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 .
Response to Amendment
2. This action is responsive to the amendments filed 2/10/2026. Claims 55, 67, and 79 have been amended. No claims were canceled or newly added.
Response to Arguments
Applicant’s arguments filed on 2/10/2026 with respect to the art rejections for claims 55-86 for the newly added limitation of “an adaptive threshold” have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Claim Rejections - 35 USC § 103
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.
5. Claims 67-78 are rejected under 35 U.S.C 103 as being unpatentable over Friedman et al. (International Publication No.: WO 2015/048514 A1, – Previously Cited) and further in view of Arnold et al. (US Patent No.: 5,713,367, – Previously Cited) and further in view of Mottaiyan et al. (US Pub.: 2014/0088450 A1) and further in view of Donnelly et al. (US Pub.: 2010/0298899 A1, – Previously Cited).
Regarding claim 67, Friedman teaches an analyte level prediction apparatus comprising: a memory (e.g. Fig. 15 – element 1504; paragraph 0113);
and a processor (e.g. Fig. 15 – element 1502; paragraph 0113), operatively coupled to the memory (e.g. Fig. 15; paragraph 0113), wherein the processor is configured to:
obtain electrocardiogram (ECG) data comprising a plurality of ECG signals of a subject (e.g. paragraphs 0028, 0035); receive a seeding blood test, wherein the seeding blood test is patient-specific (e.g. paragraph 0041, 0042, – drawing a blood sample from the patient to create a personalized template is construed as a seeding blood test);
process the ECG data and the seeding blood test (e.g. paragraphs 0041-0042) to generate one or more features of the ECG data (e.g. paragraphs 0070, 0091 – an exemplary feature is the slope); enter the one or more features into a statistical model (e.g. paragraphs 0070, 00104 – linear regression is a type of statistical model. Furthermore, machine learning models are also a type of statistical model);
predict, from the statistical model (e.g. paragraphs 0031, 00104, – linear regression is a type of statistical model. Furthermore, machine learning models are also a type of statistical model), a level of an analyte of the subject; and provide the predicted level of the analyte of the subject (e.g. paragraphs 0031, 0034).
However, Friedman does not explicitly teach detect, using linear filtering techniques, artifacts and remove segments of an ECG signal of the ECG data containing those artifacts by: generating a plot of automatic scoring of the ECG data, wherein the plot includes an initial ECG signal and several derived signals used in detecting an artifact in the initial ECG signal; and rejecting a segment of the ECG signal containing the artifact that exceeds an adaptive threshold value by processing data collected by a respiratory belt worn by the subject.
Arnold, in a same field of endeavor of electrocardiogram (ECG) devices, discloses detecting, using linear filtering techniques, artifacts and removing segments of an ECG signal of the ECG data containing those artifacts (e.g. column 17, lines 50-55; column 41, lines 1-10) by: generating a plot of automatic scoring of the ECG data, wherein the plot includes an initial ECG signal and several derived signals used in detecting an artifact in the initial ECG signal (e.g. Fig. 35-35D, 39; column 38, lines 54-65, – calculation of the alternans); and rejecting a segment of the ECG signal containing the artifact that exceeds a threshold value by processing data (e.g. column 41, lines 1-10).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the apparatus of Friedman to include detecting, using linear filtering techniques, artifacts and removing segments of an ECG signal of the ECG data containing those artifacts by: generating a plot of automatic scoring of the ECG data, wherein the plot includes an initial ECG signal and several derived signals used in detecting an artifact in the initial ECG signal; and rejecting a segment of the ECG signal containing the artifact that exceeds a threshold value by processing data, as taught and suggested by Arnold, in order to provide the predictable results that the clinician could consider such data in interpreting the significance of the test results (Arnold, column 41, lines 9-10).
However, Friedman in view of Arnold does not explicitly teach an adaptive threshold and processing data collected by a respiratory belt worn by the subject.
Mottaiyan, in a same field of endeavor of electrocardiogram (ECG) devices, discloses an adaptive threshold (e.g. paragraph 0032).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Friedman and Arnold to include an adaptive threshold, as taught and suggested by Mottaiyan, in order to improve the accuracy of detecting the time instants of the R-peaks in the signal under different noisy conditions (Mottaiyan, paragraph 0032).
However, Friedman in view of Arnold in view of Mottaiyan does not explicitly teach processing data collected by a respiratory belt worn by the subject.
Donnelly, in a same field of endeavor of electrocardiogram (ECG) devices, discloses processing data collected by a respiratory belt worn by the subject (e.g. paragraph 0075).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Friedman, Arnold, and Mottaiyan to include processing data collected by a respiratory belt worn by the subject, as taught and suggested by Donnelly, in order to collect data in a manner that is comfortable for the patient as well as to further improve the quality of the ECG signal.
Regarding claim 68, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly teaches the analyte level prediction apparatus of claim 67 as discussed above, and Friedman further teaches wherein the processor (e.g. Fig. 15 – element 1502; paragraph 0113) is further configured to: identify a plurality of beats in the ECG data (e.g. Fig. 2; paragraphs 0007, 0029); and determine, for each beat in the plurality of beats, a value for a first feature of each beat (e.g. paragraphs 0024-0025, – determining a slope for a plurality of beats), wherein to predict, from the statistical model, the level of the analyte, (e.g. paragraphs 0031, 0034) the processor is further configured to enter the value for the first feature of each beat into the statistical model (e.g. paragraphs 0025-0026).
Regarding claim 69, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly teaches the analyte level prediction apparatus of claim 68 as discussed above, and Friedman further teaches wherein to determine the value for the first feature of each beat (e.g. paragraphs 0024-0025), the processor (e.g. Fig. 15 – element 1502; paragraph 0113) is further configured to calculate, for each beat in the plurality of beats, a slope of at least a portion of a T-wave in each beat between a peak of the T-wave and an end of the T- wave (e.g. paragraphs 0092-0093).
Regarding claim 70, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly teaches the analyte level prediction apparatus of claim 68 as discussed above, and Friedman further teaches wherein to determine the value for the first feature of each beat, the processor is further configured to calculate (e.g. paragraphs 0025-0026), for each beat in the plurality of beats, a magnitude of a peak of a T-wave in each beat (e.g. paragraphs 0024-0025).
Regarding claim 71, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly teaches the analyte level prediction apparatus of claim 68 as discussed above, and Friedman further teaches wherein the processor (e.g. Fig. 15 – element 1502; paragraph 0113) is further configured to: determine a second value for a second feature of each beat (e.g. paragraph 0031, – there are multiple characteristics i.e. features of the electrogram data), wherein to predict, from the statistical model, the level of the analyte (e.g. paragraphs 0031, 0034), the processor is further configured to enter the value for the first feature of each beat and the second value for the second feature of each beat into the statistical model (e.g. paragraphs 0025-0026).
