Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Status of Claims
This office action for the 18/993031 application is in response to the communications filed December 11, 2025.
Claims 1-11 were initially submitted January 10, 2025.
Claims 1-11 were cancelled December 11, 2025.
Claims 12-33 were added as new December 11, 2025.
Claims 12-33 are currently pending and considered below.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 12-33 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
As per claim 12,
Step 1: The claim recites subject matter within a statutory category as a process.
Step 2A is a two-prong inquiry, in which Prong 1 determines whether a claim recites a judicial exception. Prong 2 determines if the additional limitations of the claim integrates the recited judicial exception into a practical application. If the additional elements of the claim fail to integrate the judicial exception into a practical application, claim is directed to the recited judicial exception, see MPEP 2106.04(II)(A).
Step 2A Prong 1: The claim contains subject matter that recites an abstract idea, with the steps of a method of facilitating diagnosis of myocardial infarction of a subject based on signal processing analysis of electrocardiogram signal data, the method comprising: accessing electrocardiogram (ECG) signal data obtained from a subject; accessing data obtained from a plurality of other subjects, the data comprising ECG signal data and coronary angiography data; analyzing the ECG signal data obtained from the plurality of other subjects; analyzing the coronary angiography data obtained from the plurality of other subjects to assign a degree of stenosis of one or more coronary arteries of the plurality of other subjects; and correlations between the analyzed ECG signal data obtained from the plurality of other subjects and the degree of stenosis of the one or more coronary arteries of the plurality of other subjects; and generating an indication representing a likelihood or progression of myocardial infarction for the subject based on a predicted degree of stenosis in one or more coronary arteries of the subject. These steps, as drafted, under the broadest reasonable interpretation recite:
certain methods of organizing human activity (e.g., fundamental economic principles or practices including: hedging; insurance; mitigating risk; etc., commercial or legal interactions including: agreements in the form of contracts; legal obligations; advertising, marketing or sales activities or behaviors; business relations; etc., managing personal behavior or relationships or interactions between people including: social activities; teaching; following rules or instructions; etc.) but for recitation of generic computer components. That is, other than reciting steps as performed by the generic computer components, nothing in the claim element precludes the step from being directed to certain methods of organizing human activity. The identified abstract idea, law of nature, or natural phenomenon identified above, in the context of this claim, encompasses a certain method of organizing human activity, namely, managing personal behavior or relationships or interactions between people. This is because each of the limitations of the abstract idea recite a list of rules or instructions that a human person can follow in the course of their personal behavior. If a claim limitation, under its broadest reasonable interpretation, covers at least the recited methods of organizing human activity above, but for the recitation of generic computer components, then it falls within the “Certain Methods of Organizing Human Activity” grouping of abstract ideas. Accordingly, the claim recites an abstract idea. See MPEP 2106.04(a).
Step 2A Prong 2: The claim does not recite additional elements that integrate the judicial exception into a practical application. In particular, the additional elements do not integrate the abstract idea into a practical application, other than the abstract idea per se, because the additional elements amount to no more than limitations which:
amount to mere instructions to apply an exception, see MPEP 2106.05(f), such as:
“computer-implemented”, “by a computer system”, “the deep learning model trained by”, “training”, “training the deep learning model based on”, “by the computer system” and “derived from an output of the deep learning model, wherein the computer system comprises a computer processor and an electronic storage medium” which corresponds to merely using a computer as a tool to perform an abstract idea. Page 9 Lines 5-17 of the as-filed specification describes that the hardware that implements the steps of the abstract idea amount to nothing more than a generic computer. Implementing an abstract idea on a generic computer, does not integrate the abstract idea into a practical application in Step 2A Prong Two or add significantly more in Step 2B, similar to how the recitation of the computer in the claim in Alice amounted to mere instructions to apply the abstract idea of intermediated settlement on a generic computer.
add insignificant extra-solution activity to the abstract idea, see MPEP 2106.05(g), such as:
“inputting, by the computer system, the accessed ECG signal data into a deep learning model” which corresponds to mere data gathering and/or output.
Accordingly, this claim is directed to an abstract idea.
Step 2B: The claim does not recite additional elements that amount to significantly more than the judicial exception. As discussed above with respect to discussion of integration of the abstract idea into a practical application, the additional elements amount to no more than mere instructions to apply an exception, add insignificant extra-solution activity to the abstract idea, and/or generally link the abstract idea to a particular technological environment or field of use. Additionally, the additional limitations, identified as insignificant extra-solution activity to the abstract idea, amount to no more than limitations which amount to elements that have been recognized as well-understood, routine, and conventional activity in particular fields such as:
computer functions that have been identified by the courts as well‐understood, routine, and conventional functions when they are claimed in a merely generic manner (e.g., at a high level of generality) or as insignificant extra-solution activity, see MPEP 2106.05(d)(II), such as:
“inputting, by the computer system, the accessed ECG signal data into a deep learning model” which corresponds to receiving or transmitting data over a network.
Looking at the limitations of the claim as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely recite an abstract idea and/or provide conventional computer implementation which does not impose a meaningful limit to integrate the abstract idea into a practical application and/or amount to no more than limitations which amount to elements that have been recognized as well-understood, routine, and conventional activity in particular fields.
As per claim 13,
Claim 13 depends from claim 12 and inherits all the limitations of the claim from which it depends. Claim 13 merely further defines the abstract idea and/or introduces additional elements that are insufficient to provide a practical application or something significantly more:
“correlations between the analyzed ECG signal data obtained from the plurality of other subjects and the degree of stenosis of each classification of the one or more coronary arteries of the plurality of other subjects, the classification of the one or more coronary arteries comprising one or more of a right coronary artery (RCA), left anterior descending artery (LAD), or left circumflex artery (LCx)” further describes the abstract idea. This claim limitation is still directed to “Certain Methods of Organizing Human Activity” and therefore continues to recite an abstract idea.
