DETAILED ACTIONS
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 .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on
08 /22 /2025 has been entered.
Response to Amendment
This office action is in response to the amendments/arguments submitted by the Applicant(s) on 08/22/2025.
Status of the Claims
Claims 1, 3, 5-8, 10, 13-14 are pending.
Claims 1 is amended.
Claim 2 is canceled.
Response to Arguments
Rejections Under 35 U.S.C. §102(a)(1):
Applicant's argument, see pages 5-7 of the remark filed on 08/22/2025 with respect to the rejection(s) of Claims under 35 U.S.C. § 102(a)(1) have been fully considered, and are moot. New rejections with a newfound prior art have been set forth below.
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, 3, 5-8, 10, 13-14 are rejected under 35 U.S.C. 103 as being unpatentable over by Hussain Abbas. (US 2018/0300639 A1, hereinafter Abbas) and in view of Nicolas Pipard (WO 2018/060142 A1, hereinafter Pipard). evidence from Qi, Y. (2012). “Random Forest for Bioinformatics”. In: Zhang, C., Ma, Y. (eds) Ensemble Machine Learning. Springer, New York, NY.
Regarding claim 1, Abbas teaches, A computer-implemented method (Abbas, Figure 1, [0007] In some embodiments, a computing system and/or computer-implemented method is disclosed for predicting pipe leaks) comprising:
a second step of extracting a sample of pipe; (Abbas, Figure 2, [0029], In some embodiments, the pipe leak prediction system may utilize various machine learning algorithms. In particular, the pipe leak prediction system may use supervised machine learning techniques, such that existing data on known pipe leaks is used to train a predictive model. Examples of such supervised machine learning techniques include-but are not limited to (..) Random forests, ensembles of classifiers, ordinal classification, data pre-processing, and statistical relational learning. [0040], The training dataset 220 may be a subset of a larger dataset, such as pipe database 108, containing data for a variety of pipes. The larger dataset may be randomly split into the training dataset 220 used to train the predictive model, as well as a validation dataset used to validate the results of the trained predictive model” NOTE: prior art Abbas teaches the use of supervised machine learning random forest method and data pre-processing and statistical relational learning. Which implies that data has been classified into groups and based on certain properties and statistical relations data classified. Training dataset is the small group/ subgroup of homogeneous pipes that leaks (see fig. 4). And 108 are database with all types of pipes. the training data/sample data is smaller group out of main database 108. The sample training data is grouped based on pipes has leaks (homogeneous subgroup, see figure 4). A small number of pipes with leaks will be enough for training predictive model. If the sample data is homogeneous the number of training data required will be small, and if the sample is larger and inhomogeneous, it will require a larger number of data to train the predictive model. Abbas teaches the use of Random Forest tree algorithm and data preprocessing and statistical relational learning. It is evident that for Random Forest three main hyperparameters need to be set before training. These include node size, the number of trees, and the number of features sampled. From there, the random forest classifier can be used to solve for regression or classification problems. Thus, considering training data set size based on the correlation between sample homogeneity is part of data preprocessing. Please see reference evidence above section 2.2. (Qi, Y. (2012). “Random Forest for Bioinformatics”. In: Zhang, C., Ma, Y. (eds) Ensemble Machine Learning. Springer, New York, NY).
a third step of obtaining, for each pipe section of the sample, one or more pipe condition scores determined by a condition assessment procedure, (Abbas, [0007] “In some embodiments, a computing system and/or computer-implemented method is disclosed for predicting pipe leaks. The method may include accessing a training dataset including first data items and known leaks associated with respective pipes of a first plurality of pipes, wherein the first data items include characteristics of the respective pipes”); and
a fourth step of performing an estimation of the one or more pipe condition scores for a pipe-sections that does not belong to the sample, the estimation being based on the pipe parameters of the pipe section that does not belong to the sample (Abbas, Figure 1 [0038], The pipe leak prediction system 110 may also be able to make predictions on pipes outside of pipe database 108, such as for newly installed pipes that have not yet been introduced to the pipe database 108. For example, after the pipe installer 102 installs a pipe (or even immediately prior to installation) and provides the input information associated with that pipe to the pipe leak prediction system 110, the system may be able to predict or estimate whether that pipe is likely to fail, when the pipe is likely to fail, the probability that the pipe will likely fail, and so forth),
the estimation being parameterized with the one or more(Abbas, A supervised machine learning technique may be applied to this training dataset to generate a predictive model configured to determine a leak prediction of a pipe by training the predictive model based on the first data items associated with respective pipes of the first plurality of pipes),
wherein performing an estimation of the one or more pipe condition scores for the pipe sections that does not belong to the sample (Abbas, Figure 1 [0038], The pipe leak prediction system 110 may also be able to make predictions on pipes outside of pipe database 108) comprises:
training a supervised machine learning engine that predicts pipe condition scores based on pipe parameters, by using the one or more pipe condition scores and the pipe parameters of the pipe sections that belongs to the sample (Abbas, Figure 2, Figure 12 [0040], “The training dataset 220 may contain information regarding the characteristics of various pipes”, “the training dataset 220 may also contain information regarding which of those pipes are known to have leaks. Thus, the predictive model can use the data in the training dataset 220 in order to determine the patterns associated with pipes without leaks and the patterns associated with pipes with leaks. The model may be able to learn and use those patterns to predict leaks in pipes when provided with the characteristics of those pipes. The predictive model may be generated using various machine learning algorithms”) and
