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 .
Formal Matters
Applicant’s response, filed 12/29/2025, has been fully considered. The following rejections and/or objections are either reiterated or newly applied. They constitute the complete set presently being applied to the instant application.
Status of Claims
Claims 1-5 and 17 are currently pending and have been examined.
Claims 1, 7-9, and 16-17 have been amended.
Claim 6 has been canceled.
Claims 1-5 and 17 have been rejected.
Information Disclosure Statement
The information disclosure statement filed 09/06/2024 fails to comply with 37 CFR 1.98(a)(2), which requires a legible copy of each cited foreign patent document; each non-patent literature publication or that portion which caused it to be listed; and all other information or that portion which caused it to be listed. It has been placed in the application file, but the information referred to therein that has been crossed through has not been considered.
Claim Objections
Claims 8-9 objected to because of the following informalities:
Claim 8 discloses, “corresponding to most recent ECG data…” The claim limitation should read as, “corresponding to the most recent ECG data…” Claim 9 discloses, “corresponding most recent ECG data…” The claim limitation should read as, “corresponding to the most recent ECG data…”
Appropriate correction is required.
Claim Rejections - 35 USC § 112(b)
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 9 is rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention. Claim 9 is dependent on Claim 1, which does not disclose a first time point or first predetermined time period. As such, claim 9 is indefinite and should disclose a first and a second time point and predetermined time period instead of a second and third since claim 8 and 9 are separate embodiments of the invention.
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 1-5 and 7-17 are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter. The claimed invention is directed to an abstract idea without significantly more. Claims 1-5 and 7-17 are directed to a system, method, or product which are one of the statutory categories of invention. (Step 1: YES).
Independent Claim 1 discloses a method for analyzing electrocardiogram (ECG) data performed by a computing device, the method comprising: determining, by using a first model having ECG data acquired by ECG measurement as an input, whether an event related to at least one of atrial fibrillation or atrial flutter is found in the acquired ECG data; generating a first diagnosis result for the acquired ECG data when the event is found in the acquired ECG data; and generating a second diagnosis result for estimating a risk for at least one of the atrial fibrillation or the atrial flutter by using a second model having the acquired ECG data as an input when the event is not found in the acquired ECG data; wherein the second model is pre-trained based on: distinguishing normal ECG data in which the even does not occur and abnormal ECG data in which the event occurs in a dataset with a plurality of ECG data acquired for a first subject, and temporally aligning the distinguished normal ECG data and abnormal ECG data, and generating, based on time information corresponding to each of the distinguished normal ECG data and abnormal ECG data, training data to which a label corresponding to a first class representing a high-risk group or a second class representing a low-risk group is allocated with respect to the normal ECG data.
Independent Claim 16 discloses a computing device for analyzing electrocardiogram (ECG) data, comprising: at least one processor: and a memory, wherein the at least one processor: determines, by using a first model having ECG data acquired by ECG measurement as an input, whether an event related to at least one of atrial fibrillation or atrial flutter is found in the acquired ECG data, generates a first diagnosis result for the acquired ECG data when the event is found in the acquired ECG data, and generates a second diagnosis result for estimating a risk for at least one of the atrial fibrillation or the atrial flutter by using a second model having the acquired ECG data as an input when the event is not found in the acquired ECG data, wherein the second model is pre-trained based on: distinguishing normal ECG data in which the even does not occur and abnormal ECG data in which the event occurs in a dataset with a plurality of ECG data acquired for a first subject, and temporally aligning the distinguished normal ECG data and abnormal ECG data, and generating, based on time information corresponding to each of the distinguished normal ECG data and abnormal ECG data, training data to which a label corresponding to a first class representing a high-risk group or a second class representing a low-risk group is allocated with respect to the normal ECG data.
Independent Claim 17 discloses a computer program stored in a computer readable storage medium, wherein when the computer program is executed by at least one processor, the computer program allows the at least one processor to perform a method for analyzing electrocardiogram (ECG) data, and the method includes: determining, by using a first model having ECG data acquired by ECG measurement as an input, whether an event related to at least one of atrial fibrillation or atrial flutter is found in the acquired ECG data, generating a first diagnosis result for the acquired ECG data when the event is found in the acquired ECG data; and generating a second diagnosis result for estimating a risk for at least one of the atrial fibrillation or the atrial flutter by using a second model having the acquired ECG data as an input when the event is not found in the acquired ECG data, wherein the second model is pre-trained based on: distinguishing normal ECG data in which the even does not occur and abnormal ECG data in which the event occurs in a dataset with a plurality of ECG data acquired for a first subject, and temporally aligning the distinguished normal ECG data and abnormal ECG data, and generating, based on time information corresponding to each of the distinguished normal ECG data and abnormal ECG data, training data to which a label corresponding to a first class representing a high-risk group or a second class representing a low-risk group is allocated with respect to the normal ECG data.
The examiner is interpreting the above bolded limitations as additional elements as further discussed below. The remaining un-bolded limitations are merely directed to analyzing patient ECG data and outputting a result for a doctor. The series of steps recited above describe managing personal behavior or relationships or interactions between people and thus are grouped as certain methods of organizing human activity which is an abstract idea. (Step 2A- Prong 1: YES. The claims are abstract).
This judicial exception is not integrated into a practical application. Limitations that are not indicative of integration into a practical application include: (1) Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea (MPEP 2106.05.f), (2) Adding insignificant extra- solution activity to the judicial exception (MPEP 2106.05.g), (3) Generally linking the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05.h).
Independent Claim 1 discloses the following additional elements:
A computing device
Independent Claim 16 discloses the following additional elements:
A computing device
A processor
A memory
Independent Claim 17 discloses the following additional elements:
A computer program stored in a computer readable storage medium
A processor
In particular, the computing device (of claim 1 and 16), processor (of claim 16 and 17), memory (of claim 16), and computer program stored in a computer readable storage medium (of claim 17) are recited at a high-level of generality such that it amounts to no more than mere instructions to implement an abstract idea by adding the words ‘apply it’ (or an equivalent) with the judicial exception.
Applicant’s specification states in para 51 discloses, “The computing device 100 according to an embodiment of the present disclosure may include a processor 110 and a memory 130.” Para 53 discloses, “The computing device 100 may mean any type of user terminal or any type of server.” Para 58 discloses, “Further, the computer program executed in the computing device according to an embodiment of the present disclosure may be a CPU, GPGPU, or TPU executable program.” Para 59 discloses “the processor 110 processes data, information, signals, and the like input or output through the components included in the computing device 100 or drives the application program stored in a storage unit to provide or process information or a function appropriate for the user.” Thus, the components are performing as expected: the processor of the computing device is processing data, information, signals, and the like. Accordingly, these additional elements, when considered separately and as an ordered combination, do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea.
Accordingly, claim(s) 1 and 16-17 are directed to an abstract idea(s) without a practical application. (Step 2A-Prong 2: NO: the additional claimed elements are not integrated into a practical application).
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements of the computing device (of claim 1 and 16), processor (of claim 16 and 17), memory (of claim 16), and computer program stored in a computer readable storage medium (of claim 17) to perform the noted steps amounts to no more than mere instructions to apply the exception using a generic computer component. Mere instructions to apply an exception using a generic computer component cannot provide an inventive concept ("significantly more’). MPEP2106.05(I)(A) indicates that merely saying "apply it” or equivalent to the abstract idea cannot provide an inventive concept ("significantly more"). Accordingly, even in combination, this additional element does not provide significantly more. As such the independent claims 1 and 16-17 are not patent eligible. (Step 2B: NO. The claims do not provide significantly more).
