Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Drawings
The drawings are objected to as failing to comply with 37 CFR 1.84(p)(5) because they do not include the following reference signs mentioned in the description:
Regarding Figure 1, [0039]-[0040] of the applicant’s specification includes reference signs 120 and 125, however, they are not in Figure 1.
Corrected drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
The drawings are objected to as failing to comply with 37 CFR 1.84(p)(5) because they include the following reference characters not mentioned in the description: Regarding Figure 1, reference signs 105 and 110. Corrected drawing sheets in compliance with 37 CFR 1.121(d), or amendment to the specification to add the reference character(s) in the description in compliance with 37 CFR 1.121(b) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
The drawings are objected to as failing to comply with 37 CFR 1.84(p)(4) because reference character “660” has been used to designate both the servers and network in Figure 6. Based on Applicant’s specification para. [0088], the examiner believes the servers labeled as 660 should be labeled as 670. Corrected drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
Claim Objections
Claims 1 and 11 are objected to because of the following informalities: In line 11 of claim 1 and line 12 of claim 11, “an cost function” should read “a cost function”. Appropriate correction is required.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Independent Claims
Step 1 analysis:
Claim 1 is drawn to a method (i.e., process) and Claim 11 is drawn to a non-transitory machine-readable medium (i.e., manufacture), which are all within the four statutory categories. (Step 1 – Yes, the claim falls into one of the statutory categories).
Step 2A analysis – Prong One:
Claim 1 recites:
A method for limiting an eligible population for a randomized controlled trial, the method comprising:
generating, using a set of one or more generative models, panel data for a plurality of digital subjects, wherein the panel data for a given digital subject of the plurality of digital subjects comprises at least one pre-trial characteristic corresponding to the given digital subject, to be tracked in a virtual randomized controlled trial (RCT);
deriving a preliminary estimate for inclusion criteria used in the virtual RCT, wherein the preliminary estimate comprises at least one preliminary upper boundary and at least one preliminary lower boundary on the at least one pre-trial characteristic for the plurality of digital subjects;
combining, to create an cost function:
an inclusion function, wherein, for the given digital subject, the inclusion function approximates the preliminary estimate for the inclusion criteria as a set of one or more soft constraints; and
an interest function, wherein the interest function maps a conditional distribution of potential values for the plurality of digital subjects to an interest quantity that is a pre-determined scalar; and
updating the preliminary estimate to derive an updated estimate for the inclusion criteria, wherein:
updating the preliminary estimate comprises optimizing the cost function with respect to the preliminary estimate; and
the updated estimate comprises at least one updated upper boundary and at least one updated lower boundary.
The series of steps, including: Generating panel data, wherein the panel data comprises at least one pre-trial characteristic; deriving a preliminary estimate that comprises at least one preliminary upper boundary and lower boundary; combining, to create an cost function, an inclusion function, wherein, for the given digital subject, the inclusion function approximates the preliminary estimate for the inclusion criteria as a set of one or more soft constraints and an interest function, wherein the interest function maps a conditional distribution of potential values for the plurality of digital subjects to an interest quantity that is a pre-determined scalar; and updating the preliminary estimate to derive an updated estimate for the inclusion criteria, wherein updating the preliminary estimate comprises optimizing the cost function with respect to the preliminary estimate and the updated estimate comprises at least one updated upper boundary and at least one updated lower boundary, falls within the “mental processes” grouping of abstract ideas, and describes concepts that can be performed in the human mind through observation, evaluation, judgement, and opinion. Therefore, the claim recites an abstract idea of a mental process.
Claim 11 recites/describes nearly identical steps as claim 1 (and therefore also recites limitations that fall within this subject matter grouping of abstract ideas), and these claims are therefore determined to recite an abstract idea under the same analysis.
Step 2A analysis – Prong 2:
This judicial exception is not integrated into a practical application. Specifically, independent claims 1 and 11 recite the following additional elements beyond the abstract idea: using a set of one or more generative models, a plurality of digital subjects, virtual randomized controlled trial (RCT), a non-transitory machine-readable medium, and a processor. These limitations are recited at a high level of generality and amount to no more than mere instructions to apply the exception using generic computer components. The limitations do not impose any meaningful limits on practicing the abstract idea, and therefore do not integrate the abstract idea into a practical application (see MPEP 2106.05(f)).
Specifically, generative models can include (but are not limited to) traditional statistical models, generative adversarial networks, recurrent neural networks, Gaussian processes, autoencoders, autoregressive models, variational autoencoders, and/or other types of probabilistic generative models (see specification [0061]), which are known algorithms. In addition, the use of the models to carry out the abstract idea amounts to using a mathematical algorithm to apply the abstract idea, which amounts to mere instructions to apply the exception, as per MPEP 2106.05(f)(2). The processor can include (but is not limited to) a processor, microprocessor, controller, or a combination of processors, microprocessor, and/or controllers that performs instructions stored in the memory to manipulate data stored in the memory (see specification [0093]).
