Remote Vital Sign Extraction Using Disturbance Projection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for extracting physiological information from electromagnetic radiation data streams, such as those from image frames, face challenges in poor signal-to-noise ratios and varying ambient conditions, making it difficult to accurately extract vital signs like heart rate and respiration rate without obtrusive devices.
Innovation Solution
An unobtrusive remote monitoring system that uses a video camera to detect electromagnetic radiation and employs multivariate statistics to extract characteristic index elements, projecting them onto a disturbance-reduced index element to minimize noise and enhance signal quality, even under poor ambient conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If signal processing is performed under varying ambient conditions, then measurement robustness is improved, but measurement precision deteriorates due to noise from changing luminance and object movement
Solution Approach 1:
The patent extracts only the relevant physiological signal components from the image data while separating and eliminating disturbing signal components. This is achieved by identifying characteristic patterns of physiological signals (e.g., periodic variations corresponding to heart rate) and isolating them from the overall image signal, thereby removing the influence of ambient disturbances such as luminance changes and object movements.
Solution Approach 2:
The patent transforms the image data into a different parameter space or representation where physiological signals can be distinguished from noise. By changing the analysis parameters (e.g., transforming spatial-temporal image data into frequency domain or extracting specific optical density parameters), the system can filter out ambient disturbances and enhance the visibility of physiological signals.
2Measurement precision
If disturbing signal components are removed through signal processing, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system uses the inherent characteristics of the physiological signals themselves to perform noise reduction. By exploiting the periodic nature and specific temporal patterns of physiological signals (e.g., heart rate variability), the algorithm automatically distinguishes and extracts these signals from the background noise without requiring complex external processing or additional hardware components.
Solution Approach 2:
The patent applies signal processing to only the necessary portions of the image data that contain physiological information, rather than processing the entire image stream in detail. By focusing computational resources on specific regions of interest or specific temporal-frequency components where physiological signals are present, the system achieves effective noise reduction with reduced overall processing complexity.
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
The system effectively improves the signal-to-noise ratio, allowing for accurate extraction of vital signs like heart rate and respiration rate without requiring laboratory-like conditions, using a video camera to detect electromagnetic radiation and applying multivariate statistics for noise reduction.
Implementation Method 1
an interface for receiving a data stream comprising a sequence of image frames derivable from electromagnetic radiation emitted or reflected by an object
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
The present invention relates to a device and a method for extracting information from detected characteristic signals. A data stream (76, 78, 80, 82) derivable from electromagnetic radiation (14) emitted or reflected by an object (11) is received and a plurality of characteristic index elements (50) varying over time can be extracted therefrom. The index elements (50) comprise physiological information (48) indicative of at least one at least partially periodic vital signal (12), and a disturbing signal component (58). For eliminating the disturbing signal component (58) to a great extent, the characteristic index elements (50) can be projected to a disturbance-reduced index element (64) having a distinct orientation in relation to a presumed orientation of the disturbing signal component (58). The disturbance-reduced index element (64) is chosen so as to reflect a dominant main orientation and length of the disturbing signal component (58) over time. Consequently, the mainly genuine physiological information (48) extracted from the data stream (76, 78, 80, 82) in this way can be utilized for determining the at least one at least partially periodic vital signal (12).