“wherein the deep learning model is further trained based on” further defines an additional element that was insufficient to provide a practical application and/or significantly more. The claim with this further defining limitation still corresponds to merely using a computer as a tool to perform an abstract idea.
Looking at the limitations of the claim as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely recite an abstract idea and/or provide conventional computer implementation which does not impose a meaningful limit to integrate the abstract idea into a practical application and/or amount to no more than limitations which amount to elements that have been recognized as well-understood, routine, and conventional activity in particular fields.
As per claim 14,
Claim 14 depends from claim 13 and inherits all the limitations of the claim from which it depends. Claim 14 merely further defines the abstract idea and/or introduces additional elements that are insufficient to provide a practical application or something significantly more:
“wherein the degree of stenosis of the one or more coronary arteries is assigned differently depending on the classification of the one or more coronary arteries.” further describes the abstract idea. This claim limitation is still directed to “Certain Methods of Organizing Human Activity” and therefore continues to recite an abstract idea.
Looking at the limitations of the claim as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely recite an abstract idea and/or provide conventional computer implementation which does not impose a meaningful limit to integrate the abstract idea into a practical application and/or amount to no more than limitations which amount to elements that have been recognized as well-understood, routine, and conventional activity in particular fields.
As per claim 15,
Claim 15 depends from claim 12 and inherits all the limitations of the claim from which it depends. Claim 15 merely further defines the abstract idea and/or introduces additional elements that are insufficient to provide a practical application or something significantly more:
“calculating an error using a loss function… until the calculated error is within a predetermined threshold” further describes the abstract idea. This claim limitation is still directed to “Certain Methods of Organizing Human Activity” and therefore continues to recite an abstract idea.
“wherein the deep learning model is further trained by” and “repeatedly adjusting parameters of the deep learning model” further defines an additional element that was insufficient to provide a practical application and/or significantly more. The claim with this further defining limitation still corresponds to merely using a computer as a tool to perform an abstract idea.
Looking at the limitations of the claim as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely recite an abstract idea and/or provide conventional computer implementation which does not impose a meaningful limit to integrate the abstract idea into a practical application and/or amount to no more than limitations which amount to elements that have been recognized as well-understood, routine, and conventional activity in particular fields.
As per claim 16,
Claim 16 depends from claim 12 and inherits all the limitations of the claim from which it depends. Claim 16 merely further defines the abstract idea and/or introduces additional elements that are insufficient to provide a practical application or something significantly more:
“classifying the…data into one or more…data groups based on one or more criteria…correlations between the analyzed ECG signal data and the degree of stenosis of the one or more coronary arteries from the one or more … data groups; generating a predicted degree of stenosis in one or more coronary arteries for the one or more…data groups; deriving a verification result for each of the one or more … data groups” further describes the abstract idea. This claim limitation is still directed to “Certain Methods of Organizing Human Activity” and therefore continues to recite an abstract idea.
“wherein the deep learning model is further trained by”, “training”, “training the deep learning model based on” and “performing additional training for the one or more training data groups when the derived verification result is below a predetermined threshold” further defines an additional element that was insufficient to provide a practical application and/or significantly more. The claim with this further defining limitation still corresponds to merely using a computer as a tool to perform an abstract idea.
Looking at the limitations of the claim as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely recite an abstract idea and/or provide conventional computer implementation which does not impose a meaningful limit to integrate the abstract idea into a practical application and/or amount to no more than limitations which amount to elements that have been recognized as well-understood, routine, and conventional activity in particular fields.
As per claim 17,
Claim 17 depends from claim 16 and inherits all the limitations of the claim from which it depends. Claim 17 merely further defines the abstract idea and/or introduces additional elements that are insufficient to provide a practical application or something significantly more:
“wherein one or more criteria for classifying the…data into the one or more…data groups comprises demographics, presence of a risk factor for disease onset, adverse medical history, time to onset of symptoms, type of chest pain, type of myocardial infarction, type of acute coronary syndrome, type of coronary artery, or type of ECG” further describes the abstract idea. This claim limitation is still directed to “Certain Methods of Organizing Human Activity” and therefore continues to recite an abstract idea.
“training” further defines an additional element that was insufficient to provide a practical application and/or significantly more. The claim with this further defining limitation still corresponds to merely using a computer as a tool to perform an abstract idea.
Looking at the limitations of the claim as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely recite an abstract idea and/or provide conventional computer implementation which does not impose a meaningful limit to integrate the abstract idea into a practical application and/or amount to no more than limitations which amount to elements that have been recognized as well-understood, routine, and conventional activity in particular fields.
As per claim 18,
Claim 18 depends from claim 16 and inherits all the limitations of the claim from which it depends. Claim 18 merely further defines the abstract idea and/or introduces additional elements that are insufficient to provide a practical application or something significantly more:
“wherein the additional training comprises few-shot training.” further defines an additional element that was insufficient to provide a practical application and/or significantly more. The claim with this further defining limitation still corresponds to merely using a computer as a tool to perform an abstract idea.
Looking at the limitations of the claim as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely recite an abstract idea and/or provide conventional computer implementation which does not impose a meaningful limit to integrate the abstract idea into a practical application and/or amount to no more than limitations which amount to elements that have been recognized as well-understood, routine, and conventional activity in particular fields.
