DETAILED ACTION
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
The amendments and remarks filed on 06JAN2026 have been entered and considered.
Claims 1-20 are currently pending
Claims 1, 8, & 15 have been amended.
No claims have been canceled, withdrawn, or added.
No new matter has been added.
Claims 1-20 are under examination.
Response to Arguments
Applicant's arguments filed 06JAN2026 regarding the rejections under 35 USC 101 have been fully considered and have been found to be not persuasive. Parts deemed not persuasive discussed below:
Applicant states (see Page 9 of the Remarks):
Applicant has amended the independent claims to additionally recite features that specify "the course of treatment comprising a modification to diet or medication prescribed in another course of treatment for another user." Applicant submits that these features amount to the "additional elements" of Step 2A, Prong Two that integrate the allegedly abstract idea into a practical application, making the amended claims patent-eligible.
The examiner is not persuaded as the amendments amount to merely a task the clinician is capable of performing without the use of the device, as no active treatment regime changes are occurring based on the determinations. According to MPEP 2106.04(d)(2), in order to qualify as a "treatment" or "prophylaxis" limitation for purposes of this consideration, the claim limitation in question must affirmatively recite an action that effects a particular treatment or prophylaxis for a disease or medical condition. An example of such a limitation is a step of "administering amazonic acid to a patient" or a step of "administering a course of plasmapheresis to a patient." If the limitation does not actually provide a treatment or prophylaxis, e.g., it is merely an intended use of the claimed invention or a field of use limitation, then it cannot integrate a judicial exception under the "treatment or prophylaxis" consideration. For example, a step of "prescribing a topical steroid to a patient with eczema" is not a positive limitation because it does not require that the steroid actually be used by or on the patient, and a recitation that a claimed product is a "pharmaceutical composition" or that a "feed dispenser is operable to dispense a mineral supplement" are not affirmative limitations because they are merely indicating how the claimed invention might be used. Evidence is not provided as to how this function amounts to additional element that integrates the abstract idea and therefore, the rejection is being maintained.
Applicant's arguments filed 06JAN2026 regarding the rejections under 35 USC 103 have been fully considered and have been found to be not persuasive. Parts deemed not persuasive discussed below:
Applicant states (see Pages 9-10 of the Remarks):
In rejecting claim 1, the Office cited Gopalakrishnan, in particular " [0015] and [0022], as allegedly teaching "prescribing a course of treatment of the LVH individually tailored to the first user." Office action, 12-13. Without acquiescing to the analysis in the Office action, Applicant has amended the independent claims to additionally require that the individually tailored course of treatment comprise a modification to diet or medication prescribed "in another course of treatment for another user.
Different from this limitation, Gopalakrishnan teaches improving the generated cardiac health score of a user based on the user's own biometric data, while being silent on modifying diet or medication prescribed for another user. Gopalakrishnan, [0015]. Therefore, Gopalakrishnan fails to disclose or otherwise suggest the limitations at issue. With respect to Kwong and Angelaki, the Office has not shown any disclosure of these references about the features at issue, e.g., prescribing a course of treatment for a user based on modifying diet or medication prescribed for another user.
The examiner is not persuaded as the limitation may be found in Gopalakrishnan ¶0076 “The patient may avoid a future healthcare issue, as instructed or recommended by the system, by modifying their behavior, habits or by taking any course of action, including but not limited to taking a medication, drug or adhering to a diet or exercise program, which may be a predetermined course of action recommended by the system independent of any analysis of the ECG data, and/or may also result from insights learned through this system and method as described herein.”. This citation describes the prescription of a modified diet or medication plan. Additionally, it is stated that the modifications “may be a predetermined course of action recommended by the system independent of any analysis of the ECG data, and/or may also result from insights learned through this system and method as described herein”. This shows that a secondary user is benefiting from patterns and treatments established for other users which were used in the training set to originally train the learning model of Gopalakrishnan such as cited in ¶0106. Therefore, the examiner is maintaining that the references teach the claim limitations as cited 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 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1
The claims recite:
Claim 1 is to a method.
Claim 8 is to a system.
Claim 15 is to a method.
Therefore, claims 1-20 are directed to a statutory category of invention.
Step 2A, Prong One
Regarding claims 1, 8, & 15 the limitations of “analyzing, using the ML model, the ECG data and the blood pressure data for each of the number of individuals to generate an output”; “analyzing, the user data to determine whether (LVH) is present in the first user based on the first ECG data and the first blood pressure data” ; “analyze the first ECG data to obtain an ECG analysis result”, “determine whether LVH is present in the first user based on the combination of the first ECG data and the first blood pressure data”, “augment the ECG analysis result with the first blood pressure data to obtain a combination of the first ECG data and the first blood pressure data”, “and in response to the indication, prescribing, by the ML model and based on the group, a course of treatment of the LVH individually tailored to the first user, the course of treatment comprising a modification to diet or medication prescribed in another course of treatment for another user”, “determine, based on analyzing the user data, a group to which the first user belongs;”, and “augmenting the ECG analysis result with the first blood pressure data to obtain a combination of the first ECG data and the first blood pressure data” are mental processes. The limitations as drafted, covers performance of the limitations that can be performed by a human using a pen and paper under the broadest reasonable interpretation standard. For example, determining heart disease based on the user’s biometrics encompasses nothing more than a user plotting ECG data and blood pressure on a piece of paper or other biometric data sets, and comparing to one another for abnormalities with various analysis. Similarly, indicating this finding is nothing more than finding abnormalities as graphed, and verbally or manually alerting the user of such findings. If claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in human mind or by a human using a pen and paper, then it falls within the “Mental Processes” grouping of abstract ideas. See MPEP 2106.04(a)(2)(III). Limitation “backpropagation to update one or more weights of the ML model based on a difference between the output for each of the number of individuals and the corresponding LVH diagnosis label” is no more than a mathematical concept. The mathematical concepts grouping is defined as mathematical relationships, mathematical formulas or equations, and mathematical calculations. MPEP 2106.04(a)(2)(I)
Step 2A, Prong Two
This judicial exception is not integrated into a practical application. In particular, the Claims 1, 8, & 15 recites additional elements of: “outputting an indication that the LVH is present in the first user and a severity of the LVH in the first user”, which is insignificant post-solution activity of data reporting. Elements “providing a training data set comprising electrocardiogram (ECG) data, blood pressure data and a corresponding left ventricle hypertrophy (LVH) diagnosis label for each a number of individuals”, “receiving user data of a first user, the user data comprising first ECG data and first blood pressure data of the first user”, “the user data comprising (first ECG) data and first blood pressure data of the first user”, and “receiving user data of a first user” amount to insignificant, extra-solution activity (data gathering) and “processing device” (Claim 1) is recited at high levels of generality that is amounts to generic computer implementation of the abstract idea.). This pre-solution activity of obtaining user data comprising electrocardiogram (ECG) data and blood pressure data using biosensors and all uses of the recited judicial exception require the insignificant pre-solution activity of data gathering. The claim elements “ML model”, amounts to generic computer implementation of the abstract idea, since an ML model is a computer automation that mimics the human thinking process. Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea, and the ML model; Cloud storage system; and non-transitory computer-readable medium are computer structures for generic computer implementation of the abstract idea additionally does not integrate into a practical application. The claim is directed to an abstract idea.
Step 2B
The independent claims and claims 1, 8, & 15 do not include additional elements that are sufficient to amount to significantly more than the judicial exception. Claims 1, 8, & 15 recites additional elements of: “outputting an indication that the LVH is present in the first user and a severity of the LVH in the first user”, which is insignificant post-solution activity of data reporting. Elements “providing a training data set comprising electrocardiogram (ECG) data, blood pressure data and a corresponding left ventricle hypertrophy (LVH) diagnosis label for each a number of individuals”, “receiving user data of a first user, the user data comprising first ECG data and first blood pressure data of the first user”, “the user data comprising (first ECG) data and first blood pressure data of the first user”, “receiving user data of a first user” and “processing device” (Claim 1) are recited at high levels of generality (i.e. The commands themselves do not have any supporting structures to perform such commands). This pre-solution activity of obtaining user data comprising electrocardiogram (ECG) data and blood pressure data using biosensors and all uses of the recited judicial exception require the insignificant pre-solution activity of data gathering. The claim elements “ML model”, amounts to generic computer implementation of the abstract idea, since an ML model is a computer automation that mimics the human thinking process. Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea, and the ML model; Cloud storage system; and non-transitory computer-readable medium are computer structures for generic computer implementation of the abstract idea additionally does not amount to significantly more. The claim is directed to an abstract idea.
Regarding dependent claims 2-7, as discussed above with respect to integration of the abstract idea into a practical application, the additional elements as claimed amounts to no more than mere pre-solution activity of data gathering (Claim 4) , generic computer implementation of the abstract idea (Claim 7), and further defining an abstract idea (Claims 2-3 & 5-6), which does not amount of an inventive concept. Therefore, the claims are not patent eligible.