Dependent claim(s) 2-5 and 7-15 are similarly rejected because they either further define/narrow the abstract idea and/or do not further limit the claim to a practical application or provide an inventive concept such that the claims are subject matter eligible even when considered individually or as an ordered combination. Dependent claims 3-4 do further disclose the additional element(s) of an artificial intelligence-based model (claims 3-5), a rule-based model (claim 3), deep learned based first sub-model (claim 12).
In particular, the artificial intelligence based model (of claims 3-5), rule based model (of claim 3), and deep learned based first sub-model (of claim 12) are recited at a high-level of generality such that it amounts to no more than mere instructions to implement an abstract idea by adding the words ‘apply it’ (or an equivalent) with the judicial exception. Accordingly, these additional elements, when considered separately and as an ordered combination, do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea.
Accordingly, even in combination, these additional elements do not integrate the abstract idea into a practical application.
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements of the artificial intelligence based model (of claims 3-5), rule based model (of claim 3), and deep learned based first sub-model (of claim 12) amounts to no more than mere instructions to apply the exception using a generic computer component. Mere instructions to apply an exception using a generic computer component cannot provide an inventive concept ("significantly more’). MPEP2106.05(I)(A) indicates that merely saying "apply it” or equivalent to the abstract idea cannot provide an inventive concept ("significantly more").
Therefore, the dependent claims are also directed to an abstract idea.
Thus, Claims 1-5 and 7-17 are rejected under 35 U.S.C. 101 as being directed to non-statutory subject matter.
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.
Claim(s) 1-4, 10-12, and 14-17 are rejected under 35 U.S.C. 103 as being unpatentable over Attia (WO 2020/086865 A1) in view of Chakravarthy (KR 2022/0006080 A).
Regarding Claim 1, Attia discloses:
A method for analyzing electrocardiogram (ECG) data performed by a computing device, the method comprising:
generating a second diagnosis result for estimating a risk for at least one of the atrial fibrillation or the atrial flutter by using a second model having the acquired ECG data as an input when the event is not found in the acquired ECG data. (Para 40 discloses the atrial fibrillation prediction 120 can indicate a likelihood that the patient 102… is susceptible to developing atrial fibrillation. The prediction 120 can be expressed as a probability or confidence score representing a probability or confidence that the patient 102… is susceptible to developing atrial fibrillation. Para 53 discloses this example describes the results of a study in which an artificial intelligence (Al)-model including a convolutional neural network was developed and tested to detect the electrocardiographic signature of atrial fibrillation present during normal sinus rhythm. The model was developed to process an ECG signature for a patient using a standard 10-second, 12- lead ECG recording. The example implementation was trained based on ECGs acquired from a set of patients aged 18 years or older having at least one digital, normal sinus rhythm, standard 10-second, 12-lead ECG acquired in the supine position at the MAYO CLINIC ECG laboratory between December 31, 1993, and July 21, 2017, with rhythm labels validated by trained personnel under cardiologist supervision. See further: Para 5.)
wherein the second model is pre-trained based on: distinguishing normal ECG data in which the event does not occur and abnormal ECG data in which the event occurs in a dataset with a plurality of ECG data acquired for a first subject, and temporally aligning the distinguished normal ECG data and abnormal ECG data, and generating, based on time information corresponding to each of the distinguished normal ECG data and abnormal ECG data, training data to which a label corresponding to a first class representing a high-risk group or a second class representing a low-risk group is allocated with respect to the normal ECG data. (Para 40 discloses the prediction 120 is expressed as a selection of a particular classification from multiple possible classifications that represents a most likely condition of the patient 102. For example, a binary classification can be made indicating whether there is at least a threshold probability or confidence level that the patient 102…is susceptible to developing atrial fibrillation [wherein over the threshold is high risk of being susceptible to an AF, low risk is below the threshold].Para 49 discloses a training engine 210 analyzes the predictions 208 and compares the predictions 208 to labels in the training examples 204 that indicate target predictions for each training example 204. The training engine 210 then generates updated model parameter values 214 by using an appropriate updating technique, e.g., stochastic gradient descent with backpropagation... the first component can represent an ECG of a patient under normal sinus rhythm, and the label can indicate whether that particular patient is known to have actually experienced atrial fibrillation at another time. In this way, the neural network 108 can be trained using sinus rhythm ECGs obtained in patients known and validated atrial fibrillation versus patients with no known atrial fibrillation... After training is complete, the training system 200 can provide a final set of parameter values 218 to the system 100 for use in making atrial fibrillation predictions 120. Para 57 discloses for patients who had had at least one atrial fibrillation rhythm recorded, the first recorded atrial fibrillation ECG was defined as the index ECG and the first day of the window of interest was defined as 31 days before the date of the index ECG. This window of interest was chosen with the assumption that the structural changes associated with atrial fibrillation would be present before the first recorded atrial fibrillation episode; a relatively short time interval was chosen as a conservative measure to avoid using ECGs before any structural changes developed. For patients with no ECGs with atrial fibrillation recorded, the index ECG was defined as the date of the first ECG available for that patient in the MAYO CLINIC Digital Data Vault. During training, all the ECGs in the window of interest were used to allow the network to have more samples; for the testing and validation sets, only the first normal sinus rhythm ECG within the window of interest was used to avoid repeated measurements and to mimic a real screening scenario.)
While Attia discloses the above limitations, Chakravarthy discloses the following limitations:
determining, by using a first model having ECG data acquired by ECG measurement as an input, whether an event related to at least one of atrial fibrillation or atrial flutter is found in the acquired ECG data; (Para 5 discloses the computing device receives the patient's electrocardiogram data sensed by the implantable medical device. The computing device obtains a first classification of the patient's arrhythmia through the feature-based description of the electrocardiogram data. The computing device applies a machine learning model to the received electrocardiogram data to obtain a second classification of the patient's arrhythmia. As an example, the computing device uses the first and second classifications to determine whether the patient has an arrhythmic episode. See further: Para 49)
generating a first diagnosis result for the acquired ECG data when the event is found in the acquired ECG data; and (Para 6 discloses in response to determining that an arrhythmic symptom has occurred in the patient, the computing device outputs a report indicating that an arrhythmic symptom has occurred and one or more cardiac features consistent with the arrhythmic symptom.)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the prediction of a likelihood that a patient has or will experience atrial fibrillation of the neural networks for atrial fibrillation screening as taught by Attia with the arrhythmia detection by feature description and machine learning as taught by Chakravarthy in order to determine whether the patient has an arrhythmic episode (Para 5) [and not just a risk, likelihood, or prediction] and apply feature delineation to classify the detected cardiac arrhythmia symptoms into … a particular type.. such as atrial fibrillation (Para 43) to improve patient welfare based on the determination such as by making “adjustments to the implantable medical device” upon determining an arrhythmic symptom has occurred.
Regarding Claim 2, this claim recites the limitations of Claim 1 and as to those limitations is rejected for the same basis and reasons as disclosed above. The combination of Attia and Chakravarthy further discloses the following limitation:
The method of claim 1, wherein the first model and the second model are trained based on different training data, and trained based on different training methods. (The combination of Attia and Chakravarthy discloses this limitation of claim 2. See Attia Para 53 as disclosed below (claim 4) where the model is trained on normal ECG data. See Chakravarthy Paras 14, 16, 17, and 60 as disclosed below (claim 3) where the model is trained on data including at least one of the atrial fibrillation or the atrial flutter.)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the prediction of a likelihood that a patient has or will experience atrial fibrillation of the neural networks for atrial fibrillation screening as taught by Attia with the arrhythmia detection by feature description and machine learning as taught by Chakravarthy in order to determine whether the patient has an arrhythmic episode (Para 5) [and not just a risk, likelihood, or prediction] and apply feature delineation to classify the detected cardiac arrhythmia symptoms into … a particular type.. such as atrial fibrillation (Para 43) to improve patient welfare based on the determination such as by making “adjustments to the implantable medical device” upon determining an arrhythmic symptom has occurred.
