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 .
This office action is in response to submission of application on 1/5/2024.
Claims 1-27 are presented for examination.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-27 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1: Is the claim to a process, machine, manufacture or composition of matter?
Claims 1-13 are directed to a method (i.e., a process); claims 14-26 are directed
to an apparatus (i.e., a machine/apparatus); and claim 27 is directed to an article of
manufacture (i.e., a product); therefore, all pending claims are directed to one of the four
categories of invention.
Independent Claims
Step 2A Prong One: Does the claim recite an abstract idea, law of nature, or natural phenomenon?
Yes, independent claim 1 recites an abstract idea in the form of mental processes. A mental process is a process that “can be performed in the human mind, or by a human using a pen and paper” (MPEP§ 2106.04(a)(2)(III), paragraph 1). Examples of mental processes include “observations, evaluations, judgments, and opinions” (MPEP § 2106.04(a)(2)(III), paragraph 2).
The following limitations of claim 1 are mental processes:
(b) generating a feature set comprising a plurality of coefficients by at least (i) selecting a set of basis functions from a plurality of predetermined learner functions indicative of properties of the electrochemical charge transfer at a sensor interface of the sensor platform, and (ii) generating the plurality of coefficients by at least projecting the current measurement data on the set of basis functions; [This is a mental process that can be performed by observations, evaluations, judgments, and opinions. No specific methodology for selecting a basis function or generating coefficients by projecting is recited in the claim; therefore, it broadly encompasses processing that can be performed as a mental process.]
(c) selecting a first Machine Learning (ML) model type from a predetermined set of ML model types, the selecting based on the received user-selected analysis; and [This is a mental process that can be performed by observations, evaluations, judgments, and opinions.]
Therefore, the independent claims recite a judicial exception.
Step 2A Prong Two: Does the claim recite additional elements that integrate the judicial exception into a practical application?
No. The judicial exception recited in the above discussed claims is not integrated into a
practical application.
(a) receiving data comprising current and voltage measurement data associated with a first sample by at least a sensor platform, metadata associated with the sensor platform, and a user-selected analysis to be performed on the current measurement data, wherein the current measurement data includes current measurement signal data as a function of voltage applied by the sensor platform on the first sample and a measurement time and voltage measurement data includes voltage measurement signal as function of applied set point voltage and a measurement time [receiving data is sending data, which is insignificant, extra-solution activity. See MPEP 2106.05(g). Transmitting data is well-understood, routine, and conventional. See MPEP 2106.05(d)(II)(i).]
(d) providing the feature set to an ML model characterized by the selected ML model type, the first ML model configured to characterize the first sample. [This additional element does no more than generally link the use of a judicial exception to a particular technological environment or field of use (MPEP § 2106.05(h)). This element merely indicates a field of use or technological environment in which a judicial exception is applied, namely the feature set is provided to an ML model.].
Therefore, under MPEP 2106.04(d), the additional elements of the claims do not integrate
the judicial exception into a practical application.
Step 2B: Does the claim recite additional elements that amount to significantly more than the judicial exception?
No. The claims do not include additional elements that are sufficient for the claims to
amount to significantly more than the judicial exception.
Additional elements that are mere instructions to apply an exception or merely generally
linking or generally linking the use of a judicial exception to a particular technological
environment or field of use do not constitute significantly more than a judicial exception under
MPEP§2106.05(I)(A). Since the additional elements in the independent claims are all are mere
instructions to apply an exception or are merely generally linking or generally linking the use of
a judicial exception to a particular technological environment or field of use, they do not
constitute significantly more than a judicial exception.
Therefore, the additional elements identified in the Step 2A Prong Two analysis do not
constitute significantly more than a judicial exception.
Independent claims 14 and 27 recite the same relevant limitations and a similar analysis
applies. Claim 14 recites the additional elements of " A system comprising: at least one data processor; memory coupled to the at least one data processor, the memory storing instructions to cause the at least one data processor to perform operations comprising:" [A system comprising a processor and memory are components recited at a high level are construed as generic computer components used to implement the abstract idea. See MPEP 2106.05(f)(2). As such, the limitations do not integrate the abstract idea into a practical application. Nor to do they amount to significantly more.] Claim 27 recites the additional limitations of "A computer program product comprising a non-transitory machine-readable medium storing instructions that, when executed by at least one programmable processor that comprises at least one physical core and a plurality of logical cores, cause the at least one programmable processor to perform operations comprising:" – [A non-transitory machine-readable medium are components recited at a high level are construed as generic computer components used to implement the abstract idea. See MPEP 2106.05(f)(2). As such, the limitations do not integrate the abstract idea into a practical application. Nor to do they amount to significantly more.]
Therefore, the independent claims are not patent eligible.
Dependent Claims
The remaining dependent claims being rejected do not recite additional elements, whether considered individually or in combination, that are sufficient to integrate the judicial exception into a practical application or amount to significantly more than the judicial exception.