using supervised machine learning engine to predict the one or more pipe condition scores of the pipe section that does not belong to the sample, based on the pipe parameters of the pipe sections that does not belong to the sample. (Abbas, figure 2, [0042], [0042] At block 212, once the predictive model has been validated, the predictive model may then be applied to new cohorts (e.g., pipes not in the training or validation datasets). Figure12, [0111], Pipe Leak Prediction System 1220 can access data from a Pipe Database 1234, which contains pipe characteristic data for pipes existing in the field and knowledge (e.g., known leaks) of which of those pipes have leaked in the past. This data is instrumental in building a model capable of predicting leaks in pipes. In addition, the Pipe Leak Prediction System 1220 may access Pipeline Data 1230, which may include new or updated pipe characteristics data for pipes in the field. The pipe characteristic data may be new in the case of newly installed pipes, and in some cases that new pipe characteristic data may be manually entered into the Pipe Leak Prediction System 1220. Leak Prediction System 1220 may apply the predictive model to the data in the Pipeline Data 1230 in order to output leak predictions about the pipes in the Pipeline Data 1230).
And wherein the method further comprises raising an alert for pipe sections whose pipe condition scores match an alert condition and each alert for a pipe section automatically triggers at least one action chosen from a group comprising: (Abbas, Figure 12, [0115] Once the Pipe Leak Prediction System 1220 has identified the pipes that are most likely to leak or are likely to have already leaked, members of the Executive Team 1290 may be able to view reports on those identified pipes through an interface 1282 on a device 1280. The device 1280 may receive those reports generated from the Pipe Leak Prediction System 1220).
a further condition assessment procedure of the pipe section, a safeguard measure, and a repair of the pipe section. (Abbas, Figure 12, [0115], “The device 1280 may receive those reports generated from the Pipe Leak Prediction System 1220. Furthermore, the members of the Executive Team 1290 may be able to pull up and view the pipe leak predictions for any given pipe (e.g., based on an ID associated with the pipe), as well as a breakdown of the calculations used to generate those leak predictions. Thus, the members of the Executive Team 1290 may be able to further verify and confirm any leak predictions. That member could then directly indicate within the interface 1282 to instruct Field Staff 1210 to physically go to the pipe's location to inspect the pipe for any leaks. [0116] The Device 1280 will then send the instruction to the Pipe Leak Prediction System 1220, which will then determine the closest Field Staff 1210 (e.g., geographically) to the pipe's location”.)
Abbas teaches extracting pipe parameters and grouping the pipes based on parameters. (Abbas, Figure 3, [0035], FIG. 3. The pipe integration system 106 may store this data for the pipe in a pipe database 108, which may contain data for various pipes installed by the utility company (e.g., pipes 120-1 through 120-N)” pipes 120-1 to 120-n represent different types of pipes or pipe categories/ clusters of pipes based on characteristics and other pipe parameter)
Abbas is silent on a first step of clustering into a number of classes, and, said number of classes being a predefined number of classes, and the first step further comprises applying a Gaussian Mixture Model (GMM) to the pipes for clustering the pipe sections into said predefined number of classes; and, for each class of said number of classes:
However, Pipard teaches a first step of clustering into a number of classes, and, said number of classes being a predefined number of classes, and the first step further comprises applying a Gaussian Mixture Model (GMM) to the pipes for clustering the pipe sections into said predefined number of classes; and, for each class of said number of classes:(Pipard, Page 2, [0035], “Determining similarities and dissimilarities between the detected objects; Clustering the objects into groups based on the determined similarities of the objects”. Lines 27-28; page 3, lines 8-17, Preferably the classification of the digital images is based on a learning algorithm. The program used to implement the classification method is using a learning algorithm of the type supervised learning (when the training examples of the clusters have known response labels). Such learning algorithms can be for example a "decision tree learning", a "support vector machine" or an "artificial neural network". It is preferred when the clustering is based on a learning algorithm of the type unsupervised learning (when the training examples in this case have no response labels). Such learning algorithms can be for example "density-based" (DBSCAN), model-based ("Gaussian mixture models") or "grid-based" (STING or CLIQUE) which are clustering methods”. NOTE: Pipard discloses the first step of clustering the samples using Gaussian mixture into groups with similarities of objects and using a decision learning algorithm when the training of the clusters has known response label. Pipard clustering method is pertinent to the clustering method of the claim.):
It would have been obvious to a person having ordinary skill in the art before the effective filing date to modify Abbas’s method in view of Pipard’ s method to incorporate clustering pipes into groups using the Gaussian mixture models as taught by Pipard and obtain the clustering of similar objects into a single group which adequately represents the set of individual objects. Therefore, this resolves the problems created by the imbalance between detections of the different types of defects, an issue very common with current inspection systems (Pipard, Page 2, lines 21-24). It would have been obvious to a person of ordinary skill to include the well-known Gaussian Mixture model along with the other machine learning network, in order to yield the predicted results of clustering and classification of samples, yet with higher accuracy (KSR).
