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 .
Drawings
Figure 1 should be designated by a legend such as --Prior Art-- because only that which is old is illustrated. See MPEP § 608.02(g). Fig. 1 of the instant application is identical to Fig. 1 of prior art Schmidt et al. (US 2019/0175072).
PNG
media_image1.png
490
681
media_image1.png
Greyscale
PNG
media_image2.png
490
681
media_image2.png
Greyscale
Corrected drawings in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. The replacement sheet(s) should be labeled “Replacement Sheet” in the page header (as per 37 CFR 1.84(c)) so as not to obstruct any portion of the drawing figures. If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
The drawings are objected to as failing to comply with 37 CFR 1.84(p)(4) because reference character “38” has been used to designate two different points as the “as the first fiducial point”. Since a single fiducial is used to calculate the first signal interval, they cannot both be that single fiducial within a single cardiac cycles.
PNG
media_image3.png
367
652
media_image3.png
Greyscale
Corrected drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
Specification
The disclosure is objected to because of the following informalities: In [0065] the first signal interval (36 in Fig. 4) is described as essentially covers the diastole of the mean segment, however the interval is calculated from and encompasses the Gs point in the systole and in Fig. 4 is shown to encompass the time of the S1 heart sound, a marker of the beginning of systole. The second signal interval (40 in Fig. 4) is calculated from and encompasses Dd point in the diastole (42 in Fig. 4) and in Fig. 4 is shown to encompass neither the time of the S1 heart sound nor the Gs point in the systole. These positionings are inconsistent with the assertion that the first interval essentially covers the diastole of the mean segment and the second interval essentially covers the systole of the mean segment, and suggests that the opposite is the case.
PNG
media_image4.png
770
644
media_image4.png
Greyscale
Appropriate correction is required.
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 15-34 rejected under 35 U.S.C. 101 because the claimed invention is directed to abstract ideas without significantly more.
Step 1:
Claims 15-24 recite a method and therefore a process.
Claims 25-26 and 27-35 recite a system and a non-transitory computer-readable medium respectively and are therefore products
Therefore, the claims fall within the statutory categories.
Step 2A, Prong 1:
Claims 15, 25, and 27 recite “obtaining a first signal interval”, “inputting the first signal interval into a first autoencoder”, “determining a first correlation”, “determining an indication of heart failure based on the first correlation”, “receiving a source signal”, “receiving the reconstructed first signal”, “deriving a first signal interval”.
The limitations, as drafted, describe a process that, under its broadest reasonable interpretation, includes performance of the limitation in the mind or a mathematical calculation except for the recitation of “an accelerometer”, “a processor”, and “A non-transitory, computer-readable medium”. That is, other than reciting that “a non-transitory computer-readable medium” and “a processor” are storing instructions for and performing these tasks, nothing in the claim precludes the steps from practically being performed in the human mind or being considered as methods of organizing human activity, or considered a mathematical calculation. MPEP 2106.04(a)(2)(I) states that the sub-grouping “mathematical calculations” includes MPEP 2106.04(a)(2)(II) states that the sub-grouping "managing personal behavior or relationships or interactions between people" include social activities, teaching, and following rules or instructions and MPEP 2106.04(a)(2)(III) states that the courts consider a mental process (thinking) that "can be performed in the human mind, or by a human using a pen and paper" to be an abstract idea. For example, aside from the recitations of “an accelerometer”, “a processor”, and “A non-transitory, computer-readable medium” language, the claim encompasses the user analyzing physiological signals, performing a mathematical calculation on those signals, and drawing a conclusion based on the result of that calculation, or in other words, the claim encompasses a medical professional receiving the results of a test, performing a diagnostic algorithm on those results, and making a diagnosis.
Step 2A, Prong 2:
The claims recite “a non-transitory computer-readable medium” and “a processor” to store instruction for and perform the abstract steps. These components read on a computer implemented system and method and are recited at a high level of generality, i.e., as a generic memory and processor, performing the generic computer functions of storing and processing data. These generic computer limitation are no more than mere instructions to apply the exception using a generic computer components. Accordingly, this additional limitation does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea.
The claims recite “an accelerometer” as the source of the signal that is being obtained and analyzed. This constitutes no more than the insignificant pre-solution activity of data gathering. Accordingly, this additional limitation does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea.
Step 2B:
As discussed with respect to Step 2A Prong Two, the additional elements in the claim amount to no more than mere instructions to apply the exception using a generic computer component. The same analysis applies here in 2B, i.e., mere instructions to apply an exception on a generic computer cannot integrate a judicial except into a practical application at Step 2A or provide an inventive concept in Step 2B.