Regarding claim 72, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly teaches the analyte level prediction apparatus of claim 68 as discussed above, and Friedman further teaches wherein to predict, from the statistical model, the level of the analyte (e.g. paragraphs 0031, 0034) the processor is further configured to fit a distribution of the values for the first feature of at least some of the plurality of beats to a probability distribution function (e.g. paragraphs 0029, 0041).
Regarding claim 73, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly teaches the analyte level prediction apparatus of claim 72 as discussed above, and Friedman further teaches wherein the probability distribution function is a normal probability distribution function, a gamma probability distribution function, or a Gaussian probability distribution function (e.g. paragraphs 0031, 00104).
Regarding claim 74, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly teaches the analyte level prediction apparatus of claim 68 as discussed above, and Friedman further teaches wherein to predict, from the statistical model, the level of the analyte (e.g. paragraphs 0031, 0034), the processor is further configured to compare the values for the first feature of at least a subset of the plurality of beats to a pre-defined template (e.g. paragraphs 0029, 0041).
Regarding claim 75, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly teaches the analyte level prediction apparatus of claim 74 as discussed above, and Friedman further teaches wherein the pre-defined template is generated based on assessments of the level of the analyte within a population (e.g. paragraph 0029).
Regarding claim 76, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly teaches the analyte level prediction apparatus of claim 74 as discussed above, and Friedman further teaches wherein the pre-defined template is generated based on assessments of the level of the analyte of the subject from which the ECG data was obtained (e.g. paragraph 0029, 0040).
Regarding claim 77, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly teaches the analyte level prediction apparatus of claim 67 as discussed above, and Friedman further teaches wherein the statistical model (e.g. paragraphs 0070, 00104) comprises a signal template (e.g. paragraphs 0029, 0067), and wherein to predict, from the statistical model, the level of the analyte (e.g. paragraphs 0031, 0034), the processor is further configured to compare the ECG data to the signal template to obtain an indication of the level of the analyte in the subject (e.g. Fig. 15 – element 1502; paragraphs 0029, 0113).
Regarding claim 78, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly teaches the analyte level prediction apparatus of claim 77 as discussed above, and Friedman further teaches wherein the signal template is generated based on assessments of the level of the analyte within a population other than the subject from which the ECG data was obtained (e.g. paragraphs 0029, 0042).
Claims 55-66 and 79-86 are rejected under 35 U.S.C 103 as being unpatentable over Friedman et al. (International Publication No.: WO 2015/048514 A1, – Previously Cited) and further in view of Arnold et al. (US Patent No.: 5,713,367, – Previously Cited) and further in view of Mottaiyan et al. (US Pub.: 2014/0088450 A1) and further in view of Donnelly et al. (US Pub.: 2010/0298899 A1, – Previously Cited) and further in view of Soykan et al. (US Pub.: 2013/0274642 A1, – Previously Cited).
Regarding claim 55, Friedman teaches an analyte level prediction apparatus comprising (e.g. paragraph 0004): a memory (e.g. Fig. 15 – element 1504; paragraph 0113); and a processor (e.g. Fig. 15 – element 1502; paragraph 0113), operatively coupled to the memory (e.g. Fig. 15; paragraph 0113), wherein the processor is configured to:
obtain electrocardiogram (ECG) data comprising a plurality of ECG signals of a subject (e.g. paragraphs 0028, 0035);
receive a seeding blood test, wherein the seeding blood test is patient-specific (e.g. paragraph 0041, 0042, – drawing a blood sample from the patient to create a personalized template is construed as a seeding blood test);
enter the ECG data into a machine learning model trained to classify (e.g. paragraphs 0029, 0031, – can be learned by supervised and unsupervised machine learning classification and clustering techniques), wherein training the machine learning model comprises utilizing the seeding blood test to train the machine learning model (e.g. paragraphs 0041-0042); predict, from the machine learning model, a level of an analyte of the subject (e.g. paragraphs 0028, 0031); and provide the predicted level of the analyte of the subject (e.g. paragraphs 0031, 0034).
However, Friedman does not explicitly teach detecting, using linear filtering techniques, artifacts and removing segments of an ECG signal of the ECG data containing those artifacts by: generating a plot of automatic scoring of the ECG data, wherein the plot includes an initial ECG signal and several derived signals used in detecting an artifact in the initial ECG signal; and rejecting a segment of the ECG signal containing the artifact that exceeds an adaptive threshold value by processing data collected by a respiratory belt worn by the subject and that the machine learning model classifies ECG segments to different analyte levels.
Arnold, in a same field of endeavor of electrocardiogram (ECG) devices, discloses detecting, using linear filtering techniques, artifacts and removing segments of an ECG signal of the ECG data containing those artifacts (e.g. column 17, lines 50-55; column 41, lines 1-10) by: generating a plot of automatic scoring of the ECG data, wherein the plot includes an initial ECG signal and several derived signals used in detecting an artifact in the initial ECG signal (e.g. Fig. 35-35A, 39; column 38, lines 54-65, – calculation of the alternans); and rejecting a segment of the ECG signal containing the artifact that exceeds a threshold value by processing data (e.g. column 41, lines 1-10).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the apparatus of Friedman to include detecting, using linear filtering techniques, artifacts and removing segments of an ECG signal of the ECG data containing those artifacts by: generating a plot of automatic scoring of the ECG data, wherein the plot includes an initial ECG signal and several derived signals used in detecting an artifact in the initial ECG signal; and rejecting a segment of the ECG signal containing the artifact that exceeds a threshold value by processing data, as taught and suggested by Arnold, in order to provide the predictable results that the clinician could consider such data in interpreting the significance of the test results (Arnold, column 41, lines 9-10).
However, Friedman in view of Arnold does not explicitly teach an adaptive threshold and processing data collected by a respiratory belt worn by the subject and that the machine learning model classifies ECG segments to different analyte levels.
Mottaiyan, in a same field of endeavor of electrocardiogram (ECG) devices, discloses an adaptive threshold (e.g. paragraph 0032).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Friedman and Arnold to include an adaptive threshold, as taught and suggested by Mottaiyan, in order to improve the accuracy of detecting the time instants of the R-peaks in the signal under different noisy conditions (Mottaiyan, paragraph 0032).
However, Friedman in view of Arnold in view of Mottaiyan does not explicitly teach processing data collected by a respiratory belt worn by the subject and that the machine learning model classifies ECG segments to different analyte levels.
