Intelligently, continuously and physiologically controlled pacemaker and method of operation of the same
Inventors
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12427321-B2
Publication Date
2025-09-30
Expiration Date
Abstract
A pacemaker control system includes a pacemaker; a plurality of sensors which are internal to the pacemaker, a plurality of sensors which are external to the pacemaker, a circuit for entering patient reports; and a circuit for using artificial intelligence to process outputs from the plurality sensors internal and external to the pacemaker and from the circuit for entering patient reports, which are collectively identified as a labeled dataset, to reiteratsvely learn a function which determines the labeled dataset most likely to provide optimal pacemaker function for the patient. The means for using artificial intelligence comprises a database of archive outputs from the plurality sensors internal and external to the pacemaker and from the means for entering patient reports for the patient used for optimization of rate modulation to intelligently, continuously and physiologically control the pacemaker.
Core Innovation
The invention provides a system for cardiac pacing that uses a heart pacing device together with at least one memory and at least one processor executing instructions to perform operations with a machine learning model. The operations receive an ideal dataset comprising a plurality of ideal patient parameters for a patient, receive a plurality of archived datasets, and receive current physiological variables from one or more sensors.
The machine learning model identifies an optimal archived dataset from the plurality of archived datasets based on threshold-based matching, including a first threshold match between the current physiological variables and optimal historical physiological variables and a second threshold match between the ideal patient parameters and optimal historical patient parameters. In one variant, ideal patient blood pressure is used as the ideal input, and current physiological variables are received from a wearable sensor.
After the optimal archived dataset is identified, the machine learning model outputs new rate modulation parameters to the heart pacing device based on optimal historical rate modulation parameters of the identified optimal archived dataset. The new rate modulation parameters increase or decrease a heart rate of the patient, and the heart pacing device stimulates the heart of the patient at the new rate modulation parameters.
A corresponding method for cardiac pacing is also provided, where the machine learning model receives the ideal dataset and archived datasets, receives current physiological variables from sensors, identifies an optimal archived dataset using a first threshold match and a second threshold match, and outputs new rate modulation parameters to a heart pacing device. The claims further describe that the rate modulation parameters can include pacing-related parameters such as pacing rate and timing/trajectory components, and that sensors can include wearable sensors, cardiac pacing device sensors, or lead sensors.
Claims Coverage
Across the disclosed independent claims, the core coverage centers on an AI/machine-learning selection of an optimal archived dataset using threshold matches between current physiological variables and historical physiological variables and between ideal patient parameters, including ideal blood pressure in one claim, and historical patient parameters, including historical blood pressure. The selected optimal archived dataset then drives output of new rate modulation parameters that increase or decrease heart rate, followed by cardiac stimulation.
Threshold-based optimal archived dataset identification from current physiological variables and ideal patient parameters
Receiving an ideal dataset comprising a plurality of ideal patient parameters for a patient at a machine learning model; receiving a plurality of archived datasets at the machine learning model; receiving current physiological variables from one or more sensors; and identifying an optimal archived dataset based on a first threshold match between the current physiological variables and optimal historical physiological variables and a second threshold match between the ideal patient parameters and optimal historical patient parameters.
Output of new rate modulation parameters derived from the identified optimal archived dataset to adjust heart rate
Outputting, by the machine learning model, to the heart pacing device, new rate modulation parameters based on optimal historical rate modulation parameters of the identified optimal archived dataset, wherein the new rate modulation parameters increase or decrease a heart rate of the patient, and stimulating the heart of the patient at the new rate modulation parameters.
Ideal patient blood pressure threshold matching using wearable-sensor current physiological variables
Receiving an ideal patient blood pressure at a machine learning model; receiving a plurality of archived datasets that comprise historical physiological variables, corresponding historical blood pressure values, or corresponding historical rate modulation parameters; receiving current physiological variables from a wearable sensor; and identifying an optimal archived dataset based on a first threshold match between the current physiological variables and optimal historical physiological variables and a second threshold match between the ideal blood pressure and optimal historical blood pressure.
Method for cardiac pacing using threshold-based optimal archived dataset identification and subsequent stimulation
Receiving an ideal dataset, receiving archived datasets, receiving current physiological variables from sensors, identifying an optimal archived dataset using first and second threshold matches, outputting new rate modulation parameters to a heart pacing device, and stimulating the heart at the new rate modulation parameters.
The claim set is primarily structured around a machine learning model that selects an optimal archived dataset using first and second threshold matches between current physiological variables and historical physiological variables and between ideal patient parameters, including ideal blood pressure in one independent claim, and historical patient parameters. The system or method then outputs new rate modulation parameters derived from the identified optimal archived dataset to adjust heart rate and stimulates the heart accordingly.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Not explicitly described in patent.
Interested in licensing this patent?