Detailed Action
Information Disclosure
The information disclosure statements (IDS) submitted on 04/27/2026 are in accordance with the provisions of 37 CFR 1.97 and are considered by the Examiner.
Request for Continuation Examination
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 04/27/2026 has been entered.
Status of the Claims
The status of the claims as of the response filled 04/27/2026: Claims 1-20 were initially pending.
Claim 1, 11 and 20 were amended and have been considered below.
Response to Arguments 35 U.S.C. § 101
Applicant’s arguments with respect to the rejection of claims 1-20 under 35 U.S.C. § 101 page 11-12, filed on 04/27/2026, have been fully considered but are not persuasive for the reasons below:
Applicant argues that claim 20, “the patient-specific metabolic model comprises a plurality of organ-specific metabolic models,” “each organ-specific metabolic model is trained,” and “each organ-specific metabolic model communicates with at least one other organ-specific metabolic model,” is no longer directed to a mental process because the amended claim recites a specific multi-component computational architecture that a clinician could not perform mentally.
The Examiner respectfully disagrees because claim 20 still recites the abstract treatment-evaluation workflow of identifying an intervention parameter, generating a sensitive cohort based on correlations, separating control and test cohorts, predicting treatment effects, determining a representative effect, and comparing that effect to control-cohort metabolic states. These are information-collection, evaluation, prediction, grouping, and comparison steps. MPEP § 2106 requires the claim to be construed under BRI before eligibility is evaluated, because “the BRI sets the boundaries of the coverage sought by the claim.” Under that BRI, the amended model language does not remove the judicial exception from the claim; it is an additional element considered under Step 2A, Prong Two and Step 2B. Therefore, Applicant’s argument is not persuasive, and the rejection is maintained.
Applicant argues that the organ-specific models are "a specific technical improvement over conventional single-output prediction models because [the architecture] disaggregates the complex, multi-dimensional nature of human metabolism into organ-level components, each trained on the biosignals most relevant to that organ system."
The Examiner respectfully disagrees. The claim requires only a plurality of organ-specific metabolic models trained to predict effects on distinct organ systems. The specification supports digital-twin modeling, but it describes the patient’s digital twin broadly as “a digital model capturing the biology and metabolism of the patient's body” and explains that it generates predictions from relationships between health factors and metabolic health. The specification also states that the digital twin module implements “one or more machine-learned, metabolic models” and that the prediction is a function of many metabolic factors, but it does not define a claimed improvement to computer architecture, processor operation, memory operation, or network communication. Thus, the claimed improvement is directed to the medical prediction content, not to the functioning of a computer or another technology. Applicant’s argument is not persuasive, and the rejection is maintained.
Applicant argues that claim 20, “each organ-specific metabolic model communicates with at least one other organ-specific metabolic model to capture cross-organ interaction effects,” is a non-abstract technical improvement because interconnected organ-specific models capture physiological dependencies.
The Examiner respectfully disagrees. The specification states that “each module of the health twin module 610 is connected to and communicates with other modules (par. 0086)” to capture complex interaction effects, including blood-pressure dynamics driven by glucose dynamics, heart function, nutrition, exercise, and sleep trends. That disclosure supports the existence of communicating modules, but the claim recites the communication at a result-oriented level. The claim does not require a communication protocol, data structure, dependency graph, propagation rule, weighting technique, iterative convergence condition, or particular algorithm for capturing cross-organ effects. Under MPEP § 2106.05(f), merely applying an abstract idea with computer implementation does not create a practical application where the claim recites the outcome without the technical details of how the solution is accomplished. Therefore, Applicant’s argument is not persuasive, and the rejection is maintained.
Applicant argues that claims 1-20 integrate the alleged abstract idea into a practical application because the claimed simulator generates individual predicted effects, aggregates those effects into a representative effect, compares the representative effect to control-cohort metabolic states, and identifies candidate treatment recommendations for physical validation.
The Examiner respectfully disagrees. The specification confirms that the simulator identifies sensitive patients, inputs candidate treatment recommendations into patient-specific metabolic models, predicts efficacy, identifies a shortlist, and may define instructions for later physical experiments. Those operations produce analytical outputs: predicted effects, representative effects, comparisons, confidence measures, and shortlisted recommendations. The claim does not require administering a treatment, changing a dosage, operating a medical device, transforming a biological sample, or causing a physical change in the patient. MPEP § 2106.04(d) requires a meaningful limit that applies the exception in a practical application, not merely a computer implementation of the abstract analysis. MPEP form paragraph guidance likewise requires the Examiner to explain why the judicial exception is not integrated into a practical application and why the additional elements do not amount to significantly more. Here, the additional model and computer elements are used to automate predictive clinical-trial analysis. Applicant’s argument is not persuasive, and the rejection is maintained.