Donnelly, in a same field of endeavor of electrocardiogram (ECG) devices, discloses processing data collected by a respiratory belt worn by the subject (e.g. paragraph 0075).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Friedman, Arnold, and Mottaiyan to include processing data collected by a respiratory belt worn by the subject, as taught and suggested by Donnelly, in order to collect data in a manner that is comfortable for the patient as well as to further improve the quality of the ECG signal.
However, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly does not explicitly teach that the machine learning model classifies ECG segments to different analyte levels.
Soykan, in a same field of endeavor of potassium analyte measurement methods, discloses classifying ECG segments to different analyte levels (e.g. Fig. 59-60; paragraphs 0258, 0345, 0367, – potassium is an analyte. R waves and T waves are ECG segments).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Friedman, Arnold, Mottaiyan, and Donnelly to incorporate the step of classifying ECG segments to different analyte levels, as taught and suggested by Soykan, for the purpose of researchers/clinicians being able to have an earlier indicator of hyperkalemia and potassium abnormality in the body in order to mitigate/prevent cardiac death of the patient (Soykan, paragraph 0367).
Regarding claim 56, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly in view of Soykan teaches the analyte level prediction apparatus of claim 55 as discussed above, and Friedman further teaches wherein the level is a classification into a category selected from the group consisting of low, normal and high (e.g. paragraph 00110).
Regarding claim 57, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly in view of Soykan teaches the analyte level prediction apparatus of claim 55 as discussed above, and Friedman further teaches wherein the machine learning model comprises one or more of: a feedforward neural network, a deep convolutional neural network, a support vector machine, a Gaussian mixture model, a hidden Markov model, Bayesian decision rules, logistic regression, nearest neighbor model, and decision trees (e.g. paragraph 0031).
Regarding claim 58, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly in view of Soykan teaches the analyte level prediction apparatus of claim 55 as discussed above, and Friedman further teaches wherein to enter the ECG data into the machine learning model (e.g. paragraphs 0029, 0031) the processor is further configured to: identify a plurality of beats in the ECG data (e.g. paragraphs 0007, 0029); and enter the plurality of beats into the machine learning model (e.g. paragraphs 0029, 00126).
Regarding claim 59, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly in view of Soykan teaches the analyte level prediction apparatus of claim 58 as discussed above, and Friedman further teaches wherein entering the ECG data comprises: averaging the plurality of beats to form a representative beat (e.g. paragraph 0077); and entering the representative beat into the machine learning model (e.g. paragraph 0081).
Regarding claim 60, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly in view of Soykan teaches the analyte level prediction apparatus of claim 55 as discussed above, and Friedman further teaches wherein the ECG data is from only a single lead (e.g. paragraph 0072).
Regarding claim 61, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly in view of Soykan teaches the analyte level prediction apparatus of claim 55 as discussed above, and Friedman further teaches wherein the ECG data is from a portable ECG device that measures less than 12 leads (e.g. paragraph 0069).
Regarding claim 62, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly in view of Soykan teaches the analyte level prediction apparatus of claim 55 as discussed above, and Friedman further teaches wherein the analyte is potassium (e.g. paragraph 0026).
Regarding claim 63, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly in view of Soykan teaches the analyte level prediction apparatus of claim 55 as discussed above, and Friedman further teaches wherein the analyte is selected from a group consisting of: potassium, magnesium, phosphorous, calcium, bicarbonate, hydrogen ion, and glucose (e.g. paragraph 0026).
Regarding claim 64, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly in view of Soykan teaches the analyte level prediction apparatus of claim 55 as discussed above, and Friedman further teaches wherein the analyte level is a predicted serum analyte concentration (e.g. paragraphs 0047, 0082).
Regarding claim 65, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly in view of Soykan teaches the analyte level prediction apparatus of claim 55 as discussed above, and Friedman further teaches wherein the machine learning model is trained on training ECG data of a population other than the subject from which the ECG data was obtained (e.g. paragraphs 0029, 0042).
Regarding claim 66, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly in view of Soykan teaches the analyte level prediction apparatus of claim 55 as discussed above, and Friedman further teaches wherein the machine learning model is trained on training ECG data of the subject from which the ECG data was obtained (e.g. paragraph 0042, – “A personalized template can be developed for individual patients, such as by supervised machine learning techniques”).
Regarding claim 79, Friedman teaches a computer-implemented method for assessing a level of an analyte (e.g. paragraph 0067), the method comprising:
obtaining electrocardiogram (ECG) data comprising a plurality of ECG signals of a subject (e.g. paragraphs 0028, 0035); receiving a seeding blood test, wherein the seeding blood test is patient-specific (e.g. paragraph 0041, 0042, - drawing a blood sample from the patient to create a personalized template is construed as a seeding blood test);
extracting, for each of one or more analytes to be measured, a corresponding feature set from the ECG data (e.g. paragraphs 0082-0083); entering the one or more feature sets into a machine learning model trained to classify (e.g. paragraphs 0029, 0031 – can be learned by supervised and unsupervised machine learning classification and clustering techniques), wherein training the machine learning model comprises utilizing the seeding blood test to train the machine learning model (e.g. paragraphs 0041-0042); predicting, using the machine learning model, a level of each of the one or more analytes within the subject (e.g. paragraphs 0028, 0031); and providing the predicted level of each of the one or more analytes within the subject (e.g. paragraphs 0031, 0034).
However, Friedman does not explicitly teach detect, using linear filtering techniques, artifacts and remove segments of an ECG signal of the ECG data containing those artifacts by: generating a plot of automatic scoring of the ECG data, wherein the plot includes an initial ECG signal and several derived signals used in detecting an artifact in the initial ECG signal; and rejecting a segment of the ECG signal containing the artifact that exceeds an adaptive threshold value by processing data collected by a respiratory belt worn by the subject and that the machine learning model classifies ECG segments to different analyte levels.
Arnold, in a same field of endeavor of electrocardiogram (ECG) devices, discloses detecting, using linear filtering techniques, artifacts and removing segments of an ECG signal of the ECG data containing those artifacts (e.g. column 17, lines 50-55; column 41, lines 1-10) by: generating a plot of automatic scoring of the ECG data, wherein the plot includes an initial ECG signal and several derived signals used in detecting an artifact in the initial ECG signal (e.g. Fig. 35-35A, 39; column 38, lines 54-65, – calculation of the alternans); and rejecting a segment of the ECG signal containing the artifact that exceeds a threshold value by processing data (e.g. column 41, lines 1-10).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Friedman to include detecting, using linear filtering techniques, artifacts and removing segments of an ECG signal of the ECG data containing those artifacts by: generating a plot of automatic scoring of the ECG data, wherein the plot includes an initial ECG signal and several derived signals used in detecting an artifact in the initial ECG signal; and rejecting a segment of the ECG signal containing the artifact that exceeds a threshold value by processing data, as taught and suggested by Arnold, in order to provide the predictable results that the clinician could consider such data in interpreting the significance of the test results (Arnold, column 41, lines 9-10).