The additional elements do not show an improvement to the functioning of a computer or to any other technology, rather the additional elements perform general computing functions and do not indicate how the particular combination improves any technology or provides a technical solution to a technical problem. Accordingly, these additional elements, when considered separately and as an ordered combination, do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. Therefore, Claims 1 and 11 are directed to an abstract idea without practical application. (Step 2A – Prong 2: No, the additional elements are not integrated into a practical application).
Step 2B analysis:
As discussed above in “Step 2A analysis – Prong 2”, the identified additional elements in Independent Claims 1 and 11 are equivalent to adding the words “apply it” on a generic computer, and/or generally link the use of the judicial exception to a particular technological environment or field of use. Therefore, the claims as a whole do not amount to significantly more than the judicial exception itself.
For the role of a computer in a computer implemented invention to be deemed meaningful in the context of this analysis, it must involve more than performance of “well- understood, routine, [and] conventional activities previously known to the industry.” Further, “the mere recitation of a generic computer cannot transform a patent ineligible abstract idea into a patent-eligible invention.”
The applicant’s specification discloses: generative models can include (but are not limited to) traditional statistical models, generative adversarial networks, recurrent neural networks, Gaussian processes, autoencoders, autoregressive models, variational autoencoders, and/or other types of probabilistic generative models (see specification [0061]). The processor can include (but is not limited to) a processor, microprocessor, controller, or a combination of processors, microprocessor, and/or controllers that performs instructions stored in the memory to manipulate data stored in the memory (see specification [0093]).
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because, as discussed above with respect to integration of the abstract idea into a practical application, using the additional elements to perform the steps for optimizing clinical trials amount to no more than using computer related devices to implement the abstract idea.
The use of a computer or processor to merely automate or implement the abstract idea cannot provide significantly more than the abstract idea itself. (See MPEP 2106.05(f) where mere instructions to apply an exception does not render an abstract idea patent eligible). There is no indication that the additional limitations alone or in combination improves the functioning of a computer or any other technology, improves another technology or technical field, or effects a transformation or reduction of a particular article to a different state or thing. Therefore, the claims are not patent eligible. The Examiner has therefore determined that no additional element, or combination of additional claims elements is/are sufficient to ensure the claims amount to significantly more than the abstract idea identified above (Step 2B: Independent claims - NO).
Dependent Claims
Dependent Claims 2-10 and 12-20 are directed towards elements used to describe the generative model and the functions. These elements include: (claim 2 and 12) wherein a given generative model is a neural network trained on a set of historical data; (claim 3 and 13)wherein the inclusion function comprises a sigmoid function and a scalar temperature value; (claim 4 and 14) deriving a gradient of the cost function, (claim 5 and 15) wherein deriving a gradient of the cost function with respect to the inclusion criteria is done using rejection sampling; (claim 6 and 16) wherein the clinical subjects are selected according to the updated estimate for the inclusion criteria; (claim 7 and 17) wherein the inclusion function is differentiable with respect to the at least one preliminary upper boundary and the at least one preliminary lower boundary; (claim 8 and 18) wherein the cost function parameterizes a tradeoff between optimizing an eligible population size for the RCT and optimizing interest quantity; (claim 9 and 19) evaluates a treatment effect applied to an eligible population; and (claim 10 and 20) wherein the interest quantity is a function of at least one of an average value for a treatment effect, a variability value for the treatment effect, or a step function.
The elements as recited in claims 3-10 and 13-20, fall within the “mental processes” grouping of abstract Ideas as previously set forth for independent claims 1 and 11, and describes concepts that can be performed in the human mind through observation, evaluation, judgement, and opinion. All tasks recited in claims 3-10 and 13-20 can be performed in the human mind. Therefore, the dependent claims recite an abstract idea of a mental process.
The elements recited in claims 3-5, 7-8, 10, 13-15, 17-18, and 20 above also fall within the mathematical concepts grouping of abstract ideas. The following mathematical concepts are recited in the dependent claims: a sigmoid function and a scalar temperature value, deriving a gradient of the cost function using rejection sampling, the inclusion function is differentiable, the cost function parameterizes a tradeoff between optimizing an eligible population size and optimizing interest quantity, and the interest quantity is a function of at least one of an average value. These limitations merely further specify or limit the elements of the independent claims with some mathematical concepts and calculations, and therefore dependent claims 3-5, 7-8, 10, 13-15, 17-18, and 20 are directed to the mathematical concepts grouping of abstract ideas.
This judicial exception is not integrated into a practical application. Specifically, dependent claims 4 and 14 recite the following additional element beyond the abstract idea: using a reinforcement learning gradient algorithm. This limitation amounts to applying the algorithm to the abstract idea which amounts to mere instructions to apply the exception, and therefore do not integrate the abstract idea into a practical application (see MPEP 2106.05(f)(2)).
