Arrhythmia detection with feature delineation and machine learning
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.
1. A device comprising processing circuitry and a storage medium, wherein the processing circuitry is configured to:
apply a machine learning model, trained using electrocardiogram (ECG) data for a plurality of patients, to sensed ECG data of a patient to determine, based on the machine learning model, that an episode of arrhythmia has occurred in the patient;
perform feature-based delineation of the sensed ECG data to obtain cardiac features present in the sensed ECG data; and
in response to determining that the episode of arrhythmia has occurred in the patient, generate data comprising an indication that the episode of arrhythmia has occurred in the patient and one or more of the cardiac features that coincide with the episode of arrhythmia.
2. The device of claim 1 , wherein, to perform feature-based delineation, the processing circuitry is configured to perform QRS detection.
3. The device of claim 1 , wherein the cardiac features present in the sensed ECG data comprise one or more of a mean heartrate of the patient, a minimum heartrate of the patient, a maximum heartrate of the patient, a variability of heartrate of the patient, one or more amplitudes of one or more features of the sensed ECG of the patient, or an interval between the or more features of the ECG of the patient.
4. The device of claim 1 , wherein the processing circuitry is configured to apply the machine learning model to the sensed ECG data of the patient to determine, based on the machine learning model, that that an episode of at least one of bradycardia, tachycardia, or atrial fibrillation has occurred in the patient.
5. The device of claim 1 , wherein the processing circuitry is further configured to:
receive an adjustment to the feature-based delineation of the ECG data in response to the data comprising the indication that the episode of arrhythmia has occurred in the patient and one or more of the cardiac features that coincide with the episode of arrhythmia; and
perform, in accordance with the adjustment, feature-based delineation of the ECG data to obtain second cardiac features present in the ECG data.
6. The device of claim 1 , wherein, to generate the data comprising the indication that the episode of arrhythmia has occurred in the patient and one or more of the cardiac features that coincide with the episode of arrhythmia, the processing circuitry is configured to:
identify a subsection of the ECG of the patient, wherein the subsection comprises ECG data for a first time period prior to the episode of arrhythmia, a second time period during the episode of arrhythmia, and a third time period after the episode of arrhythmia, and wherein a length of time of the ECG of the patient is greater than the first, second, and third time periods;
identify one or more of the cardiac features that coincide with the first, second, and third time periods; and
include, in the data, the subsection of the ECG and the one or more of the cardiac features that coincide with the first, second, and third time periods.
7. The device of claim 1 , wherein the processing circuitry is configured to:
process the sensed ECG data to generate an intermediate representation of the sensed ECG data; and
to apply the machine learning model, apply a machine learning model, trained using intermediate representations of ECG data for a plurality of patients, to the intermediate representation of the sensed ECG data and the cardiac features present in the sensed ECG data to determine, based on the machine learning model, that the episode of arrhythmia has occurred in the patient.
8. The device of claim 1 , wherein the processing circuitry is configured to:
determine a first classification of arrhythmia in the patient based on the feature-based delineation of the sensed ECG data;
determine a second classification of arrhythmia based on the application of the machine learning model to the sensed ECG data; and
determine that the episode of arrhythmia occurred in the patient based on the first classification and second classification.
9. The device of claim 8 , wherein, to determine that the episode of arrhythmia has occurred in the patient based on the first classification and second classification, the processing circuitry is configured to:
determine a degree of similarity of the first classification and the second classification; and
based on the degree of similarity of the first classification and the second classification, determine that the episode of arrhythmia has occurred in the patient.
10. The device of claim 8 , wherein, to apply the machine learning model to the sensed ECG data, the processing circuitry is configured to apply the machine learning model to the cardiac features identified by the feature-based delineation of the sensed ECG data.
11. The device of claim 1 , wherein the data comprising the indication that the episode of arrhythmia has occurred in the patient and one or more of the cardiac features that coincide with the episode of arrhythmia comprises at least one of an estimate of the certainty or likelihood that the episode of arrhythmia has occurred in the patient.
12. The device of claim 1 , wherein the device comprises a cloud computing device.
13. A system comprising:
an implantable cardiac monitoring device configured to:
sense electrocardiogram (ECG) data of a patient; and
perform feature-based delineation of the sensed ECG data to obtain cardiac features present in the sensed ECG data; and
a computing device configured to:
apply a machine learning model, trained using ECG data for a plurality of patients, to the sensed ECG data of the patient to determine, based on the machine learning model, that an episode of arrhythmia has occurred in the patient;
in response to determining that the episode of arrhythmia has occurred in the patient, generate data comprising an indication that the episode of arrhythmia has occurred in the patient and one or more of the cardiac features that coincide with the episode of arrhythmia.
14. The system of claim 13 , wherein, to perform feature-based delineation, the implantable cardiac monitoring device is configured to perform QRS detection.
15. The system of claim 13 , wherein the cardiac features present in the sensed ECG data comprise one or more of a mean heartrate of the patient, a minimum heartrate of the patient, a maximum heartrate of the patient, a variability of heartrate of the patient, one or more amplitudes of one or more features of the sensed ECG data of the patient, or an interval between the or more features of the sensed ECG of the patient.