Under 2019 PEG, a conclusion that an additional element is insignificant extra-solution activity in Step 2A should be re-evaluated in Step 2B to determine if it is more than what is well-understood, routine, conventional activity in the field. The specification in [0060] does not provide any indication that the computer is anything other than a generic, off-the-shelf computer component. Court decisions cited in MPEP 2106.05(d)(II) indicate that computer‐implemented processes not to be significantly more than an abstract idea (and thus ineligible) where the claim, as a whole, amounts to nothing more than generic computer functions merely used to implement an abstract idea, such as an idea that could be done by a human analog (i.e., by hand or by merely thinking). Accordingly, a conclusion that the generic computer functions merely being used to implement an abstract idea is well-understood, routine, conventional activity is supported under Berkheimer Option 2.
In [0003] applicant explains that “SSG {Seismocardiography} is typically measured using an accelerometer.” Therefore, by applicant’s own admission, the use of an accelerometer to measure cardiac signals for SSG analysis is well-understood, routine, and conventional, and does not amount to an inventive concept.
Dependent claims 16-23, 26. & 28-34 further limit the abstract idea already indicated in independent claims 15, 25, and 27 and they are ineligible for the same reasons provided for claims 15, 25, and 27 above.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claim(s) 15-16, 19, 21-28, 31, & 33-34 is/are rejected under 35 U.S.C. 103 as being unpatentable over Inan et al. (WO/2022/115228, published June-02 2022, hereinafter Inan. Paragraph numbers below refer to equivalent document US 2023/0293082.) in view of Melaet et al. (US 2024/0389947, effectively filed Aug-26 2021, hereinafter Melaet).
Regarding claim 15, Inan teaches a method for determining an indication of heart failure of a subject (“the disclosure provides a wearable, minimally obtrusive system to remotely monitor HF [heart failure] patients using wearable seismocardiography (SCG)” in [0157] and “the disclosed technology includes methods for assessing heart health… the method can be performed by the wearable device” in [0180]), wherein the method comprises: obtaining a first signal interval from a source signal (Fig. 6 A-c and “we detected R-peaks in the ECG signal and subsequently performed beat segmentation in SCG signals” in [0214]) recorded by an accelerometer placed on the chest of the subject (Fig. 1 A-E and “the second sensor can be contained in a wearable housing configured to be worn on a chest of the user… the second sensor can include an accelerometer” in [0022]-[0023]), wherein the first signal interval corresponds to a first subinterval of a heart cycle (“ beat segmentation of SCG signal was carried out in the following way: 150 ms before the R-peak (to include ventricular diastole) and 15th percentile of RR intervals after the R-peak (to include only the current beat)” in [0300]); and determining an indication of heart failure (“The method can include assessing, based on filling characteristics of the heart, heart health. For example, the assessment of heart health can include a classification of a clinical status of heart failure (HF) in the user. Alternatively, or in addition, the assessment of heart health can include data indicative of changes in filling characteristics of the heart of a user. For example, changes in hemodynamics, filling pressure, pulmonary artery (PA) pressure, pulmonary capillary wedge pressure (PCWP), and the like, or any combination thereof” in [0184]).
Inan does not teach inputting the first signal interval into a first autoencoder that has been trained on corresponding first signal intervals obtained from healthy subjects, and that outputs a reconstructed first signal interval; determining a first correlation between the first signal interval and the reconstructed first signal interval; and an indication based on the first correlation.
However, attention is drawn to the Melaet reference. Melaet teaches a device and a corresponding computer-implemented method are provided for training a fetal heart rate anomaly detection ([0006]) by determining an anomaly score for a fetal heart rate segment. The inventors envisaged to use machine learning to let a trainable model learn to predict normal responses of the fetal heart rate to uterine contractions. By comparing the actual response to the predicted normal response, an anomaly score may be determined that is indicative of whether the fetus shows an abnormal reaction of the fetal heart rate to uterine activity, and thereby indicative of fetal health ([0007]).
The fetal heart rate prediction model is preferably trained on data of healthy fetuses only. Consequently, the prediction model may learn to predict representations of fetal heart segments for healthy fetuses, in particular its response to uterine activity. The actual fetal heart rate response of the fetus can then be compared with the predicted response. When the two representations are similar, it is likely the fetus comes from the healthy population. If the representations show a significant difference, this can indicate that the condition of the fetus is compromised, so the clinician can be warned for example to take additional diagnostic or intervention action. In other words, the anomaly score is determined based on the expectation that the FHR prediction is less accurate in case the health of the fetus is compromised and the fetus no longer responds in the same way as a healthy fetus would do ([0009]). A trained fetal heart rate decoder model corresponding to the encoder model may be applied to the predicted representation of the second fetal heart rate segment to obtain a predicted second fetal heart rate segment that can be output ([0018]) (outputs a reconstructed first signal interval).
