US 5,732,696 AGranted

Polysomnograph scoring

AssigneeNew York University

InventorsDavid M. Rapoport, Robert G. Norman

Life of this patent

Estimating the term…

Loading patent…

Abstract

A method and apparatus for scoring EEG, EMG and EOG data, for sleep stages, in order to simulate the scoring results achieved by human scorers in which a general determination is initially made of the presence or absence of a plurality of elemental events. This determination is employed to develop the probabilities of the occurrence of the elemental events. These probabilities are then employed as a base to determine the probabilities of the occurrence of Rechtschaffen and Kales events. The Rechtschaffen and Kales event probabilities may be modified to account for the stage 2 rule and the REM rule. These Rechtschaffen and Kales event probabilities are then employed to determine the probability distribution of possible sleep stages for each epoch of the polysomnograph and choose the most likely score.

Claims7

Dependent claims

A method for computer analysis of polysomnograph waveforms to score epochs of such polysomnograph waveforms, said score constituting one of a plurality of sleep stages using standard Rechtschaffen and Kales (R&K) sleep stage scoring rules, comprising the steps:

(a) computer separating said polysomnograph signals into a plurality of sequential epochs, each epoch possibly having associated with it elemental events which can be used to associate with R&K events, said R&K events being usable to determine a score for each of the epochs, said elemental events comprising a member selected from the group consisting of:
A1:
run of alpha (.alpha.) waves
A2:
sleep spindle
A3:
delta (.delta.) wave
A4:
K complex
A5:
low EMG
A6:
very low EMG
A7:
high EMG
A8:
transient increase in EMG
A9:
in-phase eye movement
A10:
out-of-phase eye movement
A11:
slow rolling eye movement,
said R&K events comprising a member selected from the group consisting of:
C1:
alpha (.alpha.) waves <25% of epoch
C2:
alpha waves in 25-50% of epoch
C3:
alpha waves in >50% of epoch
C4:
single sleep spindle in epoch
C5:
multiple sleep spindles in epoch
C6:
delta (.delta.) waves in <20% of epoch
C7:
delta waves in 20-50% of epoch
C8:
delta waves in >50% of epoch
C9:
k complexes in epoch
C10:
low EMG for most of epoch
C11:
very low EMG for most of epoch
C12:
high EMG for most of epoch
C13:
transient increase in EMG during epoch
C14:
in-phase eye movements
C15:
out-of-phase eye movements (REMS)
C16:
slow rolling eye movements
C17:
5 second arousal
C18:
significant artifacts,
(b) computer determining whether there is a possibility of the existence of each of the elemental events in the polysomnograph waveforms of each of said epochs, the elemental events which have been determined to be possibly present constituting elemental event candidates having a certain probability of being present,
(c) then computer determining the probabilities of the existence of each of said elemental event candidates in each epoch of the polysomnograph waveforms,
(d) then, computer employing said probabilities determined in step (c), computer producing first signals representing the probability of an R&K event in each epoch of the polysomnograph waveforms,
(e) then, computer producing from the first signals a second signal representing a probability distribution of sleep stages based on said probability of an R&K event, said second signal determining the score constituting a standard R&K sleep stage, wherein said standard R&K sleep stage is selected from the group consisting of:
Non-REM (NREM) Stage 1
Non-REM (NREM) Stage 2
Non-REM (NREM) Stage 3
Non-REM (UREM) Stage 4
Rapid Eye Movement (REM) Stage
Awake
Movement Time
Artifact,
(f) outputting the score determined in step (e).

70 more elements

Apparatus for computer analysis of polysomnograph waveforms to score epochs of such polysomnograph waveforms, said score constituting one of a plurality of sleep stages using standard Rechtschaffen and Kales (R&K) sleep stage scoring rules, comprising:

(a) computer means for separating said polysomnograph signals into a plurality of sequential epochs, each epoch possibly having associated with it elemental events which can be used to associate with R&K events, said R&K events being usable to determine a score for each of the epochs, said elemental events comprising a member selected from the group consisting of:
A1:
run of alpha (.alpha.) waves
A2:
sleep spindle
A3:
delta (.delta.) wave
A4:
K complex
A5:
low EMG
A6:
very low EMG
A7:
high EMG
A8:
transient increase in EMG
A9:
in-phase eye movement
A10:
out-of-phase eye movement
A11:
slow rolling eye movement,
said R&K events comprising a member selected from the group consisting of:
C1:
alpha (.alpha.) waves <25% of epoch
C2:
alpha waves in 25-50% of epoch
C3:
alpha waves in >50% of epoch
C4:
single sleep spindle in epoch
C5:
multiple sleep spindles in epoch
C6:
delta (.delta.) waves in <20% of epoch
C7:
delta waves in 20-50% of epoch
C8:
delta waves in >50% of epoch
C9:
k complexes in epoch
C10:
low EMG for most of epoch
C11:
very low EMG for most of epoch
C12:
high EMG for most of epoch
C13:
transient increase in EMG during epoch
C14:
in-phase eye movements
C15:
out-of-phase eye movements (REMS)
C16:
slow rolling eye movements
C17:
5 second arousal
C18:
significant artifacts,
(b) computer means for determining whether there is a possibility of the existence of each of the elemental events in the polysomnograph waveforms of each of said epochs, the elemental events which have been determined to be possibly present constituting elemental event candidates having a certain probability of being present,
(c) computer means for determining the probabilities of the existence of each of said elemental event candidates in each of said epochs of the polysomnograph waveforms,
(d) computer means for employing said probabilities determined in step (c) and for producing first signals representing the probability of an R&K event in each of said epochs of the polysomnograph waveforms,
(e) computer means for producing from the first signals a second signal representing a probability distribution of sleep stages based on said probability of an R&K event, said second signal determining the score constituting a standard R&K sleep stage, wherein said standard R&K sleep stage is selected from the group consisting of:
Non-REM (NREM) Stage 1
Non-REM (NREM) Stage 2
Non-REM (NREM) Stage 3
Non-REM (NREM) Stage 4
Rapid Eye Movement (REM) Stage
Awake
Movement Time
Artifact.

69 more elements

Dependencies are inferred from the claim text.

Facts

Application: 08/271,212Filed 5 Jul 1994 · continuation
Priority: 17 Mar 1992
Granted: 31 Mar 1998
Term ends: Not estimatedThe source records don’t support an estimate.

Classification5 codes

A61B 5/4812
Sleep evaluation: detecting sleep stages or cycles
A61B 5/372
Electroencephalography [EEG]: analysis of electroencephalograms
A61B 5/389
Modalities, i.e. specific diagnostic methods: electromyography [EMG]
A61B 5/398
Modalities, i.e. specific diagnostic methods: electrooculography [EOG], e.g. detecting nystagmus; Electroretinography [ERG]