US-12622650-B2 - System and a method for electrocardiographic prediction of computed tomography-based high coronary calcium score (CAC)

Abstract

Provided herein are methods, systems, and computer program products for the detection and evaluation of coronary artery calcium (CAC) (e.g., CAC scoring) comprising receiving voltage-time data of a plurality of leads of an electrocardiograph subject; generating a feature vector from the voltage-time data; providing the feature vector to a pretrained learning system; and receiving from the pretrained learning system an indication of the level of coronary artery calcium in the subject.

Inventors

• Itzhak Zachi Attia

Assignees

• MAYO FOUNDATION FOR MEDICAL EDUCATION AND RESEARCH

Dates

Publication Date

20260512

Application Date

20240422

Claims (20)

1. 1 . A method for electrocardiographic prediction of computed tomography-based high coronary calcium score, wherein the method comprises: receiving, using at least a processor, voltage-time data of a subject, wherein the voltage-time data comprises image data of voltage data from a plurality of leads of an electrocardiograph characterizes electrocardiogram (ECG) signals; generating, using the at least a processor, a feature vector as a function of the image data of the voltage-time data, wherein the feature vector comprises one or more time-series of values indicating an amplitude of an for the plurality of leads; and identifying, using the at least a processor, an indication of a level of coronary artery calcium (CAC) in the subject using a learning system comprising a convolutional neural network (CNN) by: receiving a training set of a plurality of voltage-time data from a plurality of CAC patients, wherein the plurality of voltage-time data comprises a plurality of image data characterizing a plurality of ECG signals from the plurality of CAC patients correlated to indications comprising a probability value, indicating a likelihood of the level of CAC in a given patient; generating, automatedly, a training matrix by converting pixels in the training set representative of each ECG of the plurality of image data of the plurality of voltage-time data to associated training matrix elements to form a training matrix of a predetermined size, wherein a first dimension of the training matrix represents spatial leads and a second dimension of the training matrix represents a time series of a predetermined time duration at a predetermined frequency; training the CNN using the automatedly generated training matrix, wherein training the CNN comprises: feeding the training matrix, representative of the training set, to an input layer of the CNN; tiling the training matrix at the input layer into a plurality of tiles representative of the training matrix; applying learnable kernels of parameters in at least one convolutional layer of the CNN across pixel data in the plurality of tiles which is representative of the plurality of image data characterizing the plurality of ECG signals in the training set, wherein a predetermined number of parameters in the at least one convolutional layer are learned irrespective of image size; and populating an output layer of the CNN based on results generated by application of the learnable kernels to the pixel data; updating the feature vector as a function of demographic data associated with the subject and genomic data associated with the subject; inputting, to the trained CNN, the updated feature vector representative of the subject's image data, the subject's demographic data and the subject's genomic data; and outputting, from the trained CNN, as a function of the updated feature vector, the indication of the level of CAC in the subject.
2. 2 . The method of claim 1 , wherein generating the feature vector comprises generating a spectrogram as a function of the voltage data.
3. 3 . The method of claim 1 , wherein generating the feature vector comprises grouping the voltage data from each lead of the plurality of leads into a plurality of subsets.
4. 4 . The method of claim 1 , wherein the method further comprises: receiving, using the at least a processor, the demographic data associated with the subject; and generating, using the at least a processor, the feature vector as a function of the demographic data and the voltage-time data.
5. 5 . The method of claim 1 , wherein the method further comprises: receiving, using the at least a processor, the genomic data associated with the subject; and generating, using the at least a processor, the feature vector as a function of the genomic data and the voltage-time data.
6. 6 . The method of claim 1 , wherein the voltage-time data of the subject is received from an electronic medical record.
7. 7 . The method of claim 1 , wherein the feature vector comprises a matrix, wherein the matrix comprises: a plurality of rows corresponding to a temporal dimension; and a plurality of columns corresponding to a spatial dimension.