Claims 2 and 15
wherein the metadata associated with the sensor platform includes physical properties of the sensor platform indicative of the electrochemical charge transfer at the sensor interface and/or operational properties of the sensor platform associated with detection of the current measurement signal. [This additional element does no more than generally link the use of a judicial exception to a particular technological environment or field of use (MPEP § 2106.05(h)). This element merely indicates a field of use or technological environment in which a judicial exception is applied, namely the metadata associated with the sensor platform includes physical and operational properties.]
Claims 3 and 16
wherein the received data further includes one or more of (a) data of the source of the first sample, (b) quantitative information associated with analyte species determined from other analysis methods; (c) date and time of first sample collection, storage and re-thaw; (d) one or more quality controls applied to the first sample during collection, storage; (e) any quality control applied to first sample just before analysis; (f) information about co-morbidities of first sample source; (g) disease-relevant phenotype for first sample. [This additional element does no more than generally link the use of a judicial exception to a particular technological environment or field of use (MPEP § 2106.05(h)). This element merely indicates a field of use or technological environment in which a judicial exception is applied, namely the different types of information included in the data..]
Claims 4 and 17
wherein selecting the set of basis functions includes: selecting a first set of learner functions and a second set of learner functions from the plurality of predetermined learner functions; fitting the current measurement signal data with the first set of learner functions and the second set of learner function; and calculating a first prediction error and a second prediction error associated with the fitting of the current measurement signal with the first set of learner function and the second set of learner function, respectively. [These further limitations merely further define the mental process recited in the parent claim and are therefore considered to be part of the mental process of the parent claim. This claim does not recite any additional non-abstract elements for purposes of Step 2A Prong Two and Step 2B analysis.]
Claims 5 and 18
selecting one of the first set of learner functions and the second set of learner functions based on the first prediction error and the second prediction error. [Selecting a learner function based on prediction error are mere instructions to apply the abstract idea. Mere recitation that a judicial exception is to be performed using generic class of computer algorithms in their ordinary capacity, cannot meaningfully integrate the judicial exception into a practical application. See MPEP 2106.05(f)]
Claims 6 and 19
selecting the first set of learner functions wherein the first prediction error is smaller than the second prediction error. [Selecting a learner function based on the first prediction error being smaller than the second are mere instructions to apply the abstract idea. Mere recitation that a judicial exception is to be performed using generic class of computer algorithms in their ordinary capacity, cannot meaningfully integrate the judicial exception into a practical application. See MPEP 2106.05(f)]
Claims 7 and 20
selecting a first ML model having the first ML model type, wherein the first trained ML model is characterized by the first model type; determining that the first ML model does not require further training; and generating an output by the first ML model configured to receive the feature set and user defined metadata as an input. [These further limitations merely further define the mental process recited in the parent claim and are therefore considered to be part of the mental process of the parent claim. This claim does not recite any additional non-abstract elements for purposes of Step 2A Prong Two and Step 2B analysis.]
Claims 8 and 21
wherein the user specified analysis includes assigning a class to an analyte in the first sample and wherein the first ML model is a classifier configured to assign the class to the analyte. [This additional element does no more than generally link the use of a judicial exception to a particular technological environment or field of use (MPEP § 2106.05(h)). This element merely indicates a field of use or technological environment in which a judicial exception is applied, namely the user specified analysis including assigning a class to an analyte and a ML model assigns the class to the analyte].
Claims 9 and 22
wherein the user-specified analysis includes quantification of concentration of an analyte in the first sample. [The user-specified analysis includes quantification of concentration of an analyte are mere instructions to apply the abstract idea. Mere recitation that a judicial exception is to be performed using generic class of computer algorithms in their ordinary capacity, cannot meaningfully integrate the judicial exception into a practical application. See MPEP 2106.05(f)]
Claims 10 and 23
selecting a second ML model having the first ML model type, wherein the first trained ML model is characterized by the first model type; determining that the second ML model requires further training; training, using a training model, the second ML model based on training data including one or more of first sample data, metadata associated with detection of current measurement signal and previously generated output of the second ML model; generating an output by the second ML model configured to receive the feature set and user defined metadata as an input. [These further limitations merely further define the mental process recited in the parent claim and are therefore considered to be part of the mental process of the parent claim. This claim does not recite any additional non-abstract elements for purposes of Step 2A Prong Two and Step 2B analysis.]
