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 .
DETAILED ACTION
This Office Action is sent in response to Application’s Communication received on 11/30/2023 for application number 18/524989. The Office hereby acknowledges receipt of the following and placed of record in file: Specification, Drawing, Abstract, Oath/Declaration, and Claims.
Claims (1-17), 18 and 19 are presented for examination.
Information Disclosure Statement
The information disclosure statements (IDS) submitted on 11/25/2025 were filed prior to current Office Action. The submission is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Claim Objections
Claims 4 and 6 are objected to because of the following informalities: the claims are using colon instead of semi colon. 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 therefore, subject to the conditions and requirements of this title.
Claims (1-17), 18 and 19 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
Step 1: Claims (1-17), 18 and 19 are drawn to a method each of which is within the four statutory categories (e.g., a process, a machine).
Step 2A - Prong One: In prong one of step 2A, the claims are analyzed to evaluate whether they recite a judicial exception.
Claim 1.
A method of generating a predictive model based on a classification of patterns of time-series data to which a pre-processing pipeline has been applied, the method being performed by an electronic device and comprising:
receiving single time-series data or multiple types of time-series data that are collected in a specific domain;
drawing time-series data corresponding to an identical domain and an identical time interval, among the received single time-series data or multiple types of timeseries data;
pre-processing the drawn time-series data by applying a pre-processing pipeline built by applying at least one pre-processing module to the drawn time-series data;
generating a pattern classification model for classifying patterns of the preprocessed time-series data based on a clustering of the pre-processed time-series data;
generating a predictive model for predicting feature information of the drawn time-series data based on a cluster that is generated as results of the clustering of the pre-processed time-series data; and
storing the pattern classification model and the predictive model.
Claim 1 recites an abstract idea including mathematical concepts and mental process. Specifically, the limitations of clustering pre-processed times-series data, generating a pattern classification model based on the clustering, and generating a predictive model based on a result cluster recite mathematical relationships and mathematical calculations because the limitations analyze relationships among numerical time-series data to group the data and derive classification and predictive relationships therefrom. Mathematical relationships and mathematical calculations are mathematical concepts identified as abstract ideas under MPEP $2106.04(A)(2)(I).
The claim further recites receiving time series information, selecting information corresponding to an identical domain and time interval, classifying patterns in the information, and predicting feature information based upon the classification. At the level of generality claimed, these limitations encompass obtaining information and evaluating the information according to specified criteria to identify patterns/categories and derive a prediction, which are observations and evaluations that can practically be performed in the human mind. Accordingly, these limitations additionally recite a mental process under MPEP $2106.04(a)(2)(III).
Step 2A Prong 2:
Claim 1 recites “performed by an electronic device”, “receiving the time-series”, “the pre-processing pipeline and at least one pre-processing module” and “storing the pattern classification model and predictive model”. “ performed by an electronic device” limitation doe appear to integrate the abstract idea into a practical application. The claim merely requires the mathematical/data analysis operations to be carried out using an unspecified “electronic device”. There is no processor architecture, sensor system, specialized computing device, or other machine configuration recited. In other words, the electronic device appears to function as the tool on which the abstract data-analysis operations are performed (electronic device -> receives data ->processes data -> performs clustering/classification/prediction -> stores results. The MPEP specifically identifies merely using a computer as a tool to perform an abstract idea as an indication that the exception has not been integrated into a practical application. The claim also does not appear to require the electronic device to perform differently because of the generated predictive model, for example, it does not require: generate prediction -> use prediction to control a machine/process ->alter operation of the device. Instead, it stops after generating and storing information models. Therefore, the electronic device appears to provide merely the technological environment in which the abstract idea is executed, rather than a meaningful application of the abstract idea.
The limitation of “receiving the time-series data” does not provide practical integration. The received information serves as the input to the subsequent mathematical/data analysis operations (receive data -> analyze data). Nothing in the claim appears to require a particular sensor, measurement apparatus, physical monitoring system, or specialized data-acquisition technology to generate data. Thus, receiving the data appears to constitute insignificant extra-solution activity necessary to supply information to the abstract idea analysis, rather than a practical application of the analysis.
The limitation of “the pre-processing…” is cited at high level of generality and it does not appear to identify what technological improvement the pipeline achieves. Instead, the preprocessing pipeline appears to prepare the data for the subsequent: clustering ->classification ->prediction. The pre-processing merely makes the information suitable for the mathematical analysis that follows, then it is closely tied to carrying out the abstract idea rather than applying the abstract idea to improve another technology.