Regarding dependent claims 9-14, as discussed above with respect to integration of the abstract idea into a practical application, the additional elements as claimed amounts to no more than mere pre-solution activity of data gathering (Claim 11) , generic computer implementation of the abstract idea (Claim 14), and further defining an abstract idea (Claims 9-10 & 12-13), which does not amount of an inventive concept. Therefore, the claims are not patent eligible.
Regarding dependent claims 16-20, as discussed above with respect to integration of the abstract idea into a practical application, the additional elements as claimed amounts to no more than mere pre-solution activity of data gathering (Claim 18) , and further defining an abstract idea (Claims 16-17 & 19-20), which does not amount of an inventive concept. Therefore, the claims are not patent eligible.
Therefore, claims 1-20 are directed to an abstract idea without significantly more.
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.
Claims 1-20 are rejected under 35 U.S.C. 103 as being anticipated by Gopalakrishnan et al. (US Publication No. 20150164349; Previously Cited), in view of Kwong (US Patent No. 5792066; Previously Cited), and Angelaki E et al.. Detection of abnormal left ventricular geometry in patients without cardiovascular disease through machine learning: An ECG-based approach. J Clin Hypertens (Greenwich). 2021 May;23(5):935-945. doi: 10.1111/jch.14200. Epub 2021 Jan 28. PMID: 33507615; PMCID: PMC8678829.; Previously Cited). Rejection Maintained.
Regarding claim 1, Gopalakrishnan discloses a method comprising: providing a training data set comprising electrocardiogram (ECG) data, blood pressure data (Gopalakrishnan ¶0016 “The population data may be collected from a plurality of the heart rate monitoring enabled portable computing devices or accessories provided to a plurality of users”; ¶0074 “As described herein, various parameters may be included in the database along with the ECG data. These may include the patient's age, gender, weight, blood pressure, medications, behaviors, habits, activities, food consumption, drink consumption, drugs, medical history and other factors that may influence a patient's ECG signal. The additional parameters may or may not be used in the comparison of the changes in ECG signal over time and circumstances.”); training a machine learning model (ML) by: analyzing, using the ML model, the ECG data and the blood pressure data for each of the number of individuals to generate an output for each of the number of individuals (Gopalakrishnan ¶0016 “the data may be used to train the machine learning algorithm to extract one or more features from any continuously measured heart rate data and identify atrial fibrillation or other conditions therefrom. After the machine learning algorithm has been trained, the machine learning algorithm may recognize atrial fibrillation from the continuously measured heart rate data of a new user who has not yet been identified as having atrial fibrillation or other heart conditions. One or more of training population data or the trained machine learning algorithm may be provided on a central computing device (e.g. be stored on a non-transitory computer readable medium of a server) which is in communication with the local computing devices of the users and the application executed thereon (e.g. through an Internet or an intranet connection.)”); receiving user data of a first user, the user data comprising first ECG data and first blood pressure data of the first user (Gopalakrishnan ¶0015 “The biometric data may comprise one or more of an electrocardiogram (ECG), dietary information, stress level, activity level, gender, height, weight, age, body fat percentage, blood pressure, results from imaging scans, blood chemistry values, or genotype data.”, ¶0102 ); analyzing, by a processing device executing a model, the user data to determine whether an abnormality is present in the first user based on the first ECG data and the first blood pressure data (Gopalakrishnan ¶0103 “In a step 912, a cardiac health score is generated. The cardiac health score can be generated by considering and weighing one or more influencing factors including the incidence of atrial fibrillation or arrhythmia as detected by the handheld ECG monitor, the heart rate of the user or subject, the activity of the user or subject, hours of sleep and rest of the user or subject, blood pressure of the user or subject, etc. Often, the incidence of atrial fibrillation or arrhythmia will be weighed the most. The cardiac health score may be generated by a physician or a machine learning algorithm provided by the remote server or cloud-based service 113, for example”, ¶0102, Figure 6; ¶0010 “These parameters and further parameters may be analyzed to detect and/or predict one or more of atrial fibrillation, tachycardia, bradycardia, bigeminy, trigeminy, or other cardiac conditions. A quantitative heart health score may also be generated from the determined parameters. One or more of the heart health score, detected heart conditions, or recommended user action items based on the heart health score may be displayed to the user through a display of the portable computing device.”); determine, based on analyzing the user data, a group to which the first user belongs (Gopalakrishnan ¶0022 “The user may also input user-related health data such as age, height, weight, body mass index (BMI), diet, sleep levels, rest levels, or stress levels. One or more of these physiological signals and/or parameters may be combined with the heart rate data to detect atrial fibrillation or other conditions. The machine learning algorithm may be configured to identify atrial fibrillation or other conditions in response to heart rate data in combination with one or more of the other physiological signals and/or parameters for instance”; ¶0083 “Depending on the severity of the arrhythmia detected, the heart score may be calculated or assigned within the ranges according to the table below in Table 2.”; ¶0078 “In step 408, the subset of ECG data can be analyzed using a machine learning algorithm, which can assign a risk level to the ECG data in step 410.”); and in response to determining that an abnormality is present in the first user, outputting an indication that the abnormality is present in the first user and a severity of the abnormality in the first user (Gopalakrishnan ¶0012 “The dashboard includes a heart score that can be calculated in response to data from the user such as their ECG and other personal information such as age, gender, height, weight, body fat, disease risks, etc.”, Table 2 additionally shows the heart score with a range of defined values to show severity of detected signals as it relates to the values.); and in response to the indication, prescribing, by the ML model and based on the group, a course of treatment individually tailored to the first user (Gopalakrishnan ¶0015 “Another aspect of the present disclosure provides a method for managing cardiac health. Biometric data of a user may be received. A cardiac health score may be generated in response to the received biometric data. One or more recommendations or goals for improving the generated cardiac health score may be displayed to the user. The biometric data may comprise one or more of an electrocardiogram (ECG), dietary information, stress level, activity level, gender, height, weight, age, body fat percentage, blood pressure, results from imaging scans, blood chemistry values, or genotype data.”; ¶0022 “The user may also input user-related health data such as age, height, weight, body mass index (BMI), diet, sleep levels, rest levels, or stress levels. One or more of these physiological signals and/or parameters may be combined with the heart rate data to detect atrial fibrillation or other conditions. The machine learning algorithm may be configured to identify atrial fibrillation or other conditions in response to heart rate data in combination with one or more of the other physiological signals and/or parameters for instance. Triggers or alerts may be provided to the user in response to the measured physiological signals and/or parameters. Such triggers or alerts may notify the user to take corrective steps to improve their health or monitor other vital signs or physiological parameters”)the course of treatment comprising a modification to diet or medication prescribed in another course of treatment for another user (Gopalakrishnan ¶0076 “The patient may avoid a future healthcare issue, as instructed or recommended by the system, by modifying their behavior, habits or by taking any course of action, including but not limited to taking a medication, drug or adhering to a diet or exercise program, which may be a predetermined course of action recommended by the system independent of any analysis of the ECG data, and/or may also result from insights learned through this system and method as described herein. “); wherein analyzing the user data based on the first ECG data and the first blood pressure data comprises: analyzing the first ECG data to obtain an ECG analysis result (Gopalakrishnan ¶0052 “The machine learning based algorithm(s) may allow software application(s) to identify patterns and/or features of the R-R interval data and/or the raw heart rate signals or data to predict and/or detect atrial fibrillation or other arrhythmias.”).; augmenting the ECG analysis result with the first blood pressure data to obtain a combination of the first ECG data and the first blood pressure data (Gopalakrishnan ¶0103 “In a step 912, a cardiac health score is generated. The cardiac health score can be generated by considering and weighing one or more influencing factors including the incidence of atrial fibrillation or arrhythmia as detected by the handheld ECG monitor, the heart rate of the user or subject, the activity of the user or subject, hours of sleep and rest of the user or subject, blood pressure of the user or subject, etc. Often, the incidence of atrial fibrillation or arrhythmia will be weighed the most. The cardiac health score may be generated by a physician or a machine learning algorithm provided by the remote server or cloud-based service 113, for example”; Showing the weighted predictions which can involve both ECG data and blood pressure data); and determining whether an abnormality is present in the first user based on the combination of the first ECG data and the first blood pressure data (Gopalakrishnan ¶0103 “In a step 912, a cardiac health score is generated. The cardiac health score can be generated by considering and weighing one or more influencing factors including the incidence of atrial fibrillation or arrhythmia as detected by the handheld ECG monitor, the heart rate of the user or subject, the activity of the user or subject, hours of sleep and rest of the user or subject, blood pressure of the user or subject, etc. Often, the incidence of atrial fibrillation or arrhythmia will be weighed the most. The cardiac health score may be generated by a physician or a machine learning algorithm provided by the remote server or cloud-based service 113, for example”, ¶0102, Figure 6; ¶0010 “These parameters and further parameters may be analyzed to detect and/or predict one or more of atrial fibrillation, tachycardia, bradycardia, bigeminy, trigeminy, or other cardiac conditions. A quantitative heart health score may also be generated from the determined parameters. One or more of the heart health score, detected heart conditions, or recommended user action items based on the heart health score may be displayed to the user through a display of the portable computing device.”).