A fetal heart rate decoder model that can for example be trained together with the fetal heart rate encoder model as an autoencoder (inputting the first signal interval into a first autoencoder). ([0160]). Since the FHR predictor model is typically only trained on healthy fetuses (autoencoder that has been trained on corresponding first signal intervals obtained from healthy subjects), essentially modelling the response of a healthy fetus to UC, a large divergence may indicate a discrepancy between the two distributions and therefore possibly an anomalous event. Interestingly, data of unhealthy fetuses, i.e., anomalies, is not needed ([0179]).
The predicted encoding may be used for anomaly detection by performing a comparison of the predicted representation of the second fetal heart rate segment to the second fetal heart rate segment that was actually measured (determining a first correlation between the first signal interval and the reconstructed first signal interval), based on which an anomaly score may be determined ([0174]). The comparison may be computed for example as a mean squared error reflecting the difference between the predicted segment and actual segment ([0153]) (an indication based on the first correlation.). Although this anomaly score is not directly taught as an indicator of heart failure, it is an indicator of cardiac dysfunction, and the same method of assessing cardiac signals could be used to assess heart health, including a classification of a clinical status of heart failure.
It would have been obvious to one of ordinary skill in the art at the time of filing of the instant application to use an autoencoder to reconstruct cardiac signals and compare the reconstructed signal to the received signal, and use that comparison as an indicator of cardiac dysfunction as taught by Melaet to the system and methods for assessing heart health taught by Inan for the purpose of differentiating between a healthy and an unhealthy response without the need to train a model on specifically unhealthy data.
Regarding Claims 16, 19, 21-24 Inan in view of Melaet teaches the method of claim 15.
Inan further teaches wherein the first signal interval covers a portion of a systole of the heart cycle (Fig. 14 and “Three intervals were defined as illustrated in FIG. 14A (decompensated signal) and FIG. 14B (compensated signal): ventricular diastolic 1 (dias1 from −150 ms to 0 ms); systolic (sys, from 0 ms to 300 ms); and ventricular diastolic 2 (dias2, from 300 ms to 650 ms).” in [0281]),
PNG
media_image5.png
306
309
media_image5.png
Greyscale
wherein the first correlation is based on one a Root-Mean-Square Error (RMSE) (As a performance metric, we used root mean squared error (RMSE) between the regression output and the ground truth value measured during the RHC procedure” in [0227]).wherein the method further comprises: obtaining a second signal interval from the source signal recorded with an accelerometer placed on the chest of a subject, wherein the second signal interval corresponds to a second subinterval of a heart cycle; inputting the second signal interval into a second autoencoder that has been trained on corresponding second signal intervals obtained from healthy subjects, and that outputs a reconstructed second signal interval; and determining a second correlation between the second signal interval and the reconstructed second signal interval(Fig. 14 “For a given beat, these features can be computed over different physiologically defined time intervals corresponding to the phases of systole and diastole” in [0281]), wherein determining the indication of heart failure is further based on the second correlation (“The predicted encoding 352′ may be used for anomaly detection by performing a comparison 370 of the predicted representation 352′ of the second fetal heart rate segment to the second fetal heart rate segment 342 that was actually measured,” in [0174]), wherein obtaining the first signal comprises: recording the source signal with the accelerometer placed on the chest of a subject (Fig. 1 A-E and “the second sensor can be contained in a wearable housing configured to be worn on a chest of the user… the second sensor can include an accelerometer” in [0022]-[0023]), wherein the source signal is recorded over a period covering a plurality of cardiac cycles of the subject; dividing the source signal into a plurality of signal segments wherein each signal segment covers a single cardiac cycle (Fig. 6 and “For detection, 12 second windows were used to detect the R-peaks. Using the R-peaks, beat segmentation of SCG signal was carried out in the following way: 150 ms before the R-peak (to include ventricular diastole) and 15th percentile of RR intervals after the R-peak (to include only the current beat) was delimited as the start and end of a beat [0299]-[0300]);
PNG
media_image6.png
410
536
media_image6.png
Greyscale
aligning the plurality of signal segments (“we leveraged DTW and compared the DTW distances from different portions… the DTW is a time-series analysis method to align signals and find similarities between signals” in [0250]); determining a mean segment from the plurality of signal segments (“types of features that are examined here, based on the foundational prior work in SCG signal analysis in patients with HF: (1) ensemble averaged beat features from the diastolic and systolic time intervals of the signal” in [0279]); and determining the first signal interval in the mean segment wherein determining the first signal interval in the mean segment comprises: identifying a first fiducial point in the mean segment; and positioning the first signal interval relative to the first fiducial point (“For a given beat, these features can be computed over different physiologically defined time intervals corresponding to the phases of systole and diastole, which can be indexed based on the ECG signal by defining the location of the ECG R-peak as time zero” in [0281]), wherein the method further comprises displaying the indication of heart failure (Fig. 11 and “The method 1100 can include assessing 1108, based on filling characteristics of the heart, heart health. For example, the assessment of heart health can include a classification of a clinical status of heart failure (HF) in the user. Alternatively, or in addition, the assessment of heart health can include data indicative of changes in filling characteristics of the heart of a user. For example, changes in hemodynamics, filling pressure, pulmonary artery (PA) pressure, pulmonary capillary wedge pressure (PCWP), and the like, or any combination thereof” in [1840])
Inan does not teach the element wherein the first correlation is a correlation measure between the first signal interval and the reconstructed first signal interval (of claim 19).