Claims 11 and 24
training the second ML model to assign a class type associated with the first sample, wherein the second ML model is a classifier configured to assign the class to an analyte, wherein the training data is based on one or more samples assigned the class type, wherein training the classifier includes determining classifier boundary; and assigning the class type to the analyte in the first sample using the trained second ML to assign a class to the sample. [Training the ML model and assigning a class type to an analyte are mere instructions to apply the abstract idea. Mere recitation that a judicial exception is to be performed using generic class of computer algorithms in their ordinary capacity, cannot meaningfully integrate the judicial exception into a practical application. See MPEP 2106.05(f)]
Claims 12 and 25
defining calibration analyte samples; analyzing the calibration analyte samples; training the second ML algorithm based on a Scattered Component Analysis (SCA) to determine a projection vector that maximizes similarity to analyte-specific reference sample data while minimizing similarity to matrix-specific reference data and/or similarity to chemically and structurally similar analyte reference data, to digitally subtract the contribution of the background and other similar analytes to the signal; and determining a concentration of the analyte by at least projecting, by the trained second ML algorithm, the sample data onto the projection vector. [Defining and analyzing an analyte, training a ML algorithm, and determining a concentration of an analyte are mere instructions to apply the abstract idea. Mere recitation that a judicial exception is to be performed using generic class of computer algorithms in their ordinary capacity, cannot meaningfully integrate the judicial exception into a practical application. See MPEP 2106.05(f). Additionally, this is a description of how the abstract idea is performed, using Scattered Component Analysis (SCA). As such, this merely describes a technological environment. See MPEP 2106.05(h).]
Claims 13 and 26
determining that an ML model having the first ML model type does not exist; identifying a second sample based on a predetermined relationship with the first sample; identifying a third ML model and second training data associated with the second sample, [This is a mental process that can be performed by observations, evaluations, judgments, and opinions. No specific methodology for determining an ML model does not exist and identifying a sample, ML model, and training data is recited in the claim; therefore, it broadly encompasses processing that can be performed as a mental process.]
the second training data including one or more of the second sample data, metadata associated with detection of a current measurement signal associated with the second sample and previously generated output of the third ML model; [This additional element does no more than generally link the use of a judicial exception to a particular technological environment or field of use (MPEP § 2106.05(h)). This element merely indicates a field of use or technological environment in which a judicial exception is applied, namely the training data including sample data, metadata, and previous output.].
training, using a training model, the third ML model based on the second training data; and generating an output by the third ML model configured to receive the feature set and user defined metadata as an input. [Training and generating output by an ML model are mere instructions to apply the abstract idea. Mere recitation that a judicial exception is to be performed using generic class of computer algorithms in their ordinary capacity, cannot meaningfully integrate the judicial exception into a practical application. See MPEP 2106.05(f). Additionally, this is a description of how the abstract idea is performed, using a training model. As such, this merely describes a technological environment. See MPEP 2106.05(h).]
The prior art used for rejections are provided below:
1. US7090764B2 (August 15, 2006) to Iyengar et al. (hereinafter Iyengar)
2. WO2019200410A1 (October 17, 2019) to Drake et al. (hereinafter Drake).
3. US10413228B2 (September 17, 2019) to Iyengar et al. (hereinafter Iyengar 2019).
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1, 3-14, and 16-27 are rejected under 35 U.S.C. 103 as being unpatentable over Iyengar in view of Drake.
Per claim 1, Iyengar discloses (b) generating a feature set comprising a plurality of coefficients by at least (i) selecting a set of basis functions from a plurality of predetermined learner functions indicative of properties of the electrochemical charge transfer at a sensor interface of the sensor platform, and (ii) generating the plurality of coefficients by at least projecting the current measurement data on the set of basis functions [Iyengar, column 2, line 46 “resolving a signal contribution from the select analyte in the generated signal by a vector projection method with an analyte vector comprising a plurality of real and imaginary parts of one or more Fourier coefficients at one or more frequencies of a reference current signal for the select analyte.”. (note: Fourier basis functions are predetermined learner functions indicative of properties of the electrochemical charge transfer at the sensor interface. The Fourier coefficients are the plurality of coefficients generated by projecting the current measurement data on the set of basis functions. The real and imaginary parts of Fourier coefficients constitute the feature set.); column 3, line 4 “computing at least one Fourier coefficient of a desired frequency component of all or some portion of the generated signal;”. (note: computing Fourier coefficients is projecting the current signal onto sinusoidal basis functions. The Fourier transform is a projection operation.); column 8, line 23 “Process 24 uses a set of equations constructed based on the known spectral characteristics of each ESS from data source 26 (see step 135). Data source 26 may include for example equations that describe how each ESS interacts with other ESSs in the sample, data about the known spectral characteristics of each ESS that may be gathered by electrochemical assay…”. (note: known spectral characteristics are predetermined learner functions indicative of properties of the electrochemical charge transfer. Equations are the basis functions used to generate the feature set.)];