Applicant argues that claim 20 recites significantly more than the abstract idea because the amended organ-specific model architecture is not a generic mathematical tool.
The Examiner respectfully disagrees. Under Step 2B, the claim does not add significantly more because the amended limitation is recited functionally and broadly. The claim requires organ-specific models that are trained and communicate to capture cross-organ effects, but it does not require a specific unconventional model architecture, training technique, optimization method, communication protocol, physiological dependency graph, or propagation algorithm. The specification describes generic computing components, including a chipset, processor, volatile memory, network adapter, input-output device, storage device, and display, and states that the storage device may be a hard drive, CD-ROM, DVD, or solid-state memory device (par. 0042-0043). The specification further states that program modules may be stored, loaded into memory, and executed by the processor. These disclosures support the Examiner’s finding that the processor and non-transitory computer-readable medium are generic computer components used for their ordinary functions. The amended model language therefore does not supply an inventive concept, and the rejection is maintained. Refer to eligibility subject matter below for further details.
Applicant argues that the claims solve the practical problem of validating metabolic treatment recommendations without the time, cost, and patient risk of conventional clinical trials.
The Examiner respectfully disagrees. The specification identifies that conventional clinical trials are expensive, time-intensive, labor-intensive, and risky for patients, and that the digital-twin simulator can test scenarios with savings in time, expense, and labor. That stated benefit explains why the analysis is useful, but usefulness alone does not establish eligibility. MPEP § 2106 explains that a claim within a statutory category must still not be directed to a judicial exception unless the claim as a whole adds significantly more than the exception. The claim does not recite a specific technical improvement that produces the asserted savings; it recites the use of patient data, digital-twin models, cohorting, prediction, aggregation, and comparison to identify treatment candidates. Therefore, the asserted reduction in clinical-trial cost and risk is an intended benefit of the abstract simulation, not a claimed technological improvement. Applicant’s argument is not persuasive, and the rejection is maintained.
Response to Arguments 35 U.S.C. § 103
Applicant’s arguments with respect to the rejection of claims 1-20 under 35 U.S.C. § 103 page 13-15, filed on 04/27/2026, have been fully considered but is smoot for the reasons below:
Applicant argues that Claim 1, “the patient-specific metabolic model comprises a plurality of organ-specific metabolic models, each organ-specific metabolic model is trained to predict an effect of the intervention parameter on a distinct organ system of the patient, and each organ-specific metabolic model communicates with at least one other organ-specific metabolic model to capture cross-organ interaction effects on the metabolic state of the patient,” is not taught by Schaeffer and Constantin because Schaeffer predicts cohort treatment response and survival, while Constantin does not disclose a plurality of communicating organ-specific metabolic models.
The Examiner respectfully agreed as Constantin does not supply the missing architecture, but it no longer reaches the actual ground now relying on Thiele for that limitation. Therefore, the argument is moot.
Claim Rejections — 35 U.S.C. § 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 a judicial exception (an abstract idea) without integrating the abstract idea into a practical application, and the claims do not amount to significantly more than the abstract idea itself.
Step 1: Statutory Categories of Invention
Independent claims 1, 11, and 20 fall into the statutory categories of "process," "manufacture," and "machine," respectively. Dependent claims 2-10 and 12-19 further define these categories. Accordingly, the claims satisfy the requirements of Step 1 of the eligibility analysis.
Prong One: Does the claim recite a judicial exception?
Yes. Independent claims 1, 11, and 20 recite an abstract idea. The claims recite mental processes because they collect and evaluate patient information to predict treatment effects, including identifying intervention parameters, forming patient cohorts, predicting metabolic effects, and comparing predicted and observed metabolic states. These acts are observations, evaluations, judgments, and opinions under MPEP § 2106.04(a)(2). The amended model-communication language further recites organizing human activity because the organ-specific models coordinate organ-specific assessments to evaluate cross-organ effects, analogous to coordination among clinicians or specialists when assessing how a treatment affects different organ systems. Thus, the claims recite both a mental process and a certain method of organizing human activity.
The amended independent claim 20 recite:
Claim 20.
A system comprising: one or more processors;
and a non-transitory computer readable medium storing instruction encoded thereon that, when executed by the one or more processors, cause the one or more processors to:
identify an intervention parameter in a treatment recommendation for causing a target improvement in metabolic state, the treatment recommendation comprising instructions for adjusting the intervention parameter to cause the target improvement;
generate, from a population of patients, a cohort of patients sensitive to the intervention parameter based on correlations between changes in the metabolic state of each patient of the population and adjustments to the intervention parameter, the sensitivity of a patient representing a likelihood that adjustments to the intervention parameter will affect the metabolic state of the patient;
separate the cohort of patients into a control cohort comprising a first subset of patients and a test cohort comprising a second subset of patients;
and determine an effect of the treatment recommendation on the cohort of patients, the instructions for determining the effect of the treatment recommendation further comprise instructions that cause the processor to:
for each patient of the test cohort, input the treatment recommendation into a patient-specific metabolic model to generate an individual predicted effect of the treatment on the patient
wherein the patient-specific metabolic model comprises a plurality of organ- specific metabolic models
each organ-specific metabolic model is trained to predict an effect of the intervention parameter on a distinct organ system of the patient, andeach organ-specific metabolic model communicates with at least one other organ-specific metabolic model to capture cross-organ interaction effects on the metabolic state of the patient;
determine a representative effect of the treatment recommendation for the test cohort based on the generated individual predicted effects for the test cohort;
and compare the representative effect for the test cohort to representations of metabolic states in the control cohort.