However, as previously explained above, Melaet teach performing a comparison of the predicted representation of the second fetal heart rate segment to the second fetal heart rate segment that was actually measured ([0174]) and that “the comparison may be computed for example as a mean squared error reflecting the difference between the predicted segment and actual segment” ([0153]) (wherein the first correlation is a correlation measure between the first signal interval and the reconstructed first signal interval and wherein the first correlation is based on one of a Mean-Square Error).
Regarding claim 25, Inan teaches a system for determining an indication of heart failure of a subject, wherein the system comprises ((“the disclosure provides a wearable, minimally obtrusive system to remotely monitor HF [heart failure] patients using wearable seismocardiography (SCG)” in [0157]): an accelerometer configured, when placed on the chest of a subject, to record a source signal (Fig. 1 A-E and “the second sensor can be contained in a wearable housing configured to be worn on a chest of the user… the second sensor can include an accelerometer” in [0022]-[0023]); and a processor operatively connected to the accelerometer (“a processing system comprising one or more processors operatively coupled to the memory and one or more of the sensors” in [0009]) and configured to perform a method (“ the processing system configured to receive the ECG signal, receive at least one of the SCG signals, and generate an assessment of heart health of the user by determining one or more hemodynamic parameters based on the ECG signal and the received at least one SCG signal.” In [0009]) comprising the steps of (i) obtaining a first signal interval from the source signal (Fig. 6 A-c and “we detected R-peaks in the ECG signal and subsequently performed beat segmentation in SCG signals” in [0214]), wherein the first signal interval corresponds to a first subinterval of a heart cycle of the subject (“ beat segmentation of SCG signal was carried out in the following way: 150 ms before the R-peak (to include ventricular diastole) and 15th percentile of RR intervals after the R-peak (to include only the current beat)” in [0300]); and (iv) determining an indication of heart failure (“The method can include assessing, based on filling characteristics of the heart, heart health. For example, the assessment of heart health can include a classification of a clinical status of heart failure (HF) in the user. Alternatively, or in addition, the assessment of heart health can include data indicative of changes in filling characteristics of the heart of a user. For example, changes in hemodynamics, filling pressure, pulmonary artery (PA) pressure, pulmonary capillary wedge pressure (PCWP), and the like, or any combination thereof” in [0184]). based on the first correlation.
Inan does not teach (ii) inputting the first signal interval into a first autoencoder that has been trained on corresponding first signal intervals obtained from healthy subjects, and that outputs a reconstructed first signal interval and (iii) receiving the reconstructed first signal interval and determining a first correlation between the first signal interval and the reconstructed first signal interval.
However attention is drawn to the Melaet reference. Melaet teaches a device and a corresponding computer-implemented method are provided for training a fetal heart rate anomaly detection ([0006]) by determining an anomaly score for a fetal heart rate segment. The inventors envisaged to use machine learning to let a trainable model learn to predict normal responses of the fetal heart rate to uterine contractions. By comparing the actual response to the predicted normal response, an anomaly score may be determined that is indicative of whether the fetus shows an abnormal reaction of the fetal heart rate to uterine activity, and thereby indicative of fetal health ([0007]).