Regarding claim 3, combination of Abbas and Pipard teaches the computer-implemented method of claim 1,
Abbas further teaches, wherein the second step of extracting the sample comprises Abbas, Figure 2, lines 1-4, , The training dataset 220 may be a subset of a larger dataset, such as pipe database 108, containing data for a variety of pipes):
a fifth step of initializing a set of candidate samples of pipe sections (Abbas; [0007], “A supervised machine learning technique may be applied to this training dataset to generate a predictive model configured to determine a leak prediction of a
pipe by training the predictive model based on the first data items associated with respective pipes of the first plurality of pipes”. Figure 2, [0040] “At block 202, a predictive model may be generated using data for pipes contained within a training dataset 220. The training dataset 220 may be a subset of a larger dataset,
such as pipe database 108, containing data for a variety of pipes”).
a sixth step of iteratively modifying said set of candidate samples using:
a genetic algorithm based on an objective function comprising a minimization of the difference of average pipe parameters of the pipe sections of the sample, and the average pipe parameters of the pipe sections of the class;
a seventh step of selecting the candidate sample that optimizes said objective function. (Abbas; [0007], “Once the predictive model has been generated, the method may further include accessing a validation dataset including second data items and known leaks associated with respective pipes of a second plurality of pipes, wherein the second data items include characteristics of the respective
pipes”. Figure 2, [0041], 0041] At block 204, once the predictive model has been generated the model can be validated. The validation of the predictive model may involve one or more sub-blocks, such as sub-blocks 206, 208, and 210. At block 206, the predictive model can be used to predict pipe leaks using pipe data from the validation dataset).
Regarding claim 5, combination of Abbas and Pipard teaches the computer-implemented method of claim 1, Abbas further teaches wherein the condition assessment procedure of the third step is chosen in a group comprising one or more of:
an analysis of an electromagnetic flux applied to the pipe section; an acoustical analysis of the pipe section; the extraction, and analysis in a laboratory of a sample of the pipe section; and wherein each of the condition assessment procedure provides pipe condition scores at the same scale (Abbas, Figure 15, I/O subsystem 1508 [0162] I/O subsystem 1508 may include user interface input devices and user interface output devices. [0163] User interface input devices may also include, without limitation, “Additionally, user interface input devices may include, for example, medical imaging input devices such as computed tomography, magnetic resonance imaging, position emission tomography, medical ultrasonography devices. User interface input devices may also include, for example, audio input devices such as MIDI keyboards, digital musical instruments and the like).
Regarding claim 6, combination of Abbas and Pipard teaches the computer-implemented method of claim 5, Abbas further teaches wherein the condition assessment procedures provide two or more pipe condition scores corresponding to different parts of pipe sections and chosen in a group comprising: an inner coating condition score; an outer coating condition score; a joint condition score.(Abbas, Figure 3, 0047] In some embodiments, the data may include column 308 (shown with the header "pipe_diam") which signifies a numeric variable indicating how wide the pipe is. Any measuring unit can be specified in advance and used. In the figure, the numbers of column 308 may represent the width of each pipe in inches. 0053] In some embodiments, the data may include column 322 (shown with the header "soil_res") which signifies a numeric variable for the resistivity of the soil in which a pipe is installed. Higher values of soil resistivity may be associated with less corrosiveness. For example, column 322 shows 1138 for the soil resistivity of the first pipe and 3601 for the soil resistivity of the second pipe which may indicate that the second pipe is located in less-corrosive soil. Knowing the relative corrosiveness of the soil that each pipe is located in may allow the pipe leak prediction system to determine the impact of soil corrosiveness on the likelihood of a pipe leaking).
