DETAILED ACTION
Notice of Pre-AIA or AIA Status
1. Claims 1-22 and 24-25 are currently presented for Examination.
2. The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Continued Examination Under 37 CFR 1.114
3. 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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 12/22/2025 has been entered.
Response to Amendment
4. The amendment filed on 12/12/2025 has been entered and considered by the examiner. By the
amendment, claims 1, 16-17 and 19-21 are amended. In light of the amendment and arguments made, examiner withdraw the claim rejection under 112(a) and 112(b). Following Applicants arguments and amendments made, examiner still maintained the 101 and 103 rejections. The 103 rejection of the claims was modified. See office action for detail.
Response to 101 Arguments
Applicant arguments
Under the updated MPEP guidance, the Office is required to (1) identify the improvement described in the specification, and (2) determine whether the claim itself reflects the disclosed improvement. That is, the claim includes the components or steps of the invention that provide the improvement described in the specification. The present application fits squarely within this updated framework. As explained in Applicant's prior response and in Dr. Jacobs's declaration, the specification describes a specific technological improvement to CGM-based decision- support systems: a feature-extraction and prediction pipeline that enables long- horizon, nocturnal hypoglycemia forecasting and pre-sleep therapeutic recommendations, which conventional CGM systems could not provide. The claims reflect that improvement through their recitation of, among other things, a CGM system with a hypoglycemia prediction model formulated to predict future incognizant hypoglycemic events beyond the transient postprandial window and generate and communicate specific glucoregulatory management interventions that, when followed, prevent the predicted event. As noted above, the final Office action largely repeats the earlier characterizations of the claim at a high level of generality and does not grapple with the specific architecture or the clinical improvement it achieves, much less with the factual record supplied by Dr. Jacobs. The updated MPEP guidance makes clear that "'Examiners and panels should not evaluate claims at such a high level of generality that potentially meaningful technical limitations are dismissed without adequate explanation."
The USPTO's August 4, 2025 memorandum to Technology Centers 2100, 2600, and 3600 (Exhibit C) reinforces the same principles reflected in the updated MPEP §2106 and Desjardins. The memorandum instructs examiners (particularly those in TC 2100, which includes Art Unit 2188) to evaluate Step 2A by (1) determining whether claim limitations "cannot practically be performed in the human mind," specifically cautioning against expanding the mental-process grouping, (2) distinguishing claims that merely involve a judicial exception from those that recite one, and (3) analyzing the claim as a whole rather than at an overly generalized level that ignores technical limitations. The memorandum further emphasizes that, under Step 2A Prong Two, examiners must consider whether the claim reflects "an improvement in the functioning of a computer or any other technology or technical field," and cautions against dismissing additional elements as generic without evaluating whether they meaningfully limit the recited exception. These instructions apply directly here. The final Office action does not analyze the claim as a whole, does not address whether the claimed CGM-integrated feature-extraction and prediction architecture constitutes a technological improvement, and does not apply the memo's guidance on distinguishing mental processes from machine-implemented processes that cannot be performed in the human mind. As such, the §101 analysis in the final Office action is inconsistent with the USPTO's most recent guidance.
For at least these reasons, the § 101 rejection is inconsistent with the updated § 2106 framework and should be withdrawn.
Even if claim 1 were found to recite a judicial exception at Step 2A, the claim satisfies Step 2B. The same technical configuration that the notice and Desjardins treat as relevant to technological improvement under MPEP § 2106.05 also demonstrates that the claim as a whole amounts to significantly more than any alleged abstract idea. The notice accordingly adds new examples to MPEP § 2106.05(a): (xiii) an improved way of training a machine-learning model that preserves knowledge of prior tasks while learning new ones, and (xiv) improvements to computer component or system performance based on adjustments to parameters of a machine-learning model associated with tasks or workstreams.
Here, the claimed CGM system likewise implements a particular configuration of parameters and processing stages that changes how the system operates: instead of passively reporting near-term glucose trends, it generates long-horizon risk predictions and pre-sleep therapeutic recommendations that the prior art did not, and as the record shows, could not provide. This is not a generic application of data processing; it is a technology-based solution tailored to a specific clinical problem, analogous in structure to the model-based improvements credited in Desjardins.
Dr. Jacobs's declaration provides unrebutted evidence that, at the time of the invention, prior forecasting systems were limited to short horizons (typically 30-120 minutes), that no system performed overnight prediction 6-8 hours ahead, that Oviedo expressly identified extended nocturnal prediction as an unsolved problem, and that no CGM platform delivered pre-sleep actionable therapeutic recommendations based on such forecasts. The same declaration reports that the claimed approach, when implemented in a CGM system, achieved a 29.4% reduction in severe nocturnal hypoqlycemia (<54 mq/dL) in a controlled clinical trial, without worsening overall glycemic control. The final Office action does not address, weigh, or rebut any of these factual showings. Taken together, the claimed ordered combination (CGM-based feature extraction, a prediction model formulated for long-horizon nocturnal IH events, and a pre-incognizance therapeutic recommendation mechanism that materially improves patient outcomes) is neither routine nor conventional in the field. It represents exactly the kind of specific, technical solution to a technical problem that the updated MPEP and Desjardins identify as satisfying both Step 2A, Prong Two and, if reached, Step 2B. In view of this record, the § 101 rejection should be withdrawn.
Examiner response
Applicant argues that the claims are patent-eligible in view of Desjardins and the updated MPEP 2106 guidance, arguing the claims recite a technological improvement to CGM-based decision support systems and therefore integrate the judicial exception into practical application, however, Examiner respectfully disagrees. The claim recites limitations including receiving glucose data, extracting features, applying a prediction model, predicting future hypoglycemia and generating a recommendation. These limitations describe mathematical modeling and analysis of medical data to produce a recommendation. Thus, it falls within the abstract idea grouping of mathematical concepts and mental process. Applicant/Declaration does not dispute that the claims involve mathematical modeling. In Desjardins, the ARP found eligibility because the claims recited specific parameter preservation technique that improved continual learning models. Here, the claims do not recite catastrophic forgetting mitigation, model-parameter preservation or specialized training architecture. The instant claims recite generic prediction and recommendation functionality. Even if the specification describes such improvements, the relevant inquiry is whether the claim themselves recite those improvements. The claim recites a prediction model configured to predict a future hypoglycemic event and generating a recommendation. These limitations are recited at a high level of generality and do not recite specific technological mechanism that produces the alleged improvement. Applicant argues the claims improves patient outcomes and forecasting, however, improvement to medical decision making or clinical outcomes do not constitute improvements in computer technology. As MPEP (2106.04(a)(2)(III)(C)) states using a computer as tool to perform a mental process falls under the grouping of abstract ideas. Generic computer system performing a generic computer function such that it amounts no more than mere instructions to apply the exception using a generic computer component. (See MPEP 2106.05(f)). Claims requiring a generic computer or nominally reciting a generic computer may still recite a mental process even though the claim limitations are not performed entirely in the human mind. Thus, the claimed methods employ generic computer components performing generic computer functions. Considering all the limitations in combination, the claim do not show any inventive concept, such as improving the performance of a computer or any other technology. It is for these reasons that Applicants claimed subject matter is not patent eligible. Here the only additional element recited in the claims beyond the abstract idea are computer, processor, non-transitory computer-readable media and memories,” i.e., generic computer component. See Alice, 573 U.S. at 223 (“[T]he mere recitation of a generic computer cannot transform a patent-ineligible abstract idea into a patent-eligible invention.”). Applicant has not identified any additional elements recited in the claim that, individually or in combination, provides significantly more than the abstract idea. Moreover, when viewed as a whole with such additional elements considered as an ordered combination, the claim modified by adding a generic computer would be nothing more than a purely conventional computerized implementation and would not provide significantly more than the judicial exception itself. Here, the system receives glucose data (sensory information) from the generic sensor which is mere data gathering step and falls under insignificant extra solution activity as mentioned in MPEP 2106.05(g). The generic processor processes this data, extracting features (a form of data interpretation and analysis similar to a mental process). This data is then applied to a hypoglycemia prediction model (mimicking human reasoning and decision-making about risk). Based on the model's prediction, the system generates a recommendation (a type of advice or conclusion, again similar to human thought processes). This recommendation is communicated and presented to the subject, enabling them to prevent the event (facilitating their own decision-making and actions). Thus, claim is directed to combination of mental process and mathematical concepts of abstract ideas.