Iyengar does not expressly disclose, but Iyengar combined with Drake does teach:
A method for characterizing biological samples, the method comprising [Drake, pg. 2, para. 0007, “Described herein are methods and systems that incorporate machine learning approaches with one or more biological analytes in a biological sample for various applications…”]:
(a) receiving data comprising current and voltage measurement data associated with a first sample by at least a sensor platform, metadata associated with the sensor platform [Iyengar, column 5, line 67 “The factors generally recognized to affect an analyte signal are reaction rate and mechanism, generally referred to as the kinetics of the reaction, and the transport properties of the analyte.”. (note: these are physical properties of the sensor analyte interface, metadata associated with the sensor platform indicating how the sensor responds); Drake, pg. 7, para. 0022 “a feature module to identify a set of features corresponding to an assay that are operable to be input to the machine learning model for each of the plurality of training samples, wherein the set of features correspond to properties of molecules in the plurality of training samples,”. (note: this shows receiving metadata ssociated with an assay (properties of the measurement system and training samples) as inputs with the measurement data. This information about the assay conditions and instrument used is metadata associated with the sensor platform.); pg. 7, para. 0022 “…training vector comprises feature values of the N set of features of the corresponding assay, each feature value corresponding a feature and including to one or more measured values, wherein the training vector is formed using at least one feature from at least two of the N sets of features corresponding to a first subset of the plurality of different assays,”. (note: this shows the training vector includes features from multiple assay types, which is assay metadata about the detection platform)], and a user-selected analysis to be performed on the current measurement data, wherein the current measurement data includes current measurement signal data as a function of voltage applied by the sensor platform on the first sample and a measurement time [Iyengar, column 2, line 55 “…applying a large amplitude potential stimulus waveform to the sample to generate a nonlinear current signal; measuring the generated signal”. (note: a potential stimulus waveform is a voltage signal as a function of time, it is a time varying voltage applied to the sample. Measuring the generated signal (the resulting current) while applying a time varying voltage waveform is measuring the current signal data as a function of both applied voltage and measurement time. The waveform application indexes both the voltage and the time at each measurement point, which is current measurement signal data as a function of voltage applied by the sensor platform on the first sample and a measurement time, and voltage measurement signal as function of applied set point voltage and a measurement time.)]
and voltage measurement data includes voltage measurement signal as function of applied set point voltage and a measurement time [Iyengar, column 2, line 55 “…applying a large amplitude potential stimulus waveform to the sample to generate a nonlinear current signal; measuring the generated signal”. (note: this shows receiving current measurement data (the generated signal) and voltage measurement data (the potential stimulus waveform) from a sensor platform. The current is measured as a function of the applied voltage waveform.);][Drake, pg. 3, para. 0010 “…the disclosure provides a method of using a classifier capable of distinguishing a population of individuals comprising: a) assaying a plurality of classes of molecules in the biological sample, wherein the assaying provides a plurality of sets of measured values representative of the plurality of classes of molecules,”. (note: this shows receiving measured values from assays on biological samples, and specifies a user selected analysis type (classification or quantification))]; (c) selecting a first Machine Learning (ML) model type from a predetermined set of ML model types, the selecting based on the received user-selected analysis [Drake, pg. 4, para. 0020 “…the classifying of the biological sample is performed by a classifier trained and constructed according to one or more of: linear discriminant analysis (LDA); partial least squares (PLS); random forest; k-nearest neighbor (KNN): support vector machine (SVM) with radial basis function kernel (SVMRadial): SVM with linear basis function kernel (SVMLinear): SVM with polynomial basis function kernel (SVMPoly), decision trees, multilayer perceptron, mixture of experts, sparse factor analysis, hierarchical decomposition and combinations of linear algebra routines and statistics.”. (note: this shows a predetermined set of ML model types from which a classifier type is selected based on the analysis to be performed) ]; and
(d) providing the feature set to an ML model characterized by the selected ML model type, the first ML model configured to characterize the first sample [Drake, pg. 5, para. 0022 “a feature module to identify a set of features corresponding to an assay that are operable to be input to the machine learning model for each of the plurality of training samples, wherein the set of features correspond to properties of molecules in the plurality of training samples,”. (note: the feature set is provided (input)to the ML model. The ML model is configured to characterize the sample (classify))].
Iyengar and Drake are analogous art because they are from the same field of endeavor of electrochemical biosensors for and machine learning applied to biological sample characterization and analyte classification and quantification.
Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to use the electrochemical measurement and Fourier-coefficient feature extraction as the front end, sending the result into Drake’s ML model selection and training pipeline as the back end.
The suggestion/motivation for doing so is improved flexibility, accuracy, and adaptability to different analyte classification and quantifications, which is explicitly stated by Drake. [Drake, pg. 2, para. 0005 “…machine learning may enable large-scale statistical approaches and automated characterization of signal strength.”].
Per Claim 3, Iyengar-Drake discloses claim 1.
