Remote Photoplethysmography Signal Extraction Using Skin Pixel Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Camera-based remote photoplethysmography (rPPG) systems face challenges in accurately extracting vital signs due to the presence of non-skin pixels in the region-of-interest, leading to signal corruption and inaccuracies in heart rate and blood oxygen saturation measurements.
Innovation Solution
A device and method that weight pixel values to reduce the impact of non-skin pixels, using statistical parameters like variance or standard deviation to improve signal quality by concatenating these values over time to extract physiological parameters such as heart rate and blood oxygen saturation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If face detection and tracking are used to identify region-of-interest, then the monitoring can be performed remotely and unobtrusively, but the extracted PPG signal is corrupted by non-skin pixels
Solution Approach 1:
The patent segments the region-of-interest into skin pixels and non-skin pixels using color space analysis and clustering algorithms. By dividing the ROI based on pixel characteristics rather than treating it as a uniform region, the system can selectively process only skin-related pixels for PPG signal extraction, thereby eliminating corruption from non-skin pixels while maintaining unobtrusive remote monitoring.
2Power
If a larger region-of-interest is used to improve signal strength, then more pixels are available for PPG extraction, but more non-skin pixels are included which corrupt the signal
Solution Approach 1:
The patent applies local quality analysis by evaluating pixel characteristics individually within the ROI using color space transformations and clustering. Each pixel is assigned a quality metric based on its likelihood of being skin tissue, allowing the system to weight or exclude pixels locally rather than uniformly. This enables inclusion of all available skin pixels for signal strength while excluding non-skin pixels to maintain accuracy.
3Productivity
If conventional PPG extraction methods are used on the extracted region, then the process is simple and fast, but the presence of non-skin pixels leads to inaccurate vital sign measurements
Solution Approach 1:
The patent performs preliminary action by preprocessing the ROI to identify and separate skin pixels from non-skin pixels before PPG signal extraction. Using color space analysis, clustering, and pixel classification algorithms, the system prepares a cleaned set of skin-only pixels in advance. This preliminary segmentation enables subsequent PPG extraction to proceed quickly with high accuracy, as the harmful non-skin pixels have already been removed.
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 robustness of vital sign measurements by mitigating the effects of non-skin pixel pollution, providing a superior signal quality even in regions with significant contamination.
Implementation Method 1
Photoplethysmography (PPG) is an optical measurement technique that evaluates a time-variant change of light reflectance or transmission of an area or volume of interest. PPG is based on the principle that blood absorbs light more than surrounding tissue, so variations in blood volume with every heart beat affect transmission or reflectance correspondingly.
Implementation Method 2
By evaluating the transmittance and/or reflectivity at different wavelengths (typically red and infrared), the blood oxygen saturation can be determined.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present invention relates to fields of medical technology and camera- based vital signs monitoring using remote photoplethysmography (rPPG). A Device (10) for determining a physiological parameter of a subject (20) is presented, the device comprising: an interface (11) for receiving image data of a scene, said image data comprising a time- sequence of image frames; and a processor (12) for processing said image data, wherein the processor is configured to perform the steps of: determining, for each of said image frames, a first statistical parameter value indicative of a statistical dispersion of pixel values of said image frame (202); concatenating said first statistical parameter values over time based on the time-sequence of the corresponding image frames to obtain a first candidate signal (203); extracting a physiological parameter of the subject based on said first candidate signal (204). The present invention further relates to a corresponding system (1) and method (200).