Applicant argues that the ordered combinations amount to significantly more. The additional elements recite the claim include sensor, processor, non-transitory computer readable medium, memory, user interface and prediction model are well-understood, routine and conventional. Claims do not recite any unconventional computing architecture. The claim “in which the glucoregulatory management recommendation is that the living subject ingest an amount of carbohydrate, reduce an insulin dose, or inject glucagon is merely an intended use of the claimed invention or a field of use limitation, then it cannot integrate a judicial exception under the "treatment or prophylaxis" consideration. According to the MPEP 2106.05(d)(2)-treatment" or "prophylaxis" limitation for purposes of this consideration, the claim limitation in question must affirmatively recite an action that effects a particular treatment or prophylaxis for a disease or medical condition. An example of such a limitation is a step of “administering amazonic acid to a patient” or a step of “administering a course of plasmapheresis to a patient.” If the limitation does not actually provide a treatment or prophylaxis, e.g., it is merely an intended use of the claimed invention or a field of use limitation, then it cannot integrate a judicial exception under the “treatment or prophylaxis” consideration. For example, a step of “prescribing a topical steroid to a patient with eczema” is not a positive limitation because it does not require that the steroid actually be used by or on the patient, and a recitation that a claimed product is a “pharmaceutical composition” or that a “feed dispenser is operable to dispense a mineral supplement” are not affirmative limitations because they are merely indicating how the claimed invention might be used.
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception (inventive concept) because the additional elements when considered both individually and as an ordered combination do not amount to significantly more than the abstract idea. The disclosed and claimed invention are not an improvement to technology, but rather a mental/mathematical solution to a mental problem – i.e., an improvement in the abstract idea itself.
Applicant relies on the Jacobs Declaration to argue the prior CGM systems were limited to short prediction horizons and that the claimed invention achieves long horizon nocturnal forecasting with improved clinical outcomes. However, the 101 eligibility under Step 2A is not whether the invention is new, useful or clinically superior. Rather, it is whether the claims are directed to judicial exceptions and if so whether they integrate that judicial exception into practical application. Even assuming, the prior system was limited to 30-120 minutes horizons and that the claimed system performs 6-8 hrs. forecasting, the claim still recites the abstract idea of collecting glucose data, applying mathematical prediction model and generating a medical recommendation. Thus, extending a prediction time horizon does not change the fundamental character of the claim as being directed to mathematical modeling and medical decision making. Applicant emphasizes a 29.4% reduction in severe nocturnal hypoglycemia events. While such results may be relevant to non-obviousness under 103, they are not dispositive under 101. An abstract idea does not become patent eligible simply because it produces beneficial results. The Declaration primarily addresses clinical performance, prediction horizon and prior art forecasting capabilities. It does not establish that the claimed use of machine-learning models, feature extraction, CGM data processing, generation of recommendations was unconventional at the time of filing. The claims do not require a specific non-conventional machine learning architecture. They broadly recite applying a prediction model. Machine-learning based glucose prediction systems are well known in the art. (See Oviedo, Constantin, Davis) Thus, the Declaration does not create a factual dispute requiring withdrawal of the rejection under step 2B. Accordingly, the rejection under 101 is still maintained.
Response to 103 Arguments
Applicant arguments
The claims here require predicting a future incognizant hypoglycemic event during a period when the subject will be incognizant, and they require generating a therapeutic intervention before the subject enters that period. Oviedo does not address incognizance or sleep at all and provides no mechanism for predicting such events. Oviedo does not provide any therapeutic recommendations and certainly does not instruct a system to deliver pre-sleep guidance such as carbohydrate ingestion, insulin dose adjustment, or glucagon use. Nothing in Oviedo suggests modifying its short-horizon postprandial classifier into a long-horizon nocturnal risk forecasting system integrated into a CGM platform with an actionable intervention step.
As noted previously, the Office action's rejection also does not address the unrebutted factual evidence submitted in the Jacobs declaration. Dr. Jacobs explains that prior to this invention, forecasting hypoglycemia beyond 30-120 minutes was not feasible with existing methods, and that no CGM system provided long-horizon nocturnal prediction or pre-sleep therapeutic guidance. He further explains that
Oviedo itself confirms these limitations. The Office action has not disputed or addressed this evidence, which shows that the claimed arrangement was not routine, predictable, or suggested by the cited prior art. Even accepting the Office action's statement that, for examination purposes, "exceeding a transient postprandial window" is treated as equivalent to nocturnal hypoglycemia involving long-range forecasting, Oviedo still does not teach or suggest the claimed subject matter. Oviedo does not perform long-range forecasting at all. It expressly warns that its predictions cannot extend reliably into the nocturnal period. A reference that identifies a problem it cannot solve does not provide a motivation or reasonable expectation of success in modifying it to achieve the claimed invention. Because Oviedo does not teach or suggest the claimed long-horizon prediction of an IH event occurring during sleep, does not teach proactive therapeutic recommendations, and does not disclose or suggest the claimed predictive architecture, and because the Jacobs declaration demonstrates that the invention achieves results not predictable from Oviedo or any other cited art, the § 103 rejection should be withdrawn.
Examiner response
In view of Declaration of Dr. Peter Jacobs on May 2, 2025 under 37 C.F.R. § 1.132 for the prior art and Applicant arguments, Examiner withdraw the Oviedo reference for rejecting claim 1 and added the new references Constantin that clearly teaches time window for exceeding the transient postprandial window. Constantin address incognizance or sleep and provides mechanism for predicting such events. See office action for detail.
Claim Rejections - 35 USC §101
5. 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.
6. Claims 1-22 and 24-25 are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter. These claims are directed to an abstract idea without significantly more.
(Step 1) Is the claims to a process, machine, manufacture, or composition of matter?
Claims: 1-22 and 24-25 are directed to process or method, which falls into the one of the statutory category.
(Step 2A) (Prong 1) Is the claim directed to a law of nature, a natural phenomenon, or an abstract idea? (Judicially recognized exceptions)?
processing the glucose management data as prescribed by a feature extractor to generate a set of glucoregulatory feature values; (Under the broadest reasonable interpretation, this limitation covers mental process including an evaluation or judgement that could be performed in the human mind or with the aid of pencil and paper therefore it falls within the “Mental Process” grouping of abstract ideas. Also, this limitation also falls under mathematical concepts since feature extraction involves mathematical algorithm (such as “Mathematical transformation”). So, it falls under the mental process or mathematical concepts of abstract ideas)
applying the glucoregulatory feature values as prescribed by the feature extractor to a hypoglycemia prediction model, the hypoglycemia prediction model being formulated to predict the potential failure occurrence of the IH event; (Under the broadest reasonable interpretation, this limitation covers mental process including an evaluation or judgement that could be performed in the human mind or with the aid of pencil and paper therefore it falls within the “Mental Process” grouping of abstract ideas. Also, the mechanism relies on computational, algorithmic, and statistical analysis (e.g., mapping input parameters to a risk score) The computer serves as a tool to rapidly perform these complex, high-dimensional calculations that are too intricate for human mental calculation This is a "mental process" (evaluation, judgment) or "managing personal behavior".)
and
generating a glucoregulatory management recommendation as prescribed by the decision support recommender if the hypoglycemia prediction model predicts the occurrence of a hypoglycemic event during the period of incognizance. (Under the broadest reasonable interpretation, this limitation covers mental process including an evaluation or judgement that could be performed in the human mind or with the aid of pencil and paper therefore it falls within the “Mental Process” grouping of abstract ideas. It is the result of the above mathematical algorithm/abstract idea. The recommendation itself is a cognitive, diagnostic, or therapeutic decision (e.g., "If glucose is low, take sugar"). Providing advice on how to treat a condition is a method of organizing human activity or a mental act.)
