Systems and methods of QT interval analysis

Patent metadata

Field	Value
Publication number	US-12097035-B2
Application number	US-201917054130-A
Country	US
Kind code	B2
Filing date	May 8, 2019
Priority date	May 8, 2018
Publication date	Sep 24, 2024
Grant date	Sep 24, 2024

Abstract

Disclosed systems and method receive electrocardiogram (EKG) data of a subject. The EKG data comprises a record of at least a full beat of the subject. The EKG data is input into a machine learning model to generate an output. The output can include a segmentation of the full beat or a measurement of a QT interval of the subject.

First claim

What is claimed is:

1. A method comprising:
inputting each of a set of training electrocardiogram (EKG) measurements into a machine learning model to generate an output comprising an estimated start time of a Q-wave and an estimated end time of a T-wave for each of the set of training EKG measurements;
comparing each of the set of outputs to a corresponding labeled start time of a Q-wave and a labeled end time of a T-wave and adjusting one or more weight matrices of the machine learning model based on the comparing to train the machine learning model;
receiving EKG data of a subject, wherein the EKG data comprises at least a full beat of the subject; and
inputting, by a processing device, the EKG data into the machine learning model to generate a first output comprising a segmentation of the full beat, wherein the segmentation comprises at least a start time of a Q-wave and an end time of a T-wave of the EKG data.

2. The method of claim 1 , further comprising determining, by the processing device, whether a time difference between the start time of the Q-wave and the end time of the T-wave of the EKG data is greater than a threshold.

3. The method of claim 2 , further comprising generating an alert indicating that the time difference satisfies the threshold in response to determining that the time difference is greater than the threshold.

4. The method of claim 2 , further comprising determining a previous QT interval of the subject and generating the threshold based at least in part on the previous QT interval.

5. The method of claim 1 , wherein the machine learning model comprises a neural network comprising one or more of a convolutional layer or a recurrent layer.

6. The method of claim 2 , wherein the threshold is 500 milliseconds.

7. The method of claim 1 , wherein inputting the EKG data into the machine learning model further generates an output identifying times associated with a P-wave, an R-wave, and an S-wave.

8. The method of claim 1 , wherein the EKG data comprises a lead I measurement, a lead I and a lead II measurement, or lead I, lead II, lead III, lead aVR, lead aVL, and aVF measurements.

9. A system comprising:
a single lead electrocardiogram (EKG) sensor; and
a processing device communicably coupled to the single lead EKG sensor, wherein the processing device is to:
input each of a set of training electrocardiogram (EKG) measurements into a machine learning model to generate an output comprising an estimated start time of a Q-wave and an estimated end time of a T-wave for each of the set of training EKG measurements;
compare each of the set of outputs to a corresponding labeled start time of a Q-wave and a labeled end time of a T-wave and adjusting one or more weight matrices of the machine learning model based on the comparing to train the machine learning model;
receive EKG data from the single lead EKG sensor;
analyze, using the machine learning model, the EKG data to generate an output of a QT interval of the EKG data, the QT interval comprising a start time of a Q-wave and an end time of a T-wave of the EKG data;
determine that the QT interval satisfies a threshold QT interval value; and
generate an alert in response to determining that the QT interval satisfies the threshold value.

10. The system of claim 9 , wherein inputting the EKG data into the machine learning model further generates an output identifying timing measurements associated with a P-wave, an R-wave, and an S-wave.

11. The system of claim 9 , wherein
the QT interval corresponds to a difference between the start time of the Q-wave and the end time of the T-wave of the EKG data.

12. The system of claim 9 , wherein the processing device is further to determine a previous QT interval of the subject and generate the threshold based at least in part on the previous QT interval.

13. A non-transitory computer readable medium including instructions thereon that when executed by a processing device cause the processing device to:
receive a training set of EKGs associated with a plurality of subjects, wherein each of the training set of EKGs has labeled QT interval data;
for each EKG in the training set of EKGs:
input, by the processing device, the EKG into an untrained machine learning model to generate an estimated QT interval;
compare the estimated QT interval to the labeled QT interval data; and
update the untrained machine learning model based on the comparison; and

output the untrained machine learning model as a trained machine learning model in response to determining that the untrained machine learning model has converged within a threshold level of accuracy.

14. The non-transitory computer readable medium of claim 13 , wherein the labeled QT interval data comprises a first label indicating a start time of a Q-wave and a second label indicating an end time of a T-wave.

15. The non-transitory computer readable medium of claim 13 , wherein to compare the estimated QT interval to the labeled QT interval data, the processing device is further to:
compare an estimated Q-wave start time to a labeled Q-wave start time; and
compare an estimated T-wave end time to a labeled T-wave end time.

16. The non-transitory computer readable medium of claim 13 , wherein the processing device is further to:
receive a new EKG signal from a subject;
analyze the EKG signal to generate an output of a QT interval;
determine that the output of the QT interval satisfies a threshold value; and
generate an alert in response to determining that the output of the QT interval satisfies the threshold value.

17. The non-transitory computer readable medium of claim 16 , wherein the EKG signal was taken from a lead I measurement.

Assignees

• Alivecor Inc

Inventors

• Albert David E
• Valys Alexander Vainius
• Galloway Conner Daniel Cross

Classifications

• A61B5/349 Primary

  Detecting specific parameters of the electrocardiograph cycle · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06N3/0442

  characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] · CPC title

• G06N3/08

  Learning methods · CPC title