Techniques that include applying machine learning models to episode data, including a cardiac electrogram, stored by a medical device are disclosed. In some examples, based on the application of one or more machine learning models to the episode data, processing circuitry derives, for each of a plurality of arrhythmia type classifications, class activation data indicating varying likelihoods of the classification over a period of time associated with the episode. The processing circuitry may display a graph of the varying likelihoods of the arrhythmia type classifications over the period of time. In some examples, processing circuitry may use arrhythmia type likelihoods and depolarization likelihoods to identify depolarizations, e.g., QRS complexes, during the episode.

A method and device for dynamic device based AV delay adjustment are provided. The method provides electrodes that are configured to be located proximate to an atrial (A) site and a right ventricular (RV) site. The method utilizes one or more processors, in an implantable medical device (IMD), for detecting an atrial paced (Ap) event or atrial sensed (As) event. The method determines a measured AV interval corresponding to an interval between the Ap event or the As event and a ventricular sensed event and calculates a percentage-based (PB) offset based on the measured AV interval. The method automatically dynamically adjusting an AV delay, utilized by the IMD, based on the measured AV interval and the PB offset and manages a pacing therapy, utilized by the IMD, based on the AV delay after the adjusting operation.

Devices and methods for improving device therapy such as cardiac resynchronization therapy by determining a value for a device parameter are described. An ambulatory medical device (AMD) can include a sensor circuit to sense a physiological signal and generate two or more signal metrics, and detect an event of worsening cardiac condition using the two or more signal metrics. In response to the detection of worsening cardiac condition, the AMD can determine, for a stimulator, a value of at least one stimulation parameter based on temporal responses of two or more signal metrics. The temporal responses include near-term and long-term responses to the stimulation. The AMD can program the stimulator with the determined parameter value, and generate stimulation according to the determined parameter value to stimulate target tissue.

A computer implemented method and device for providing dual chamber sensing with a single chamber leadless implantable medical device (LIMD) are provided. The method is under control of one or more processors in the LIMD configured with specific executable instructions. The method obtains a far field (FF) cardiac activity (CA) signals for activity in a remote chamber of a heart and compares the far field CA signals to a P-wave template to identify an event of interest associated with the remote chamber. The method sets an atrial-ventricular (AV) delay based on the P-wave identified and delivers pacing pulses at a pacing site of interest to a local chamber based on the AV delay.

Techniques are disclosed for using both feature delineation and machine learning to detect cardiac arrhythmia. A computing device receives cardiac electrogram data of a patient sensed by a medical device. The computing device obtains, via feature-based delineation of the cardiac electrogram data, a first classification of arrhythmia in the patient. The computing device applies a machine learning model to the received cardiac electrogram data to obtain a second classification of arrhythmia in the patient. As one example, the computing device uses the first and second classifications to determine whether an episode of arrhythmia has occurred in the patient. As another example, the computing device uses the second classification to verify the first classification of arrhythmia in the patient. The computing device outputs a report indicating that the episode of arrhythmia has occurred and one or more cardiac features that coincide with the episode of arrhythmia.

Techniques are disclosed for using both feature delineation and machine learning to detect cardiac arrhythmia. A computing device receives cardiac electrogram data of a patient sensed by a medical device. The computing device obtains, via feature-based delineation of the cardiac electrogram data, a first classification of arrhythmia in the patient. The computing device applies a machine learning model to the received cardiac electrogram data to obtain a second classification of arrhythmia in the patient. As one example, the computing device uses the first and second classifications to determine whether an episode of arrhythmia has occurred in the patient. As another example, the computing device uses the second classification to verify the first classification of arrhythmia in the patient. The computing device outputs a report indicating that the episode of arrhythmia has occurred and one or more cardiac features that coincide with the episode of arrhythmia.

Techniques that include applying machine learning models to episode data, including a cardiac electrogram, stored by a medical device are disclosed. In some examples, based on the application of one or more machine learning models to the episode data, processing circuitry derives, for each of a plurality of arrhythmia type classifications, class activation data indicating varying likelihoods of the classification over a period of time associated with the episode. The processing circuitry may display a graph of the varying likelihoods of the arrhythmia type classifications over the period of time. In some examples, processing circuitry may use arrhythmia type likelihoods and depolarization likelihoods to identify depolarizations, e.g., QRS complexes, during the episode.

A method and device for dynamic device based AV delay adjustment is provided. The method comprises electrodes that are configured to be located proximate to an atrial (A) site and a right ventricular (RV) site. The method utilizes one or more processors for detecting an atrial paced (Ap) event or atrial sensed (As) event, and measures an AV interval corresponding to an interval between the Ap event or the As event and a sensed ventricular (Vs) event. The AV interval is associated with a current heart rate (HR). The method automatically dynamically adjusts a first AV delay based directly on the measured AV interval, identifies a scale factor associated with the current HR, calculates a second AV delay by scaling the first AV delay based on the scale factor and manages a pacing therapy, utilized by the IMD, based on the first and second AV delays.

This document discusses, among other things, systems and methods to detect an initial arrhythmia event indication and, after a threshold amount of detection window intervals detecting the initial arrhythmia event indication, adjust a set of arrhythmia parameters or at least one of a respective set of parameter thresholds to increase sensitivity of an extended arrhythmia event indication detection.

Techniques are disclosed for monitoring a patient for the occurrence of a cardiac arrhythmia. A computing system generates sample probability values by applying a machine learning model to sample patient data. The machine learning model determines a respective probability value that indicates a probability that the cardiac arrhythmia occurred during each respective temporal window. The computing system outputs a user interface comprising graphical data based on the sample probability values and receives, via the user interface, an indication of user input to select a probability threshold for a patient. The computing system receives patient data for the patient and applies the machine learning model to the patient data to determine a current probability value. In response to the determination that the current probability exceeds the probability threshold for the patient, the computing system generates an alert indicating the patient has likely experienced the occurrence of the cardiac arrhythmia.