The limitation of “specific domain” does not appear sufficient. The claim limits the received time-series data to data” “collected in a specific domain” and selects data corresponding to” “an identical domain and an identical time interval”. But the claim does not appear to recite concrete technological application of the resulting prediction. Thus, limiting the mathematical analysis to time-series data from a particular “domain” does not by itself provide practical integration.
The limitation of storing the models: “storing the pattern classification model and the predictive model” do not provide practical integration. The claim already performed the clustering, generated the classification model, and generated the predictive model, it then merely stores the resulting information. That is essentially post-solution activity. The MPEP explains that nominal activities occurring before or after the judicial exception, such as data gathering or merely outputting results may constitute extra-solution activity. Here, storage does not appear to cause anything else technologically significant to happen.
The additional elements, individually and in combination, do not integrate the recited abstract idea into a practical application. The recitation that the method is performed by an “electronic device” merely uses a generic computing device as a tool for performing mathematical/data-analysis operations. Receiving the time-series data constitutes data gathering activity that supplies information for the subsequent analysis, while storing the resulting classification and predictive models constitutes post-solution activity. Although the claim additionally recites preprocessing the selected time-series data using a preprocessing module, the claim does not recite a specific improvement to the functioning of the electronic device or another technology or technical field, rather, the preprocessing prepares the information for the subsequently recited clustering, classification and predictive analysis. Further, limiting the information to a particular domain and time interval merely limits the abstract idea to a field of use. Accordingly, the additional elements considered individually and in combination do not integrate the judicial exception into a practical application. Therefore, claim 1 is directed to judicial exception under step 2A.
Dependent claims (2-17) fail to include any additional elements. In other words, each of the limitations/elements recited in respective dependent claims (2-17) are further part of the abstract idea as identified by the Examiner for each respective dependent claim (i.e. they are part of the abstract idea recited in each respective claim).
The Examiner has therefore determined that the elements, or combination of additional elements, do not integrate the abstract idea into a practical application. Accordingly, the claims are directed to an abstract idea.
Step 2B: The claim does not provide an inventive concept (significantly more than the abstract idea). The claim is ineligible.
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. The recited electronic device merely provides generic computer implementation of abstract mathematical/data-analysis operations. Receiving the time-series amounts to data-gathering activity necessary to perform the recited analysis and storing the resulting pattern-classification model and predictive model amounts to post-solution storage of the result of the analysis. The recited preprocessing pipeline, preprocessing module likewise do not, either individually or in combination with other additional elements, provide an improvement to the functioning the electronic device, or another technology or technical field, a particular machine integral to claimed method, a transformation of particular article into different state or thing, or meaningful limitation amounting to an inventive concept. Rather, the additional elements merely implement the recited abstract data-analysis operations using computer functionality. Considered individually and as an ordered combination, the additional elements do not amount to significantly more than the recited judicial exception. Accordingly, the claim 1 does not recite patent-eligible subject matter under 35 U.S.C $101.
Dependent claims (2-17) fail to include any additional elements. In other words, each of the limitations/elements recited in respective dependent claims (2-17) are further part of the abstract idea as identified by the Examiner for each respective dependent claim (i.e. they are part of the abstract idea recited in each respective claim).
The Examiner has therefore determined that the elements, or combination of additional elements, do not integrate the abstract idea into a practical application. Accordingly, the claims are directed to an abstract idea.
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 of this title, 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 1-19 are rejected under AIA 35 U.S.C. 103(a) as being unpatentable over Jain et al. US 20250209322 A1 (hereinafter Jain) in view of Garvey et al. Foreign Patent Application publication CN 108292296 A (hereinafter Garvey).
Regarding claim 1, Jain teaches A method of generating a predictive model based on a classification of patterns of time-series data to which a pre-processing pipeline has been applied, the method being performed by an electronic device and comprising ([0064], [0082], [0097], [0115], [0125] wherein Jain allows visualizing time-series developing model as a prediction model for preprocessing data through pipelines and classify data based on patterns of time-series) receiving single time-series data or multiple types of time-series data that are collected in a specific domain ([0039-0041], [0165-0166] wherein Jain receives time-series for specific domain) drawing time-series data corresponding to an identical domain and an identical time interval, among the received single time-series data or multiple types of time-series data ([0039-0040] wherein Jain processes time-series in a continuous sequence of data points collected or recorded at successive points in time, at uniform intervals) pre-processing the drawn time-series data by applying a pre-processing pipeline built by applying at least one pre-processing module to the drawn time-series data ([0054], [0058], [0084-0085], [0093-0098] wherein Jain displays a visual representation of time-series data and preprocess it applying pipelines) generating a pattern classification model for classifying patterns of the pre-processed time-series data based on a clustering of the pre-processed time-series data ([0009], [0064], [0067-0068] wherein Jain incorporates a vector-like by extracting data features and producing patterns of similar domain) generating a predictive model for predicting feature information of the drawn time-series data based on a cluster that is generated as results of the clustering of the pre-processed time-series data ([0064], [0082], [0115], [0125], [0172-0173] wherein Jain describes developing model for prediction and evaluating dataset wherein the predictions are obtained by using the model. Jain describes generating a classifier that includes a classification method that utilizes feature similarity to analyze how closely out-of-sample-features resemble training data to classify input data to one or more clusters and/or categories of features as represented in training data).