The fetal heart rate prediction model is preferably trained on data of healthy fetuses only. Consequently, the prediction model may learn to predict representations of fetal heart segments for healthy fetuses, in particular its response to uterine activity. The actual fetal heart rate response of the fetus can then be compared with the predicted response. When the two representations are similar, it is likely the fetus comes from the healthy population. If the representations show a significant difference, this can indicate that the condition of the fetus is compromised, so the clinician can be warned for example to take additional diagnostic or intervention action. In other words, the anomaly score is determined based on the expectation that the FHR prediction is less accurate in case the health of the fetus is compromised and the fetus no longer responds in the same way as a healthy fetus would do ([0009]). A trained fetal heart rate decoder model corresponding to the encoder model may be applied to the predicted representation of the second fetal heart rate segment to obtain a predicted second fetal heart rate segment that can be output ([0018]) (outputs a reconstructed first signal interval).
A fetal heart rate decoder model that can for example be trained together with the fetal heart rate encoder model as an autoencoder (inputting the first signal interval into a first autoencoder). ([0160]). Since the FHR predictor model is typically only trained on healthy fetuses (autoencoder that has been trained on corresponding first signal intervals obtained from healthy subjects), essentially modelling the response of a healthy fetus to UC, a large divergence may indicate a discrepancy between the two distributions and therefore possibly an anomalous event. Interestingly, data of unhealthy fetuses, i.e., anomalies, is not needed (0179].
The predicted encoding may be used for anomaly detection by performing a comparison of the predicted representation of the second fetal heart rate segment to the second fetal heart rate segment that was actually measured (determining a first correlation between the first signal interval and the reconstructed first signal interval), based on which an anomaly score may be determined ([0174]). The comparison may be computed for example as a mean squared error reflecting the difference between the predicted segment and actual segment ([0153]) (an indication based on the first correlation.). Although this anomaly score is not directly taught as an indicator of heart failure, it is an indicator of cardiac dysfunction, and the same method of assessing cardiac signals could be used to assess heart health, including a classification of a clinical status of heart failure.
It would have been obvious to one of ordinary skill in the art at the time of filing of the instant application to use an autoencoder to reconstruct cardiac signals and compare the reconstructed signal to the received signal, and use that comparison as an indicator of cardiac dysfunction as taught by Melaet to the system and methods for assessing heart health taught by Inan for the purpose of differentiating between a healthy and an unhealthy response with the need to train a model on specifically unhealthy data.
Regarding claim 26, Inan in view of Melaet teaches the system of claim 25.
Inan further teaches further comprising a non-transient memory operatively connected to the processor and configured to store computer-readable program code instructions that, when executed by the processor, instruct the processor to perform the method (Fig. 1B and “The electronics 170 can include a processor and a memory. For example, the electronics 170 can include CPU, microprocessor, and the like. The memory can comprise logical instructions that, when executed by the processor, cause the processor to carry out one of more of the functions disclosed herein. Alternatively, or in addition, the memory can include a removeable memory card” in [0178]).
Regarding claim 27 Inan teaches a non-transitory, computer-readable medium storing instructions for executing a computer-implemented method (Fig. 1B and “The electronics 170 can include a processor and a memory. For example, the electronics 170 can include CPU, microprocessor, and the like. The memory can comprise logical instructions that, when executed by the processor, cause the processor to carry out one of more of the functions disclosed herein. Alternatively, or in addition, the memory can include a removeable memory card” in [0178]) for determining an indication of heart failure of a subject (“ the assessment of heart health can include a classification of a clinical status of heart failure (HF) in the user” in [0184]), the method comprising the steps of (i) obtaining a first signal interval from the source signal (Fig. 6 A-c and “we detected R-peaks in the ECG signal and subsequently performed beat segmentation in SCG signals” in [0214]), wherein the first signal interval corresponds to a first subinterval of a heart cycle of the subject (“ beat segmentation of SCG signal was carried out in the following way: 150 ms before the R-peak (to include ventricular diastole) and 15th percentile of RR intervals after the R-peak (to include only the current beat)” in [0300]); and (iv) determining an indication of heart failure (“The method can include assessing, based on filling characteristics of the heart, heart health. For example, the assessment of heart health can include a classification of a clinical status of heart failure (HF) in the user. Alternatively, or in addition, the assessment of heart health can include data indicative of changes in filling characteristics of the heart of a user. For example, changes in hemodynamics, filling pressure, pulmonary artery (PA) pressure, pulmonary capillary wedge pressure (PCWP), and the like, or any combination thereof” in [0184]). based on the first correlation.
Inan does not teach (ii) inputting the first signal interval into a first autoencoder that has been trained on corresponding first signal intervals obtained from healthy subjects, and that outputs a reconstructed first signal interval and (iii) receiving the reconstructed first signal interval and determining a first correlation between the first signal interval and the reconstructed first signal interval.