Note: The above non-bold language means abstract idea, and the bold language means additional elements further evaluated in the prong two and step 2B.
Under the broadest reasonable interpretation, these claims recite steps for identifying an intervention parameter, generating a cohort based on correlations, separating into control/test groups, and comparing predicted effects all of which are fundamental concepts that can be performed in the human mind. The specification itself describes these operations in abstract, analytical terms. For example, the process involves generating a cohort "based on correlations between changes in the metabolic state ... and adjustments to the intervention parameter" (Spec., Para. [0010]), where sensitivity represents a "likelihood" of an effect (Spec., Para. [0010]). These are cognitive operations involving observation, evaluation, and judgment.
The amendment does not overcome Step 2A, Prong One. Under BRI, the added limitation recites organ-specific metabolic models trained to predict how an intervention parameter affects distinct organ systems and to communicate with other organ-specific models to capture cross-organ effects on the patient’s metabolic state. This is still the collection and evaluation of patient information to predict treatment effects. Such prediction and assessment constitute observation, evaluation, judgment, and opinion, which fall within the mental-process grouping of abstract ideas under MPEP § 2106.04(a)(2). The recited model-to-model communication also reflects coordination of organ-specific assessments, analogous to coordinating physician and specialist evaluations regarding cross-organ effects, and therefore further falls within managing relationships or interactions under the certain-methods-of-organizing-human-activity grouping. Accordingly, the added limitation does not remove the claim from the judicial-exception analysis at Step 2A, Prong One.
When the processor, computer-readable medium, and model implementation are set aside, claim 20 reads on a clinician’s mental and organizational treatment-evaluation workflow. A clinician could identify an intervention parameter by selecting a treatment variable expected to improve metabolism, generate a sensitive cohort by reviewing patient records for correlations between parameter changes and metabolic changes, separate the cohort into control and test cohorts by dividing the patients into comparison groups, determine the treatment effect by predicting how the recommendation would affect each test patient, including organ-specific and cross-organ effects, determine a representative effect by summarizing the predicted results for the test group, and compare that representative effect with the control group’s metabolic states. Thus, the claim, apart from the additional computer elements, recites mental evaluation, prediction, grouping, and comparison of patient information, and also organizes patients into clinical comparison groups for treatment assessment under MPEP § 2106.04(a)(2).
Dependent claims 2-10 and 12-19 merely refine this core mental process. Claims 2, 3, 4, 12, and 13 add steps of labeling patient sensitivity based on historical data or thresholds, which are data-sorting tasks. Claims 5 and 14 involve filtering patients based on a metabolic state threshold, another analytical step. Claims 6 and 15 consider long-term effects, which is an extension of the same predictive reasoning. Amended claim 8, for example, further limits the cohort generation to "categorizing the population... into categories... with a shared metabolic state," "determining an overall sensitivity of the category," and "generating the cohort of patients based on the category... most sensitive." These added steps—categorizing, determining sensitivity, and comparing—are all pure mental processes of data analysis. Similarly, claims 9 and 18 involve deciding if an improvement is sufficient, and claims 10 and 19 use feature vectors, a standard data representation technique. In all cases, the core operations remain activities that can be performed by human thought, thus falling within the "mental process" abstract idea.
Because the claims recite a judicial exception, the analysis proceeds to Prong Two.
Prong Two: Does the claim integrate the judicial exception into a practical application?
No. The claims do not integrate the abstract mental process into a practical application. The additional elements merely use generic computer components to perform the abstract idea, which does not transform the claim into a patent-eligible application. As stated in MPEP § 2106.04(d), "a claim that integrates a judicial exception into a practical application will apply, rely on, or use the judicial exception in a manner that imposes a meaningful limit on the judicial exception, such that the claim is more than a drafting effort designed to monopolize the judicial exception." The additional elements here fail to impose such a limit.
The additional elements beyond the abstract idea are the use of a "patient-specific metabolic model," "one or more processors," and a "non-transitory computer readable medium."