Jain teaches and storing…the predictive model ([0128], [0139-0140], [0165]).
Jain does not teach storing the pattern classification model.
However in analogous art of generating predictive model based on classification of patterns of time-series, Garvey teaches storing the pattern classification model (page. 10, ¶ 5-6, page. 19, ¶ 2, page. 26, ¶ 3, wherein Garvey teaches saving classification of patterns).
It would have been obvious to a person in the ordinary skill in the art before the effective filing date of the claimed invention to combine Jain with Garvey by incorporating the method of a storing the pattern classification model of Garvey into the method of generating a predictive model for predicting feature information of the drawn time-series data based on a cluster that is generated as results of the clustering of the pre-processed time-series data of Jain for the purpose of incorporating techniques to predict algorithm that considers the trend and seasonal time series data in order to make the prediction relating to the future value. (Garvey: page. 2, ¶ 5).
Regarding claim 2, Jain as modified by Garvey teach wherein the drawing of the time-series data corresponding to the identical domain and the identical time interval, among the received single time-series data or multiple types of time-series data comprises drawing the time-series data that correspond to an identical domain and that are included in the identical time interval, among time-series data at a plurality of different places, which have been previously collected in the specific domain ([0038], [0083], [0089], [0090], [0116], [0134], [0165], wherein Jain receives time-series of multiple types, display visualization of the time-series based on the domain and intervals and using previous collected data).
Regarding claim 3, Jain as modified by Garvey teach wherein the drawing of the time-series data corresponding to the identical domain and the identical time interval, among the received single time-series data or multiple types of time-series data comprises drawing the time-series data by combining conditions for a plurality of time intervals, among time intervals corresponding to time-series data that have been previously collected in the specific domain ([0061], [0183] wherein Jain describes performing a lazy-learning process and/or protocol, which may alternatively be referred to as a “lazy loading” or “call-when-needed” process and/or protocol, may be a process whereby machine learning is conducted upon receipt of an input to be converted to an output, by combining the input and training set to derive the algorithm to be used to produce the output on demand).
Regarding claim 4, Jain as modified by Garvey teach wherein the pre-processing of the time-series data comprises performing at least one of: pre-processing for purifying the drawn time-series data based on a pre-determined reference period ([0076], [0091] wherein Jain preprocesses the time-series based on segments and each time-series segment may be immediately passed to interfaced machine learning models) pre-processing for processing time-series data or each time-series data included in a time interval having an abnormal value, among the drawn time-series data, as a loss value ([0043], [0065], [0067], [0079] wherein Jain preprocesses time-series based on abnormal values), ([0039-0040] wherein Jain processes time-series in a continuous sequence of data points collected or recorded at successive points in time, at uniform intervals) pre-processing for performing exclusion processing on time-series data included in a time interval in which time-series data having a preset first threshold or more have been lost and performing recovery processing on time-series data included in a time interval in which time-series data less than a second threshold smaller than the first threshold have been lost by supplementing the time-series data ([0095], [0097], [0115], [0177] wherein Jain filters segments of the time series based on a quality threshold performing categorization using threshold, wherein Jain adds labels that represents specific domain with signal lengths less than a predetermined threshold are expanded to meet this requirement).
Regarding claim 5, Jain as modified by Garvey teach segmenting an entire time interval of the time-series data into a plurality of time intervals; building the pre-processing pipeline by applying at least one of an identical type of pre-processing module and different types of pre-processing modules to a time interval to be pre-processed, among the segmented time intervals; and pre-processing the time-series data by operating the pre-processing pipeline (Abstract, [0009-0010], [0065], [0084], [0093], [0095], [0126], [0144] wherein Jain describes the model pipeline for receiving the preprocesses data, wherein the time-series segments are annotated by domain experts using labeling module that may be ground truth labels. Ground truth labels may be added to existing labeled training data. The expanded dataset may be processed by converting plurality of annotated time-series segments to canonical data format as described above. Such conversion may involve normalizing signal amplitudes, segmenting or augmenting the data into fixed-length segment, removing artifacts, reducing noise, filtering segments based on a quality threshold, and the like. Processors may load expanded dataset into training pipelines. At least a classifier may then be trained on the updated dataset. During retraining, at least a classifier may map input time-series segments to corresponding labels provided by the experts. In some cases, performance may be again evaluated after retraining. As more time-series data e.g., patient data is collected, additional time-series segments may be identified that were not previously labeled. At least a classifier may be continuously retrained to include new signal features improving model's accuracy and robustness over time).
