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 .
Claims 1, 7 and 13 have been amended. Claims 1-18 have been examined.
Response to Arguments/Amendments
The prior claim objections are withdrawn in view of the claim amendments.
The following arguments filed 6/23/2026 have been fully considered but they are not persuasive.
On p. 10 of the 6/23/2026 remarks, Applicant argues that cited art of record fails to teach or suggest “using a GPR model as a generative model to produce additional synthetic data samples that emulate the statistical properties of original training data for the purpose of data augmentation.” However, Zhu teaches GPR to produce additional synthetic data samples (see ¶ 0040 as well as Fig. 3). It should be noted that Zhu’s GPR produces posterior mean and posterior variance for components of the original data, which applies to a broad but reasonable interpretation of “statistical properties.” Therefore, the argument is not persuasive.
In the second full paragraph on p. 11 of the remarks, Applicant argues that “combination of Hershey's aircraft engine RUL concept with Zhu's vital sign interpolation framework does not yield the claimed GIDL approach, as neither reference teaches the generative use of GPR for data augmentation or the joint hyperparameter tuning recited in the amended claims.” In response to applicant's arguments against the references individually, one cannot show nonobviousness by attacking references individually where the rejections are based on combinations of references. See In re Keller, 642 F.2d 413, 208 USPQ 871 (CCPA 1981); In re Merck & Co., 800 F.2d 1091, 231 USPQ 375 (Fed. Cir. 1986). As set forth in the rejection, none of the individual references are relied upon to teach the sum total of each claim limitation. Instead, the references combine to teach the claim limitations. Zhu teaches GPR while newly cited art of record by Puri teaches hyperparameter tuning and Hershey teaches aircraft engine RUL. The rejection is based upon the combination of references.
Applicant’s remaining arguments, see pp. 10-11, filed 6/23/2026, with respect to the rejection(s) of claim(s) 1-18 under 35 USC 103 have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made in view of U.S. Patent Application Publication 20210073671 by Puri et al.
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claims 1-18 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention.
Claim 1 recites: “wherein hyperparameters of the generative GPR model and the deep learning model are jointly tuned based on the RUL prediction obtained from the validation data.” ¶ 0024 of Applicant’s originally filed specification provides that “hyperparameters of both the generative GPR model and deep learning models are tuned during the training and validating process.” Other portions of the specification provide generally similar disclosure. However, no portions of the disclosure appear to provide any indication that the models are “jointly tuned” or that they are “based on the RUL prediction obtained from the validation data.” Applicant has not indicated where this amended limitation finds support in the originally filed disclosure. For the purpose of further examination, the limitation will be interpreted according to supporting disclosure at ¶ 0024 as noted above.
Claims 7 and 13 have been amended with limitations similar to those of claim 1 and are rejected for the same reason indicated above.
Claims 2-6, 8-12 and 14-18 are rejected as including limitations of a rejected base claim.
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.
Claims 1-18 are rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent Application Publication 20220051796 by Zhu et al. ("Zhu") in view of U.S. Patent Application Publication 20180054376 by Hershey et al. ("Hershey") and U.S. Patent Application Publication 20210073671 by Puri et al. (“Puri”).
In regard to claim 1, Zhu discloses:
1. A method for remaining useful life (RUL) prediction of an aircraft engine based on gaussian process regression (GPR) integrated deep learning (GIDL), comprising: See Zhu, Fig. 1, broadly depicting a method for prediction based upon GPR and deep learning.
partitioning observation data into training data, validation data, and testing data; Zhu,¶ 0077, “We split our dataset to 70% for a training set, 15% validation set and 15% test set. We tested our method on approximately 4,000 observation windows.”
training a generative GPR model using the training data to obtain a trained GPR model; Zhu, ¶ 0006, “Thus, a method is provided in which Gaussian process regression is used to generate synthetic vital sign data at regularly spaced intervals, which is provided as input to a recurrent neural network (RNN).” Also ¶ 0051, “Since it is desired to model vital sign data of the entire patient population, log-normal distributions are applied as priors for the three hyperparameters based on clinical judgment. The model is optimized by minimizing the negative log likelihood with respect to the hyperparameters.”
using the trained GPR model as a synthetic data generator to generate synthetic data including additional data samples that emulate statistical properties of original training data; Zhu ¶ 0040, “A Gaussian process model is applied to the continuous variables and used to generate a time series of synthetic vital sign data.” Also ¶ 0042, “The output from the Gaussian process regression 303 and the step function modelling 304 is a posterior mean and a posterior variance for each of the components of the vital sign information processed.”