However attention is drawn to the Melaet reference. Melaet teaches a device and a corresponding computer-implemented method are provided for training a fetal heart rate anomaly detection ([0006]) by determining an anomaly score for a fetal heart rate segment. The inventors envisaged to use machine learning to let a trainable model learn to predict normal responses of the fetal heart rate to uterine contractions. By comparing the actual response to the predicted normal response, an anomaly score may be determined that is indicative of whether the fetus shows an abnormal reaction of the fetal heart rate to uterine activity, and thereby indicative of fetal health ([0007]).
The fetal heart rate prediction model is preferably trained on data of healthy fetuses only. Consequently, the prediction model may learn to predict representations of fetal heart segments for healthy fetuses, in particular its response to uterine activity. The actual fetal heart rate response of the fetus can then be compared with the predicted response. When the two representations are similar, it is likely the fetus comes from the healthy population. If the representations show a significant difference, this can indicate that the condition of the fetus is compromised, so the clinician can be warned for example to take additional diagnostic or intervention action. In other words, the anomaly score is determined based on the expectation that the FHR prediction is less accurate in case the health of the fetus is compromised and the fetus no longer responds in the same way as a healthy fetus would do ([0009]). A trained fetal heart rate decoder model corresponding to the encoder model may be applied to the predicted representation of the second fetal heart rate segment to obtain a predicted second fetal heart rate segment that can be output ([0018]) (outputs a reconstructed first signal interval).
A fetal heart rate decoder model that can for example be trained together with the fetal heart rate encoder model as an autoencoder (inputting the first signal interval into a first autoencoder). ([0160]). Since the FHR predictor model is typically only trained on healthy fetuses (autoencoder that has been trained on corresponding first signal intervals obtained from healthy subjects), essentially modelling the response of a healthy fetus to UC, a large divergence may indicate a discrepancy between the two distributions and therefore possibly an anomalous event. Interestingly, data of unhealthy fetuses, i.e., anomalies, is not needed (0179].
The predicted encoding may be used for anomaly detection by performing a comparison of the predicted representation of the second fetal heart rate segment to the second fetal heart rate segment that was actually measured (determining a first correlation between the first signal interval and the reconstructed first signal interval), based on which an anomaly score may be determined ([0174]). The comparison may be computed for example as a mean squared error reflecting the difference between the predicted segment and actual segment ([0153]) (an indication based on the first correlation.). Although this anomaly score is not directly taught as an indicator of heart failure, it is an indicator of cardiac dysfunction, and the same method of assessing cardiac signals could be used to assess heart health, including a classification of a clinical status of heart failure.
It would have been obvious to one of ordinary skill in the art at the time of filing of the instant application to use an autoencoder to reconstruct cardiac signals and compare the reconstructed signal to the received signal, and use that comparison as an indicator of cardiac dysfunction as taught by Melaet to the system and methods for assessing heart health taught by Inan for the purpose of differentiating between a healthy and an unhealthy response with the need to train a model on specifically unhealthy data.
Regarding claim 28 & 31 Inan in view of Melaet teaches the non-transitory computer-readable medium of claim 27.
Inan further teaches wherein, in the computer- implemented method, the first signal interval covers a portion of a systole of the heart cycle of the subject (Fig. 14 “For a given beat, these features can be computed over different physiologically defined time intervals corresponding to the phases of systole and diastole” in [0281]), wherein, in the computer- implemented method, the first correlation is a correlation measure between the first signal interval and the reconstructed first signal interval, and wherein the first correlation is based on one of a Root-Mean-Square Error (RMSE) (As a performance metric, we used root mean squared error (RMSE) between the regression output and the ground truth value measured during the RHC procedure” in [0227]).
Melaet also teach performing a comparison of the predicted representation of the second fetal heart rate segment to the second fetal heart rate segment that was actually measured ([0174]) and that “the comparison may be computed for example as a mean squared error reflecting the difference between the predicted segment and actual segment” ([0153]) (wherein the first correlation is a correlation measure between the first signal interval and the reconstructed first signal interval and wherein the first correlation is based on one of a Mean-Square Error).
Regarding claim 33 Inan in view of Melaet teaches the non-transitory computer-readable medium of claim 27.
