PPG Signal Quality Assessment Using Segmentation and Random Forest

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Photoplethysmogram (PPG) signals are prone to noise due to motion artefacts, poor blood perfusion, and changes in ambient light, which impede their usability in real-time assessment and monitoring applications, especially in heterogeneous sensor environments like smartphones and pulse oximeters.

Innovation Solution

A system and method for sensor-agnostic PPG signal quality assessment using morphological analysis, which segments PPG signals into samples, performs a signal sufficiency check, and extracts features to classify signals as clean or noisy using Random Forest models, ensuring accurate heart rate detection and noise identification.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If PPG signal is used for real-time monitoring, then vital sign assessment capability is improved, but signal quality deteriorates due to noise and motion artifacts

Engineering Contradiction:
Improvevital sign assessment capabilityVSAvoidnoise and motion artifacts
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The PPG signal is divided into multiple segments, and quality assessment is performed on each segment independently. This allows identification of clean portions within noisy signals and enables selective use of high-quality segments for vital sign monitoring, thereby maintaining reliability while managing noise impact.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs real-time quality assessment of PPG signals and uses this feedback to determine whether to proceed with vital sign extraction. When signal quality is insufficient, the system can trigger alerts or request repositioning, preventing inaccurate measurements and improving overall reliability.

Inventive Principle:
Principle #23Feedback

2Object-affected harmful factors

If signal processing is performed to remove noise, then signal quality is improved, but processing time increases

Engineering Contradiction:
Improvesignal qualityVSAvoidprocessing time
Core Design Contradiction:
Object-affected harmful factorsVSLoss of time

Solution Approach 1:

Quality assessment is performed preliminarily before full signal processing and vital sign extraction. This preliminary check identifies segments that already meet quality criteria, allowing the system to skip extensive processing for acceptable signals while applying enhanced processing only when necessary, thus reducing overall processing time.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system applies different levels of processing based on signal quality. For segments that pass basic quality checks, minimal processing is applied. For segments with moderate quality, selective processing is performed. This partial action approach maintains signal quality while avoiding unnecessary processing time expenditure.

Inventive Principle:
Principle #16Partial or excessive action

3Measurement precision

If multiple PPG signals from different sensors are used, then measurement accuracy is improved, but system complexity increases

Engineering Contradiction:
Improvemeasurement accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The quality assessment system is designed to be sensor-agnostic and can evaluate PPG signals from multiple different sensor types (optical sensors, cameras, pulse oximeters). This universal assessment framework allows the system to handle multiple sensors without requiring separate processing pipelines for each sensor type, managing complexity while maintaining measurement precision.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system evaluates each sensor's PPG signal independently based on its local quality characteristics. This allows the system to weigh and combine signals from multiple sensors according to their individual quality metrics, improving measurement accuracy while avoiding the complexity of unified signal processing for heterogeneous sensors.

Inventive Principle:
Principle #3Local quality

4Reliability

If real-time quality assessment is performed, then monitoring reliability is improved, but computational load increases

Engineering Contradiction:
Improvemonitoring reliabilityVSAvoidcomputational load
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The system performs preliminary quality checks using computationally efficient metrics before committing to more intensive processing. This tiered approach ensures monitoring reliability by assessing signal adequacy early, while minimizing computational load by avoiding unnecessary complex analysis for signals that fail basic quality criteria.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

Real-time quality assessment is implemented on segmented portions of the PPG signal rather than the entire continuous signal. This segmentation reduces the computational burden at any given moment while maintaining monitoring reliability, as each segment can be evaluated independently and processed more efficiently.

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

The method effectively assesses PPG signal quality in real-time, reducing noise artifacts and improving the accuracy of vital sign monitoring across different sensor types, enhancing the reliability of PPG signals for clinical and home care applications.

Implementation Method 1

Photoplethysmogram (PPG) signal obtained at the fingertip indicates a volumetric fluctuation of blood in finger arterioles

Methodology Applied
Scientific EffectPhotoplethysmography: Absorption (EM radiation)

Data Source

PatentEP3479763B1System and method for photoplethysmogram (PPG) signal quality assessment
Publication Date: 2023.03.01 TATA CONSULTANCY SERVICES LTD
  • EP3479763B1 patent drawingFigure 1A~1H
  • EP3479763B1 patent drawingFigure 2~3
  • EP3479763B1 patent drawingFigure 4~5

AI summary

A system and method for PPG signal quality assessment is provided. The method includes obtaining a PPG signal captured using a testing device in real-time. The PPG signal is segmented into a first plurality of PPG signal samples such that length of each of the first plurality of PPG signal samples more than a threshold length. A signal sufficiency check (SSC) is performed for each first PPG signal sample of the first plurality of PPG signal samples to obtain at least a first set of PPG signal samples complying with the SSC. A set of features is extracted from the first set of PPG signal samples. Based on the set of features, each of the set of PPG signal samples is identified as one of a noisy signal sample and a clean signal sample using a plurality of Random Forest (RF) models created during the training phase.