As per claim 19,
Claim 19 depends from claim 12 and inherits all the limitations of the claim from which it depends. Claim 19 merely further defines the abstract idea and/or introduces additional elements that are insufficient to provide a practical application or something significantly more:
“wherein the deep learning model is trained using one or more of supervised learning or self-supervised learning.” further defines an additional element that was insufficient to provide a practical application and/or significantly more. The claim with this further defining limitation still corresponds to merely using a computer as a tool to perform an abstract idea.
Looking at the limitations of the claim as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely recite an abstract idea and/or provide conventional computer implementation which does not impose a meaningful limit to integrate the abstract idea into a practical application and/or amount to no more than limitations which amount to elements that have been recognized as well-understood, routine, and conventional activity in particular fields.
As per claim 20,
Claim 20 depends from claim 12 and inherits all the limitations of the claim from which it depends. Claim 20 merely further defines the abstract idea and/or introduces additional elements that are insufficient to provide a practical application or something significantly more:
“further comprising standardizing…the ECG signal data” further describes the abstract idea. This claim limitation is still directed to “Certain Methods of Organizing Human Activity” and therefore continues to recite an abstract idea.
“by the computer system” and “prior to inputting the accessed ECG signal data into the deep learning model” further defines an additional element that was insufficient to provide a practical application and/or significantly more. The claim with this further defining limitation still corresponds to merely using a computer as a tool to perform an abstract idea.
Looking at the limitations of the claim as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely recite an abstract idea and/or provide conventional computer implementation which does not impose a meaningful limit to integrate the abstract idea into a practical application and/or amount to no more than limitations which amount to elements that have been recognized as well-understood, routine, and conventional activity in particular fields.
As per claim 21,
Claim 21 depends from claim 12 and inherits all the limitations of the claim from which it depends. Claim 21 merely further defines the abstract idea and/or introduces additional elements that are insufficient to provide a practical application or something significantly more:
“wherein the ECG signal data is obtained from a medical electrocardiogram equipment or a wearable electronic device.” further defines an additional element that was insufficient to provide a practical application and/or significantly more. The claim with this further defining limitation still corresponds to mere data gathering and/or output and receiving or transmitting data over a network.
Looking at the limitations of the claim as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely recite an abstract idea and/or provide conventional computer implementation which does not impose a meaningful limit to integrate the abstract idea into a practical application and/or amount to no more than limitations which amount to elements that have been recognized as well-understood, routine, and conventional activity in particular fields.
As per claim 22,
Claim 22 depends from claim 12 and inherits all the limitations of the claim from which it depends. Claim 22 merely further defines the abstract idea and/or introduces additional elements that are insufficient to provide a practical application or something significantly more:
“wherein analyzing the ECG signal data obtained from the plurality of other subjects comprises extracting ECG waveform features, the ECG waveform features comprising one or more of shape, size, or duration of a P wave, QRS wave, or T wave.” further describes the abstract idea. This claim limitation is still directed to “Certain Methods of Organizing Human Activity” and therefore continues to recite an abstract idea.
Looking at the limitations of the claim as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely recite an abstract idea and/or provide conventional computer implementation which does not impose a meaningful limit to integrate the abstract idea into a practical application and/or amount to no more than limitations which amount to elements that have been recognized as well-understood, routine, and conventional activity in particular fields.
As per claim 23,
Claim 23 is substantially similar to claim 12. Accordingly, claim 23 is rejected for the same reasons as claim 12.
Claim 23 further recites “A system for” and “one or more computer readable storage devices configured to store a plurality of computer executable instructions; and one or more hardware computer processors in communication with the one or more computer readable storage devices and configured to execute the plurality of computer executable instructions in order to cause the system to” which further defines an additional element that was insufficient to provide a practical application and/or significantly more. The claim with this further defining limitation still corresponds to merely using a computer as a tool to perform an abstract idea.
Looking at the limitations of the claim as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely recite an abstract idea and/or provide conventional computer implementation which does not impose a meaningful limit to integrate the abstract idea into a practical application and/or amount to no more than limitations which amount to elements that have been recognized as well-understood, routine, and conventional activity in particular fields.
As per claim 24,
Claim 24 is substantially similar to claim 13. Accordingly, claim 24 is rejected for the same reasons as claim 13.
As per claim 25,
Claim 25 is substantially similar to claim 14. Accordingly, claim 25 is rejected for the same reasons as claim 14.
As per claim 26,
Claim 26 is substantially similar to claim 15. Accordingly, claim 26 is rejected for the same reasons as claim 15.
As per claim 27,
Claim 27 is substantially similar to claim 16. Accordingly, claim 27 is rejected for the same reasons as claim 16.
As per claim 28,
Claim 28 is substantially similar to claim 17. Accordingly, claim 28 is rejected for the same reasons as claim 17.
As per claim 29,
Claim 29 is substantially similar to claim 18. Accordingly, claim 29 is rejected for the same reasons as claim 18.
As per claim 30,
Claim 30 is substantially similar to claim 19. Accordingly, claim 30 is rejected for the same reasons as claim 19.
As per claim 31,
Claim 31 is substantially similar to claim 20. Accordingly, claim 31 is rejected for the same reasons as claim 20.
As per claim 32,
Claim 32 is substantially similar to claim 21. Accordingly, claim 32 is rejected for the same reasons as claim 21.
As per claim 33,
Claim 33 is substantially similar to claim 22. Accordingly, claim 33 is rejected for the same reasons as claim 22.
Claim Rejections - 35 USC § 102
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claims 12-17, 19-28 and 30-33 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Jain et al. (US 11,568,991; herein referred to as Jain).