Patient-Specific Metabolic Model / Organ-Specific Metabolic Models: These limitations do not integrate the abstract idea into a practical application. The specification describes the patient-specific metabolic model as a digital model capturing the biology and metabolism of the patient’s body and states that it implements one or more machine-learned models (Spec., [0028-0029], [0055]). Under BRI, the model is used to calculate or predict treatment effects from patient data. The amended organ-specific model language narrows the prediction by organ system and recites model-to-model communication to account for cross-organ effects, but it still uses the models for the same abstract purpose: evaluating patient information and predicting metabolic response. The claim does not recite an improvement to model architecture, computer operation, network communication, or any other technology; nor does it require administering a particular treatment or transforming the patient’s body. It merely uses computer models as tools to perform the abstract clinical analysis. MPEP § 2106 recognizes that merely using a computer to apply an abstract idea, without a meaningful technological improvement or other practical application, does not integrate the exception into a practical application.
Non-Transitory Computer Readable Medium / One or More Processors: These limitations recite generic computer components that are invoked merely as a tool to perform the abstract idea. The specification describes these elements in purely conventional terms, such as a "storage device 230 representing a non-volatile memory, such as a hard drive, compact disk read-only memory (CD-ROM), DVD, or a solid-state memory device" (Spec., Paras. [0042]- [0043]) and a "chipset 210 coupled to at least one processor 205" (Spec., Para. [0041]). Reciting generic computer components to perform generic computer functions (like processing and storing data) amounts to "mere instructions to implement an abstract idea on a computer" and does not constitute a practical application (MPEP § 2106.05(f)).
The dependent claims similarly fail to integrate the abstract idea. The limitations regarding accessing data, labeling, comparing to thresholds (claims 2–4, 12–13), identifying patient subsets (claims 5, 14), and determining long-term effects (claims 6, 15) are all generic data manipulation steps. Even amended claim 8, which adds steps for categorizing patients and determining sensitivity, represents conventional data segmentation and analysis, not a specific technical application.
Considered individually and as a whole, the additional elements do not transform the abstract mental process into a practical application. They simply use a general-purpose computer to automate the abstract analysis. Therefore, the claims are "directed to" the abstract idea (Step 2A: YES), and the analysis must proceed to Step 2B.
Step 2B: Inventive Concept (Significantly More)
The claims do not recite additional elements that amount to "significantly more" than the abstract idea itself. This inquiry looks for an "inventive concept" by considering whether the additional elements transform the claim into a patent-eligible application by adding something more than what is well-understood, routine, and conventional in the field (MPEP § 2106.05).
The additional elements identified in Step 2A are analyzed here to determine if they constitute an inventive concept.
Patient-Specific Metabolic Model: This element lacks an inventive concept. The specification describes it broadly as a "digital twin" that "implements one or more machine-learned [models]" (Spec., Para. [0055]) for the purpose of "simulating ... clinical trials using personalized digital twins" (Spec., Para. [0028]). Using machine-learned models for prediction and simulation is a well-understood, routine, and conventional activity in data science and healthcare analytics Refer to US20210202098- Bostic par. 0026, US20210202107-Bostic par. 0026, US20190005200, par.0078. The specification provides no details on a specific, unconventional modeling technique or algorithm that would constitute an improvement in computer technology or medical modeling itself.
The amended patient-specific metabolic model limitation does not add an inventive concept. Under BRI, the claim requires organ-specific metabolic models trained to predict effects on distinct organ systems and to communicate with another organ-specific model to capture cross-organ effects. The claim does not require a particular model architecture, training technique, dependency graph, propagation rule, communication protocol, data structure, or algorithm. The added communication language only states the desired result of capturing cross-organ interaction effects; it does not claim how the communication is technically performed or how the models are improved. The ordered combination likewise remains generic computer implementation of the abstract prediction and comparison process. Therefore, the additional elements, individually and as an ordered combination, do not amount to significantly more than the judicial exception under Step 2B.
One or More Processors / Non-Transitory Computer Readable Medium: These elements are conventional. As quoted previously, the specification at paragraphs [0041]- [0043] describes a generic computer architecture, including a "chipset 210," "processor 205," "memory controller 211," and "storage device 230." These are fundamental, off-the-shelf components used for their basic functions, representing nothing more than well-understood, routine, and conventional computer hardware.
Dependent Claims: The limitations added by the dependent claims also fail to provide an inventive concept.
Cohort Generation (Claims 2-8, 12-17): These claims add limitations regarding how patients are selected. This includes accessing data where "patient data sent through the network 150 is received by the patient health management platform 130…where it is analyzed" (Spec., Para. [0033]); determining historical changes where the "patient cohort generator 920 analyzes an individual patient's historical changes" (Spec., Para. [0137]); and comparing to a threshold where, "If the historical change(s) satisfies the threshold change, the patient cohort generator 920 labels the patient as sensitive" (Spec., Para. [0138]). Amended claim 8 adds categorizing patients "based on current metabolic states" (Spec., Para. [0141]). These are all abstracts’ steps data-filtering and analysis techniques.