The additional elements do not show an improvement to the functioning of a computer or to
any other technology, rather the additional elements perform general computing functions and do not
indicate how the particular combination improves any technology or provides a technical solution to a
technical problem. Accordingly, these additional elements, when considered separately and as an
ordered combination, do not integrate the abstract idea into a practical application because they do not
impose any meaningful limits on practicing the abstract idea. Therefore, the dependent claims are
directed to an abstract idea without practical application. (Step 2A – Prong 2: No, the additional
elements are not integrated into a practical application).
The use of a computer or processor to merely automate or implement the abstract idea cannot
provide significantly more than the abstract idea itself. (See MPEP 2106.05(f) where mere instructions to apply an exception does not render an abstract idea patent eligible). There is no indication that the
additional limitations alone or in combination improves the functioning of a computer or any other
technology, improves another technology or technical field, or effects a transformation or reduction of a
particular article to a different state or thing. Therefore, the claims are not patent eligible.
The Examiner has therefore determined that no additional element, or combination of
additional claims elements is/are sufficient to ensure the claims amount to significantly more than the
abstract idea identified above (Step 2B: Dependent claims - NO).
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-2, 6, 9-12, 16, and 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over Das et al. (US 2022/0084633 A1) (hereinafter Das) in view of Brooks et al. (US 2022/0180979 A1) (Hereinafter Brooks).
Regarding Claim 1, Das teaches the following:
A method for limiting an eligible population for [clinical trials] ([0006] methods and systems in clinical trial management systems, and more specifically, using machine learning techniques to select candidate patients meet the clinical trial inclusion criteria and who are statistically likely to meet at least one of the one or more clinical trial endpoints), the method comprising:
generating, using a set of one or more generative models, panel data for a plurality of digital subjects ([0065], [0067], [0070] generating a training dataset from a corpus of clinical trial specifications. all measurements relevant to a prediction or classification task are measured frequently, at a standard interval, e.g. one measurement every hour. Supervised machine learning algorithms require labeled training data to identify the patterns in the data from which labels can be inferred), wherein the panel data for a given digital subject of the plurality of digital subjects comprises at least one pre-trial characteristic corresponding to the given digital subject ([0009]-[0010] the clinical variable data from the first plurality of patients includes information relating to at least one of a patient's a vital sign, heartrate, blood pressure, body temperature, electrocardiogram, electroencephalogram, pharmacokinetics, pharmacodynamics, toxicology, histology, cytometry, cytology, disease or condition stage, disease etiology, genetic profile, weight, age, gender, diet information, lifestyle, metabolic rate, patient demographic, measurements of vital signs, physiological monitor data, blood chemistry profile, the ward in which the patient stayed, diagnosis information, treatment information, lab test results, medication data, patient outcome information, clinical notes, proteomic profile, microbiome profile, imaging information, and patient medical history.),
deriving a preliminary estimate for inclusion criteria ([0007], [0054], [0077] determining whether a patient is eligible for participation in a clinical trial and who is likely to meet one or more clinical trial endpoints. a software-based prediction tool intended to provide advance notice how a candidate patient is likely to respond to an experimental therapy. Further, such information could be inferred and used to determine optimum dosage, optimum stage of disease that would produce the greatest drug benefit, pre-conditioning of the patient, etc. Inferring this information could incorporate knowledge of a patient's age, weight, and dietary factors, which may affect drug metabolism.), wherein the preliminary estimate comprises [a threshold] on the at least one pre-trial characteristic for the plurality of digital subjects ([0080], [0083] By placing a threshold on the score, e.g. patients whose scores are above 10 are determined to have sepsis; and those with scores below 10 do not, an algorithm can ultimately make a prediction. Generate as an output a score or confidence level that may be used to determine if a particular individual may be omitted from a clinical trial or whether the individual may be an appropriate candidate for the clinical trial (e.g., based on comparison of the output to a predetermined threshold level);
combining, to create an cost function ([0087] one or more loss functions may be used to measure the accuracy of the model.):
an inclusion function, wherein, for the given digital subject, the inclusion function approximates the preliminary estimate for the inclusion criteria as a set of one or more soft constraints ([0080] a weighting of the features which can then be used to make predictions on new examples, subject to the features first being constructed from the new data. For classifying a patient, the weighted features can be combined and will often lead to a numerical score that reflects the extent to which a given patient is believed to belong to a particular class. By placing a threshold on the score, e.g. patients whose scores are above 10 are determined to have sepsis; and those with scores below 10 do not, an algorithm can ultimately make a prediction.); and