As per claim 12,
Jain discloses a computer-implemented method of facilitating diagnosis of myocardial infarction of a subject based on signal processing analysis of electrocardiogram signal data:
(Column 15 Lines 46-54 and Column 16 Lines 35-43 of Jain. The teaching describes a diagnostic tool, wherein the diagnostic tool includes: a sensor for capturing at least one biosignal produced by a patient's heart; and a computer device that implements a deep neural network that is trained iteratively through machine learning to generate a prediction about a heart condition of the patient. In some non-limiting aspects, the coronary condition includes at least one of: a Myocardial Infarction (MI), a risk of a Major Adverse Cardiovascular Event (MACE), a coronary artery disease (CAD), a left anterior descending (LAD) coronary artery fractional flow reserve (FFR), an atherosclerotic cardiovascular disease (ACVD), cardiac hypertrophy, ventricle morphology, an abnormal ST-T wave, a conduction disorder, and a “70%” disease threshold, or combinations thereof.)
Jain further discloses accessing, by a computer system, electrocardiogram (ECG) signal data obtained from a subject:
(Column 16 Lines 3-9 of Jain. The teaching describes that the biosignal is an ECG signal, and wherein N is 12. In some non-limiting aspects, the biosignal is an ECG signal, wherein N is less than 12, and wherein the computer device further implements an autoencoder trained via machine learning to generate 12-N biosignals based, at least in part, on a reconstruction loss function.)
Jain further discloses inputting, by the computer system, the accessed ECG signal data into a deep learning model:
(Column 15 Lines 46-54 and Column 16 Lines 35-43 of Jain. The teaching describes a diagnostic tool, wherein the diagnostic tool includes: a sensor for capturing at least one biosignal produced by a patient's heart; and a computer device that implements a deep neural network that is trained iteratively through machine learning to generate a prediction about a heart condition of the patient. In some non-limiting aspects, the coronary condition includes at least one of: a Myocardial Infarction (MI), a risk of a Major Adverse Cardiovascular Event (MACE), a coronary artery disease (CAD), a left anterior descending (LAD) coronary artery fractional flow reserve (FFR), an atherosclerotic cardiovascular disease (ACVD), cardiac hypertrophy, ventricle morphology, an abnormal ST-T wave, a conduction disorder, and a “70%” disease threshold, or combinations thereof.)
(Column 16 Lines 3-9 of Jain. The teaching describes that he biosignal is an ECG signal, and wherein N is 12. In some non-limiting aspects, the biosignal is an ECG signal, wherein N is less than 12, and wherein the computer device further implements an autoencoder trained via machine learning to generate 12-N biosignals based, at least in part, on a reconstruction loss function.)
Jain further discloses the deep learning model trained by: accessing training data obtained from a plurality of other subjects, the training data comprising ECG signal data and coronary angiography data; analyzing the ECG signal data obtained from the plurality of other subjects; analyzing the coronary angiography data obtained from the plurality of other subjects to assign a degree of stenosis of one or more coronary arteries of the plurality of other subjects; and training the deep learning model based on correlations between the analyzed ECG signal data obtained from the plurality of other subjects and the degree of stenosis of the one or more coronary arteries of the plurality of other subjects:
(Column 1 Lines 45-49, Column 3 Lines 46-67, Column 4 Lines 1-17 and Column 5 Lines 9-25 of Jain. The teaching describes that the software-based diagnostic tool can be trained through machine learning to enable users to better analyze and diagnose a patient's ECG signal. The software for the tool can be used with already cleared/approved ECG devices. The tool's software is able to accept input from any ECG device, regardless of device gain and sample frequency. The HEARTio computer 104 stores and executes a software program 108 that analyses a patient's biosignals (e.g., 12-lead ECG data) and outputs a predicted level, severity, and localization of coronary stenosis or other coronary diseases that the software is trained, through machine learning, to identify. The software program 108 can include one or more neural models and, after input of biosignal data, the data is pushed through the neural model(s), sometimes referred to herein as a “HEARTio neural model” as shown in FIG. 1 , implemented by the software 108 of the HEARTio computer 104. In that connection, the HEARTio neural model 108 can be a fixed graph that has a very specific set of instructions for how to process the input data. Then HEARTio computer 104 can separate the biosignal data into more than, for example, 100 distinct layers of the graph (or neural network) by applying a specific set of mathematical operations to the data exiting the preceding layer in the neural graph. The final layer of the HEARTio neural model 108 preferably provides an array of probabilities for the question being asked (Is this biosignal representative of coronary stenosis? How severe is this coronary stenosis? What is the location of the coronary stenosis? etc.) FIG. 2 depicts the training and validation processes 200 in more detail. The patient cohort is split into two groups: a “training set” 202 and a “testing set” 204. Generally accepted deep learning practice is to utilize greater than 90% of the patient cohort for the training set 202 and the remainder to validate the accuracy statistics of the model 206. The programmer may also manually set the hyperparameters, for the training, which are parameters unrelated to the training set but will impact the training environment (e.g., learning rate, label smoothing, dropout rate). Once the patient cohort is split and before training begins, the model randomizes the model 206 parameters it will refine to build the statistical dependencies between input and output. Over time the model 206 improves upon these parameters utilizing a process known as gradient descent, in which each of the parameters searches for local minimums of a loss function (the measure of how different the prediction is from the actual value).)
