Camera-Based SpO2 Monitoring Using Physiological Activity Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current contact-based SpO2 measurement devices are impractical for long-term monitoring due to discomfort and hygiene issues, and remote deep learning-based SpO2 measurement techniques are not yet at state-of-the-art for accurate and continuous health monitoring.
Innovation Solution
A system and method using a camera to capture color video frames, apply computer vision for face detection and landmark identification, generate a Physiological Activity Image (PAI) through image and signal processing, and utilize a deep learning model for SpO2 estimation, providing a contactless and accurate SpO2 measurement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If contact-based pulse oximeters are used for SpO2 measurement, then measurement accuracy is improved, but comfort and hygiene are worsened
Solution Approach 1:
The patent replaces the mechanical contact-based pulse oximeter with a contactless camera system that uses image processing and deep learning to estimate SpO2. This substitution eliminates physical contact, thereby improving comfort and hygiene while maintaining measurement capability through computational analysis of facial images.
Solution Approach 2:
The patent introduces an intermediary processing pipeline between the camera and SpO2 estimation. This includes image processing modules that extract physiological signals from facial images, and deep learning models that translate these signals into SpO2 values. This intermediary system enables accurate measurement without direct contact.
2Duration of action of stationary object
If contact-based devices are used for long-term monitoring, then measurement continuity is improved, but comfort and usability are worsened
Solution Approach 1:
By replacing the mechanical contact device with a contactless camera-based system, the patent enables long-term monitoring without the discomfort and hygiene issues that limit the duration of use of traditional pulse oximeters. The system can operate continuously as the camera simply captures images without physical contact.
3Reliability
If contact-based monitoring is used, then measurement reliability is improved, but spread of infectious diseases is facilitated
Solution Approach 1:
The patent substitutes the contact-based measurement system with a contactless camera system that captures facial images and processes them to estimate SpO2. This eliminates the physical contact that could facilitate the spread of infectious diseases while maintaining measurement reliability through robust image processing and deep learning algorithms.
4Ease of operation
If remote deep learning-based SpO2 measurement is used, then comfort and hygiene are improved, but measurement accuracy is worsened
Solution Approach 1:
The patent introduces sophisticated intermediary processing steps between the camera and final SpO2 estimation. These include image processing modules that extract physiological signals from facial images, and deep learning models that accurately translate these signals into SpO2 values. This detailed intermediary processing ensures high accuracy despite the contactless approach.
Solution Approach 2:
The patent transforms the input data through multiple parameter changes and processing stages. Facial images are converted into physiological signals through image processing, and then these signals are transformed into SpO2 estimates through deep learning. This multi-stage parameter transformation enables accurate measurement while maintaining comfort and hygiene.
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 a comfortable and hygienic method for continuous SpO2 monitoring, leveraging deep learning for enhanced accuracy and integration into telehealth, wellness programs, and clinical applications.
Implementation Method 1
executing light intensity analysis for the color image frames of the subject
Data Source
AI summary
A system includes a camera, a color image and signal processing system, a physiological activity image processing system, a blood oxygen saturation estimator, and a report output module. The camera captures color image frames of a subject. The color image and signal processing system extracts physiological signals by analyzing light intensity, detecting and resizing facial images, interpolating facial landmarks, creating facial patches, selecting patches based on landmarks, and extracting signals from different color channels. The physiological activity image processing system generates a PAI embedding the subject's physiological information by forming 1D signals, concatenating filtered 1D signals into 2D images for each color channel, and combining the 2D images along a color channel dimension. The blood oxygen saturation estimator, using a deep learning model, receives the PAI to estimate the subject's blood oxygen saturation level. The report output module provides a readable report of the estimation.