an interest function, wherein the interest function maps a conditional distribution of potential values for the plurality of digital subjects to an interest quantity that is a pre-determined scalar (Claim 1, [0073]: maps the clinical variable data to one or more trajectories across different timepoints within a plurality of multidimensional spaces and further comprises associations between locations within the plurality of multidimensional spaces and predicted patient outcomes. Multiple different finite discrete hyperdimensional spaces (FDHS) and associated classification mechanisms can be defined for evaluation of a single condition. The multiple outcomes can then be aggregated into a single result, such as by averaging. In some embodiments, multiple different conditions can be mapped within a single FDHS, such that during evaluation, results for each condition can be identified by referencing a current patient descriptor within a single FDHS. The examiner is interpreting the finite discrete hyperdimensional spaces as functions.); and
updating the preliminary estimate to derive an updated estimate for the inclusion criteria ([0056], [0093] Learning refines the network parameters, and, by extension, the connections or weight factors associated with the connections between neurons in the network, such that the neural network behaves in a desired manner, such as by producing accurate predictions of drug response in a candidate patient. When the system detects that a candidate patient is displaying physiological signals consistent with a successful experimental drug therapy, the candidate patient’s caregiver can notify that the candidate patient is suitable for inclusion in the clinical trial.), wherein:
updating the preliminary estimate comprises optimizing the cost function ([0072], [0090] Computational optimization processes can be applied to the mapped descriptors in order to develop a classification mechanism, such as an association between location within the FDHS and patient outcome. As the disclosed algorithms run and patients are treated, more data is generated, Another training technique utilized is called online learning. Online learning allows algorithms to continually improve themselves as new data become available. In this context, the disclosed algorithms can learn from their own mistakes. Once a patient's outcome is known, that patient will become part of the training data and improve the algorithm's future predictions by comparing the algorithm's original prediction to the ultimate patient outcome and adjusting its future predictions for similar patients accordingly.) with respect to the preliminary estimate ([0072] The derived classification mechanism can then be applied within an evaluation environment to evaluate patient descriptors associated with new patients whose future outcome is yet to be determined.); and
However, Das does not disclose the following that is met by Brooks:
a randomized controlled trial ([0004] the system relies on active control of the environment for knowledge generation and application. Examples of active control techniques include randomized controlled experimentation.)
at least one pre-trial characteristic corresponding to the given digital subject to be tracked in a virtual randomized controlled trial (RCT) ([0004] the system relies on active control of the environment for knowledge generation and application. Examples of active control techniques include randomized controlled experimentation.);
[deriving a preliminary estimate for inclusion criteria] used in the virtual RCT ([0004] the system relies on active control of the environment for knowledge generation and application. Examples of active control techniques include randomized controlled experimentation.)
at least one preliminary upper boundary and at least one preliminary lower boundary ([0276], [0277] the set of internal parameters that define the data inclusion are stochastically varied. In some cases, the lower bound A and the upper bound B are fixed and the system adjusts the probabilities that are assigned to different values between the lower bound A and the upper bound B by updating the causal model as described above. In other cases, the system can vary the lower bound A and the upper bound B while also updating the causal model. the data inclusion window parameters include the lower bound of the range, the upper bound of the range, and the possible values that the data inclusion window can take within the range.)
the updated estimate comprises at least one updated upper boundary and at least one updated lower boundary ([0276], [0277] the system can adjust the range of possible values for the data inclusion window, including an upper and lower boundary, based on the likelihood that relative causal effects of different possible values of the controllable element are changing)
It would have been obvious to one of ordinary skill in the art before the effective filing date to have substituted the clinical trials as taught by Das with the randomized controlled trials (RCT) and include the upper and lower boundaries that can be updated, as taught by Brooks, because the randomized controlled experiments allow for more control over the system (See Brooks [0291]-[0297]), and having a system that can vary the lower and upper bounds of the range to adjust the possible data inclusion windows can prevent the system from exploring data inclusion windows that are too short or too long (See Brooks [0278]).
Regarding Claim 2, the combination of Das and Brooks teaches the method of claim 1, and Das further teaches:
The method of claim 1, wherein a given generative model of the set of one or more generative models is a neural network ([0020] In various embodiments, the machine learning system is one of a supervised learning system, an unsupervised learning system, or a reinforcement learning system, or a combination of the three systems. This includes a neural network system) trained on a set of historical data comprising at least one of control arm data from historical control arms, patient registries, electronic health records, or real world data ([0017] a machine learning system using training data, wherein the training data includes patient health record data).
