EEG Consciousness Assessment Using Multi-Band AI Anesthesia Tracking

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional methods for measuring the depth of anesthesia, such as the bispectral index (BIS) analysis, face challenges in accurately and quickly detecting rapid changes in anesthesia status due to algorithmic limitations and slow tracking speeds, and lack transparency in their analysis algorithms.

Innovation Solution

A method using a trained artificial neural network to analyze components of specific frequency bands from EEG signals, combined with EMG signals, to determine consciousness levels by calculating probability values and emotional status, allowing for timely and accurate anesthesia depth measurement.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If conventional BIS analysis method is used to measure depth of anesthesia, then the measurement can be performed, but the tracking speed is slow and cannot accurately detect rapid changes in anesthesia status

Engineering Contradiction:
Improvetracking speedVSAvoidaccuracy in measuring depth of anesthesia
Core Design Contradiction:
SpeedVSMeasurement precision

Solution Approach 1:

The patent changes the analysis parameters by using multiple frequency bands (delta, theta, alpha, beta, gamma) instead of the conventional single BIS index. This allows the system to capture different aspects of brain activity simultaneously, improving both tracking speed and measurement precision by monitoring multiple parameters in parallel rather than relying on a single composite index that requires complex algorithms.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If conventional BIS analysis algorithm is used, then depth of anesthesia can be measured, but the algorithm details are not disclosed and accuracy cannot be proven

Engineering Contradiction:
Improveaccuracy in measuring depth of anesthesiaVSAvoidalgorithm transparency
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The patent segments the EEG signal into multiple frequency bands (delta, theta, alpha, beta, gamma) and analyzes each band separately. This segmentation makes the analysis process transparent and interpretable, as each frequency band corresponds to specific brain states. The method discloses clear calculation procedures for each band's power spectral density and ratio, eliminating the black-box nature of conventional BIS algorithms while improving measurement accuracy through multi-band analysis.

Inventive Principle:
Principle #1Segmentation

3Loss of time

If conventional BIS analysis method is used, then depth of anesthesia measurement is available, but response time is delayed when sudden changes occur

Engineering Contradiction:
Improveresponse timeVSAvoidaccuracy in detecting sudden changes
Core Design Contradiction:
Loss of timeVSMeasurement precision

Solution Approach 1:

The patent calculates power spectral density for multiple frequency bands simultaneously and pre-establishes reference values and ratios for each band. This preliminary preparation of multiple parameters allows the system to rapidly detect sudden changes in anesthesia status by comparing real-time measurements against pre-computed references, reducing response time while maintaining high measurement precision through multi-parameter validation.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3915478B1Consciousness level determination method and computer program
Publication Date: 2026.01.28 BRAINU CO LTD
  • EP3915478B1 patent drawingFigure 1
  • EP3915478B1 patent drawingFigure 2~3
  • EP3915478B1 patent drawingFigure 4

AI summary

A method for determining a consciousness level of a patient according to an embodiment of the present disclosure may include: a step of extracting components of one or more frequency bands from a first section of an EEG; a step of calculating a first index for each of the components of one or more frequency bands, wherein the first index is calculated based on a degree to which a magnitude of each of the components of one or more frequency bands with respect to a magnitude of a predetermined reference component in the first section exceeds a predetermined threshold value; a step of calculating a probability value for each of one or more patient statuses from the first index for each of the components of one or more frequency bands using a trained artificial neural network; and a step of determining the consciousness level of the patient based on the probability value for each of the one or more calculated patient statuses.