Camera-Based rPPG Blood Pressure Estimation Without Cuffs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cuff-based methods for blood pressure measurement are uncomfortable and not suitable for long-term monitoring, while contactless methods using photoplethysmography (PPG) signals are also not pain-free and lack accuracy.
Innovation Solution
A camera-based system that uses computer vision and machine learning to extract physiological signals from video frames, generating remote photoplethysmography (rPPG) signals, which are processed to estimate systolic and diastolic blood pressure without physical contact.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If cuff-based methods are used for blood pressure measurement, then measurement accuracy is improved, but user comfort deteriorates due to discomfort and skin irritation
Solution Approach 1:
The patent replaces the mechanical cuff-based oscillometry system with an optical-based remote photoplethysmography system using standard camera sensors. This substitution eliminates mechanical contact with the skin while maintaining measurement capability through optical detection of blood volume changes in facial regions.
Solution Approach 2:
The patent introduces facial video images as an intermediary medium between the measurement system and the subject. Instead of direct mechanical contact, the system captures optical reflections from facial skin that contain physiological information, using the face as an indirect sensing surface.
2Object-affected harmful factors
If contactless methods using PPG signals are used, then user comfort is improved, but measurement accuracy deteriorates
Solution Approach 1:
The patent transitions from traditional one-dimensional PPG signal processing to multi-dimensional analysis by utilizing spatial information from facial video frames. The system processes color channel variations across different facial regions and time, adding spatial and chromatic dimensions to the measurement approach.
Solution Approach 2:
The patent changes the measurement parameters from standard PPG wavelength detection to multi-color channel analysis of facial video. By analyzing color variations in RGB channels and their temporal derivatives, the system extracts physiological information with enhanced accuracy for blood pressure estimation.
3Reliability
If cuff-based methods are used, then measurement reliability is improved, but ease of operation deteriorates due to discomfort during long-term monitoring
Solution Approach 1:
The patent replaces the mechanical cuff system with a contactless optical sensing system using standard cameras. This eliminates the need for physical attachment devices, making the system suitable for prolonged use without skin irritation while maintaining measurement reliability through robust signal processing.
Solution Approach 2:
The patent utilizes standard camera sensors and video processing technology that can be integrated into various devices (smartphones, computers, televisions). This multi-functional approach allows the same system to serve both entertainment purposes and health monitoring, enhancing ease of adoption for long-term use.
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
Enables accurate, continuous, and comfortable blood pressure monitoring by leveraging computer vision and machine learning, suitable for long-term use.
Implementation Method 1
executing light intensity analysis for the color image frames of the subject
Implementation Method 2
extracting the physiological signals from the selected region of interest... generate a rPPG signal embedding the subject's cardiovascular activity information
Data Source
AI summary
In the present invention, to extract the subject's SBP and DBP estimation, the system first starts capturing a color video of the subject. Then, computer vision techniques are applied to the video frames to locate regions of interest on the face, and these regions are tracked continuously for a while. Next, the images with located region of interest are fed into a pipeline with image processing, signal processing, and machine learning algorithms to create a signal based on the rPPG signal. The rPPG signal, along with its derivatives of different orders and an estimated HR from the rPPG signal are all fed into a machine learning model to estimate/predict the SBP and DBP of the subject.