Regarding claim 6, Jain as modified by Garvey teach wherein the building of the pre-processing pipeline by applying the at least one of the identical type of pre-processing module and the different types of pre-processing modules to the time interval to be pre-processed, among the segmented time intervals, comprises, in response to the time-series data being the multiple types of time-series data, applying a first pre-processing module to a first pre-processing target time interval of first and second single time-series data that constitute the multiple types of time-series data; and applying a second pre-processing module to a second pre-processing target time interval subsequent to the first pre-processing target time interval of the first and second single time-series data (Abstract, [0010], [0038], [0061], [0076], [0090], [0100], [0163-0164] wherein Jain uses the at least a classifier, one or more segment identifications at each time-series segment of a plurality of time-series segments subsequently identified using the detector based on continuous time-series data. Wherein Jain describes the steps for a specific domain, wherein the steps include processing time series data that may be done using one or more digital filtering techniques to enhance the quality of the recorded signals. Applying “digital filtering techniques” that are methods used to manipulate and refine digital signals to remove unwanted components. This may include removing noise or artifacts, while preserving the essential features of the signal. Digital filtering techniques may ensure that the time-series data is accurate and reliable for any subsequent processing, segmentation, and analysis. A digital filtering technique may include the application of mathematical algorithms to the raw data to isolate and remove specific frequency components. For example, low-pass filters may allow signals with frequencies below a certain threshold to pass through while attenuating higher frequency noise. Conversely, high-pass filters permit high-frequency signals to pass while reducing the impact of lower frequency interference. Band-pass filters combine these principles to isolate a specific range of frequencies, which is particularly useful for focusing on the relevant portions of the cardiac signal. Examples of digital filtering techniques may include the use of Finite Impulse Response (FIR) filters, which apply a finite sequence of weights to the signal. FIR filters are known for their stability and linear phase response, making them ideal for applications requiring precise timing, such as ECG and IEGM data analysis).
Regarding claim 7, Jain as modified by Garvey teach wherein the building of the pre-processing pipeline by applying the at least one of the identical type of pre-processing module and the different types of pre-processing modules to the time interval to be pre-processed, among the segmented time intervals, comprises, in response to the time-series data being the multiple types of time-series data ([0039-0043], [0054], [0065-0067], [0077], [0085], [0092-0093] wherein Jain Captures and filters the features of plurality of time series, wherein the features includes for example, morphological features (e.g., the shape of P wave, QRS complex, T wave in an ECG, amplitude of the wave, and duration of the wave), temporal features (e.g., time intervals between waves such as PR interval, QT interval, and RR interval, heart rate, and repetition patterns in wave occurrences), frequency features (e.g., specific frequency bands), among others. Wherein the processor is configured to segment time-series data into plurality of time-series segments if a particular signal feature, for example, and without limitation, at least a QRS complex is detected. As person skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various signal features that can be used identify plurality of time-series segments, specific cardiac conditions, abnormalities, or events).
Regarding claim 8, Jain as modified by Garvey teach wherein the pre-processing of the time-series data comprises selecting a pre-processing pipeline having a smallest difference between each of data that have been pre-processed by a plurality of built pre-processing pipelines, respectively, and correct answer data that have been previously prepared when the plurality of built pre-processing pipelines is present (Claim 8 text, [0080], [0085], [0115], [0161], [0176] wherein Jain Obtains annotated time series segments for data processing process, wherein the prediction model data in some cases, including one or more preliminary segment identifications, is corrected by the expert. Wherein overlapping is removed based on the expert labelled data, and the model prediction data. Wherein a label expansion is performed. For example, labels with signal lengths less than a predetermined threshold are expanded to meet this requirement. Wherein a script employs an ‘extend’ factor, calculated as the absolute difference between the expected signal length and the actual length, the labels are adjusted to meet this desired length while maintaining data integrity. If an adjacent label also requires expansion, the script adjusts the label to ensure a continuous signal. Furthermore, if the signal overlaps with the preceding or succeeding label, the expansion is performed to maintain signal integrity. A signal expansion may be performed for the negative categories).
