Wearable Biosensor Signal Processing for Motion Artifact Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current wearable biosensors face challenges in accurately monitoring physiological parameters like blood pressure and respiration rate due to high costs, obtrusive measurement methods, and interference from motion artifacts and RF noise, limiting their effectiveness in continuous remote monitoring, especially during patient movement.
Innovation Solution
A low-cost, energy-efficient sensor calibration model using deep learning and compressed sensing techniques to improve the classification accuracy of ECG and PPG signals, reducing motion artifacts and interference, allowing for continuous monitoring of physiological parameters like skin temperature, blood pressure, heart rate, oxygen saturation, and respiration rate on wearable devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional wearable biosensors are used to monitor physiological parameters, then measurement capability is provided, but cost is high and device complexity increases
Solution Approach 1:
The patent replaces complex mechanical sensor systems with a computational approach using compressed sensing and deep learning. Instead of using traditional complex hardware sensors to directly measure physiological parameters, the system uses simple sensors to collect raw signals and then applies compressed sensing algorithms and neural networks to extract physiological information computationally, thereby reducing hardware complexity while maintaining measurement accuracy.
Solution Approach 2:
The patent transforms the measurement approach by changing from direct physiological parameter measurement to indirect measurement through signal processing. The system measures raw biosignals (electrical, mechanical, optical) and then uses parameter transformation through compressed sensing and deep learning algorithms to derive physiological parameters, allowing accurate monitoring with simpler sensors.
2Reliability
If continuous monitoring is implemented to detect rapid health deterioration, then patient safety is improved, but energy consumption increases
Solution Approach 1:
The patent implements periodic sampling of biosignals at optimized intervals rather than continuous high-rate sampling. The compressed sensing algorithm can reconstruct physiological parameters from these periodic, subsampled measurements, maintaining detection reliability for rapid health deterioration while significantly reducing the energy consumption associated with continuous high-frequency data acquisition and processing.
Solution Approach 2:
The patent extracts only the essential physiological information from raw biosignals using compressed sensing and deep learning, rather than continuously processing and transmitting all raw data. This extraction approach focuses computational energy on deriving critical physiological parameters, maintaining monitoring reliability while reducing overall energy consumption by eliminating redundant data processing.
3Ease of operation
If motion artifacts are included in biosignal measurement during patient movement, then mobility is maintained, but measurement accuracy deteriorates
Solution Approach 1:
The patent converts motion artifacts, which are traditionally harmful to biosignal accuracy, into useful information. The deep learning model is trained to recognize patterns that distinguish between motion-induced signal variations and actual physiological changes. By learning from data collected during movement, the system can differentiate between artifacts and genuine physiological signals, maintaining both patient mobility and measurement accuracy.
Solution Approach 2:
The patent performs preliminary training of the deep learning model using data that includes motion artifacts. During this preliminary phase, the model learns to identify and compensate for motion-related distortions. When deployed, the pre-trained model can automatically correct for motion artifacts in real-time, allowing patients to move freely while maintaining measurement precision through the model's learned compensation mechanisms.
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 proposed solution enables accurate, continuous remote monitoring of multiple physiological parameters with reduced computational complexity and power consumption, addressing the limitations of existing systems by improving data quality and reducing costs, thus enhancing patient monitoring efficiency and accuracy.
Implementation Method 1
measuring a biosignal, from which the physiological parameter is deducible
Implementation Method 2
measuring a biosignal, from which the physiological parameter is deducible
Implementation Method 3
converting the signal to a vector having a plurality of different frequency components each with a corresponding magnitude coefficient
Implementation Method 4
the machine learning algorithm is configured to substantially remove from the transmitted signal noise data associated with electromagnetic interference
Implementation Method 5
the machine learning algorithm is configured to substantially remove from the transmitted signal noise data associated with motion of the wearable sensor
Data Source
AI summary
A method of collecting physiological parameter data of a monitored subject comprises measuring a biosignal from which the physiological parameter is deducible, including noise; converting the noisy measured biosignal to a vector having different frequency components with corresponding magnitude coefficients; discarding select frequency components with coefficients below a prescribed threshold; and communicating the reduced vector to a computing device for processing to deduce the physiological parameter. A method of processing physiological parameter data comprises receiving a measured biosignal with electromagnetic interference noise; obtaining from the noisy measured biosignal representative data using a machine learning algorithm; and determining the physiological parameter from the representative data. A system for monitoring a physiological parameter comprises a wearable sensor configured to measure a biosignal and to remove noise from the measured signal, and a portable computing device configured to receive a transmitted signal from the sensor and to determine the physiological parameter therefrom.