Regarding Claim 3, this claim recites the limitations of Claim 1 and as to those limitations is rejected for the same basis and reasons as disclosed above. The combination of Attia and Chakravarthy discloses the following limitation that Chakravarthy further discloses:
The method of claim 1 wherein the first model includes at least one of: an artificial intelligence-based model trained based on ECG data including at least one of the atrial fibrillation or the atrial flutter, or a rule-based model generated based on feature information for at least one of the atrial fibrillation or the atrial flutter. (Para 14 discloses feature delineation of ECG data and machine learning models processing patient data, such as machine learning systems and/or artificial intelligence (AI) algorithms, to analyze single and multi-channel patient data to detect and classify cardiac arrhythmias in patients. Techniques are disclosed for combining multiple decision mechanisms, such as state-of-the-art signal processing algorithms that perform Such patient data may include, for example, cardiac electrogram data or electrocardiogram (ECG) data. Para 16 discloses machine learning techniques can be contrasted with feature description in that feature description relies on signal processing, and machine learning systems use the relationship between features and electrocardiograms that exhibit arrhythmia symptoms on behalf of the system designer without knowledge or understanding of the symptoms of the arrhythmia. It can "learn" the basic features present in the data. Para 17 discloses machine learning and AI methods for arrhythmia detection can be applied to various targets (e.g., atrial fibrillation (AF) detection, cardiac symptoms without arrhythmias without the need for feature engineering required by expert design or feature delineation algorithms). Para 60 discloses In some examples, the machine learning model implemented by the machine learning system 450 is trained with training data that includes electrocardiogram data for a plurality of patients marked with descriptive metadata. For example, during the training phase, the machine learning system 450 processes a plurality of ECG waveforms. Generally, a plurality of ECG waveforms originate from a plurality of different patients. Each ECG waveform is marked with one or more symptoms of one or more types of arrhythmias. For example, a training ECG waveform may include a plurality of segments, each segment specifying the absence of an arrhythmia or the presence of a particular class of arrhythmia (e.g., bradycardia, tachycardia, atrial fibrillation, ventricular fibrillation, or AV block).)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the prediction of a likelihood that a patient has or will experience atrial fibrillation of the neural networks for atrial fibrillation screening as taught by Attia with the arrhythmia detection by feature description and machine learning as taught by Chakravarthy in order to determine whether the patient has an arrhythmic episode (Para 5) [and not just a risk, likelihood, or prediction] and use both feature delineation and machine learning together can improve the accuracy of detecting arrhythmias in a patient (Para 8).
Regarding Claim 4, this claim recites the limitations of Claim 1 and as to those limitations is rejected for the same basis and reasons as disclosed above. The combination of Attia and Chakravarthy discloses the following limitation that Attia further discloses:
The method of claim 1, wherein the second model is an artificial intelligence- based model trained based on normal ECG data in which the event does not occur. (Para 53 discloses this example describes the results of a study in which an artificial intelligence (Al)-model including a convolutional neural network was developed and tested to detect the electrocardiographic signature of atrial fibrillation present during normal sinus rhythm. The model was developed to process an ECG signature for a patient using a standard 10-second, 12- lead ECG recording. The example implementation was trained based on ECGs acquired from a set of patients aged 18 years or older having at least one digital, normal sinus rhythm, standard 1O-second, l2-lead ECG [wherein the claim does not limit the training to only normal ECG data, just that the two models are trained differently] acquired in the supine position at the MAYO CLINIC ECG laboratory between December 31, 1993, and July 21, 2017, with rhythm labels validated by trained personnel under cardiologist supervision. See further: Para 5.)
Regarding Claim 10, this claim recites the limitations of Claim 1 and as to those limitations is rejected for the same basis and reasons as disclosed above. The combination of Attia and Chakravarthy discloses the following limitation that Attia further discloses:
The method of claim 1, wherein the second model analyzes a structural change in an atria from normal ECG data in which the event does not occur to output the risk including a possibility of past or future occurrence of the atrial fibrillation or the atrial flutter. (Para 5 discloses due to structural irregularities of the heart, or the presence of other factors that can lead to atrial fibrillation, the patient’s ECG in normal sinus rhythm can include features, not detectable by the human eye, but which are nonetheless highly predictive of a patient that has experienced atrial fibrillation or is susceptible to atrial fibrillation. A neural network can be trained to learn these features and predict patients that have or will experience atrial fibrillation based on ECG recordings… if the prediction indicates a sufficiently high likelihood of past or expected atrial fibrillation and/or other SVTs, appropriate action may be taken such as administration of medication (e.g., anticoagulants), longer term monitoring (e.g., with an implantable loop recorder) to validate the prediction, or both. Para 8 discloses the ECG recording represented by the first neural network input can describe a normal sinus rhythm for the mammal, such that the neural network generates the atrial fibrillation prediction and/or other SVT prediction for the mammal based on features of the mammal’s normal sinus rhythm as indicated by the ECG recording.)
Regarding Claim 11, this claim recites the limitations of Claim 1 and as to those limitations is rejected for the same basis and reasons as disclosed above. The combination of Attia and Chakravarthy discloses the following limitation that Attia further discloses:
The method of claim 1, the second diagnosis result includes a result representing that at least one of the atrial fibrillation or the atrial flutter does not exist with respect to the acquired ECG data, but there is a possibility of past or future occurrence of at least one of the atrial fibrillation or the atrial flutter with respect to the acquired ECG data. (Para 5 discloses due to structural irregularities of the heart, or the presence of other factors that can lead to atrial fibrillation, the patient’s ECG in normal sinus rhythm can include features, not detectable by the human eye, but which are nonetheless highly predictive of a patient that has experienced atrial fibrillation or is susceptible to atrial fibrillation. A neural network can be trained to learn these features and predict patients that have or will experience atrial fibrillation based on ECG recordings… if the prediction indicates a sufficiently high likelihood of past or expected atrial fibrillation [past or future atrial fibrillation] and/or other SVTs, appropriate action may be taken such as administration of medication (e.g., anticoagulants), longer term monitoring (e.g., with an implantable loop recorder) to validate the prediction, or both. Para 8 discloses the ECG recording represented by the first neural network input can describe a normal sinus rhythm for the mammal, such that the neural network generates the atrial fibrillation prediction and/or other SVT prediction for the mammal based on features of the mammal’s normal sinus rhythm as indicated by the ECG recording.)
While Chakravarthy discloses the above limitations, the combination of Attia and Chakravarthy further discloses the following limitation that Chakravarthy discloses:
wherein the first diagnosis result includes a result representing that there is a possibility that at least one of the atrial fibrillation or the atrial flutter will exist with respect to the acquired ECG data, and wherein (Para 60 discloses the machine learning model implemented by the machine learning system 450 is trained with training data that includes electrocardiogram data for a plurality of patients marked with descriptive metadata. For example, during the training phase, the machine learning system 450 processes a plurality of ECG waveforms. Generally, a plurality of ECG waveforms originate from a plurality of different patients. Each ECG waveform is marked with one or more symptoms of one or more types of arrhythmias. For example, a training ECG waveform may include a plurality of segments, each segment specifying the absence of an arrhythmia or the presence of a particular class of arrhythmia (e.g., bradycardia, tachycardia, atrial fibrillation, ventricular fibrillation, or AV block). Para 61 discloses machine learning system 450 may output a preliminary determination [a possibility] that the symptom of cardiac arrhythmia represents a particular type of arrhythmia, as well as a determination indicative of confidence in the determination as well as an estimate of certainty about the determination.)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the prediction of a likelihood that a patient has or will experience atrial fibrillation of the neural networks for atrial fibrillation screening as taught by Attia with the arrhythmia detection by feature description and machine learning as taught by Chakravarthy in order to determine whether the patient has an arrhythmic episode (Para 5) [and not just a risk, likelihood, or prediction] and apply feature delineation to classify the detected cardiac arrhythmia symptoms into … a particular type.. such as atrial fibrillation (Para 43) to improve patient welfare based on the determination such as by making “adjustments to the implantable medical device” upon determining an arrhythmic symptom has occurred.