Regarding claim 9, Jain as modified by Garvey teach wherein the pre-processing of the time-series data comprises building the pre-processing pipeline based on input and output feature information of the pre-processing module and input and output feature information between a previous pre-processing module and a subsequent pre-processing module ([0054], [0058], [0084-0085], [0093-0098] wherein Jain displays a visual representation of time-series data and preprocess it applying pipelines), (Abstract, [0009-0010], [0065], [0084], [0093], [0095], [0126], [0144] wherein Jain describes the model pipeline for receiving the preprocesses data, wherein the time-series segments are annotated by domain experts using labeling module that may be ground truth labels. Ground truth labels may be added to existing labeled training data. The expanded dataset may be processed by converting plurality of annotated time-series segments to canonical data format as described above. Such conversion may involve normalizing signal amplitudes, segmenting or augmenting the data into fixed-length segment, removing artifacts, reducing noise, filtering segments based on a quality threshold, and the like. Processors may load expanded dataset into training pipelines. At least a classifier may then be trained on the updated dataset. During retraining, at least a classifier may map input time-series segments to corresponding labels provided by the experts. In some cases, performance may be again evaluated after retraining. As more time-series data e.g., patient data is collected, additional time-series segments may be identified that were not previously labeled. At least a classifier may be continuously retrained to include new signal features improving model's accuracy and robustness over time).
Regarding claim 10, Jain as modified by Garvey teach selectively outputting pre-processing modules which are applicable depending on whether the time-series data that are input correspond to a single type or multiple types; selectively applying a first pre-processing module, among the output pre-processing modules; selectively outputting pre-processing modules that are applicable after the first pre-processing module depending on whether the time-series data output by the first pre-processing module correspond to a single type or multiple types; and selectively applying a second pre-processing module, among the output pre-processing modules ([0010], [0038], [0073-0077], [0086], [0096-0098], [0196] wherein Jain describes a classification algorithm that sorts inputs in into categories or bins of data, outputting the categories or bins of data and/or labels associated therewith. Jain applies a sigmoid activation function that is used to squash the model's output between 0 and 1, which can be interpreted as probability. This is particularly useful in binary classification problems interpreting the model's output as the probability of belonging to a particular class (i.e., segment identification or label). During training, at least a classifier may generate raw output values that may be used for computing the loss, and applying the sigmoid during testing helps in obtaining probability-like values for generating segment identifications).
Regarding claim 11, Jain as modified by Garvey teach selecting at least one of the previously built pre-processing pipelines when receiving new single or multiple types of time-series data; and applying the selected pre-processing pipeline to any one interval of the segmented intervals of the time-series data ([0089], wherein Jain incorporates incorporate machine learning assistance, where incorporate machine learning assistance, where preliminary segment identifications or labels are suggested, for example, by one or more machine learning algorithms as described herein based on previously labeled data. As a non-limiting example, labeling module that may interface with at least a classifier and detector to request, upon receiving time-series data and identifying plurality of time-series segments, a machine-generated label for each time-series segment of plurality of time-series segments. In some cases, annotating each time-series segment of plurality of time-series segments that may include selectively generating, using at least a classifier, at least one segment identification as a default label for each time-series segment of plurality of time-series segments. In cases where a plurality of classifiers, each one of the plurality of classifiers implements a different classification algorithm is trained, labeling module may allow user to select, via user interface, at least a classifier of the plurality of classifiers to generate at least one segment identification), ([0054], [0058], [0084-0085], [0093-0098] wherein Jain displays a visual representation of time-series data and preprocess it applying pipelines), (Abstract, [0009-0010], [0065], [0084], [0093], [0095], [0126], [0144] wherein Jain describes the model pipeline for receiving the preprocesses data, wherein the time-series segments are annotated by domain experts using labeling module that may be ground truth labels. Ground truth labels may be added to existing labeled training data. The expanded dataset may be processed by converting plurality of annotated time-series segments to canonical data format as described above. Such conversion may involve normalizing signal amplitudes, segmenting or augmenting the data into fixed-length segment, removing artifacts, reducing noise, filtering segments based on a quality threshold, and the like. Processors may load expanded dataset into training pipelines. At least a classifier may then be trained on the updated dataset. During retraining, at least a classifier may map input time-series segments to corresponding labels provided by the experts. In some cases, performance may be again evaluated after retraining. As more time-series data e.g., patient data is collected, additional time-series segments may be identified that were not previously labeled. At least a classifier may be continuously retrained to include new signal features improving model's accuracy and robustness over time).
