PPG Signal Segmentation for Artifact Removal
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Physiological monitoring systems face challenges in extracting reliable physiological information due to corrupted signals caused by artifacts, baseline shifts, and high pulse-to-pulse variability in photoplethysmograph (PPG) signals.
Innovation Solution
A method and system that identify and exclude portions of the PPG signal with artifacts, large baseline shifts, and excessive pulse-to-pulse variability, using thresholds to determine usable segments of the signal, ensuring the remaining signal is of sufficient length for accurate physiological information determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the entire PPG signal is used for physiological information extraction, then more data is available for analysis, but the reliability of the physiological information decreases due to artifacts and noise
Solution Approach 1:
The PPG signal is divided into multiple segments based on detected artifacts, baseline shifts, and pulse-to-pulse variability. Each segment is evaluated independently to determine quality metrics, allowing the system to identify and exclude only the corrupted portions while retaining usable segments for physiological information extraction.
Solution Approach 2:
Different quality assessment criteria are applied to different portions of the PPG signal based on local characteristics. The system evaluates each segment's artifact content, baseline stability, and pulse variability separately, assigning quality metrics that reflect the specific conditions of each segment rather than treating the entire signal uniformly.
2Reliability
If signal processing filters are applied to remove artifacts, then signal quality improves, but the complexity of the processing system increases
Solution Approach 1:
The system performs preliminary quality assessment of the PPG signal by detecting artifacts, baseline shifts, and pulse-to-pulse variability before physiological information extraction. This preliminary evaluation identifies problematic segments that can be excluded from analysis, preventing corrupted data from degrading the physiological measurements without requiring complex real-time filtering during the main processing stage.
3Measurement precision
If contiguous signal portions are required for analysis, then measurement accuracy improves, but the amount of usable signal data decreases
Solution Approach 1:
The signal is segmented into contiguous portions based on artifact detection, with each segment evaluated for quality. The system identifies the longest contiguous segment that meets minimum quality criteria and length requirements, ensuring that physiological information is extracted from continuous, high-quality signal portions while maximizing the use of available usable data.
Data Source
AI summary
A patient monitoring system may determine portions of a PPG signal that correspond to artifacts, to a baseline shift that exceeds a threshold, or to a pulse-to-pulse variability that exceeds a threshold. The patient monitoring system may identify a contiguous portion of the PPG signal that does not include the determined portions. The contiguous portion of the PPG signal may be used to determine physiological information.