DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claim 20 is rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 20 has an antecedent basis problem. The claim recites "the recommended bolus dose" but this term is only introduced in claim 19. Since claim 20 depends directly from claim 1 (not from claim 19), there is no proper antecedent basis for this definite article reference. Recommended Fix: Either change claim 20 to depend from claim 19, or change "the recommended bolus dose" to "a recommended bolus dose" in claim 20.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1 and 2 are rejected under 35 U.S.C. 103 as being unpatentable over Mastrototaro (US 7,785,313) in view of Vleugels (US 11,363,986B2).
INDEPENDENT CLAIM 1 ANALYSIS
Claim 1 discloses a method of compensating for meal ingestion in a drug delivery system comprising:
receiving notification of a meal event;
calculating a safe bolus dose to be delivered to a user;
delivering the safe bolus dose;
monitoring blood glucose levels of the user; and
compensating for further excursions in the blood glucose levels of the user in accordance with a medication delivery algorithm of the drug delivery system.
Prior Art Evidentiary Support
Primary Reference: US 7,785,313 B2 (Mastrototaro)
Element 1: A method of compensating for meal ingestion in a drug delivery system.
Disclosure: Mastrototaro discloses a closed loop/semi-closed loop infusion system
that provides therapy modification for insulin delivery.
Location: Abstract
Quote: "A closed loop/semi-closed loop infusion system provides therapy modification
and safeguards against the over-delivery or under-delivery of insulin."
Element 2: receiving notification of a meal event.
Disclosure: Mastrototaro does not explicitly disclose receiving notification of a meal
event.
Missing Element: Requires secondary reference.
Element 3: calculating a safe bolus dose to be delivered to a user.
Disclosure: Mastrototaro discloses adjusting therapy delivery parameters with safety
boundaries.
Location: abstract.
Quote: "adjust a therapy delivery parameter when the alarm is triggered, wherein the
adjusted therapy delivery parameter is limited to be within a boundary".
Element 4: delivering the safe bolus dose.
Disclosure: Mastrototaro discloses delivery of insulin therapy.
Location: Abstract.
Quote: "Thereafter, a delivery system delivers therapy at the adjusted therapy delivery
parameter".
Element 5: monitoring blood glucose levels of the user.
Disclosure: Mastrototaro discloses continuous glucose monitoring.
Location: Abstract.
Quote: "A glucose sensor system is configured to obtain a measured blood glucose
value".
Element 6: compensating for further excursions in the blood glucose levels of the user
in accordance with a medication delivery algorithm.
Disclosure: Mastrototaro discloses closed-loop control that adjusts insulin delivery
based on glucose readings.
Location: Col. 2, lines 35-40.
Quote: "a closed loop system for diabetes entails a glucose sensor and an insulin
infusion pump attached to a patient, wherein the delivery of insulin is automatically
administered by a controller of the infusion pump based on the sensor's glucose value
readings".
Secondary Reference for Missing Element: US 11,363,986 B2 (Vleugels)
Element 2: receiving notification of a meal event.
Disclosure: Vleugels discloses automated detection of meal events through
gesture-based physical behavior event detection.
Location: Abstract; Col. 10, lines 64-67; and Col. 12, lines 45-57.
Quote: "detect, based on analysis of sensor readings obtained from the sensors,
occurrences of gesture-based physical behavior events" where "the physical behavior
event of interest" includes meal ingestion events [see Abstract].
Quote: "Early meal detection capabilities can provide unique insights into eating behaviors are critical components towards the realization of an autonomous artificial pancreas system”
[see Col. 10, 64-67].
Quote: "In some embodiments, sensing devices can sense, without requiring user interaction, the start/end of a food intake event, the pace of eating, the pace of drinking, the number of bites, the number of sips, the estimation of fluid intake, and/or estimation of portion sizing. Operating with less human intervention, no human intervention, or only intervention not apparent to others will allow the devices to scale well with different meal scenarios and different social situations. Sensing might include capturing details of the food before it is consumed, as well as user actions that are known to accompany eating, such as repeated rotation of an upper arm or other hand-to-mouth motions. Sensors might include an accelerometer, a gyroscope, a camera, and other sensors [see Col. 12, lines 45-57].