Iyengar does not fully disclose, but with Drake does teach the received data further includes one or more of (a) data of the source of the first sample, (b) quantitative information associated with analyte species determined from other analysis methods; (c) date and time of first sample collection, storage and re-thaw; (d) one or more quality controls applied to the first sample during collection, storage; (e) any quality control applied to first sample just before analysis; (f) information about co-morbidities of first sample source; (g) disease-relevant phenotype for first sample [Drake, pg. 3, para. 0009 “The study of non-cellular analytes in a liquid biological sample (e.g., plasma) permits deconvolution of the sample to recapitulate the molecular state of the individual's tissue and immune cells in a living cellular state”. (note: this receives data about the source of the sample (individual’s tissue and immune cells) disease-relevant phenotype (cancer type), and co-morbidities.); pg. 5, para. 0021 “…the specified property can be a clinically-diagnosed disorder”. (note: this shows disease-relevant phenotype for the first sample); pg. 20, para. 0104 “…the methods and systems are useful for predicting disease, treatment efficacy and guiding treatment decisions for affected individuals.”. (note: treatment decisions include data about co-morbidities and prior-analysis, which is quantitative information associated with analyte species determined from other analysis methods)].
Iyengar and Drake are analogous art because they are from the same field of endeavor of electrochemical biosensors for and machine learning applied to biological sample characterization and analyte classification and quantification.
Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to use the electrochemical measurement and Fourier-coefficient feature extraction as the front end, sending the result into Drake’s ML model selection and training pipeline as the back end.
The suggestion/motivation for doing so is improved flexibility, accuracy, and adaptability to different analyte classification and quantifications, which is explicitly stated by Drake. [Drake, pg. 2, para. 0005 “…machine learning may enable large-scale statistical approaches and automated characterization of signal strength.”].
Per Claim 4, Iyengar-Drake discloses claim 1.
Iyengar further teaches selecting the set of basis functions includes:
selecting a first set of learner functions and a second set of learner functions from the plurality of predetermined learner functions [Iyengar, column 6, line 46 “A parameter may be any feature of the signal, or any function of any portion (or all) of the signal. Examples of parameters include frequency-domain items such as Fourier transform coefficients at various frequencies.”. (note: this selects from a plurality of predetermined learner functions, different frequency-domain Fourier basis functions are distinct sets of learner functions)];
fitting the current measurement signal data with the first set of learner functions and the second set of learner function [Iyengar, column 6, line 62 “In steps 120 and 125 the frequency is selected that gives the best resolution of the FFT coefficients between the analytes. This is done by plotting the real vs. imaginary value of the FFT coefficient at each frequency (step 120). Thus a series of plots is created that has the real value of FFT coefficient on one axis and the imaginary value of the FFT coefficient on the other axis. Thus, if 128 frequencies are being considered, 128 different plots are constructed, one for each frequency. Each plot graphically shows the FFT value (real versus imaginary) at one frequency for all the data points (corresponding to different concentrations of each analyte).”. (note: fitting the current measurement signal data with first and second sets of learner functions is computing FFT coefficients at each candidate frequency. Each frequency represents a distinct basis function set being fitted to the signal data. This shows that multiple candidate sets are each independently fitted to the data)]; and
calculating a first prediction error and a second prediction error associated with the fitting of the current measurement signal with the first set of learner function and the second set of learner function, respectively [Iyengar, column 7, line 6 “The plots are screened to determine which frequency corresponds to the relatively greater separation between the two analyte vectors (step 125). This can be referred to as the “angle separation’ (angle or phase angle separation) between the analyte vectors.”. (note: screening the plots to find the relatively greater separation is calculating a prediction error for each set of learner functions. The angle separation between analyte vectors is the metric used to evaluate each fitted set, which is the first prediction error and second prediction error.)].
Per Claim 5, Iyengar-Drake discloses claim 4.
Iyengar further teaches selecting one of the first set of learner functions and the second set of learner functions based on the first prediction error and the second prediction error [Iyengar, column 7, line 6 “The plots are screened to determine which frequency corresponds to the relatively greater separation between the two analyte vectors (step 125).”; column 6, line 62 “In steps 120 and 125 the frequency is selected that gives the best resolution of the FFT coefficients between the analytes.”. (note: step 125 is the selection step. One of two sets of parameters, plotted at different frequencies, is selected based on which gives relatively greater separation. This is selecting one of the first or second set of learner functions based on comparison of prediction errors (separation).)].
Per Claim 6, Iyengar-Drake discloses claim 5.
Iyengar further teaches selecting the first set of learner functions wherein the first prediction error is smaller than the second prediction error [Iyengar, column 6, line 62 “In steps 120 and 125 the frequency is selected that gives the best resolution of the FFT coefficients between the analytes.”. (note: best resolution is the lowest error. When one frequency (first set of learner functions) gives better resolution (smaller prediction error) than another (second set), it is selected.)].
Per Claim 7, Iyengar-Drake discloses claim 1.