Step 2A, Prong 2: Does the claim recite additional elements that integrate the judicial exception into a practical application?
In accordance with Step 2A, Prong 2, the judicial exception is not integrated into a practical application. In particular, the claim 1 recites the additional elements of receiving, via a glucose sensor, glucose management data obtained from a living subject which is a mere data gathering step and thus it falls under insignificant pre-solution activity to the judicial exception, as discussed in MPEP § 2106.05(g). The additional elements of the glucoregulatory management recommendation is that the living subject ingest an amount of carbohydrate, reduce an insulin dose, or inject glucagon is merely an intended use of the claimed invention or a field of use limitation, as discussed in MPEP § 2106.05(h). Also, the limitation “communicating the glucoregulatory management recommendation to the living subject via a connected device equipped with a user interface, which presents the recommendation and thereby enables prevention of the predicted hypoglycemic event when the living subject adheres to the recommendation” is merely displaying the output of an algorithm and is considered generic post-solution activity as discussed in MPEP § 2106.05(g). The additional elements of processor and a non-transitory computer readable medium storing instruction thereon that, when executed by the processor are mere instructions to perform the method on a generic component or machinery. (See MPEP § 2106.05(f)) Thus, a method, performed by a continuous glucose monitoring (CGM) system including a processor and a non-transitory computer readable medium storing instructions thereon that, when executed by the processor, configure the CGM system for supporting glucoregulatory management decisions using a decision support recommender to predict for a living subject when cognizant potential future occurrence of an incognizant hypoglycemic (IH) event projected to occur during a period of incognizance extending a beyond a transient postprandial window is no more than generally linking the use of a judicial exception to a particular technological environment or field of use, as discussed in MPEP § 2106.05(h). These additional elements do not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea.
Step 2B: Does the claim recite additional elements that amount to significantly more than the judicial exception?
In accordance with Step 2B, the claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. In accordance with Step 2A, Prong 2, the judicial exception is not integrated into a practical application. In particular the claim 1 recites the additional elements of receiving, via a glucose sensor, glucose management data obtained from a living subject which is a mere data gathering step and thus it falls under insignificant extra-solution activity to the judicial exception, as discussed in MPEP § 2106.05(g) and is well-understood, routine or conventional. ((See MPEP 2106.05 (d)(II)(i))) Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); TLI Communications LLC v. AV Auto. LLC, 823 F.3d 607, 610, 118 USPQ2d 1744, 1745 (Fed. Cir. 2016) (using a telephone for image transmission); OIP Techs., Inc., v.Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015) (sending messages over a network); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) (computer receives and sends information over a network); but see DDR Holdings, LLC v. Hotels.com, L.P., 773 F.3d 1245, 1258, 113 USPQ2d 1097, 1106 (Fed. Cir. 2014) The additional elements of the glucoregulatory management recommendation is that the living subject ingest an amount of carbohydrate, reduce an insulin dose, or inject glucagon is merely an intended use of the claimed invention or a field of use limitation,, as discussed in MPEP § 2106.05(h). Also, the limitation “communicating the glucoregulatory management recommendation to the living subject via a connected device equipped with a user interface, which presents the recommendation and thereby enables prevention of the predicted hypoglycemic event when the living subject adheres to the recommendation” is merely displaying the output of an algorithm and is considered generic post-solution activity as discussed in MPEP § 2106.05(g) and is considered conventional, generic activity. The additional elements of processor and a non-transitory computer readable medium storing instruction thereon that, when executed by the processor, are mere instructions to perform the method on a generic component or machinery. (See MPEP § 2106.05(f)) Thus, a method, performed by a continuous glucose monitoring (CGM) system including a processor and a non-transitory computer readable medium storing instructions thereon that, when executed by the processor, configure the CGM system for supporting glucoregulatory management decisions using a decision support recommender to predict for a living subject when cognizant potential future occurrence of an incognizant hypoglycemic (IH) event projected to occur during a period of incognizance extending a beyond a transient postprandial window is no more than generally linking the use of a judicial exception to a particular technological environment or field of use, as discussed in MPEP § 2106.05(h).The claim does not include any additional element; thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception. Thus, claim 1 is not patent eligible.
Claim 2 further recites in which the decision support recommender is trained using a supervised training process. According to the instant specification para 23, the supervise learned process includes "Classifier" is an algorithm that implements classification or the mathematical function implemented by a classification algorithm. "Classification" is a method that identifies to which of a set of categories a new observation belongs based on a set of training data containing observations or instances whose category membership is known. Examples of classifiers that are known in the art include k-nearest neighbor (KNN) classification, Case- based reasoning classification, Decision Tree classification, Naive Bayes classification, and neural network classification. Thus, this limitation involves mathematical algorithm so it falls under mathematical concepts of the abstract ideas and also generally linking the use of a judicial exception(mathematical algorithm) to a particular technological environment or field of use (machine learning), as discussed in MPEP § 2106.05(h). The claim does not include any additional element, thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception.
Claim 3 further recites in in which the supervised training process includes one or both of a regression-based process or a classifier-based process. This limitation involves mathematical algorithm so it falls under mathematical concepts of the abstract ideas. The claim does not include any additional element, thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception
Claim 4 further recites in which the regression-based process is a support vector regression or a neural-network regression. Thus, this limitation generally linking the use of a judicial exception(mathematical algorithm) to a particular technological environment or field of use (machine learning), as discussed in MPEP § 2106.05(h). The claim does not include any additional element, thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception.
Claim 5 further recites in which the classifier-based process is at least one of a random forest process, a decision tree process, and a support vector machine process. This limitation generally linking the use of a judicial exception(mathematical algorithm) to a particular technological environment or field of use (machine learning), as discussed in MPEP § 2106.05(h). The claim does not include any additional element, thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception.
Claim 6 further recites in which the supervised training process generates a set of hyperparameter values wherein hyperparameter values in the set are optimized by a cross- validation optimization process. This limitation involves mathematical algorithm so it falls under mathematical concepts of the abstract ideas. The claim does not include any additional element, thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception.
.
. Claim 7 further recites which the set of hyperparameter values includes a tolerance hyperparameter and a regularization hyperparameter This limitation involves mathematical algorithm so it falls under mathematical concepts of the abstract ideas. The claim does not include any additional element, thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception.
Claim 8 further recites in which the supervised training process includes at least one of an insulin-metabolism impact model, a glucagon-metabolism impact model, and a nutrient- metabolism impact model. This limitation generally linking the use of a judicial exception to a particular technological environment or field of use, as discussed in MPEP § 2106.05(h). The claim does not include any additional element, thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception.
Claim 9 further recites in which the nutrient-metabolism impact model includes a carbohydrate model, a protein model, or a fats model. This limitation generally linking the use of a judicial exception to a particular technological environment or field of use, as discussed in MPEP § 2106.05(h). The claim does not include any additional element, thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception.