In relation to claim 1, Mastrototaro discloses a method of compensating for meal ingestion in
a drug delivery system, comprising delivering a safe bolus dose, monitoring blood glucose levels of the user, and compensating for further excursions in the blood glucose levels of the user in accordance with a medication delivery algorithm of the drug delivery system. Specifically, Mastrototaro discloses a closed loop/semi-closed loop infusion system that provides therapy modification for insulin delivery. The system delivers therapy at an adjusted therapy delivery parameter and monitors blood glucose levels using a glucose sensor system configured to obtain a measured blood glucose value. The system automatically administers insulin based on the sensor's glucose value readings, thus compensating for
excursions in blood glucose levels in accordance with a medication delivery algorithm, wherein the delivery of insulin is automatically administered by a controller of the infusion pump based on the sensor's glucose value readings. However, Mastrototaro does not explicitly disclose receiving notification of a meal event as recited in claim 1.
Vleugels discloses a system that automatically detects meal events. Vleugels teaches an automated medication dosing and dispensing system with sensors to detect physical movement, including gesture-based physical behavior event detection for meal ingestion events. The system can detect, based on analysis of sensor readings obtained from the sensors, occurrences of gesture-based physical behavior events, and can automatically determine if the user has ingested a meal. This provides the missing element of receiving notification of a meal event.
Based on the above teachings, it would have been obvious to one of ordinary skill in the art at the time of filing to combine the teachings of Mastrototaro and Vleugels. Both references are directed to improving diabetes management through automated insulin delivery systems. Mastrototaro teaches the need for therapy modification based on glucose levels, and Vleugels provides a solution for automatically detecting meal events that would trigger such therapy modifications. The combination would improve the system's ability to respond to meal ingestion without requiring manual user input, thereby enhancing the automation and effectiveness of the closed-loop insulin delivery system taught by Mastrototaro. One of ordinary skill in the art would have been motivated to combine these references to provide a more automated and responsive insulin delivery system that can detect and respond to meal events without requiring manual user intervention. Therefore, the combination of Mastrototaro and Vleugels teaches all the limitations of claim 1.
In relation to claim 2, claim 2 depends from claim 1 and further recites that the notification of a meal event comprises a user selection of a button in a user interface of the drug delivery system.
Mastrototaro discloses a user interface for patient intervention and acceptance of therapy parameters. Specifically, Mastrototaro teaches prompting a patient to accept the adjusted therapy delivery parameter prior to delivering the therapy (Mastrototaro, col. 3, lines 35-37), and prompting the user to decide whether calibration is to be performed (Mastrototaro, col. 3, lines 6-12). This disclosure demonstrates that Mastrototaro's system includes user interface components for patient input and acceptance of therapy decisions.
Based on the above teachings, it would have been obvious to one of ordinary skill in the art to include a button in the user interface for meal notification as a conventional means for allowing users to manually input meal events into an automated insulin delivery system. User interface buttons for inputting meal information are well-known in the art of diabetes management devices, and providing
such a button would have been a predictable use of prior art elements according to their established
functions. Therefore, claim 2 is unpatentable over the combination of Mastrototaro and Vleugels.
Claim 3 is rejected under 35 U.S.C. 103 as being unpatentable over Mastrototaro (US 7,785,313) in view of Vleugels (US 11,363,986B2) as discussed above, and in further view of Keenan et al. (US 9,364,609B2; hereinafter “Keenan”) and Campbell et al. (US 8,467,980B2; hereinafter “Campbell”).
In relation to claim 3, claim 3 depends from claim 1 and further recites that the safe bolus dose is calculated as a function of a safe nominal insulin delivery amount and the current insulin on board.
Keenan discloses insulin on board (IOB) compensation for calculating insulin infusion rates.
Specifically, Keenan teaches an electronic controller that estimates a current IOB value that indicates an amount of active insulin in the body of the user, calculates an IOB rate based at least in part on the estimated current IOB value, and determines an adjusted insulin infusion rate based at least in part on the calculated IOB rate and an uncompensated insulin infusion rate (Keenan, Abstract; col. 1, starting in line 16 to column 2, line 14).
