ECG Signal Quality Measurement Using Stability Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current patient monitoring systems face challenges in accurately determining patient parameters due to poor quality physiological signals, particularly in ECG data, which can lead to inaccurate data and potential harm, as existing methods struggle to distinguish between desired signals and noise effectively.

Innovation Solution

A system that measures signal quality using stability and complexity algorithms to automatically select high-quality ECG leads for processing, assigning weights to ensure accurate QRS detection and reduce noise impact, thereby improving the reliability of patient parameter monitoring.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If current ECG processing methods are used to detect and quantify noise, then noise detection capability is provided, but reliable QRS detection cannot be achieved when ECG signal quality is poor due to inability to distinguish between desired signal data and noise

Engineering Contradiction:
Improvenoise detection accuracyVSAvoidQRS detection reliability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent applies preliminary action by performing stability analysis on QRS complex features before attempting QRS detection. The system calculates stability metrics (such as morphological consistency) of QRS complexes in advance, and only proceeds with QRS detection when stability thresholds are met. This preliminary stability assessment ensures that the signal quality is sufficient for reliable detection, preventing false detections in noisy conditions.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces stability analysis as an intermediary step between signal acquisition and QRS detection. This intermediary process evaluates the quality and consistency of the ECG signal by analyzing feature stability (morphological characteristics) before the main detection task. The stability metric acts as a gatekeeper that determines whether the signal is suitable for reliable QRS detection, thereby resolving the contradiction between noise detection and reliable detection.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Adaptability or versatility

If multiple ECG leads are used as inputs to multi-lead algorithm for detecting arrhythmia, then detection capability is enhanced, but performance is degraded when leads with inferior quality are included

Engineering Contradiction:
Improvemulti-lead detection capabilityVSAvoidarrhythmia detection accuracy
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent applies local quality by assigning different quality assessments to different ECG leads based on their individual signal characteristics. Instead of treating all leads uniformly, the system performs stability analysis on each lead separately, evaluating the morphological consistency of QRS complexes in each lead independently. Leads with higher stability (better quality) are given greater weight or preference in the multi-lead algorithm, while leads with lower stability are downweighted or excluded, thereby maintaining high detection accuracy.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent changes the parameter of lead selection from a static binary choice (use or discard) to a dynamic quality-based weighting system. The stability metric serves as a variable parameter that adjusts the contribution of each lead to the overall detection algorithm. This parameter change allows the system to adaptively optimize the use of multiple leads, enhancing versatility while maintaining precision by dynamically adjusting lead weights based on real-time signal quality assessment.

Inventive Principle:
Principle #35Parameter changes

3Object-affected harmful factors

If noise estimation is performed on ECG input signal, then noise presence is detected, but the ECG algorithm may reject part or all of the ECG signal resulting in loss of useful data

Engineering Contradiction:
Improvenoise detectionVSAvoidECG signal data loss
Core Design Contradiction:
Object-affected harmful factorsVSLoss of information

Solution Approach 1:

The patent applies partial action by performing stability analysis selectively on specific features (QRS complex morphology) rather than rejecting the entire signal based on overall noise estimation. The system calculates stability metrics for relevant features and uses these to make informed decisions about signal usability. This partial assessment approach allows the system to retain useful signal components while identifying and mitigating the impact of noise, avoiding excessive rejection of valid ECG data.

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The patent implements feedback by using stability analysis results to inform subsequent processing decisions. The stability metric provides feedback about signal quality that is fed back into the detection algorithm, allowing it to adjust its operation accordingly. When stability is high, the algorithm proceeds with confidence; when stability is low, the algorithm can apply additional filtering, adjust parameters, or flag the segment for review, rather than simply rejecting the data. This feedback mechanism preserves useful information while accounting for noise conditions.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP2688468B1Apparatus and method for measuring physiological signal quality
Publication Date: 2020.12.23 DRAGERWERK AG
  • EP2688468B1 patent drawingFigure 1
  • EP2688468B1 patent drawingFigure 2
  • EP2688468B1 patent drawingFigure 3

AI summary

An apparatus and method for determining a signal quality of an input signal representing a repetitious phenomena derived from at least one sensor connected to a patient is provided. A detector receives the input signal and determines data representing the repetitious phenomena from the input signal for use in determining at least one patient parameter. A measurement processor is electrically coupled to the detector that determines a first signal quality value by identifying at least one feature of the repetitious phenomena data and compares the at least one feature of a first set of the determined repetitious phenomena data with a second set of the determined repetitious phenomena data to determine a feature variability value and using the feature variability value to determine a stability value representative of the quality of the input signal.