Automatic recognition of sleep spindles using short-term spectral analysis

Vo-Ngoc, B.; Poussart, D.; Langlois, J. M. Pierre

doi:10.3758/bf03208142

Behav. Res. Meth. & Instru.

1971

DOI: 10.3758/bf03208142

|View full text |Cite

Automatic recognition of sleep spindles using short-term spectral analysis

B. Vo-Ngoc

D. Poussart

J. M. Pierre Langlois

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...

Citation Types

Supporting

Mentioning

Contrasting

Year Published

1973

1985

Publication Types

Select...

Article3

Relationship

Self Cite2

Independent1

Authors

Journals

Cited by 3 publications

(1 citation statement)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From over 700 scatter diagrams observed, we have not detected any evident indication of a clear Analysis An approach to automatically recognize sleep spindles has been described previously (Vo-Ngoc, Poussart, & Langlois, 1971). This detection approach, which basically utilizes spectral analysis techniques, transforms the original electroencephalogram (EEG) into a rectangular wave signalwith two possible distinct states: presence or absence of spindles.…”

mentioning

confidence: 99%

Some stochastic properties of sleep spindles

Vo-Ngoc

Poussart

Langlois

1973

Behav. Res. Meth. & Instru.

Self Cite

View full text Add to dashboard Cite

mentioning

confidence: 99%

Some stochastic properties of sleep spindles

Vo-Ngoc

Poussart

Langlois

1973

Behav. Res. Meth. & Instru.

Self Cite

View full text Add to dashboard Cite

An optimal monitor of the electroencephalographic sigma sleep state

Kemp¹,

Jaspers²,

Franzen³

et al. 1985

Biol. Cybern.

View full text Add to dashboard Cite

show abstract

Correlation of spindle activity recorded from different brain regions

Vo-Ngoc

Poussart

Langlois

1973

Behav. Res. Meth. & Instru.

Self Cite

View full text Add to dashboard Cite

T C"(r) = -L f·(t)· [.(t + r) dt 1JT 0 1 1 fj(t) or fj(t) = 0 ---+ fj(t) • fj(t) = 0 fj(t) and fj(t) =1 ---+ fj(t) . fj(t) =1 RESULTSAn example of correlation functions obtained with the above program is illustrated in Fig. 2. Spindle activity recorded from Channel 0 appears to be in lag with respect to that recorded from Channel 1 by two time units (one time unit =0.16 sec).The global results obtained from one preparation are where fi(t) and fj(t) are the rectangular wave outputs of a spindle detection program for two EEG channels, which has been described elsewhere (Vo-Ngoc, Poussart, & Langlois, 1971).In general, we have chosen r max = 5.12 sec and T = 15 min, so that (r max/T) "i 1. In the ideal case, where T ---+ 00 and where fj(t) and fj(!) are stationary, this expression can be identified with the conventional definition of the cross-correlation function of two stochastic signals.Here, fj(t) and fj(t) can only have the values 0 or 1, corresponding, respectively, to the absence or presence of spindle activity. The product fj(t) . fj(t) becomes a simple AND logic operation. Therefore, the computation time is considerably reduced.The flow diagram of the program which computes the cross-correlation functions of spindle activities, as identified by the detection program and stored on DECtape, is shown in Fig. 1.First, one specifies to the program all required parameters, such as the starting block and the number of blocks of data stored in the DECtape obtained from the spindle detection program, and the maximum delay r max' The program then clears all counter registers and tables of results.For pratical reasons related to the computer memory capacity, the computation of correlation functions is worked out as follows: (1) For each value of r , the data on DECtape is divided into groups of two blocks each. (2) The program transfers two consecutive blocks of data to computer memory, for example, Blocks nand n + 1, computes the partial correlation functions, Cij(r), and stores the results. (3) The data transfer is repeated, this time with Blocks n + 1 and n + 2. The new partial functions, Cij(r), are then added up to the previous Cij(r). (4) The process is repeated automatically. After all specified blocks have been analyzed, the program transfers to display mode.The program is fully interactive. By a suitable command at the Teletype, one can choose the type of correlation function for display, change the scales, observe the results on the CRT display, or also obtain a duplicate of results on graph paper with the point plotter.The main hardware used in this analysis consists of PDP-8 computer, DECtapes, CRT display, and point plotter. The program uses 2.5K of core memory.

show abstract

scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.

Contact Info

hi@scite.ai

10624 S. Eastern Ave., Ste. A-614

Henderson, NV 89052, USA

Blog Terms and Conditions API Terms Privacy Policy Contact Cookie Preferences Do Not Sell or Share My Personal Information

Made with 💙 for researchers

Part of the Research Solutions Family.

Automatic recognition of sleep spindles using short-term spectral analysis

Cited by 3 publications

References 6 publications

Some stochastic properties of sleep spindles

Some stochastic properties of sleep spindles

An optimal monitor of the electroencephalographic sigma sleep state

Correlation of spindle activity recorded from different brain regions

Contact Info

Product

Resources

About