Campbell further discloses calculating a suggested bolus dose using an insulin on board
value. Campbell teaches that the extended bolus IOB value may be used to calculate a suggested bolus, wherein the IOB may be subtracted from a bolus dose that would otherwise be appropriate to avoid delivering a bolus dose that would result in too much insulin (Campbell, Abstract; col. 1, lines 56-61; col. 8, lines 26-33).
Based on the above teachings, it would have been obvious to one of ordinary skill in the art to combine the teachings of Keenan and Campbell with the meal compensation method of claim 1 to calculate a safe bolus dose as a function of a safe nominal insulin delivery amount and the current insulin on board. This combination prevents insulin stacking and over-delivery, which are well-known concerns in diabetes management. One of ordinary skill in the art would have been motivated to make this combination to improve the safety of bolus dosing by accounting for insulin already present in the body, thereby reducing the risk of hypoglycemia from excessive insulin delivery. Therefore, claim 3 is unpatentable over the cited combination of references.
Claim 4 is rejected under 35 U.S.C. 103 as being unpatentable over Mastrototaro (US 7,785,313) in view of Vleugels (US 11,363,986B2), Keenan et al. (US 9,364,609B2; hereinafter “Keenan”), Campbell et al. (US 8,467,980B2; hereinafter “Campbell”), as discussed above, and in further view of Cinar et al. (US 10,646,650; hereinafter “Cinar I”).
In relation to claim 4, claim 4 depends from claim 3 and further recites that the safe nominal insulin delivery amount is calculated as a percentage of a total daily insulin requirement (TDI) of the user.
Cinar I discloses methods for calculating insulin delivery based on patient-specific parameters including total daily insulin requirements. Cinar I teaches a multivariable artificial pancreas system that provides patient-specific therapy for management of diabetes (Cinar I, Abstract; col. 53, lines 41-45). The system calculates insulin delivery based on patient physiological parameters and insulin requirements. Cinar I explicitly discloses the calculation of a correction bolus based of TDD (total daily dose) (see Cinar I; col. 49, lines 38-43).
Based on the above teachings, it would have been obvious to one of ordinary skill in the art to calculate the safe nominal insulin delivery amount as a percentage of total daily insulin requirement (TDI) because this is a standard practice in diabetes management for determining appropriate bolus doses (correction doses). Using TDI or TDD as a basis for calculating bolus doses ensures that the bolus doses are proportional to the patient's overall insulin needs and provides a personalized approach to insulin dosing. One of ordinary skill in the art would have been motivated to use this well-established method to improve the accuracy and safety of the bolus dose calculations. Therefore, claim 4 is unpatentable over the cited combination of references.
Claims 5-8 are rejected under 35 U.S.C. 103 as being unpatentable over Mastrototaro (US 7,785,313) in view of Vleugels (US 11,363,986B2), Keenan et al. (US 9,364,609B2; hereinafter “Keenan”), Campbell et al. (US 8,467,980B2; hereinafter “Campbell”), Cinar et al. (US 10,646,650; hereinafter “Cinar I”), as discussed above, and in further view of Cinar et al. (US 8,690,820; hereinafter “Cinar II”).
In relation to claim 5, claim 5 depends from claim 4 and further recites that the safe bolus dose is modified by one or more of: a blood glucose correction factor that takes into account the current blood glucose levels of the user; a trend correction factor, wherein the trend correction factor corrects for trends in the user's blood glucose values for a past predetermined period of time; a rebound from hypo factor, wherein the rebound from hypo factor reduces the safe bolus dose by a predetermined amount if the user has experienced a hypoglycemic event within a predetermined time period prior to delivery of the safe bolus dose; or a nighttime factor, wherein the nighttime factor reduces the safe bolus dose by a calculated amount if the bolus dose is being administered during a nighttime period.
Mastrototaro discloses adjusting therapy delivery based on measured blood glucose values.
Specifically, Mastrototaro teaches adjusting a therapy delivery parameter when an alarm is triggered based on a measured blood glucose value (Mastrototaro, col. 3, lines 1-5; col. 3, lines 52-61). This teaches the blood glucose correction factor.