Regarding claim 12, Jain as modified by Garvey teach wherein the pre-processing of the time-series data comprises pre-processing the time-series data by segmenting and applying the built pre-processing pipeline based on a physical characteristic interval (Abstract, [0009-0010], [0065], [0084], [0093], [0095], [0126], [0144] wherein Jain describes the model pipeline for receiving the preprocesses data, wherein the time-series segments are annotated by domain experts using labeling module that may be ground truth labels. Ground truth labels may be added to existing labeled training data. The expanded dataset may be processed by converting plurality of annotated time-series segments to canonical data format as described above. Such conversion may involve normalizing signal amplitudes, segmenting or augmenting the data into fixed-length segment, removing artifacts, reducing noise, filtering segments based on a quality threshold, and the like. Processors may load expanded dataset into training pipelines. At least a classifier may then be trained on the updated dataset. During retraining, at least a classifier may map input time-series segments to corresponding labels provided by the experts. In some cases, performance may be again evaluated after retraining. As more time-series data e.g., patient data is collected, additional time-series segments may be identified that were not previously labeled. At least a classifier may be continuously retrained to include new signal features improving model's accuracy and robustness over time).
Regarding claim 13, Jain as modified by Garvey teach wherein the generating of the pattern classification model for classifying the patterns of the pre-processed time-series data based on the clustering of the pre-processed time-series data comprises generating the pattern classification model that classifies the patterns of the time-series data by a pre-designated clustering number or an automatically adjusted clustering number ([0009], [0037], [0064], [0067-0068], [0081-0082], [0087], [0093], [0172] wherein Jain describes a system and method for data management and visualization of the cardiac signals. The system incorporates machine learning models to improve the identification and classification of cardiac signals. Specifically, a machine learning module is integrated into the system, enabling adaptive learning from real-time cardiac signal data. The system continuously refines its labelling capabilities based on evolving patterns and variations in the cardiac signals. Furthermore, the present disclosure discloses a method and a script for real-time simulation of the cardiac signals and real-time segmentation of the cardiac signals. In comparison with traditional labelling techniques, the disclosed system may label the cardiac signals quickly and accurately. Wherein Jain identifies a plurality of time-series segments 132 from received time-series data 112. As used in this disclosure, a “time-series segment” refers to a contiguous subset of time-series data that is defined by specific start and end points within the overall dataset. Each time-series segment of the plurality of time-series segments 132 may capture, for instance, and without limitation, a discrete portion of the time-series data 112 and is used for analysis, processing, and/or labeling to identify one or more specific events or patterns within time-series data 112. Processor 104 may identify and characterizing cardiac events by examining, for example, and at least in part, the activity of the heart over a defined time window. Wherein the system includes a classification method that utilizes feature similarity to analyze how closely out-of-sample-features resemble training data to classify input data to one or more clusters and/or categories of features as represented in training data).
Regarding claim 14, Jain as modified by Garvey teach wherein the generating of the pattern classification model for classifying the patterns of the pre-processed time-series data based on the clustering of the pre-processed time-series data comprises generating the pattern classification model for classifying the patterns of the time-series data based on the drawn time-series data and a label corresponding to the drawn time-series data ([0009], [0037], [0064], [0067-0068], [0081-0082], [0087], [0093], [0172] wherein Jain describes a system and method for data management and visualization of the cardiac signals. The system incorporates machine learning models to improve the identification and classification of cardiac signals. Specifically, a machine learning module is integrated into the system, enabling adaptive learning from real-time cardiac signal data. The system continuously refines its labelling capabilities based on evolving patterns and variations in the cardiac signals. Furthermore, the present disclosure discloses a method and a script for real-time simulation of the cardiac signals and real-time segmentation of the cardiac signals. In comparison with traditional labelling techniques, the disclosed system may label the cardiac signals quickly and accurately. Wherein Jain identifies a plurality of time-series segments 132 from received time-series data 112. As used in this disclosure, a “time-series segment” refers to a contiguous subset of time-series data that is defined by specific start and end points within the overall dataset. Each time-series segment of the plurality of time-series segments 132 may capture, for instance, and without limitation, a discrete portion of the time-series data 112 and is used for analysis, processing, and/or labeling to identify one or more specific events or patterns within time-series data 112. Processor 104 may identify and characterizing cardiac events by examining, for example, and at least in part, the activity of the heart over a defined time window. Wherein the system includes a classification method that utilizes feature similarity to analyze how closely out-of-sample-features resemble training data to classify input data to one or more clusters and/or categories of features as represented in training data).