Claim 10 further recites in which the supervised training process includes an exercise- metabolism impact model. This limitation generally linking the use of a judicial exception to a particular technological environment or field of use, as discussed in MPEP § 2106.05(h). The claim does not include any additional element, thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception.
Claim 11 further recites in which the supervised training process includes a stress- metabolism impact model. This limitation generally linking the use of a judicial exception to a particular technological environment or field of use, as discussed in MPEP § 2106.05(h). The claim does not include any additional element, thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception.
Claim 12 further recites in which the supervised training process includes a sleep- metabolism impact model. This limitation generally linking the use of a judicial exception to a particular technological environment or field of use, as discussed in MPEP § 2106.05(h). The claim does not include any additional element, thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception.
Claim 13 further recites in which the supervised training process uses virtual patient data, the virtual patient data including single or multiple hormone virtual patient data. The limitation amounts to data selecting for training and thus, falls under the insignificant extra solution activity as discussed in MPEP § 2106.05(f). The claim does not include any additional element, thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception.
Claim 14 further recites in which the virtual patient data are generated using a glucoregulatory model. The limitation amounts to data gathering or generating and thus, falls under the insignificant extra solution activity as discussed in MPEP § 2106.05(f). The claim does not include any additional element, thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception.
Claim 15 further recites in which the supervised training process uses validated or actual patient data. The limitation amounts to data selecting for training and thus, falls under the insignificant extra solution activity as discussed in MPEP § 2106.05(f). The claim does not include any additional element, thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception.
Claim 16 further recites in which the hypoglycemia threshold value is selected using a decision under theoretic process for decision under uncertainty. Under the broadest reasonable interpretation, this limitation covers mental process including an evaluation or judgement that could be performed in the human mind or with the aid of pencil and paper therefore it falls within the “Mental Process” grouping of abstract ideas. The clinician or algorithm must balance two uncertain, competing risks: False Negative (High Risk): Missing a low-glucose event (leading to seizure/coma). False Positive (Low Risk/High Inconvenience): Treating a non-existent low-glucose event (leading to anxiety, unnecessary carbs, weight gain). The claim does not include any additional element, thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception.
Claim 17 further recites in which the decision theoretic process calculates an outcome benefit. Under the broadest reasonable interpretation, this limitation covers mental process including an evaluation or judgement that could be performed in the human mind or with the aid of pencil and paper therefore it falls within the “Mental Process” grouping of abstract ideas. Also, it also falls under the “Mathematical concepts” of abstract idea since it is the mathematical calculation. The claim does not include any additional element, thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception.
Claim 18 further recites in which the outcome benefit is calculated using a blood glucose index. Under the broadest reasonable interpretation, this limitation covers mental process including an evaluation or judgement that could be performed in the human mind or with the aid of pencil and paper therefore it falls within the “Mental Process” grouping of abstract ideas. Also, it also falls under the “Mathematical concepts” of abstract idea since it is the mathematical calculation. The claim does not include any additional element, thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception.
Claim 19 further recites in which the decision theoretic process has first and second expected net benefit values wherein the first and second expected net benefit values represent, respectively, the expected net benefit of predicting hypoglycemia and predicting the absence of hypoglycemia, and the hypoglycemia threshold value correlates with the first expected net benefit value being greater than the second expected net benefit value. . Under the broadest reasonable interpretation, this limitation covers mental process including an evaluation or judgement that could be performed in the human mind or with the aid of pencil and paper therefore it falls within the “Mental Process” grouping of abstract ideas. The claim does not include any additional element, thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception.
Claim 20 further recites in which the decision theoretic process comprises:
selecting a BTP value, the BTP value representing a benefit associated with a true positive (TP) prediction of the IH event; selecting a BTN value, the BTN value representing a benefit associated with a true negative (TN) prediction of the IH event; selecting a BFP value, the BFP value representing a benefit associated with a false positive (FP) prediction of the IH event; selecting a BFN value, the BFN value representing a benefit associated with a false negative (FN) prediction of the IH event; calculating a p(IH)Crit value, the p(IH)Crit value representing an expected net benefit for predicting the IH event that is greater than the expected net benefit of predicting the absence of IH event, wherein: p(IH)Crit > ((BTN -BTP)/( BFP - BFP + BTP - BFN)); calculating a minimum nocturnal glucose prediction threshold value ( gTH) wherein p(IH)Crit = 0.5 x [1 + erf(x - (gTH + pe)/l2Ge)] where pe and ae are the average and standard deviation of errors made by the hypoglycemia prediction model, and where x is a glucose concentration threshold used to define IH; and setting the hypoglycemia threshold value equal to the gTH. Under the broadest reasonable interpretation, this limitation covers mental process including an evaluation or judgement that could be performed in the human mind or with the aid of pencil and paper therefore it falls within the “Mental Process” grouping of abstract ideas. Also, this limitations also falls under mathematical concepts since it involves mathematical calculation. So, it falls under the combination of mental process and mathematical concepts of abstract ideas. The claim does not include any additional element, thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception.
Claim 21 further in which the BTP, BTN, BFP, and BFN values are selected using low blood glucose index LBGI and high blood glucose index HBGI benefit estimates wherein Bj = (LBGIni – LBGIi) + (HBGIni- HBGIi) where subscript j represents a category from {TP, FP, FN, TN}corresponding respectively to true positive, false positive, false negative and true negative outcomes and superscripts ni and i correspond to metrics calculated when no intervention is performed and when an intervention is performed based on predictions for several decision thresholds, respectively. This limitation also falls under mathematical concepts since it involves mathematical calculation or equation. So, it falls under the mathematical concepts of abstract ideas. The claim does not include any additional element; thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception.
Claim 22 further recites in which the glucose management data includes one or more of the following: historical values related to insulin therapy, glucagon therapy, nutrients, meals, physical activity, and sleep of the living subject. The limitation amounts to data gathering and thus, falls under the insignificant extra solution activity as discussed in MPEP § 2106.05(f). The claim does not include any additional element, thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception.
Claim 24 further recites in which the amount of carbohydrate, the reduction of insulin, or amount of glucagon is based on an estimated net benefit calculation. Under the broadest reasonable interpretation, this limitation covers mental process including an evaluation or judgement that could be performed in the human mind or with the aid of pencil and paper therefore it falls within the “Mental Process” grouping of abstract ideas. The claim does not include any additional element, thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception.
Claim 25 further recites in which the amount of carbohydrate, the reduction of insulin, or amount of glucagon is based on a linear regression model that correlates the amount of the carbohydrate, the reduction of insulin, or the amount of glucagon with a predicted risk of hypoglycemia. Under the broadest reasonable interpretation, this limitation covers mental process including an evaluation or judgement that could be performed in the human mind or with the aid of pencil and paper therefore it falls within the “Mental Process” grouping of abstract ideas. The claim does not include any additional element; thus, it does not integrate the judicial exception into a practical application nor amount to significantly more than the judicial exception.
