Automated system for controlling blood glucose levels

Abstract

The present disclosure relates to a glycemic control system comprising a processing and control unit (105) configured to implement an automated glycemic control method, the adjustment method taking into account at least one hyperparameter having a default value, the value of the at least one hyperparameter being adjustable on the fly by the processing and control unit by an adjustment function, the processing and control unit being configured to estimate, after an adjustment period, the performance of the adjustment method by at least one performance indicator and to adjust on the fly the value of the at least one hyperparameter according to the at least one performance indicator.

Description

Technical Field

[0001] The present disclosure relates to the field of automated blood glucose regulation systems, also referred to as artificial pancreases.

Background Art

[0002] An artificial pancreas is a system that enables automatic adjustment of insulin input for a diabetic patient or patient based on the patient's blood glucose (i.e., blood sugar) history, diet history, and insulin injection history.

[0003] Examples of this type of regulation system are specifically described in International Patent Applications No. WO2018 / 055283 (DD16959 / B15018), No. WO2018 / 055284 (DD17175 / B15267), and No. WO2019 / 016452 (DD17609 / B15860), filed previously by the applicant, as well as French Patent Applications No. 18 / 52354 (DD18479 / B16770) of March 20, 2018, No. 18 / 56016 (DD18587 / B16893) of June 29, 2018, No. 18 / 00492 (DD18480 / B16894) of May 22, 2018, No. 18 / 00493 (DD18588 / B16895) of May 22, 2018, and No. 18 / 73812 (DD18986 / B17521) of December 21, 2018.

Summary of the Invention

[0004] It is desirable to at least partially improve certain aspects of known artificial pancreases.

[0005] Accordingly, one embodiment is a blood glucose regulation system comprising a processing and control unit configured to implement an automated blood glucose regulation method, the regulation method comprising at least one Height parameter (hereinafter also referred to as "hyperparameter").Taking into account, the value of at least one hyperparameter is adjustable on-the-fly by the processing and control unit through an adjustment function, and after the adjustment period, the processing and control unit estimates the performance of the adjustment method based on a performance indicator and is configured to adjust the value of at least one hyperparameter on-the-fly according to the value of at least one performance indicator, providing a blood glucose regulation system.

[0006] According to an embodiment of the present invention, the system further comprises a blood glucose sensor and an insulin injection device, and the blood glucose regulation method implemented by the processing and control unit comprises controlling the insulin injection device by taking into account the measurement value provided by the blood glucose sensor.

[0007] According to an embodiment of the present invention, the performance indicator is an indicator from the group comprising the proportion of past time of hypoglycemia or the number of occurrences of hypoglycemia during the adjustment period, the proportion of past time of hyperglycemia or the number of occurrences of hyperglycemia during the adjustment period, the proportion of past time of euglycemia during the adjustment period, an amount representing the variation of blood glucose during the adjustment period, and the number of glucose administrations recommended to the user during the adjustment period.

[0008] According to an embodiment of the present invention, the value of at least one hyperparameter is kept constant by the processing and control unit during the adjustment period.

[0009] According to an embodiment of the present invention, at least one hyperparameter is used multiple times by the processing and control unit during the adjustment period.

[0010] According to an embodiment of the present invention, at least one hyperparameter is used at least 20 times by the processing and control unit during the adjustment period.

[0011] According to one embodiment of the present invention, at least one hyperparameter is a blood glucose threshold that, when fallen below, is regarded by the user as hypoglycemia by the processing and control unit, a coefficient used by the processing and control unit to determine the dose of insulin injected into the user, an allowable threshold of the error between the blood glucose prediction made by the processing and control unit using a mathematical model and the actual blood glucose measured by the sensor, a suppression period between two consecutive glucose administration recommendations made to the user by the processing and control unit, and a value of an increase in the blood glucose target applied by the processing and control unit before the physical activity declared by the user, and is a hyperparameter from the group comprising.

[0012] According to one embodiment of the present invention, at least one hyperparameter is a coefficient used by the processing and control unit to determine the dose of insulin injected into the user, at least one performance indicator is the proportion of past time of hyperglycemia or the number of occurrences of hyperglycemia during the adjustment period, and in the step of on-the-fly adjustment of the value of at least one hyperparameter by at least one performance indicator, the processing and control unit increases the value of the coefficient when the value of the indicator is greater than the threshold.

Brief Description of the Drawings

[0013] The foregoing features and advantages, as well as others, are described in detail in the following description of specific embodiments given by way of illustration and not limitation with reference to the accompanying drawings.

