Personalized Heart Rate Prediction From Motion During PPG Dropout
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing heart rate monitoring technologies, particularly those using low power optical sensors, struggle with signal quality issues due to movement, poor contact, and interference, leading to unreliable results, especially during strenuous activities.
Innovation Solution
A personalized heart rate prediction system that utilizes kinematic information from sensors to predict heart rate by training models with high confidence measurements, employing 2D lookup tables and auto-regressive filtering to fill in gaps in low signal quality regions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If low power optical sensors are used for heart rate monitoring, then energy consumption is reduced and convenience is improved, but signal quality deteriorates due to movement, poor contact, and interference
Solution Approach 1:
The patent introduces an intermediary prediction model that mediates between the unreliable PPG signal and the required heart rate output. When PPG signal quality degrades below a threshold, the system switches to using a prediction model trained on kinematic data (accelerometer, gyroscope) to estimate heart rate, rather than directly relying on the degraded optical signal. This intermediary approach maintains reliability while preserving the low-power benefit of optical sensors.
Solution Approach 2:
The system dynamically changes the parameter of heart rate measurement reliability by monitoring PPG signal quality metrics. When signal quality parameters fall below acceptable thresholds, the system transitions from direct PPG-based measurement to model-based prediction, effectively adapting the measurement approach based on real-time signal conditions to maintain overall system reliability.
2Device complexity
If conventional frequency tracking is used for heart rate measurement, then the system remains simple and low-cost, but accuracy fails when PPG signals deteriorate during strenuous activities
Solution Approach 1:
The patent implements a dynamic heart rate monitoring system that adapts its measurement strategy based on real-time PPG signal quality assessment. The system continuously evaluates signal quality metrics and dynamically switches between direct frequency tracking (when signal is good) and prediction model-based estimation (when signal degrades). This dynamic adaptation maintains measurement precision across varying activity intensities without requiring permanently complex hardware.
Solution Approach 2:
The system performs preliminary training offline to build personalized prediction models using periods of high-quality PPG signals and corresponding kinematic data. During actual monitoring, when PPG signals deteriorate, the pre-trained model is already available to immediately provide accurate predictions without requiring real-time complex processing. This preliminary preparation enables accurate heart rate measurement during strenuous activities while keeping real-time computational requirements manageable.
3Duration of action of moving object
If heart rate monitoring continues over extended periods during athletic activities, then continuous data is collected, but reliability fails when PPG signals deteriorate over time
Solution Approach 1:
The patent prepares for future signal degradation by continuously training and updating personalized prediction models during periods of high-quality PPG signals. This beforehand cushioning ensures that when PPG signals inevitably deteriorate during extended athletic activities (due to sweat, movement, sensor displacement), the system already has robust prediction models ready to maintain reliable heart rate estimation. The continuous model improvement acts as a buffer against future reliability issues.
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
Accurately predicts heart rate during periods of low signal quality, providing reliable monitoring and improving HR-related statistics like VO2 max and recovery.
Implementation Method 1
heart rate (HR) assessment using low power optical sensors
Implementation Method 2
PPG signals can deteriorate to the extent that conventional frequency tracking is impossible
Data Source
AI summary
A device for determining a personalized heart rate, including one or more heart rate sensors configured to obtain heart rate values for a user for a time period, one or more movement sensors configured to obtain movement values for the user for the time period, one or more processors configured to execute a personalized heart rate prediction model, trained on user-specific movement data and user-specific heart rate data associated with signal qualities greater than the signal quality threshold, to receive the heart rate values and the activity type or the movement values, and generate, based on the heart rate values and the activity type or the movement values, a user-specific predicted heart rate for at least a part of the time period, and a user interface configured to output the user-specific predicted heart rate as the user heart rate for at least the part of the time period.