Multi-Sensor PPG Signal Quality via Segment Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Direct photoplethysmography (PPG) signals obtained from multiple sensors in wearable devices often suffer from low quality due to artefacts like motion, incorrect skin positioning, and ambient light interference, leading to inaccurate physiological parameter estimation, especially when static selection of optimal regions or colors results in inversions and bad quality sections.
Innovation Solution
A computer-implemented method for direct PPG that combines PPG signals from multiple sensors by identifying and removing bad quality segments, and combining good quality segments temporally corresponding across sensors to generate a multi-sensor PPG signal, which dynamically adjusts to variations in sensor quality and color, using wavelet transformation and neural networks for quality assessment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If multiple sensors are used to obtain PPG signals, then the availability of signal data increases, but the signal quality deteriorates due to artefacts from motion, positioning errors, and ambient light interference
Solution Approach 1:
The patent divides the PPG signal into multiple segments along the time axis and evaluates each segment independently for quality metrics. This allows identification and exclusion of specific low-quality segments while retaining good-quality segments from the same sensor, thereby maintaining signal quantity while improving overall reliability.
Solution Approach 2:
The patent combines multiple PPG signals from different sensors after individual quality assessment. By merging signals segment-by-segment based on their quality metrics, the system maintains the benefits of multiple sensors while filtering out artefacts and low-quality portions, thus improving signal reliability without losing data availability.
2Device complexity
If static selection of optimal regions or colors is applied, then processing complexity is reduced, but signal quality deteriorates due to inversions and bad quality sections
Solution Approach 1:
The patent implements dynamic selection of optimal sensor regions and color channels by evaluating quality metrics for each segment and adapting the selection accordingly. This dynamic approach allows the system to respond to changing signal conditions, avoiding static selections that may include inverted or low-quality sections, thereby improving signal reliability while maintaining manageable processing complexity.
Solution Approach 2:
The patent changes evaluation parameters such as signal inversion detection, quality thresholds, and sensor selection criteria based on segment-specific characteristics. By adapting parameters dynamically rather than using fixed static values, the system improves its ability to identify and exclude bad quality sections while controlling processing complexity through efficient parameter adjustment.
3Loss of information
If all PPG signal segments are retained for analysis, then data completeness is maintained, but measurement precision deteriorates due to inclusion of artefacts and low quality portions
Solution Approach 1:
The patent applies local quality assessment to individual signal segments rather than treating the entire signal uniformly. By evaluating and weighting segments based on their local quality characteristics, the system preserves complete temporal information while excluding or down-weighting specific artefact-contaminated portions, thereby improving measurement precision without sacrificing overall data completeness.
Solution Approach 2:
The patent implements feedback mechanisms where quality metrics of signal segments are continuously evaluated and used to adjust the inclusion or exclusion of segments in the final analysis. This feedback loop ensures that only high-quality segments contribute to physiological parameter estimation, improving measurement precision while maintaining data completeness through systematic quality-based filtering.
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
This approach enhances the accuracy and reliability of physiological parameter estimation by eliminating bad quality segments and maintaining inverted segments as good quality, resulting in improved signal quality and more reliable physiological parameter monitoring.
Implementation Method 1
PPG makes use of light absorption by blood to track these volumetric changes
Implementation Method 2
A light sensor then converts these variations in light reflection into a digital signal
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A computer-implemented method (100) for direct photoplethysmography or direct PPG comprises: - obtaining (101) during a time interval plural PPG signals (311-314; 401-404; 501-504; 601-608; 701-708) for respective sensors (321, 322, 323, 324) in a wearable device (301); and - combining (105) the plural PPG signals (311-314; 401-404; 501-504; 601-608; 701-708) to thereby obtain a multi-sensor PPG signal (405; 505; 609; 709).