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.

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.

Methods and systems are provided for automatically diagnosing an electrocardiogram (ECG) using a hybrid system comprising a rule-based system and one or more deep neural networks. In one embodiment, by mapping ECG data to a plurality of features using a convolutional neural network, mapping the plurality of features to a preliminary diagnosis using a decision network, and determining a diagnosis based on the ECG data and the preliminary diagnosis using the rule-based system, a more accurate diagnosis may be determined. In another example, by incorporating both a rule-based system and one or more deep neural networks into the hybrid system, the hybrid system may be more easily adapted for use in various contexts/communities, as the one or more deep learning networks may be trained using context/community specific ECG data.

Methods and systems are provided for automatically diagnosing an electrocardiogram (ECG) using a hybrid system comprising a rule-based system and one or more deep neural networks. In one embodiment, by mapping ECG data to a plurality of features using a convolutional neural network, mapping the plurality of features to a preliminary diagnosis using a decision network, and determining a diagnosis based on the ECG data and the preliminary diagnosis using the rule-based system, a more accurate diagnosis may be determined. In another example, by incorporating both a rule-based system and one or more deep neural networks into the hybrid system, the hybrid system may be more easily adapted for use in various contexts/communities, as the one or more deep learning networks may be trained using context/community specific ECG data.

Provided is a biofeedback device using an electrocardiogram, and the biofeedback device using an electrocardiogram may include: a signal acquisition unit acquiring an electrocardiogram signal of a user from an electrocardiograph; an analysis unit analyzing the acquired electrocardiogram signal and providing a state diagnosis result of the user as an analysis result; and a control unit controlling at least one stimulation unit among a plurality of stimulation units for state control of the user based on the state diagnosis result, and a stimulation corresponding to the at least one stimulation unit may be provided to the user by the control of the control unit.

The present invention relates to a method to provide a mean temporal spatial isochrone (TSI) path relating to an ECG feature (wave form) of interest, such as the activation of the heart from a single point (QRS), relative to the heart in a torso while using an ECG measurement from an ECG recording device. The method includes: receiving ECG measuring data from the ECG recording device; determining vector cardiogram (VCG) data; receiving a model of the heart, preferably with torso, as an input, preferably based on a request including request parameters; determining mean TSI data values representing the TSI path relating to an electrophysiological phase representing the ECG feature, the mean TSI providing a location within the heart representing the mean location of the ECG feature at the corresponding time; positioning the mean TSI path and preferably the vector cardiogram data points in the model of the heart and/or torso at an initial position; and rendering the model of the heart, preferably with torso, with the mean TSI path, preferably with VCG data related to the TSI, for displaying on a display screen for interpretation of the displayed rendering.

The present invention relates to a method to provide a mean temporal spatial isochrone (TSI) path relating to an ECG feature (wave form) of interest, such as the activation of the heart from a single point (QRS), relative to the heart in a torso while using an ECG measurement from an ECG recording device. The method includes: receiving ECG measuring data from the ECG recording device; determining vector cardiogram (VCG) data; receiving a model of the heart, preferably with torso, as an input, preferably based on a request including request parameters; determining mean TSI data values representing the TSI path relating to an electrophysiological phase representing the ECG feature, the mean TSI providing a location within the heart representing the mean location of the ECG feature at the corresponding time; positioning the mean TSI path and preferably the vector cardiogram data points in the model of the heart and/or torso at an initial position; and rendering the model of the heart, preferably with torso, with the mean TSI path, preferably with VCG data related to the TSI, for displaying on a display screen for interpretation of the displayed rendering.

A computer system for use in monitoring blood glucose level monitoring, the computer system configured, in response to receiving electrocardiogram data measured over a given period of time for a given subject, to classify the electrocardiogram data using at least one neural network and a personalised model which is specific to the given subject so as to identify whether a low blood glucose level condition is present wherein blood glucose level falls below a predefined level and, upon identifying the presence of the low blood glucose level condition, to flag an alarm condition.

Systems, methods, devices, and techniques for estimating a heart disease prediction of a mammal. An electrocardiogram (ECG) procedure is performed on a mammal, and a computer system obtains ECG data that describes results of the ECG over a period of time. The system provides a predictive input that is based on the ECG data to a predictive model, such as a neural network or other machine-learning model. In response, the predictive model processes the input to generate an estimated heart disease predictive characteristic of the mammal. The system outputs the estimated heart disease prediction of the mammal for presentation to a user.

Systems, methods, devices, and techniques for estimating a heart disease prediction of a mammal. An electrocardiogram (ECG) procedure is performed on a mammal, and a computer system obtains ECG data that describes results of the ECG over a period of time. The system provides a predictive input that is based on the ECG data to a predictive model, such as a neural network or other machine-learning model. In response, the predictive model processes the input to generate an estimated heart disease predictive characteristic of the mammal. The system outputs the estimated heart disease prediction of the mammal for presentation to a user.