Cinar II discloses predicting future glucose levels based on past glucose trends using recursive modeling and time-series analysis. Cinar II teaches automatically predicting a future glucose level using data measured by the glucose sensor, wherein the model is recursively updated at each sampling time to dynamically capture the subject's glucose variation (Cinar II, Abstract; col. 13, lines 50-65 [time-series]; col. 19, lines 33-46 [recursive modeling]). This teaches the trend correction factor.
Mastrototaro also discloses safeguards against over-delivery and under-delivery of insulin
with alarm systems for hypoglycemia. The system provides safeguards against the over delivery or under-delivery of insulin and triggers alarms based on glucose values (Mastrototaro, Abstract; col. 1, lines 21-26). This teaches the rebound from hypo factor.
Cinar I discloses detection of sleep and its stages for adjusting insulin delivery. Cinar I teaches detection of sleep and its stages and assessment of sleep stages on glucose control for adaptive control of insulin delivery (Cinar I, Abstract). This teaches the nighttime factor.
Based on the teachings above, it would have been obvious to one of ordinary skill in the art to incorporate multiple correction factors into the safe bolus dose calculation to account for various physiological conditions that affect insulin requirements. Each of these factors addresses known
challenges in diabetes management: current glucose levels require immediate correction, glucose trends predict future needs, hypoglycemia rebound requires reduced dosing to prevent further low glucose events, and nighttime periods typically require reduced insulin due to decreased activity and different metabolic states. One of ordinary skill in the art would have been motivated to combine these factors to provide more accurate and safer insulin dosing that accounts for the complex physiological variables affecting glucose control. Therefore, claim 5 is unpatentable over the cited combination of references.
In relation to claim 6, claim 6 depends from claim 5 and further recites that the trend correction factor is based on a slope and a curvature of the user's blood glucose values for the past predetermined
period of time.
As discussed above, Cinar II discloses recursive time series modeling that analyzes glucose trends including rate of change. Cinar II teaches a recursive modeling strategy that uses time-series model identification techniques to dynamically capture the subject's glucose variation and predict future glucose levels (Cinar II, Abstract; col. 13, lines 50-65 [time-series]; col. 19, lines 33-46 [recursive modeling]).
The use of slope and curvature for trend analysis would have been obvious to one of ordinary skill in the art as standard mathematical techniques for analyzing time-series data to predict future values and trends. Slope represents the first derivative (rate of change) and curvature represents the second derivative (rate of change of the rate of change) of the glucose time series. These are conventional mathematical tools for characterizing trends in time-series data and would have been obvious methods for implementing the trend correction factor disclosed in the prior art. Therefore, claim 6 is unpatentable over the cited combination of references.
In relation to claim 7, claim 7 depends from claim 5 and further recites that nighttime mode is automatically detected using a motion profile provided by an IMU and time zone information.
Cinar I discloses detection of sleep and its stages using physiological variables, as previously discussed. Vleugels discloses using sensors to detect physical movement of a user, including inertial measurement unit (IMU) sensors for detecting gesture-based physical behavior events (Vleugels, col. 5, lines 53-67 to col. 8, lines 1-47).
Based on the above teachings, it would have been obvious to one of ordinary skill in the art to use motion sensors (IMU) and time zone information to automatically detect nighttime mode because reduced motion during nighttime hours is a well-known indicator of sleep, and time zone information provides context for when nighttime typically occurs. The combination of motion data from an IMU with time zone information would have enabled automatic detection of nighttime periods without requiring manual user input. One of ordinary skill in the art would have been motivated to make this combination to enable automatic adjustment of insulin delivery during sleep periods, thereby improving the automation and user-friendliness of the system. Therefore, claim 7 is unpatentable over the cited combination of references.
In relation to claim 8, claim 8 depends from claim 5 and further recites that a current value of the nighttime factor is lowered if, in past occurrences, the current value created an additional risk of
hypoglycemia or is raised if, in past occurrences, the current value caused blood glucose values of the user to be elevated.
Cinar II discloses adaptive model-based control that recursively updates based on patient
data. Cinar II teaches an adaptive model-based control strategy for blood glucose regulation that can dynamically respond to unpredicted glycemic variations, wherein the model is recursively updated at each sampling time to dynamically capture the subject's glucose variation (Cinar II, Abstract; col. 2, lines 24-46).