Regarding Claim 6, the combination of Das and Brooks teaches the method of claim 1, and Das further teaches:
The method of claim 1, further comprising implementing the virtual RCT, wherein clinical subjects to the virtual RCT are selected from the plurality of digital subjects according to the updated estimate for the inclusion criteria ([0072]: The derived classification mechanism can then be applied within an evaluation environment to evaluate patient descriptors associated with new patients whose future outcome is yet to be determined. [0089], [0090]: The system can select a candidate patient for enrollment in a clinical trial from a plurality of candidate patients by analyzing a dataset of clinical trial inclusion criteria, clinical trial endpoints, and patient health record data via a model. The model can be trained to allow the algorithms to continually improve themselves as new data becomes available, and improve the algorithm’s future predictions.).
Regarding Claim 9, The combination of Das and Brooks teaches the method of claim 1, and Das further teaches:
The method of claim 1, wherein the [clinical trial] evaluates a treatment effect applied to an eligible population taken from the plurality of digital subjects ([0020], [0077] determine the latency between drug administration and the appearance of the drug's effects, from which the drug administration time could be inferred. The method may be used to effectively treat as many patients as possible or to maximize the effect of a drug in a patient population.).
However, Das does not disclose the following that is met by Brooks:
virtual RCT ([0004] the system relies on active control of the environment for knowledge generation and application. Examples of active control techniques include randomized controlled experimentation.)
It would have been obvious to one of ordinary skill in the art before the effective filing date to have substituted the clinical trials as taught by Das with the randomized controlled trials (RCT), as taught by Brooks because the randomized controlled experiments allow for more control over the system (See Brooks [0291]-[0297])
Regarding Claim 10, the combination of Das and Brooks teaches the method of claim 9, and Das further teaches:
The method of claim 9, wherein the interest quantity is a function of at least one of an average value for a treatment effect, a variability value for the treatment effect; or a step function ([0072]-[0073] the classification component maps patient descriptors comprising patient physiological data, each associated with one or more known outcomes, into one or more finite multidimensional spaces, such as finite discrete hyperdimensional spaces (FDHS). Multiple different finite discrete hyperdimensional spaces (FDHS) and associated classification mechanisms can be defined for evaluation of a single condition. The multiple outcomes can then be aggregated into a single result, such as by averaging.).
Regarding Claim 11, Das teaches the following:
A non-transitory machine-readable medium comprising instructions that, when executed, are configured to cause a processor to perform a process for limiting an eligible population [for a clinical trial] ([0017] a non-transitory computer readable medium comprising instructions that, when executed, causes the at least one processor to automatically select or identify a candidate patient for enrollment in a clinical trial), the process comprising:
generating, using a set of one or more generative models, panel data for a plurality of digital subjects ([0065], [0067], [0070] generating a training dataset from a corpus of clinical trial specifications. all measurements relevant to a prediction or classification task are measured frequently, at a standard interval, e.g. one measurement every hour. Supervised machine learning algorithms require labeled training data to identify the patterns in the data from which labels can be inferred), wherein the panel data for a given digital subject of the plurality of digital subjects comprises at least one pre-trial characteristic corresponding to the given digital subject, to be tracked in [a clinical trial] ([0009]-[0010] the clinical variable data from the first plurality of patients includes information relating to at least one of a patient's a vital sign, heartrate, blood pressure, body temperature, electrocardiogram, electroencephalogram, pharmacokinetics, pharmacodynamics, toxicology, histology, cytometry, cytology, disease or condition stage, disease etiology, genetic profile, weight, age, gender, diet information, lifestyle, metabolic rate, patient demographic, measurements of vital signs, physiological monitor data, blood chemistry profile, the ward in which the patient stayed, diagnosis information, treatment information, lab test results, medication data, patient outcome information, clinical notes, proteomic profile, microbiome profile, imaging information, and patient medical history.);
deriving a preliminary estimate for inclusion criteria ([0007], [0054], [0077] determining whether a patient is eligible for participation in a clinical trial and who is likely to meet one or more clinical trial endpoints. a software-based prediction tool intended to provide advance notice how a candidate patient is likely to respond to an experimental therapy. Further, such information could be inferred and used to determine optimum dosage, optimum stage of disease that would produce the greatest drug benefit, pre-conditioning of the patient, etc. Inferring this information could incorporate knowledge of a patient's age, weight, and dietary factors, which may affect drug metabolism.), wherein the preliminary estimate comprises [a threshold] on the at least one pre-trial characteristic for the plurality of digital subjects ([0080], [0083] By placing a threshold on the score, e.g. patients whose scores are above 10 are determined to have sepsis; and those with scores below 10 do not, an algorithm can ultimately make a prediction. Generate as an output a score or confidence level that may be used to determine if a particular individual may be omitted from a clinical trial or whether the individual may be an appropriate candidate for the clinical trial (e.g., based on comparison of the output to a predetermined threshold level);
combining, to create an cost function ([0087] one or more loss functions may be used to measure the accuracy of the model.):
