PPG Pulse Classification via Morphology and SNR Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for monitoring cardiac arrhythmias using photoplethysmography (PPG) struggle to accurately distinguish atrial fibrillation (AF) from other cardiac arrhythmias and cardiac dysfunctions, as they rely primarily on inter-beat interval features without effectively capturing pulse morphology variations.
Innovation Solution
A method that utilizes pulse wave analysis to extract time-related, normalized amplitude-related, and signal-to-noise ratio (SNR)-related features from PPG signals, combined with a machine learning model, to classify PPG pulses as 'normal', 'pathological', or 'non-physiological', enabling improved separation of AF episodes from other arrhythmias and monitoring of cardiovascular parameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If inter-beat interval features are used for arrhythmia detection, then sinus rhythm and AF episodes can be distinguished, but other cardiac arrhythmias cannot be accurately separated
Solution Approach 1:
The patent transitions from one-dimensional temporal analysis (inter-beat intervals) to multi-dimensional analysis by incorporating pulse morphology features (amplitude, shape characteristics) and signal quality metrics (SNR). This dimensional expansion enables the system to differentiate between various arrhythmia types that share similar heart rate patterns but exhibit distinct pulse wave characteristics.
Solution Approach 2:
The patent changes the analytical parameters from solely temporal (RR intervals) to include morphological parameters (pulse amplitude, waveform shape) and signal quality parameters (SNR). This parameter diversification allows the machine learning model to capture subtle differences between arrhythmia types, improving both detection accuracy and differentiation capability simultaneously.
2Adaptability or versatility
If PPG pulse morphology features are extracted and combined with machine learning, then various cardiac arrhythmias can be separated, but computational complexity increases
Solution Approach 1:
The patent segments the PPG signal analysis into distinct feature extraction components (temporal features, morphological features, SNR features) that can be processed independently and then integrated. This segmentation allows the system to manage computational complexity by processing features in modular stages rather than analyzing the entire signal simultaneously, making the complex analysis feasible for wearable devices.
3Measurement precision
If multiple feature types are extracted from PPG signals, then classification accuracy improves, but processing time increases
Solution Approach 1:
The patent performs preliminary processing of the PPG signal to extract multiple feature types in parallel or pre-computed manner before classification. By preparing temporal, morphological, and SNR features in advance through efficient signal processing pipelines, the system reduces the computational burden during actual classification, thereby minimizing processing time while maintaining high accuracy.
Data Source
Figure 1(a)~2(b)
Figure 3~4
Figure 5(a)~5(c)
AI summary
A computer program comprising instructions for implementing a method based on pulse wave analysis for monitoring cardiovascular vital signs, the method comprising: measuring a photoplethysmography (PPG) signal during a measurement time period such as to obtain a time series of PPG pulses; and during said measurement time period, identifying individual PPG pulses in the PPG signal, each PPG pulse corresponding to a PPG pulse cycle. For each PPG pulse, the method uses a pulse-wave analysis technique to determine, within the pulse cycle, at least one of: a time-related feature comprising a time duration and a normalized amplitude-related pulse-related feature and a SNR-related pulse-related. For each PPG pulse, a machine learning model is used in combination with the determined time-related, normalized amplitude-related and SNR-related features, to classify each PPG pulse in the pre-processed PPG signal as "normal", "pathological" or "non-physiological" such as to output a time series of pulse classes.