Based on the above teachings, it would have been obvious to one of ordinary skill in the art to adaptively adjust the nighttime factor based on historical outcomes because adaptive control systems that learn from past performance are well-known in the art of automated insulin delivery. This approach improves safety and efficacy by personalizing the nighttime factor to each individual's response patterns. One of ordinary skill in the art would have been motivated to implement this adaptive adjustment to optimize the nighttime factor for each patient based on their historical glucose responses, thereby improving both safety (reducing hypoglycemia risk) and efficacy (preventing elevated glucose levels). Therefore, claim 8 is unpatentable over the cited combination of references.
Claims 9-10 are rejected under 35 U.S.C. 103 as being unpatentable over Mastrototaro (US 7,785,313) in view of Vleugels (US 11,363,986B2), as discussed above, and in further view of Kamen et al. (US 10,010,669; hereinafter “Mandro”), Campbell et al. (US 8,140,275; hereinafter “Campbell II”) and Keenan et al. (US 9,364,609; hereinafter “Keenan”).
In relation to claim 9, claim 9 depends from claim 1 and further recites that delivering the safe bolus dose further comprises splitting the safe bolus dose into multiple staged doses delivered at
predetermined times after notification of the meal event.
Mandro discloses extended bolus delivery over time and staged delivery schedules. Mandro
teaches an extended bolus trajectory and a delivery schedule for that trajectory, wherein the extended bolus is delivered over a given period of time and the delivery schedule may be determined by first calculating the optimal schedule for delivery of the extended bolus (Mandro, col. 73, starting in line 65 to column 74, lines 1-3; col. 75, lines 14-20).
Campbell II further discloses extended bolus delivery over time periods. Campbell II teaches that an extended bolus may include a bolus in which a portion of the insulin is delivered immediately and a portion of the insulin is delivered over an extended period of time (Campbell II, col. 2, lines 34-41).
Based on the above teachings, it would have been obvious to one of ordinary skill in the art to split a meal bolus into multiple staged doses delivered at predetermined times to better match the glucose absorption profile from meals, particularly for meals with high fat or protein content that cause extended glucose elevation. This approach is well-known in diabetes management for improving postprandial glucose control. One of ordinary skill in the art would have been motivated to implement staged bolus delivery to provide better matching between insulin action and glucose absorption from meals, thereby improving glucose control and reducing the risk of both hyperglycemia and hypoglycemia. Therefore, claim 9 is unpatentable over the cited combination of references.
In relation to claim 10, claim 10 depends from claim 9 and further recites that the safe bolus dose is recalculated at each staged delivery.
Keenan discloses continuous recalculation of insulin delivery rates based on current IOB
and glucose levels. Keenan teaches that the system continuously estimates a current IOB
value and determines an adjusted insulin infusion rate at each sampling time (Keenan, Abstract; col. 1, starting in line 66 to col. 2, line 25).
Based on the above teachings, it would have been obvious to one of ordinary skill in the art to recalculate the safe bolus dose at each staged delivery to account for changes in glucose levels, insulin on board (IOB), and other factors that may have changed since the initial calculation. This approach provides more accurate and safer insulin dosing by responding to real-time conditions. One of ordinary skill in the art would have been motivated to implement this recalculation to improve the accuracy of staged bolus delivery by accounting for the actual glucose response and insulin on board (IOB) at each stage, rather than relying solely on the initial calculation. Therefore, claim 10 is unpatentable over the cited combination of references.
Claims 11-13 are rejected under 35 U.S.C. 103 as being unpatentable over Mastrototaro (US 7,785,313) in view of Vleugels (US 11,363,986B2), as discussed above, and in further view of Cinar et al. (US 8,690,820; hereinafter “Cinar II”) and Keenan et al. (US 9,364,609; hereinafter “Keenan”).
In relation to claim 11, claim 11 depends from claim 1 and further recites that the notification of a meal event comprises an automatic detection of ingestion of a meal based on blood glucose levels of the user received from a continuous glucose monitor.