Regarding Claim 12, this claim recites the limitations of Claim 1 and as to those limitations is rejected for the same basis and reasons as disclosed above. The combination of Attia and Chakravarthy further discloses the following limitation that Attia discloses:
The method of claim 1, wherein the second model includes a deep learned based first sub-model which is trained based on a plurality of normal ECG data aligned in order of acquisition time and outputs the risk for at least one of the atrial fibrillation or the atrial flutter in response to the acquired ECG data. (Para 5 discloses a machine-learning model such as a deep neural network can be trained to process a short-recording of ECG data from a patient to generate a prediction indicating a likelihood that the patient has or will experience atrial fibrillation (e.g., paroxysmal atrial fibrillation) and/or other supraventricular tachycardia (e.g., atrial flutter, atrial tachycardia)... the neural networks described herein can detect a likelihood of atrial fibrillation and/or other supraventricular tachycardia (SVT) in a patient from an ECG recording that nominally represents a normal sinus rhythm. Due to structural irregularities of the heart, or the presence of other factors that can lead to atrial fibrillation, the patient’s ECG in normal sinus rhythm can include features, not detectable by the human eye, but which are nonetheless highly predictive of a patient that has experienced atrial fibrillation or is susceptible to atrial fibrillation. Para 23-25 discloses the ECG recording of the mammal can be recorded over a first time interval, and the method can further include: obtaining a second neural network input, the second neural network input representing a second ECG recording of the mammal that was recorded over a second time interval, the first time interval and the second time interval separated by a third time interval; and processing the first neural network input along with the second neural network input with the neural network to generate the atrial fibrillation prediction and/or other SVT prediction for the mammal… The neural network can further process, along with the first neural network input and the second neural network input, a third neural network input that indicates a length of the third time interval between the first and second time intervals when the ECG recording and the second ECG recordings were recorded, respectively. Para 75 discloses various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable sub-combination. Moreover, although features may be described above as acting in certain combinations and even initially be claimed as such, one or more features from a claimed combination can in some cases be excised from the combination, and the claimed combination may be directed to a sub-combination or variation of a sub-combination [thus it would have been obvious to utilize sub models of the embodiments of Attia].)
Regarding Claim 14, this claim recites the limitations of Claim 12 and as to those limitations is rejected for the same basis and reasons as disclosed above. The combination of Attia and Chakravarthy discloses the following limitation that Attia further discloses:
The method of claim 12, wherein the second model further includes a third sub- model which changes [the first threshold] to a second threshold, where the second threshold is lower than the first threshold, and wherein the third sub-model outputs the risk for at least one of the atrial fibrillation or the atrial flutter from the acquired ECG data. (Para 66 discloses it is noted that the threshold for a positive result (i.e., a positive classification of atrial fibrillation) could be altered to suit the purposes of different clinical applications. The current binary cutoff was chosen to balance sensitivity and specificity, but a more sensitive cutoff point might be useful in excluding patients who do not need monitoring of atrial fibrillation after stroke or a more specific cutoff point could be used for screening of otherwise healthy people with a low pretest probability of atrial fibrillation, for instance.)
While Attia discloses the above limitation, the combination of Attia and Chakravarthy disclose the following limitation that Chakravarthy further discloses:
a first threshold used in the first model which determines whether at least one of the atrial fibrillation or the atrial flutter is found in the acquired ECG data (Para 61 discloses the machine learning system 450 may output a preliminary determination that the symptom of cardiac arrhythmia represents a particular type of arrhythmia, as well as a determination indicative of confidence in the determination as well as an estimate of certainty about the determination. In response to determining that the certainty of the determination estimate is greater than a predetermined threshold (e.g., 50%, 75%, 90%, 95%, 99%), the computing device 400 determines the specific type of cardiac arrhythmia symptom.)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the prediction of a likelihood that a patient has or will experience atrial fibrillation of the neural networks for atrial fibrillation screening as taught by Attia with the arrhythmia detection by feature description and machine learning as taught by Chakravarthy in order to determine whether the patient has an arrhythmic episode (Para 5) [and not just a risk, likelihood, or prediction] and apply feature delineation to classify the detected cardiac arrhythmia symptoms into … a particular type.. such as atrial fibrillation (Para 43) to improve patient welfare based on the determination such as by making “adjustments to the implantable medical device” upon determining an arrhythmic symptom has occurred.
Regarding Claim 15, this claim recites the limitations of Claim 12 and as to those limitations is rejected for the same basis and reasons as disclosed above. The combination of Attia and Chakravarthy discloses the following limitation that Attia further discloses:
The method of claim 12, wherein the second model further includes a fourth sub-model generated based on analyzing ECG characteristic indicators including at least one of a PR-interval, an RR-interval, QRS duration, a QT interval, or a QT-corrected interval for high-risk group ECG data and low-risk group ECG data, and wherein the fourth sub-model outputs the risk for at least one of the atrial fibrillation or the atrial flutter from the acquired ECG data. (Para 5 discloses if the prediction indicates a sufficiently high likelihood of… expected atrial fibrillation [high risk group] and/or other SVTs, appropriate action may be taken such as administration of medication (e.g., anticoagulants), longer term monitoring (e.g., with an implantable loop recorder) to validate the prediction, or both. Para 40 discloses the prediction 120 is expressed as a selection of a particular classification from multiple possible classifications that represents a most likely condition of the patient 102. For example, a binary classification can be made indicating whether there is at least a threshold probability or confidence level that the patient 102…is susceptible to developing atrial fibrillation [wherein over the threshold is high risk of being susceptible to an AF, low risk is below the threshold]. Para 44 discloses the atrial fibrillation detection neural network 108 is configured to process additional (auxiliary) information in generating atrial fibrillation prediction 120. For example, the network 108 may process a second neural network input 124 in addition to the first neural network input 118 to generate the atrial fibrillation prediction 120. The second neural network input 124 represents morphological features of the patient’s ECG. Figure 5, for instance, depicts a waveform or tracing 500 for a single beat from a patient’s ECG. The waveform includes several segments including a P-wave, a QRS-complex, and a T-wave. The interface 106 can provide the ECG recording 116 to a morphological feature extractor 110 for analysis, and the extractor 110 can measure various morphological features of one or more beats (or a composite or averaged beat) from the ECG recording 116. The morphological features are parameters that describe attributes of the shape of the beat, including attributes of individual segments of the beat and attributes between segments. A number of morphological features that may be employed for atrial fibrillation prediction are labeled in Figure 5, such as a duration of the QRS-complex, am amplitude of the P-wave, R-wave, or T-wave, an area of the P-wave, QRS-complex, or T-wave, slopes of any of the waves, distances between the waves, and centers- of-gravity of the waves. The morphological feature extractor 110 provides values for the morphological features 122 to the interface 106, and the interface 106 formats them into an acceptable form for processing by atrial fibrillation detection neural network 108 as second neural network input 124. The atrial fibrillation detection neural network 108 processes the first and second inputs 118, 124 to generate the atrial fibrillation prediction 120. Para 75 discloses various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable sub-combination [and thus a sub model would have been obvious].)