an inclusion function, wherein, for the given digital subject, the inclusion function approximates the preliminary estimate for the inclusion criteria as a set of one or more soft constraints ([0080] a weighting of the features which can then be used to make predictions on new examples, subject to the features first being constructed from the new data. For classifying a patient, the weighted features can be combined and will often lead to a numerical score that reflects the extent to which a given patient is believed to belong to a particular class. By placing a threshold on the score, e.g. patients whose scores are above 10 are determined to have sepsis; and those with scores below 10 do not, an algorithm can ultimately make a prediction.); and
an interest function, wherein the interest function maps a conditional distribution of potential values for the plurality of digital subjects to an interest quantity that is a pre-determined scalar (Claim 1, [0073]: maps the clinical variable data to one or more trajectories across different timepoints within a plurality of multidimensional spaces and further comprises associations between locations within the plurality of multidimensional spaces and predicted patient outcomes. Multiple different finite discrete hyperdimensional spaces (FDHS) and associated classification mechanisms can be defined for evaluation of a single condition. The multiple outcomes can then be aggregated into a single result, such as by averaging. In some embodiments, multiple different conditions can be mapped within a single FDHS, such that during evaluation, results for each condition can be identified by referencing a current patient descriptor within a single FDHS. The examiner is interpreting the finite discrete hyperdimensional spaces as functions.); and
updating the preliminary estimate to derive an updated estimate for the inclusion criteria ([0056], [0093] Learning refines the network parameters, and, by extension, the connections or weight factors associated with the connections between neurons in the network, such that the neural network behaves in a desired manner, such as by producing accurate predictions of drug response in a candidate patient. When the system detects that a candidate patient is displaying physiological signals consistent with a successful experimental drug therapy, the candidate patient’s caregiver can notify that the candidate patient is suitable for inclusion in the clinical trial.), wherein:
updating the preliminary estimate comprises optimizing the cost function ([0072], [0090] Computational optimization processes can be applied to the mapped descriptors in order to develop a classification mechanism, such as an association between location within the FDHS and patient outcome. As the disclosed algorithms run and patients are treated, more data is generated, Another training technique utilized is called online learning. Online learning allows algorithms to continually improve themselves as new data become available. In this context, the disclosed algorithms can learn from their own mistakes. Once a patient's outcome is known, that patient will become part of the training data and improve the algorithm's future predictions by comparing the algorithm's original prediction to the ultimate patient outcome and adjusting its future predictions for similar patients accordingly.) with respect to the preliminary estimate ([0072] The derived classification mechanism can then be applied within an evaluation environment to evaluate patient descriptors associated with new patients whose future outcome is yet to be determined.); and
However, Das does not disclose the following that is met by Brooks:
a randomized controlled trial ([0004] the system relies on active control of the environment for knowledge generation and application. Examples of active control techniques include randomized controlled experimentation.)
at least one pre-trial characteristic corresponding to the given digital subject to be tracked in a virtual randomized controlled trial (RCT) ([0004] the system relies on active control of the environment for knowledge generation and application. Examples of active control techniques include randomized controlled experimentation.);
[deriving a preliminary estimate for inclusion criteria] used in the virtual RCT ([0004] the system relies on active control of the environment for knowledge generation and application. Examples of active control techniques include randomized controlled experimentation.)
at least one preliminary upper boundary and at least one preliminary lower boundary ([0276], [0277] the set of internal parameters that define the data inclusion are stochastically varied. In some cases, the lower bound A and the upper bound B are fixed and the system adjusts the probabilities that are assigned to different values between the lower bound A and the upper bound B by updating the causal model as described above. In other cases, the system can vary the lower bound A and the upper bound B while also updating the causal model. the data inclusion window parameters include the lower bound of the range, the upper bound of the range, and the possible values that the data inclusion window can take within the range.)
the updated estimate comprises at least one updated upper boundary and at least one updated lower boundary ([0276], [0277] the system can adjust the range of possible values for the data inclusion window, including an upper and lower boundary, based on the likelihood that relative causal effects of different possible values of the controllable element are changing)
It would have been obvious to one of ordinary skill in the art before the effective filing date to have substituted the clinical trials as taught by Das with the randomized controlled trials (RCT) and include the upper and lower boundaries that can be updated, as taught by Brooks, because the randomized controlled experiments allow for more control over the system (See Brooks [0291]-[0297]), and having a system that can vary the lower and upper bounds of the range to adjust the possible data inclusion windows can prevent the system from exploring data inclusion windows that are too short or too long (See Brooks [0278]).
Regarding Claim 12, the combination of Das and Brooks teaches the non-transitory machine-readable medium of claim 11, and Das further teaches:
The non-transitory machine-readable medium of claim 11, wherein a given generative model of the set of one or more generative models is a neural network ([0020] In various embodiments, the machine learning system is one of a supervised learning system, an unsupervised learning system, or a reinforcement learning system, or a combination of the three systems. This includes a neural network system) trained on a set of historical data comprising at least one of control arm data from historical control arms, patient registries, electronic health records, or real world data ([0017] a machine learning system using training data, wherein the training data includes patient health record data).