8. 8 . The method of claim 7 , wherein each row of the plurality of rows corresponds to at least one lead of the plurality of leads.
9. 9 . The method of claim 7 , wherein each column of the plurality of columns corresponds to at least one timestamp.
10. 10 . The method of claim 7 , wherein the temporal dimension has a resolution of 500 Hz.
11. 11 . A system for electrocardiographic prediction of computed tomography-based high coronary calcium score (CAC), wherein the system comprises: at least a processor; a computer readable storage medium communicatively connected to the at least a processor, wherein the computer readable storage medium contains instructions configuring the at least a processor: receive voltage-time data of a subject, wherein the voltage-time data comprises image data of voltage data from a plurality of leads of an electrocardiograph and characterizes electrocardiogram (ECG) signals; generate a feature vector as a function of the image data of the voltage-time data, wherein the feature vector comprises one or more time-series of values indicating an amplitude of an ECG for the plurality of leads; and identify an indication of a level of coronary artery calcium (CAC) in the subject using a learning system comprising a convolutional neural network (CNN) by: receiving a training set of a plurality of voltage-time data from a plurality of CAC patients, wherein the plurality of voltage-time data comprises a plurality of image data characterizing a plurality of ECG signals from the plurality of CAC patients correlated to indications comprising a probability value, indicating a likelihood of the level of CAC in a given patient; generating, automatedly, a training matrix by converting pixels in the training set representative of each ECG of the plurality of image data of the plurality of voltage-time data to associated training matrix elements to form a training matrix of a predetermined size, wherein a first dimension of the training matrix represents spatial leads and a second dimension of the training matrix represents a time series of a predetermined time duration at a predetermined frequency; training the CNN using the automatedly generated training matrix, wherein training the CNN comprises: feeding the training matrix, representative of the training set, to an input layer of the CNN; tiling the training matrix at the input layer into a plurality of tiles representative of the training matrix; applying learnable kernels of parameters in at least one convolutional layer of the CNN across pixel data in the plurality of tiles which is representative of the plurality of image data characterizing the plurality of ECG signals in the training set, wherein a predetermined number of parameters in the at least one convolutional layer are learned irrespective of image size; and populating an output layer of the CNN based on results generated by application of the learnable kernels to the pixel data; updating the feature vector as a function of demographic data associated with the subject and genomic data associated with the subject; inputting, to the trained CNN, the updated feature vector representative of the subject's image data, the subject's demographic data and the subject's genomic data; and outputting, from the trained CNN, as a function of the updated feature vector, the indication of the level of CAC in the subject.
12. 12 . The system of claim 11 , wherein generating the feature vector comprises generating a spectrogram as a function of the voltage data.
13. 13 . The system of claim 11 , wherein generating the feature vector comprises grouping the voltage data from each lead of the plurality of leads into a plurality of subsets.
14. 14 . The system of claim 11 , wherein the computer readable storage medium contains further instructions configuring the at least a processor to: receive the demographic data associated with the subject; and generate the feature vector as a function of the demographic data and the voltage-time data.
15. 15 . The system of claim 11 , wherein the computer readable storage medium contains further instructions configuring the at least a processor to: receive the genomic data associated with the subject; and generate the feature vector as a function of the genomic data and the voltage-time data.
16. 16 . The system of claim 11 , wherein the voltage-time data of the subject is received from an electronic medical record.
17. 17 . The system of claim 11 , wherein the feature vector comprises a matrix, wherein the matrix comprises: a plurality of rows corresponding to a temporal dimension; and a plurality of columns corresponding to a spatial dimension.
18. 18 . The system of claim 17 , wherein each row of the plurality of rows correspond to at least one lead of the plurality of leads.
19. 19 . The system of claim 17 , wherein each column of the plurality of columns corresponds to at least one timestamp.