Regarding claim 15, Jain as modified by Garvey teach wherein the generating of the predictive model for predicting the feature information of the drawn time-series data comprises generating predictive models having a number corresponding to all clusters generated as the results of the clustering ([0172] wherein Jain generates a classifier using a K-nearest neighbors (KNN) algorithm. A “K-nearest neighbors algorithm” as used in this disclosure, includes a classification method that utilizes feature similarity to analyze how closely out-of-sample-features resemble training data to classify input data to one or more clusters and/or categories of features as represented in training data; this may be performed by representing both training data and input data in vector forms, and using one or more measures of vector similarity to identify classifications within training data, and to determine a classification of input data. K-nearest neighbors algorithm may include specifying a K-value, or a number directing the classifier to select the k most similar entries training data to a given sample, determining the most common classifier of the entries in the database, and classifying the known sample; this may be performed recursively and/or iteratively to generate a classifier that may be used to classify input data as further samples. For instance, an initial set of samples may be performed to cover an initial heuristic and/or “first guess” at an output and/or relationship, which may be seeded, without limitation, using expert input received according to any process as described herein. As a non-limiting example, an initial heuristic may include a ranking of associations between inputs and elements of training data. Heuristic may include selecting some number of highest-ranking associations and/or training data elements).
Regarding claim 16, Jain as modified by Garvey teach wherein the generating of the predictive model for predicting the feature information of the drawn time-series data comprises: generating predictive models having a number corresponding to upper n (n is a natural number) selected clusters each having a large number of time-series data included in each of all clusters generated as the results of the clustering; and generating the pattern classification model again based on the selected clusters so that the selected clusters correspond to the number of generated predictive models ([0168] wherein Jain describes training data that contains correlations that a machine-learning process may use to model relationships between two or more categories of data elements. For instance, and without limitation, training data 1004 may include a plurality of data entries, also known as “training examples,” each entry representing a set of data elements that were recorded, received, and/or generated together; data elements may be correlated by shared existence in a given data entry, by proximity in a given data entry, or the like. Multiple data entries in training data 1004 may evince one or more trends in correlations between categories of data elements; for instance, and without limitation, a higher value of a first data element belonging to a first category of data element may tend to correlate to a higher value of a second data element belonging to a second category of data element, indicating a possible proportional or other mathematical relationship linking values belonging to the two categories. Multiple categories of data elements may be related in training data 1004 according to various correlations; correlations may indicate causative and/or predictive links between categories of data elements, which may be modeled as relationships such as mathematical relationships by machine-learning processes as described in further detail below. Training data 1004 may be formatted and/or organized by categories of data elements, for instance by associating data elements with one or more descriptors corresponding to categories of data elements. As a non-limiting example, training data 1004 may include data entered in standardized forms by persons or processes, such that entry of a given data element in a given field in a form may be mapped to one or more descriptors of categories. Elements in training data 1004 may be linked to descriptors of categories by tags, tokens, or other data elements; for instance, and without limitation, training data 1004 may be provided in fixed-length formats, formats linking positions of data to categories such as comma-separated value (CSV) formats and/or self-describing formats such as extensible markup language (XML), JavaScript Object Notation (JSON), or the like, enabling processes or devices to detect categories of data).
Regarding claim 17, Jain as modified by Garvey teach receiving time-series data that are collected at a new place and that have a domain identical with the specific domain ([0089], wherein Jain incorporates incorporate machine learning assistance, where incorporate machine learning assistance, where preliminary segment identifications or labels are suggested, for example, by one or more machine learning algorithms as described herein based on previously labeled data. As a non-limiting example, labeling module that may interface with at least a classifier and detector to request, upon receiving time-series data and identifying plurality of time-series segments, a machine-generated label for each time-series segment of plurality of time-series segments. In some cases, annotating each time-series segment of plurality of time-series segments that may include selectively generating, using at least a classifier, at least one segment identification as a default label for each time-series segment of plurality of time-series segments. In cases where a plurality of classifiers, each one of the plurality of classifiers implements a different classification algorithm is trained, labeling module may allow user to select, via user interface, at least a classifier of the plurality of classifiers to generate at least one segment identification) classifying the received time-series data as a corresponding cluster by inputting the received time-series data to the pattern classification model ([0009], [0064], [0067-0068] wherein Jain incorporates a vector-like by extracting data features and producing patterns of similar domain) selecting a predictive model corresponding to the classified cluster ([0062], [0089] wherein Jain incorporates filtering technique that decomposes the signal into components at different scales, allowing for the selective filtering of noise while retaining important signal features and allows for selectively generating, using at least a classifier, at least one segment identification as a default label for each time-series segment of plurality of time-series segments. In cases where a plurality of classifiers, each one of the plurality of classifiers implements a different classification algorithm is trained, labeling module may allow user to select, via user interface, at least a classifier of the plurality of classifiers to generate at least one segment identification) and outputting results of prediction of feature information of the received time-series data based on the selected predictive model ([0128], [0140], [0189] wherein Jain outputs the prediction result).
