Method and system for using artificial intelligence to determine a user's progress during interval training
Inventors
Bissonnette, Michael • Berga, Luis • Mason, Steven • Powers, Philip • Steidl, James D.
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12102878-B2
Publication Date
2024-10-01
Expiration Date
Abstract
A method is disclosed for using an artificial intelligence engine to perform a control action. The control action is based on measurements from a wearable device. The method includes generating, by the artificial intelligence engine, a machine learning model trained to receive the measurements as input, and outputting, based on the measurements, a control instruction that causes the control action to be performed. The method includes receiving the measurements from the wearable device being worn by a user, determining whether the measurements indicate, during an interval training session, that one or more characteristics of the user are within a desired target zone, and responsive to determining that the measurements indicate the one or more characteristics of the user are not within the desired target zone during the interval training session, performing the control action.
Core Innovation
The invention uses an artificial intelligence engine to generate a machine learning model trained to receive one or more measurements as input and output, based on the one or more measurements, a control instruction that causes a control action to be performed. The one or more measurements are received from a wearable device being worn by a user, and the control action is generated in connection with an interval training session.
The invention determines, based on a desired physical activity to be performed subsequent to completing an interval training session, a desired target zone. Determining the desired target zone includes determining one or more attainment levels for physical abilities of the user for the user to be able to perform the desired physical activity, and determining whether the one or more measurements indicate, during the interval training session, that one or more characteristics of the user are within the desired target zone.
Responsive to determining that the one or more measurements indicate the one or more characteristics of the user are not within the desired target zone during the interval training session, the invention performs a control action. The control action comprises updating a resistance force of a cycling machine, and the invention also receives data associated with the user and predicts, via the machine learning model, a medical condition associated with the user based on the data, the one or more measurements, and a history of user conditions identified based on prior interval training sessions or data associated with prior users.
Claims Coverage
Independent claims are provided in three formats, all focused on an AI-engine-generated machine learning model that drives control instructions during interval training and predicts a medical condition using user data and prior-session history. Across the independent claims, five inventive features are present.
AI engine machine learning model for control instructions
Generating, by the artificial intelligence engine, a machine learning model trained to receive the one or more measurements as input; outputting, based on the one or more measurements, a control instruction that causes the control action to be performed.
Wearable measurement based interval training feedback loop
Receiving the one or more measurements from a wearable device being worn by a user; determining whether the one or more measurements indicate, during the interval training session, that one or more characteristics of the user are within the desired target zone.
Desired target zone with attainment levels for a subsequent physical activity
Determining, based on a desired physical activity to be performed subsequent to completing an interval training session, a desired target zone, wherein determining the desired target zone includes determining one or more attainment levels for physical abilities of the user for the user to be able to perform the desired physical activity.
Conditional cycling machine resistance force update when outside the target zone
Responsive to determining that the one or more measurements indicate the one or more characteristics of the user are not within the desired target zone during the interval training session, performing the control action, wherein the control action comprises updating a resistance force of a cycling machine.
Medical condition prediction from user data and prior session history
Receiving data associated with the user; based on the data, the one or more measurements of the user, and a history of user conditions identified based on one or more prior interval training sessions undertaken by the user or data associated with prior users, predicting, via the machine learning model, a medical condition associated with the user.
The independent claims cover an AI-engine-based machine learning model that uses wearable device measurements during interval training to determine a desired target zone, compare measured characteristics to that target zone, conditionally update cycling-machine resistance when the user is not within the zone, and predict a medical condition using current user data and historical information from prior interval training sessions or prior users.
Stated Advantages
Documented Applications
No documented applications found
Interested in licensing this patent?