20. 20 . The system of claim 17 , wherein the temporal dimension has a resolution of 500 Hz.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS This application is a continuation of Patent Cooperation Treaty (PCT) Application No. PCT/US2023/022325, filed on May 16, 2023, and entitled “DEEP LEARNING ENABLED ELECTROCARDIOGRAPHIC PREDICTION OF COMPUTER TOMOGRAPHY-BASED HIGH CORONARY CALCIUM SCORE (CAC)”, which claims the benefit of priority to U.S. Provisional Patent Application Ser. No. 63/342,275, filed on May 16, 2022, and titled “DEEP LEARNING ENABLED ELECTROCARDIOGRAPHIC PREDICTION OF COMPUTER TOMOGRAPHY-BASED HIGH CORONARY CALCIUM SCORE (CAC)”, which is incorporated by reference herein in its entirety. FIELD OF THE INVENTION The present invention generally relates to the field of artificial intelligence. In particular, the present invention is directed to a system and a method for electrocardiographic prediction of computed tomography-based high coronary calcium score (CAC). BACKGROUND Embodiments of the present disclosure relate to methods for the detection and evaluation of coronary artery calcium and treatment of subjects identified for cardiovascular risk. Computed tomography-based coronary artery calcium (CAC) scoring is recommended in adults with unclear cardiovascular risk to inform preventative strategies and statin prescription. CAC has major limitations as it exposes the individual to radiation, is costly, not readily available everywhere and requires scoring by an expert radiologist. Thus, there is a long felt and unmet need for improved methods to detect and evaluate CAC. SUMMARY OF THE DISCLOSURE According to embodiments of the present disclosure, methods of and computer program products for the detection of coronary artery calcium and prediction of CAC score from 12-lead electrocardiograms (ECG-AI). In some aspects of the invention, disclosed herein are methods comprising receiving voltage-time data of a subject, the voltage-time data comprising voltage data of a plurality of leads of an electrocardiograph; a feature vector from the voltage-time data; providing the feature vector to a pretrained learning system; and receiving from the pretrained learning system an indication of the level of CAC in the subject. Aspects of the invention, as disclosed herein, also include a system comprising: an electrocardiograph comprising a plurality of leads; a computing node comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor of the computing node to cause the processor to perform a method comprising: receiving voltage-time data of a subject from the echocardiograph, the voltage-time data comprising voltage data of the plurality of leads; generating a feature vector from the voltage-time data; providing the feature vector to a pretrained learning system; and receiving from the pretrained learning system an indication of the level of CAC in the subject. In certain aspects of the invention, disclosed herein is a computer program product for evaluation of CAC, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform a method comprising: receiving voltage-time data of a subject from the echocardiograph, the voltage-time data comprising voltage data of the plurality of leads; generating a feature vector from the voltage-time data; providing the feature vector to a pretrained learning system; and receiving from the pretrained learning system an indication of the level of CAC in the subject. BRIEF DESCRIPTION OF THE DRAWINGS For the purpose of illustrating the invention, the drawings show aspects of one or more embodiments of the invention. However, it should be understood that the present invention is not limited to the precise arrangements and instrumentalities shown in the drawings, wherein: FIG. 1 is a schematic view of an exemplary embodiment of a system for detecting or otherwise predicting coronary artery calcium (CAC). FIG. 2 is an exemplary flowchart illustrating a method of detecting or otherwise predicting CAC according to embodiments of the present disclosure. FIG. 3 is an exemplary depiction of a computing node according to an embodiment of the present disclosure. FIG. 4 is an exemplary depiction of a receiver operating characteristic curve (ROC) of the convolutional neural network used to identify patients with a CAC score 2: 300. FIG. 5 is an exemplary depiction of a relationship between the AI-ECG probability and the Calcium score level. FIG. 6 is a table of network ROC and sensitivity and specificity across subgroups. The drawings are not necessarily to scale and may be illustrated by phantom lines, diagrammatic representations, and fragmentary views. In certain instances, details that are not necessary for an understanding of the embodiments or that render other details difficult to perceive may have been omitted. DETAILED DESCRIPTION Approximately half of