Claim Rejections - 35 USC § 102
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 the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
7. Claim(s) 1-3, 8-18, 22 and 24-25 is/are rejected under 35 U.S.C. 102(a)(2) as being anticipated by Constantin et al. (US20190252079A1)
Regarding claim 1
Constantin teaches a method, performed by a continuous glucose monitoring (CGM) system including a processor and a non-transitory computer readable medium storing instructions thereon that, when executed by the processor, configure the CGM system for supporting glucoregulatory management decisions using a decision support recommender to predict for a living subject when cognizant potential future occurrence of an incognizant hypoglycemic (IH) event projected to occur during a period of incognizance extending a beyond a transient postprandial window, (see para 300- Decision support tools may, for example, help a patient respond to a problem in real time by predicting hypoglycemia or hyperglycemia events or trends, providing treatment recommendations to address occurring or potential hypoglycemia or hyperglycemia events or trends, and monitor how the glycemic, physiologic, or behavioral response in real time. see para 314-315-An example real-time decision support system may include a source of real-time information about a patient, such as a continuous glucose monitor (“CGM”, further described below). A continuous glucose monitor may provide information about a glucose level of a patient (“CGM data”) at periodic intervals, such as once a minute or once every five minutes, or upon request. The information may be pushed out by the CGM, or may be pulled by an external device, e.g. by a smart phone or other hand-held device that is held or swept over a sensor. An example real-time decision support system may also include a processing system that receives CGM data and other information such as insulin delivery, caloric consumption, activity, wellness or sickness status, and other sensed or user-entered information about the status, environment, behavior, or wellness of the patient. See para 331-In various examples, the decision support engine 104 may reside on a remote resource 109, a local device 108, or a combination thereof (e.g. a “hybrid configuration”). See para 343-In the severe case, the patient's ability to administer or participate in following guidance may diminish, as the patient may lose cognitive capacity or consciousness with very low glucose levels. See para 415-In some examples, the system may provide nighttime readiness guidance to provide a user comfort at night that they will not trend to low glucose levels at night, to reduce nighttime alerts, and to reduce nighttime hypo events. To develop long range (e.g., 6-10 hours) or high confidence predictions. See para 436-In some examples, a decision support system may predict blood glucose levels in advance for a specified time period (e.g., 30 minutes, several hours, or six hours. see also para 334)) comprising:
receiving, via a glucose sensor, glucose management data obtained from a living subject; (see para 306-307 Exemplary embodiments disclosed herein relate to the use of a glucose sensor that measures a concentration of glucose or a substance indicative of the concentration or presence of another analyte. The glucose sensor can use any known detection method, including invasive, minimally invasive, and non-invasive sensing techniques, to provide a data stream indicative of the concentration of the analyte in a host. The data stream is typically a raw data signal that is used to provide a useful value of the analyte to a user, such as a patient or health care professional (HCP, e.g., doctor, physician, nurse, caregiver), who may be using the sensor. Also see para 325-326)
processing, with the processor of the CGM system, the glucose management data as prescribed by a feature extractor to generate a set of glucoregulatory feature values; (see para 576- generate transformed sensor data and displayable sensor information, e.g., via a processor module. For example, the processor module may transform sensor data into one or more of the following: filtered sensor data (e.g., one or more filtered analyte concentration values), raw sensor data, calibrated sensor data (e.g., one or more calibrated analyte concentration values), rate of change information, trend information, rate of acceleration/deceleration information, sensor diagnostic information, location information, alarm/alert information, calibration information such as may be determined by calibration algorithms, smoothing and/or filtering algorithms of sensor data, and/or the like. See also para 526- Yet another type of input involves derived data. Derived data, which also may also in some cases constitute real-time data, includes an analyte rate of change or acceleration, various types of sensitivities, including sensitivities to insulin, sleep, meal, and/or exercise, a hierarchy within a risk stratification (described in greater detail below), glucose variability, user stress/emotions/feeling data, a user desire for interaction with the device, sensor accuracy/confidence/signal quality, pattern data, and the like.)
applying, with the processor of the CGM system, the glucoregulatory feature values as prescribed by the feature extractor to a hypoglycemia prediction model, the hypoglycemia prediction model being formulated to predict the potential future occurrence of the IH event. (See para 329- The decision support engine may include one or more models 110. see para 436-437- In some examples, a decision support system may predict blood glucose levels in advance for a specified time period (e.g., 30 minutes, several hours, or six hours.) The prediction of blood glucose levels may, for example, be based upon dietary information (e.g., amount consumed, time of consumption, carbohydrate and fat or protein content, absorption rate (fast, medium, slow carbs), confidence in food estimate), bolus information (time, insulin action, and bolus type (regular or dual-wave), basal rates, exercise, CGM data, blood glucose (e.g., figure stick data), alcohol consumption, stress factors (e.g., based on behavior or a calendar), or time of day. The decision support system may predict glucose levels in real-time to provide advance notice of potential hypoglycemic or hyperglycemic events before they happen. The predicted glucose level may also be used to deliver early guidance to deliver a correction bolus before a hyperglycemic event occurs or before the patient becomes insulin resistant due to high glucose levels.)
generating with the processor of the CGM system, a glucoregulatory management recommendation as prescribed by the decision support recommender if the hypoglycemia prediction model predicts the potential future occurrence of the IH event. (see para 334- n an example, the system may promote healthy or uninterrupted sleep by determining that a problem is likely to occur at a time when a user is likely to be sleeping (e.g., 2 AM) and calculating a time to deliver guidance when the user is likely to be available (e.g., 9 PM). The system may recommend an action, such as eating a pre-sleep snack to avoid a night-time low, to avoid the need to wake the user during the night. In an example, the system may receive a glucose value and determine, based on a learned model of the patient, that the patient will likely have a hypoglycemic event when the user (caregiver or patient) is likely to be sleeping, and determine guidance and a time to deliver the guidance to avoid interrupting sleep of a patient or caregiver. See para 437- The predicted glucose level may also be used to deliver early guidance to deliver a correction bolus before a hyperglycemic event occurs or before the patient becomes insulin resistant due to high glucose levels. See para 347- The guidance may include alerts, recommendations, or other useful information. See para 409- The probability of state transitions may be used to determine guidance, or whether and when to deliver guidance. For example, when a probability of a transition to an undesirable state is detected, guidance may be determined and delivered.)
communicating the glucoregulatory management recommendation to the living subject via a connected device equipped with a user interface, which presents the recommendation and thereby enables prevention of the predicted hypoglycemic event when the living subject adheres to the recommendation, (see para 300- In some examples, delivered guidance may assist patients, caregivers, or healthcare providers improve lifestyle or clinical/patient outcomes by meeting a variety of challenges, such as overnight glucose control (e.g., reduce incidence of hypoglycemic events or hyperglycemic excursions), glucose control during and after meals (e.g. use historical information and trends to increase glycemic control), hyperglycemia corrections (e.g., increase time in target zone while avoiding hypoglycemic events from over-correction), hypoglycemia treatments (e.g., address hypoglycemia while avoiding “rebound” hyperglycemia), exercise, and other health factors. see para 324-The decision support engine 104 may produce a variety of outputs, which may be provided to the patient 102 as guidance via a user interface 108, or may be used to determine guidance, or may be used to determine the timing of guidance. The guidance may, for example, include diabetes therapy guidance. See also para 349- In various examples or guidance decision-support requests may use text, email, app notifications, phone calls, or other forms of communication.)
in which the glucoregulatory management recommendation is that the living subject ingest an amount of carbohydrate, reduce an insulin dose, or inject glucagon. (See para 302- For example, a system may use an algorithm or model to determine whether a hyperglycemia or hypoglycemia event is possible or likely and develop guidance, which may for example include a pre-sleep action item such as insulin delivery, eating a food item (e.g., fast carbs, slow carbs, or carbs in combination with protein), or setting an alarm to check a status or check for guidance at a particular time. See para 359- n some examples, the system may receive a user request for decision support for a future activity (e.g., exercise, driving, meal, or sleep), and the system may provide suggested therapy adjustments or preparatory steps as guidance. For example, the guidance may include changes in basal insulin, bolus insulin calculations, and suggested snack amounts)
Regarding claim 2
Constantin further teaches in which the decision support recommender is trained using a supervised training process. (see para 399-402-In some examples, the system may enter a “learning mode” in which the system observes physiologic or behavior data. For example, the system may collect CGM values, blood glucose measurement, insulin doses, food, activity, medication, stress levels, sleep patterns, hormone cycles, location (via GPS, etc.) and other information to build an information set from which patterns may be deduced. In some example, the system may request that the user eat specified foods or ask the user to consume food with a nutritional label and take a picture to confirm. Because the carb content is established with confidence and other sources of glucose variability is minimized, the system can determine an ICR for the patient. The system may then ask the user to take a correction bolus. The system may then monitor the glucose response and use the response to determine an insulin sensitivity factor. See para 322-Machine learning may also be used to identify sets of input variables that lead to a consistent insulin sensitivity model that can be applied when in that state.)