Physical Experiment & Data Input (Claims 9-10, 18-19): These claims add steps such as generating instructions for a physical experiment, where the "physical experiment generator 950 generates a shortlist of candidate treatment recommendations" (Spec., Para. [0154]), and encoding data into a feature vector, where "inputs…may be encoded into a vector representation" (Spec., Para. [0099]). These are routine pre- and post-processing steps in data analysis and experimental design.
In summary, the additional elements, both individually and as an ordered combination, do not supply an inventive concept. They amount to nothing more than the performance of well-understood, routine, and conventional activities in the field of data analysis, performed on a generic computer. The claims therefore do not amount to "significantly more" than the abstract idea of collecting, analyzing, and predicting information.
Conclusion
For the foregoing reasons, claims 1-20 are directed to patent-ineligible subject matter under 35 U.S.C. 101.
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.
Claim(s) 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over US20220044826A1-Schaeffer, and further in view of US20190252079A1-Constantin and in further view of Thiele, refer to PTO-892-U.
Claim 1.
Schaeffer teaches, A method comprising:
, identifying an intervention parameter in a treatment recommendation for causing a target improvement in the intervention parameter to cause the target improvement wherein the ; (See at least, par. 0238, 0080… particular organ or a patient's future probability of metastasis to yet another organ in…, 0434, 0439), Schaeffer teaches a treatment with an intervention parameter rule as anastrozal with lotinib medication therapy to improve probability PFS in 12 months.
based on correlations between changes generating, from a population of patients, a cohort of patients sensitive to the intervention parameter, in the ; Schaeffer, teaches in par. 0005 and par. 0238 a cohort according to certain treatment.
separating the cohort of patients into a control cohort comprising a first subset of patients and a test cohort comprising a second subset of patients; (See at least, 00303, 0005, 0008), Schaeffer teaches a control and treatment cohort.
and determining an effect of the treatment recommendation on the cohort of patients, the determination comprising:
for each patient of the test cohort, inputting the treatment recommendation into a patient-specific ; (Schaeffer, par. 0005-0008, 0126, 0157-0158, 0172-0173, 0251, 0224)
Schaeffer describes a prediction framework that analyzes individual patients within a defined population, using a model that processes patient-specific data, including treatments, to forecast an expected outcome such as survival or response for each patient. This is disclosed where Schaeffer describes a system to predict an expected response of a particular patient population or cohort when provided with a certain treatment and performs operations for each of the plurality of patients (for each patient of the test cohort). Schaeffer further details using a prediction model implemented based on information about the patients' medical history and treatments (inputting the treatment recommendation into a patient-specific model). Finally, Schaeffer states an objective is determining the likelihood of each patient surviving longer than Y years, which constitutes an individual predicted effect (to generate an individual predicted effect of the treatment on the patient).
determining a representative effect of the treatment recommendation for the test cohort based on the generated individual predicted effects for the test cohort; (Schaeffer, par. 0005-0008, 0126, 0129, 0139-0142, 0303)
Schaeffer discloses calculating an aggregate outcome for a patient group that is derived from the individual data points of the patients within that group. This is taught where Schaeffer’s system is designed to predict an expected response of a particular patient population or cohort when provided with a certain treatment, which constitutes determining a representative effect for the test cohort. This representative effect is based on individual predicted effects, as Schaeffer further describes processes such as calculating an average survival rate for the cohort of patients and generating time until event… curves… for sub-groups of the filtered cohort of interest, both of which are aggregate metrics (representative effects) necessarily derived from the individual outcomes and data of each patient constituting the cohort. and comparing the representative effect for the test cohort to representations of (Scheffer, par. 0005-0007, 0126, 0139-0142, 0303)
This limitation is explicitly taught by Schaeffer, which describes multiple ways of comparing a specific cohort's results against a baseline or control group.
In summarize Schaeffer teaches all limitations of evaluating treatment efficacy through comparative analysis of predicted treatment outcomes against a control group, except for the specificity of metabolic status and model. However, Constantin invention is for diabetic a metabolic health disease (refer to abstract Constantin) and by first real-time datum associated with a patient, determining a state of the patient (Refer to par. 0010, 0532). Therefore, describe a model that predictive a metabolic state for a patient. It therefore would have been obvious to one of ordinary skill in the art, before the effective filling date of the claimed invention, to add in Schaeffer invention the Constantin specificity metabolic state and model to provide support for treatment decision-making (refer to par. 0002 Constantin). Schaeffer also speak about metabolic panels paragraph 0391, that is a rational to combine Schaeffer with Constantin to develop a specific in a metabolic state improvement.
