IMU-Based Heart Health Detection With Device-Specific Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing heart monitoring technologies, such as those using electrocardiograms (ECG), ballistocardiography (BCG), and seismocardiography (SCG), struggle with calibration accuracy when IMU measurements are not placed at the center of mass, leading to unreliable health state detection, particularly for conditions like congestive heart failure (CHF).
Innovation Solution
A system and method utilizing an Inertial Measurement Unit (IMU) on a device, like a smartphone, to measure six-dimensional signals, including linear and rotational dimensions, and calibrate based on device characteristics like weight and moment of inertia, enabling robust heart health state detection without relying on ECG, by processing cardiac kinetic parameters like energy and power.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If calibration is based on user characteristics (center of mass), then BCG measurements are accurate, but SCG measurements placed apart from center of mass cannot be properly calibrated
Solution Approach 1:
The patent changes the calibration parameter from user-specific characteristics (center of mass location) to device-specific characteristics (mass and moment of inertia). This allows the same calibration approach to work regardless of where the device is placed on the body, resolving the contradiction between measurement precision and placement flexibility.
Solution Approach 2:
The patent introduces device characteristics (mass and moment of inertia) as an intermediary calibration parameter that bridges the gap between the measurement signal and the actual cardiac kinetic parameters. This intermediary allows accurate calibration without requiring knowledge of the user's center of mass or precise placement location.
2Measurement precision
If IMU is placed at center of mass, then calibration works well, but device placement becomes restricted
Solution Approach 1:
The patent changes the calibration parameter from user-specific characteristics (center of mass location) to device-specific characteristics (mass and moment of inertia). This allows the same calibration approach to work regardless of where the device is placed on the body, resolving the contradiction between measurement precision and placement flexibility.
3Device complexity
If only one or few dimensions are measured, then device complexity is reduced, but measurement reliability deteriorates
Solution Approach 1:
The patent extends the measurement from one or few dimensions to six dimensions (three linear accelerations and three angular velocities). This multi-dimensional measurement captures both translational and rotational cardiac movements, significantly improving the reliability of health state detection while using a standard IMU sensor.
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 provides accurate and reliable heart health state detection, including CHF, even when the IMU is not at the center of mass, using machine learning to analyze cardiac kinetic features, improving detection accuracy and reducing reliance on orientation-dependent measurements.
Implementation Method 1
the IMU is configured to measure a measurement signal when the device is placed on a body of the user
Implementation Method 2
with the gyroscope three rotational dimensions to obtain the rotation kinetic energy of the movement of the heart. The kinetic energy is calculated based on the moment of inertia and the mass of the user
Implementation Method 3
The kinetic energy is calculated based on the moment of inertia and the mass of the user
Data Source
AI summary
System for detecting a health state of a heart of a user (10) comprising: a device (1) including an IMU (1), wherein the IMU (1) is configured to measure a measurement signal when the device is placed on a body of the user (10); a processing means (2) configured to determine a characteristic of the device (11), to determine the health state of the heart of the user (10) based on the measurement signal and based on the characteristic of the device (11), and to output an output information based on the determined health state FIG. 1.