Iyengar does not fully disclose, but with Drake does teach selecting a first ML model having the first ML model type, wherein the first trained ML model is characterized by the first model type; determining that the first ML model does not require further training; and generating an output by the first ML model configured to receive the feature set and user defined metadata as an input [Drake, pg. 3, para. 0010 “…the machine learning model comprising the classifier, the machine learning model trained using training vectors obtained from training biological samples, a first subset of the trnining biological samples identified as having a specified property and a second subset of the training biological samples identified as not having the specified property,”. (note: a pre-trained ML model, which does not require further training, is selected and used to generate output from the feature vector.); pg. 6, para. 0023 “…a system for classifying subjects based on multi analyte analysis in a biological sample composition comprising: (a) a computer-readable medium comprising a classifier operable to classify the subjects based on the multi-analyte analysis; and (b) one or more processors for executing insirnctions stored the computer-readable medium.”. (note: the pre-trained classifier is loaded and used to generate an output classification, without requiring further training, using the feature set as input)].
Iyengar and Drake are analogous art because they are from the same field of endeavor of electrochemical biosensors for and machine learning applied to biological sample characterization and analyte classification and quantification.
Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to use the electrochemical measurement and Fourier-coefficient feature extraction as the front end, sending the result into Drake’s ML model selection and training pipeline as the back end.
The suggestion/motivation for doing so is improved flexibility, accuracy, and adaptability to different analyte classification and quantifications, which is explicitly stated by Drake. [Drake, pg. 2, para. 0005 “…machine learning may enable large-scale statistical approaches and automated characterization of signal strength.”].
Per Claim 8, Iyengar-Drake discloses claim 7.
Iyengar does not fully disclose, but with Drake does teach the user specified analysis includes assigning a class to an analyte in the first sample and wherein the first ML model is a classifier configured to assign the class to the analyte [Drake, pg. 3, para. 0010 “…the disclosure provides a method of using a classifier capable of distinguishing a population of individuals comprising: a) assaying a plurality of classes of molecules in the biological sample…”. (note: the ML model is a classifier that assigns a class (disease positive or negative) to an analyte in the biological sample); pg. 5, para. 0021 “…the specified property can be a clinically-diagnosed disorder”. (note: classifying the sample into a clinically-diagnosed disorder is assigning a class)].
Iyengar and Drake are analogous art because they are from the same field of endeavor of electrochemical biosensors for and machine learning applied to biological sample characterization and analyte classification and quantification.
Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to use the electrochemical measurement and Fourier-coefficient feature extraction as the front end, sending the result into Drake’s ML model selection and training pipeline as the back end.
The suggestion/motivation for doing so is improved flexibility, accuracy, and adaptability to different analyte classification and quantifications, which is explicitly stated by Drake. [Drake, pg. 2, para. 0005 “…machine learning may enable large-scale statistical approaches and automated characterization of signal strength.”].
Per Claim 9, Iyengar-Drake discloses claim 8.
Iyengar further teaches the user-specified analysis includes quantification of concentration of an analyte in the first sample [Iyengar, column 3, line 6 “…determining a concentration of the select analyte in the mixed sample by use of the at least one Fourier coefficient to resolve an estimation equation based on analyte vectors for each of the select and interfering analytes.”. (note: this shows quantification (determining a concentration) of an analyte as a user-specified analysis)].
Per Claim 10, Iyengar-Drake discloses claim 1.
Iyengar does not fully disclose, but with Drake does teach selecting a second ML model having the first ML model type, wherein the first trained ML model is characterized by the first model type; determining that the second ML model requires further training;
training, using a training model, the second ML model based on training data including one or more of first sample data, metadata associated with detection of current measurement signal and previously generated output of the second ML model;
generating an output by the second ML model configured to receive the feature set and user defined metadata as an input [Drake, pg. 5, para. 0022 “a labeling module to inform the system on the training vectors using parameters of the machine learning model to obtain output labels for the plurality of training samples”. (note: previously generated output labels are used in subsequent training, which is previously generated output of the second ML model being used as training data); pg. 6, para. 0022 “a training module to iteratively search for optimal values of the parameters as part of training the machine learning model based on the comparing the output labels to the known labels of the training samples…”. (note: a training module performs further training of the ML model based on training data that includes sample data and previously generated output labels)].
Iyengar and Drake are analogous art because they are from the same field of endeavor of electrochemical biosensors for and machine learning applied to biological sample characterization and analyte classification and quantification.
Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to use the electrochemical measurement and Fourier-coefficient feature extraction as the front end, sending the result into Drake’s ML model selection and training pipeline as the back end.
The suggestion/motivation for doing so is improved flexibility, accuracy, and adaptability to different analyte classification and quantifications, which is explicitly stated by Drake. [Drake, pg. 2, para. 0005 “…machine learning may enable large-scale statistical approaches and automated characterization of signal strength.”].
Per Claim 11, Iyengar-Drake discloses claim 10.