Zhu does not expressly disclose: performing an averaging process to integrate the synthetic data and the training data to obtain integrated data wherein the averaging process combines the original training data and the GPR-generated synthetic data to oversample a ground truth model …; This is taught by Puri. See Puri, ¶ 0019, “By generating such a combined synthetic training sample, the digital synthetic data system can augment existing training repositories without distorting class distribution within the training data or over-fitting.” ¶ 0078, “For instance, in some embodiments, the digital synthetic data system 106 can utilize the average vector value between the modified training sample values.” ¶ 0083, “Indeed, the digital synthetic data system 106 by combining one or more existing training samples using a combination ratio, the digital synthetic data system 106 can increase the number of available training samples without substantially affecting a class distribution within a set of training samples and increasingly train a machine learning model for accuracy.” Also ¶ 0089, “the digital synthetic data system 106 can utilize both combined synthetic training samples and existing/organic training samples to train a machine learning model.” Note that Puri’s synthetic data generation effectively provides oversampling according to the terms of a broad but reasonable interpretation of the plain language of the “averaging process” which is set forth in the claim. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to Puri’s synthetic data integration with Zhu’s GPR data in order to increase the number of available training samples without substantially affecting a class distribution within a set of training samples and increasingly train a machine learning model for accuracy, as suggested by Puri.
Zhu does not expressly disclose: model … of an aircraft engine degradation process. This is taught by Hershey. See Hershey, ¶ 0001, “For example, it may be helpful to predict a Remaining Useful Life (“RUL”) of an electro-mechanical system, such as an aircraft engine, to help plan when the system should be replaced.” It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Hershey’s aircraft engine RUL with Zhu’s RNN model in order to help plan when a system should be replaced as suggested by Hershey.
generating a plurality of data minibatches from the integrated data; feeding the plurality of data minibatches into a deep learning model to train the deep learning model; Zhu, ¶ 0091, “RNNs All of the RNNs used in step S4 of FIG. 1 were trained for 200 epochs with early stopping using the validation set to avoid overfitting, 50 steps per epoch and a batch size of 50 sequences of the same length.” Also see Puri, ¶ 0089, “For example, the digital synthetic data system 106 can utilize a batch of training data that includes both organic/existing training samples and combined synthetic training samples to train a machine learning model.”
obtaining … prediction from the trained deep learning model based on the validation data; and Zhu, ¶ 0035, “Each EWS may, for example, comprise a binary output indicating whether an observation set of a patient is within 24 hours of a composite outcome …” Also ¶ 0084, “The scaling and shifting operations are obtained through the training set and then applied to the validation and test sets.”
Zhu does not expressly disclose: RUL prediction. This is taught by Hershey. See Hershey, ¶ 0001, “For example, it may be helpful to predict a Remaining Useful Life (“RUL”) of an electro-mechanical system, such as an aircraft engine, to help plan when the system should be replaced.”
using the RUL prediction for further parameter training of … the deep learning model. Zhu, ¶ 0129, “All hyperparameters of the model were optimised empirically using a balanced training and validation set, referred to as DO,1B.”
Zhu does not expressly disclose … further parameter training of the generative GPR model and wherein hyperparameters of the generative GPR model and the deep learning model are jointly tuned based on the RUL prediction obtained from the validation data. This is taught by Puri. See Puri, ¶ 0087-0088, e.g. “In one or more embodiments, the digital synthetic data system 106 utilizes the machine learning model 604 to repeatedly generate predicted values from combined synthetic training samples, compare the predicted values with combined synthetic ground truth labels, and alters parameters of the machine learning model 604 to minimize calculated loss. In some embodiments, the digital synthetic data system 106 repeats this process until a termination condition (e.g., the calculated loss 612 is minimized past a threshold, a threshold number of iterations has been satisfied, or a threshold time has passed) to generate the trained machine learning model 614.” It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use Puri’s repeated training in order to generate a trained model as suggested by Puri.
In regard to claim 2, Zhu also discloses:
2. The method according to claim 1, further including:
obtaining sensing data from sensors Zhu, Fig. 2 and ¶ 0036, “… vital sign information may be provided on an automatic basis by a sensor system 12 …”
Zhu does not expressly disclose: of the aircraft engine; This is taught by Hershey. See Hershey ¶ 0027, “For example, it may be helpful to predict a Remaining Useful Life (“RUL”) of an electro-mechanical system, such as an aircraft engine, to help plan when the system should be replaced.”
Zhu also discloses:
inputting the sensing data into the trained deep learning model to provide RUL prediction of the aircraft engine; and Zhu, Fig. 1, elements S2-S5, depicting event prediction.