Regarding claim 3
Constantin further teaches in which the supervised training process includes one or both of a regression-based process or a classifier-based process. see para 399-402-In some examples, the system may enter a “learning mode” in which the system observes physiologic or behavior data. For example, the system may collect CGM values, blood glucose measurement, insulin doses, food, activity, medication, stress levels, sleep patterns, hormone cycles, location (via GPS, etc.) and other information to build an information set from which patterns may be deduced. In some example, the system may request that the user eat specified foods or ask the user to consume food with a nutritional label and take a picture to confirm. Because the carb content is established with confidence and other sources of glucose variability is minimized, the system can determine an ICR for the patient. The system may then ask the user to take a correction bolus. The system may then monitor the glucose response and use the response to determine an insulin sensitivity factor. See para 322-Machine learning may also be used to identify sets of input variables that lead to a consistent insulin sensitivity model that can be applied when in that state.)
Regarding claim 8
Constantin further teaches in which the supervised training process includes at least one of an insulin-metabolism impact model, a glucagon-metabolism impact model, and a nutrient- metabolism impact model. (See para 385-In an example, the physiology model 112 may characterize the flow of energy in the human body, which may include the effect of meals, exercise, glucose and insulin, as well as other hormones such as cortisol and adrenaline. See fig 2B)
Regarding claim 9
Constantin further teaches in which the nutrient-metabolism impact model includes a carbohydrate model, a protein model, or a fats model. (See para 374-Physiologic Model-Food consumption may be provided as an input to the physiology model 112. The food consumption information may include information about meals, snacks, and beverages, such as the size, content (carbohydrate, fat, protein), sequence of consumption, and time of consumption)
Regarding claim 10
Constantin further teaches in which the supervised training process includes an exercise- metabolism impact model. (See para 385- In an example, the physiology model 112 may characterize the flow of energy in the human body, which may include the effect of meals, exercise, glucose and insulin, as well as other hormones such as cortisol and adrenaline. See para 441- Exercise can affect blood sugar level quickly (e.g., almost immediately), and for up to 48 hours after exercise is completed. The system may also suggest a possible temporary basal change if exercise will (or does) last for longer than one hour. Because exercise can either increase blood glucose (e.g., due to adrenaline secretion which triggers release of glycogen) or decrease blood glucose (due to activity), the system may account for the type of exercise (e.g., competitive vs. training))
Regarding claim 11
Constantin further teaches in which the supervised training process includes a stress- metabolism impact model. (see para 317-In some examples, a physiologic state of the patient may be determined by applying one or more inputs to a physiologic model. A determined physiologic state may also be or include activity level (e.g. exercise detected from an accelerometer, heart rate sensor, or respiration sensor), metabolic drive (e.g. glucose consumption rate), and insulin resistance (e.g. in response to stress hormone secretion, sickness, or high blood sugar levels), insulin sensitivity (discussed in greater detail below), insulin on board, insulin time to action, insulin action time, carb to insulin ratio. See para 399- In some examples, the system may enter a “learning mode” in which the system observes physiologic or behavior data. For example, the system may collect CGM values, blood glucose measurement, insulin doses, food, activity, medication, stress levels, sleep patterns, hormone cycles, location (via GPS, etc.) and other information to build an information set from which patterns may be deduced)
Regarding claim 12
Constantin further teaches in which the supervised training process includes a sleep- metabolism impact model. (see para 399- In some examples, the system may enter a “learning mode” in which the system observes physiologic or behavior data. For example, the system may collect CGM values, blood glucose measurement, insulin doses, food, activity, medication, stress levels, sleep patterns, hormone cycles, location (via GPS, etc.) and other information to build an information set from which patterns may be deduced.)
Regarding claim 13
Constantin further teaches in which the supervised training process uses virtual patient data, the virtual patient data including single or multiple hormone virtual patient data. (See para 505- Hormone sensors may provide another source of entered data, and in one example may also be a source of data related to the menstrual cycle. Other hormone sensors may include those that sense cortisone and/or epinephrine. Hormone sensors or the like may also be employed to provide a measure of “energy in” versus “energy out”, particularly with regard to a parameterized system model of the patient.)
Regarding claim 14
Constantin further teaches in which the virtual patient data are generated using a glucoregulatory model. (See para 505- Hormone sensors may provide another source of entered data, and in one example may also be a source of data related to the menstrual cycle. Other hormone sensors may include those that sense cortisone and/or epinephrine. Hormone sensors or the like may also be employed to provide a measure of “energy in” versus “energy out”, particularly with regard to a parameterized system model of the patient.)
Regarding claim 15
Constantin further teaches in which the supervised training process uses validated or actual patient data. (see para 399-400- In some examples, the system may enter a “learning mode” in which the system observes physiologic or behavior data. For example, the system may collect CGM values, blood glucose measurement, insulin doses, food, activity, medication, stress levels, sleep patterns, hormone cycles, location (via GPS, etc.) and other information to build an information set from which patterns may be deduced. A system may receive an input from a user to initiate a basal test. The system may then evaluate blood glucose concentration levels for a period of time (e.g., three hours) after the last meal. If the blood glucose levels are stable through the period of time, the system may declare that the basal rate(s) are correct. If there are periods of time when the BG is increasing or decreasing, then basal rates affecting that period of time may be modified by small, safe increments, until the blood glucose is stable over the whole period. In some examples, this process may be repeated until a full 24-hour period (or longer, e.g. a week) has been covered by a successful basal test.)
Regarding claim 16
Constantin further teaches in which a hypoglycemia threshold value is selected using a decision theoretic process for decision under uncertainty. (See para 304- A decision support system may use a variety of sources of information to determine guidance, such as patterns or other information relating to concentration thresholds (e.g., below 70 mg/dL, below 50 mg/dL), percentage of time above one or more glucose concentrations threshold (e.g. above 180 mg/DL, above 250 mg/DL. See page 604- The user interface 3301 may show a low glucose concentration threshold (e.g., 70 mg/dL) 3302, a high glucose concentration threshold (e.g., 200 mg/dL) 3304, and a target 3310 graphed against a timeline (e.g., on the x-axis.) A first trend portion 3306 shows a trend that progresses into a hypoglycemic event 3312 where blood glucose is below the low threshold 3302. See 409-410-The probability of state transitions may be used to determine guidance, or whether and when to deliver guidance. For example, when a probability of a transition to an undesirable state is detected, guidance may be determined and delivered. For example, guidance may be delivered when both a glucose state and an exercise state satisfy specified conditions, e.g. guidance may be delivered when a low glucose state is present or likely, and a patient is about to exercise, or just completed exercising. See para 508-Goal data may include an indication of priorities in terms of glucose control, e.g., hypoglycemic versus hyperglycemic avoidance. Different control types may be provided as preprogrammed profiles that a user may choose in selecting a goal. For example, a user may choose a goal of hypo-minimizing control, hyper minimizing control, frequent versus infrequent correction boluses, and so on, so as to fit different treatment styles and goals.)