Regarding claim 18, Jain as modified by Garvey teach An electronic device comprising: a processor configured to receive single or multiple types of time-series data that are collected in a specific domain, build a pre-processing pipeline by applying at least one pre-processing module to the time-series data, and pre-process the time-series data by applying the built pre-processing pipeline (Abstract, [0010], [0038], [0061], [0076], [0090], [0100], [0163-0164] wherein Jain uses the at least a classifier, one or more segment identifications at each time-series segment of a plurality of time-series segments subsequently identified using the detector based on continuous time-series data. Wherein Jain describes the steps for a specific domain, wherein the steps include processing time series data that may be done using one or more digital filtering techniques to enhance the quality of the recorded signals. Applying “digital filtering techniques” that are methods used to manipulate and refine digital signals to remove unwanted components. This may include removing noise or artifacts, while preserving the essential features of the signal. Digital filtering techniques may ensure that the time-series data is accurate and reliable for any subsequent processing, segmentation, and analysis. A digital filtering technique may include the application of mathematical algorithms to the raw data to isolate and remove specific frequency components. For example, low-pass filters may allow signals with frequencies below a certain threshold to pass through while attenuating higher frequency noise. Conversely, high-pass filters permit high-frequency signals to pass while reducing the impact of lower frequency interference. Band-pass filters combine these principles to isolate a specific range of frequencies, which is particularly useful for focusing on the relevant portions of the cardiac signal. Examples of digital filtering techniques may include the use of Finite Impulse Response (FIR) filters, which apply a finite sequence of weights to the signal. FIR filters are known for their stability and linear phase response, making them ideal for applications requiring precise timing, such as ECG and IEGM data analysis), ([0064], [0082], [0097], [0115], [0125] wherein Jain allows visualizing time-series developing model as a prediction model for preprocessing data through pipelines and classify data based on patterns of time-series).
Regarding claim 19, Jain as modified by Garvey teach a processor configured to draw time-series data corresponding to an identical domain and an identical time interval, among time-series data that have been previously collected in a specific domain, generate a pattern classification model for classifying patterns of the time-series data based on a clustering of the drawn time-series data, store the generated pattern classification model, generate a predictive model for predicting feature information of the drawn time-series data based on a cluster that is generated as results of the clustering of the time-series data, and store the generated predictive model (Abstract, [0010], [0038], [0061], [0076], [0090], [0100], [0163-0164] wherein Jain uses the at least a classifier, one or more segment identifications at each time-series segment of a plurality of time-series segments subsequently identified using the detector based on continuous time-series data. Wherein Jain describes the steps for a specific domain, wherein the steps include processing time series data that may be done using one or more digital filtering techniques to enhance the quality of the recorded signals. Applying “digital filtering techniques” that are methods used to manipulate and refine digital signals to remove unwanted components. This may include removing noise or artifacts, while preserving the essential features of the signal. Digital filtering techniques may ensure that the time-series data is accurate and reliable for any subsequent processing, segmentation, and analysis. A digital filtering technique may include the application of mathematical algorithms to the raw data to isolate and remove specific frequency components. For example, low-pass filters may allow signals with frequencies below a certain threshold to pass through while attenuating higher frequency noise. Conversely, high-pass filters permit high-frequency signals to pass while reducing the impact of lower frequency interference. Band-pass filters combine these principles to isolate a specific range of frequencies, which is particularly useful for focusing on the relevant portions of the cardiac signal. Examples of digital filtering techniques may include the use of Finite Impulse Response (FIR) filters, which apply a finite sequence of weights to the signal. FIR filters are known for their stability and linear phase response, making them ideal for applications requiring precise timing, such as ECG and IEGM data analysis), ([0064], [0082], [0097], [0115], [0125] wherein Jain allows visualizing time-series developing model as a prediction model for preprocessing data through pipelines and classify data based on patterns of time-series), ([0064], [0082], [0115], [0125], [0172-0173] wherein Jain describes developing model for prediction and evaluating dataset wherein the predictions are obtained by using the model. Jain describes generating a classifier that includes a classification method that utilizes feature similarity to analyze how closely out-of-sample-features resemble training data to classify input data to one or more clusters and/or categories of features as represented in training data).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to HASSAN MRABI whose telephone number is (571)272-8875. The examiner can normally be reached on Monday-Friday, 7:30am-5pm. Alt, Friday, EST.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Viker Lamardo can be reached on 571-270-5871. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/HASSAN MRABI/Examiner, Art Unit 2144
Prosecution Insights