Manufacturing control using process measurements for sensor calibration

Abstract

Medical devices, systems, and methods are provided. One method involves obtaining process measurement data for a plurality of instances of a sensing element; obtaining reference output measurement data from the plurality of instances in response to a reference stimulus; determining a predictive model for a measured output of the sensing element as a function of a process measurement variable based on a relationship between the process measurement data and the reference output measurement data; generating a simulated output measurement distribution across a range of the process measurement variable using the predictive model; identifying a performance threshold for the measured output based on the simulated output measurement distribution; obtaining output measurement data from an instance of the sensing element in response to the reference stimulus; and verifying that the output measurement data satisfies the performance threshold prior to calibrating subsequent instances of the sensing element.

Description

TECHNICAL FIELD

[0001] Embodiments of the subject matter described herein relate generally to medical devices, and more particularly, embodiments of the subject matter relate to calibrating sensing elements for use with medical devices. BACKGROUND

[0002] Infusion pump devices and systems are quite well known in the medical arts for delivering or dispensing a reagent, such as insulin or another prescribed medication, to a patient. Control schemes have been developed to allow an insulin infusion pump to monitor and regulate a patient's blood glucose level in a substantially continuous and autonomous manner. For purposes of continuous glucose monitoring (CGM) or determining operational commands for the infusion pump, rather than employing continuous sampling and monitoring of a user's blood glucose level, which can compromise battery life, intermittent sensed glucose data samples are typically utilized.

[0003] Many continuous glucose monitoring (CGM) sensors measure glucose in interstitial fluid (ISF). Typically, to achieve desired levels of accuracy and reliability and to reduce the effects of noise and other spurious signals, sensor data is calibrated using known good blood glucose values, typically obtained via so-called "finger stick measurements" using a blood glucose meter that measures blood glucose in capillaries. However, performing such calibration measurements increases patient burden and perceived complexity, and can be inconvenient, uncomfortable, or otherwise disliked by patients. Moreover, ISF glucose measurements lag blood glucose measurements based on the time it takes for glucose to diffuse from capillaries to interstitial space, where glucose is measured by CGM sensors, requiring signal processing (e.g., filtering) or other techniques to compensate for the physiological lag. Furthermore, various factors can cause transient changes in sensor output that can affect the accuracy of calibration. Degradation of sensor performance over time or manufacturing variations can further exacerbate these issues. Thus, it would be desirable to provide sensor calibration in a manner that mitigates patient burden and improves overall user experience without compromising accuracy or reliability. SUMMARY

[0004] Medical devices and related systems and methods of operation are provided. An embodiment of a method of calibrating an instance of a sensing element capable of providing an electrical signal affected by a physiological condition in a user's body is provided. The method involves obtaining process-of-fabrication measurement data from a substrate on which the instance of the sensing element is fabricated, obtaining a calibration model associated with the sensing element, determining calibration data associated with the instance of the sensing element based on the process-of-fabrication measurement data using the calibration model, the calibration data used to convert the electrical signal to a calibrated measurement parameter, and storing the calibration data in a data storage element associated with the instance of the sensing element.

[0005] In another embodiment, a method of operating a sensing device is provided. The method involves: obtaining, by a control module of the sensing device, one or more electrical signals from a sensing element of the sensing device, wherein the one or more electrical signals are affected by a physiological condition in a user's body; obtaining, by the control module, from a data storage element of the sensing device, calibration data associated with the sensing element; determining, by the control module, a calibrated measurement parameter based on the one or more measurement signals using the calibration data; obtaining a performance model associated with the sensing element; obtaining personal data associated with the patient; and determining a calibrated output value indicative of the physiological condition based on the personal data and the calibrated measurement parameter using the performance model.

[0006] In another embodiment, a method of calibrating an interstitial glucose sensing element is provided. The method involves: obtaining process measurement data from a substrate on which an instance of the interstitial glucose sensing element is fabricated; obtaining a calibration model associated with the interstitial glucose sensing element; determining, using the calibration model, a calibration factor associated with the instance of the interstitial glucose sensing element based on the process measurement data, the calibration factor being used to convert an uncalibrated value of a measurement parameter determined based on an output signal from the interstitial glucose sensing element to a calibrated value; and storing the calibration factor in a data storage element associated with the instance of the sensing element.

[0007] In yet another embodiment, a method of fabricating an instance of a sensing element capable of providing an electrical signal affected by a physiological condition in a user's body is provided. The method involves: obtaining process measurement data for a plurality of instances of the sensing element; obtaining reference output measurement data from the plurality of instances of the sensing element in response to a reference stimulus; determining a predictive model for a measurement output of the sensing element as a function of a process measurement variable based on a relationship between the process measurement data and the reference output measurement data; generating a simulated output measurement distribution across a range of the process measurement variable using the predictive model; identifying a performance threshold for the measurement output based on the simulated output measurement distribution; obtaining output measurement data from the instance of the sensing element in response to the reference stimulus; and verifying that the output measurement data satisfies the performance threshold prior to calibrating the instance of the sensing element.

[0008] In another embodiment, a method is provided for calibrating an instance of a sensing element capable of providing an electrical signal influenced by physiological conditions in a user's body. The method involves: obtaining a performance threshold associated with the sensing element's response to a reference stimulus, wherein the performance threshold is derived using a predictive model of the sensing element's measurement output as a function of a processing measurement variable; obtaining a current value of the measurement output from the instance of the sensing element in response to the reference stimulus; and, after verifying that the current value is within the performance threshold range, obtaining processing measurement data associated with the instance of the sensing element; obtaining a calibration model associated with the instance of the sensing element; using the calibration model based on the processing measurement data to determine calibration data associated with the instance of the sensing element, the calibration data being used to convert the electrical signal into calibrated measurement parameters; and storing the calibration data in a data storage element associated with the instance of the sensing element.

[0009] In yet another embodiment, a testing system is provided. The testing system includes: a data storage element for maintaining a performance threshold associated with the response of a sensing element to a reference stimulus, wherein the performance threshold is derived using a predictive model of the measured output of the sensing element as a function of a processing measurement variable; and a processing system coupled to the data storage element to: obtain output measurement data from an instance of the sensing element in response to a reference stimulus; and verify that the output measurement data meets the performance threshold before calibrating the instance of the sensing element according to a calibration model.