However, Friedman in view of Arnold does not explicitly teach an adaptive threshold and processing data collected by a respiratory belt worn by the subject and that the machine learning model classifies ECG segments to different analyte levels.
Mottaiyan, in a same field of endeavor of electrocardiogram (ECG) devices, discloses an adaptive threshold (e.g. paragraph 0032).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Friedman and Arnold to include an adaptive threshold, as taught and suggested by Mottaiyan, in order to improve the accuracy of detecting the time instants of the R-peaks in the signal under different noisy conditions (Mottaiyan, paragraph 0032).
However, Friedman in view of Arnold in view of Mottaiyan does not explicitly teach processing data collected by a respiratory belt worn by the subject and that the machine learning model classifies ECG segments to different analyte levels.
Donnelly, in a same field of endeavor of electrocardiogram (ECG) devices, discloses processing data collected by a respiratory belt worn by the subject (e.g. paragraph 0075).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Friedman, Arnold, and Mottaiyan to include processing data collected by a respiratory belt worn by the subject, as taught and suggested by Donnelly, in order to collect data in a manner that is comfortable for the patient as well as to further improve the quality of the ECG signal.
However, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly does not explicitly teach that the machine learning model classifies ECG segments to different analyte levels.
Soykan, in a same field of endeavor of potassium analyte measurement methods, discloses classifying ECG segments to different analyte levels (e.g. Fig. 59-60; paragraphs 0258, 0345, 0367, – potassium is an analyte. R waves and T waves are ECG segments).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Friedman, Arnold, Mottaiyan, and Donnelly to incorporate the step of classifying ECG segments to different analyte levels, as taught and suggested by Soykan, for the purpose of researchers/clinicians being able to have an earlier indicator of hyperkalemia and potassium abnormality in the body in order to mitigate/prevent cardiac death of the patient (Soykan, paragraph 0367).
Regarding claim 80, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly in view of Soykan teaches the computer-implemented method of claim 79 as discussed above, and Friedman further teaches wherein the predicted level of at least an analyte of the one or more analytes is a classification into a category selected from the group consisting of low, normal and high (e.g. paragraph 00110).
Regarding claim 81, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly in view of Soykan teaches the computer-implemented method of claim 79 as discussed above, and Friedman further teaches wherein the machine learning model comprises one or more of: a feedforward neural network, a deep convolutional neural network, a support vector machine, a Gaussian mixture model, a hidden Markov model, Bayesian decision rules, logistic regression, nearest neighbor model, and decision trees (e.g. paragraph 0031).
Regarding claim 82, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly in view of Soykan teaches the computer-implemented method of claim 79 as discussed above, and Friedman further teaches further comprising: identifying a plurality of beats in the ECG data (e.g. Fig. 2; paragraphs 0007, 0029); and extracting the corresponding feature set for each of the one or more analytes to be measured from the plurality of beats (e.g. Fig. 2; paragraphs 0007, 0029, 0082-0083).
Regarding claim 83, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly in view of Soykan teaches the computer-implemented method of claim 79 as discussed above, and Friedman further teaches further comprising: identifying a plurality of beats in the ECG data (e.g. Fig. 2; paragraphs 0007, 0029); averaging the plurality of beats to form a representative beat (e.g. paragraph 0077); and extracting the corresponding feature set for each of the one or more analytes to be measured from the representative beat (e.g. Fig. 6; paragraphs 0074, 0081-0083).
Regarding claim 84, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly in view of Soykan teaches the computer-implemented method of claim 79 as discussed above, and Friedman further teaches wherein the analyte level is a potassium serum concentration (e.g. paragraphs 0047, 0082).
Regarding claim 85, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly in view of Soykan teaches the computer-implemented method of claim 79 as discussed above, and Friedman further teaches wherein the machine learning model is trained on training ECG data of a population other than the subject from which the ECG data was obtained (e.g. paragraphs 0029, 0042).
Regarding claim 86, Friedman in view of Arnold in view of Mottaiyan in view of Donnelly in view of Soykan teaches the computer-implemented method of claim 79 as discussed above, and Friedman further teaches wherein the machine learning model is trained on training ECG data of the subject from which the ECG data was obtained (e.g. paragraph 0029, 0040).
Claims 55-66 and 79-86 are rejected under 35 U.S.C 103 as being unpatentable over Dillon et al. (NPL reference, “Noninvasive Potassium Determination Using a Mathematically Processed ECG”, – Previously Cited) and further in view of Arnold et al. (US Patent No.: 5,713,367, – Previously Cited) and further in view of Mottaiyan et al. (US Pub.: 2014/0088450 A1) and further in view of Donnelly et al. (US Pub.: 2010/0298899 A1 – Previously Cited) and further in view of Friedman et al. (US Pub.: 2013/0184599 A1, hereafter referred to as “Friedman ‘599”, – Previously Cited) and further in view of Soykan and further in view of Huang (NPL reference, “Kernel Based Algorithms…Supervised, Semi-Supervised, and Unsupervised Learning:” – Previously Cited).
Regarding claim 55, Dillon teaches an analyte level prediction apparatus comprising: a memory; and a processor, operatively coupled to the memory (e.g. pg. 3, Methods Section, – a computing environment inherently has a processor and memory), wherein the processor is configured to: obtain electrocardiogram (ECG) data comprising a plurality of ECG signals of a subject (e.g. pg. 3, Data Acquisition Section); enter the ECG data into a machine learning model (e.g. pg. 4, paragraph 1 – fuzzy clustering is a machine learning model; pg. 6); predict, from the machine learning model, a level of an analyte of the subject; and provide the predicted level of the analyte of the subject (e.g. pg. 6 – Discussion Section; pg. 7, paragraph 1).
However, Dillon does not explicitly teach detect, using linear filtering techniques, artifacts and remove segments of an ECG signal of the ECG data containing those artifacts by: generating a plot of automatic scoring of the ECG data, wherein the plot includes an initial ECG signal and several derived signals used in detecting an artifact in the initial ECG signal; and rejecting a segment of the ECG signal containing the artifact that exceeds an adaptive threshold value by processing data collected by a respiratory belt worn by the subject; receive a seeding blood test, wherein the seeding blood test is patient-specific and that the machine learning model classifies ECG segments to different analyte levels as well as wherein training the machine learning model comprises utilizing the seeding blood test to train the machine learning model. Dillon does teach using a machine learning model to cluster ECG data (e.g. pg. 4, paragraph 1; pg. 6).
