Physiological Signal Normalization via Location-Specific Parameters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current biomedical data analytics face challenges in normalizing physiological signals, which differ significantly between individuals and over time due to various factors, leading to high errors when not properly calibrated, especially requiring expensive equipment or frequent recalibration.
Innovation Solution
A device and method that calculates location-specific normalization parameters for physiological signals using a data input module, activity recognition, location recognition, and a mathematical model, allowing for continuous, automatic recalibration without specific exercise tests, accounting for environmental and individual changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If standard normalization techniques are applied, then data can be processed, but the procedure is not extendable to new subjects and requires complete protocol repetition
Solution Approach 1:
The system performs preliminary calibration with a small subset of subjects to establish initial normalization parameters. These pre-computed parameters are then stored and can be directly applied to new subjects without requiring them to undergo the complete calibration protocol, thus enabling quick adaptation to new individuals.
Solution Approach 2:
The system creates a library of normalization parameters from calibrated subjects. When a new subject arrives, the system copies and applies the most relevant pre-computed parameters from the library rather than performing complete recalibration, significantly reducing the time and effort required for new subject integration.
2Measurement precision
If individual calibration is performed, then measurement precision improves, but it requires expensive equipment and frequent recalibration
Solution Approach 1:
The system enables devices to self-calibrate by automatically detecting activity levels and computing normalization parameters from the device's own sensor data. This eliminates the need for expensive external calibration equipment and complex manual procedures, allowing the device to perform calibration autonomously using its built-in sensors and processors.
Solution Approach 2:
The system replaces mechanical/equipment-based calibration methods with computational approaches. Instead of using expensive physiological measurement equipment, the system uses algorithms that process data from standard wearable sensors (accelerometers, gyroscopes) to derive normalization parameters, substituting complex hardware with software-based solutions.
3Ease of operation
If physiological signals are not normalized, then data collection is simple, but high error between individuals is reported
Solution Approach 1:
The system automatically transforms physiological signal parameters by computing activity-specific normalization factors. These parameters are derived from the relationship between sensor readings and known activity levels, then applied to normalize the signals. This parameter transformation enables accurate comparisons between individuals while maintaining the simplicity of continuous data collection.
Data Source
AI summary
An example device includes: a data input module configured to receive information about a living being's physiological signals, coordinates, and motion intensity; an activity recognition module configured to calculate, from information received about the living being's motion intensity, a living being's activity; a location recognition module, configured to calculate, from information received about the living being's coordinates, a living being's location; a memory storage configured to store information about the living being's physiological signals and activity in association with the location; a normalization parameters estimator module configured to use a mathematical model to calculate a plurality of normalization parameters for a plurality of detected activities and locations; and a model selector module configured to determine, based on the plurality of normalization parameters and the living being's location, a set of location-specific normalization parameters used to further calculate normalized physiological signals for the living being.