Gopalakrishnan does not disclose using backpropagation to update one or more weights of the ML model based on a difference between the output for each of the number of individuals and the corresponding diagnosis label. Kwong in a similar field of cardiology also teaches using backpropagation to update one or more weights of the ML model based on a difference between the output for each of the number of individuals and the corresponding LVH diagnosis label (Kwong Column 9 Lines 33-49 “Method step 78 shows a decision point. In this decision point it is determined whether the one or more wave amplitude ratios created in method step 74 used in conjunction with the generated decision matrices generated in method step 76 yields accurate enough results to be useful in a clinical sense. If the answer is yes, then one proceeds to method step 80 which is the end of the process. Otherwise, one proceeds back to method step 74 wherein another one or more wave amplitude ratios is created and again used to generate another statistical decision matrix in method step 76. The loop comprised of method steps 74, 76 and 78 continues until the process yields at least one wave amplitude ratio-decision matrix combination which will generate diagnosis of sufficient accuracy to be useful in a clinical sense, as is depicted in method step 78. Once this criterion has been satisfied one proceeds to method step 80 and the process stops.” Which describes the methods of performing backpropagation; Figure 4).
Before the effective filing date, it would have been obvious to a person of skill in the art to modify Gopalakrishnan in view Kwong by combining the cardiac analysis system of Gopalakrishnan with Kwong’s analysis system using backpropagation to update one or more weights of the ML model based on a difference between the output for each of the number of individuals and the corresponding diagnosis label. It would have been obvious to one having ordinary skill in the art at the time the invention was made to use explicitly LHV data and backpropagation to update one or more weights of the system’s ML model based on a difference between the output for each of the number of individuals and the corresponding diagnosis label for the purpose of optimized cardiac health monitoring. The addition of backpropagation techniques as taught in Kwong serves a method for reducing diagnosis and monitoring errors that may come from noisy data sets and the initial training of a machine learning model. This will provide and optimized diagnosis system that can self-optimize as the system takes cardiac data and uses it as new training sets over time.
Gopalakrishnan in view of Kwong does not disclose providing a corresponding left ventricle hypertrophy (LVH) diagnosis label for each a number of individuals. Angelaki in a similar field of ECG based left ventricular hypertrophy detection teaches providing a corresponding left ventricle hypertrophy (LVH) diagnosis label for each a number of individuals (Angelaki Introduction “This study was designed to test the hypothesis that a 12-lead ECG, a routine and inexpensive screening procedure, can provide further accuracy in detecting abnormal LVG, using ML methods, even at the early stages before the onset of LVH, in a population without established CVD. We also seek to understand which features contribute to the ML model's decisions, by calculating global feature importance and feature interactions”). Before the effective filing date, it would have been obvious to a person of skill in the art to modify Gopalakrishnan in view of Kwong with Angelaki by combining the cardiac analysis and abnormality detection system with Angelaki’ s analysis system ECG based left ventricular hypertrophy detection system for providing a corresponding left ventricle hypertrophy (LVH) diagnosis label for each a number of individuals to provide and optimized diagnosis system that can provide a diagnosis and/or detect a wider range of abnormalities.
Regarding claim 2, Gopalakrishnan additionally discloses wherein the user data of the first user further comprises characteristics of the first user and the ML model determines whether the abnormality is present in the first user based further on the characteristics of the first user (Gopalakrishnan ¶0015 “The biometric data may comprise one or more of an electrocardiogram (ECG), dietary information, stress level, activity level, gender, height, weight, age, body fat percentage, blood pressure, results from imaging scans, blood chemistry values, or genotype data.”, ¶0074 “As described herein, various parameters may be included in the database along with the ECG data. These may include the patient's age, gender, weight, blood pressure, medications, behaviors, habits, activities, food consumption, drink consumption, drugs, medical history and other factors that may influence a patient's ECG signal. The additional parameters may or may not be used in the comparison of the changes in ECG signal over time and circumstances”, ¶0012).
Gopalakrishnan in view of Kwong does not disclose detecting left ventricle hypertrophy (LVH) for an individual. Angelaki in a similar field of ECG based left ventricular hypertrophy detection teaches detecting left ventricle hypertrophy (LVH) for an individual (Angelaki Introduction “This study was designed to test the hypothesis that a 12-lead ECG, a routine and inexpensive screening procedure, can provide further accuracy in detecting abnormal LVG, using ML methods, even at the early stages before the onset of LVH, in a population without established CVD. We also seek to understand which features contribute to the ML model's decisions, by calculating global feature importance and feature interactions”).
Before the effective filing date, it would have been obvious to a person of skill in the art to modify Gopalakrishnan in view of Kwong with Angelaki by combining the cardiac analysis and abnormality detection system with Angelaki’ s analysis system ECG based left ventricular hypertrophy detection system for detecting left ventricle hypertrophy (LVH) for an individual to provide and optimized diagnosis system that can provide a diagnosis and/or detect a wider range of abnormalities.
Regarding claim 3, Gopalakrishnan additionally discloses wherein determining whether an abnormality is present in the first user comprises: comparing the first ECG data to ECG criteria comprising a set of ranges, wherein the set of ranges indicate whether the abnormality is present and a severity of the abnormality if it is present (Gopalakrishnan ¶0070 “For example, the P wave and heart rate can be extracted and analyzed to identify atrial fibrillation, where the absence of P waves and/or an irregular heart rate may indicate atrial fibrillation.”, ¶0071); comparing the first blood pressure data to blood pressure criteria defining a set of systolic and diastolic pressure ranges, each of the set of systolic and diastolic pressure ranges indicating a likelihood that the abnormality is present (Gopalakrishnan Page 6 Table 1 shows thresholds that are of concern including blood pressure.); and determining whether the abnormality is present in the first user and a severity of the abnormality if it is present, based on a range that the ECG data is within and a systolic and diastolic pressure range that the blood pressure data is within (Gopalakrishnan ¶0074 “As described herein, various parameters may be included in the database along with the ECG data. These may include the patient's age, gender, weight, blood pressure, medications, behaviors, habits, activities, food consumption, drink consumption, drugs, medical history and other factors that may influence a patient's ECG signal. The additional parameters may or may not be used in the comparison of the changes in ECG signal over time and circumstances”).
Gopalakrishnan in view of Kwong does not disclose detecting left ventricle hypertrophy (LVH) for an individual. Angelaki in a similar field of ECG based left ventricular hypertrophy detection teaches detecting left ventricle hypertrophy (LVH) for an individual (Angelaki Introduction “This study was designed to test the hypothesis that a 12-lead ECG, a routine and inexpensive screening procedure, can provide further accuracy in detecting abnormal LVG, using ML methods, even at the early stages before the onset of LVH, in a population without established CVD. We also seek to understand which features contribute to the ML model's decisions, by calculating global feature importance and feature interactions”).
Before the effective filing date, it would have been obvious to a person of skill in the art to modify Gopalakrishnan in view of Kwong with Angelaki by combining the cardiac analysis and abnormality detection system with Angelaki’ s analysis system ECG based left ventricular hypertrophy detection system for detecting left ventricle hypertrophy (LVH) for an individual to provide and optimized diagnosis system that can provide a diagnosis and/or detect a wider range of abnormalities.
Regarding claim 4, Gopalakrishnan additionally discloses monitoring, by the ML model, further user data of the first user, the further user data captured while the first user is undergoing the course of treatment from among a plurality of courses of treatment for abnormality and monitoring, by the ML model, further user data of each of a plurality of second users, wherein the further user data of each second user is monitored while the second user is undergoing one of the plurality of courses of treatment for abnormality (Gopalakrishnan ¶0073 “Analysis of the time following the abnormality, adverse event or disease state can provide information regarding the efficacy of treatments and/or provide the patient or physician information regarding disease progression, such as whether the patient's condition in improving, worsening or staying the same. The diagnosis and determination can also be used for indexing by, for example, including it in the metadata associated with the corresponding ECG data.”, ¶103 “A plurality of users and subject may concurrently use the cardiac health and/or rhythm management system 100 and the machine learning algorithm may, for example, consider population data and trends to generate an individual user or subject's cardiac health score.”, ¶0104-¶0105).
Gopalakrishnan in view of Kwong does not disclose detecting left ventricle hypertrophy (LVH) for an individual. Angelaki in a similar field of ECG based left ventricular hypertrophy detection teaches detecting left ventricle hypertrophy (LVH) for an individual (Angelaki Introduction “This study was designed to test the hypothesis that a 12-lead ECG, a routine and inexpensive screening procedure, can provide further accuracy in detecting abnormal LVG, using ML methods, even at the early stages before the onset of LVH, in a population without established CVD. We also seek to understand which features contribute to the ML model's decisions, by calculating global feature importance and feature interactions”).