Schaeffer discloses predict an expected response of a particular patient population or cohort when provided with a certain treatment and asks What is the likelihood of each patient surviving longer than Y years in para. 0005 and para. 0158. Schaeffer further discloses that the prior features may include various features related to a patient's medical condition and or treatment and that for each patient in the cohort having the anchor point, the prediction model may be provided with the selected subset of the plurality of forward features in para. 0186 and para. 0251. These teachings reasonably show a patient-specific predictive model that receives treatment-related inputs and produces an individual predicted treatment outcome for each patient in the cohort.
Schaeffer does not teach wherein the patient-specific metabolic model comprises a plurality of organ-specific metabolic models, each organ-specific metabolic model communicates with at least one other organ-specific metabolic model to capture cross-organ interaction effects on the metabolic state of the patient.
Thiele article teaches that missing architecture. Thiele discloses three organ-specific metabolic models were connected through a blood compartment and further discloses whole-body metabolic reconstructions can be converted into personalized WBM models and started from a meta-reconstruction with one copy of Recon3D for each of its 28 organs, tissues, and cell types connected through anatomically consistent biofluid compartments. Refer to Abstract page 1, introduction/result page 2
Thiele also states in the Abstract on page 1 that the reconstructions capture the metabolism of 26 organs and could recapitulate known inter-organ metabolic cycles. Those disclosures teach a patient-personalized multi-organ metabolic model made of distinct organ-specific models that exchange information through shared compartments so cross-organ metabolic interactions are represented.
A POSITA would have combined Schaeffer with Thiele to improve the physiologic specificity of Schaeffer’s patient-level treatment-effect prediction. The specific modification would have been to implement Schaeffer’s patient-specific predictive model using Thiele’s personalized whole-body organ-resolved metabolic architecture, so the treatment input is propagated through multiple interacting organ models rather than a single undifferentiated patient model. The reason to combine is supported by Schaeffer’s express goal of predicting patient-specific treatment response and survival for treatment decision-making, and by Thiele’s teaching that personalized organ-resolved metabolic models capture inter-organ metabolic cycles and provide more realistic metabolic prediction. The result would have been predictably successful because using known organ-specific submodels linked through known biofluid connections is a straightforward application of prior-art modeling structure to Schaeffer’s existing treatment-response prediction task.
Claim 2.
Schaeffer in further view of Constantin and Thiele teaches,
The method of claim 1, wherein generating the cohort of patients further comprises: accessing patient data for the population of patients, the patient data comprising labels describing the sensitivity of each patient of the population of patients to the intervention parameter; (See at least, [par.0125, 0324], Schaeffer describes retrieving patient data that includes pre-categorized disease response information. The combination of retrieving pre-categorized data and selecting based on that data describes the limitation.) and generating the cohort of patients based on patients sensitive to the intervention parameter in the treatment recommendation based on the accessed patient data. (See at least, [abstract, 0005, 0324, 0358], Schaeffer describes the way to select the patients, based on specific feature that could be easy substitute with treatment for example and sensitivity.)
Claim 3.
Schaeffer in further view of Constantin and Thiele teaches,
The method of claim 2, wherein assigning the label describing the sensitivity of a patient to the intervention parameter to the patient comprises: determining historical changes in a metabolic state of the patient caused by previous adjustments to the intervention parameter; (See at least, [0251, 0324, 0152], Schaeffer describes the uses historical patient data, specifically “disease response” labels derived from physician notes and medical records, to categorize patients based on their sensitivity to past treatments. ) and assigning the patient to either a first subset of patients sensitive to the intervention parameter or a second subset of patients insensitive to the intervention parameter based the historical changes. (See at least, [0152, 0324]-Schaeffer)
Claim 4.
Schaeffer in further view of Constantin and Thiele teaches,
The method of claim 3, further comprising: comparing the historical changes in the metabolic state to a threshold change; (See at least, [0152, 0136], Schaeffer describe a threshold-based comparison “survival time to a particular time” to assign patients to one of two subgroups.)
Claim 5.
Schaeffer in further view of Constantin and Thiele teaches,
The method of claim 1, wherein generating the cohort of patients further comprises: identifying, from the population of patients, a subset of patients whose metabolic state is below a threshold metabolic state; (See at least, [0152, 0358, 0324], The Schaeffer paragraphs describe selecting patient who did not survive with a particular time (Threshold). This “not surviving” is a below a threshold. This selection forms a subgroup which, is equivalent to a cohort that are generated based on specific features and values. and generating the cohort of patients from the identified subset of the population of patients. (See at least, [0324-Schaeffer])
Claim 6.