Regarding Claim 16, Attia discloses:
A computing device for analyzing electrocardiogram (ECG) data, comprising:
at least one processor; and a memory, wherein the at least one processor: (Para 67 discloses embodiments of the subject matter described in this specification can be implemented as one or more computer programs, i.e., one or more modules of computer program instructions encoded on a tangible non-transitory storage medium for execution by, or to control the operation of, data processing apparatus. Para 68 discloses the term “data processing apparatus” refers to data processing hardware and encompasses all kinds of apparatus, devices, and machines for processing data, including by way of example a programmable processor, a computer, or multiple processors or computers.)
generates a second diagnosis result for estimating a risk for at least one of the atrial fibrillation or the atrial flutter by using a second model having the acquired ECG data as an input when the event is not found in the acquired ECG data. (Para 40 discloses the atrial fibrillation prediction 120 can indicate a likelihood that the patient 102… is susceptible to developing atrial fibrillation. The prediction 120 can be expressed as a probability or confidence score representing a probability or confidence that the patient 102… is susceptible to developing atrial fibrillation. Para 53 discloses this example describes the results of a study in which an artificial intelligence (Al)-model including a convolutional neural network was developed and tested to detect the electrocardiographic signature of atrial fibrillation present during normal sinus rhythm. The model was developed to process an ECG signature for a patient using a standard 10-second, 12- lead ECG recording. The example implementation was trained based on ECGs acquired from a set of patients aged 18 years or older having at least one digital, normal sinus rhythm, standard 10-second, 12-lead ECG acquired in the supine position at the MAYO CLINIC ECG laboratory between December 31, 1993, and July 21, 2017, with rhythm labels validated by trained personnel under cardiologist supervision. See further: Para 5.)
wherein the second model is pre-trained based on: distinguishing normal ECG data in which the even does not occur and abnormal ECG data in which the event occurs in a dataset with a plurality of ECG data acquired for a first subject, and temporally aligning the distinguished normal ECG data and abnormal ECG data, and generating, based on time information corresponding to each of the distinguished normal ECG data and abnormal ECG data, training data to which a label corresponding to a first class representing a high-risk group or a second class representing a low-risk group is allocated with respect to the normal ECG data. (Para 40 discloses the prediction 120 is expressed as a selection of a particular classification from multiple possible classifications that represents a most likely condition of the patient 102. For example, a binary classification can be made indicating whether there is at least a threshold probability or confidence level that the patient 102…is susceptible to developing atrial fibrillation [wherein over the threshold is high risk of being susceptible to an AF, low risk is below the threshold].Para 49 discloses a training engine 210 analyzes the predictions 208 and compares the predictions 208 to labels in the training examples 204 that indicate target predictions for each training example 204. The training engine 210 then generates updated model parameter values 214 by using an appropriate updating technique, e.g., stochastic gradient descent with backpropagation... the first component can represent an ECG of a patient under normal sinus rhythm, and the label can indicate whether that particular patient is known to have actually experienced atrial fibrillation at another time. In this way, the neural network 108 can be trained using sinus rhythm ECGs obtained in patients known and validated atrial fibrillation versus patients with no known atrial fibrillation... After training is complete, the training system 200 can provide a final set of parameter values 218 to the system 100 for use in making atrial fibrillation predictions 120. Para 57 discloses for patients who had had at least one atrial fibrillation rhythm recorded, the first recorded atrial fibrillation ECG was defined as the index ECG and the first day of the window of interest was defined as 31 days before the date of the index ECG. This window of interest was chosen with the assumption that the structural changes associated with atrial fibrillation would be present before the first recorded atrial fibrillation episode; a relatively short time interval was chosen as a conservative measure to avoid using ECGs before any structural changes developed. For patients with no ECGs with atrial fibrillation recorded, the index ECG was defined as the date of the first ECG available for that patient in the MAYO CLINIC Digital Data Vault. During training, all the ECGs in the window of interest were used to allow the network to have more samples; for the testing and validation sets, only the first normal sinus rhythm ECG within the window of interest was used to avoid repeated measurements and to mimic a real screening scenario.)
While Attia discloses the above limitations, Chakravarthy discloses the following limitations:
at least one processor; and a memory, wherein the at least one processor: (Para 49 discloses computing device 400 is an example implementation of computing system 24 of FIG. 1 . In one example, computing device 400 includes processing circuitry 402 for executing applications 424 including machine learning system 450 or any other applications described herein. Para 52 discloses storage device 408 is used to store program instructions for execution by processing circuitry 402. See Further: Paras 58, 140.)
determines, by using a first model having ECG data acquired by ECG measurement as an input, whether an event related to at least one of atrial fibrillation or atrial flutter is found in the acquired ECG data, (Para 5 discloses the computing device receives the patient's electrocardiogram data sensed by the implantable medical device. The computing device obtains a first classification of the patient's arrhythmia through the feature-based description of the electrocardiogram data. The computing device applies a machine learning model to the received electrocardiogram data to obtain a second classification of the patient's arrhythmia. As an example, the computing device uses the first and second classifications to determine whether the patient has an arrhythmic episode.)
generates a first diagnosis result for the acquired ECG data when the event is found in the acquired ECG data, and (Para 6 discloses in response to determining that an arrhythmic symptom has occurred in the patient, the computing device outputs a report indicating that an arrhythmic symptom has occurred and one or more cardiac features consistent with the arrhythmic symptom.)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the prediction of a likelihood that a patient has or will experience atrial fibrillation of the neural networks for atrial fibrillation screening as taught by Attia with the arrhythmia detection by feature description and machine learning as taught by Chakravarthy in order to determine whether the patient has an arrhythmic episode (Para 5) [and not just a risk, likelihood, or prediction] and apply feature delineation to classify the detected cardiac arrhythmia symptoms into … a particular type.. such as atrial fibrillation (Para 43) to improve patient welfare based on the determination such as by making “adjustments to the implantable medical device” upon determining an arrhythmic symptom has occurred.
Regarding Claim 17,
A computer program stored in a computer readable storage medium, wherein when the computer program is executed by at least one processor, the computer program allows the at least one processor to perform a method for analyzing electrocardiogram (ECG) data, and the method includes: (Para 67 discloses embodiments of the subject matter described in this specification can be implemented as one or more computer programs, i.e., one or more modules of computer program instructions encoded on a tangible non-transitory storage medium for execution by, or to control the operation of, data processing apparatus. Para 68 discloses the term “data processing apparatus” refers to data processing hardware and encompasses all kinds of apparatus, devices, and machines for processing data, including by way of example a programmable processor, a computer, or multiple processors or computers.)
generating a second diagnosis result for estimating a risk for at least one of the atrial fibrillation or the atrial flutter by using a second model having the acquired ECG data as an input when the event is not found in the acquired ECG data. (Para 40 discloses the atrial fibrillation prediction 120 can indicate a likelihood that the patient 102… is susceptible to developing atrial fibrillation. The prediction 120 can be expressed as a probability or confidence score representing a probability or confidence that the patient 102… is susceptible to developing atrial fibrillation. Para 53 discloses this example describes the results of a study in which an artificial intelligence (Al)-model including a convolutional neural network was developed and tested to detect the electrocardiographic signature of atrial fibrillation present during normal sinus rhythm. The model was developed to process an ECG signature for a patient using a standard 10-second, 12- lead ECG recording. The example implementation was trained based on ECGs acquired from a set of patients aged 18 years or older having at least one digital, normal sinus rhythm, standard 10-second, 12-lead ECG acquired in the supine position at the MAYO CLINIC ECG laboratory between December 31, 1993, and July 21, 2017, with rhythm labels validated by trained personnel under cardiologist supervision. See further: Para 5.)