Before the effective filing date, it would have been obvious to a person of skill in the art to modify Gopalakrishnan in view of Kwong with Angelaki by combining the cardiac analysis and abnormality detection system with Angelaki’ s analysis system ECG based left ventricular hypertrophy detection system for detecting left ventricle hypertrophy (LVH) for an individual to provide and optimized diagnosis system that can provide a diagnosis and/or detect a wider range of abnormalities.
Regarding claim 5, Gopalakrishnan additionally discloses identifying, based on the monitoring of the further user data of the first user and the further user data of each of the plurality of second users, one or more of the plurality of courses of treatment that are successful in reducing abnormality, each of the one or more courses of treatment successfully reducing abnormality in a subset of the plurality of second users (Gopalakrishnan ¶0020 “[0020] The system may further comprise a remote server receiving the biometric data from the local computing device. One or more of the local computing devices or the remote server may comprise a machine learning algorithm which generates one or more of the cardiac health score or the one or more recommendations or goals for the user. The remote server may be configured for access by a medical professional. Alternatively, or in combination, one or more of the cardiac health score or one or more recommendations or goals may be generated by the medical professional and provided to the local computing device through the remote server”, ¶0104-¶0106).
Gopalakrishnan in view of Kwong does not disclose detecting left ventricle hypertrophy (LVH) for an individual. Angelaki in a similar field of ECG based left ventricular hypertrophy detection teaches detecting left ventricle hypertrophy (LVH) for an individual (Angelaki Introduction “This study was designed to test the hypothesis that a 12-lead ECG, a routine and inexpensive screening procedure, can provide further accuracy in detecting abnormal LVG, using ML methods, even at the early stages before the onset of LVH, in a population without established CVD. We also seek to understand which features contribute to the ML model's decisions, by calculating global feature importance and feature interactions”).
Before the effective filing date, it would have been obvious to a person of skill in the art to modify Gopalakrishnan in view of Kwong with Angelaki by combining the cardiac analysis and abnormality detection system with Angelaki’ s analysis system ECG based left ventricular hypertrophy detection system for detecting left ventricle hypertrophy (LVH) for an individual to provide and optimized diagnosis system that can provide a diagnosis and/or detect a wider range of abnormalities.
Regarding claim 6, Gopalakrishnan additionally discloses in response to determining that an abnormality is present in a third user based on user data of the third user, recommending, by the ML model, a course of treatment of the one or more courses of treatment based on the user data of the third user (Gopalakrishnan ¶0017 “The machine learning algorithm may generate the cardiac health score of the user and/or the recommendations and/or goals in response to biometric data from a plurality of users.”, ¶0016, ¶0093, ¶103).
Gopalakrishnan in view of Kwong does not disclose detecting left ventricle hypertrophy (LVH) for an individual. Angelaki in a similar field of ECG based left ventricular hypertrophy detection teaches detecting left ventricle hypertrophy (LVH) for an individual (Angelaki Introduction “This study was designed to test the hypothesis that a 12-lead ECG, a routine and inexpensive screening procedure, can provide further accuracy in detecting abnormal LVG, using ML methods, even at the early stages before the onset of LVH, in a population without established CVD. We also seek to understand which features contribute to the ML model's decisions, by calculating global feature importance and feature interactions”). Before the effective filing date, it would have been obvious to a person of skill in the art to modify Gopalakrishnan in view of Kwong with Angelaki by combining the cardiac analysis and abnormality detection system with Angelaki’ s analysis system ECG based left ventricular hypertrophy detection system for detecting left ventricle hypertrophy (LVH) for an individual to provide and optimized diagnosis system that can provide a diagnosis and/or detect a wider range of abnormalities.
Regarding claim 7, Gopalakrishnan additionally discloses wherein the ML model comprises a neural network model. (Gopalakrishnan ¶0053 “Any number of machine learning algorithms or methods may be trained to identify atrial fibrillation or other conditions such as arrhythmias. These may include the use of decision tree learning such as with a random forest, association rule learning, artificial neural network, inductive logic programming, support vector machines, clustering, Bayesian networks, reinforcement learning, representation learning, similarity and metric learning, sparse dictionary learning, or the like.”).
Regarding claim 8, Gopalakrishnan discloses a system comprising: an electrocardiogram (ECG) monitor to record the first ECG data of a first user; a blood pressure monitor to record the first blood pressure data of the first user (Gopalakrishnan ¶0021 “The sensor may comprise one or more of a hand-held electrocardiogram (ECG) sensor, a wrist-worn activity sensor, a blood pressure monitor, a personal weighing scale, a body fat percentage sensor, a personal thermometer, a pulse oximeter sensor, or any mobile health monitor or sensor.”); and a cloud storage system (Gopalakrishnan ¶0077 “The ECG data and the associated metadata and other related data as described herein can be stored in a central database, a cloud database, or a combination of the two.”) to: store a training data set comprising ECG data and blood pressure data for each a number of individuals (Gopalakrishnan ¶0016 “The population data may be collected from a plurality of the heart rate monitoring enabled portable computing devices or accessories provided to a plurality of users”; ¶0074 “As described herein, various parameters may be included in the database along with the ECG data. These may include the patient's age, gender, weight, blood pressure, medications, behaviors, habits, activities, food consumption, drink consumption, drugs, medical history and other factors that may influence a patient's ECG signal. The additional parameters may or may not be used in the comparison of the changes in ECG signal over time and circumstances.”); train a machine learning (ML) model by: analyzing, using the ML model, the ECG data and the blood pressure data for each of the number of individuals to generate an output for each of the number of individuals (Gopalakrishnan ¶0016 “he data may be used to train the machine learning algorithm to extract one or more features from any continuously measured heart rate data and identify atrial fibrillation or other conditions therefrom. After the machine learning algorithm has been trained, the machine learning algorithm may recognize atrial fibrillation from the continuously measured heart rate data of a new user who has not yet been identified as having atrial fibrillation or other heart conditions. One or more of training population data or the trained machine learning algorithm may be provided on a central computing device (e.g. be stored on a non-transitory computer readable medium of a server) which is in communication with the local computing devices of the users and the application executed thereon (e.g. through an Internet or an intranet connection.)”); receive user data of a first user, the user data comprising the first (ECG) data and the first blood pressure data of the first user (Gopalakrishnan ¶0077 “The ECG data and the associated metadata and other related data as described herein can be stored in a central database, a cloud database, or a combination of the two.”); analyze, by a (ML) model, the user data to determine whether an abnormality is present in the first user based on the first ECG data and the first blood pressure data (Gopalakrishnan ¶0103 “In a step 912, a cardiac health score is generated. The cardiac health score can be generated by considering and weighing one or more influencing factors including the incidence of atrial fibrillation or arrhythmia as detected by the handheld ECG monitor, the heart rate of the user or subject, the activity of the user or subject, hours of sleep and rest of the user or subject, blood pressure of the user or subject, etc. Often, the incidence of atrial fibrillation or arrhythmia will be weighed the most. The cardiac health score may be generated by a physician or a machine learning algorithm provided by the remote server or cloud-based service 113, for example”, ¶0102, ¶0010 “These parameters and further parameters may be analyzed to detect and/or predict one or more of atrial fibrillation, tachycardia, bradycardia, bigeminy, trigeminy, or other cardiac conditions. A quantitative heart health score may also be generated from the determined parameters. One or more of the heart health score, detected heart conditions, or recommended user action items based on the heart health score may be displayed to the user through a display of the portable computing device.” Figure 6); determine, based on analyzing the user data, a group to which the first user belongs (Gopalakrishnan ¶0022 “The user may also input user-related health data such as age, height, weight, body mass index (BMI), diet, sleep levels, rest levels, or stress levels. One or more of these physiological signals and/or parameters may be combined with the heart rate data to detect atrial fibrillation or other conditions. The machine learning algorithm may be configured to identify atrial fibrillation or other conditions in response to heart rate data in combination with one or more of the other physiological signals and/or parameters for instance”; ¶0083 “Depending on the severity of the arrhythmia detected, the heart score may be calculated or assigned within the ranges according to the table below in Table 2.”; ¶0078 “In step 408, the subset of ECG data can be analyzed using a machine learning algorithm, which can assign a risk level to the ECG data in step 410.”); and in response to determining that the abnormality is present in the first user, output an indication that the abnormality is present in the first user and a severity of the abnormality in the first user (Gopalakrishnan ¶0012 “The dashboard includes a heart score that can be calculated in response to data from the user such as their ECG and other personal information such as age, gender, height, weight, body fat, disease risks, etc.”, Table 2 additionally shows the heart score with a range of defined values to show severity of detected signals as it relates to the values.); and in response to the indication, prescribing, by the ML model and based on the group, a course of treatment of the LVH individually tailored to the first user (Gopalakrishnan ¶0015 “Another aspect of the present disclosure provides a method for managing cardiac health. Biometric data of a user may be received. A cardiac health score may be generated in response to the received biometric data. One or more recommendations or goals for improving the generated cardiac health score may be displayed to the user. The biometric data may comprise one or more of an electrocardiogram (ECG), dietary information, stress level, activity level, gender, height, weight, age, body fat percentage, blood pressure, results from imaging scans, blood chemistry values, or genotype data.”; ¶0022 “The user may also input user-related health data such as age, height, weight, body mass index (BMI), diet, sleep levels, rest levels, or stress levels. One or more of these physiological signals and/or parameters may be combined with the heart rate data to detect atrial fibrillation or other conditions. The machine learning algorithm may be configured to identify atrial fibrillation or other conditions in response to heart rate data in combination with one or more of the other physiological signals and/or parameters for instance. Triggers or alerts may be provided to the user in response to the measured physiological signals and/or parameters. Such triggers or alerts may notify the user to take corrective steps to improve their health or monitor other vital signs or physiological parameters”)the course of treatment comprising a modification to diet or medication prescribed in another course of treatment for another user (Gopalakrishnan ¶0076 “The patient may avoid a future healthcare issue, as instructed or recommended by the system, by modifying their behavior, habits or by taking any course of action, including but not limited to taking a medication, drug or adhering to a diet or exercise program, which may be a predetermined course of action recommended by the system independent of any analysis of the ECG data, and/or may also result from insights learned through this system and method as described herein. “)wherein, to analyze the user data based on the first ECG data and the first blood pressure data, the cloud storage system is further configured to: analyze the first ECG data to obtain an ECG analysis result (Gopalakrishnan ¶0052 “The machine learning based algorithm(s) may allow software application(s) to identify patterns and/or features of the R-R interval data and/or the raw heart rate signals or data to predict and/or detect atrial fibrillation or other arrhythmias.”); wherein system is further configured to: augment the ECG analysis result with the first blood pressure data to obtain a combination of the first ECG data and the first blood pressure data (Gopalakrishnan ¶0103 “In a step 912, a cardiac health score is generated. The cardiac health score can be generated by considering and weighing one or more influencing factors including the incidence of atrial fibrillation or arrhythmia as detected by the handheld ECG monitor, the heart rate of the user or subject, the activity of the user or subject, hours of sleep and rest of the user or subject, blood pressure of the user or subject, etc. Often, the incidence of atrial fibrillation or arrhythmia will be weighed the most. The cardiac health score may be generated by a physician or a machine learning algorithm provided by the remote server or cloud-based service 113, for example”; Showing the weighted predictions which can involve both ECG data and blood pressure data); and determine whether the abnormality is present in the first user based on the combination of the first ECG data and the first blood pressure data (Gopalakrishnan ¶0103, ¶0102, ¶0010; Figure 6).
