Smart Wearable Heart Rate Detection with Neural-Frequency Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional heart rate detection methods using smart wearable devices suffer from low accuracy due to noise introduced by factors such as ambient light, baseline drift, and workout artifacts, leading to signal distortion.
Innovation Solution
A heart rate detection method that combines a deep sequence neural network and a frequency tracking algorithm to fuse heart rate data, using weighted summation to improve accuracy during workouts by compensating for sudden changes in heart rate.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a PPG sensor is used to detect heart rate in smart wearable devices, then the measurement can be performed continuously and non-invasively, but the detection accuracy deteriorates due to noise from ambient light, baseline drift, and workout artifacts
Solution Approach 1:
The patent segments the heart rate detection process into two distinct algorithmic approaches: a deep sequence neural network for general heart rate prediction and a frequency tracking algorithm for rapid heart rate change detection. By dividing the detection task into multiple specialized components, the system can address different noise conditions and signal characteristics separately, thereby improving overall measurement precision while maintaining continuous monitoring capability
Solution Approach 2:
The patent merges the outputs of two different algorithm models through a fusion mechanism. The deep sequence neural network provides robust heart rate prediction under normal conditions, while the frequency tracking algorithm quickly captures sudden heart rate changes. By combining these complementary approaches, the system achieves high accuracy across diverse workout scenarios, resolving the contradiction between continuous monitoring and measurement precision
2Measurement precision
If a deep sequence neural network model is used for heart rate prediction, then the prediction accuracy is improved under normal conditions, but the response speed deteriorates when sudden heart rate changes occur
Solution Approach 1:
The patent introduces a frequency tracking algorithm as an intermediary mechanism that acts as a complementary detector for rapid heart rate changes. While the deep sequence neural network processes general heart rate prediction, the frequency tracking algorithm specifically monitors for sudden changes and provides quick updates. This intermediary approach allows the system to maintain high prediction accuracy while significantly improving response speed to acute cardiac events
Solution Approach 2:
The patent implements a dynamic algorithm selection and fusion mechanism that adapts to real-time signal characteristics. When the signal shows signs of rapid heart rate change, the system dynamically increases reliance on the frequency tracking algorithm, which is optimized for speed. Under normal conditions, the system relies more on the deep sequence neural network for accurate prediction. This dynamic adjustment resolves the contradiction between prediction accuracy and response speed
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
Enhances heart rate prediction accuracy by leveraging the quick tracking capability of the frequency tracking algorithm to complement the deep sequence neural network, particularly in scenarios with sudden heart rate changes.
Implementation Method 1
a photoplethysmography PPG sensor... The LED lamp continuously projects light onto the skin, and the light passes through the skin tissue and is absorbed by the blood flow. At the same time, the photosensor receives reflected light signals.
Data Source
AI summary
This application provides a heart rate detection method and an electronic device. Through the solution, when it is detected that a user wearing a smart wearable device is making a first workout, a PPG signal is acquired through a PPG sensor in the smart wearable device; first heart rate data is obtained based on the PPG signal and a first deep sequence neural network model; second heart rate data is obtained based on the PPG signal and a first frequency tracking algorithm model; and the first heart rate data and the second heart rate data are fused to obtain a target heart rate of the user. Because a frequency tracking algorithm can quickly track a heart rate change, thus the solution can compensate for a scenario in which a deep sequence neural network cannot implement timely tracking in the case of a sudden change of heart rate.