[0014]

Figure 1

Figure 2

Mode for Carrying Out the Invention

[0015] Like features are designated by like reference numerals in the various figures. In particular, structural and / or functional features that are common among the various embodiments may have the same reference numeral and may be provided with the same structural, dimensional, and material properties.

[0016] For clarity, only steps and elements that are useful for understanding the embodiments described herein are illustrated and described in detail. In particular, the blood glucose measurement device and insulin injection device of the described regulation system are not described in detail, and the described embodiments are compatible with all or most known blood glucose measurement devices and insulin injection devices. Further, the hardware implementation of the processing and control unit of the described regulation system is not described in detail, and the formation of such processing and control units is within the ability of those skilled in the art based on the functional disclosure herein.

[0017] Unless otherwise specified, the expressions “about,” “approximately,” “substantially,” and “on the order of” mean within 10%, preferably within 5%.

[0018] FIG. 1 schematically shows, in block form, an example of an embodiment of an automated system for regulating the blood glucose of a patient or diabetic user.

[0019] The system of FIG. 1 includes a sensor 101 (CG) adapted to measure the blood glucose of a patient. In normal operation, the sensor 101 can be permanently placed on and inside the patient's body, for example, at the height of the abdomen. The sensor 101 is, for example, a CGM type (“Continuous Glucose Monitoring”) sensor, i.e., a sensor capable of measuring the patient's blood glucose continuously or at a relatively high frequency (e.g., at least once every 20 minutes, preferably at least once every 5 minutes). The sensor 101 is, for example, a subcutaneous blood glucose sensor.

[0020] The system of FIG. 1 further comprises an insulin injection device 103 (PMP), for example, a subcutaneous injection device. The device 103 is, for example, an insulin pump type automatic injection device comprising an insulin reservoir connected to an injection needle implanted under the patient's skin, and the pump can be electrically controlled to automatically inject a determined insulin dose at a determined time. In normal operation, the injection device 103 can be permanently placed on and inside the patient's body, for example, at the height of the abdomen.

[0021] The system of FIG. 1 further comprises a processing and control unit 105 (CTRL) connected to the blood glucose sensor 101, on the one hand, for example, by a wired link or a wireless (wireless) link, and to the injection device 103, on the other hand, for example, by a wire or a wireless link. During operation, the processing and control unit 105 can receive data related to the patient's blood glucose measured by the sensor 101 and electrically control the device 103 to inject a determined insulin dose into the patient at a determined time. In this example, the processing and control unit 105 can further receive, although not described in detail, data cho(t) representing the time variation of the amount of glucose ingested by the patient via a user interface. The processing and control unit 105 can be further adapted to receive possible complementary data, for example, data related to the user's physical activity and / or stress state, or data related to their health state.

[0022] The processing and control unit 105 can determine the insulin dose to be injected into the patient, in particular, by taking into account the history of blood glucose measured by the sensor 101, the history of insulin injected by the device 103, and the history of glucose intake by the patient, as well as possible supplementary data, for example, data related to the patient's physical activity and / or stress state. To achieve this, the processing and control unit 105 comprises a digital computing circuit (not described in detail), for example, a microprocessor. The processing and control unit 105 is, for example, a mobile device carried by the patient during the day and / or night. As an example, the processing and control unit 105 is firmly assembled to the insulin injection device 103 or the sensor 101. As a variant, the processing and control unit 105 is a device independent of the injection device 103 and the sensor 101, for example, a smartphone-type device.

[0023] The processing and control unit 105 is configured to implement an automated regulation method that can comprise a plurality of separate regulation blocks or modules corresponding to respective separate regulation modes.

[0024] In particular, the regulation method implemented by the processing and control unit 105 can comprise a block that implements an MPC-type ("Model Predictive Control") regulation method, also called a predictive control method, where the regulation of the administered insulin dose takes into account the prediction of the future trend of the patient's blood glucose over time obtained from a mathematical model, for example, a physiological model that describes the assimilation of insulin by the patient's body and its effect on the patient's blood glucose. In this operating mode, the actual blood glucose data measured by the sensor 101 is mainly used for the purpose of calibrating the mathematical model.