Jain further discloses generating, by the computer system, an indication representing a likelihood or progression of myocardial infarction for the subject based on a predicted degree of stenosis in one or more coronary arteries of the subject derived from an output of the deep learning model:
(Column 15 Lines 46-54 and Column 16 Lines 35-43 of Jain. The teaching describes a diagnostic tool, wherein the diagnostic tool includes: a sensor for capturing at least one biosignal produced by a patient's heart; and a computer device that implements a deep neural network that is trained iteratively through machine learning to generate a prediction about a heart condition of the patient. In some non-limiting aspects, the coronary condition includes at least one of: a Myocardial Infarction (MI), a risk of a Major Adverse Cardiovascular Event (MACE), a coronary artery disease (CAD), a left anterior descending (LAD) coronary artery fractional flow reserve (FFR), an atherosclerotic cardiovascular disease (ACVD), cardiac hypertrophy, ventricle morphology, an abnormal ST-T wave, a conduction disorder, and a “70%” disease threshold, or combinations thereof.)
Jain further discloses wherein the computer system comprises a computer processor and an electronic storage medium:
(Column 13 Lines 13-39 of Jain. The teaching describes a HEARTio computer device, such as the computer device 104 (FIG. 1 ), according to various aspects of the present invention. The illustrated computer device 2400 includes multiple processor units 2402A-B that each includes, in the illustrated aspect, multiple (N) sets of processor cores 2404A-N. Each processor unit 2402A-B may include onboard memory (ROM or RAM) (not shown) and off-board memory 2406A-B. The onboard memory may include primary, volatile, and/or non-volatile storage (e.g., storage directly accessible by the processor cores 2404A-N). The off-board memory 2406A-B may include secondary, non-volatile storage (e.g., storage that is not directly accessible by the processor cores 2404A-N), such as ROM, HDDs, SSD, flash, etc. The processor cores 2404A-N may be CPU cores, GPU cores and/or Al accelerator cores. GPU cores operate in parallel (e.g., a general-purpose GPU (GPGPU) pipeline) and, hence, can typically process data more efficiently that a collection of CPU cores, but all the cores of a GPU execute the same code at one time. Al accelerators are a class of microprocessor designed to accelerate artificial neural networks. They typically are employed as a coprocessor in a device with a host processor 2410 as well. An Al accelerator typically has tens of thousands of matrix multiplier units that operate at lower precision than a CPU core, such as 8-bit precision in an Al accelerator versus 64-bit precision in a CPU core.)
As per claim 13,
Jain discloses the limitations of claim 12.
Jain further discloses wherein the deep learning model is further trained based on correlations between the analyzed ECG signal data obtained from the plurality of other subjects and the degree of stenosis of each classification of the one or more coronary arteries of the plurality of other subjects, the classification of the one or more coronary arteries comprising one or more of a right coronary artery (RCA), left anterior descending artery (LAD), or left circumflex artery (LCx):
(Column 1 Lines 45-49, Column 3 Lines 46-67, Column 4 Lines 1-17 and Column 5 Lines 9-25 of Jain. The teaching describes that the software-based diagnostic tool can be trained through machine learning to enable users to better analyze and diagnose a patient's ECG signal. The software for the tool can be used with already cleared/approved ECG devices. The tool's software is able to accept input from any ECG device, regardless of device gain and sample frequency. The HEARTio computer 104 stores and executes a software program 108 that analyses a patient's biosignals (e.g., 12-lead ECG data) and outputs a predicted level, severity, and localization of coronary stenosis or other coronary diseases that the software is trained, through machine learning, to identify. The software program 108 can include one or more neural models and, after input of biosignal data, the data is pushed through the neural model(s), sometimes referred to herein as a “HEARTio neural model” as shown in FIG. 1 , implemented by the software 108 of the HEARTio computer 104. In that connection, the HEARTio neural model 108 can be a fixed graph that has a very specific set of instructions for how to process the input data. Then HEARTio computer 104 can separate the biosignal data into more than, for example, 100 distinct layers of the graph (or neural network) by applying a specific set of mathematical operations to the data exiting the preceding layer in the neural graph. The final layer of the HEARTio neural model 108 preferably provides an array of probabilities for the question being asked (Is this biosignal representative of coronary stenosis? How severe is this coronary stenosis? What is the location of the coronary stenosis? etc.) FIG. 2 depicts the training and validation processes 200 in more detail. The patient cohort is split into two groups: a “training set” 202 and a “testing set” 204. Generally accepted deep learning practice is to utilize greater than 90% of the patient cohort for the training set 202 and the remainder to validate the accuracy statistics of the model 206. The programmer may also manually set the hyperparameters, for the training, which are parameters unrelated to the training set but will impact the training environment (e.g., learning rate, label smoothing, dropout rate). Once the patient cohort is split and before training begins, the model randomizes the model 206 parameters it will refine to build the statistical dependencies between input and output. Over time the model 206 improves upon these parameters utilizing a process known as gradient descent, in which each of the parameters searches for local minimums of a loss function (the measure of how different the prediction is from the actual value).)
(Column 15 Lines 46-54 and Column 16 Lines 35-43 of Jain. The teaching describes a diagnostic tool, wherein the diagnostic tool includes: a sensor for capturing at least one biosignal produced by a patient's heart; and a computer device that implements a deep neural network that is trained iteratively through machine learning to generate a prediction about a heart condition of the patient. In some non-limiting aspects, the coronary condition includes at least one of: a Myocardial Infarction (MI), a risk of a Major Adverse Cardiovascular Event (MACE), a coronary artery disease (CAD), a left anterior descending (LAD) coronary artery fractional flow reserve (FFR), an atherosclerotic cardiovascular disease (ACVD), cardiac hypertrophy, ventricle morphology, an abnormal ST-T wave, a conduction disorder, and a “70%” disease threshold, or combinations thereof.)
As per claim 14,
Jain discloses the limitations of claim 13.