Arnold, in a same field of endeavor of electrocardiogram (ECG) devices, discloses detecting, using linear filtering techniques, artifacts and removing segments of an ECG signal of the ECG data containing those artifacts (e.g. column 17, lines 50-55; column 41, lines 1-10) by: generating a plot of automatic scoring of the ECG data, wherein the plot includes an initial ECG signal and several derived signals used in detecting an artifact in the initial ECG signal (e.g. Fig. 35-35A, 39; column 38, lines 54-65, – calculation of the alternans); and rejecting a segment of the ECG signal containing the artifact that exceeds a threshold value by processing data (e.g. column 41, lines 1-10).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the apparatus of Dillon to include detecting, using linear filtering techniques, artifacts and removing segments of an ECG signal of the ECG data containing those artifacts by: generating a plot of automatic scoring of the ECG data, wherein the plot includes an initial ECG signal and several derived signals used in detecting an artifact in the initial ECG signal; and rejecting a segment of the ECG signal containing the artifact that exceeds a threshold value by processing data, as taught and suggested by Arnold, in order to provide the predictable results that the clinician could consider such data in interpreting the significance of the test results (Arnold, column 41, lines 9-10).
However, Dillon in view of Arnold does not explicitly teach an adaptive threshold and processing data collected by a respiratory belt worn by the subject; receive a seeding blood test, wherein the seeding blood test is patient-specific and that the machine learning model classifies ECG segments to different analyte levels as well as wherein training the machine learning model comprises utilizing the seeding blood test to train the machine learning model.
Mottaiyan, in a same field of endeavor of electrocardiogram (ECG) devices, discloses an adaptive threshold (e.g. paragraph 0032).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Dillon and Arnold to include an adaptive threshold, as taught and suggested by Mottaiyan, in order to improve the accuracy of detecting the time instants of the R-peaks in the signal under different noisy conditions (Mottaiyan, paragraph 0032).
However, Friedman in view of Arnold in view of Mottaiyan does not explicitly teach processing data collected by a respiratory belt worn by the subject; receive a seeding blood test, wherein the seeding blood test is patient-specific and that the machine learning model classifies ECG segments to different analyte levels as well as wherein training the machine learning model comprises utilizing the seeding blood test to train the machine learning model.
Donnelly, in a same field of endeavor of electrocardiogram (ECG) devices, discloses processing data collected by a respiratory belt worn by the subject (e.g. paragraph 0075).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Dillon, Arnold, and Mottaiyan to include processing data collected by a respiratory belt worn by the subject, as taught and suggested by Donnelly, in order to collect data in a manner that is comfortable for the patient as well as to further improve the quality of the ECG signal.
However, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly does not explicitly teach receive a seeding blood test, wherein the seeding blood test is patient-specific and that the machine learning model classifies ECG segments to different analyte levels as well as wherein training the machine learning model comprises utilizing the seeding blood test to train the machine learning model.
Friedman ‘599, in a same field of endeavor of analyte measurement systems/methods, discloses receive a seeding blood test, wherein the seeding blood test is patient-specific (e.g. paragraph 0007; paragraph 0026, – clinical data such as blood sample can be taken from the subject. Use of a blood sample as an input in an algorithmic system/hidden Markov model is construed as a seeding blood test) as well as wherein training the machine learning model comprises utilizing the seeding blood test to train the machine learning model (e.g. paragraphs 0007, 0026).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Dillon, Arnold, Mottaiyan, and Donnelly to include the step of receiving a seeding blood test, wherein the seeding blood test is patient-specific as well as wherein training the machine learning model comprises utilizing the seeding blood test to train the machine learning model, as taught and suggested by Friedman ‘599, for the purpose of improving the accuracy of the system (Friedman ‘599, paragraphs 0007, 0026).
However, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 does not explicitly teach that the machine learning model classifies ECG segments to different analyte levels.
Soykan, in a same field of endeavor of potassium analyte measurement methods, discloses classifying ECG segments to different analyte levels (e.g. Fig. 59-60; paragraphs 0258, 0345, 0367, – potassium is an analyte. R waves and T waves are ECG segments).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Dillon, Arnold, Mottaiyan, Donnelly and Friedman ‘599 to incorporate the method of classifying ECG segments to different analyte levels, as taught and suggested by Soykan, for the purpose of researchers/clinicians being able to have an earlier indicator of hyperkalemia and potassium abnormality in the body in order to mitigate/prevent cardiac death of the patient (Soykan, paragraph 0367).
However, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 in view of Soykan does not explicitly teach using a machine learning model trained to classify data.
Huang, in a same field of endeavor of machine learning, discloses that it was well established, before the effective filing date of the claimed invention, to train a machine learning model, using supervised learning techniques, to classify data due to their ability in providing more accurate results in various applications (Huang, pg. 2, 22). Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination of Dillon, Arnold, Mottaiyan, Donnelly, Friedman ‘599, and Soykan to include training a machine learning model, using supervised learning techniques, to classify data, as taught and suggested by Huang, for the purpose of yielding more accurate results from provided data (Huang, pg. 2, 22).
Regarding claim 56, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 in view of Soykan in view of Huang teaches the analyte level prediction apparatus of claim 55 as discussed above, and Dillon further teaches wherein the level is a classification into a category selected from the group consisting of low, normal and high (e.g. pg. 6 – remote-monitoring system has alerts for high, low, and normal potassium levels).
Regarding claim 57, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 in view of Soykan in view of Huang teaches the analyte level prediction apparatus of claim 55 as discussed above, and Friedman ‘599 further teaches wherein the machine learning model comprises one or more of: a feedforward neural network, a deep convolutional neural network, a support vector machine, a Gaussian mixture model, a hidden Markov model, Bayesian decision rules, logistic regression, nearest neighbor model, and decision trees (e.g. paragraphs 0007, 0026, – use of hidden Markov model).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Dillon, Arnold, Mottaiyan, Donnelly, Friedman ‘599, Soykan, and Huang to include a hidden Markov model, as taught and suggested by Friedman ‘599, for the purpose of being able to incorporate additional clinical variables besides ECG data when calculating a prediction for the analyte level so that the final results are more accurate (Friedman ‘599, paragraphs 0007, 0026).
Regarding claim 58, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 in view of Soykan in view of Huang teaches the analyte level prediction apparatus of claim 55 as discussed above, and Dillon further teaches wherein to enter the ECG data into the machine learning model, the processor is further configured to: identify a plurality of beats in the ECG data; and enter the plurality of beats into the machine learning model (e.g. pg. 5, Results Section).
