Camera-Based Vital Sign Discrimination Using Direction Change Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for non-invasive respiratory rate monitoring struggle to accurately differentiate respiratory signals from noise, especially in cases of shallow breathing, due to the reliance on expensive motion estimation techniques and frequency analysis, which can be ineffective for non-periodic signals and require large temporal windows.
Innovation Solution
A camera-based system that analyzes raw input image frames to determine the number of direction changes and time intervals within a region of interest, using projection profiles to distinguish between vital sign information and noise by setting thresholds for direction changes and time distances, allowing for robust discrimination of respiratory signals from noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If motion estimation techniques are used to detect respiratory signals, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent extracts only the essential feature needed for respiratory detection - the number of direction changes in projection profiles - while discarding complex motion estimation algorithms. This extraction approach maintains measurement precision by focusing on the most discriminative feature while significantly reducing device complexity
Solution Approach 2:
The patent replaces expensive, complex motion estimation techniques with a simple, computationally inexpensive method based on counting direction changes in projections. This simpler approach achieves comparable or superior precision for shallow breathing detection without requiring sophisticated processing resources
2Measurement precision
If frequency analysis is used to differentiate respiratory signals from noise, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The patent performs preliminary analysis by examining direction changes in projection profiles before committing to full frequency analysis. This preliminary step quickly identifies whether a signal is present, avoiding the time-consuming requirement of large temporal windows needed for traditional frequency analysis methods
3Measurement precision
If large temporal windows are used for frequency analysis, then measurement precision is improved, but productivity decreases
Solution Approach 1:
The patent applies partial action by using only the necessary minimum processing - counting direction changes in projection profiles - rather than performing complete frequency analysis over large temporal windows. This partial approach maintains sufficient measurement precision for clinical purposes while dramatically improving processing speed and productivity
4Productivity
If only region of interest is analyzed, then productivity is improved, but measurement precision worsens
Solution Approach 1:
The patent transitions from analyzing only the spatial region of interest to analyzing the temporal dimension by examining direction changes over time in projection profiles. This dimensional shift allows the system to maintain high productivity through focused ROI analysis while improving measurement precision through temporal pattern recognition that distinguishes true respiratory signals from noise
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 enables simple, low-cost, and accurate discrimination of respiratory signals from noise, even in cases of shallow breathing, by focusing on raw input values and observing direction changes and time intervals, improving the reliability and accuracy of vital sign monitoring.
Implementation Method 1
an imaging unit, in particular a camera, for remotely detecting electromagnetic radiation emitted or reflected from the subject
Data Source
Figure 1~2
Figure 3~5
Figure 6~7
AI summary
The present invention relates to a device and method for obtaining a vital sign of a subject despite motion of the subject, in particular for discriminating a vital sign such as a respiratory information signal from noise in a projection based vital signs registration. The proposed device comprises an interface (124) for receiving a set of image frames of a subject, an analysis unit (128) for determining the amount of direction changes and/or the time distances between direction changes within a region of interest in a subset of image frames comprising a number of image frames of said set, a direction change indicating a change of the direction of motion appearing within said region of interest, an evaluation unit (132) for determining if said region of interest within said subset of image frames comprises a vital sign information and/or noise by use of the determined amount of direction changes and/or the time distances for said subset of image frames, and a processor (136) for determining the desired vital sign of the subject from said region of interest within said subset of image frames if it is determined that said region of interest within said subset of image frames comprises a vital sign information.