[0010] This summary is provided to introduce, in a simplified form, some concepts that will be further described in the following detailed description. This summary is not intended to identify the principal or essential features of the claimed subject matter, nor is it intended to help determine the scope of the claimed subject matter. Attached Figure Description

[0011] The subject matter can be more fully understood by taking into account the following figures, and by referring to the detailed description and claims, wherein the same reference numerals indicate similar elements throughout the figures, which are shown for simplicity and clarity, but are not necessarily drawn to scale.

[0012] Figure 1 An exemplary implementation of the infusion system is shown;

[0013] Figure 2 It is suitable for Figure 1 A block diagram of an exemplary embodiment of a sensing device used in an infusion system;

[0014] Figure 3 The following diagram shows the process and calibration suitable for use in... Figure 2 The processing system for sensing elements used in sensing devices;

[0015] Figure 4 is a cross-section of an electrode of an interstitial glucose sensing element suitable for processing with a processing system of Figure 3 for use in a sensing device of Figure 2 ;

[0016] Figure 5 is a flowchart of an exemplary process model development process suitable for use with a processing system of Figure 3 in one or more exemplary embodiments;

[0017] Figure 6 is a flowchart of an exemplary sensor initialization process suitable for use with a sensing device of Figure 5 in one or more exemplary embodiments in conjunction with the process model development process of Figure 2 ;

[0018] Figure 7 is a flowchart of an exemplary performance model development process suitable for use with a sensing device in one or more exemplary embodiments;

[0019] Figure 8 is a flowchart of an exemplary measurement process suitable for use with a sensing device in one or more exemplary embodiments in conjunction with the sensor initialization process of Figure 6 and the performance model development process of Figure 7 ;

[0020] Figure 9 is a block diagram of a data management system suitable for use with a sensing device in conjunction with one or more of the processes of Figures 5 to 8 ; and

[0021] Figure 10 is a flowchart of an exemplary sensor initialization process suitable for use with a sensing device of Figures 5 to 8 in one or more exemplary embodiments in conjunction with one or more of the processes of Figure 2 . DETAILED DESCRIPTION

[0022] The following detailed description is merely illustrative in nature and is not intended to limit or restrict the embodiments of the subject matter or the application and uses of such embodiments. As used herein, the word “exemplary” means “serving as an example, instance, or illustration.” Any implementation described herein as exemplary is not necessarily to be construed as preferred or advantageous over other implementations. Furthermore, there is no intention to be bound by any expressed or implied theory presented in the preceding technical field, background, summary or the following detailed description.

[0023] Exemplary embodiments of the subject matter described herein generally relate to calibrating sensing elements and related sensing devices and apparatuses that provide an output indicative of and / or influenced by one or more characteristics or conditions sensed, measured, detected, or otherwise quantified by the sensing element. While the subject matter described herein is not necessarily limited to any particular type of sensing application, exemplary embodiments are primarily described herein in the context of sensing elements that generate or otherwise provide an electrical signal indicative of and / or influenced by a physiological condition in a human user or patient, such as, for example, an interstitial glucose sensing element.

[0024] As described in detail below, calibration data is determined based on a calibration model associated with a sensing element using process measurement data associated with an instance of the sensing element, the calibration data used to convert electrical signals output by the instance of the sensing element to one or more calibrated measurement parameters. In this regard, the calibration model maps one or more process measurements corresponding to a zone or region of a substrate on which a particular instance of the sensing element was manufactured to a calibration factor to determine one or more calibration measurement parameters for a current instance of the sensing element. In exemplary embodiments, the calibration data is determined and stored or otherwise maintained in association with the instance of the sensing element after processing of the sensing element but prior to deployment. Thereafter, during operation, the calibration data can be utilized to convert electrical signals output by the instance of the sensing element to one or more calibrated measurement parameters. In exemplary embodiments, the calibrated measurement parameters are converted to calibrated output values indicative of a physiological condition sensed by a patient using a performance model associated with the sensing element using personal data associated with the patient or other data characterizing properties or modes of operation of the sensing element. In this way, calibrated measurement values of a physiological condition of a patient are obtained without the need for so-called "finger stick measurements" or other reference measurements.

[0025] For purposes of explanation, the exemplary embodiments of the subject matter are described herein in conjunction with implementation in conjunction with a medical device such as a portable electronic medical device. While there can be many different applications, the following description can focus on embodiments in conjunction with a fluid infusion device (or infusion pump) as part of an infusion system deployment. That is, the subject matter can be implemented in an equivalent manner in the context of other medical devices such as continuous glucose monitoring (CGM) devices, injection pens (e.g., smart injection pens), and the like. For the sake of brevity, conventional techniques related to infusion system operation, insulin pump and / or infusion set operation, and other functional aspects of the systems (and the individual operating components of the systems) can not be described in detail herein. That is, the subject matter described herein can be used in the context of overall diabetes management or in other physiological conditions independent of or without the use of infusion devices or other medical devices (e.g., when using oral medications), and the subject matter described herein is not limited to any particular type of medication. In this regard, the subject matter is not limited to medical applications and can be implemented in any device or application that includes or incorporates sensing elements.

[0026] Infusion System Overview

[0027] Figure 1 An exemplary embodiment of an infusion system 100 is shown that includes, but is not limited to, a fluid infusion device (or infusion pump) 102, a sensing arrangement 104, a command control device (CCD) 106, and a computer 108. The components of the infusion system 100 can be implemented using different platforms, designs, and configurations, and Figure 1 The illustrated embodiment is not exhaustive or limiting. In practice, the infusion device 102 and the sensing arrangement 104 are secured at desired locations on a user’s (or patient’s) body, as Figure 1 In this regard, the infusion device 102 and the sensing arrangement 104 are secured to locations on a patient’s body in Figure 1 are provided as merely representative, non-limiting examples. The elements of the infusion system 100 can be similar to those described in U.S. Patent 8,674,288, the subject matter of which is hereby incorporated by reference in its entirety.

