System and methods for consciousness evaluation in non-communicating subjects
Inventors
RAIMONDO, Federico • SITT, Jacobo D • Naccache, Lionel • FERNANDEZ SLEZAK, Diego
Assignees
Icm Institut Du Cerveau Et de la Moelle Epiniere • Centre National de la Recherche Scientifique CNRS • Assistance Publique Hopitaux de Paris APHP • Institut National de la Sante et de la Recherche Medicale INSERM • Sorbonne Universite • Institut du Cerveau et de La Moelle Epiniere ICM
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-11559244-B2
Publication Date
2023-01-24
Expiration Date
Abstract
Disclosed is a method for the generation of a consciousness indicator for a non-communicating subject, including the steps of generating an auditory stimulation, receiving an electrocardiographic signal of the subject obtained from a recording during the generation of the auditory stimulation, extracting at least one feature from the electrocardiographic signal and deducing a consciousness indicator from an analysis of the electrocardiographic feature.
Core Innovation
The invention relates to a computer-implemented method for the generation of a consciousness indicator for a non-communicating subject, for evaluation of a state of consciousness. The method generates an auditory stimulation with multiple auditory trials having a predefined intertrial interval, each auditory trial including N consecutive auditory stimuli with a predefined time duration and a predefined gap between auditory stimuli onsets, with N equal or superior to 2.
The auditory stimulation comprises a first percentage of local standard trials and a second percentage of locally deviant trials. Local standard trials include N identical auditory stimuli, while locally deviant trials include the first N−1 identical auditory stimuli and the Nth auditory stimulus different from the preceding N−1 auditory stimuli. The stimulation is adapted to be heard by the non-communicating subject so as to prompt cognitive processes.
The method receives an electrocardiographic signal and an electroencephalographic signal obtained during the generation of the auditory stimulation. It extracts EKG features representative of the occurrence of cognitive processes by selecting at least one time interval involving the onset of the Nth stimulus and a following R peak, or involving an R peak preceding the Nth stimulus onset. It also extracts EEG features representative of the occurrence of cognitive processes.
The at least one EEG feature and the at least one EKG feature are used as input to a classifier that has been previously trained using a machine learning technique, to generate the consciousness indicator.
Claims Coverage
The independent claims are clm-00001, clm-00005, and clm-00006. Across these independent claims, the inventive features center on generating a structured local-global auditory stimulation, extracting cognition-representative EKG and EEG features, and producing a consciousness indicator via a previously trained classifier.
Structured auditory stimulation with local standard and local deviant trials
Generating an auditory stimulation comprising multiple auditory trials with a predefined intertrial interval, each auditory trial formed by N consecutive auditory stimuli with a predefined time duration and a predefined gap between auditory stimuli onsets, wherein the auditory stimulation comprises a first percentage of local standard trials comprising N identical auditory stimuli and a second percentage of locally deviant trials comprising the first N−1 identical auditory stimuli and the Nth auditory stimulus different from the preceding N−1 auditory stimuli, wherein N is equal or superior to 2, and adapting the auditory stimulation to be heard by the non-communicating subject so as to prompt cognitive processes.
Receiving electrocardiographic and electroencephalographic signals during stimulation
Receiving an electrocardiographic signal and an electroencephalographic signal of the subject obtained during the generation of the auditory stimulation.
EKG feature extraction using timing relative to Nth stimulus and R peak
Extracting from the electrocardiographic signal at least one electrocardiographic EKG feature representative of the occurrence of cognitive processes, wherein at least one of the at least one electrocardiographic EKG feature is selected among a time interval between the onset of the Nth stimulus of one local standard trial and a following R peak; a time interval between the onset of the Nth stimulus of one locally deviant trial and the following R peak; a time interval between an R peak preceding the Nth stimulus of one local standard trial and the onset of the Nth stimulus of one local standard trial; and a time interval between an R peak preceding the Nth stimulus of one locally deviant trial and the onset of the Nth stimulus of said one locally deviant trial.
EEG feature extraction representative of cognitive processes
Extracting from the electroencephalographic signal at least one electroencephalographic EEG feature representative of the occurrence of cognitive processes.
Classifier-based generation of a consciousness indicator using previously trained machine learning
Using the at least one EEG feature and said at least one EKG feature as input to a classifier to generate said consciousness indicator, the classifier having been previously trained using a machine learning technique.
Together, the independent claims define structured auditory stimulation with local standard and locally deviant trials for prompting cognitive processes, acquisition of electrocardiographic and electroencephalographic signals during stimulation, extraction of EKG timing-based features and EEG cognition-representative features, and generation of a consciousness indicator by inputting those features into a previously trained classifier.
Stated Advantages
Documented Applications
No documented applications found
Interested in licensing this patent?