Regarding Claim 16, the combination of Das and Brooks teaches the non-transitory machine-readable medium of claim 11, and Das further teaches:
The non-transitory machine-readable medium of claim 11, further comprising implementing the virtual RCT, wherein clinical subjects to the virtual RCT are selected from the plurality of digital subjects according to the updated estimate for the inclusion criteria ([0072] The derived classification mechanism can then be applied within an evaluation environment to evaluate patient descriptors associated with new patients whose future outcome is yet to be determined. [0089], [0090]: The system can select a candidate patient for enrollment in a clinical trial from a plurality of candidate patients by analyzing a dataset of clinical trial inclusion criteria, clinical trial endpoints, and patient health record data via a model. The model can be trained to allow the algorithms to continually improve themselves as new data becomes available, and improve the algorithm’s future predictions.).
Regarding Claim 19, the combination of Das and Brooks teaches the non-transitory machine-readable medium of claim 11, and Das further teaches:
The non-transitory machine-readable medium of claim 11, wherein the [clinical trial] evaluates a treatment effect applied to an eligible population taken from the plurality of digital subjects ([0020], [0077] determine the latency between drug administration and the appearance of the drug's effects, from which the drug administration time could be inferred. The method may be used to effectively treat as many patients as possible or to maximize the effect of a drug in a patient population.).
However, Das does not disclose the following that is met by Brooks:
virtual RCT ([0004] the system relies on active control of the environment for knowledge generation and application. Examples of active control techniques include randomized controlled experimentation.)
It would have been obvious to one of ordinary skill in the art before the effective filing date to have substituted the clinical trials as taught by Das with the randomized controlled trials (RCT), as taught by Brooks because the randomized controlled experiments allow for more control over the system (See Brooks [0291]-[0297]).
Regarding Claim 20, the combination of Das and Brooks teaches the non-transitory machine-readable medium of claim 19, and Das further teaches:
The non-transitory machine-readable medium of claim 19, wherein the interest quantity is a function of at least one of an average value for a treatment effect, a variability value for the treatment effect; or a step function ([0072]-[0073] the classification component maps patient descriptors comprising patient physiological data, each associated with one or more known outcomes, into one or more finite multidimensional spaces, such as finite discrete hyperdimensional spaces (FDHS). Multiple different finite discrete hyperdimensional spaces (FDHS) and associated classification mechanisms can be defined for evaluation of a single condition. The multiple outcomes can then be aggregated into a single result, such as by averaging.).
Claims 4 and 14 are rejected under 35 U.S.C. 103 as being unpatentable over Das et al. (US 2022/0084633 A1) (hereinafter Das), in view of Brooks et al. (US 2022/0180979 A1) (Hereinafter Brooks), in further view of Manrai et al. (US 2021/0225513 A1) (Hereinafter Manrai).
Regarding Claim 4, the combination of Das and Brooks teaches the method of claim 1, however, Das and Brooks do not teach the following that is met by Manrai:
The method of claim 1, wherein optimizing the cost function with respect to the preliminary estimate comprises deriving a gradient of the cost function with respect to the inclusion criteria using a reinforcement learning gradient algorithm ([0163]-[0164] the convolutional neural network uses a stochastic gradient descent (SGD) to calculate
the cost function. An SGD approximates the gradient with respect to the weights in the loss function. the convolutional neural network uses different loss functions such as Euclidean loss and softmax loss. In a further implementation, an ADAM stochastic optimizer is used by the convolutional neural network.).
It would have been obvious to one of ordinary skill in the art to have combined the method including creating a cost function, as taught by Das, with the gradient of the cost function created by a machine learning gradient algorithm (i.e., SGD) because it allows for a prediction error to be calculated and allows for the weights or coefficients of the model/function to be adjusted so the prediction error is reduced (See Manrai [0134]).
Regarding Claim 14, the combination of Das and Brooks teaches the non-transitory machine-readable medium of claim 11, however, Das and Brooks do not teach the following that is met by Manrai:
The non-transitory machine-readable medium of claim 11, wherein optimizing the cost function with respect to the preliminary estimate comprises deriving a gradient of the cost function with respect to the inclusion criteria using a reinforcement learning gradient algorithm ([0163]-[0164] the convolutional neural network uses a stochastic gradient descent (SGD) to calculate the cost function. An SGD approximates the gradient with respect to the weights in the loss function. the convolutional neural network uses different loss functions such as Euclidean loss and softmax loss. In a further implementation, an ADAM stochastic optimizer is used by the convolutional neural network.).
