Remote PPG Image Region Selection for Reliable Vital Sign Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for measuring physiological parameters using remote photoplethysmography (PPG) face challenges in accuracy and reliability due to noise from additional objects, subject movement, and environmental changes, which can lead to erroneous measurements.
Innovation Solution
A method involving image partitioning and signal quality assessment is employed, where a series of images are divided into multiple regions, and the signal quality is determined and classified for each region, allowing for the extraction of a reliable physiological parameter measurement from the highest quality image regions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If remote photoplethysmography is performed using a webcam to automatically monitor an individual, then the convenience and automation of health monitoring is improved, but measurement accuracy deteriorates when additional objects or people are in the image, or when the subject moves or environmental factors change
Solution Approach 1:
The image is divided into multiple regions of interest (ROIs), each corresponding to different skin regions of the subject. Multiple physiological signals are extracted from each ROI and then combined to produce the final measurement. This segmentation allows the system to select and combine signals from multiple locations, improving robustness against noise from movement, environmental changes, or occlusions by other objects.
Solution Approach 2:
The system extracts physiological signals from multiple ROIs and combines them using a fusion algorithm that weights each signal based on its quality. This feedback mechanism allows the system to dynamically adjust which regions contribute most to the final measurement, ensuring high accuracy even when some regions are affected by noise from subject movement or environmental factors.
2Quantity of substance
If the entire image is used for physiological parameter measurement, then more data is available for analysis, but noise from background objects and subject movement increases, leading to erroneous measurements
Solution Approach 1:
The image is divided into multiple regions of interest (ROIs), each corresponding to different skin regions of the subject. Multiple physiological signals are extracted from each ROI and then combined to produce the final measurement. This segmentation allows the system to select and combine signals from multiple locations, improving robustness against noise from movement, environmental changes, or occlusions by other objects.
Solution Approach 2:
The system extracts only the relevant physiological signals from specific skin regions (ROIs) while discarding background and non-skin areas. By isolating and processing only the useful signal sources, the system eliminates noise from background objects and focuses computational resources on the most reliable measurement areas.
3Measurement precision
If multiple image regions are processed to improve signal quality, then measurement accuracy is improved, but computational complexity increases
Solution Approach 1:
The image is divided into multiple regions of interest (ROIs), each corresponding to different skin regions of the subject. Multiple physiological signals are extracted from each ROI and then combined to produce the final measurement. This segmentation allows the system to select and combine signals from multiple locations, improving robustness against noise from movement, environmental changes, or occlusions by other objects.
Solution Approach 2:
The system combines multiple physiological signals extracted from different ROIs using a fusion algorithm. By merging signals from multiple regions and weighting them according to their quality, the system achieves more accurate and reliable measurements while managing computational complexity through efficient signal integration.
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 measurements by minimizing noise and disturbances, ensuring a strong and clean signal is used for analysis.
Implementation Method 1
techniques for performing remote photoplethysmography (PPG or rPPG) have been developed. In WO 2013/027027, a method of automatically monitoring an individual via a webcam is described. In this document, a person under test sits in front of the webcam and their pulse is identified from the slight variation in skin colour which occurs when the heart beats.
Data Source
AI summary
A method of performing a measurement of a physiological parameter of a person from a series of images is provided, the method comprising: acquiring a series of images of a person, each image of the series of images including at least a skin exposed region of the person; partitioning at least a portion of each image of the series of images into a plurality of image regions; computing target colour values for each image region of each image of the series of images; determining a signal quality indicator of each of the plurality of image regions for the series of images from the target colour values; classifying the plurality of image regions in accordance with the signal quality indicator which has been determined; and performing a measurement of a physiological parameter of the person using one or more of the plurality of image regions based on a result of the classification.