[0025] The adjustment method implemented by the processing and control unit 105 may further comprise a block implementing a security capping algorithm, also called a hypoglycemia minimization (HM) unit, which has the function of predicting and preventing impending hypoglycemia by blocking the insulin flow administered by the device 103 and / or by recommending glucose administration to the patient, i.e. carbohydrate intake. In fact, in certain situations, the predictions made by the mathematical model of the MPC block may not be sufficiently reliable, so that the control of the insulin injection device 103 based only on the predictions made by the mathematical model of the MPC block may not be able to correctly regulate the patient's blood glucose. The hypoglycemia minimization capping algorithm can predict the imminent risk of hypoglycemia and, if such a risk is detected, can reduce or interrupt the flow of insulin injected into the patient or even recommend glucose administration to the patient in order to avoid hypoglycemia.

[0026] The adjustment method implemented by the processing and control unit 105 may further comprise a block implementing an adjustment method of the decision matrix type (MD) that can be used as an alternative to the predictive control adjustment algorithm of the MPC block, for example when it is determined that the predictions made by the mathematical model of the MPC block are not sufficiently reliable.

[0027] The adjustment method implemented by the processing and control unit 105 may further comprise a post-meal management block (PMM) that implements a specific adjustment method during the post-meal phase declared by the user.

[0028] The adjustment method implemented by the processing and control unit 105 may further comprise a block implementing a bolus adjustment algorithm and various sensitivity parameters, for example by using a decision tree.

[0029] The adjustment method implemented by the processing and control unit 105 uses a number of parameters. Some of these parameters are fixed, i.e., they cannot be modified without completely recompiling the adjustment software, which means interrupting the adjustment to perform their update. Other parameters, called hyperparameters, can be modified on-the-fly, or instantaneously, i.e., without interrupting the adjustment. Each hyperparameter has a default value and can be adjusted by the processing and control unit 105 by an adjustment function between a minimum value and a maximum value.

[0030] As a non-limiting example, the adjustment method may use a hyperparameter called PATIENT_HYPO_LIMIT, which corresponds to a blood glucose threshold below which the user is considered hypoglycemic by the hypo-minimization (HM) security brick. This parameter is used in particular by the capping algorithm implemented by the HM brick to decide to interrupt the insulin flow administered by the device 103 and / or to recommend glucose administration to the user. This parameter has a default value on the order of 70 mg / dl, for example, but can be modified between a minimum value on the order of 60 mg / dl and a maximum value on the order of 85 mg / dl without interrupting the adjustment. By setting this parameter, the number of glucose administrations recommended to the patient and / or the number of interruptions of the insulin flow administered to the patient can be finely controlled.

[0031] Another example of a hyperparameter that can be used by the adjustment method is a coefficient hereinafter referred to as MD_BOLUS_FACTOR, which is used in the decision matrix (MD) block to determine the size (insulin dose) of the bolus to be injected into the patient at the end of the decision stage. This parameter has a default value on the order of 0.7, for example, and can be modified without interruption of adjustment between a minimum value on the order of 0.3 and a maximum value on the order of 1.3, for example. By setting this parameter, for a given patient, the amount of insulin injected when adjustment is performed by the decision matrix block can be finely controlled.

[0032] Another example of a hyperparameter that can be used by the adjustment method is an error threshold, which is hereinafter referred to as MODEL_MISMATCH and is used to estimate the reliability of the prediction made by the mathematical model of the MPC block, and to determine whether to switch from the MPC block (predictive control adjustment based on a mathematical model) to the MD block (decision matrix adjustment) or not. More specifically, the processing and control unit 105 can be configured to calculate a digital indicator representing the error between the blood glucose estimated from the model and the actual blood glucose measured by the sensor 101 during the stage of estimating the reliability of the mathematical model of the MPC block, and to compare this indicator with the threshold MODEL_MISMATCH. If the calculated error is smaller than the threshold, the adjustment continues to be implemented by the MPC predictive control adjustment block. If the calculated error is larger than the threshold, the MPC block is temporarily deactivated and the adjustment is implemented by the decision matrix MD adjustment block. The parameter MODEL_MISMATCH has a default value and can be modified without interruption of adjustment between a minimum value and a maximum value. By setting this parameter, ultimately, for a given patient, the ratio of the past time of MPC adjustment to the past time of MD adjustment can be controlled.

[0033] Another example of a hyperparameter that can be used by the adjustment method is the suppression period between two glucose administration recommendations, which is hereinafter referred to as SNACK_INHIB_DURATION. It is the minimum period after a glucose administration declared by the user, during which the Brick HM is not allowed to recommend a new glucose administration. This parameter has a default value and can be modified without interruption of adjustment between a minimum value and a maximum value. By setting this parameter, ultimately, the number of glucose administrations recommended to the patient at a given time interval for a given user can be controlled.

