ECG Q and J Point Detection Using Waveform Length Transform

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional methods for measuring the ST-segment level in electrocardiograms face challenges in reliably locating the Iso-electric and J points, particularly in real-time and ambulatory monitoring, due to high computational load, sensitivity to QRS morphology changes, and incorrect noise estimation, leading to false alarms and missed events.

Innovation Solution

A system and method that determine the Q and J points of an electrocardiogram using a processor to receive beat-cycle waveforms from multiple leads, assess signal quality, and combine waveforms with good quality using a Waveform Length Transform (WLT) to identify these points, ensuring robust lead selection and noise tolerance for accurate ST-segment level measurements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If manual determination of fixed time intervals u and z is used for each patient, then the approach introduces extra work load to clinicians, but the fixed intervals do not adapt to changes of QRS morphology causing measurement errors

Engineering Contradiction:
Improvework loadVSAvoidST measurement accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The patent implements dynamic adaptation of time intervals u and z based on detected QRS morphology characteristics. The system automatically adjusts these intervals for each beat rather than using fixed manual values, allowing the measurement parameters to adapt to changing cardiac conditions while reducing clinician workload.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system performs automatic detection and adjustment of time intervals without requiring manual intervention. The algorithm autonomously identifies QRS morphology changes and recalibrates the measurement parameters, enabling the system to self-adjust and eliminating the need for continuous manual configuration by clinicians.

Inventive Principle:
Principle #25Self-service

2Extent of automation

If automatic detection of Iso-electric point and J point is performed from averaged beat-cycle ECG waveform, then the computational load increases, but the approach provides automated searching capability

Engineering Contradiction:
Improveautomatic detectionVSAvoidcomputational load
Core Design Contradiction:
Extent of automationVSUse of energy by moving object

Solution Approach 1:

The patent divides the ECG signal analysis into distinct segments: QRS complex detection, P wave detection, and ST segment analysis. By segmenting the waveform and focusing computational resources on specific regions of interest, the system achieves automatic detection while reducing overall computational load compared to analyzing the entire beat cycle.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary detection of QRS complexes and identifies candidate regions for Iso-electric and J point detection before conducting the full automatic search. This preliminary action narrows down the search space and reduces the computational effort required for the subsequent automatic detection phase.

Inventive Principle:
Principle #10Preliminary action

3Reliability

If noise detection is performed using Karhunen-Loeve transform feature vectors for ST segment and QRS complex, then the amount of calculation is relatively high, but the approach enables noise detection capability

Engineering Contradiction:
Improvenoise detection accuracyVSAvoidcalculation amount
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent extracts only the essential feature vectors from the ECG signal that are most indicative of noise - specifically focusing on ST segment and QRS complex characteristics. By extracting and analyzing only these critical features rather than processing the entire signal, the system achieves reliable noise detection with reduced computational requirements.

Inventive Principle:
Principle #2Taking out (Extraction)

4Object-affected harmful factors

If beat averaging is performed to reduce ECG noise level, then the ECG noise level is reduced, but the approach requires information of neighboring beats which may not be available in real-time monitoring

Engineering Contradiction:
ImproveECG noise levelVSAvoidtime window requirement
Core Design Contradiction:
Object-affected harmful factorsVSLoss of time

Solution Approach 1:

The patent introduces an intermediary approach where instead of requiring full beat averaging over a time window, the system uses real-time noise estimation based on current signal characteristics and applies adaptive filtering. This intermediary method reduces noise effectively while operating with minimal time delay, making it suitable for real-time monitoring where neighboring beats may not be available.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP2744402B1System and method for determining the q and j points of an electrocardiogram
Publication Date: 2020.12.09 KONINKLIJKE PHILIPS NV
  • EP2744402B1 patent drawingFigure 1~2
  • EP2744402B1 patent drawingFigure 3~4
  • EP2744402B1 patent drawingFigure 5~6

AI summary

A system for determining the Q and J points of an electrocardiogram (ECG)combines a WLT-based Q, J detection algorithm with signal quality assessment for lead selection. A Q, J detector (24) receives a beat-cycle waveform for the beat under consideration from each of a plurality (N) of ECG leads, and assesses signal quality for each lead using signal quality assessor (SQA) components 261, 262,... 26N. The leads with "good" signal qualities are employed for a multichannel waveform length transform (WLT), which yields a combined waveform length signal (CWLS). The Q and J points are then determined from the CWLS.