Third-Order Derivative PPG Signal Feature Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing feature extraction methods for photoplethysmography (PPG) signals face challenges in accurately identifying feature points, particularly in cases of non-standard morphology, leading to missing or ambiguous features, which complicates the decomposition of PPG signals and results in significant deviations in estimating health indicators like blood pressure and vascular age.
Innovation Solution
The method involves calculating a third-order derivative photoplethysmography (TDPPG) signal to impute missing feature points and resolve ambiguous points in the second-order derivative photoplethysmography (SDPPG) signal by analyzing zero-crossing points and regional extreme points, allowing for precise determination of feature points within a cardiac cycle.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If only second-order derivative photoplethysmography (SDPPG) signal is used for feature extraction, then the calculation process is simple, but feature points may be missing or ambiguous in non-standard morphology cases
Solution Approach 1:
The patent segments the feature extraction process into multiple stages: first using SDPPG for initial feature detection, then applying TDPPG specifically for resolving ambiguous or missing features. This segmented approach allows each derivative order to perform its optimal function without unnecessary complexity.
Solution Approach 2:
The patent transitions from a single-dimensional SDPPG analysis to a multi-dimensional approach by introducing TDPPG as an additional analytical dimension. This enables the system to resolve feature ambiguities that cannot be detected by SDPPG alone, improving identification accuracy without proportionally increasing complexity.
2Measurement precision
If third-order derivative photoplethysmography (TDPPG) signal is always calculated to resolve all ambiguous features, then feature point identification accuracy is improved, but the calculation complexity increases
Solution Approach 1:
The patent applies partial action by calculating TDPPG only when and where needed—specifically when SDPPG features are ambiguous or missing—rather than always calculating it. This selective application improves accuracy for problematic cases without unnecessarily increasing complexity for all cases.
Solution Approach 2:
The patent applies local quality by using different levels of derivative analysis in different situations: SDPPG for clear features and TDPPG for ambiguous features. This localized approach ensures high precision where needed while maintaining simplicity where sufficient.
3Reliability
If feature points are missing or ambiguous in SDPPG signal, then the PPG signal decomposition into Gaussian component waves becomes difficult, but calculating higher-order derivatives increases processing complexity
Solution Approach 1:
The patent performs preliminary action by using SDPPG to identify obvious features first, then preparing TDPPG calculation only for regions where features are ambiguous or missing. This preliminary assessment avoids unnecessary complex calculations while ensuring decomposition accuracy.
Solution Approach 2:
The patent uses TDPPG as an intermediary tool to resolve the difficulty of PPG signal decomposition when SDPPG features are insufficient. Rather than directly attempting difficult decomposition with incomplete features, the TDPPG intermediary provides the additional information needed for accurate decomposition.
Data Source
AI summary
A feature extraction method for photoplethysmography includes obtaining a photoplethysmography (PPG) signal; calculating a first-order derivative photoplethysmography (FDPPG) signal, a second-order derivative photoplethysmography (SDPPG) signal and a third-order derivative photoplethysmography (TDPPG) signal from the PPG signal, wherein the SDPPG signal has multiple feature points; and performing a feature extraction operation that exploits the property of the TDPPG signal to impute the missing feature points of the SDPPG signal. Moreover, the feature extraction operation further includes: resolving the ambiguous feature points of the SDPPG signal.