[0034] Another example of a hyperparameter that can be used by the adjustment method is the value of the increase in the target blood glucose level, which is applied before the physical activity declared by the user so that the entry of physical activity into the adjustment is taken into account. This parameter, hereinafter referred to as BEFORE_PA_TARGET_MAJORATION, has a default value and can be adjusted without interruption of adjustment between a minimum value and a maximum value. By setting this parameter, ultimately, the risk of hypoglycemia associated with physical activity can be controlled for a given user.

[0035] More generally, many other hyperparameters are likely to be used in the adjustment method implemented by the processing and control unit 105.

[0036] According to one aspect of an embodiment, the values of one or more hyperparameters, such as the hyperparameters from the group comprising PATIENT_HYPO_LIMIT, MD_BOLUS_FACTOR, MODEL_MISMATCH, SNACK_INHIB_DURATION, and BEFORE_PA_TARGET_MAJORATION, for example the above-mentioned parameters, are automatically adjusted according to one or more performance indicators of the adjustment system.

[0037] For this purpose, at the end of the adjustment stage implementing the hyperparameter(s) to be adjusted, the processing and control unit calculates one or more metrics representing the performance of the adjustment system during said adjustment stage and is configured to adjust, on the fly (i.e., without interrupting the adjustment), the value(s) of the hyperparameter(s) to be considered according to the calculated performance metrics.

[0038] Figure 2 shows an example of an automated adjustment method for a blood glucose regulation method that can be implemented by the system of Figure 1.

[0039] The method of Figure 2 comprises step 201 in which the system automatically adjusts the user's blood glucose. For this purpose, the processing and control unit 105 implements an adjustment method based, for example, on one or more of the above adjustment blocks. During this adjustment stage, the processing and control unit 105 determines the insulin dose to be injected into the patient, taking into account in particular the history of blood glucose measured by the sensor 101, the history of insulin injected by the device 103, and the history of glucose intake by the patient, as well as possible supplementary data, such as data relating to the patient's physical activity and / or stress state. The processing and control unit 105 further controls the insulin injection unit 103 in order to administer the determined insulin dose to the user.

[0040] During adjustment stage 201, the value of each hyperparameter for which adjustment is desired is maintained constant. The duration of adjustment stage 201 is selected such that each of the hyperparameters to be adjusted is used at least once, preferably a plurality of times, during adjustment stage 201. As an example, the duration of adjustment stage 201 is selected to comprise at least 20 occurrences of the event implementing the hyperparameter to be considered for each of the hyperparameters for which adjustment is desired.

[0041] The method of FIG. 2 further comprises, at the end of the adjustment stage 201, a step 202 of estimating the performance of the adjustment carried out in step 201. For this purpose, the processing and control unit 105 calculates or determines one or more indicators representing the performance of the adjustment implemented during stage 201.

[0042] The performance indicators are, for example, the ratio of the past time of hypoglycemia or the number of occurrences of hypoglycemia during the adjustment stage 201, the ratio of the past time of hyperglycemia or the number of occurrences of hyperglycemia during the adjustment stage 201, the ratio of the past time of normoglycemia during the adjustment stage 201, the variation of blood glucose during the adjustment stage 201, and the number of glucose administrations requested from the user during the adjustment stage 201 and are one or more indicators from the group comprising.

[0043] Hypoglycemia, a state in which the patient's blood glucose is lower than a predetermined low threshold, Hyperglycemia, a state in which the patient's blood glucose is higher than a predetermined high threshold, and Normoglycemia, a state in which the patient's blood glucose is between the hypoglycemia threshold and the hyperglycemia threshold It should be noted that they are thus defined.

[0044] The method of FIG. 2 further comprises, at the end of step 202, step 203 of adjusting the value of the hyperparameter(s) to be considered, taking into account the performance indicator(s) determined in step 202. During this step, the processing and control unit 105 modifies the value of the hyperparameter(s) to be considered according to a predetermined rule in order to improve the performance of the adjustment system. As an example, if the performance indicator comprises the percentage of past time with hypoglycemia or the percentage of past time with hyperglycemia, the adjustment of the hyperparameter(s) may attempt to reduce this percentage. If the performance indicator comprises the percentage of past time with euglycemia, the adjustment of the hyperparameter may attempt to increase this percentage. If the performance indicator comprises glucose variability, the adjustment of the hyperparameter(s) may aim to reduce this variability. If the performance indicator comprises the number of glucose administrations requested from the user during a given time period, the adjustment of the hyperparameter(s) may aim to reduce this number.