Iyengar does not fully disclose, but with Drake does teach training the second ML model to assign a class type associated with the first sample, wherein the second ML model is a classifier configured to assign the class to an analyte, wherein the training data is based on one or more samples assigned the class type, wherein training the classifier includes determining classifier boundary; and assigning the class type to the analyte in the first sample using the trained second ML to assign a class to the sample [Drake, pg. 4, para. 0020 “…the classifying of the biological sample is performed by a classifier trained and constructed according to one or more of: linear discriminant analysis (LDA); partial least squares (PLS); random forest; k-nearest neighbor (KNN): support vector machine (SVM) with radial basis function kernel (SVMRadial)…”. (note: training a classifier (LDA and others listed) includes determining a classifier boundary. The trained classifier assigns a class to the analyte in the sample.); pg. 5, para. 0022 “a comparator module to compare the output labels to the known labels of the training samples,”. (note: comparing output labels to known labels is determining the classifier boundary)].
Iyengar and Drake are analogous art because they are from the same field of endeavor of electrochemical biosensors for and machine learning applied to biological sample characterization and analyte classification and quantification.
Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to use the electrochemical measurement and Fourier-coefficient feature extraction as the front end, sending the result into Drake’s ML model selection and training pipeline as the back end.
The suggestion/motivation for doing so is improved flexibility, accuracy, and adaptability to different analyte classification and quantifications, which is explicitly stated by Drake. [Drake, pg. 2, para. 0005 “…machine learning may enable large-scale statistical approaches and automated characterization of signal strength.”].
Per Claim 12, Iyengar-Drake discloses claim 10.
Iyengar further teaches defining calibration analyte samples; analyzing the calibration analyte samples [Iyengar, column 3, line 16 “…applying the waveform to samples containing multiple different concentrations of each of the select and interfering analytes alone, and measuring the resulting reference current signals”, (note: this shows both the defining and analyzing steps. “…samples containing multiple different concentrations of each of the select and interfering analytes along…” refers to the calibration analyte samples, they are selected and prepared (defined) for the purpose of calibrating the system. “…applying the waveform to…and measuring the resulting reference current signals…” is the step of analyzing the calibration analyte samples.)];
training the second ML algorithm based on a Scattered Component Analysis (SCA) to determine a projection vector that maximizes similarity to analyte-specific reference sample data while minimizing similarity to matrix-specific reference data [Iyengar , pg. 21, line 46 “…resolving a signal contribution from the select analyte in the generated signal by a vector projection method with an analyte vector comprising a plurality of real and imaginary parts of one or more Fourier coefficients at one or more frequencies of a reference current signal for the select analyte”. (note: this is interpreted as and is analogous to the SCA. The analyte vector is made from Fourier coefficients of the reference current signal for the select analyte, it is built to be maximally aligned with the analyte-specific reference data. The vector projection method then resolves the analyte’s signal contribution by projecting the measured signal onto this analyte vector. This is the mathematical equivalent to the projection vector that maximizes similarity to analyte-specific reference sample data. The analyte vector is derived specifically from the target analyte’s reference data, making it most similar to that analyte’s signal and least similar to other signal sources.)] and/or similarity to chemically and structurally similar analyte reference data, to digitally subtract the contribution of the background and other similar analytes to the signal [Iyengar, column 5, line 30 “DSP 17 performs mathematical operations on the measured signal to mathematically filter out some or essentially all of the current signal 7 generated from interfering analyte ascorbic acid and allows the contribution from the desired analyte signal 3 to be quantified”. (note: this shows the digital subtraction step. The DSP performs mathematical operations, the vector projection, that filter out (digitally subtract) the signal contribution from the interfering analyte (ascorbic acid), which is a chemically and structurally similar analyte to the target. The result is that only the desired analyte’s signal contribution remains quantifiable. By minimizing similarity to the interfering/matrix signal components, it effectively subtracts their contribution from the total measured signal.)]; and
determining a concentration of the analyte by at least projecting, by the trained second ML algorithm, the sample data onto the projection vector [Iyengar, column 12, line 52 “…an electro chemical method of determining concentration of a select analyte in a mixed sample with an interfering analyte is provided that includes applying a large amplitude potential stimulus waveform to the sample to generate a nonlinear current signal; measuring the generated signal; computing at least one parameter of all or some portion of the generated signal; and determining a concentration of the select analyte in the mixed sample by resolving an estimation equation based on analyte vectors for each of the select and interfering analytes and the at least one parameter.” (note: “Resolving an estimation equation based on analyte vectors…and the at least one parameter” is the mathematical operation of projecting the measured signal (the parameter computed from the sample data) onto the analyte vectors (projection vectors) to determine concentration. The at least one parameter computed from the sample is the sample data being projected, and the analyte vectors are the projection vectors. Resolving the estimation equation by substituting the measured parameter into an equation built around the analyte vectors is the step of projecting the sample data onto the projection vector to determine concentration. The output of this projection operation is directly the analyte concentration value.)].
Per Claim 13, Iyengar-Drake discloses claim 1.