[0028] In Figure 1In example implementations, the infusion device 102 is designed as a portable medical device adapted to infuse a fluid, liquid, gel, or other medicament into a user's body. In example implementations, the fluid infused is insulin, but many other fluids can be administered by infusion, such as but not limited to HIV medications, medications to treat pulmonary hypertension, iron chelation medications, pain medications, anti-cancer therapy medications, vitamins, hormones, etc. In some implementations, the fluid can include nutritional supplements, dyes, tracking media, saline media, hydration media, etc. Generally, the fluid infusion device 102 includes a motor or other actuation means for linearly displacing a plunger (or stopper) of a reservoir disposed within the fluid infusion device to deliver a dose of fluid, such as insulin, to the patient's body. The dose command controlling operation of the motor can be generated in an automated manner according to a delivery control scheme associated with a particular operating mode, and the manner in which the dose command is generated can be influenced by current (or recent) measurements of physiological conditions in the patient's body. For example, in a closed-loop operating mode, the dose command can be generated based on a difference between a current (or recent) measurement of interstitial fluid glucose levels in the user's body and a target (or reference) glucose value. In this regard, the rate of infusion can vary with fluctuations in the difference between the current measurement and the target measurement. For purposes of illustration, the subject matter is described herein in the context of the infusion fluid being insulin for regulating the glucose levels of a user (or patient); however, it should be understood that many other fluids can be administered by infusion, and the subject matter described herein is not necessarily limited to use with insulin.

[0029] The sensing device 104 generally represents another medical device that includes components of the infusion system 100 that are configured to sense, detect, measure, or otherwise quantify a physiological condition of a patient, and can include sensors, monitors, and the like for providing data indicative of the condition sensed, detected, measured, or otherwise monitored by the sensing device. In this regard, the sensing device 104 can include electronics and enzymes that are reactive to a biological condition of a patient, such as a blood glucose level, and provide data indicative of the blood glucose level to the infusion device 102, the CCD 106, and / or the computer 108. For example, the infusion device 102, the CCD 106, and / or the computer 108 can include a display for presenting information or data to the patient based on sensor data received from the sensing device 104, such as a current glucose level of the patient, a graph or chart of the glucose level of the patient over time, a device status indicator, an alert message, and the like. In other embodiments, the infusion device 102, the CCD 106, and / or the computer 108 can include electronics and software configured to analyze the sensor data and operate the infusion device 102 to deliver fluid to the patient’s body based on the sensor data and / or a preprogrammed delivery plan. Thus, in example embodiments, one or more of the infusion device 102, the sensing device 104, the CCD 106, and / or the computer 108 include a transmitter, a receiver, and / or other transceiving electronics that allow for communication with other components of the infusion system 100, such that the sensing device 104 can transmit sensor data or monitor data to one or more of the infusion device 102, the CCD 106, and / or the computer 108. While the subject matter is described herein in the context of glucose sensing, it should be understood that the subject matter described herein is not necessarily limited to glucose sensing, and can be implemented in an equivalent manner for any number of other different enzyme species such as, for example, lactate, beta-hydroxybutyrate, creatinine, and the like.

[0030] Still referring to Figure 1 In various embodiments, the sensing device 104 can be affixed to the patient’s body or embedded in the patient’s body at a location that is remote from a location at which the infusion device 102 is affixed to the patient’s body. In various other embodiments, the sensing device 104 can be incorporated within the infusion device 102. In other embodiments, the sensing device 104 can be separate and apart from the infusion device 102, and can be part of, for example, the CCD 106. In such embodiments, the sensing device 104 can be configured to receive a biological sample, an analyte, or the like to measure a condition of the patient.

[0031] In some embodiments, the CCD 106 and / or computer 108 can include electronics and other components configured to perform processing, deliver routine doses, and control the infusion device 102 in a manner affected by sensor data measured and / or received from the sensing device 104. By including control functionality in the CCD 106 and / or computer 108, the infusion device 102 can be made with more simplified electronics. However, in other embodiments, the infusion device 102 can include all control functionality and can operate without the CCD 106 and / or computer 108. In various embodiments, the CCD 106 can be a portable electronic device. Additionally, in various embodiments, the infusion device 102 and / or sensing device 104 can be configured to transmit data to the CCD 106 and / or computer 108 for display or processing of the data by the CCD 106 and / or computer 108.

[0032] In some embodiments, the CCD 106 and / or computer 108 can provide information to the patient to facilitate subsequent use of the infusion device 102 by the patient. For example, the CCD 106 can provide information to the patient to allow the patient to determine a rate or dose of a medication to be administered into the patient's body. In other embodiments, the CCD 106 can provide information to the infusion device 102 to autonomously control a rate or dose of a medication to be administered into the patient's body. In some embodiments, the sensing device 104 can be integrated into the CCD 106. Such embodiments can allow the patient to assess his or her condition to monitor a condition by, for example, providing a sample of his or her blood to the sensing device 104. In some embodiments, the glucose level in the patient's blood and / or bodily fluids can be determined using the sensing device 104 and the CCD 106 without using or requiring a wired or cable connection between the infusion device 102 and the sensing device 104 and / or the CCD 106.

[0033] In some embodiments, the sensing device 104 and / or the infusion device 102 are cooperatively configured to utilize a closed-loop system to deliver fluid to a patient. Examples of sensing devices and / or infusion pumps that utilize a closed-loop system can be found in, but are not limited to, U.S. Patent Nos. 6,088,608; 6,119,028; 6,589,229; 6,740,072; 6,827,702; 7,323,142; and 7,402,153, or U.S. Patent Application Publication No. 2014 / 0066889, all of which are incorporated herein by reference in their entireties. In such embodiments, the sensing device 104 is configured to sense or measure a condition of the patient, such as a blood glucose level, among others. The infusion device 102 is configured to deliver fluid in response to the condition sensed by the sensing device 104. In turn, the sensing device 104 continues to sense or otherwise quantify the current condition of the patient, thereby allowing the infusion device 102 to continuously deliver fluid in response to the condition currently (or most recently) sensed by the sensing device 104. In some embodiments, the sensing device 104 and / or the infusion device 102 can be configured to utilize the closed-loop system only for a portion of the day, for example, only when the patient is asleep or awake.