wherein the second model is pre-trained based on: distinguishing normal ECG data in which the even does not occur and abnormal ECG data in which the event occurs in a dataset with a plurality of ECG data acquired for a first subject, and temporally aligning the distinguished normal ECG data and abnormal ECG data, and generating, based on time information corresponding to each of the distinguished normal ECG data and abnormal ECG data, training data to which a label corresponding to a first class representing a high-risk group or a second class representing a low-risk group is allocated with respect to the normal ECG data. (Para 40 discloses the prediction 120 is expressed as a selection of a particular classification from multiple possible classifications that represents a most likely condition of the patient 102. For example, a binary classification can be made indicating whether there is at least a threshold probability or confidence level that the patient 102…is susceptible to developing atrial fibrillation [wherein over the threshold is high risk of being susceptible to an AF, low risk is below the threshold].Para 49 discloses a training engine 210 analyzes the predictions 208 and compares the predictions 208 to labels in the training examples 204 that indicate target predictions for each training example 204. The training engine 210 then generates updated model parameter values 214 by using an appropriate updating technique, e.g., stochastic gradient descent with backpropagation... the first component can represent an ECG of a patient under normal sinus rhythm, and the label can indicate whether that particular patient is known to have actually experienced atrial fibrillation at another time. In this way, the neural network 108 can be trained using sinus rhythm ECGs obtained in patients known and validated atrial fibrillation versus patients with no known atrial fibrillation... After training is complete, the training system 200 can provide a final set of parameter values 218 to the system 100 for use in making atrial fibrillation predictions 120. Para 57 discloses for patients who had had at least one atrial fibrillation rhythm recorded, the first recorded atrial fibrillation ECG was defined as the index ECG and the first day of the window of interest was defined as 31 days before the date of the index ECG. This window of interest was chosen with the assumption that the structural changes associated with atrial fibrillation would be present before the first recorded atrial fibrillation episode; a relatively short time interval was chosen as a conservative measure to avoid using ECGs before any structural changes developed. For patients with no ECGs with atrial fibrillation recorded, the index ECG was defined as the date of the first ECG available for that patient in the MAYO CLINIC Digital Data Vault. During training, all the ECGs in the window of interest were used to allow the network to have more samples; for the testing and validation sets, only the first normal sinus rhythm ECG within the window of interest was used to avoid repeated measurements and to mimic a real screening scenario.)
While Attia discloses the above limitations, Chakravarthy discloses the following limitations:
a computer program stored in a computer readable storage medium, wherein when the computer program is executed by at least one processor, the computer program allows the at least one processor to perform a method for analyzing electrocardiogram (ECG) data, (Para 49 discloses computing device 400 is an example implementation of computing system 24 of FIG. 1 . In one example, computing device 400 includes processing circuitry 402 for executing applications 424 including machine learning system 450 or any other applications described herein. Para 52 discloses storage device 408 is used to store program instructions for execution by processing circuitry 402. Para 58 discloses application 422 may also include program instructions and/or data executable by computing device 400. Exemplary application(s) 422 executable by computing device 400 may include machine learning system 450. Para 140 discloses the functions may be stored as one or more instructions or code in a computer-readable medium and executed by a hardware-based processing device. A computer-readable medium is a data storage medium (e.g., RAM, ROM, EEPROM, flash memory, or any program code that can be used to store desired program code in the form of instructions or data structures that can be accessed by a computer. other media) and non-transitory computer-readable media that are tangible media.)
determining, by using a first model having ECG data acquired by ECG measurement as an input, whether an event related to at least one of atrial fibrillation or atrial flutter is found in the acquired ECG data, (Para 5 discloses the computing device receives the patient's electrocardiogram data sensed by the implantable medical device. The computing device obtains a first classification of the patient's arrhythmia through the feature-based description of the electrocardiogram data. The computing device applies a machine learning model to the received electrocardiogram data to obtain a second classification of the patient's arrhythmia. As an example, the computing device uses the first and second classifications to determine whether the patient has an arrhythmic episode.)
generating a first diagnosis result for the acquired ECG data when the event is found in the acquired ECG data; and (Para 6 discloses in response to determining that an arrhythmic symptom has occurred in the patient, the computing device outputs a report indicating that an arrhythmic symptom has occurred and one or more cardiac features consistent with the arrhythmic symptom.)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the prediction of a likelihood that a patient has or will experience atrial fibrillation of the neural networks for atrial fibrillation screening as taught by Attia with the arrhythmia detection by feature description and machine learning as taught by Chakravarthy in order to determine whether the patient has an arrhythmic episode (Para 5) [and not just a risk, likelihood, or prediction] and apply feature delineation to classify the detected cardiac arrhythmia symptoms into … a particular type.. such as atrial fibrillation (Para 43) to improve patient welfare based on the determination such as by making “adjustments to the implantable medical device” upon determining an arrhythmic symptom has occurred.
Claims 5 and 8-9 are rejected under 35 U.S.C. 103 as being unpatentable over Attia (WO 2020/086865 A1) in view of Chakravarthy (KR 2022/0006080 A), further in view of Fornwalt (US PG Pub 2021/0076960 A1).
Regarding Claim 5, this claim recites the limitations of Claim 4 and as to those limitations is rejected for the same basis and reasons as disclosed above. While Attia Para 5 discloses “if the prediction indicates a sufficiently high likelihood of… expected atrial fibrillation and/or other SVTs, appropriate action may be taken such as administration of medication (e.g., anticoagulants), longer term monitoring (e.g., with an implantable loop recorder) to validate the prediction, or both,” para 40 discloses, ”a binary classification can be made indicating whether there is at least a threshold probability or confidence level that the patient 102… is susceptible to developing atrial fibrillation,” and the second model as presented above under Claim 1 and 4, it does not fully disclose the following limitation that Fornwalt discloses:
The method of claim 4, wherein the second model is an artificial intelligence- based model trained based on first training data in which first normal ECG data is labeled with a first class representing a high-risk group, and second training data in which second normal ECG data is labeled with a second class representing a low-risk group, and wherein the normal ECG data, in which abnormal ECG data including at least one of the atrial fibrillation or the atrial flutter within a predetermined time period from an acquisition time point of the normal ECG data exists, is determined as the first normal ECG data. (Abstract discloses the method includes receiving electrocardiogram data associated with the patient, providing at least a portion of the electrocardiogram data to a trained model, receiving a risk score indicative of the likelihood the patient will suffer from atrial fibrillation within a predetermined period of time from when the electrocardiogram data was generated, and outputting the risk score to at least one of a memory or a display for viewing by a medical practitioner or healthcare administrator. Para 97 discloses the training data database 124 can include a number of ECGs and clinical data. In some embodiments, the clinical data can include outcome data, such as whether or not a patient developed AF in a time period following the day that the ECG was taken. Exemplary time periods may include 1 month [ranging to]… 10 years. Para 133 discloses an operating point can be the threshold of the model risk that was used to classify high or low risk for developing incident AF. For example, an operating point of 0.7 would indicate that model risk scores equal to and above 0.7 are considered high risk, and risk scores below 0.7 are low risk. Para 150 discloses a deep neural network that, trained on 12-lead resting ECG data, can predict incident AF within 1 year, in patients without a history of AF, with high performance (AUROC=0.85).)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination of the prediction of a likelihood that a patient has or will experience atrial fibrillation of the neural networks for atrial fibrillation screening as taught by Attia and the arrhythmia detection by feature description and machine learning as taught by Chakravarthy with the ECG based atrial fibrillation predictor systems and methods as taught by Fornwalt in order to differentiate between a high and low risk for developing incident AF to more accurately determine a patient’s risk for developing AF within a certain time period and thus improve patient outcomes by prescribing a treatment plan designed to help avoid the condition in the future (See Fornwalt Para 6).