Schaeffer in further view of Constantin and Thiele teaches,
The method of claim 1, wherein generating the cohort of patients further comprises: determining a long-term effect of adjustments to the intervention parameter on each patient of the population of patients based on historical changes in the metabolic state of the patient; (See at least, [0002, 0324, 0136, 0152-Schaeffer], Schaeffer analyzes historical patient data, including treatments and outcomes. It identifies features that have a significant impact on prediction, and these features can include past treatments. The system evaluates levels of response and uses time periods in its analysis, indicating an assessment of lasting impact. The disease response categories are direct measures of the impact of treatments of patient data.)
and generating the cohort of patients based on the long-term effect of adjustments to the intervention parameter determined for each patient of the population of patients. (See at least, [0152, 0324, 0136, 0160], Schaeffer describes creating subgroups of patients. This subgroup creation is based on the lasting impact of past events as evidenced by the use of survival time, Disease response categories and survival profile. The system evaluates each patient’s outcome individually to determine a subgroup membership.)
Claim 7.
Schaeffer in further view of Constantin and Thiele teaches,
The method of claim 1, wherein the treatment recommendation comprises instructions or adjusting a plurality of intervention parameters and generating the cohort of patients comprises: for each patient of the population of patients, determining a sensitivity of each patient to each intervention parameter of the plurality; (See at least, [0358, 0069, 0107, 0369, 0324, 00002, 0363, 0107], Schaeffer allows the user to select multiple criteria and features. These criteria can include various aspects of treatment. Selecting different values for these features is equivalent to “adjusting a plurality of intervention parameters. Also, the outcome category, includes sub-categories.) and determining an overall sensitivity of the patient to the treatment recommendation based on the sensitivity of the patient to each intervention parameter of the plurality; (See at least, [par. 0136, 0160, 0152, 0171, 0270]: Schaeffer analyzing how different combinations of factors affect patient outcomes. The process of creating subgroups and determining differences in survival inherently involves combining the effects of multiple factors. The ability to work with unlabeled data and various machine learning algorithms, add flexibility in how this combination is performed and the greatest weight it is implies an overall sensitivity.) and generating the cohort of patients based on the overall sensitivity of each patient of the population of patients. (See at least, [par. 0387, 0152], Schaeffer is designed to generate cohorts based on selected criteria. Overall sensitivity can be established as a criterion as established in previous limitations, then the system can generate a cohort based on it.)
Claim 8.
Schaeffer in further view of Constantin and Thiele teaches,
The method of claim 1, wherein generating the cohort of patients comprises: categorizing the population of patients into categories of patients with a shared metabolic state; (See at least, [par. 0233-0234,], Schaeffer categorizes patients using similarity metrics and dimensionality reduction, which aligns with categorizing patients. ) for each category of patients, predicting an effect of the treatment recommendation on each patient of the category by inputting the treatment recommendation to a patient-specific metabolic model of the patient; (See at least, [par. 0218,-0219], Schaeffer describes a predictive model trained on patient-specific features. ) and determining an overall sensitivity of the category of patients to the treatment recommendation based on the predicted effect of the treatment recommendation on each patient of the category; (See at least, [par. 0228-0232], Schaeffer describes measuring patient similarity through decision trees.) and determining a category of patients most sensitive to the treatment recommendation based on a comparison of the overall sensitivity of each category of patients; (See at least, [par. 0228-0230 ], Schaeffer describes selecting features that create the most significant differences between patient groups. ) and generating the cohort of patients based on the category of patients most sensitive to the treatment recommendation. (See at least, [par.0229-0230], Schaeffer describes forming patient subgroups based on distinguishing features, which aligns with generating a patient cohort using the most responsive category)
Claim 9.
Schaeffer in further view of Constantin and Thiele teaches,
The method of claim 1, further comprising: determining that the effect of the treatment recommendation on the cohort of patients satisfies a threshold improvement in a metabolic state of each patient of the cohort of patients; (See at least, [0387, 0390-0391, 0394], Schaeffer describes a system that determines a treatment effect by identifying and analyzing patient cohorts based on predefined health criteria, aligning with determining the impact of treatment recommendations. Schaeffer quantifies the effect of genetic variations on disease states, which is analogous to assessing whether a treatment leads to a threshold improvement in health.) and generating instructions for performing a physical experiment to validate the treatment recommendation. (See at least, [par. 0418, 0421, 0436-0437], Schaeffer describes generating reports and patient similarity indicators for clinical trials.)
Claim 10.
Schaeffer in further view of Constantin and Thiele teaches,
The method of claim 1, wherein determining the effect of the treatment recommendation on the cohort of patients further comprises: encoding a feature vector representation of the treatment recommendation; (See at least, [par. 0218],Schaeffer describes a system where treatment predictions are generated using machine learning based on structured patient features analogous to feature vector representation of the treatment recommendation.) and inputting the feature vector representation to the patient-specific metabolic model of each patient of the test cohort. (See at least, [0265, 0220], Schaeffer maps medical treatment features into structured categories for predictive analytics, which aligns with inputting a feature vector representation into a patient-specific model.)