[0034] Figure 2 An example embodiment of a sensing device 200 suitable for use as a sensing device 104 in an infusion system in accordance with one or more embodiments is shown. The illustrated sensing device 200 includes, but is not limited to, a control module 204, a sensing element 202, an output interface 208, and a data storage element (or memory) 206. The control module 204 is coupled to the sensing element 202, the output interface 208, and the memory 206, and the control module 204 is suitably configured to support the operations, tasks, and / or processes described herein. Figure 1

[0035] The sensing element 202 generally represents a component of the sensing device 200 that is configured to generate, produce, or otherwise output one or more electrical signals indicative of a condition sensed, measured, or otherwise quantified by the sensing device 200. In this regard, a physiological condition of the user will affect a characteristic of the electrical signal output by the sensing element 202 such that the characteristic of the output signal corresponds to or is otherwise related to the physiological condition to which the sensing element 202 is sensitive. The sensing element 202 can be implemented as a glucose sensing element that generates an output electrical signal having a current (or voltage) associated therewith that is related to an interstitial fluid glucose level sensed or otherwise measured in the patient by the sensing device 104, 200.

[0036] Still referring to Figure 2 ​The control module 204 generally represents the hardware, circuitry, logic, firmware, and / or other components of the sensing device 200 that are coupled to the sensing element 202 to receive electrical signals output by the sensing element 202 and perform various additional tasks, operations, functions, and / or processes described herein. For example, the control module 204 can filter, analyze, or otherwise process electrical signals received from the sensing element 202 to obtain a measurement value for conversion to a calibrated measurement of the interstitial fluid glucose level. Further, in one or more embodiments, the control module 204 also implements or otherwise executes a calibration application module that uses calibration data associated with the sensing element 202 stored or otherwise maintained in the memory 206 to calculate or otherwise determine a calibrated measurement parameter based on the measurement value, as described in detail below. The calibrated measurement parameter can then be utilized to obtain a calibrated measurement of the patient's interstitial glucose level, as described in detail below.

[0037] Depending on the embodiment, the control module 204 can be implemented or realized with a general purpose processor, a microprocessor, a controller, a microcontroller, a state machine, a content addressable memory, a programmable logic device, an application specific integrated circuit, a field programmable gate array, any suitable programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof, designed to perform the functions described herein. In this regard, the methods or algorithm described in connection with the embodiments disclosed herein can be embodied directly in hardware, in firmware, in a software module executed by the control module 204, or in any practical combination thereof. In an example embodiment, the control module 204 includes or otherwise has access to a data storage element or memory 206. The memory 206 can be realized with any suitable kind of RAM, ROM, flash memory, registers, hard disks, removable disks, magnetic or optical mass storage devices, short or long term storage media, or any other non-transitory computer readable medium that stores programming instructions, code, or other data for execution by the control module 204. The computer executable programming instructions, when read and executed by the control module 204, cause the control module 204 to perform the tasks, operations, functions, and processes described in detail below.

[0038] In some embodiments, the control module 204 includes an analog-to-digital converter (ADC) or another similar sampling device that samples or otherwise converts the output electrical signal received from the sensing element 202 to a corresponding digital measurement data value related to the interstitial fluid glucose level sensed by the sensing element 202. In other embodiments, the sensing element 202 can incorporate an ADC and output a digital measurement value. In one or more embodiments, the electrical current of the electrical signal output by the sensing element 202 is affected by the user's interstitial fluid glucose level, and the digital measurement data value is realized as a current measurement value provided by the ADC based on the analog electrical output signal from the sensing element 202.

[0039] Output interface 208 generally represents the hardware, circuitry, logic, firmware, and / or other components of sensing device 200 that are coupled to control module 204 for outputting data and / or information from / to sensing device 200 (e.g., to / from infusion device 102, CCD 106, and / or computer 108). In this regard, in example embodiments, output interface 208 is implemented as a communication interface configured to support communication to / from sensing device 200. In such embodiments, the communication interface can include or otherwise be coupled to one or more transceiver modules capable of supporting wireless communication between sensing device 200 and another electronic device (e.g., infusion device 102 or another electronic device 106, 108 in infusion system 100). Alternatively, the communication interface can be implemented as or otherwise coupled to a port adapted to receive a wireless adapter that includes one or more transceiver modules and / or other components that support the operation of sensing device 200 as described herein. In other embodiments, the communication interface can be configured to support wired communication to / from sensing device 200. In other embodiments, output interface 208 can include or otherwise be implemented as an output user interface element for providing notifications or other information to a user, such as a display element (e.g., light emitting diode, etc.), a display device (e.g., liquid crystal display, etc.), a speaker or other audio output device, a haptic feedback device, etc. In such embodiments, the output user interface can be integrated with (e.g., within a common housing) or separately implemented from sensing device 104, 200.

[0040] It should be understood that the configurations and / or approaches described herein are Figure 2 is a simplified representation of sensing device 200. In this regard, although Figure 2 Various elements are shown as being located within sensing device 200, one or more elements of sensing device 200 can be distinct or otherwise separate from other elements of sensing device 200. For example, sensing element 202 can be separate and / or physically distinct from control module 204 and / or communication interface. Moreover, features and / or functionality described herein as being implemented by control module 204 can alternatively be implemented at infusion device 102 or another device 106, 108 within infusion system 100.

[0041] Sensor processing

[0042] Figure 3An exemplary embodiment of a processing system 300 is shown for developing a calibration model for sensing elements processed on a substrate 302. In this regard, multiple instances of a sensing element can be processed on the substrate 302, which are subsequently diced into smaller discrete portions (or dies) containing respective instances of the sensing element. In the exemplary embodiment, different instances of an electrochemical sensing element are simultaneously processed on or within a region 304 (alternatively referred to herein as a sensing region) of the substrate 302, while a process control monitor (PCM) is simultaneously processed on or within another region 306 of the substrate 302 adjacent to or otherwise near the sensing region 304. For example, in the illustrated embodiment, the sensing regions 304 are arranged in vertically oriented columns on the substrate 302, with the PCM regions 306 implemented as vertically oriented columns interposed between adjacent sensing regions 304. In this regard, the PCM regions 306 can include multiple PCMs extending vertically over an entire length of the PCM regions 306, while the sensing regions 304 include multiple instances of the sensing element extending vertically over an entire length of the sensing regions 304.

[0043] After and / or during processing, one or more process measurement systems 310 are used to analyze the PCMs processed within the PCM regions 306 to obtain process measurements for each of the PCMs processed on the substrate 302. In this regard, the process measurement systems 310 are capable of measuring biological, chemical, electrical, and / or physical properties of the respective PCMs. The process measurement data obtained for each of the PCMs can include, for example, glucose oxidase (GOx) thickness, GOx activity, glucose limiting membrane (GLM) thickness, working electrode (WE) platinum dummy impedance, counter electrode (CE) platinum dummy impedance, and human serum albumin (HSA) concentration. That being said, it should be noted that the subject matter described herein is not intended to be limited to any particular type or number of process measurements, and the process measurements can include measurements of any number of different properties or characteristics (e.g., dielectric properties, permeability, diffusivity, etc.). Additionally or alternatively, in some embodiments, the process measurements can be obtained by directly measuring characteristics of the sensing elements on the sensing regions 304.

