Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2015:2015:594-7.
doi: 10.1109/EMBC.2015.7318432.

Pattern recognition with adaptive-thresholds for sleep spindle in high density EEG signals

Jessica Gemignani, Jacopo Agrimi, Enrico Cheli, Angelo Gemignani, Marco Laurino, Paolo Allegrini, Alberto Landi, Danilo Menicucci

Pattern recognition with adaptive-thresholds for sleep spindle in high density EEG signals

Jessica Gemignani et al. Annu Int Conf IEEE Eng Med Biol Soc. 2015.

Abstract

Medicine and Surgery, University of Pisa, via Savi 10, 56126, Pisa, Italy Sleep spindles are electroencephalographic oscillations peculiar of non-REM sleep, related to neuronal mechanisms underlying sleep restoration and learning consolidation. Based on their very singular morphology, sleep spindles can be visually recognized and detected, even though this approach can lead to significant mis-detections. For this reason, many efforts have been put in developing a reliable algorithm for spindle automatic detection, and a number of methods, based on different techniques, have been tested via visual validation. This work aims at improving current pattern recognition procedures for sleep spindles detection by taking into account their physiological sources of variability. We provide a method as a synthesis of the current state of art that, improving dynamic threshold adaptation, is able to follow modification of spindle characteristics as a function of sleep depth and inter-subjects variability. The algorithm has been applied to physiological data recorded by a high density EEG in order to perform a validation based on visual inspection and on evaluation of expected results from normal night sleep in healthy subjects.

PubMed Disclaimer

Figures

Fig. 1
Fig. 1
Flow diagram of the applied procedure.
Fig. 2
Fig. 2
Examples of spindle detection based on dynamic thresholds. Each panel has two traces: (top) The EEG signal, filtered in [0.5 – 40Hz] as usually seen during visual scoring - (bottom) EEG signal, filtered in [12 – 15Hz]. In red, the rectified signal, calculated via Hilbert Transform. In green, lower and upper thresholds. Panel A shows two full-fledged spindles. The first one is following a K-complex, the second one is an temporally isolated one. Both spindles would be recognized by visual scoring of any sleep expert. Panel B show a spindle warped by a slower oscillation and difficult to be identified during visual scoring.
Fig. 3
Fig. 3
Results - Group-averaged topological maps of spindle features.
See this image and copyright information in PMC

References

    1. Wolpert EA. A manual of standardized terminology, techniques and scoring system for sleep stages of human subjects. Archives of General Psychiatry. 1969;20(2):246. - PubMed
    1. Menicucci D, Piarulli A, Allegrini P, Laurino M, Mastorci F, Sebastiani L, Bedini R, Gemignani A. Fragments of wake-like activity frame down-states of sleep slow oscillations in humans: New vistas for studying homeostatic processes during sleep. International Journal of Psychophysiology. 2013;89(2):151–157. - PubMed
    1. Steriade M. Grouping of brain rhythms in corticothalamic systems. Neuroscience. 2006;137(4):1087–1106. - PubMed
    1. Nir Y, Staba RJ, Andrillon T, Vyazovskiy VV, Cirelli C, Fried I, Tononi G. Regional slow waves and spindles in human sleep. Neuron. 2011;70(1):153–169. - PMC - PubMed
    1. De Gennaro L, Ferrara M. Sleep spindles: An overview. Sleep medicine reviews. 2003;7(5):423–440. - PubMed

Publication types

LinkOut - more resources