Regarding claim 17
Constantin further teaches in which the decision under uncertainty theoretic calculates an outcome benefit. (see para 479- The user request may, for example, be associated with data entry of a planned activity. The calculated insight may indicate a user act calculated by one or more of the models to result in a desired outcome associated with the glucose concentration value, such as maintaining the glucose concentration level (or a rate of change thereof) in a target range, or above or below a certain level. The planned activity may, for example, be a meal, and the calculated insight may be a calculated or predicted effect of the meal on the glucose concentration value, or a strategy (e.g., insulin delivery or activity or both) to manage glucose concentration levels after the meal. In some examples, the calculated insight may include an interactive recommendation and at least one factor used in determining the interactive recommendation.)
Regarding claim 18
Constantin further teaches in which the outcome benefit is calculated using a blood glucose index. (see para 479-480- The user request may, for example, be associated with data entry of a planned activity. The calculated insight may indicate a user act calculated by one or more of the models to result in a desired outcome associated with the glucose concentration value, such as maintaining the glucose concentration level (or a rate of change thereof) in a target range, or above or below a certain level. The planned activity may, for example, be a meal, and the calculated insight may be a calculated or predicted effect of the meal on the glucose concentration value, or a strategy (e.g., insulin delivery or activity or both) to manage glucose concentration levels after the meal. In some examples, the calculated insight may include an interactive recommendation and at least one factor used in determining the interactive recommendation. In some examples, the causing a display of the insight may be initiated by an occurrence of an event that matches a predetermined condition, such as a blood glucose level)
Regarding claim 22
Constantin further teaches in which the glucose management data includes one or more of the following: historical values related to insulin therapy, glucagon therapy, nutrients, meals, physical activity, and sleep of the living subject. (see para 382-383-In some examples, model inputs may be inferred, e.g. from one or more of historical user inputs (e.g. a meal diary), geolocation, insulin dosing, CGM features (rising blood sugar), or time of day. In some examples, the model may operate in a learning mode, where a database of information (historical or real-time) is collected. The activity level state may, for example, include four states: sleeping, resting, active, exercising. Insulin sensitivity may be determined using historical data, real-time data, or a combination thereof, and may, for example, be based upon food consumption, insulin delivery, and resulting glucose levels. Insulin on board may be determined using insulin delivery information, as well as known or learned (e.g. from patient data) insulin time action profiles, which may account for both basal metabolic rate (update of insulin to maintain operation of the body) and insulin usage driven by activity or food consumption)
Regarding claim 24
Constantin further teaches in which the amount of carbohydrate, the reduction of insulin, or amount of glucagon is based on an estimated net benefit calculation. (see para 298-The present inventors have recognized, among other things, that intensive insulin users can benefit from real-time guidance that is determined or delivered at a time that is calculated to be useful to a for a patient or caregiver. See para 532-Meal data may also be a form of derived data as the same may be derived from food patterns, food libraries, and so on. In another way of deriving meal data, retrospective analytics may be employed to use other data to estimate carb content for a given meal. Such other data may include, e.g., insulin delivered, glucose levels, exercise, health, meal time, and the like. By collecting such data, the decision-support application/functionality may be enabled to determine a user's typical meal and typical glucose response to that meal while mitigating the effect of other factors on their glucose. Such functionality has several benefits, including improving accuracy of bolus calculators, the ability to optimize alerts if the system notices an atypical glucose response, assistance to users in understanding their eating patterns and the effects of different meals on their glucose, and more accurate calculation of insulin to carb ratios.)
Regarding claim 25
Constantin further teaches in which the amount of carbohydrate, the reduction of insulin, or amount of glucagon is based on a linear regression model that correlates the amount of the carbohydrate, the reduction of insulin, or the amount of glucagon with a predicted risk of hypoglycemia. (See para 371 and fig 2B- The physiology model 112 or behavior model 114 may also receive information received from the patient, such as calendar information from a computer system such as a smart phone, or user input about specific activity or exercise, or user input about food consumed. A detailed description of the behavior model inputs and outputs is provided below in relation to FIG. 2B, and numerous inputs are shown and described in FIGS. 14-16. A measurement model 154 may optionally be included in the system. Data from the sensor system 152 may be provided to the measurement model 154, which may process the sensor data to assess the accuracy or precision of the data, e.g. to determine a likelihood that a measured glucose level matches a true glucose level. The sensor system may use statistical methods such as variability or variance of sensor data points, trend information, historical information, as well information provided by the behavior model, physiology model, and other sensors, to assess whether one or more sensor data points are likely accurate or inaccurate. For example, when successive blood glucose data vary widely or reveal patterns that do not reflect normal physiological patterns (e.g. 90 mg/dl, 112 mg/dl, 96 mg/dl, 121 mg/dl in successive five-minute increments), the measurement model may determine that sensor data is relatively inaccurate. see para 399-In some examples, the system may enter a “learning mode” in which the system observes physiologic or behavior data. For example, the system may collect CGM values, blood glucose measurement, insulin doses, food, activity, medication, stress levels, sleep patterns, hormone cycles, location (via GPS, etc.) and other information to build an information set from which patterns may be deduced. See para 322-Machine learning may also be used to identify sets of input variables that lead to a consistent insulin sensitivity model that can be applied when in that state. See also para 399-402))
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 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.
8. The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied 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.
9. This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
10. Claim 4 is rejected under 35 U.S.C. 103 as being unpatentable over Constantin et al. (US20190252079A1) in view of Davis et al. (US20170220751A1)
Regarding claim 4
Constantin does not teach in which the regression-based process is a support vector regression or a neural-network regression.
In the related field of invention, Davis further teaches in which the regression-based process is a support vector regression or a neural-network regression. (See para 145-In this implementation, some of the models/algorithm architectures that may be used include algorithms and models that take a layered approach, e.g., where each additional data available is added to the model to improve its ability to classify or predict. For example, logistic regression models allow easily expanding variables within the model to help predict outcomes better when more information is available. Decision trees and neural networks may also be used, as may fuzzy logic models to interpret subjective information or user-entered information.)
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method for supporting glucoregulatory management decisions using a decision support recommender to predict an occurrence of a hypoglycemic event as disclosed by Constantin to include in which the regression-based process is a support vector regression or a neural-network regression as taught by Davis in the system of Constantin in order to minimize the impact of the vagaries of diabetes on individuals, i.e., by looking for patterns and tendencies of an individual and customizing the management to that individual. Consequently, the same reduces the uncertainty that diabetes typically is associated with and improves quality of life. (See Abstract, Davis)
11. Claims 5-7 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Constantin et al. (US20190252079A1) in view of Oviedo et al. (Risk-based postprandial hypoglycemia forecasting using supervised learning." International journal of medical informatics 126 (2019): 1-8.)
Regarding claim 5
Constantin does not teach in which the classifier-based process is at least one of a random forest process, a decision tree process, and a support vector machine process.
In the related field of invention, Oviedo teaches in which the classifier-based process is at least one of a random forest process, a decision tree process, and a support vector machine process. (See section 2.2- Support vector machines are widely used supervised machine learning algorithms that model a separating hyperplane in a multidimensional space to solve a given classification task. Given a training set of N instances of features x and their corresponding labels y ∈ − 1, 1, the parameters ω, b are used to model the classifier.)
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method for supporting glucoregulatory management decisions using a decision support recommender to predict an occurrence of a hypoglycemic event as disclosed by Constantin to include in which the classifier-based process is at least one of a random forest process, a decision tree process, and a support vector machine process as taught by Oviedo a in the system of Constantin for predicting postprandial hypoglycemia using machine learning techniques personalized to each patient. The proposed system enables on-line therapeutic decision making for patients using a sensor augmented pump therapy. The results demonstrated that it is feasible to anticipate hypoglycemic events with a reasonable false-positive rate. The accuracy of the results and the trade-off between performance metrics allow its use in decision support systems for patients who wear insulin pumps. Another motivation is to generate a bolus reduction suggestion as the scaled weighted sum of the predictions. We evaluated the general and postprandial glycemic outcomes to assess the systems capability of avoiding hypoglycemia. (See Abstract, Oviedo)
Regarding claim 6
Constantin does not teach in which the supervised training process generates a set of hyperparameter values wherein hyperparameter values in the set are optimized by a cross- validation optimization process.