[0044] Figure 4A cross-section of a working electrode 400 suitable for processing on a substrate 302 within the sensing region 304 is shown, which is used in an interstitial glucose sensing element. Further, in some embodiments, a dummy version of the working electrode 400 can be processed within the PCM region 306 for obtaining process of manufacture measurements. The working electrode 400 includes a substrate or base layer 402 (e.g., polyimide) and a covered plated metal layer 404 (e.g., chromium and gold). A plated layer 406 (e.g., platinum) is disposed on the metal layer 404 between portions of an insulating layer 408 (e.g., polyimide). A glucose oxidase layer 410 is formed on the layer 406 by depositing a glucose oxidase solution (e.g., via slot coating), and a human serum albumin (HSA) layer 412 is formed on the glucose oxidase layer 410. An adhesive layer 414 is disposed on the HSA layer 412 to affix a glucose limiting membrane (GLM) layer 416 on the working electrode 400. A counter electrode of the interstitial glucose sensing element can be similar or substantially identical to the working electrode 400, but lacks the glucose oxidase layer 410.

[0045] To obtain process of manufacture measurements, in example embodiments, a dummy impedance of the working electrode (and similarly, a dummy impedance of the counter electrode) is measured after the platinum plating process to form the layer 406. GOx solution activity measurements can be obtained during or after the GOx solution preparation process prior to deposition, while GOx thickness (e.g., thickness of the layer 410) is measured after the slot coating and selective patterning processes on the working electrode 400. HSA concentration can be measured during or after the solution preparation process prior to spraying the HSA layer 412 on the substrate, and GLM thickness (e.g., thickness of the layer 416) can be measured prior to applying the GLM layer 416. In one or more embodiments, these measurements are performed with respect to a sacrificial or monitoring instance of the working electrode 400 processed within the PCM region 306 on the substrate 302. In some embodiments, additional process of manufacture measurements of the working electrode 400 can be obtained during or after processing (e.g., via interferometry), such as surface roughness or other topography measurements

[0046] It should be appreciated that, Figure 4 A simplified representation of one example embodiment of the working electrode 400 is shown, and actual embodiments can include any number of additional and / or alternative layers (e.g., a high density amine (HDA) layer, etc.). Thus, the subject matter described herein is not intended to be limited to the embodiments shown in Figure 4 FIG. 1.

[0047] Referring again to Figure 3Processing measurements obtained by the process measurement system 310 are provided to a modeling system 330. In one or more embodiments, the modeling system 330 interpolates and / or extrapolates processing measurement data for different PCMs to obtain representative processing measurement data for a given instance of a sense element processed on the substrate 302. In this regard, the modeling system 330 can maintain an association between locations (e.g., coordinate locations) of respective PCMs on the substrate 302 and corresponding processing measurements obtained for the respective PCMs. Thereafter, based on a location of a respective sense element processed on the substrate 302, the modeling system 330 can: identify a subset of PCMs that are adjacent, proximate, or otherwise proximal to the respective sense element; obtain processing measurement data for the identified subset of PCMs; and then average or otherwise combine the processing measurement data for the subset of different PCMs based on a relationship between the location of the respective sense element relative to the locations of the different PCMs to obtain representative processing measurement data on the substrate 302 corresponding to the location where the respective sense element was processed.

[0048] In example embodiments, one or more test systems 320 are used to analyze each of the different sense elements processed within the sense region 304 to obtain reference measurement outputs for each sense element processed on the substrate 302 in response to one or more known reference inputs. For example, in example embodiments, the sense elements processed within the sense region 304 are implemented as electrochemical interstitial glucose sense elements exposed to a known concentration of glucose, where the test system 320 includes a glucose sensor transmitter, a logger, an amperemeter, a voltmeter, or a suitable measurement instrument capable of measuring a characteristic of a resulting output signal generated or otherwise provided by the glucose sense elements. In this regard, the reference output measurement parameters obtained for each sense element can include one or more of a current output by the sense element in response to: a reference glucose concentration, an electrochemical impedance spectroscopy (EIS) value (for one or more frequencies), or other measurements indicative of a characteristic impedance associated with the sense element in response to a reference glucose concentration, a counter electrode voltage (Vctr) (e.g., a difference between a counter electrode potential and a working electrode potential), and the like. For example, the glucose sensor transmitter can include potentiostat hardware and firmware that are cooperatively configured to collect current measurements corresponding to a current through a working electrode resulting from an applied bias potential and a reaction of a glucose oxidase layer of the working electrode of the sense element to a reference glucose concentration, while also calculating a counter electrode voltage (Vctr) by measuring a difference between a counter electrode potential and a working electrode potential. The glucose sensor transmitter can also be configured to perform electrochemical impedance spectroscopy analysis at different time intervals and at multiple frequencies with respect to current and voltage at the working electrode.

[0049] The reference output measurement parameters obtained by the test system 320 are provided to a modeling system 330 that maintains the reference output measurement parameters in association with respective instances of the sense elements processed on the substrate 302. In this regard, the modeling system 330 maintains an association between a reference output measurement parameter for a respective sense element processed on the substrate 302 and a representative process-of- fabrication measurement for that respective sense element processed on the substrate 302. As detailed below, based on the relationships between the process-of-fabrication measurement data and the reference measurement output data for various different instances of the sense elements, the modeling system 330 determines a calibration model for calculating or otherwise predicting an output measurement parameter for a sense element as a function of one or more process-of-fabrication measurement parameters associated with that sense element. In this regard, the output measurement parameters determined using the calibration model are effectively calibrated to account for process-of-fabrication variations and are therefore alternatively referred to herein as calibrated measurement parameters.

