Physiological Signal Quality Evaluation with Adaptive Shape Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for evaluating the quality of periodic or quasi-periodic physiological signals, such as PPG, ECG, and ICG, are not robust and accurate due to their inability to adapt to changes in environment and activity level, leading to unreliable determination of physiological parameters without requiring calibration or individualized measurements.
Innovation Solution
A method and system that evaluate signal quality by comparing the morphology of temporally offset segments of the physiological signal within adaptive temporal windows, using a transformation function like SRVF to determine a shape difference and compute a quality index, integrating changes in environment and activity level, and optionally incorporating gyroscope data to detect motion artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional quality evaluation methods are used for physiological signals, then the system is simple to operate, but the measurement precision and reliability are insufficient due to inability to adapt to environmental changes and motion artifacts
Solution Approach 1:
The patent implements dynamic adaptation by continuously adjusting the reference signal based on incoming physiological signal segments. The reference signal is updated using a forgetting factor that weights recent segments more heavily, allowing the system to adapt to changing environmental conditions and activity levels without requiring manual recalibration or complex device architecture
Solution Approach 2:
The patent changes the parameter of the reference signal dynamically by incorporating new signal segments with varying weights. The forgetting factor alpha controls the rate of parameter change, allowing the system to adapt to environmental changes while maintaining computational efficiency through a simple weighted update mechanism rather than complex reprocessing
2Adaptability or versatility
If fixed predetermined values are used for quality assessment, then the computation is simple, but the adaptability to different environments and activity levels is poor
Solution Approach 1:
The system performs self-calibration by automatically updating its own reference signal using the physiological signal segments it receives. The reference signal adapts to the subject's specific characteristics, environment, and activity level without requiring external calibration equipment, manual intervention, or complex initialization procedures
Solution Approach 2:
The patent performs preliminary adaptation by continuously building and updating the reference signal before final quality assessment. The reference signal is prepared in advance through cumulative segmentation and weighted averaging, enabling the system to adapt to individual subject characteristics and environmental conditions before evaluating new signal quality
3Measurement precision
If signal segments are compared without transformation, then the computation is faster, but the shape difference measurement is less accurate
Solution Approach 1:
The patent substitutes direct mechanical comparison of raw signal segments with a transformation-based approach using SRVF (square root velocity function). This mathematical transformation converts the comparison problem into a more accurate metric space where Euclidean distance better reflects true shape differences, while the transformation itself is computationally efficient
Data Source
Figure 1~2
Figure 3
Figure 4~5
AI summary
This method, intended for the evaluation of the quality of at least one periodic or quasi-periodic physiological signal, comprises steps of: - segmenting the physiological signal temporally into a plurality of signal segments, - for each given signal segment, determining a distance representative of a shape difference between the given signal segment and at least one signal segment temporally offset relative to the given signal segment, - determining a quality index of the given signal segment according to the distance determined for the given signal segment.