Vleugels discloses automated detection of eating events through gesture-based physical behavior event detection, as previously discussed. Cinar II discloses using continuous glucose monitoring data for detecting disturbances. Cinar II teaches that the system uses continuous glucose monitoring (CGM) device data and includes a change detection method to ensure faster response and convergence of model parameters in presence of disturbances (Cinar II, col. 2, lines 30-40). The system can detect unpredicted glycemic variations due to physiological or external perturbations (Cinar II, col. 13, lines 59-65).
Based on the above teachings, it would have been obvious to one of ordinary skill in the art to combine gesture-based meal detection with glucose-based meal detection to provide redundant or complementary detection methods. Meals cause characteristic rises in blood glucose levels, and detecting these patterns in CGM data would provide an additional or alternative method for
identifying meal events. One of ordinary skill in the art would have been motivated to implement glucose-based meal detection because it provides an independent verification of meal ingestion and can detect meals even when gesture-based detection fails or is unavailable. Therefore, claim 11 is unpatentable over the cited combination of references.
In relation to claim 12, claim 12 depends from claim 11 and further recites that the automatic detection of the ingestion of a meal is based on detection of a significant increase in the user's blood
glucose levels and a significant increase in automatic delivery of a liquid drug as calculated by a medication delivery algorithm of the drug delivery system.
Cinar II discloses change detection methods for glucose variations, as previously discussed. Keenan discloses automatic adjustment of insulin infusion rates by the medication delivery algorithm. Keenan teaches that the system determines an adjusted insulin infusion rate based at least in part on the calculated IOB rate and an uncompensated insulin infusion rate, wherein the system automatically adjusts delivery (Keenan, Abstract; col. 1, line 66 to col. 3, line 25).
Based on the above teachings, it would have been obvious to one of ordinary skill in the art to detect meal ingestion by monitoring both glucose increases and corresponding increases in automatic insulin delivery because both are characteristic responses to meal ingestion in a closed-loop system. The combination of these two indicators provides more reliable meal detection than either indicator alone. One of ordinary skill in the art would have been motivated to use both indicators because the combination reduces false positives (glucose increases from other causes) and false negatives (meals that do not immediately raise glucose), thereby improving the accuracy and reliability of automatic meal detection. Therefore, claim 12 is unpatentable over the cited combination of references.
In relation to claim 13, claim 13 depends from claim 12 and further recites that the increase in the user's blood glucose levels and the increase in the automatic delivery of the liquid drug are evaluated for a moving window of a predetermined width evaluated each time a new blood glucose
reading is received from the continuous glucose monitor.
Cinar II discloses recursive updating at each sampling time using CGM data. Cinar II teaches
that the model is recursively updated at each sampling time to dynamically capture the subject's glucose variation using continuous glucose monitoring (CGM) device data (Cinar II, Abstract; col. 13, lines 50-65 [time-series]; col. 19, lines 33-46 [recursive modeling]). The use of a moving window for time-series analysis would have been obvious to one of ordinary skill in the art as a conventional technique in signal processing and time-series data analysis. A moving window allows the system to analyze recent trends while discarding older data that may no longer be relevant. One of ordinary skill in the art would have been motivated to use a moving window approach to focus the meal detection analysis on recent data that is most relevant to detecting a current meal event, while avoiding interference from older data that may represent previous meals or other disturbances. Therefore, claim 13 is unpatentable over the cited combination of references.
Claims 14-18 are rejected under 35 U.S.C. 103 as being unpatentable over Mastrototaro (US 7,785,313) in view of Vleugels (US 11,363,986B2), Cinar et al. (US 8,690,820; hereinafter “Cinar II”) and Keenan et al. (US 9,364,609; hereinafter “Keenan”), as discussed above, and in further view of Constantin et al. (EP3749183B1; hereinafter “Constantin”).
In relation to claim 14, claim 14 depends from claim 11 and further recites that the automatic detection of the ingestion of a meal is based on an analysis of the plot of the user's blood glucose levels by one or more machine learning models trained as meal detection classifiers.
Constantin discloses a system with stored models for decision support based on glucose
data. Constantin teaches a memory circuit including a stored model that includes a plurality of parameters for glucose-based decision support (Constantin, e.g., paragraphs [0037, 0087, 0171, 0190, 0206, 0211]).