Melaet further teaches wherein the computer- implemented method further comprises the steps of: (vii) obtaining a second signal interval from the source signal recorded with an accelerometer placed on the chest of a subject, wherein the second signal interval corresponds to a second subinterval of a heart cycle of the subject; (Fig. 14 and “Three intervals were defined as illustrated in FIG. 14A (decompensated signal) and FIG. 14B (compensated signal): ventricular diastolic 1 (dias1 from −150 ms to 0 ms); systolic (sys, from 0 ms to 300 ms); and ventricular diastolic 2 (dias2, from 300 ms to 650 ms).” in [0281]), (viii) inputting the second signal interval into a second autoencoder that has been trained on corresponding second signal intervals obtained from healthy subjects; (ix) outputting a reconstructed second signal interval by the second autoencoder ((Fig. 14 “For a given beat, these features can be computed over different physiologically defined time intervals corresponding to the phases of systole and diastole” in [0281])); and (x) determining a second correlation between the second signal interval and the reconstructed second signal interval ((“The predicted encoding 352′ may be used for anomaly detection by performing a comparison 370 of the predicted representation 352′ of the second fetal heart rate segment to the second fetal heart rate segment 342 that was actually measured,” in [0174])), wherein determining the indication based on the second correlation.
Regarding claim 34 Inan in view of Melaet teaches the non-transitory computer-readable medium of claim 27,
Inan further teaches wherein, in the computer- implemented method, the step of (ii) deriving the first signal interval comprises the substeps of: (ii)(a) recording the source signal with the accelerometer placed on the chest of the subject (Fig. 1 A-E and “the second sensor can be contained in a wearable housing configured to be worn on a chest of the user… the second sensor can include an accelerometer” in [0022]-[0023]), wherein the source signal is recorded over a period covering a plurality of cardiac cycles of the subject; (Fig. 6 and “For detection, 12 second windows were used to detect the R-peaks. Using the R-peaks, beat segmentation of SCG signal was carried out in the following way: 150 ms before the R-peak (to include ventricular diastole) and 15th percentile of RR intervals after the R-peak (to include only the current beat) was delimited as the start and end of a beat [0299]-[0300]);(ii)(b) dividing the source signal into a plurality of signal segments, wherein each signal segment covers a single cardiac cycle of the subject; (ii)(c) aligning the plurality of signal segments (“we leveraged DTW and compared the DTW distances from different portions… the DTW is a time-series analysis method to align signals and find similarities between signals” in [0250]); (ii)(d) determining a mean segment from the plurality of signal segments (“types of features that are examined here, based on the foundational prior work in SCG signal analysis in patients with HF: (1) ensemble averaged beat features from the diastolic and systolic time intervals of the signal” in [0279]); and (ii)(e) determining the first signal interval in the mean segment (“For a given beat, these features can be computed over different physiologically defined time intervals corresponding to the phases of systole and diastole, which can be indexed based on the ECG signal by defining the location of the ECG R-peak as time zero” in [0281]).
Claim(s) 17-18 & 28-29 is/are rejected under 35 U.S.C. 103 as being unpatentable over Inan in view of Melaet, in further view of Amos et al. (US 2021/0378597, hereinafter Amos)
Regarding Claims 17-18 & 28-29, Inan in view of Melaet teaches the method of claim 15 and the non-transitory computer-readable medium of claim 27.
Inan in view of Melaet does not teach wherein the first autoencoder compresses the first signal interval to between 5 and 15 nodes and wherein the first autoencoder is a single layer autoencoder.
However, attention is drawn to the Amos reference. Amos teaches a system and method for reducing noise of intracardiac electrocardiograms using an autoencoder and utilizing and refining intracardiac and body surface electrocardiograms using one or more deep learning training loss functions ([0002]. The method includes receiving raw signal data including signal noise and encoding, by a denoised autoencoder, the raw signal data. The encoding includes performing a denoising autoencoder operation on the raw signal data to produce a latent representation. The method also includes decoding, by the denoised autoencoder, the latent representation to produce clean signal data reconstructed without the signal noise ([0005]). The autoencoder encodes the raw signal data by performing a denoising autoencoder operation (e.g., passing the raw signal data through a deep neural network, such as a convolutional neural network (CNN) architecture) to reduce a dimensionality of the raw signal data and retain only important information therein ([0030]). Then the autoencoder further decodes the latent representation to produce, as an output, a reconstruction of a clean version of the raw signal data. That is, in this decoding stage, the autoencoder reconstructs a clean signal data from the latent representation ([0031]). In Fig. 6B, Amos shows the use of a single layer autoencoder (wherein the first autoencoder is a single layer autoencoder) with 4 nodes. Amos also teaches that if the denoising autoencoder implements the convolutional neural network architecture, other configurable aspects of the architecture can include a number of filters at each stage, kernel size, a number of kernels per layer ([01131]). Although Fig. 6 only shows 4 nodes, Amos teaches that both the number of layers and the number of kernels which generate nodes in each layer is a configurable element (i.e. a result-effective variable. The specification of the instant application lists the number of nodes that may be used in the autoencoder, and lists multiple ranges, including the claimed range of 5 to 15, and less than 15, less than 12, or less than 10, or in the range 6 to 12, or 7 to 9. It provides no evidence of the criticality of the claimed range. As such, one of ordinary skill in the art could arrive at the claimed range of 5 to 12 nodes through the routine optimization of the teachings of Amos.