Jain further discloses wherein the degree of stenosis of the one or more coronary arteries is assigned differently depending on the classification of the one or more coronary arteries:
(Column 1 Lines 45-49, Column 3 Lines 46-67, Column 4 Lines 1-17 and Column 5 Lines 9-25 of Jain. The teaching describes that the software-based diagnostic tool can be trained through machine learning to enable users to better analyze and diagnose a patient's ECG signal. The software for the tool can be used with already cleared/approved ECG devices. The tool's software is able to accept input from any ECG device, regardless of device gain and sample frequency. The HEARTio computer 104 stores and executes a software program 108 that analyses a patient's biosignals (e.g., 12-lead ECG data) and outputs a predicted level, severity, and localization of coronary stenosis or other coronary diseases that the software is trained, through machine learning, to identify. The software program 108 can include one or more neural models and, after input of biosignal data, the data is pushed through the neural model(s), sometimes referred to herein as a “HEARTio neural model” as shown in FIG. 1 , implemented by the software 108 of the HEARTio computer 104. In that connection, the HEARTio neural model 108 can be a fixed graph that has a very specific set of instructions for how to process the input data. Then HEARTio computer 104 can separate the biosignal data into more than, for example, 100 distinct layers of the graph (or neural network) by applying a specific set of mathematical operations to the data exiting the preceding layer in the neural graph. The final layer of the HEARTio neural model 108 preferably provides an array of probabilities for the question being asked (Is this biosignal representative of coronary stenosis? How severe is this coronary stenosis? What is the location of the coronary stenosis? etc.) FIG. 2 depicts the training and validation processes 200 in more detail. The patient cohort is split into two groups: a “training set” 202 and a “testing set” 204. Generally accepted deep learning practice is to utilize greater than 90% of the patient cohort for the training set 202 and the remainder to validate the accuracy statistics of the model 206. The programmer may also manually set the hyperparameters, for the training, which are parameters unrelated to the training set but will impact the training environment (e.g., learning rate, label smoothing, dropout rate). Once the patient cohort is split and before training begins, the model randomizes the model 206 parameters it will refine to build the statistical dependencies between input and output. Over time the model 206 improves upon these parameters utilizing a process known as gradient descent, in which each of the parameters searches for local minimums of a loss function (the measure of how different the prediction is from the actual value).)
(Column 15 Lines 46-54 and Column 16 Lines 35-43 of Jain. The teaching describes a diagnostic tool, wherein the diagnostic tool includes: a sensor for capturing at least one biosignal produced by a patient's heart; and a computer device that implements a deep neural network that is trained iteratively through machine learning to generate a prediction about a heart condition of the patient. In some non-limiting aspects, the coronary condition includes at least one of: a Myocardial Infarction (MI), a risk of a Major Adverse Cardiovascular Event (MACE), a coronary artery disease (CAD), a left anterior descending (LAD) coronary artery fractional flow reserve (FFR), an atherosclerotic cardiovascular disease (ACVD), cardiac hypertrophy, ventricle morphology, an abnormal ST-T wave, a conduction disorder, and a “70%” disease threshold, or combinations thereof.)
As per claim 15,
Jain discloses the limitations of claim 12.
Jain further discloses wherein the deep learning model is further trained by calculating an error using a loss function and repeatedly adjusting parameters of the deep learning model until the calculated error is within a predetermined threshold:
(Column 5 Lines 26-40 of Jain. The teaching describes that the training the model 206 is iterative (as shown in FIG. 3 ). The number of epochs (a measure of how many forward and backward passes the model 206 makes as part of the iterative gradient descent training), and the number of steps per epoch (a measure of how many data points are evaluated within the epoch) may be set by the programmer. In each epoch, a training set is randomly pulled from the training cohort 204. The model 206 then makes predictions about the patient based on patient biosignals, trying to identify important parameters that make the correct diagnosis. Once the forward pass (forward propagation) is complete, the predictions are sent to the loss function, which compares the predictions against the actual patient conditions and determines the overall error. This error then gets back-propagated through the model 206 and helps adjusts parameters to correct for the model 206 error.)
As per claim 16,
Jain discloses the limitations of claim 12.
Jain further discloses wherein the deep learning model is further trained by: classifying the training data into one or more training data groups based on one or more criteria; training the deep learning model based on correlations between the analyzed ECG signal data and the degree of stenosis of the one or more coronary arteries from the one or more training data groups; generating a predicted degree of stenosis in one or more coronary arteries for the one or more training data groups; deriving a verification result for each of the one or more training data groups; and performing additional training for the one or more training data groups when the derived verification result is below a predetermined threshold:
(Column 1 Lines 45-49, Column 3 Lines 46-67, Column 4 Lines 1-17 and Column 5 Lines 9-25 of Jain. The teaching describes that the software-based diagnostic tool can be trained through machine learning to enable users to better analyze and diagnose a patient's ECG signal. The software for the tool can be used with already cleared/approved ECG devices. The tool's software is able to accept input from any ECG device, regardless of device gain and sample frequency. The HEARTio computer 104 stores and executes a software program 108 that analyses a patient's biosignals (e.g., 12-lead ECG data) and outputs a predicted level, severity, and localization of coronary stenosis or other coronary diseases that the software is trained, through machine learning, to identify. The software program 108 can include one or more neural models and, after input of biosignal data, the data is pushed through the neural model(s), sometimes referred to herein as a “HEARTio neural model” as shown in FIG. 1 , implemented by the software 108 of the HEARTio computer 104. In that connection, the HEARTio neural model 108 can be a fixed graph that has a very specific set of instructions for how to process the input data. Then HEARTio computer 104 can separate the biosignal data into more than, for example, 100 distinct layers of the graph (or neural network) by applying a specific set of mathematical operations to the data exiting the preceding layer in the neural graph. The final layer of the HEARTio neural model 108 preferably provides an array of probabilities for the question being asked (Is this biosignal representative of coronary stenosis? How severe is this coronary stenosis? What is the location of the coronary stenosis? etc.) FIG. 2 depicts the training and validation processes 200 in more detail. The patient cohort is split into two groups: a “training set” 202 and a “testing set” 204. Generally accepted deep learning practice is to utilize greater than 90% of the patient cohort for the training set 202 and the remainder to validate the accuracy statistics of the model 206. The programmer may also manually set the hyperparameters, for the training, which are parameters unrelated to the training set but will impact the training environment (e.g., learning rate, label smoothing, dropout rate). Once the patient cohort is split and before training begins, the model randomizes the model 206 parameters it will refine to build the statistical dependencies between input and output. Over time the model 206 improves upon these parameters utilizing a process known as gradient descent, in which each of the parameters searches for local minimums of a loss function (the measure of how different the prediction is from the actual value).)