Regarding Claim 8, this claim recites the limitations of Claim 1 and as to those limitations is rejected for the same basis and reasons as disclosed above. The combination of Attia and Chakravarthy does not fully disclose the following limitation that Fornwalt discloses:
The method of claim 1 most recent (Para 97 discloses the training data database 124 can include a number of ECGs and clinical data. In some embodiments, the clinical data can include outcome data, such as whether or not a patient developed AF in a time period following the day that the ECG was taken [wherein if the patient did not develop AF then there is no abnormal ECG data and the model outputs no risk if there is no abnormal ECG data thus reading on the non-risk group ECG of Fig. 4 of the instant application which is the low risk group]. Exemplary time periods may include 1 month, 2 months, 3 months, 4 months, 5 months, 6 months, 7 months, 8 months, 9 months, 10 months, 11 months 12 months, 1 year, 2 years, 3 years, 4 years, 5 years, 6 years, 7 years, 8 years, 9 years, or 10 years. The ECGs and clinical data can be used for training a model to generate AF risk scores. In some embodiments, the training data database 124 can include multi-lead ECGs taken over a period of time (such as ten seconds) and corresponding clinical data. In some embodiments, the trained models database 128 can include a number of trained models that can receive raw ECGs and output AF risk scores.)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination of the prediction of a likelihood that a patient has or will experience atrial fibrillation of the neural networks for atrial fibrillation screening as taught by Attia and the arrhythmia detection by feature description and machine learning as taught by Chakravarthy with the ECG based atrial fibrillation predictor systems and methods as taught by Fornwalt in order to differentiate between a high and low risk for developing incident AF to more accurately determine a patient’s risk for developing AF within a certain time period and thus improve patient outcomes by prescribing a treatment plan designed to help avoid the condition in the future (See Fornwalt Para 6).
Regarding Claim 9, this claim recites the limitations of Claim 1 and as to those limitations is rejected for the same basis and reasons as disclosed above. The combination of Attia and Chakravarthy does not fully disclose the following limitation that Fornwalt discloses:
The method of claim 1 most recent (Para 97 discloses the training data database 124 can include a number of ECGs and clinical data. In some embodiments, the clinical data can include outcome data, such as whether or not a patient developed AF in a time period following the day that the ECG was taken [wherein if the patient did not develop AF then there is abnormal ECG data and the model outputs risk if there is abnormal ECG data thus reading on the risk group ECG of Fig. 4 of the instant application which is the high risk group]. Exemplary time periods may include 1 month, 2 months, 3 months, 4 months, 5 months, 6 months, 7 months, 8 months, 9 months, 10 months, 11 months 12 months, 1 year, 2 years, 3 years, 4 years, 5 years, 6 years, 7 years, 8 years, 9 years, or 10 years. The ECGs and clinical data can be used for training a model to generate AF risk scores. In some embodiments, the training data database 124 can include multi-lead ECGs taken over a period of time (such as ten seconds) and corresponding clinical data. In some embodiments, the trained models database 128 can include a number of trained models that can receive raw ECGs and output AF risk scores.)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination of the prediction of a likelihood that a patient has or will experience atrial fibrillation of the neural networks for atrial fibrillation screening as taught by Attia and the arrhythmia detection by feature description and machine learning as taught by Chakravarthy with the ECG based atrial fibrillation predictor systems and methods as taught by Fornwalt in order to differentiate between a high and low risk for developing incident AF to more accurately determine a patient’s risk for developing AF within a certain time period and thus improve patient outcomes by prescribing a treatment plan designed to help avoid the condition in the future (See Fornwalt Para 6).
Claim 7 is rejected under 35 U.S.C. 103 as being unpatentable over Attia (WO 2020/086865 A1) in view of Chakravarthy (KR 2022/0006080 A) further in view of Nadarajah (Nadarajah, R., Wu, J., Frangi, A. F., Hogg, D., Cowan, C., & Gale, C. (2021, November 1). Predicting patient-level new-onset atrial fibrillation from population-based nationwide electronic health records: Protocol of find-AF for developing a precision medicine prediction model using Artificial Intelligence. BMJ Open. https://bmjopen.bmj.com/content/11/11/e052887).
Regarding Claim 7, this claim recites the limitations of Claim 6 and as to those limitations is rejected for the same basis and reasons as disclosed above. The combination of Attia and Chakravarthy discloses the following limitation that Attia further discloses:
The method of claim 6, the training data to which the label corresponding to the first class representing the high-risk group or the second class representing the low-risk group is allocated with respect to the normal ECG data. (Para 5 discloses the neural networks described herein can detect a likelihood of atrial fibrillation and/or other supraventricular tachycardia (SVT) in a patient from an ECG recording that nominally represents a normal sinus rhythm. Para 40 discloses the prediction 120 is expressed as a selection of a particular classification from multiple possible classifications that represents a most likely condition of the patient 102. For example, a binary classification can be made indicating whether there is at least a threshold probability or confidence level that the patient 102…is susceptible to developing atrial fibrillation [wherein over the threshold is high risk of being susceptible to an AF, low risk is below the threshold].)
While Attia discloses the above limitation, the combination of Attia and Chakravarthy does not fully disclose the following limitation that Nadarajah discloses:
wherein the generating of the training data includes: generating, additionally based on at least one of prescription record information or surgery history information allocated to the distinguished normal ECG data, (Predictor variables discloses a systematic review has highlighted 22 predictor variables included in varying combinations by 10 preceding prediction models developed to detect incident AF in the community… The potential predictors may include the following: sociodemographic variables including age, sex, ethnicity and indices of multiple deprivation…all disease conditions during follow-up, including hospitalised diseases and procedures, such as other cardiovascular diseases, diabetes mellitus, chronic lung disease, renal disease, inflammatory disease, cancer, hypothyroidism and surgical procedures… Clinical assessments… Medications prescribed including antihypertensives, statins, antidepressants, anxiolytics/hypnotics and antipsychotics… lifestyle factors, and biomarkers. Development and external validation of model discloses preprocessed patient-level data in CPRD-GOLD will be randomly split into an 80:20 ratio to create training and testing samples. The split ratio is not a significant factor, given the volume of the sample size. The model parameters and dropout rate will be chosen through a grid search and 10-fold cross-validation will be used (i.e., 10% of the training data will be randomly selected as the cross-validation set).)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination of the prediction of a likelihood that a patient has or will experience atrial fibrillation of the neural networks for atrial fibrillation screening as taught by Attia and the arrhythmia detection by feature description and machine learning as taught by Chakravarthy with predictor variables as taught by Nadarajah in order to capture the full potential of deep learning (Nadarajah predictor variables).
Claim 13 is rejected under 35 U.S.C. 103 as being unpatentable over Attia (WO 2020/086865 A1) in view of Chakravarthy (KR 2022/0006080 A) further in view of Chang (Chang, C.-H., Lin, C.-S., Luo, Y.-S., Lee, Y.-T., & Lin, C. (2022, February 8). Electrocardiogram-based heart age estimation by a deep learning model provides more information on the incidence of cardiovascular disorders. Frontiers in cardiovascular medicine. https://pmc.ncbi.nlm.nih.gov/articles/PMC8860826/.)