Regarding claim 59, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 in view of Soykan in view of Huang teaches the analyte level prediction apparatus of claim 58 as discussed above, and Dillon further teaches wherein entering the ECG data comprises: averaging the plurality of beats to from a representative beat; and entering the representative beat into the machine learning model (e.g. pg. 5, Results Section, – the mean heart rates are ECG parameters).
Regarding claim 60, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 in view of Soykan in view of Huang teaches the analyte level prediction apparatus of claim 55 as discussed above, and Dillon further teaches wherein the ECG data is from only a single lead (e.g. pg.5, Results Section, – lead V4).
Regarding claim 61, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 in view of Soykan in view of Huang teaches the analyte level prediction apparatus of claim 55 as discussed above, and Dillon further teaches wherein the ECG data is from a portable ECG device that measures less than 12 leads (e.g. pg. 7, paragraph 3).
Regarding claim 62, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 in view of Soykan in view of Huang teaches the analyte level prediction apparatus of claim 55 as discussed above, and Dillon further teaches wherein the analyte is potassium (e.g. pg. 3, paragraph 4).
Regarding claim 63, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 in view of Soykan in view of Huang teaches the analyte level prediction apparatus of claim 55 as discussed above, and Dillon further teaches wherein the analyte is selected from a group consisting of: potassium, magnesium, phosphorous, calcium, bicarbonate, hydrogen ion, and glucose (e.g. pg. 7, paragraph 2).
Regarding claim 64, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 in view of Soykan in view of Huang teaches the analyte level prediction apparatus of claim 55 as discussed above, and Dillon further teaches wherein the analyte level is a predicted serum analyte concentration (e.g. pg. 2, paragraph 2).
Regarding claim 65, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 in view of Soykan in view of Huang teaches the analyte level prediction apparatus of claim 55 as discussed above, and Dillon further teaches wherein the machine learning model is trained on training ECG data of a population other than the subject from which the ECG data was obtained (e.g. pg. 4, paragraph 1).
Regarding claim 66, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 in view of Soykan in view of Huang teaches the analyte level prediction apparatus of claim 55 as discussed above, and Dillon further teaches wherein the machine learning model is trained on training ECG data of the subject from which the ECG data was obtained (e.g. pg. 2, Introduction Section).
Regarding claim 79, Dillon teaches a computer-implemented method for assessing a level of an analyte, the method comprising: obtaining electrocardiogram (ECG) data comprising a plurality of ECG signals of a subject (e.g. pg. 3, Data Acquisition Section); extracting, for each of one or more analytes to be measured, a corresponding feature set from the ECG data (e.g. pg. 3,7); entering the one or more feature sets into a machine learning model (e.g. pg. 4, 6); predicting, using the machine learning model, a level of each of the one or more analytes within the subject; and providing the predicted level of each of the one or more analytes within the subject (e.g. pg. 5, Results Section; pg.6-7).
However, Dillon does not explicitly teach detect, using linear filtering techniques, artifacts and remove segments of an ECG signal of the ECG data containing those artifacts by: generating a plot of automatic scoring of the ECG data, wherein the plot includes an initial ECG signal and several derived signals used in detecting an artifact in the initial ECG signal; and rejecting a segment of the ECG signal containing the artifact that exceeds an adaptive threshold value by processing data collected by a respiratory belt worn by the subject; receiving a seeding blood test, wherein the seeding blood test is patient-specific and that the machine learning model classifies ECG segments to different analyte levels as well as wherein training the machine learning model comprises utilizing the seeding blood test to train the machine learning model. Dillon does teach using a machine learning model to cluster ECG data (e.g. pg. 4, paragraph 1; pg. 6).
Arnold, in a same field of endeavor of electrocardiogram (ECG) devices, discloses detecting, using linear filtering techniques, artifacts and removing segments of an ECG signal of the ECG data containing those artifacts (e.g. column 17, lines 50-55; column 41, lines 1-10) by: generating a plot of automatic scoring of the ECG data, wherein the plot includes an initial ECG signal and several derived signals used in detecting an artifact in the initial ECG signal (e.g. Fig. 35-35A, 39; column 38, lines 54-65, – calculation of the alternans); and rejecting a segment of the ECG signal containing the artifact that exceeds a threshold value by processing data (e.g. column 41, lines 1-10).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Dillon to include detecting, using linear filtering techniques, artifacts and removing segments of an ECG signal of the ECG data containing those artifacts by: generating a plot of automatic scoring of the ECG data, wherein the plot includes an initial ECG signal and several derived signals used in detecting an artifact in the initial ECG signal; and rejecting a segment of the ECG signal containing the artifact that exceeds a threshold value by processing data, as taught and suggested by Arnold, in order to provide the predictable results that the clinician could consider such data in interpreting the significance of the test results (Arnold, column 41, lines 9-10).
However, Dillon in view of Arnold does not explicitly teach an adaptive threshold and processing data collected by a respiratory belt worn by the subject; receiving a seeding blood test, wherein the seeding blood test is patient-specific and that the machine learning model classifies ECG segments to different analyte levels as well as wherein training the machine learning model comprises utilizing the seeding blood test to train the machine learning model.
Mottaiyan, in a same field of endeavor of electrocardiogram (ECG) devices, discloses an adaptive threshold (e.g. paragraph 0032).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Dillon and Arnold to include an adaptive threshold, as taught and suggested by Mottaiyan, in order to improve the accuracy of detecting the time instants of the R-peaks in the signal under different noisy conditions (Mottaiyan, paragraph 0032).
However, Friedman in view of Arnold in view of Mottaiyan does not explicitly teach processing data collected by a respiratory belt worn by the subject; receiving a seeding blood test, wherein the seeding blood test is patient-specific and that the machine learning model classifies ECG segments to different analyte levels as well as wherein training the machine learning model comprises utilizing the seeding blood test to train the machine learning model.
Donnelly, in a same field of endeavor of electrocardiogram (ECG) devices, discloses processing data collected by a respiratory belt worn by the subject (e.g. paragraph 0075).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Dillon, Arnold, and Mottaiyan to include processing data collected by a respiratory belt worn by the subject, as taught and suggested by Donnelly, in order to collect data in a manner that is comfortable for the patient as well as to further improve the quality of the ECG signal.
However, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly does not explicitly teach receiving a seeding blood test, wherein the seeding blood test is patient-specific and that the machine learning model classifies ECG segments to different analyte levels as well as wherein training the machine learning model comprises utilizing the seeding blood test to train the machine learning model.