(Column 5 Lines 26-40 of Jain. The teaching describes that the training the model 206 is iterative (as shown in FIG. 3 ). The number of epochs (a measure of how many forward and backward passes the model 206 makes as part of the iterative gradient descent training), and the number of steps per epoch (a measure of how many data points are evaluated within the epoch) may be set by the programmer. In each epoch, a training set is randomly pulled from the training cohort 204. The model 206 then makes predictions about the patient based on patient biosignals, trying to identify important parameters that make the correct diagnosis. Once the forward pass (forward propagation) is complete, the predictions are sent to the loss function, which compares the predictions against the actual patient conditions and determines the overall error. This error then gets back-propagated through the model 206 and helps adjusts parameters to correct for the model 206 error.)
As per claim 17,
Jain discloses the limitations of claim 16.
Jain further teaches wherein one or more criteria for classifying the training data into the one or more training data groups comprises demographics, presence of a risk factor for disease onset, adverse medical history, time to onset of symptoms, type of chest pain, type of myocardial infarction, type of acute coronary syndrome, type of coronary artery, or type of ECG:
(Column 15 Lines 46-54 and Column 16 Lines 35-43 of Jain. The teaching describes a diagnostic tool, wherein the diagnostic tool includes: a sensor for capturing at least one biosignal produced by a patient's heart; and a computer device that implements a deep neural network that is trained iteratively through machine learning to generate a prediction about a heart condition of the patient. In some non-limiting aspects, the coronary condition includes at least one of: a Myocardial Infarction (MI), a risk of a Major Adverse Cardiovascular Event (MACE), a coronary artery disease (CAD), a left anterior descending (LAD) coronary artery fractional flow reserve (FFR), an atherosclerotic cardiovascular disease (ACVD), cardiac hypertrophy, ventricle morphology, an abnormal ST-T wave, a conduction disorder, and a “70%” disease threshold, or combinations thereof.)
As per claim 19,
Jain discloses the limitations of claim 12.
Jain further discloses wherein the deep learning model is trained using one or more of supervised learning or self-supervised learning:
(Column 4 Lines 18-36 of Jain. The teaching describes he HEARTio neural model 108 is not determined by a physician, programmer, or expert, but by the HEARTio computer 104 itself. A typical software program implements a known flow of information through a fixed set of code to perform processes on the data, but those processes are designed by the programmer. With the diagnostic tool 100 of the present invention, the determination of these instructions is done algorithmically, meaning the HEARTio computer 104 itself determines what mathematical operations are used to operate on data exiting one layer of the graph and being delivered to the next layer. A training set of data is used to train the neural model 108 using, for example, supervised, unsupervised, and/or semi-supervised training. The graph is initially set up to use a randomly generated set of instructions to deliver the next layer of data from the preceding layer, with the first layer being the “training set” of biosignal data, singularly, for each patient in the “training set”.)
As per claim 20,
Jain discloses the limitations of claim 12.
Jain further discloses further comprising standardizing, by the computer system, the ECG signal data prior to inputting the accessed ECG signal data into the deep learning model:
(Column 8 Lines 17-27 of Jain. The teaching describes a biosignal in tensor form the following steps may be followed. First, clip a biosignal such that it represents (for example) 10 seconds and is in the form N×M, where M is the number of the leads and N is the number of samples. N is equal to the sampling rate of the biosignal multiplied by 10. Second, using fast Fourier transform (FFT), resample N to 1000, in effect reducing the sampling rate to 100 Hz. Third, bandpass Butterworth filter with a passband starting at 2 Hz and extending to 40 Hz. This will try to make sure only the information within this band of frequencies are retained while others are removed. This FFT and Butterworth filter process is considered to be a standardization of the ECG data.)
As per claim 21,
Jain discloses the limitations of claim 12.
Jain further discloses wherein the ECG signal data is obtained from a medical electrocardiogram equipment or a wearable electronic device:
(Column 3 Lines 29-34 of Jain. The teaching describes an ECG machine with one or more leads, it shall be appreciated that the sensor 102 can include any number of devices configured to capture any number of biosignals associated with the patient. For example, in other non-limiting aspects, the sensor 102 can include an electroencephalogram (EEG) machine. In still other non-limiting aspects, the sensor 102 can include a wearable device (e.g., a smart watch, a smart ring, a smart phone, or smart glasses, etc.) configured to generate either ECG or EEG signals. HEARTio)
As per claim 22,
Jain discloses the limitations of claim 12.