Gopalakrishnan does not discloses using backpropagation to update one or more weights of the ML model based on a difference between the output for each of the number of individuals and the corresponding diagnosis label. Kwong in a similar field of cardiology teaches using backpropagation to update one or more weights of the ML model based on a difference between the output for each of the number of individuals and the corresponding LVH diagnosis label (Kwong Column 9 Lines 33-49 “Method step 78 shows a decision point. In this decision point it is determined whether the one or more wave amplitude ratios created in method step 74 used in conjunction with the generated decision matrices generated in method step 76 yields accurate enough results to be useful in a clinical sense. If the answer is yes, then one proceeds to method step 80 which is the end of the process. Otherwise, one proceeds back to method step 74 wherein another one or more wave amplitude ratios is created and again used to generate another statistical decision matrix in method step 76. The loop comprised of method steps 74, 76 and 78 continues until the process yields at least one wave amplitude ratio-decision matrix combination which will generate diagnosis of sufficient accuracy to be useful in a clinical sense, as is depicted in method step 78. Once this criterion has been satisfied one proceeds to method step 80 and the process stops.” Which describes the methods of performing backpropagation; Figure 4).
Before the effective filing date, it would have been obvious to a person of skill in the art to modify Gopalakrishnan in view Kwong by combining the cardiac analysis system of Gopalakrishnan with Kwong’s analysis system which uses backpropagation to update one or more weights of the ML model based on a difference between the output for each of the number of individuals and the corresponding diagnosis label. It would have been obvious to one having ordinary skill in the art at the time the invention was made to use backpropagation to update one or more weights of the system’s ML model based on a difference between the output for each of the number of individuals and the corresponding diagnosis label for the purpose of optimized cardiac health monitoring. The addition of backpropagation techniques as taught in Kwong serves a method for reducing diagnosis and monitoring errors that may come from noisy data sets and the initial training of a machine learning model. This will provide and optimized diagnosis system that can self-optimize as the system takes cardiac data and uses it as new training sets over time.
Gopalakrishnan in view of Kwong does not disclose providing a corresponding training dataset of left ventricle hypertrophy (LVH) diagnosis labels for each a number of individuals. Angelaki in a similar field of ECG based left ventricular hypertrophy detection teaches providing a corresponding left ventricle hypertrophy (LVH) diagnosis label for each a number of individuals (Angelaki Introduction “This study was designed to test the hypothesis that a 12-lead ECG, a routine and inexpensive screening procedure, can provide further accuracy in detecting abnormal LVG, using ML methods, even at the early stages before the onset of LVH, in a population without established CVD. We also seek to understand which features contribute to the ML model's decisions, by calculating global feature importance and feature interactions”). Before the effective filing date, it would have been obvious to a person of skill in the art to modify Gopalakrishnan in view of Kwong with Angelaki by combining the cardiac analysis and abnormality detection system with Angelaki’ s analysis system ECG based left ventricular hypertrophy detection system for providing a corresponding left ventricle hypertrophy (LVH) diagnosis label for each a number of individuals to provide and optimized diagnosis system that can provide a diagnosis and/or detect a wider range of abnormalities.
Regarding claim 9, Gopalakrishnan additionally discloses wherein the user data of the first user further comprises characteristics of the first user and the ML model determines whether the abnormality is present in the first user based further on the characteristics of the first user. (Gopalakrishnan ¶0015 “The biometric data may comprise one or more of an electrocardiogram (ECG), dietary information, stress level, activity level, gender, height, weight, age, body fat percentage, blood pressure, results from imaging scans, blood chemistry values, or genotype data.”, ¶0074 “As described herein, various parameters may be included in the database along with the ECG data. These may include the patient's age, gender, weight, blood pressure, medications, behaviors, habits, activities, food consumption, drink consumption, drugs, medical history and other factors that may influence a patient's ECG signal. The additional parameters may or may not be used in the comparison of the changes in ECG signal over time and circumstances”, ¶0012).
Gopalakrishnan in view of Kwong does not disclose detecting left ventricle hypertrophy (LVH) for an individual. Angelaki in a similar field of ECG based left ventricular hypertrophy detection teaches detecting left ventricle hypertrophy (LVH) for an individual (Angelaki Introduction “This study was designed to test the hypothesis that a 12-lead ECG, a routine and inexpensive screening procedure, can provide further accuracy in detecting abnormal LVG, using ML methods, even at the early stages before the onset of LVH, in a population without established CVD. We also seek to understand which features contribute to the ML model's decisions, by calculating global feature importance and feature interactions”). Before the effective filing date, it would have been obvious to a person of skill in the art to modify Gopalakrishnan in view of Kwong with Angelaki by combining the cardiac analysis and abnormality detection system with Angelaki’ s analysis system ECG based left ventricular hypertrophy detection system for detecting left ventricle hypertrophy (LVH) for an individual to provide and optimized diagnosis system that can provide a diagnosis and/or detect a wider range of abnormalities.
Regarding claim 10, Gopalakrishnan additionally discloses wherein to determine whether the LVH is present in the first user, the cloud storage system is to: compare the first ECG data to ECG criteria comprising a set of ranges, wherein the set of ranges indicate whether the abnormality is present and a severity of the abnormality if it is present (Gopalakrishnan ¶0070 “For example, the P wave and heart rate can be extracted and analyzed to identify atrial fibrillation, where the absence of P waves and/or an irregular heart rate may indicate atrial fibrillation.”, ¶0071); compare the first blood pressure data to blood pressure criteria defining a set of systolic and diastolic pressure ranges, each of the set of systolic and diastolic pressure ranges indicating a likelihood that the LVH is present (Gopalakrishnan Page 6 Table 1 shows thresholds that are of concern including blood pressure.); and determine whether the abnormality is present in the first user and a severity if it is, based on a range that the first ECG data is within and a systolic and diastolic pressure range that the first blood pressure data is within (Gopalakrishnan ¶0074 “As described herein, various parameters may be included in the database along with the ECG data. These may include the patient's age, gender, weight, blood pressure, medications, behaviors, habits, activities, food consumption, drink consumption, drugs, medical history and other factors that may influence a patient's ECG signal. The additional parameters may or may not be used in the comparison of the changes in ECG signal over time and circumstances”).
