PPG Scatter Diagram Classification via CNN for Heart Rhythm Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional ECG signal acquisition methods are not suitable for long-time monitoring, requiring at least half an hour for accurate analysis of heart rhythm characteristics, and existing PPG signal processing methods do not efficiently generate scatter diagrams for heart rhythm analysis.

Innovation Solution

A method and device that extracts inter-beat interval data from PPG signals to generate scatter diagrams, utilizing a convolutional neural network and fully connected neural network for classification, reducing scatter diagram generation difficulty and enriching health monitoring applications.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional ECG signal acquisition method is used, then accurate analysis of heart rhythm characteristics can be achieved, but long-time continuous acquisition (at least half an hour) is required which is not suitable for monitoring at any time

Engineering Contradiction:
Improveaccuracy of heart rhythm analysisVSAvoidacquisition time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces the mechanical ECG signal acquisition system with a PPG (photoplethysmography) signal acquisition system that uses optical means to detect blood volume changes. This substitution enables shorter acquisition times while maintaining the capability to analyze heart rhythm characteristics through scatter diagram generation from PPG inter-beat intervals.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If the number of scatter points is increased to improve analysis accuracy, then more obvious typical image and accurate analysis result are obtained, but long-time continuous acquisition is required

Engineering Contradiction:
Improveaccuracy of scatter diagram analysisVSAvoidacquisition duration
Core Design Contradiction:
Measurement precisionVSDuration of action of moving object

Solution Approach 1:

The patent changes the fundamental parameter of signal acquisition from ECG to PPG, which has different physiological characteristics. PPG signals can provide sufficient inter-beat interval data for scatter diagram generation in shorter durations, thereby reducing the required acquisition time while maintaining analysis accuracy.

Inventive Principle:
Principle #35Parameter changes

3Ease of operation

If PPG signal is used for scatter diagram generation, then non-invasive long-time monitoring is enabled, but efficient generation of scatter diagrams for heart rhythm analysis needs to be achieved

Engineering Contradiction:
Improvenon-invasive monitoring capabilityVSAvoidscatter diagram generation efficiency
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The patent segments the PPG signal processing into distinct stages: signal acquisition, inter-beat interval extraction, scatter point coordinate generation, and scatter diagram construction. This segmentation enables efficient processing and generation of scatter diagrams from PPG signals, improving productivity while maintaining non-invasive monitoring capabilities.

Inventive Principle:
Principle #1Segmentation

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Enables efficient and accurate classification of heart rhythm patterns from PPG signals, facilitating long-time monitoring and improving the analysis of heart rate variability and arrhythmia diagnosis.

Implementation Method 1

Photoplethysmography (PPG) is a non-invasive method to detect the change of blood volume in viable tissue by photoelectric means

Methodology Applied
Scientific EffectPhotoelectric Effect: Photoelectric Effect

Data Source

PatentEP4227852B1Scatter diagram classification method and apparatus for photoplethysmography signal
Publication Date: 2025.08.13 LEPU MEDICAL TECH (BEIJING) CO LTD
  • EP4227852B1 patent drawingFigure 1
  • EP4227852B1 patent drawingFigure 2
  • EP4227852B1 patent drawingFigure 3

AI summary

The embodiments of the present invention relate to a scatter diagram classification method and apparatus for a photoplethysmography (PPG) signal. The method comprises: acquiring a PPG signal; performing signal sampling processing on the PPG signal so as to generate a sampled PPG signal; according to the sampled PPG signal, a sampling frequency and the maximum value of an inter-beat interval, generating a corresponding scatter diagram two-dimensional tensor, using an artificial intelligence network to perform convolutional and pooling calculation, fully connected calculation and normalized exponential calculation on the scatter diagram two-dimensional tensor, so as to obtain a normalized two-dimensional tensor; and performing classification confirmation processing according to the normalized two-dimensional tensor, so as to generate confirmation data, wherein the confirmation data comprises information indicative of rational classification and information indicative of irrational classification. By means of the embodiments of the present invention, the difficulty of generating a scatter diagram is reduced, and the application scenarios of photoplethysmography in the field of health monitoring are diversified.