[0050] Manufacturing calibration

[0051] Figure 5 An exemplary embodiment of a process-of-fabrication model development process 500 is shown for developing a calibration model that maps process-of-fabrication measurements for a sense element to corresponding calibrated measurement parameters for that sense element. The various tasks performed in connection with the process-of-fabrication model development process 500 can be performed by hardware, firmware, software executed by processing circuitry, or any combination thereof. For illustrative purposes, the following description can refer to elements mentioned above in connection with the Figures 1 to 3 elements mentioned above in connection with the Figure 5 elements mentioned above in connection with the

[0052] The example process model development process 500 first performs operations to receive or otherwise obtain process measurements from different PCM regions on a substrate on which sensing elements are processed (task 502). For example, as described above, the process measurement system 310 analyzes the PCM regions 306 on the substrate 302 to obtain, for each PCM region 306, one or more measurements of a physical characteristic of the respective PCM region 306. The measurements of the physical characteristics of the PCM regions 306 are provided to the modeling system 330, which maintains an association between the respective measurements and the respective locations on the substrate 302 of the respective PCM regions 306. The process model development process 500 also receives or otherwise obtains measurement signal outputs from different sensing elements processed on the substrate (task 504). For example, as described above, the test system 320 tests or otherwise analyzes the different sensing elements 304 processed on the substrate 302 to obtain, for each sensing element 304, one or more reference measurement outputs generated or otherwise provided by the respective sensing element 304 in response to one or more known reference inputs. The reference measurement outputs are provided to the modeling system 330, which maintains an association between the respective reference measurement outputs and the respective locations on the substrate 302 of the respective sensing elements 304.

[0053] In the example implementation, the process model development process 500 continues to perform operations to assign the process measurements to each of the sensing elements processed on the substrate and to maintain an association between the assigned process measurements and the reference measurement outputs for each sensing element (tasks 506, 508). For example, as described above, using the coordinate locations at which the respective sensing elements 304 are processed on the substrate 302, the modeling system 330 can calculate or otherwise determine estimated process measurements for the coordinate locations based on the process measurements from different PCM regions 306 around the coordinate locations. In this regard, interpolation techniques (e.g., multivariate interpolation) can be employed to derive estimates of what the physical characteristics of the respective sensing elements 304 can be based on the process measurements associated with the neighboring PCM regions 306 in a manner that accounts for the spatial relationships between the coordinate locations of the sensing elements 304 relative to the respective coordinate locations of the neighboring PCM regions 306. For each respective sensing element 304, the modeling system 330 can maintain an association between a representative or estimated process measurement assigned to the respective sensing element 304, the reference measurement outputs obtained from the respective sensing element 304, and the coordinate location on the substrate 302 at which the respective sensing element 304 is processed. Additionally or alternatively, some implementations can obtain process measurements directly from the respective sensing elements 304, which can be used alone or in combination with the estimated process measurements derived from the PCM regions 306. Thus, the subject matter described herein is not necessarily limited to any particular location from which the process measurements are obtained.

[0054] Still referring to Figure 5 , the process model development process 500 utilizes the relationships between the reference measurement outputs of the different sensing elements and the process measurements to calculate or otherwise determine a predictive model for determining the calibrated measurement parameter as a function of the process measurements (task 510). In this regard, for each different measurement parameter, the modeling system 330 can utilize machine learning or artificial intelligence techniques to: determine which combination of process measurement parameters has a correlation or is predictive of the respective calibration measurement parameter; and then determine a corresponding formula, function, or model for calculating the calibration factor (or scaling factor) to thereby determine the valid calibration value of the parameter of interest based on this set of input variables. Thus, this model is able to characterize or map a particular combination of one or more process measurement parameters to a calibration factor to determine the valid calibration value of the calibration parameter of interest (e.g., current output, EIS value, etc.).

[0055] For example, the interstitial sensing elements can be designed to produce a particular amount of current in response to the reference glucose concentration used by the test system 320, alternatively referred to herein as a design current. For each sensing element, the modeling system 330 can divide the measured reference current output of the respective sensing element in response to the reference glucose concentration by the design current to determine an output current calibration factor for each sensing element. Thereafter, the modeling system 330 can utilize machine learning to: identify which combination of process measurement parameters has a correlation or is predictive of the output current calibration factor; and then determine a corresponding formula, function, or model to calculate the output current calibration factor based on this subset of process measurement input variables. Similarly, the measured reference EIS value of the respective sensing element can be divided by the design EIS value to determine an EIS calibration factor for each sensing element, which in turn is used to determine a corresponding formula, function, or model to calculate the EIS calibration factor based on the subset of associated process measurement input variables.

[0056] As another example, a neural network model can be developed using linear regression and appropriate activation functions, which can vary depending on the calibration parameter of interest. The process measurement inputs and calibration parameter outputs are structured into a corresponding matrix or vector, which is then fed into a loss function with initial values for the cost, weights, and biases for mapping the input matrix to the output matrix. The initial values are then input to the linear formula and activation portion of the neuron to initialize the neural network. The cost is then calculated and gradient descent is performed to determine updated weights and updated biases as a result of the gradient descent and a characteristic learning rate of the optimization. The process is then repeated iteratively for a desired number of iterations (e.g., 1000 iterations) to “learn” the weights and biases that will be used as part of the predictive model for the calibration parameter.

[0057] It should be noted that the subset of process measurement parameters that are predictive or correlated with a particular calibration measurement parameter can differ from other calibration measurement parameters. Moreover, the relative weight applied to the respective process measurement parameters of this predictive subset can also differ from other calibration measurement parameters that can have a common predictive subset based on the different correlations between the particular process measurement variables and the reference measurement data for that calibration parameter. In this regard, each measurement has a particular weight depending on the degree of influence (or lack thereof) with respect to the particular measurement parameter. For example, the current output can be most closely correlated with GLM thickness and GOx thickness. It should also be noted that any number of different machine learning techniques can be utilized to determine which process measurement parameters are predictive of a calibration measurement parameter of interest, such as, for example, genetic programming, support vector machines, Bayesian networks, probabilistic machine learning models or other Bayesian techniques, fuzzy logic, heuristically derived combinations, etc. Moreover, in practice, the aforementioned tasks 502, 504, 506, 508 of the process model development process 500 can be performed multiple times for a number of different substrates prior to model development, until a sufficient number of sensing elements and corresponding data sets have been obtained to achieve a desired level of accuracy or reliability of the resulting model.