Based on the above teachings, it would have been obvious to one of ordinary skill in the art to use machine learning models trained as meal detection classifiers because machine learning is a well-known technique for pattern recognition in time-series data. The characteristic glucose rise pattern following meal ingestion would be suitable for machine learning classification. One of ordinary skill in the art would have been motivated to employ machine learning models because they can learn complex patterns in glucose data that may not be easily captured by rule-based algorithms, and they can adapt to individual patient characteristics through training on patient-specific data. Therefore, claim 14 is unpatentable over the cited combination of references.
In relation to claim 15, claim 15 depends from claim 14 and further recites that the one or more machine learning models are trained to detect post-prandial rises in blood glucose levels of the user
beginning at various times prior to the current time.
As discussed above, Cinar II discloses prediction and detection using historical glucose data. Cinar II teaches automatically predicting a future glucose level using data measured by the glucose sensor using time-series model identification techniques and analyzing glucose variations over time (Cinar II, Abstract; col. 13, lines 50-65 [time-series]; col. 19, lines 33-46 [recursive modeling]).
Based on the above teachings, it would have been obvious to one of ordinary skill in the art to train machine learning models to detect rises beginning at various historical time points as a method to detect meals with different absorption rates and to account for delays between meal ingestion and glucose rise. Different types of meals (high carbohydrate, high fat, high protein, mixed composition) cause glucose rises with different onset times and durations. One of ordinary skill in the art would have been motivated to train models to detect rises at various historical time points to improve detection accuracy across different meal types and to enable earlier detection of meal events. Therefore, claim 15 is unpatentable over the cited combination of references.
In relation to claim 16, claim 16 depends from claim 15 and further recites that the one or more machine learning models comprise a 10-minute detector, a 15-minute detector and a 20-minute detector, wherein the one or more machine learning models are cascaded to provide a final
determination of a meal ingestion.
The specific time intervals (10, 15, 20 minutes) and cascaded architecture represent design
choices that would have been obvious to one of ordinary skill in the art. As discussed above, Cinar II discloses recursive modeling at different time scales using a recursive modeling strategy with timeseries model identification that can analyze data at different time scales. Using multiple detectors at different time intervals and cascading them for final determination would have been obvious as a method to improve detection accuracy and account for variability in meal absorption rates. Cascaded classifiers are a well-known technique in machine learning for improving classification performance, wherein multiple
classifiers are combined in sequence to make a final decision. One of ordinary skill in the art would have been motivated to use cascaded detectors at multiple time intervals to improve the robustness of meal detection by capturing meals with different glucose rise onset times, and the specific intervals of 10, 15, and 20 minutes represent reasonable design choices for capturing the typical range of meal-induced glucose rise onset times. Therefore, claim 16 is unpatentable over the cited combination of references.
In relation to claim 17, claim 17 depends from claim 11 and further recites that the automatic detection of the ingestion of a meal is based on an analysis of the plot of the user's blood glucose levels by one or more decision trees that use multiple delay detectors to recognize rising blood glucose values with an assumption that a rise in the blood glucose levels started at various times prior to the current time.
Cinar II discloses analyzing glucose trends and detecting changes with time delays. Cinar II
teaches a change detection method to ensure faster response and convergence of model parameters in presence of disturbances (Cinar II, col. 13, lines 59-65), and time-delay compensators are included in the closed-loop algorithm (Cinar II, col. 14, lines 58-63). Decision trees would have been obvious to one of ordinary skill in the art as a conventional machine learning technique for classification tasks.1 Using multiple delay detectors would have been obvious for detecting meal-induced glucose rises that may have started at different times in the past. One of ordinary skill in the art would have been motivated to use decision trees with multiple delay detectors because decision trees provide interpretable
classification rules and can efficiently handle multiple input features (such as glucose values at different historical time points), and the multiple delay detectors enable detection of meals with varying onset times. Therefore, claim 17 is unpatentable over the cited combination of references.
In relation to claim 18, claim 18 depends from claim 17 and further recites that the results of the multiple delay detectors are pooled to synthesize a final meal detection decision.