Iyengar does not fully disclose, but with Drake does teach determining that an ML model having the first ML model type does not exist; identifying a second sample based on a predetermined relationship with the first sample; identifying a third ML model and second training data associated with the second sample, the second training data including one or more of the second sample data, metadata associated with detection of a current measurement signal associated with the second sample and previously generated output of the third ML model; training, using a training model, the third ML model based on the second training data; and generating an output by the third ML model configured to receive the feature set and user defined metadata as an input [Drake, pg. 20, para. 0104 “…methods and systems are useful for predicting disease, treatment efficacy and guiding treatment decisions for affected individuals”. (note: when a model for one disease type does not exist, the system adapts by using related disease data (second sample from a related condition) with a related model, which is identifying a second sample based on predetermined relationship and training a third ML model on that data.); pg. 6, para. 0022 “a training module to iteratively search for optimal values of the parameters as part of training the machine learning model based on the comparing the output labels to the known labels of the training samples…”. (note: the training module trains a third ML model on second training data and generates an output from the trained model)].
Iyengar and Drake are analogous art because they are from the same field of endeavor of electrochemical biosensors for and machine learning applied to biological sample characterization and analyte classification and quantification.
Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to use the electrochemical measurement and Fourier-coefficient feature extraction as the front end, sending the result into Drake’s ML model selection and training pipeline as the back end.
The suggestion/motivation for doing so is improved flexibility, accuracy, and adaptability to different analyte classification and quantifications, which is explicitly stated by Drake. [Drake, pg. 2, para. 0005 “…machine learning may enable large-scale statistical approaches and automated characterization of signal strength.”].
Claims 14 and 16-26 are substantially similar in scope and spirit to claims 1 and 3-13, respectively. Therefore, the rejection of claim 1 and 3-13 is applied accordingly. Drake further shows the method being implemented by an apparatus [Drake, pg. 10, para. 0023 " …a system for classifying subjects based on multianalyte analysis in a biological sample composition comprising: (a) a computer-readable medium comprising a classifier operable to classify the subjects based on the multi-analyte analysis; and (b) one or more processors for executing instructions stored the computer-readable medium.”].
Claim 27 is substantially similar in scope and spirit to claim 1. Therefore, the rejection of claim 1 is applied accordingly. Drake shows the method being implemented by an article of manufacture [Drake, pg. 6, para. 0026 “Another aspect of the present disclosure provides a non-transitory computer-readable medium comprising machine-executable code that, upon execution by one or more computer processors, implements any of the methods above or elsewhere herein”].
Claim 2 is rejected under 35 U.S.C. 103 as being unpatentable over Iyengar in view of
Drake, in further view of Iyengar 2019.
Per Claim 2, Iyengar-Drake discloses claim 1.
Iyengar combined with Drake does not fully disclose, but with Iyengar 2019 does teach the metadata associated with the sensor platform includes physical properties of the sensor platform indicative of the electrochemical charge transfer at the sensor interface and/or operational properties of the sensor platform associated with detection of the current measurement signal. [Iyengar 2019, column 1, line 17 “The signal that is generated at the electrode can depend on many factors and properties of the electrochemical system. Examples of properties of the sample that affect the transport of the analyte include viscosity, temperature, density, and ionic strength”. (note: this shows that electrochemical system properties (metadata) influence the signal.); column 1, line 25 “…the properties of the electrode itself can affect the transport of the analytes and/or the kinetics of any reactions that may generate the measured electrochemical signals”. (note: this directly shows properties of the electrode, corresponding to physical properties of the sensor platform indicative of the electrochemical charge transfer)].
Iyengar, Drake, and Iyengar 2019 are analogous art because they are from the same field of endeavor of electrochemical biosensors for analyte detection and measurement in biological samples.
Before the effective filing date of the claimed invention, it would have been obvious to a
person of ordinary skill in the art to add the physical properties of the sensor platform from Iyengar2 to the generated feature vector.
The suggestion/motivation for doing so would have been improved measurement accuracy, which is explicitly stated by Iyengar 2019. [Iyengar 2019, column 3, line 54 “Thus, when electrochemical means of detection are used, the environmental factors—including the properties of the sample that contain the analyte—may heavily influence the signal that is measured. Such factors may introduce inaccuracies into the measurement, including but not limited to, change in calibration and change in sensitivity.”].
Claim 15 is substantially similar in scope and spirit to claim 2. Therefore, the rejection of claim 2 is applied accordingly. Drake further shows the method being implemented by an apparatus [Drake, pg. 10, para. 0023 " …a system for classifying subjects based on multianalyte analysis in a biological sample composition comprising: (a) a computer-readable medium comprising a classifier operable to classify the subjects based on the multi-analyte analysis; and (b) one or more processors for executing instructions stored the computer-readable medium.”].
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Sayed M Shah whose telephone number is (571)272-9406. The examiner can normally be reached Monday-Friday 6:00 am - 2:00 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, Miranda Huang can be reached at (571) 270-7092. 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.
/SAYED MUNEER SHAH/Examiner, Art Unit 2124
/MIRANDA M HUANG/Supervisory Patent Examiner, Art Unit 2124
Prosecution Insights