[0058] Figure 6 An exemplary embodiment of a sensor initialization process 600 is shown that utilizes the calibration model from the process model development process 500 to configure the sensing elements after processing and prior to deployment. The various tasks performed in connection with the sensor initialization process 600 can be performed by hardware, firmware, software executed by processing circuitry, or any combination thereof. For illustrative purposes, the following description can refer to elements mentioned above in connection with the process model development process 500. It should be appreciated that the sensor initialization process 600 can include any number of additional or alternative tasks, the tasks need not be performed in the illustrated order and / or the sensor initialization process 600 can be incorporated into a more comprehensive procedure or process having additional functionality not Figures 1 to 3 described herein. In addition, one or more of the tasks shown and described in the context of the sensor initialization process 600 can be omitted from a practical implementation of the sensor initialization process 600 as is Figure 6 appreciated by those of ordinary skill in the art.

[0059] In an example implementation, the sensor initialization process 600 is performed with respect to sensing elements that are processed after the calibration model has been developed, to allow for the assignment of calibration factors or scaling factors to the sensing elements based on the process measurement data without the need for testing, to empirically determine the calibration data for the respective sensing elements. Similar to the process model development process 500, the sensor initialization process 600 first performs the following operations: receiving or otherwise obtaining process measurement data from different PCM regions on a substrate on which sensing elements are processed; identifying locations of the respective sensing elements on the substrate; and determining representative process measurement data for the respective sensing elements based on their locations (tasks 602, 604, 606). As described above, the substrate 302 is provided to the process measurement system 310 for measurement of the physical characteristics of the different PCM regions 306 on the substrate 302. Based on the coordinate locations at which the respective sensing elements are processed, the estimated process measurements for the sensing elements are calculated or otherwise determined, for example, through multivariate interpolation of the process measurements associated with the adjacent PCM regions 306, based on the spatial relationship of the respective sensing elements relative to the adjacent PCM regions 306.

[0060] After obtaining the process measurement parameters for the current instance of the sensing element of interest, the sensor initialization process 600 continues by applying the calibration model developed for that sensing element to the estimated process measurements to determine the calibration factors or scaling factors for the current instance of the sensing element (task 608). In this regard, for each respective calibration measurement parameter, the relevant subset of the estimated process measurements for that respective calibration measurement parameter is input or otherwise provided to the calibration model for that respective calibration measurement parameter to calculate the corresponding calibration factor for converting the output from the sensing element to the calibrated value for that respective calibration measurement parameter. Thus, for an interstitial glucose sensing element, a first calibration factor for converting the current output from the interstitial glucose sensing element to a calibrated current output can be determined, a second calibration factor for converting the EIS value to a calibrated EIS value can be determined, and so on.

[0061] After determining the calibration factors for the different calibration measurement parameters, the sensor initialization process 600 continues by storing or otherwise maintaining the calibration data as associated with the sensing elements (task 610). In this regard, in the example embodiment, for each respective calibration measurement parameter, the corresponding calibration factor is stored or otherwise maintained in the memory 206 of the sensing device 200 that includes the respective sensing element 202. Thereafter, when the sensing device 200 is in use, the control module 204 utilizes the calibration factors stored in the memory 206 to convert different measured values of the calibration measurement parameter determined based on the output (e.g., current output, EIS value, etc.) of the sensing element 202 into calibrated values. For example, the control module 204 can determine raw or uncalibrated EIS values based on the output signals provided by the sensing element 202, and then multiply or otherwise convert the EIS values to calibrated EIS values using the model-derived EIS calibration factors stored in the memory 206. In this way, the sensing device 200 can be configured to output validly calibrated measurement parameter values that account for variations in the fabrication process without requiring testing of the sensing element 202.

[0062] Figure 7 An example embodiment of a performance model development process 700 is shown that is used to develop one or more models that map calibration measurement parameters provided by a sensing device to calibrated measurement values. The various tasks performed in connection with the performance model development process 700 can be performed by hardware, firmware, software executed by processing circuitry, or any combination thereof. For illustrative simplicity, the following description can refer to the elements mentioned above in connection with the Figures 1 to 3 It should be appreciated that the performance model development process 700 can include any number of additional or alternative tasks, the tasks need not be performed in the illustrated order and / or the tasks can be performed concurrently, and / or the performance model development process 700 can be incorporated into a more comprehensive procedure or process having additional functionality not Figure 7 described herein. In addition, one or more of the tasks shown and described in the context of the performance model development process 700 can be omitted from a practical implementation of the performance model development process 700 as is

[0063] In the illustrated embodiment, the performance model development process 700 obtains patient data for a plurality of different patients, obtains one or more sets of calibrated measurement parameters and corresponding reference measurement values for each patient, and maintains an association between the patient data, calibrated measurement parameters, and reference measurement values for individual patients (tasks 702, 704, 706, 708). In example embodiments, the patient data includes one or more of the patient's age, gender, height, weight, body mass index (BMI), demographic data, and / or other parameters characterizing the patient. For each patient, at least one set of calibrated measurement parameters (e.g., output current measurements, EIS values, etc.) is also obtained and maintained in association with contemporaneous and / or corresponding reference measurement values of the patient's physiological condition. For example, contemporaneous or current calibrated measurement parameters output by the interstitial sensing device 104, 200 can be stored or otherwise maintained in association with reference blood glucose measurements for the patient's finger stick or other reference blood glucose measurements to develop a model for calculating or otherwise predicting blood glucose measurements as a function of the one or more calibrated measurement parameters.

[0064] The performance model development process 700 continues by calculating or otherwise determining a model for calculating or otherwise determining a measurement value as a function of the patient data and one or more calibration measurement parameters (task 710). For example, machine learning can be utilized to determine which combinations of patient data parameters and calibration measurement parameters have a correlation or predictive value with respect to reference blood glucose measurement values, and then determine a corresponding formula, function, or model for calculating blood glucose measurement values based on that set of input variables. Thus, the sensor performance model is able to characterize or map a particular combination of patient data and calibrated measurement parameters to an effectively calibrated blood glucose measurement value without the need for a finger stick or other reference measurement to calibrate an instance of the sensing device 104, 200. Depending on the embodiment, the sensor performance model can be stored at the sensing device 104, 200 (e.g., in the memory 206) or at another remote device or database.