[0045] As an example, in the case of the above parameter PATIENT_HYPO_LIMIT, at step 203, if it is considered that the number of hypoglycemic episodes following a recommendation or decision based on the use of this parameter is too high, increasing the value of the threshold PATIENT_HYPO_LIMIT may be provided, for example, increasing it from 70 mg / dl to 75 mg / dl for the continuation of the adjustment.

[0046] In the case of the parameter MD_BOLUS_FACTOR, at step 203, if it is considered that the number of hyperglycemic episodes after a bolus or insulin dose injection calculated based on this parameter is too high, increasing the value of the parameter MD_BOLUS_FACTOR by, for example, 10% for the remaining adjustment may be provided.

[0047] More generally, determining the auto - tuning rules applied in step 203 to improve the performance of the adjustment system according to the hyperparameters and performance metrics considered is within the ability of those skilled in the art.

[0048] The adjustment of the hyperparameter(s) in step 203 is performed on - the - fly, that is, without interrupting the ht adjustment, by the adjustment function implemented by the processing and control unit 105.

[0049] And it is understood that steps 201 to 203 can be repeated, and the adjustment rules implemented in step 203 can take into account the change in the performance metric(s) between successive iterations to determine whether the system performance changes in the correct way.

[0050] Various embodiments and variations are described. Those skilled in the art can combine specific features of these embodiments and understand that other variations will readily occur to those skilled in the art. In particular, the described embodiments are not limited to specific examples of hyperparameters or specific examples of the performance metrics mentioned in this description. More generally, the provided method of automated on - the - fly adjustment of hyperparameters to improve the performance of the adjustment system can be implemented for hyperparameters other than those mentioned above and based on performance metrics other than those shown above.

[0051] This patent application claims the priority of French Patent Application No. FR19 / 08457, which is incorporated herein by reference.

Claims

1. A blood glucose regulation system comprising a processing and control unit configured to implement an automated blood glucose regulation method, wherein the regulation method uses at least one altitude parameter having a default value, and the value of the at least one altitude parameter can be instantaneously adjusted by an adjustment function without interruption of the adjustment by the processing and control unit, after the adjustment period, the processing and control unit estimates the performance of the regulation method by at least one performance indicator, and instantaneously adjusts the value of the at least one altitude parameter according to the at least one performance indicator without interruption of the adjustment A blood glucose regulation system configured as such.

2. a blood glucose sensor, and an insulin injection device further comprising, wherein the blood glucose regulation method implemented by the processing and control unit comprises controlling the insulin injection device by using the measurement values provided by the blood glucose sensor, The system according to claim 1.

3. The at least one performance indicator is the proportion of past time of hypoglycemia or the number of hypoglycemic states during the regulation period, the proportion of past time of hyperglycemia or the number of hyperglycemic states during the regulation period, the proportion of past time of normoglycemia during the regulation period, an amount representing the fluctuation of blood glucose during the regulation period, and the number of glucose administrations recommended to the user during the regulation period An indicator from the group comprising, the system according to claim 1 or 2.

4. The system according to any one of claims 1 to 3, wherein the value of the at least one advanced parameter is kept constant by the processing and control unit during the adjustment period.

5. The system according to any one of claims 1 to 4, wherein the at least one advanced parameter is used a plurality of times by the processing and control unit during the adjustment period.

6. The system according to claim 5, wherein the at least one advanced parameter is used at least 20 times by the processing and control unit during the adjustment period.

7. The at least one advanced parameter is a blood glucose threshold value below which the user is considered hypoglycemic by the processing and control unit, a coefficient used by the processing and control unit to determine the dose of insulin to be injected into the user, an allowable threshold value of the error between the blood glucose prediction made by the processing and control unit using a mathematical model and the actual blood glucose measured by the sensor, a suppression period between two consecutive glucose administration recommendations made to the user by the processing and control unit, and a value of increase in the blood glucose target applied by the processing and control unit before physical activity declared by the user and is an advanced parameter from the group consisting of: The system according to any one of claims 1 to 6.

8. The at least one advanced parameter is a coefficient used by the processing and control unit to determine the dose of insulin to be injected into the user, and the at least one performance indicator is the percentage of past time of hyperglycemia or the number of hyperglycemic states during the adjustment period. In the step of instantaneously adjusting without interrupting the adjustment of the value of the at least one advanced parameter by the at least one performance indicator, the processing and control unit increases the value of the coefficient when the value of the indicator is greater than a threshold value. The system according to claim 7, dependent on claim 3.

Images