Dynamic Cardiovascular Waveform Template Adaptation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing algorithms for filtering and evaluating cardiovascular waveforms often incorrectly classify atypical electrical activity as noise or artifacts, missing true cardiovascular abnormalities due to mismatch with existing templates.

Innovation Solution

A method and system for template-based analysis and classification of cardiovascular waveforms that updates templates with newly-identified features, allowing for better identification and reclassification of periodic components in both electrical and hemodynamic waveforms, ensuring accurate classification of cardiovascular abnormalities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If existing template-based algorithms are used to classify cardiovascular waveforms, then classification speed is maintained, but classification accuracy deteriorates due to mismatch with abnormal waveforms

Engineering Contradiction:
Improvewaveform classification accuracyVSAvoidtemplate adaptability to abnormal waveforms
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent implements dynamic template adaptation where the template library is continuously updated with newly identified abnormal waveform patterns. The system transitions from static templates to dynamic templates that evolve with new discoveries, allowing the classification algorithm to adapt to previously unrecognized abnormal patterns while maintaining classification performance

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system performs self-learning by automatically identifying new abnormal waveform patterns from clinical data and incorporating them into the template library without requiring manual template creation. The algorithm autonomously expands its knowledge base by detecting patterns it previously couldn't classify, enabling continuous improvement of classification accuracy

Inventive Principle:
Principle #25Self-service

2Reliability

If traditional filtering algorithms are applied to cardiovascular waveforms, then noise reduction is achieved, but true abnormalities are lost due to over-filtering

Engineering Contradiction:
Improveabnormality detection reliabilityVSAvoidfalse noise classification
Core Design Contradiction:
ReliabilityVSObject-generated harmful factors

Solution Approach 1:

The system uses feedback from hemodynamic waveform analysis to validate or correct ECG-based classifications. When the hemodynamic waveform shows abnormalities that contradict the ECG classification, the system investigates further and may reclassify the signal, preventing false noise classification and improving abnormality detection reliability

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent introduces hemodynamic waveform analysis as an intermediary validation layer between ECG signal processing and final classification. This intermediary step cross-checks ECG-based classifications against independent hemodynamic measurements, reducing false positives caused by over-filtering or artifact misclassification

Inventive Principle:
Principle #24Intermediary (Mediator)

3Measurement precision

If manual waveform analysis is performed by physicians, then detection accuracy improves, but analysis time increases significantly

Engineering Contradiction:
Improveabnormality detection accuracyVSAvoidwaveform analysis time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs automated self-analysis of cardiovascular waveforms using machine learning algorithms that continuously learn from clinical data. The algorithm autonomously identifies, classifies, and flags abnormal patterns without requiring manual physician review for every waveform, dramatically reducing analysis time while maintaining high detection accuracy through continuous learning

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP3474746B1Template-based analysis and classification of cardiovascular waveforms
Publication Date: 2024.09.04 KONINKLIJKE PHILIPS NV
  • EP3474746B1 patent drawingFigure 1
  • EP3474746B1 patent drawingFigure 2
  • EP3474746B1 patent drawingFigure 3

AI summary

In various embodiments, a first classification assigned to a periodic component of an electrical waveform that represents electrical activity in a patient's heart may be identified (302). A corresponding periodic component of a hemodynamic waveform that represents hemodynamic activity in the patient's cardiovascular system may be analyzed (306, 318, 328). The corresponding periodic component may be causally related to the periodic component of the electrical waveform. Based on the analysis, the previously-assigned classification may be assigned (312, 324) to the corresponding periodic component of the hemodynamic waveform in response to a determination, based on the analyzing, that the previously-assigned classification also applies to the corresponding periodic component. In a database (130) of hemodynamic templates, a hemodynamic template associated with the previously-assigned classification may be updated (314) to include one or more features of the corresponding periodic component of the hemodynamic waveform.