Friedman ‘599, in a same field of endeavor of analyte measurement systems/methods, discloses receiving a seeding blood test, wherein the seeding blood test is patient-specific (e.g. paragraph 0007; paragraph 0026, – clinical data such as blood sample can be taken from the subject. Use of a blood sample as an input in an algorithmic system/hidden Markov model is construed as a seeding blood test) as well as wherein training the machine learning model comprises utilizing the seeding blood test to train the machine learning model (e.g. paragraphs 0007, 0026).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Dillon, Arnold, Mottaiyan, and Donnelly to include the step of receiving a seeding blood test, wherein the seeding blood test is patient-specific as well as wherein training the machine learning model comprises utilizing the seeding blood test to train the machine learning model, as taught and suggested by Friedman ‘599, for the purpose of improving the accuracy of the system (Friedman ‘599, paragraphs 0007, 0026).
However, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 does not explicitly teach that the machine learning model classifies ECG segments to different analyte levels.
Soykan, in a same field of endeavor of potassium analyte measurement methods, discloses classifying ECG segments to different analyte levels (e.g. Fig. 59-60; paragraphs 0258, 0345, 0367, – potassium is an analyte. R waves and T waves are ECG segments).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Dillon, Arnold, Mottaiyan, Donnelly, and Friedman ‘599 to incorporate the method of classifying ECG segments to different analyte levels, as taught and suggested by Soykan, for the purpose of researchers/clinicians being able to have an earlier indicator of hyperkalemia and potassium abnormality in the body in order to mitigate/prevent cardiac death of the patient (Soykan, paragraph 0367).
However, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 in view of Soykan does not explicitly teach using a machine learning model trained to classify data.
Huang, in a same field of endeavor of machine learning, discloses that it was well established, before the effective filing date of the claimed invention, to train a machine learning model, using supervised learning techniques, to classify data due to their ability in providing more accurate results in various applications (Huang, pg. 2, 22). Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination of Dillon, Arnold, Mottaiyan, Donnelly, Friedman ‘599, and Soykan to include training a machine learning model, using supervised learning techniques, to classify data, as taught and suggested by Huang, for the purpose of yielding more accurate results from provided data (Huang, pg. 2, 22).
Regarding claim 80, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 in view of Soykan in view of Huang teaches the computer-implemented method of claim 79 as discussed above, and Dillon further teaches wherein the predicted level of at least an analyte of the one or more analytes is a classification into a category selected from the group consisting of low, normal and high (e.g. pg. 6 – remote-monitoring system has alerts for high, low, and normal potassium levels).
Regarding claim 81, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 in view of Soykan in view of Huang teaches the computer-implemented method of claim 79 as discussed above, and Friedman ‘599 further teaches wherein the machine learning model comprises one or more of: a feedforward neural network, a deep convolutional neural network, a support vector machine, a Gaussian mixture model, a hidden Markov model, Bayesian decision rules, logistic regression, nearest neighbor model, and decision trees (e.g. paragraphs 0007, 0026, – use of hidden Markov model).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Dillon, Arnold, Mottaiyan, Donnelly, Friedman ‘599, Soykan, and Huang to include a hidden Markov model, as taught and suggested by Friedman ‘599, for the purpose of being able to incorporate additional clinical variables besides ECG data when calculating a prediction for the analyte level so that the final results are more accurate (Friedman ‘599, paragraphs 0007, 0026).
Regarding claim 82, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 in view of Soykan in view of Huang teaches the computer-implemented method of claim 79 as discussed above, and Dillon further teaches further comprising: identifying a plurality of beats in the ECG data (e.g. pg. 3-5); and extracting the corresponding feature set for each of the one or more analytes to be measured from the plurality of beats (e.g. pg. 3, 5, 7).
Regarding claim 83, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 in view of Soykan in view of Huang teaches the computer-implemented method of claim 79 as discussed above, and Dillon further teaches further comprising: identifying a plurality of beats in the ECG data (e.g. pg. 3-5); averaging the plurality of beats to form a representative beat (e.g. pg. 3-5); and extracting the corresponding feature set for each of the one or more analytes to be measured from the representative beat (e.g. pg. 3-5).
Regarding claim 84, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 in view of Soykan in view of Huang teaches the computer-implemented method of claim 79 as discussed above, and Dillon further teaches wherein the analyte level is a potassium serum concentration (e.g. pg. 2, paragraph 2).
Regarding claim 85, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 in view of Soykan in view of Huang teaches the computer-implemented method of claim 79 as discussed above, and Dillon further teaches wherein the machine learning model is trained on training ECG data of a population other than the subject from which the ECG data was obtained (e.g. pg. 4, paragraph 1).
Regarding claim 86, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 in view of Soykan in view of Huang teaches the computer-implemented method of claim 79 as discussed above, and Dillon further teaches wherein the machine learning model is trained on training ECG data of the subject from which the ECG data was obtained (e.g. pg. 2, Introduction Section).
Claims 67-78 are rejected under 35 U.S.C 103 as being unpatentable over Dillon and further in view of Arnold and further in view of Mottaiyan and further in view of Donnelly and further in view of Friedman ‘599.
Regarding claim 67, Dillon teaches an analyte level prediction apparatus comprising: a memory; and a processor, operatively coupled to the memory (e.g. pg. 3, Methods Section, – a computing environment inherently has a processor and memory), wherein the processor is configured to: obtain electrocardiogram (ECG) data comprising a plurality of ECG signals of a subject; process the ECG data to generate one or more features of the ECG data; enter the one or more features into a statistical model; predict, from the statistical model, a level of an analyte of the subject; and provide the predicted level of the analyte of the subject (e.g. pg. 5, Results Section).
However, Dillon does not explicitly teach detect, using linear filtering techniques, artifacts and remove segments of an ECG signal of the ECG data containing those artifacts by: generating a plot of automatic scoring of the ECG data, wherein the plot includes an initial ECG signal and several derived signals used in detecting an artifact in the initial ECG signal; and rejecting a segment of the ECG signal containing the artifact that exceeds an adaptive threshold value by processing data collected by a respiratory belt worn by the subject; receive a seeding blood test, wherein the seeding blood test is patient-specific and to process the seeding blood test.