Gopalakrishnan in view of Kwong does not disclose detecting left ventricle hypertrophy (LVH) for an individual. Angelaki in a similar field of ECG based left ventricular hypertrophy detection teaches detecting left ventricle hypertrophy (LVH) for an individual (Angelaki Introduction “This study was designed to test the hypothesis that a 12-lead ECG, a routine and inexpensive screening procedure, can provide further accuracy in detecting abnormal LVG, using ML methods, even at the early stages before the onset of LVH, in a population without established CVD. We also seek to understand which features contribute to the ML model's decisions, by calculating global feature importance and feature interactions”).
Before the effective filing date, it would have been obvious to a person of skill in the art to modify Gopalakrishnan in view of Kwong with Angelaki by combining the cardiac analysis and abnormality detection system with Angelaki’ s analysis system ECG based left ventricular hypertrophy detection system for detecting left ventricle hypertrophy (LVH) for an individual to provide and optimized diagnosis system that can provide a diagnosis and/or detect a wider range of abnormalities.
Regarding claim 11, Gopalakrishnan additionally discloses wherein the cloud storage system is further to: monitor, by the ML model, further user data of the first user, wherein the further user data captured while the first user is undergoing the course of treatment from among a plurality of courses of treatment for an abnormality; and monitor, by the ML model, further user data of each of a plurality of second users, further user data of each second user is monitored while the second user is undergoing one of the plurality of courses of treatment for an abnormality (Gopalakrishnan ¶0073 “Analysis of the time following the abnormality, adverse event or disease state can provide information regarding the efficacy of treatments and/or provide the patient or physician information regarding disease progression, such as whether the patient's condition in improving, worsening or staying the same. The diagnosis and determination can also be used for indexing by, for example, including it in the metadata associated with the corresponding ECG data.”, ¶103 “A plurality of users and subject may concurrently use the cardiac health and/or rhythm management system 100 and the machine learning algorithm may, for example, consider population data and trends to generate an individual user or subject's cardiac health score.”, ¶0104-¶0105).
Gopalakrishnan in view of Kwong does not disclose detecting left ventricle hypertrophy (LVH) for an individual. Angelaki in a similar field of ECG based left ventricular hypertrophy detection teaches detecting left ventricle hypertrophy (LVH) for an individual (Angelaki Introduction “This study was designed to test the hypothesis that a 12-lead ECG, a routine and inexpensive screening procedure, can provide further accuracy in detecting abnormal LVG, using ML methods, even at the early stages before the onset of LVH, in a population without established CVD. We also seek to understand which features contribute to the ML model's decisions, by calculating global feature importance and feature interactions”).
Before the effective filing date, it would have been obvious to a person of skill in the art to modify Gopalakrishnan in view of Kwong with Angelaki by combining the cardiac analysis and abnormality detection system with Angelaki’ s analysis system ECG based left ventricular hypertrophy detection system for detecting left ventricle hypertrophy (LVH) for an individual to provide and optimized diagnosis system that can provide a diagnosis and/or detect a wider range of abnormalities.
Regarding claim 12, Gopalakrishnan additionally discloses wherein the cloud storage system is further to: identify, based on the monitoring of the further user data of each of the first user and the further user data of the plurality of second users, one or more of the plurality of courses of treatment that are successful in reducing abnormality, each of the one or more courses of treatment successfully reducing abnormality in a subset of the plurality of second users. (Gopalakrishnan ¶0020 “[0020] The system may further comprise a remote server receiving the biometric data from the local computing device. One or more of the local computing devices or the remote server may comprise a machine learning algorithm which generates one or more of the cardiac health score or the one or more recommendations or goals for the user. The remote server may be configured for access by a medical professional. Alternatively, or in combination, one or more of the cardiac health score or one or more recommendations or goals may be generated by the medical professional and provided to the local computing device through the remote server”, ¶0104-¶0106).
Gopalakrishnan in view of Kwong does not disclose detecting left ventricle hypertrophy (LVH) for an individual. Angelaki in a similar field of ECG based left ventricular hypertrophy detection teaches detecting left ventricle hypertrophy (LVH) for an individual (Angelaki Introduction “This study was designed to test the hypothesis that a 12-lead ECG, a routine and inexpensive screening procedure, can provide further accuracy in detecting abnormal LVG, using ML methods, even at the early stages before the onset of LVH, in a population without established CVD. We also seek to understand which features contribute to the ML model's decisions, by calculating global feature importance and feature interactions”). Before the effective filing date, it would have been obvious to a person of skill in the art to modify Gopalakrishnan in view of Kwong with Angelaki by combining the cardiac analysis and abnormality detection system with Angelaki’ s analysis system ECG based left ventricular hypertrophy detection system for detecting left ventricle hypertrophy (LVH) for an individual to provide and optimized diagnosis system that can provide a diagnosis and/or detect a wider range of abnormalities.
Regarding claim 13, Gopalakrishnan additionally discloses wherein the cloud storage system is further to: in response to determining that abnormality is present in a third user based on user data of the third user, recommending, by the ML model, a course of treatment of the one or more courses of treatment based on the user data of the third user. (Gopalakrishnan ¶0017 “The machine learning algorithm may generate the cardiac health score of the user and/or the recommendations and/or goals in response to biometric data from a plurality of users.”, ¶0016, ¶0093, ¶103).
Gopalakrishnan in view of Kwong does not disclose detecting left ventricle hypertrophy (LVH) for an individual. Angelaki in a similar field of ECG based left ventricular hypertrophy detection teaches detecting left ventricle hypertrophy (LVH) for an individual (Angelaki Introduction “This study was designed to test the hypothesis that a 12-lead ECG, a routine and inexpensive screening procedure, can provide further accuracy in detecting abnormal LVG, using ML methods, even at the early stages before the onset of LVH, in a population without established CVD. We also seek to understand which features contribute to the ML model's decisions, by calculating global feature importance and feature interactions”). Before the effective filing date, it would have been obvious to a person of skill in the art to modify Gopalakrishnan in view of Kwong with Angelaki by combining the cardiac analysis and abnormality detection system with Angelaki’ s analysis system ECG based left ventricular hypertrophy detection system for detecting left ventricle hypertrophy (LVH) for an individual to provide and optimized diagnosis system that can provide a diagnosis and/or detect a wider range of abnormalities.
Regarding claim 14, Gopalakrishnan additionally discloses wherein the ML model comprises a neural network model. (Gopalakrishnan ¶0053 “Any number of machine learning algorithms or methods may be trained to identify atrial fibrillation or other conditions such as arrhythmias. These may include the use of decision tree learning such as with a random forest, association rule learning, artificial neural network, inductive logic programming, support vector machines, clustering, Bayesian networks, reinforcement learning, representation learning, similarity and metric learning, sparse dictionary learning, or the like.”).