Note: Claims 11-20 are rejected with the same analysis above, for being very similar to claims 1-10.
Conclusion
Relevant Prior Arts:
US 20160335412 A1
Abstract
The method and system of this invention provides for the use of the Simcyp Simulator to identify the characteristics of a Virtual Twin to a real patient based on physiological data and demographic characteristics of the real patient. The Virtual Twin can be used to estimate appropriate dosage levels for a real patient undergoing pharmaceutical treatment and to indicate drug interactions that can occur during the administration of multiple drugs.
[0018] FIG. 2. System Pharmacokinetics—an area of pharmacology that integrates information on system specific factors (demography, physiology, genetics) with drug specific factors (physicochemical properties, in vitro data on binding to proteins, membrane permeability, and enzyme and transporter kinetics) to predict pharmacokinetic (PK) behavior (summarized by clearance (CL), volume of distribution (V), elimination half-life (t.sub.1/2) and the likely extent of drug-drug interactions (DDI)), which can then be linked to pharmacodynamic (PD) outcomes.
US 20230020925 A1
[0020] Furthermore, the invention in particular implements a feedback loop between outputs of each digital twin and the input of the other digital twin. In other words, outputs of each respective digital twin are provided as inputs to the other of the digital twins….
US 20150154375 A1
Par. 0006… In several embodiments, the one or more mutations are unique to the subject (or to the disease that the subject is affected by), and thus provide a possible specific target for a drug therapy. In several embodiments, the methods comprise identifying one or more drugs (from a pool of candidate drugs) that may have an increased therapeutic efficacy against cells having the one or more mutations identified from the sample from the subject. In other words, the subject is screened to identify disease targets that are unique and then a drug (or drugs) are identified that will be effective in treating (e.g., eliminating or reducing the activity of) cells or tissues bearing the one or more identified mutations. Some embodiments, also comprise evaluating the pharmacogenomics profile of the subject in order to identify and/or classify the subject with respect to his or her ability to absorb, distribute, excrete or otherwise metabolize each of the drugs identified. In this manner, the potential for the subject to react adversely to a drug (e.g., if they metabolize a drug very slowly, a standard dose may lead to adverse side effects in that subject) can be identified. Likewise, the possibility of the subject needed a particularized dose, such as a dosing regimen employing greater or more frequent dosing, can be identified if, for example, the subject metabolizes a drug particularly rapidly. Moreover, in several embodiments, the methods also evaluate the possibility of drug-drug interactions between each of the identified candidate drugs as well as optionally the possible interaction between any of the identified candidate drugs and other drugs that the subject may already be receiving (e.g., other drugs for the disease or ailment in question or another disease or ailment). In this manner, the potential for adverse effects ….
US 20210202103 A1
Abstract
Systems and methods are provided for simulating a patient health state by determining one or more relationships between patient data and historical data, creating enriched data elements based on the determined relationships, and using a machine learning module to compute a current health state for a patient and to simulate a future health state of the patient.
[0011] In embodiments, the future wellness state is a predicted illness. In embodiments, the machine learning simulation includes pharmaceutical data to simulate the future wellness state contingent upon the patient taking a stated medication. In embodiments, the machine learning simulation includes treatment plan data to simulate the future wellness state contingent upon the patient receiving a stated treatment. In embodiments, the machine learning simulation uses a digital twin of the patient. In embodiments, the digital twin of the patient is matched to a digital twin representing a population of patients sharing a patient health attribute. In embodiments, the digital twin of the patient is matched to a plurality of digital twins, each representing a population of patients receiving a stated treatment, wherein each stated treatment is indicated for the patient. In embodiments, the digital twin of the patient is matched to a plurality of digital twins, each representing a population of patients receiving a stated medication, wherein each stated medication is indicated for the patient. In embodiments, the machine learning simulation uses a plurality of digital twins of the patient.
US 20210241909 A1
Abstract
A method of assessing the impact of a certain treatment strategy for a patient on a digital twin or virtual model of that patient. It is recognized that the impact on a virtual twin could determine whether or not a treatment strategy is selected, as clinicians are becoming increasingly reliant on virtual twins to perform long-term monitoring of a patient's condition.
[0004] A virtual model is a digital representation of at least part of the anatomy and/or a bodily/psychological process of the patient, i.e. a biological model. The virtual model processes input data, which may provide characteristics or measured parameters of the patient, to generate output data. The output data may comprise other predicted characteristics of the patient.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JOSHUA DAMIAN RUIZ whose telephone number is (571)272-0409. The examiner can normally be reached 0800-1800.
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, Shahid Merchant can be reached at (571) 270-1360. 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.
/JOSHUA DAMIAN RUIZ/Examiner, Art Unit 3684
/Shahid Merchant/Supervisory Patent Examiner, Art Unit 3684
Prosecution Insights