It would have been obvious to one of ordinary skill in the art to have combined the method including creating a cost function, as taught by Das, with the gradient of the cost function created by a machine learning gradient algorithm (i.e., SGD) because it allows for a prediction error to be calculated and allows for the weights or coefficients of the model/function to be adjusted so the prediction error is reduced (See Manrai [0134]).
Claims 3 and 13 are rejected under 35 U.S.C. 103 as being unpatentable over Das et al. (US 2022/0084633 A1) (hereinafter Das) in view of Brooks et al. (US 2022/0180979 A1) (Hereinafter Brooks), in further view of Manrai et al. (US 2021/0225513 A1) (Hereinafter Manrai), in further view of NPL Ferrell: Stability and Approximator Convergence in Nonparametric Nonlinear Adaptive Control.
Regarding Claim 3, The combination of Das and Brooks teaches the method of claim 1, however, Das and Brooks do not teach the following that is met by Manrai:
The method of claim 1, wherein the inclusion function comprises:
a sigmoid function ([0153]-[0154]: the neural network includes a sigmoid function); and
It would have been obvious to one of ordinary skill in the art before the effective filing date to have combined the method as taught by Das and Brooks with the sigmoid function as taught by Manrai because it allows for the neural network to compute errors across its layers, thus optimizing the function (See Manrai [0153]-[0161]).
However, Das, Brooks, and Manrai do not disclose the following that is met by Farrell:
a scalar temperature value, where the scalar temperature value is pre-determined to control sharpness of the inclusion criteria (Farrell pg. 8, Col. 2, par. 1-3 discloses estimating parameters using an iterative approach cannot guarantee convergence of the estimate unless the function is defined by some n > 0 and for all k > 0. Thus, the parameter estimation routing requires the scalar value n > 0 to control the convergence of the estimate).
It would have been obvious to one of ordinary skill in the art before the effective filing date to have modified the combination of Das, Brooks, and Manrai to include the scalar value which controls sharpness of the inclusion criteria, as taught by Farrell, because it guarantees the convergence of the estimate, and when the data is noisy, the accuracy is dependent on the addition of the scalar and regressor covariance matrix (See Farrell Pg. 8, section A: Thought Example).
Regarding Claim 13, the combination of Das and Brooks teaches the non-transitory machine-readable medium of claim 11, however, Das and Brooks do not teach the following that is met by Manrai:
The non-transitory machine-readable medium of claim 11, wherein the inclusion function comprises:
a sigmoid function ([0153]-[0154]: the neural network includes a sigmoid function); and
It would have been obvious to one of ordinary skill in the art before the effective filing date to have combined the method as taught by Das and Brooks with the sigmoid function as taught by Manrai because it allows for the neural network to compute errors across its layers, thus optimizing the function (See Manrai [0153]-[0161]).
However, Das, Brooks, and Manrai do not disclose the following that is met by Farrell:
a scalar temperature value, where the scalar temperature value is pre-determined to control sharpness of the inclusion criteria (Farrell pg. 8, Col. 2, par. 1-3 discloses estimating parameters using an iterative approach cannot guarantee convergence of the estimate unless the function is defined by some n > 0 and for all k > 0. Thus, the parameter estimation routing requires the scalar value n > 0 to control the convergence of the estimate).
It would have been obvious to one of ordinary skill in the art before the effective filing date to have modified the combination of Das, Brooks, and Manrai to include the scalar value which controls sharpness of the inclusion criteria, as taught by Farrell, because it guarantees the convergence of the estimate, and when the data is noisy, the accuracy is dependent on the addition of the scalar and regressor covariance matrix (See Farrell Pg. 8, section A: Thought Example).
Subject Matter Free of the Prior Art
The following is an examiner’s statement of subject matter free of the prior art: The limitations in Claims 5, 7-8, 15, and 17-18 stating the following is free of the prior art:
(claim 5 and 15) wherein deriving a gradient of the cost function with respect to the inclusion criteria is done, in part, using rejection sampling based on the inclusion function.
(claim 7 and 17) wherein the inclusion function is differentiable with respect to the at least one preliminary upper boundary and the at least one preliminary lower boundary.
(claim 8 and 18) wherein the cost function parameterizes a tradeoff between optimizing an eligible population size for the virtual RCT; and optimizing the interest quantity.
Relevant Prior Art of Record Not Currently Being Applied
The relevant art made of record and not relied upon is considered pertinent to applicant’s disclosure.
Jain et al. (US patent no. 11,328,796 B1) describes receiving data indicating selection criteria for a cohort and determining, using machine learning, a set of candidates classified as having attributes that satisfy the selection criteria.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ALEXIS K VAN DUZER whose telephone number is (571)270-5832. The examiner can normally be reached Monday thru Thursday 8-5 CT.
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, Fonya Long can be reached at (571) 270-5096. 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.
/A.K.V./Examiner, Art Unit 3682
/EVANGELINE BARR/Primary Examiner, Art Unit 3682