Arnold, in a same field of endeavor of electrocardiogram (ECG) devices, discloses detecting, using linear filtering techniques, artifacts and removing segments of an ECG signal of the ECG data containing those artifacts (e.g. column 17, lines 50-55; column 41, lines 1-10) by: generating a plot of automatic scoring of the ECG data, wherein the plot includes an initial ECG signal and several derived signals used in detecting an artifact in the initial ECG signal (e.g. Fig. 35-35A, 39; column 38, lines 54-65, – calculation of the alternans); and rejecting a segment of the ECG signal containing the artifact that exceeds a threshold value by processing data (e.g. column 41, lines 1-10).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the apparatus of Dillon to include detecting, using linear filtering techniques, artifacts and removing segments of an ECG signal of the ECG data containing those artifacts by: generating a plot of automatic scoring of the ECG data, wherein the plot includes an initial ECG signal and several derived signals used in detecting an artifact in the initial ECG signal; and rejecting a segment of the ECG signal containing the artifact that exceeds a threshold value by processing data, as taught and suggested by Arnold, in order to provide the predictable results that the clinician could consider such data in interpreting the significance of the test results (Arnold, column 41, lines 9-10).
However, Dillon in view of Arnold does not explicitly teach an adaptive threshold and processing data collected by a respiratory belt worn by the subject; receive a seeding blood test, wherein the seeding blood test is patient-specific and to process the seeding blood test.
Mottaiyan, in a same field of endeavor of electrocardiogram (ECG) devices, discloses an adaptive threshold (e.g. paragraph 0032).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Dillon and Arnold to include an adaptive threshold, as taught and suggested by Mottaiyan, in order to improve the accuracy of detecting the time instants of the R-peaks in the signal under different noisy conditions (Mottaiyan, paragraph 0032).
However, Dillon in view of Arnold in view of Mottaiyan does not explicitly teach processing data collected by a respiratory belt worn by the subject; receive a seeding blood test, wherein the seeding blood test is patient-specific and to process the seeding blood test.
Donnelly, in a same field of endeavor of electrocardiogram (ECG) devices, discloses processing data collected by a respiratory belt worn by the subject (e.g. paragraph 0075).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Dillon, Arnold, and Mottaiyan to include processing data collected by a respiratory belt worn by the subject, as taught and suggested by Donnelly, in order to collect data in a manner that is comfortable for the patient as well as to further improve the quality of the ECG signal.
However, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly does not explicitly teach receive a seeding blood test, wherein the seeding blood test is patient-specific and to process the seeding blood test.
Friedman ‘599, in a same field of endeavor of analyte measurement systems/methods, discloses receiving a seeding blood test, wherein the seeding blood test is patient-specific and to process the seeding blood test (e.g. paragraph 0007).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Dillon, Arnold, Mottaiyan, and Donnelly to include receiving a seeding blood test, wherein the seeding blood test is patient-specific and to process the seeding blood test, as taught and suggested by Friedman ‘599, for the purpose of improving the accuracy of the system (Friedman ‘599, paragraphs 0007, 0026).
Regarding claim 68, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 teaches the analyte level prediction apparatus of claim 67 as discussed above, and Dillon further teaches wherein the processor is further configured to: identify a plurality of beats in the ECG data; and determine, for each beat in the plurality of beats (e.g. pg. 5, Results Section), a value for a first feature of each beat, wherein to predict, from the statistical model, the level of the analyte, the processor is further to enter the value for the first feature of each beat into the statistical model (e.g. pg. 3, paragraph 5 to pg. 4, paragraph 1).
Regarding claim 69, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 teaches the analyte level prediction apparatus of claim 68 as discussed above, and Dillon further teaches wherein to determine the value for the first feature of each beat, the processor is further configured to calculate, for each beat in the plurality of beats, a slope of at least a portion of a T-wave in each beat between a peak of the T-wave and the end of an T- wave (e.g. pg. 5, Results Section, – right slope of the T-wave).
Regarding claim 70, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 teaches the analyte level prediction apparatus of claim 68 as discussed above, and Dillon further teaches wherein to determine the value for the first feature of each beat, the processor is further configured to calculate, for each beat in the plurality of beats, a magnitude of a peak of a T-wave in each beat (e.g. pg. 5, Results Section, – T amplitude is construed as the peak of a T-Wave).
Regarding claim 71, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 teaches the analyte level prediction apparatus of claim 68 as discussed above, and Dillon further teaches wherein the processor is further configured to: determine a second value for a second feature of each beat, wherein to predict, from the statistical model, the level of the analyte, the processor is further configured to enter the value for the first feature of each beat and the second value for the second feature of each beat into the statistical model (e.g. pg. 3, paragraph 5 to pg. 4, paragraph 1).
Regarding claim 72, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 teaches the analyte level prediction apparatus of claim 68 as discussed above, and Dillon further teaches wherein to predict, from the statistical model, the level of the analyte, the processor is further configured to fit a distribution of the values for the first feature of at least some of the plurality of beats to a probability distribution function (e.g. pg. 4, paragraph 2).
Regarding claim 73, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 teaches the analyte level prediction apparatus of claim 72 as discussed above, and Dillon further teaches wherein the probability distribution function is a normal probability distribution function, a gamma probability distribution function, or a Gaussian probability distribution function (e.g. pg. 4, paragraph 2).
Regarding claim 74, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 teaches the analyte level prediction apparatus of claim 68 as discussed above, and Dillon further teaches wherein to predict, from the statistical model, the level of the analyte, the processor is further configured to compare the values for the first feature of at least a subset of the plurality of beats to a pre-defined template (pg. 6, paragraph 2).
Regarding claim 75, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 teaches the analyte level prediction apparatus of claim 74 as discussed above, and Dillon further teaches wherein the pre-defined template is generated based on assessments of the level of the analyte within a population (pg. 4, paragraph 4).
Regarding claim 76, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 teaches the analyte level prediction apparatus of claim 74 as discussed above, and Dillon further teaches wherein the pre-defined template is generated based on assessments of the level of the analyte of the subject from which the ECG data was obtained (pg. 6, paragraph 2).
Regarding claim 77, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 teaches the analyte level prediction apparatus of claim 67 as discussed above, and Dillon further teaches wherein the statistical model comprises a signal template, and wherein to predict, from the statistical model, the level of the analyte the processor is further to compare the ECG data to the signal template to obtain an indication of the level of the analyte in the subject (pg. 6, paragraph 2).
Regarding claim 78, Dillon in view of Arnold in view of Mottaiyan in view of Donnelly in view of Friedman ‘599 teaches the analyte level prediction apparatus of claim 77 as discussed above, and Dillon further teaches wherein the signal template is generated based on assessments of the level of the analyte within a population other than the subject from which the ECG data was obtained (e.g. pg. 3, paragraph 4).
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DANIEL TEHRANI whose telephone number is (571)270-0697. The examiner can normally be reached 9:00am-5:00pm.
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, Benjamin Klein can be reached at 571-270-5213. 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.
/D.T./Examiner, Art Unit 3792
/Benjamin J Klein/Supervisory Patent Examiner, Art Unit 3792