Regarding claim 15, Gopalakrishnan discloses a non-transitory computer-readable medium having instructions stored thereon which, when executed by a processing device, cause the processing device to: provide a training data set comprising electrocardiogram (ECG) data and blood pressure data for each a number of individuals (Gopalakrishnan ¶0016 “The population data may be collected from a plurality of the heart rate monitoring enabled portable computing devices or accessories provided to a plurality of users”; ¶0074 “As described herein, various parameters may be included in the database along with the ECG data. These may include the patient's age, gender, weight, blood pressure, medications, behaviors, habits, activities, food consumption, drink consumption, drugs, medical history and other factors that may influence a patient's ECG signal. The additional parameters may or may not be used in the comparison of the changes in ECG signal over time and circumstances.”); train a machine learning (ML) model by: analyzing, using a machine learning (ML) model, the ECG data and the blood pressure data for each of the number of individuals to generate an output for each of the number of individuals (Gopalakrishnan ¶0016 “he data may be used to train the machine learning algorithm to extract one or more features from any continuously measured heart rate data and identify atrial fibrillation or other conditions therefrom. After the machine learning algorithm has been trained, the machine learning algorithm may recognize atrial fibrillation from the continuously measured heart rate data of a new user who has not yet been identified as having atrial fibrillation or other heart conditions. One or more of training population data or the trained machine learning algorithm may be provided on a central computing device (e.g. be stored on a non-transitory computer readable medium of a server) which is in communication with the local computing devices of the users and the application executed thereon (e.g. through an Internet or an intranet connection.)”); receive user data of a first user (Gopalakrishnan ¶0010 “The portable computing device may have loaded onto (e.g. onto a non-transitory computer readable medium of the computing device) and executing thereon (e.g. by a processor of the computing device) an application for one or more of receiving the continuously measured physiological signal(s), analyzing the physiological signal(s), sending the physiological signal(s) to a remote computer for further analysis and storage, and displaying to the user analysis of the physiological signal(s).”) , the user data comprising (ECG) data and blood pressure data of the first user (Gopalakrishnan ¶0015 “The biometric data may comprise one or more of an electrocardiogram (ECG), dietary information, stress level, activity level, gender, height, weight, age, body fat percentage, blood pressure, results from imaging scans, blood chemistry values, or genotype data.”, ¶0102 ); analyze, by the processing device executing a machine learning (ML) model, the user data to determine whether abnormality is present in the first user based on the first ECG data and the first blood pressure data (Gopalakrishnan ¶0103 “In a step 912, a cardiac health score is generated. The cardiac health score can be generated by considering and weighing one or more influencing factors including the incidence of atrial fibrillation or arrhythmia as detected by the handheld ECG monitor, the heart rate of the user or subject, the activity of the user or subject, hours of sleep and rest of the user or subject, blood pressure of the user or subject, etc. Often, the incidence of atrial fibrillation or arrhythmia will be weighed the most. The cardiac health score may be generated by a physician or a machine learning algorithm provided by the remote server or cloud-based service 113, for example”, ¶0102, ¶0010 “These parameters and further parameters may be analyzed to detect and/or predict one or more of atrial fibrillation, tachycardia, bradycardia, bigeminy, trigeminy, or other cardiac conditions. A quantitative heart health score may also be generated from the determined parameters. One or more of the heart health score, detected heart conditions, or recommended user action items based on the heart health score may be displayed to the user through a display of the portable computing device.” Figure 6); determine, based on analyzing the user data, a group to which the first user belongs (Gopalakrishnan ¶0022 “The user may also input user-related health data such as age, height, weight, body mass index (BMI), diet, sleep levels, rest levels, or stress levels. One or more of these physiological signals and/or parameters may be combined with the heart rate data to detect atrial fibrillation or other conditions. The machine learning algorithm may be configured to identify atrial fibrillation or other conditions in response to heart rate data in combination with one or more of the other physiological signals and/or parameters for instance”; ¶0083 “Depending on the severity of the arrhythmia detected, the heart score may be calculated or assigned within the ranges according to the table below in Table 2.”; ¶0078 “In step 408, the subset of ECG data can be analyzed using a machine learning algorithm, which can assign a risk level to the ECG data in step 410.”); and in response to determining that the abnormality is present in the first user, output an indication that the abnormality is present in the first user and a severity of the abnormality in the first user (Gopalakrishnan ¶0012 “The dashboard includes a heart score that can be calculated in response to data from the user such as their ECG and other personal information such as age, gender, height, weight, body fat, disease risks, etc.”, Table 2 additionally shows the heart score with a range of defined values to show severity of detected signals as it relates to the values.); and in response to the indication, prescribing, by the ML model and based on the group, a course of treatment of the LVH individually tailored to the first user (Gopalakrishnan ¶0015 “Another aspect of the present disclosure provides a method for managing cardiac health. Biometric data of a user may be received. A cardiac health score may be generated in response to the received biometric data. One or more recommendations or goals for improving the generated cardiac health score may be displayed to the user. The biometric data may comprise one or more of an electrocardiogram (ECG), dietary information, stress level, activity level, gender, height, weight, age, body fat percentage, blood pressure, results from imaging scans, blood chemistry values, or genotype data.”; ¶0022 “The user may also input user-related health data such as age, height, weight, body mass index (BMI), diet, sleep levels, rest levels, or stress levels. One or more of these physiological signals and/or parameters may be combined with the heart rate data to detect atrial fibrillation or other conditions. The machine learning algorithm may be configured to identify atrial fibrillation or other conditions in response to heart rate data in combination with one or more of the other physiological signals and/or parameters for instance. Triggers or alerts may be provided to the user in response to the measured physiological signals and/or parameters. Such triggers or alerts may notify the user to take corrective steps to improve their health or monitor other vital signs or physiological parameters”)the course of treatment comprising a modification to diet or medication prescribed in another course of treatment for another user (Gopalakrishnan ¶0076 “The patient may avoid a future healthcare issue, as instructed or recommended by the system, by modifying their behavior, habits or by taking any course of action, including but not limited to taking a medication, drug or adhering to a diet or exercise program, which may be a predetermined course of action recommended by the system independent of any analysis of the ECG data, and/or may also result from insights learned through this system and method as described herein. “) wherein, to analyze the user data based on the first ECG data and the first blood pressure data, the processing device is further to: analyze the first ECG data to obtain an ECG analysis result (Gopalakrishnan ¶0052 “The machine learning based algorithm(s) may allow software application(s) to identify patterns and/or features of the R-R interval data and/or the raw heart rate signals or data to predict and/or detect atrial fibrillation or other arrhythmias.”); augment the ECG analysis result with the first blood pressure data to obtain a combination of the first ECG data and the first blood pressure data (Gopalakrishnan ¶0103 “In a step 912, a cardiac health score is generated. The cardiac health score can be generated by considering and weighing one or more influencing factors including the incidence of atrial fibrillation or arrhythmia as detected by the handheld ECG monitor, the heart rate of the user or subject, the activity of the user or subject, hours of sleep and rest of the user or subject, blood pressure of the user or subject, etc. Often, the incidence of atrial fibrillation or arrhythmia will be weighed the most. The cardiac health score may be generated by a physician or a machine learning algorithm provided by the remote server or cloud-based service 113, for example”; Showing the weighted predictions which can involve both ECG data and blood pressure data); and determine whether abnormality is present in the first user based on the combination of the first ECG data and the first blood pressure data (Gopalakrishnan ¶0103, ¶0102, ¶0010; Figure 6).
Gopalakrishnan does not discloses using backpropagation to update one or more weights of the ML model based on a difference between the output for each of the number of individuals and the corresponding diagnosis labels. Kwong in a similar field of cardiology teaches using backpropagation to update one or more weights of the ML model based on a difference between the output for each of the number of individuals and the corresponding diagnosis labels (Kwong Column 9 Lines 33-49 “Method step 78 shows a decision point. In this decision point it is determined whether the one or more wave amplitude ratios created in method step 74 used in conjunction with the generated decision matrices generated in method step 76 yields accurate enough results to be useful in a clinical sense. If the answer is yes, then one proceeds to method step 80 which is the end of the process. Otherwise, one proceeds back to method step 74 wherein another one or more wave amplitude ratios is created and again used to generate another statistical decision matrix in method step 76. The loop comprised of method steps 74, 76 and 78 continues until the process yields at least one wave amplitude ratio-decision matrix combination which will generate diagnosis of sufficient accuracy to be useful in a clinical sense, as is depicted in method step 78. Once this criterion has been satisfied one proceeds to method step 80 and the process stops.” Which describes the methods of performing backpropagation; Figure 4). Before the effective filing date, it would have been obvious to a person of skill in the art to modify Gopalakrishnan in view Kwong by combining the cardiac analysis system of Gopalakrishnan with Kwong’s analysis system using backpropagation to update one or more weights of the ML model based on a difference between the output for each of the number of individuals and the corresponding diagnosis label. It would have been obvious to one having ordinary skill in the art at the time the invention was made to use backpropagation to update one or more weights of the system’s ML model based on a difference between the output for each of the number of individuals and the corresponding diagnosis label for the purpose of optimized cardiac health monitoring. The addition of backpropagation techniques as taught in Kwong serves a method for reducing diagnosis and monitoring errors that may come from noisy data sets and the initial training of a machine learning model. This will provide and optimized diagnosis system that can self-optimize as the system takes cardiac data and uses it as new training sets over time.
Gopalakrishnan in view of Kwong does not disclose providing a corresponding left ventricle hypertrophy (LVH) diagnosis label for each a number of individuals. Angelaki in a similar field of ECG based left ventricular hypertrophy detection teaches providing a corresponding left ventricle hypertrophy (LVH) diagnosis label for each a number of individuals (Angelaki Introduction “This study was designed to test the hypothesis that a 12-lead ECG, a routine and inexpensive screening procedure, can provide further accuracy in detecting abnormal LVG, using ML methods, even at the early stages before the onset of LVH, in a population without established CVD. We also seek to understand which features contribute to the ML model's decisions, by calculating global feature importance and feature interactions”). Before the effective filing date, it would have been obvious to a person of skill in the art to modify Gopalakrishnan in view of Kwong with Angelaki by combining the cardiac analysis and abnormality detection system with Angelaki’ s analysis system ECG based left ventricular hypertrophy detection system for providing a corresponding left ventricle hypertrophy (LVH) diagnosis label for each a number of individuals to provide and optimized diagnosis system that can provide a diagnosis and/or detect a wider range of abnormalities.
Regarding claim 16, Gopalakrishnan additionally discloses wherein the user data of the first user further comprises characteristics of the first user and the ML model determines whether the abnormality is present in the first user based further on the characteristics of the first user. (Gopalakrishnan ¶0015 “The biometric data may comprise one or more of an electrocardiogram (ECG), dietary information, stress level, activity level, gender, height, weight, age, body fat percentage, blood pressure, results from imaging scans, blood chemistry values, or genotype data.”, ¶0074 “As described herein, various parameters may be included in the database along with the ECG data. These may include the patient's age, gender, weight, blood pressure, medications, behaviors, habits, activities, food consumption, drink consumption, drugs, medical history and other factors that may influence a patient's ECG signal. The additional parameters may or may not be used in the comparison of the changes in ECG signal over time and circumstances”, ¶0012).
