Method for Classifying Electroencephalography Data with the Use of Artificial Neural Networks

Simple SummaryContent extracted from patent full text and abstract with AI.

This invention presents a system and method that uses artificial neural networks (ANNs) to automatically classify electroencephalography (EEG) data obtained from subjects (typically patients) who are exposed to specific sequences of auditory stimuli. The central application described is predicting whether a comatose patient will awaken, using a single EEG recording session combined with standardized sound stimuli. The EEG signals are pre-processed and analyzed by a deep learning architecture, which outputs an objective classification result (e.g., survivor vs. non-survivor). The method streamlines and automates interpretation of brain activity responses to stimuli, focusing especially on challenging prognostic situations like post-cardiac arrest coma.

Use CasesContent extracted from patent full text and abstract with AI.

• Predicting awakening outcomes for comatose patients in intensive care units using bedside EEG and sound stimulation.
• Automated and objective assessment of neurological prognosis after cardiac arrest.
• Supporting clinicians in making decisions about patient care and resource allocation in coma cases.
• Potential diagnosis and classification of various sleep-wake disorders based on EEG responses to auditory stimuli.
• Research into other brain conditions by classifying EEG responses to tailored stimulation protocols.
• Automation in clinical EEG analysis for faster, more reproducible results.

BenefitsContent extracted from patent full text and abstract with AI.

• Provides a fast, objective, and reproducible method for predicting coma outcomes, reducing subjectivity from human interpretation.
• Requires only a single EEG recording, minimizing patient handling and streamlining workflow compared to multi-day assessments.
• Adds new prognostic information by incorporating standardized auditory stimulation, which is not routine in current clinical practices.
• Enables fully automated processing and classification using trained neural networks, which reduces labor and expertise required.
• Shows high predictive performance, as demonstrated in clinical test groups, improving confidence in patient prognoses.
• Flexible architecture could be adapted to other neurological conditions or stimulation paradigms.
• Reduces the number of patients with indeterminate prognosis, improving care planning and communication with families.

Technical Classifications (CPCs)

Inventors & Applicants

Patent Abstract

A sensor system is proposed in one example for analysing electroencephalography sensor signals obtained by a plurality of sensors connected to an object to be stimulated with a sequence of stimulation signals. The sensor system comprises means for: acquiring a set of sensor signals in response to stimulating a plurality of objects with a plurality of sequences of stimulation signals comprising a set of standard stimulation signals and a set of deviant stimulation signals; pre-processing the acquired set of sensor signals to obtain a set of pre-processed sensor signals; preparing an artificial neural network for data analysis by using a first set of the pre-processed sensor signals, the preparation comprising the steps of training and validating the network by using the first set of the pre-processed sensor signals; feeding a second set of the pre-processed sensor signals to the network as trained and validated; and combining output signals of the network as trained and validated in response to feeding the second set of the pre-processed sensor signals to the network to obtain a single classification result for a respective object and for a respective sequence of stimulation signals.