PNG
media_image7.png
506
627
media_image7.png
Greyscale
It would have been obvious to one of ordinary skill in the art at the time of filing of the instant application to use the single layer autoencoder with 5 to 12 nodes taught by Amos as the autoencoder used in the system, method, and non-transitory computer-readable medium taught by Inan in view of Melaet for the purpose of reconstructing clean signal data from the latent representation of the raw signal data.
Claim(s) 20 & 32 is/are rejected under 35 U.S.C. 103 as being unpatentable over Inan in view of Melaet, in further view of Rao and Garcia. (US 2023/0240922, effectively filed Feb-02, 2022, hereinafter Rao)
Regarding claims 20 & 32 Inan in view of Melaet teaches the method of claim 15 and the non-transitory computer-readable medium of claim 27.
Inan in view of Melaet does not teach wherein the indication of heart failure is a probability score for heart failure, wherein the probability score is based on a logistic regression model and the first correlation.
However, attention is drawn to the Rao reference. Rao teaches s ystems, methods, and techniques for predicting risk of congestive heart failure (CHF) and/or other cardiac failure in users of bed systems, such as smart beds, and that the disclosed technology can be used to quantify risk of CHF as a probability measurement, not merely identification of a risk of CHF. A logistic regression model (wherein the probability score is based on a logistic regression model) can be used to determine the probability measurement (wherein the indication of heart failure is a probability score for heart failure) ([0004]). The user data can include ballistocardiogram (BCG) signals. The logistic regression model could have been trained with a training dataset of historic user data associated with the user or the model could have been trained with a training dataset of historic user data associated with different users of the bed systems ([0008]). Rao explains that a logistic regression is a process of modeling probability of a discrete outcome given one or more input variables. A logistic regression model can output a binary value, such as 0 or 1, true or false, and/or yes or no. In such a scenario, the computer system can normalize the output from the regression model by multiplying the output by a predetermined factor to generate the probability measurement of CHF risk ([0057]).
It would have been obvious to one of ordinary skill in the art at the time of filing of the instant application to apply the logistic regression model for determining a probably of heart failure taught by Rao to the system and system, method, and non-transitory computer-readable medium taught by Inan in view of Melaet for the purpose of generating a probability model rather than just a risk indicator of heart failure based on multiple input variables, including ballistocardiogram signals.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
US 2021/0127983
Abstract
A system and method for myocardial performance determination is provided. The present invention provided for generating a first dataset representing a set of events associated with a pre-defined parameter of a biomarker extracted from physiological parameters of a subject. The set of events is determined by processing the pre-defined parameter at a first level and a second level of a multi-level artificial neural network architecture recursively for a pre-defined number of times. Further, generating second dataset representing characteristics associated with the set of events by processing first dataset at third level and fourth level of multi-level artificial neural network architecture. Further, computing set of values associated with set of events by processing second dataset at fifth level of multi-level artificial neural network architecture. Further, computing myocardial performance index based on set of values. The myocardial performance index is representative of myocardial performance of the subject.
Description of Invention
The present invention discloses a system and a method for contactless monitoring of cardiac activity for myocardial performance determination. In particular, the present invention discloses a system and a method for contactless monitoring of cardiac activity for myocardial performance determination based on computation of myocardial performance index (MPI) associated with heartbeats of a subject. The present invention provides for a system and a method for cardiac activity monitoring and detection with precision and accuracy without the need of a physician. The present invention provides for a system and a method for MPI computation based on detection of ballistocardiographic (BCG) signals associated with a person or a subject. The present invention further provides for a system and a method with in-built intelligent mechanism for myocardial performance index (MPI) computation based on artificial intelligence techniques and machine learning techniques. The present invention provides for a system and a method for detection of myocardial performance indicators in an unobtrusive manner. Further, the present invention provides for a system and a method for regular and effective monitoring of a person's cardiac health. Furthermore, the present invention provides for a preemptive system and a method for monitoring cardiac activity of not only a person with cardiac disease history, but for a healthy person as a preventive measure. Furthermore, the present invention provides for a system and a method which is easily deployable, easily implementable, easily portable and cost effective.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to WILLIAM P ADAMS whose telephone number is (571)270-0136. The examiner can normally be reached 9am-6pm M-Th.
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, Unsu Jung can be reached at (571)272-8506. 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.
/W.P.A./ Examiner, Art Unit 3792
/AMANDA L STEINBERG/ Examiner, Art Unit 3792