Jain further discloses wherein analyzing the ECG signal data obtained from the plurality of other subjects comprises extracting ECG waveform features, the ECG waveform features comprising one or more of shape, size, or duration of a P wave, QRS wave, or T wave:
(Column 15 Lines 46-54 and Column 16 Lines 35-43 of Jain. The teaching describes a diagnostic tool, wherein the diagnostic tool includes: a sensor for capturing at least one biosignal produced by a patient's heart; and a computer device that implements a deep neural network that is trained iteratively through machine learning to generate a prediction about a heart condition of the patient. In some non-limiting aspects, the coronary condition includes at least one of: a Myocardial Infarction (MI), a risk of a Major Adverse Cardiovascular Event (MACE), a coronary artery disease (CAD), a left anterior descending (LAD) coronary artery fractional flow reserve (FFR), an atherosclerotic cardiovascular disease (ACVD), cardiac hypertrophy, ventricle morphology, an abnormal ST-T wave, a conduction disorder, and a “70%” disease threshold, or combinations thereof.)
As per claim 23,
Claim 23 is substantially similar to claim 12. Accordingly, claim 23 is rejected for the same reasons as claim 12.
Jain further discloses a system for and one or more computer readable storage devices configured to store a plurality of computer executable instructions; and one or more hardware computer processors in communication with the one or more computer readable storage devices and configured to execute the plurality of computer executable instructions in order to cause the system to:
(Column 13 Lines 13-39 of Jain. The teaching describes a HEARTio computer device, such as the computer device 104 (FIG. 1 ), according to various aspects of the present invention. The illustrated computer device 2400 includes multiple processor units 2402A-B that each includes, in the illustrated aspect, multiple (N) sets of processor cores 2404A-N. Each processor unit 2402A-B may include onboard memory (ROM or RAM) (not shown) and off-board memory 2406A-B. The onboard memory may include primary, volatile, and/or non-volatile storage (e.g., storage directly accessible by the processor cores 2404A-N). The off-board memory 2406A-B may include secondary, non-volatile storage (e.g., storage that is not directly accessible by the processor cores 2404A-N), such as ROM, HDDs, SSD, flash, etc. The processor cores 2404A-N may be CPU cores, GPU cores and/or Al accelerator cores. GPU cores operate in parallel (e.g., a general-purpose GPU (GPGPU) pipeline) and, hence, can typically process data more efficiently that a collection of CPU cores, but all the cores of a GPU execute the same code at one time. Al accelerators are a class of microprocessor designed to accelerate artificial neural networks. They typically are employed as a coprocessor in a device with a host processor 2410 as well. An Al accelerator typically has tens of thousands of matrix multiplier units that operate at lower precision than a CPU core, such as 8-bit precision in an Al accelerator versus 64-bit precision in a CPU core.)
As per claim 24,
Claim 24 is substantially similar to claim 13. Accordingly, claim 24 is rejected for the same reasons as claim 13.
As per claim 25,
Claim 25 is substantially similar to claim 14. Accordingly, claim 25 is rejected for the same reasons as claim 14.
As per claim 26,
Claim 26 is substantially similar to claim 15. Accordingly, claim 26 is rejected for the same reasons as claim 15.
As per claim 27,
Claim 27 is substantially similar to claim 16. Accordingly, claim 27 is rejected for the same reasons as claim 16.
As per claim 28,
Claim 28 is substantially similar to claim 17. Accordingly, claim 28 is rejected for the same reasons as claim 17.
As per claim 30,
Claim 30 is substantially similar to claim 19. Accordingly, claim 30 is rejected for the same reasons as claim 19.
As per claim 31,
Claim 31 is substantially similar to claim 20. Accordingly, claim 31 is rejected for the same reasons as claim 20.
As per claim 32,
Claim 32 is substantially similar to claim 21. Accordingly, claim 32 is rejected for the same reasons as claim 21.
As per claim 33,
Claim 33 is substantially similar to claim 22. Accordingly, claim 33 is rejected for the same reasons as claim 22.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 18 and 29 are rejected under 35 U.S.C. 103 as being unpatentable over Jain in view of Kim (US 2024/0212843).
As per claim 18,
Jain discloses the limitations of claim 16.
Jain does not explicitly disclose wherein the additional training comprises few-shot training.
However, Kim teaches a machine learning model that predicts the state of a patient given ECG data wherein this machine learning model is provided additional training which comprises few-shot training;
(Paragraph [0228] of Kim. The teaching describes pretraining and concurrent training of the above-mentioned supervised, self-supervised, and unsupervised learning increases the universality of the numerical vectors (embedding vectors) output by the ECG encoder by ensuring that the vectors simultaneously add and include morphological information corresponding to the clinician-defined patterns seen in the electrocardiogram and its own unrelated morphological information (supervised/self-supervised learning). Also, by efficiently rearranging the vector space in which the numerical vectors are placed (unsupervised learning), the ECG encoder can be efficiently used for other types of downstream tasks that are not initially planned. That is, there is a further benefit to implementing few-shot, one-shot learning.)
It would have been obvious to one of ordinary skill in the art before the time of filing to add to the teaching of Jain, the machine learning methods of Kim. Paragraph [0228] of Kim teaches that there is a benefit in efficiency to train a model with few-shot learning. One in possession of Jain would have looked to Kim to achieve such an improvement. One of ordinary skill in the art would have added to the teaching Jain, the teaching of Kim based on this incentive.
As per claim 29,
Claim 29 is substantially similar to claim 18. Accordingly, claim 29 is rejected for the same reasons as claim 18.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to CHAD A NEWTON whose telephone number is (313)446-6604. The examiner can normally be reached M-F 8:00AM-4:00PM (EST).
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, PETER H. CHOI can be reached at (469) 295-9171. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/CHAD A NEWTON/Primary Examiner, Art Unit 3681