Regarding claim 7, combination of Abbas and Pipard teaches the computer-implemented method of claim 6, Abbas further teaches wherein a single pipe condition score is obtained from the two or more pipe condition scores corresponding to different parts of pipe sections, using a weighted or orthogonal sum. (Abbas, Figure 3, [0066] In some embodiments, the random forest model may be used to predict whether a pipe will leak based on a combination of 11 factors, such as the factors or pipe attributes shown in FIG. 3).
Regarding claim 8, Abbas teaches the computer-implemented method of claim 5, Abbas further teaches wherein the one or more pipe condition scores are associated with one or more reliability indexes. (Abbas, Figure 3, Abbas, Figure 3, 0066] In some embodiments, the random forest model may be used to predict whether a pipe will leak based on a combination of 11 factors, such as the factors or pipe attributes shown in FIG. 3).
Regarding claim 10, combination of Abbas and Pipard teaches the computer-implemented method of claim 1, Abbas further teaches wherein said supervised machine learning engine is a random forest machine learning engine. (Abbas, Figure 5, [0032] In some embodiments, the pipe leak prediction system may utilize a combination of different predictive
models or machine learning techniques [0033] In the present disclosure, embodiments of a pipe leak prediction system are disclosed that use random forest).
Regarding claim 13, combination of Abbas and Pipard teaches the computer-implemented method of claim 1, Abbas further teaches A computer program product, stored on a non-transitory computer-readable medium, said computer program product comprising code instructions for executing a method according to claim 1. (Abbas, Figure 15, 0169] Computer-readable storage media 1522 containing code, or portions of code, can also include any appropriate media known or used in the art, including storage media and communication media, such as but not limited to, volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage and/or transmission of information. This can include tangible, non-transitory computer-readable storage media such as RAM, ROM, electronically erasable programmable ROM (EEPROM), flash memory or other memory technology).
Regarding claim 14, combination of Abbas and Pipard teaches the computer-implemented method of claim 1, Abbas further teaches A device comprising a processor configured to execute a method according to claim. (Abbas, Figure 15, [0161] In various embodiments, processing unit 1504 can execute a variety of programs in response to program code and can maintain multiple concurrently executing programs or processes. At any given time, some or all of the program code to be executed can be resident in processor(s) 1504 and/or in storage subsystem 1518. Through suitable programming, processor(s) 1504 can provide various functionalities described above. Computer system 1500 may additionally include a processing acceleration unit 1506, which can include a digital signal processor (DSP), a special purpose processor, and/or the like.).
Conclusion
Citation of Pertinent Prior Art
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
IDE et al. ( US 2018/0095004 A1) recites “A method for detecting early indications of equipment failure in an industrial system includes receiving sensor training data collected from industrial equipment under normal conditions and identifying periods of time in the sensor training data when the equipment was functioning normally; finding a pattern for each identified period of time to initialize a plurality of mixture models; learning weighting factors, mean and variance of each of the plurality of mixture models, and removing unimportant models from the plurality of mixture models; detern1ining a Gaussian Markov random field model from surviving mixture models by calculating gating functions for each of the variables and individual mixture models; determining a threshold value of an anomaly score for each variable from the sensor training data; and deploying the model to monitor sensor data from industrial equipment using the threshold values to detect anomalous sensor data values indicative of an impending system failure” (Abstract).
Sina et al. “A Gaussian mixture model-based discretization algorithm for associative classification of medical data”, Expert System with Applications, Volume 58, 1 October 2016, Pages 119-129.
Knowledge-based systems such as expert systems are of particular interest in medical applications as extracted if-then rules can provide interpretable results. Various rule induction algorithms have been proposed to effectively extract knowledge from data, and they can be combined with classification methods to form rule-based classifiers. However, most of the rule-based classifiers cannot directly handle numerical data such as blood pressure. A data processing step called discretization is required to convert such numerical data into a categorical format. Existing discretization algorithms do not take into account the multimodal class densities of numerical variables in datasets, which may degrade the performance of rule-based classifiers. In this paper, a new Gaussian Mixture Model based Discretization Algorithm (GMBD) is proposed that preserve the most frequent patterns of the original dataset by taking into account the multimodal distribution of the numerical variables. The effectiveness of GMBD algorithm was verified using six publicly available medical datasets. According to the experimental results, the GMBD algorithm outperformed five other static discretization methods in terms of the number of generated rules and classification accuracy in the associative classification algorithm. Consequently, our proposed approach has a potential to enhance the performance of rule-based classifiers used in clinical expert systems” (Abstract)
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DILARA SULTANA whose telephone number is (571)272-3861. The examiner can normally be reached Mon-Fri, 9 AM-5:30 PM.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s
supervisor, EMAN ALKAFAWI can be reached on (571) 272-4448. 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.
/DILARA SULTANA/Examiner, Art Unit 2858
/EMAN A ALKAFAWI/Supervisory Patent Examiner, Art Unit 2858 10/23/2025