However, Oviedo further teaches in which the supervised training process generates a set of hyperparameter values wherein hyperparameter values in the set are optimized by a cross-validation optimization process. (See page 3 col 2 and fig 1-The general flow of the method is presented in Fig. 1. After data preprocessing, each SVC prediction model required the optimization of the hyperparameter C and the kernel parameter γ by via grid search using stratified fivefold cross-validation and MCC as a scorer. After fixing the optimized hyperparameters for each of the eight approaches (C1–C8) in Table 2, the models were tested for every patient. Stratification based on cross-validation ensured that each fold represented the overall distribution of the data. Therefore, both classes were considered to be similarly represented in every fold. In addition, the following two tasks were performed in order to test the robustness of the models. First, we randomly shuffled the data five times. Second, we trained and tested five different models with 80% and 20%. The calculated average performance metrics are reported in the Results section. It should be noted that the cross-validation and testing data subsets were identical across cases (C1–C8).)
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method for supporting glucoregulatory management decisions using a decision support recommender to predict an occurrence of a hypoglycemic event as disclosed by Constantin to include in which the supervised training process generates a set of hyperparameter values wherein hyperparameter values in the set are optimized by a cross-validation optimization process as taught by Oviedo a in the system of Constantin for predicting postprandial hypoglycemia using machine learning techniques personalized to each patient. The proposed system enables on-line therapeutic decision making for patients using a sensor augmented pump therapy. The results demonstrated that it is feasible to anticipate hypoglycemic events with a reasonable false-positive rate. The accuracy of the results and the trade-off between performance metrics allow its use in decision support systems for patients who wear insulin pumps. Another motivation is to generate a bolus reduction suggestion as the scaled weighted sum of the predictions. We evaluated the general and postprandial glycemic outcomes to assess the systems capability of avoiding hypoglycemia. (See Abstract, Oviedo)
Regarding claim 7
Constantin does not teach in which the set of hyperparameter values includes a tolerance hyperparameter and a regularization hyperparameter.
However, Oviedo further teaches in which the set of hyperparameter values includes a tolerance hyperparameter and a regularization hyperparameter. (see page 2-where C > 0 is the regularization parameter that controls the trade-off between the generalization capability and the training error. See also section 2.4 and page 3-The Matthews correlation coefficient (MCC) is used to merge the confusion matrix into a single metric that correlates the target and the predicted binary outcomes, which is desirable because it allows a unique metric to be optimized when training each model. The MCC index returns a value in the range of [− 1, 1], where 1 is a perfect prediction and -1 is an erroneous classification. These metrics are defined in Table 3. The MCC was used in the hyper-parameter tuning grid search and the selection of the best model from C1-C8.)
Regarding claim 19
Constantin does not teach in which the decision theoretic process has first and second expected net benefit values wherein the first and second expected net benefit values represent, respectively, the expected net benefit of predicting hypoglycemia and predicting the absence of hypoglycemia, and the hypoglycemia threshold value correlates with the first expected net benefit value being greater than the second expected net benefit value.
However, Oviedo further teaches in which the decision theoretic process has first and second expected net benefit values wherein the first and second expected net benefit values represent, respectively, the expected net benefit of predicting hypoglycemia and predicting the absence of hypoglycemia, and the hypoglycemia threshold value correlates with the first expected net benefit value being greater than the second expected net benefit value. (see page 7 and table 3-4 and 6-In addition, patient-specific bolus insulin dosage optimization demands assessment of the pre-prandial calculation of the postprandial glucose levels based on an accurate prediction. An online dosage optimization tool is a target application that could be developed based on the results of this study. For instance, an online personalized hypoglycemia predictive model can compute the outcome from the insulin bolus at mealtime. If the prediction is positive, then this will be interpreted as an excessive amount of insulin and the bolus should be reduced by a fraction. This new bolus amount will be part of a new computation in the predictive model together with the other unaffected inputs. This process should be repeated until the prediction is a negative class, i.e., a glucose level not below 70 mg/dL (Level 1) or 54 mg/dL (Level 2). Other approaches could use the risk assessment to simultaneously adjust the bolus and basal insulin for the next few hours. Using this method, many hypoglycemic events could be avoided but without increasing the postprandial peak.)
PNG
media_image1.png
280
525
media_image1.png
Greyscale
PNG
media_image2.png
426
1094
media_image2.png
Greyscale
Examiner note: Examiner consider the SE and SP are the first and second expected net benefit values of predicting hypoglycemia and predicting the absence of hypoglycemia. The hypoglycemia threshold value (70 or 54mg/dL) correlates with the first expected net benefit value being greater than the second expected net benefit value.
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method for supporting glucoregulatory management decisions using a decision support recommender to predict an occurrence of a hypoglycemic event as disclosed by Constantin to include in which the decision theoretic process has first and second expected net benefit values wherein the first and second expected net benefit values represent, respectively, the expected net benefit of predicting hypoglycemia and predicting the absence of hypoglycemia, and the hypoglycemia threshold value correlates with the first expected net benefit value being greater than the second expected net benefit value as taught by Oviedo a in the system of Constantin for predicting postprandial hypoglycemia using machine learning techniques personalized to each patient. The proposed system enables on-line therapeutic decision making for patients using a sensor augmented pump therapy. The results demonstrated that it is feasible to anticipate hypoglycemic events with a reasonable false-positive rate. The accuracy of the results and the trade-off between performance metrics allow its use in decision support systems for patients who wear insulin pumps. Another motivation is to generate a bolus reduction suggestion as the scaled weighted sum of the predictions. We evaluated the general and postprandial glycemic outcomes to assess the systems capability of avoiding hypoglycemia. (See Abstract, Oviedo)
Allowable Subject Matter
11. Claim 20-21 is objected to, but would be allowable if rewritten to overcome the 101 rejections and claim objection of the claims.
PNG
media_image3.png
790
670
media_image3.png
Greyscale
The closest pieces of prior art are the Davis et al. (US 20170220751 A1) in view of Oviedo et al. (Risk-based postprandial hypoglycemia forecasting using supervised learning." International journal of medical informatics 126 (2019): 1-8.) references. The closest references alone and in combination do not teach the claimed features as provided by the equation as claimed in the claim 20-21. Therefore, claims 20-21 as drafted, are rendered neither obvious nor anticipated by the prior art of the record and the available field of prior art. The claims would be allowable if rewritten to overcome the 101 rejections and claim objection of the claims.
Conclusion
11. Claims 1-22 and 24-25 is/are rejected.
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Zhong, et al. " US 20170053552 A1.
Discussing the method of managing use of a diabetes management. An embodiment of the method obtains a number of glycemic insight messages for delivery to a user device associated with a user of the diabetes management device, each of the glycemic insight messages conveying information regarding a relationship between an insight event derived from patient-specific historical input data and a glycemic outcome. The method continues by culling and prioritizing the number of glycemic insight messages to identify a group of insight messages intended for delivery, queuing the group of insight messages based on the culling and prioritizing, and communicating at least one of the queued insight messages to the user device.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to PURSOTTAM GIRI whose telephone number is (469)295-9101. The examiner can normally be reached 7:30-5:30 PM, Monday to Friday.
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, RENEE CHAVEZ can be reached at (571)-270-1104. 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.
/PURSOTTAM GIRI/Examiner, Art Unit 2186
/RENEE D CHAVEZ/Supervisory Patent Examiner, Art Unit 2186