Zhu does not expressly disclose:
determining a scheduling strategy for maintenance of the aircraft engine according to the RUL prediction of the aircraft engine, wherein the maintenance of the aircraft engine is performed according to the scheduling strategy. This is taught by Hershey. See ¶ 0001, “For example, it may be helpful to predict a Remaining Useful Life (“RUL”) of an electro-mechanical system, such as an aircraft engine, to help plan when the system should be replaced.” Also ¶ 0029, “A digital twin may estimate a remaining useful life of a twinned physical system using sensors, communications, modeling, history, and computation.” Also ¶ 0079, “The process may bring the system off-line in a scheduled, orderly, and cost-beneficial manner thereby reducing any unscheduled down time.” It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Hershey’s aircraft engine maintenance scheduling with Zhu’s prediction in order to help plan when the system should be replaced as suggested by Hershey.
In regard to claim 3, Zhu also discloses:
3. The method according to claim 1, wherein: the generative GPR model is first trained with initial hyperparameters and further tuned empirically using the training data and the testing data. Zhu, ¶ 0051, “Since it is desired to model vital sign data of the entire patient population, log-normal distributions are applied as priors for the three hyperparameters based on clinical judgment. The model is optimized by minimizing the negative log likelihood with respect to the hyperparameters. The GPR models may be built for example using GPy, which is a GP framework written in python.” Also ¶ 0129, “All hyperparameters of the model were optimised empirically using a balanced training and validation set, …”
In regard to claim 4, Zhu also discloses:
4. The method according to claim 1, wherein: training the generative GPR model using the training data includes obtaining a posterior distribution based on standard Bayesian update. Zhu, ¶ 0042, “In some embodiments, the generation of the EWS in step S4 uses the posterior variances generated by the pre-processing of step S3 in addition to the posterior means generated by the pre-processing of step S3. Thus, the mean and variance of each component of the vital sign information generated by the Gaussian process model at each time point tin the assessment window may be used as input to step S4.” Not that posterior mean and variance are derived from a posterior distribution.
In regard to claim 5, Zhu also discloses:
5. The method according to claim 4, after obtaining the posterior distribution, further including: sampling data from the posterior distribution. See Zhu, ¶ 0042, as cited above. Sampling from a posterior distribution is inherent in calculation of posterior mean and variance.
In regard to claim 6, Zhu and Hershey also teach:
6. The method according to claim 1, wherein: RUL is calculated as a first passage time when a health status value of the aircraft engine exceeds a predefined failure threshold. Hershey ¶ 0029, “It may provide an answer in a time frame that is useful, that is, meaningfully prior to a projected occurrence of a failure event or suboptimal operation.” Also ¶ 0053 “If differences between the sensor values at time=t and the UPM predictions fall outside of the tolerance envelopes, then a report issues at 360.” Also ¶ 0107, “predicting RUL or the time to failure of a twinned physical system.”
In regard to claim 7, Zhu discloses:
7. A system, comprising: a memory, configured to store program instructions for performing a method for remaining useful life (RUL) prediction of an aircraft engine based on gaussian process regression (GPR) integrated deep learning (GIDL); and a processor, coupled with the memory and, when executing the program instructions, configured for: Zhu, ¶ 0034, “The one or more computer programs may be provided in the form of media or data carriers, optionally non-transitory media, storing computer readable instructions. When the computer readable instructions are read by the computer, the computer performs the required method steps.” Also Fig. 2 and ¶ 0037, “In the schematic configuration of FIG. 2, the vital sign data is received by a data receiving unit 8 of the data processing apparatus 5. The data processing apparatus 5 may further comprise a processor 10 configured to carry out steps of the method.”
All further limitations of claim 7 have been addressed in the above rejection of claim 1.
In regard to claims 8-12, parent claim 7 is addressed above.
All further limitations of claims 8-12 have been addressed in the above rejections of claims 2-6, respectively.
In regard to claim 13, Zhu discloses:
13. A non-transitory computer-readable storage medium, containing program instructions for, when being executed by a processor, performing a method for remaining useful life (RUL) prediction of an aircraft engine based on gaussian process regression (GPR) integrated deep learning (GIDL), the method comprising: See Zhu, ¶ 0034, “The one or more computer programs may be provided in the form of media or data carriers, optionally non-transitory media, storing computer readable instructions. When the computer readable instructions are read by the computer, the computer performs the required method steps.” Also Fig. 2 and ¶ 0037, “In the schematic configuration of FIG. 2, the vital sign data is received by a data receiving unit 8 of the data processing apparatus 5. The data processing apparatus 5 may further comprise a processor 10 configured to carry out steps of the method.”
All further limitations of claim 13 have been addressed in the above rejection of claim 1.
In regard to claims 14-18, parent claim 13 is addressed above.
All further limitations of claims 14-18 have been addressed in the above rejections of claims 2-6, respectively.
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 James D Rutten whose telephone number is (571)272-3703. The examiner can normally be reached M-F 9:00-5:30 ET.
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, Li B Zhen can be reached at (571)272-3768. 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.
/James D. Rutten/Primary Examiner, Art Unit 2121
Prosecution Insights