16. The system of claim 13 , wherein the computing device is configured to apply the machine learning model to the sensed ECG data of the patient to determine, based on the machine learning model, that that an episode of at least one of bradycardia, tachycardia, or atrial fibrillation has occurred in the patient.
17. The system of claim 13 , wherein the computing device is configured to receive, from a user, an adjustment to the feature-based delineation of the cardiac electrogram data in response to the data comprising the indication that the episode of arrhythmia has occurred in the patient and one or more of the cardiac features that coincide with the episode of arrhythmia, and the implantable cardiac monitoring device is configured to perform, in accordance with the adjustment, feature-based delineation of the sensed ECG data to obtain second cardiac features present in the cardiac electrogram data.
18. The system of claim 13 , wherein, to generate the data comprising the indication that the episode of arrhythmia has occurred in the patient and one or more of the cardiac features that coincide with the episode of arrhythmia, the computing device is configured to:
identify a subsection of the ECG of the patient, wherein the subsection comprises ECG data for a first time period prior to the episode of arrhythmia, a second time period during the episode of arrhythmia, and a third time period after the episode of arrhythmia, and wherein a length of time of the ECG of the patient is greater than the first, second, and third time periods;
identify one or more of the cardiac features that coincide with the first, second, and third time periods; and
include, in the data, the subsection of the ECG and the one or more of the cardiac features that coincide with the first, second, and third time periods.
19. The system of claim 13 , wherein the computing device is configured to:
process the sensed ECG data to generate an intermediate representation of the sensed ECG data; and
to apply the machine learning model, apply a machine learning model, trained using intermediate representations of ECG data for a plurality of patients, to the intermediate representation of the sensed ECG data and the cardiac features present in the sensed ECG data to determine, based on the machine learning model, that the episode of arrhythmia has occurred in the patient.
20. The system of claim 13 , wherein the implantable cardiac monitoring device is configured to determine a first classification of arrhythmia in the patient based on the feature-based delineation of the sensed ECG data, and the computing device is configured to:
determine a second classification of arrhythmia based on the application of the machine learning model to the sensed ECG data; and
determine that the episode of arrhythmia occurred in the patient based on the first classification and second classification.
21. The system of claim 20 , wherein, to determine that the episode of arrhythmia has occurred in the patient based on the first classification and second classification, the computing device is configured to:
determine a degree of similarity of the first classification and the second classification; and
based on the degree of similarity of the first classification and the second classification, determine that the episode of arrhythmia has occurred in the patient.
22. The system of claim 20 , wherein, to determine the first classification of arrhythmia in the patient based on the feature-based delineation of the sensed ECG data the implantable cardiac monitoring device is configured to:
extract one or more features from the sensed ECG based on the feature-based delineation; and
apply a model to the one or more extracted features.
23. The system of claim 13 , wherein, to apply the machine learning model to the received cardiac electrogram data, the computing device is configured to apply the machine learning model to the cardiac features identified by the feature-based delineation of the received cardiac electrogram data.
24. The system of claim 13 , wherein the data comprising the indication that the episode of arrhythmia has occurred in the patient and one or more of the cardiac features that coincide with the episode of arrhythmia comprises at least one of an estimate of the certainty or likelihood that the episode of arrhythmia has occurred in the patient.
25. The system of claim 13 , wherein the computing device comprises a cloud computing device.
26. A non-transitory computer-readable storage medium comprising instructions that, when executed by processing circuitry, cause the processing circuitry to:
apply a machine learning model, trained using electrocardiogram (ECG) data for a plurality of patients, to sensed ECG data of a patient to determine, based on the machine learning model, that an episode of arrhythmia has occurred in the patient;
perform feature-based delineation of the sensed ECG data to obtain cardiac features present in the sensed ECG data; and
in response to determining that the episode of arrhythmia has occurred in the patient, generate data comprising an indication that the episode of arrhythmia has occurred in the patient and one or more of the cardiac features that coincide with the episode of arrhythmia.
27. The non-transitory computer-readable storage medium of claim 26 , wherein the instructions that cause the processing circuitry to perform feature-based delineation comprise instructions that cause the processing circuitry to perform QRS detection.
28. The non-transitory computer-readable storage medium of claim 26 , wherein the cardiac features present in the sensed ECG data comprise one or more of a mean heartrate of the patient, a minimum heartrate of the patient, a maximum heartrate of the patient, a variability of heartrate of the patient, one or more amplitudes of one or more features of the sensed ECG of the patient, or an interval between the or more features of the sensed ECG of the patient.
29. The non-transitory computer-readable storage medium of claim 26 , wherein the instructions that cause the processing circuitry to apply the machine learning model to the sensed ECG data of the patient comprise instructions that cause the processing circuitry to determine, based on the machine learning model, that that an episode of at least one of bradycardia, tachycardia, or atrial fibrillation has occurred in the patient.
30. A system comprising:
processing circuitry configured to:
apply a machine learning model, trained using ECG data for a plurality of patients, to sensed ECG data of the patient to determine, based on the machine learning model, that an episode of arrhythmia has occurred in the patient; and
perform feature-based delineation of the sensed ECG data to obtain cardiac features present in the sensed ECG data; and
means for generating, in response to the determination that the episode of arrhythmia has occurred in the patient, data comprising an indication that the episode of arrhythmia has occurred in the patient and one or more of the cardiac features that coincide with the episode of arrhythmia.