A computer implemented method and system for labeling types of heart arrhythmias based on cardiac activity are provided. The method is under control of one or more processors of an implantable medical device (IMD) configured with specific executable instruction. The method obtains cardiac activity (CA) signals at electrodes of the IMD during cardiac beats, declares a ventricular tachycardia (VT) episode based on the CA signals and obtains acceleration signatures, at an accelerometer of the IMD, indicative of heart sounds generated during the cardiac beats. The method analyzes an S1 characteristic of interest (COI) from the acceleration signature to identify the VT episode as a stable or non-stable VT episode and labels the VT episode as stable or non-stable based on the analyzing operation.

A system for cardiac monitoring and therapy includes a mother device configured to receive signals indicative of cardiac electrical activity in a patient's heart. The mother device includes a mother wireless communications module configured to transmit and receive information to and from the mother device. The system also includes a satellite device configured to receive the signals indicative of the cardiac electrical activity in the patient's heart from a remote location relative to the mother device and includes a satellite wireless communications module configured to transmit from and receive communications sent to the satellite device to at least communicate with the mother wireless communications module. The system also includes a processor configured to receive the signals indicative of the cardiac electrical activity in the heart received by the mother device and the satellite device and, based thereon, control delivery of electrical therapy to the patient's heart.

Conventionally, Atrial Fibrillation (AF) has been detected using atrial analyses which is vulnerable to background noise. Again there is a dependency on statistical features which are extracted from R-R intervals of long ECG recordings. The present disclosure addresses AF detection from single lead short ECG recordings of less than one minute wherein automatic detection of P-R and P-Q intervals is difficult, which introduces error in feature computing from the segregated intervals and compromises the performance of the classifier. In the present disclosure, a Recurrent Neural Network (RNN) based architecture comprising two Long Short Term Memory (LSTM) networks is provided for temporal analysis of R-R intervals and P wave regions in an ECG signal respectively. Output sates of the two LSTM networks are merged at a dense layer along with a set of hand-crafted statistical features to create a composite feature set for classification of the AF.

A method of detecting increasing fluid in a lung of a subject, the method comprising: at a first time, determining a first difference between a time of arrival of a first feature of a received electrical signal of a subject's heart beat and a time of arrival of a second feature of a received acoustic signal of the subject's heart beat; at a second later time, determining a second difference between a time of arrival of the first feature of a received electrical signal of a subject's subsequent heart beat and a time of arrival of the second feature of a received acoustic signal of the subject's subsequent heart beat; and if the second difference is less than the first difference by more than a threshold value, producing a fluid detection alert.

A method for accurately extracting an abnormal potential within a QRS, comprising: in an ideal electrocardiographic signal pre-estimation stage, pre-estimating an ideal electrocardiographic signal using a non-linear transformation technology;

according to the pre-estimated ideal electrocardiographic signal, further estimating the ideal electrocardiographic signal by using a spline method, so as to accurately estimate the ideal electrocardiographic signal; and according to the accurately estimated ideal electrocardiographic signal, accurately extracting an abnormal potential within the QRS by means of a mobile standard deviation analysis technology. The method can be used not only on an average electrocardiographic signal after multiple superimposition, but also on a single beat electrocardiographic signal.

A medical device includes at least one electrode to sense an electrocardiogram (ECG) signal of a patient, and a controller coupled to the at least one electrode. The controller is configured to generate a first ECG template based on a first ECG signal of the patient received during a first baselining operation. The controller is configured to determine that the patient has been administered a therapeutic shock, and responsive to the determination that the patient has been administered the therapeutic shock, the controller is configured to initiate a second baselining operation and generate a second ECG template based on a second ECG signal of the patient received during the second baselining operation. The controller is configured to determine whether the patient is experiencing a cardiac event based on a comparison of the second ECG template to a real time ECG signal received during real time monitoring of the patient.

Systems and methods of arrhythmia detection and associated apparatus that utilize machine learning techniques that allow for the consideration individual characteristics and the tailoring/personalization of biometric data allow for early detection and treatment, especially of cardiac arrhythmias and other abnormalities.

According to at least one example, an ambulatory medical device is provided. The device includes a plurality of electrodes disposed at spaced apart positions about a patient's body and a control unit. The control unit includes a sensor interface, a memory and a processor. The sensor interface is coupled to the plurality of electrodes and configured to receive a first ECG signal from a first pairing of the plurality of electrodes and to receive a second ECG signal from a second pairing of the plurality of electrodes. The memory stores information indicating a preferred pairing, the preferred pairing being either the first pairing or the second pairing. The processor is coupled to the sensor interface and the memory and is configured to resolve conflicts between interpretations of first ECG signal and the second ECG signal in favor of the preferred pairing.

The present disclosure relates to a wearable monitor device and methods and systems for using such a device. In certain embodiments, the wearable monitor records cardiac data from a mammal and extracts particular features of interest. These features are then transmitted and used to provide health-related information about the mammal.

A system and a method are provided for using a mobile computing system comprising that has a communication connection configured to communicate with an electrocardiographic (ECG) apparatus to acquire ECG signals from the subject through a plurality of ECG leads. The system includes a processor configured to receive the ECG signals through the communications connection and process the ECG signals to estimate at least one of a respiratory rate of the subject, a tidal volume of the subject, or an ischemic index or repolarization alternans of the subject, from the ECG signals. The processor is further configured to generate an alert upon determining at least one of the ischemic index or repolarization alternans is above a threshold value or a change in respiratory rate or tidal volume indicative of an abnormal respiratory event. The system also includes a display configured to display the alert.