Gopalakrishnan in view of Kwong does not disclose detecting left ventricle hypertrophy (LVH) for an individual. Angelaki in a similar field of ECG based left ventricular hypertrophy detection teaches detecting left ventricle hypertrophy (LVH) for an individual (Angelaki Introduction “This study was designed to test the hypothesis that a 12-lead ECG, a routine and inexpensive screening procedure, can provide further accuracy in detecting abnormal LVG, using ML methods, even at the early stages before the onset of LVH, in a population without established CVD. We also seek to understand which features contribute to the ML model's decisions, by calculating global feature importance and feature interactions”). Before the effective filing date, it would have been obvious to a person of skill in the art to modify Gopalakrishnan in view of Kwong with Angelaki by combining the cardiac analysis and abnormality detection system with Angelaki’ s analysis system ECG based left ventricular hypertrophy detection system for detecting left ventricle hypertrophy (LVH) for an individual to provide and optimized diagnosis system that can provide a diagnosis and/or detect a wider range of abnormalities.
Regarding claim 17, Gopalakrishnan additionally discloses wherein to determine whether the abnormality is present in the first user, the processing device is to: compare the first ECG data to ECG criteria comprising a set of ranges, wherein the set of ranges indicate whether the abnormality is present and a severity of the abnormality if it is present (Gopalakrishnan ¶0070 “For example, the P wave and heart rate can be extracted and analyzed to identify atrial fibrillation, where the absence of P waves and/or an irregular heart rate may indicate atrial fibrillation.”, ¶0071); compare the first blood pressure data to blood pressure criteria defining a set of systolic and diastolic pressure ranges, each of the set of systolic and diastolic pressure ranges indicating a likelihood that the abnormality is present (Gopalakrishnan Page 6 Table 1 shows thresholds that are of concern including blood pressure.); and determine whether the abnormality is present in the first user and a severity if it is, based on a range that the first ECG data is within and a systolic and diastolic pressure range that the first blood pressure data is within (Gopalakrishnan ¶0074 “As described herein, various parameters may be included in the database along with the ECG data. These may include the patient's age, gender, weight, blood pressure, medications, behaviors, habits, activities, food consumption, drink consumption, drugs, medical history and other factors that may influence a patient's ECG signal. The additional parameters may or may not be used in the comparison of the changes in ECG signal over time and circumstances”).
Gopalakrishnan in view of Kwong does not disclose detecting left ventricle hypertrophy (LVH) for an individual. Angelaki in a similar field of ECG based left ventricular hypertrophy detection teaches detecting left ventricle hypertrophy (LVH) for an individual (Angelaki Introduction “This study was designed to test the hypothesis that a 12-lead ECG, a routine and inexpensive screening procedure, can provide further accuracy in detecting abnormal LVG, using ML methods, even at the early stages before the onset of LVH, in a population without established CVD. We also seek to understand which features contribute to the ML model's decisions, by calculating global feature importance and feature interactions”). Before the effective filing date, it would have been obvious to a person of skill in the art to modify Gopalakrishnan in view of Kwong with Angelaki by combining the cardiac analysis and abnormality detection system with Angelaki’ s analysis system ECG based left ventricular hypertrophy detection system for detecting left ventricle hypertrophy (LVH) for an individual to provide and optimized diagnosis system that can provide a diagnosis and/or detect a wider range of abnormalities.
Regarding claim 18, Gopalakrishnan additionally discloses wherein the processing device is further to: monitor, by the ML model, further user data of the first user, the further user data captured while the first user is undergoing the course of treatment from among a plurality of courses of treatment for abnormality; and monitor, by the ML model, wherein the further user data of each of a plurality of second users, further user data of each second user is monitored while the second user is undergoing one of the plurality of courses of treatment for abnormality (Gopalakrishnan ¶0073 “Analysis of the time following the abnormality, adverse event or disease state can provide information regarding the efficacy of treatments and/or provide the patient or physician information regarding disease progression, such as whether the patient's condition in improving, worsening or staying the same. The diagnosis and determination can also be used for indexing by, for example, including it in the metadata associated with the corresponding ECG data.”, ¶103 “A plurality of users and subject may concurrently use the cardiac health and/or rhythm management system 100 and the machine learning algorithm may, for example, consider population data and trends to generate an individual user or subject's cardiac health score.”, ¶0104-¶0105).
Gopalakrishnan in view of Kwong does not disclose detecting left ventricle hypertrophy (LVH) for an individual. Angelaki in a similar field of ECG based left ventricular hypertrophy detection teaches detecting left ventricle hypertrophy (LVH) for an individual (Angelaki Introduction “This study was designed to test the hypothesis that a 12-lead ECG, a routine and inexpensive screening procedure, can provide further accuracy in detecting abnormal LVG, using ML methods, even at the early stages before the onset of LVH, in a population without established CVD. We also seek to understand which features contribute to the ML model's decisions, by calculating global feature importance and feature interactions”). Before the effective filing date, it would have been obvious to a person of skill in the art to modify Gopalakrishnan in view of Kwong with Angelaki by combining the cardiac analysis and abnormality detection system with Angelaki’ s analysis system ECG based left ventricular hypertrophy detection system for detecting left ventricle hypertrophy (LVH) for an individual to provide and optimized diagnosis system that can provide a diagnosis and/or detect a wider range of abnormalities.
Regarding claim 19, Gopalakrishnan additionally discloses wherein the processing device is further to: identify, based on the monitoring of the further user data of the first user and the further user data of each of the plurality of second users, one or more of the plurality of courses of treatment that are successful in reducing abnormality, each of the one or more courses of treatment successfully reducing abnormality in a subset of the plurality of second users. (Gopalakrishnan ¶0020 “[0020] The system may further comprise a remote server receiving the biometric data from the local computing device. One or more of the local computing devices or the remote server may comprise a machine learning algorithm which generates one or more of the cardiac health score or the one or more recommendations or goals for the user. The remote server may be configured for access by a medical professional. Alternatively, or in combination, one or more of the cardiac health score or one or more recommendations or goals may be generated by the medical professional and provided to the local computing device through the remote server”, ¶0104-¶0106).
Gopalakrishnan in view of Kwong does not disclose detecting left ventricle hypertrophy (LVH) for an individual. Angelaki in a similar field of ECG based left ventricular hypertrophy detection teaches detecting left ventricle hypertrophy (LVH) for an individual (Angelaki Introduction “This study was designed to test the hypothesis that a 12-lead ECG, a routine and inexpensive screening procedure, can provide further accuracy in detecting abnormal LVG, using ML methods, even at the early stages before the onset of LVH, in a population without established CVD. We also seek to understand which features contribute to the ML model's decisions, by calculating global feature importance and feature interactions”). Before the effective filing date, it would have been obvious to a person of skill in the art to modify Gopalakrishnan in view of Kwong with Angelaki by combining the cardiac analysis and abnormality detection system with Angelaki’ s analysis system ECG based left ventricular hypertrophy detection system for detecting left ventricle hypertrophy (LVH) for an individual to provide and optimized diagnosis system that can provide a diagnosis and/or detect a wider range of abnormalities.
Regarding claim 20, Gopalakrishnan additionally discloses wherein the processing device is further to: in response to determining that abnormality is present in a third user based on user data of the third user, recommending, by the ML model, a course of treatment of the one or more courses of treatment based on the user data of the third user. (Gopalakrishnan ¶0017 “The machine learning algorithm may generate the cardiac health score of the user and/or the recommendations and/or goals in response to biometric data from a plurality of users.”, ¶0016, ¶0093, ¶103).
Gopalakrishnan in view of Kwong does not disclose detecting left ventricle hypertrophy (LVH) for an individual. Angelaki in a similar field of ECG based left ventricular hypertrophy detection teaches detecting left ventricle hypertrophy (LVH) for an individual (Angelaki Introduction “This study was designed to test the hypothesis that a 12-lead ECG, a routine and inexpensive screening procedure, can provide further accuracy in detecting abnormal LVG, using ML methods, even at the early stages before the onset of LVH, in a population without established CVD. We also seek to understand which features contribute to the ML model's decisions, by calculating global feature importance and feature interactions”). Before the effective filing date, it would have been obvious to a person of skill in the art to modify Gopalakrishnan in view of Kwong with Angelaki by combining the cardiac analysis and abnormality detection system with Angelaki’ s analysis system ECG based left ventricular hypertrophy detection system for detecting left ventricle hypertrophy (LVH) for an individual to provide and optimized diagnosis system that can provide a diagnosis and/or detect a wider range of abnormalities.
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 extension fee 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 date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MEGAN FEDORKY whose telephone number is (571)272-2117. The examiner can normally be reached M-F 9:30-4:30.
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, Jennifer McDonald can be reached on M-F 9:30-4:30. 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.
/MEGAN T FEDORKY/
Examiner, Art Unit 3796
/Jennifer Pitrak McDonald/Supervisory Patent Examiner, Art Unit 3796
Prosecution Insights