Machine Learning Functional Threshold Power Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies lack effective methods for accurately predicting functional threshold power (FTP) and generating personalized training power zones for users, especially on devices with limited computational resources, using machine learning techniques that integrate various physiological and activity data sources.
Innovation Solution
The implementation of a machine learning system that utilizes neural networks and physiological state equations to predict FTP by analyzing user activity data, demographic information, and sensor data from devices like heart rate monitors and power meters, allowing for the generation of training power zones tailored to individual fitness levels and goals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machine learning techniques are used to predict FTP and generate personalized training zones, then prediction accuracy and personalization improve, but device complexity and computational resource requirements increase
Solution Approach 1:
The patent segments the machine learning model into multiple components: a training phase that processes large datasets offline to create pre-trained models, and an inference phase that runs on the wearable device using the pre-trained model. This segmentation allows complex computational work to be done during training while keeping the runtime device requirements manageable.
Solution Approach 2:
The patent performs preliminary actions by pre-training the machine learning model offline using extensive physiological and activity data before deployment. The pre-trained model parameters and weights are stored on the wearable device, enabling accurate FTP predictions without requiring the device to perform complex training computations in real-time.
2Reliability
If multiple data sources (physiological measurements, activity data, sensor data) are integrated for FTP prediction, then prediction reliability improves, but data processing complexity increases
Solution Approach 1:
The patent merges multiple data sources including physiological measurements (heart rate, blood oxygen), activity data (workout intensity, duration), and sensor data into a unified machine learning model. The model integrates these diverse inputs to produce a comprehensive FTP prediction that leverages the strengths of each data type.
Solution Approach 2:
The machine learning model is designed as a universal system that can process multiple types of inputs (physiological data, activity data, sensor data) through a single integrated framework. This multi-functional approach allows the same model to handle various data sources without requiring separate processing pipelines for each type.
Data Source
AI summary
The subject technology provides a framework for generating physiological predictions for a user of an electronic device. The physiological predictions may include user-specific predictions of functional threshold power (FTP) that may occur if the user engages in a future activity, such as a future workout. The FTP predictions may be generated by a machine learning model that utilizes user-specific and user-agnostic physiological measures such as VO2 max to generate multiple FTP estimates according to different physiological approaches in calculating the FTP estimates and arbitrating between the FTP estimates to make the prediction with the best FTP estimate.