.

。
。 If we can more accurately characterize the individual differences observed in neurohealth and disease, we can work on improving diagnosis and treatment
.

。
。 This study expands the way
we look at brain activity during sleep.

。
。 By moving beyond the traditional notion of dividing the complex sleep continuum into specific categories and waveform categories, we can uncover new signal types and dynamics that may be important
for understanding brain health and disease.
”