Pooling results from multiple detectors to synthesize a final decision would have been obvious to one of ordinary skill in the art as a conventional ensemble method in machine learning and signal processing. Cinar II discloses integrating multiple sources of information for control decisions, teaching that the system integrates CGM data with metabolic/physiological information collected from a body monitoring system and the insulin infusion rate from the pump (Cinar II, col. 2, lines 52-55). One of ordinary skill in the art would have been motivated to pool results from multiple delay detectors because ensemble methods that combine multiple classifiers typically provide better performance than individual classifiers by reducing variance and improving robustness. Therefore, claim 18 is unpatentable over the cited combination of references.
Claims 19 is rejected under 35 U.S.C. 103 as being unpatentable over Mastrototaro (US 7,785,313) in view of Vleugels (US 11,363,986B2), as discussed above, and in further view of Campbell et al. (US 8,467,980B2; hereinafter “Campbell”).
In relation to claim 19, claim 19 depends from claim 1 and further recites that, in response to the notification of the meal event, the method further comprises informing the user of the unannounced detection of a meal and providing a recommended bolus dose to the user. Vleugels discloses outputting medication administration messages to the user. Vleugels teaches outputting the medication administration message to the user as to a medication administration need (Vleugels, col. 3, lines 28-29; col. 4, lines 22-23).
Campbell discloses calculating and providing suggested bolus doses. Campbell teaches that the extended bolus IOB value may be used to calculate a suggested bolus (Campbell, Abstract; col. 2, lines 31-34). It would have been obvious to one of ordinary skill in the art to inform the user of an automatically detected meal and provide a recommended bolus dose to allow the user to confirm or modify the automatic detection and dosing decision. This provides a safety mechanism and allows user oversight of the automated system. One of ordinary skill in the art would have been motivated to implement this user notification and recommendation feature to provide a semi-automated system that combines the benefits of automatic detection with user supervision, thereby improving safety while maintaining automation benefits. Therefore, claim 19 is unpatentable over the cited combination of references.
Claim 20 is rejected under 35 U.S.C. 103 as being unpatentable over Mastrototaro (US 7,785,313) in view of Vleugels (US 11,363,986B2), as discussed above, and in further view of Cinar et al. (US 8,690,820; hereinafter “Cinar II”) and Keenan et al. (US 9,364,609B2; hereinafter “Keenan”)
In relation to claim 20, claim 20 depends from claim 1 and further recites that, in response to the notification of the meal event, the method further comprises adjusting a medication delivery algorithm of the drug delivery system to respond more aggressively to glucose deviations based on the recommended bolus dose.
Cinar II discloses adaptive control that dynamically responds to glycemic variations. Cinar II
teaches an adaptive model-based control strategy for blood glucose regulation that can dynamically respond to unpredicted glycemic variations due to physiological or external perturbations (Cinar II, col. 2, lines 24-28; col. 2, lines 40-51).
Keenan further discloses adjusting insulin infusion rates based on glucose deviations. Keenan teaches that the system determines an adjusted insulin infusion rate and selects a final insulin infusion rate based on current conditions (Keenan, col. 2, lines 10-14; col. 2, lines 24-31).
Based on the above teachings, it would have been obvious to one of ordinary skill in the art to adjust the medication delivery algorithm to respond more aggressively to glucose deviations following a meal because meals cause rapid glucose excursions that require more aggressive insulin delivery than basal conditions. The prior art teaches adaptive control systems that adjust their response based on detected disturbances such as meals. One of ordinary skill in the art would have been motivated to implement more aggressive algorithm response following meal detection to better control postprandial glucose excursions, which are known to be more rapid and larger in magnitude than glucose variations during fasting periods. Therefore, claim 20 is unpatentable over the cited combination of references.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MANUEL A MENDEZ whose telephone number is (571)272-4962. The examiner can normally be reached Mon-Fri 7:00 AM-5:00 PM.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Bhisma Mehta can be reached at 571-272-3383. 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.
Respectfully submitted,
/MANUEL A MENDEZ/ Primary Examiner, Art Unit 3783
1 See Blockeel et al., “Decision Trees: from efficient prediction to responsible AI”, page 2, second column, paragraph titled: “2.2 Recursive Partitioning”… “Decision trees became prominent in machine learning and data analysis around 1980s”.