EEG Stacked Ensemble Stroke Detection for Rapid LVO Triage

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing technologies face challenges in rapidly and accurately identifying large vessel occlusion (LVO) strokes in the pre-hospital setting due to the impracticality of dense electrode arrays and the need for complex, time-consuming data analysis, which delays timely interventions.

Innovation Solution

A portable EEG-based stroke detection device using a stacked ensemble classification model (ESCM) that processes EEG data from a dry electrode headset to automatically detect LVO strokes with high accuracy, enabling rapid triage and routing to appropriate healthcare facilities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If dense electrode arrays are used for stroke detection, then measurement precision is improved, but device complexity and ease of operation deteriorate

Engineering Contradiction:
Improvestroke detection accuracyVSAvoidelectrode array complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts and removes unnecessary electrodes from the dense electrode array, retaining only a simplified subset of electrodes that are sufficient for accurate stroke detection. This extraction approach maintains measurement precision while significantly reducing device complexity and ease of operation.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent employs a simplified, disposable EEG cap with minimal electrodes rather than a complex, reusable dense electrode array. This approach trades some long-term durability for immediate simplicity and ease of use, while maintaining sufficient detection accuracy for the intended application.

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

2Measurement precision

If complex data analysis methods are used, then measurement precision is improved, but loss of time increases

Engineering Contradiction:
Improvestroke detection accuracyVSAvoiddata analysis time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary feature extraction and data preprocessing during the data acquisition phase, organizing raw EEG signals into standardized formats with key features already identified. This preliminary action reduces the computational burden during analysis, maintaining high accuracy while significantly reducing the time required for the actual detection process.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements a two-stage analysis approach where a rapid initial assessment filters obvious cases, and only ambiguous cases proceed to full complex analysis. This partial application of complex methods reduces overall analysis time while maintaining precision for cases that require it.

Inventive Principle:
Principle #16Partial or excessive action

3Productivity

If rapid stroke detection is implemented, then productivity is improved, but measurement precision deteriorates

Engineering Contradiction:
Improvedetection speedVSAvoidstroke detection accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent segments the detection process into multiple specialized stages: rapid signal acquisition, preliminary feature extraction, initial classification, and confirmatory analysis. Each segment is optimized for its specific function, with the majority of cases resolved in the faster initial segments while maintaining the option for deeper analysis when needed, thus achieving both speed and accuracy.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4460238B1Stroke prediction multi-architecture stacked ensemble supermodel
Publication Date: 2025.12.17 ASTERION AI INC
  • EP4460238B1 patent drawingFigure 1A
  • EP4460238B1 patent drawingFigure 1B
  • EP4460238B1 patent drawingFigure 1C

AI summary

Apparatus and associated methods relate to emergency stroke detection and classification. In an illustrative example, a stroke detection device may include an ensemble stroke classification model (ESCM). The ESCM may, for example, include class-specific model sets applicable for at least four classes of features, and a general model set applicable for all classes of features. Each model set, for example, may be stacked with multiple class-specific models for each of a corresponding group of architectures. The stroke detection device may, for example, extract predetermined features from a rolling window of a first predetermined duration of EEG data. The predetermined features are extracted and combined into a 1-D input vector. By applying the input vector, the stroke detection device may generate a binary stroke prediction result. Various embodiments may advantageously accurately predict whether a patient is experiencing a stroke within a finite time to assist an emergency service personnel.