Regarding Claim 13, this claim recites the limitations of Claim 12 and as to those limitations is rejected for the same basis and reasons as disclosed above. Attia discloses in para 75, “various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable sub-combination [and thus a sub model would have been obvious],” the combination of Attia and Chakravarthy does not fully disclose the following limitation that Chang discloses:
The method of claim 12, wherein the second model further includes a second sub-model which estimates the heart age from the acquired ECG data and compares the estimated heart age with an actual age of a subject to output the risk for at least one of the atrial fibrillation or the atrial flutter. (Introduction discloses we developed a deep learning model (DLM) to predict the biological age via ECG to explore its contribution to future cardiovascular diseases (CVDs). Results discloses the higher difference between ECG-age and chronological age was related to more comorbidities and abnormal ECG rhythm. The cases with the difference of more than 7 years had higher risk on the all-cause mortality… AF (HR: 2.38, 95% CI: 1.86–3.04… The external validation sets also validated that an ECG-age >7 years compare to chronologic age had 3.16-fold risk (95% CI: 1.72–5.78) and 1.59-fold risk (95% CI: 1.45–1.74) on all-cause mortality in SaMi-Trop and CODE15 cohorts. The ECG-age significantly contributed additional information on heart failure, stroke, coronary artery disease, and atrial fibrillation predictions after considering all the known risk factors. Fig. 7A discloses the high residual group had an increased risk for the all-cause mortality (HR 1.61, 95%CI 1.23–2.12), CV-caused mortality (HR 3.49, 95%CI 1.74–7.01), newly onset HF (HR 2.79, 95%CI 2.25–3.45), and newly onset AF (HR 2.38, 95%CI 1.86–3.04).)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination of the prediction of a likelihood that a patient has or will experience atrial fibrillation of the neural networks for atrial fibrillation screening as taught by Attia and the arrhythmia detection by feature description and machine learning as taught by Chakravarthy with the ECG age as taught by Chang to further identify the parameters that contribute to future cardiovascular diseases (CVDs)… such as atrial fibrillation (See Chang Introduction).
Response to Arguments
Applicant’s arguments filed 12/29/2025 with respect to 35 U.S.C. § 112(b) have been fully considered. In regards to the “latest ECG” of claims 8 and 9, the Applicant has amended the claim language to read, “most recent ECG” and as such, this previous 112(b) rejection has been withdrawn.
In regards to claim 9, the Applicant amended the dependency so that claim 9 now depends on Claim 1. However, as previously presented, the claim from which claim 9 depends on (previously claim 6, now amended to be claim 1) does not disclose a first time point or a first predetermined time period. As such, claim 9 should disclose a first and second time point and predetermined time period instead of a second and third time point and predetermined time period since claims 8 and 9 are separate embodiments of the invention. Therefore, the previous 112(b) rejection is maintained with the claim from which claim 9 is dependent on having been updated.
Applicant’s arguments filed 12/29/2025 with respect to 35 U.S.C. § 101 have been fully considered, but are not persuasive.
Applicant argues that the claims are not directed to a mental process or a mathematical concept. The Examiner submits that the abstract idea was not characterized as being directed to a mental process or a mathematical concept. The claimed invention was characterized as falling under Certain Methods of Organizing Human activity (see Non-Final Office Action dated 09/29/2025 at Pg. 5). As such, this argument cannot be persuasive.
Further, the Applicant argues that the amended independent claims implement a technical improvement, thereby providing a practical application. The Examiner respectfully disagrees. MPEP 2106.04(d)(1) states “the word ‘improvements’ in the context of this consideration is limited to improvements to the functioning of a computer or any other technology/technical field, whether in Step 2A Prong Two or in Step 2B.” Here, there is no improvement to the computer (the computer is performing as expected) nor is there an improvement to another technology. The Applicant argues that the claims present a specific technical procedure for pre-training data so that the model can effectively learn subtle signs of abnormality existing within the electrocardiogram. However, this is not an improvement to how the model is functioning nor to how the model is being trained. Training a model with labeled data does not present an inventive concept. The claimed invention is using a computer as a tool to run a model to implement an abstract idea and any improvement present is an improvement to the abstract idea of, to paraphrase, analyzing patient ECG data and outputting a result for a doctor. Because neither type of improvement is present in the claims, an improvement to technology is not present and there is no practical application.
Applicant’s arguments filed 12/29/2025 with respect to 35 U.S.C. § 103 have been fully considered, but are not persuasive. Applicant argues that the cited references do not read on the “specific pre-processing process or calculating the ‘temporal distance’ between normal electrocardiogram and abnormal electrocardiogram within the data of a single subject, and dynamically allocating high-risk/low-risk labels for each individual normal electrocardiogram data accordingly to generate training data,” It is noted that “calculating the ‘temporal distance’ between normal electrocardiogram and abnormal electrocardiogram within the data of a single subject,” is not claimed and as such this is not persuasive.
Specifically, the Applicant argues that, “Attia corresponds to the model performing binary classification… after the training of the model is completely finished…” The Examiner respectfully disagrees. As previously presented, Attia Para 49 discloses, “The training neural network subsystem 206 can generate, for each training example 204, an atrial fibrillation prediction 208. A training engine 210 analyzes the predictions 208 and compares the predictions 208 to labels in the training examples 204 that indicate target predictions for each training example 204. The training engine 210 then generates updated model parameter values 214 by using an appropriate updating technique [thus disclosing the model was trained with labeled data, and then updated by using an appropriate updating technique by comparing the predictions 208 to labels and as such does not disclose outputting a risk level only after the training of the model is completely finished], e.g., stochastic gradient descent with backpropagation. The training engine 210 can then update the collection of model parameter values 214 using the updated model parameter values 212. For example, each training example 204 can include a first component representing a single- or multi-lead ECG recording of a patient and a label indicating a target atrial fibrillation prediction. The first component can represent an ECG of a patient under normal sinus rhythm, and the label can indicate whether that particular patient is known to have actually experienced atrial fibrillation at another time. In this way, the neural network 108 can be trained using sinus rhythm ECGs obtained in patients known and validated atrial fibrillation versus patients with no known atrial fibrillation [thus disclosing generating training data].” As previously presented, para 40 discloses, “a binary classification can be made indicating whether there is at least a threshold probability or confidence level that the patient 102… is susceptible to developing atrial fibrillation [wherein over the threshold is high risk of being susceptible to an AF, low risk is below the threshold]”. Thus, the data is labelled or trained as either being high risk (over the threshold) or low risk (below the threshold), which is consistent to the claimed “high-risk group” and “low-risk group”
To clarify, para 43 discloses, “The atrial fibrillation detection neural network 108 is configured to process the first neural network input 118 and to generate atrial fibrillation prediction 120 based on the first neural network input 118. The neural network 108 can include multiple layers of operations that have been trained to discern an atrial fibrillation condition of a patient based on ECG recordings of a patient’s normal sinus rhythm,” wherein para 49 discloses the training of the neural network 108. See further the Attia Abstract which discloses, “processing the first neural network input with a neural network to generate an atrial fibrillation prediction for the mammal,” as such the prediction of Attia is further clarified as a prediction for atrial fibrillation. Para 50 discloses the training system 200 can provide a final set of parameter values 218 to the system 100 for use in making atrial fibrillation predictions 120.” As such, the predictions 208 reads on distinguishing normal ECG data in which the event does not occur and abnormal ECG data in which the event occurs in a dataset with a plurality of ECG data acquired for a first subject as presented above.
The Applicant then argues that the remaining prior art references do not disclose the amended claim language (from the previous claim 6). These arguments are rendered moot as the Examiner cites Attia to read on the amended claim language, as previously presented under the now canceled claim 6, and as now clarified above.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SARA J MORICE DE VARGAS whose telephone number is (703)756-4608. The examiner can normally be reached M-F 8:30-5:30 pm.
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.
/SARA JESSICA MORICE DE VARGAS/Examiner, Art Unit 3681
/PETER H CHOI/Supervisory Patent Examiner, Art Unit 3681