[0065] In example embodiments, the time (or time stamp) associated with the patient data parameters and calibration measurement parameters can also be used as an input to the sensor performance model. For example, the output from the sensing device 104, 200 can be time stamped to allow determination of the time elapsed since sensor insertion, the time of day, or other time variables, which can in turn be used as an input variable related to sensor performance. In this way, the sensor performance model can account for time-dependent signal changes or variations that can be specific to a particular patient (or subset of patients), the manufacturing process measurement, and / or combinations thereof.

[0066] Figure 8 An example embodiment of a measurement process 800 is shown, which is used to use the sensor performance model developed using the process 700 to determine a blood glucose measurement value for a patient based on a set of output measurement parameters from the sensing device 104, 200. Figure 5calibration model developed by the process 500 and utilizing Figure 7 the sensor performance model developed by the process 700 to determine calibrated measurements of the patient physiological condition without requiring the patient to perform any additional calibration procedures. The various tasks performed in connection with the measurement process 800 can be performed by hardware, firmware, software executed by processing circuitry, or any combination thereof. For illustrative purposes, the following description can refer to elements mentioned above in connection with the process 500. It should be appreciated that the measurement process 800 can include any number of additional or alternative tasks, the tasks need not be performed in the illustrated order and / or the tasks can be performed concurrently, and / or the measurement process 800 can be incorporated into a more comprehensive procedure or process having additional functionality not Figures 1 to 3 described in detail herein. In addition, one or more of the tasks shown and described in the context of the process 500 above can be omitted from the measurement process 800 as is Figure 8 described from the actual implementation. Still yet, tasks shown and described above in the context of the process 500 can be added to the measurement process 800.

[0067] The illustrative measurement process 800 first performs the following operations: receiving or otherwise obtaining one or more measurement signals from a sensing element; and converting the measurement signals to calibrated measurement parameters utilizing calibration data associated with the sensing element (tasks 802, 804, 806). For example, the control module 204 can sample or otherwise obtain a measurement signal output by the interstitial glucose sensing element 202 that is influenced by the patient's interstitial glucose level, and based thereon, determine a raw or uncalibrated value of the output current by the sensing element 202, an EIS value characterizing the impedance of the sensing element 202, and so on. Thereafter, the control module 204 obtains the stored calibration factor associated with the sensing element 202 from the memory 206, and utilizes the stored calibration factor to convert the uncalibrated value to a calibrated output current, a calibrated EIS value, and so on.

[0068] Further, the measurement process 800 receives or otherwise obtains patient data associated with or otherwise characterizing the current patient, and utilizes the sensor performance model to determine a calibrated measurement value using the current patient data and the calibrated measurement parameter (tasks 808, 810, 812). For example, along with the sensor performance model associated with the type or configuration of the sensing element 202 and / or sensing device 104, 200 currently being utilized, the patient's age, gender, height, weight, body mass index (BMI), demographic data, and / or other parameters characterizing the patient can be stored or otherwise maintained in the memory 206 of the sensing device 104, 200 (or alternatively, at another device 102, 106, 108 in the infusion system 100). Along with the subset of patient data previously identified as being relevant to the calibrated measurement value, the current value of the calibrated measurement parameter (previously identified as an input variable of the sensor performance model relevant to the calibrated measurement value) is input or otherwise provided to the sensor performance model. In this way, the control module 204 at the sensing device 104, 200 (or alternatively, at another device 102, 106, 108 in the infusion system 100) utilizes the formula or function provided by the sensor performance model and its associated input variable weights to calculate or otherwise determine a calibrated sensor glucose measurement value based on one or more of the calibrated output current, calibrated EIS value, etc., in combination with one or more of the patient data parameters. The resulting calibrated sensor glucose measurement value can then be utilized to generate a corresponding dose command for operating the infusion device 102 or to perform other actions involving patient glucose level management. For example, a closed loop operating mode for controlling the infusion device 102 can calculate or otherwise determine a dose command based on a difference between the calibrated sensor glucose measurement value and a target glucose value for the patient, and autonomously operate a motor or other actuation means of the infusion device 102 to deliver the commanded dose of insulin to the patient.

[0069] Figure 9 An example embodiment of a data management system 900 suitable for implementing the subject matter described herein is shown. The data management system 900 includes, without limitation, a computing device coupled to a database 904 that is also communicatively coupled to one or more electronic devices 906 over a communication network 908, such as, for example, the Internet, a cellular network, a wide area network (WAN), etc. It should be appreciated that the data management system 900 is shown for illustrative purposes and is not intended to limit the subject matter described herein in any way, Figure 9 A simplified representation of a patient data management system 900 is shown.

[0070] In example embodiments, the electronic devices 906 include one or more medical devices, such as, for example, an infusion device, a sensing device, a monitoring device, etc. Further, the electronic devices 906 can include any number of non-medical client electronic devices, such as, for example, a mobile phone, a smartphone, a tablet computer, a smart watch, or other similar mobile electronic device, or any kind of electronic device capable of communicating with the computing device via the network 908, such as a laptop or notebook computer, a desktop computer, etc. In this regard, the electronic devices 906 can also include one or more components of the process measurement system 310, the test system 320, and / or the modeling system 330 configured to support the subject matter described herein. One or more of the electronic devices 906 can include or be coupled to a display device capable of graphically presenting data and / or information related to a patient's physiological condition, such as a monitor, a screen, or another conventional electronic display. Further, one or more of the electronic devices 906 also include or are otherwise associated with a user input device capable of receiving input data and / or other information from a user of the electronic device 906, such as a keyboard, a mouse, a touch screen, a microphone, etc.

[0071] In example embodiments, one or more of the electronic devices 906 transmit, upload, or otherwise provide data or information to the computing device for processing at the computing device and / or storage in the database 904. For example, when the electronic devices 906 are implemented as a sensing device, a monitoring device, or other device including a sensing element, and the electronic devices are inserted into a patient's body or otherwise worn by the patient to obtain measurement data indicative of a physiological condition in the patient's body, the electronic devices 906 can periodically upload or otherwise transmit the measurement data to the computing device. In other embodiments, the client electronic devices 906 can be used by the patient to manually define, input, or otherwise provide data or information characterizing the patient, and then transmit, upload, or otherwise provide such patient data to the computing device. In other embodiments, when the electronic devices 906 are implemented as components of the process measurement system 310, the test system 320, and / or the modeling system 330, the electronic devices 906 can upload process measurement data, test data, and / or other modeling data to the computing device for processing at the computing device and / or storage in the database 904 (e.g., modeling data 920). For example, in some embodiments, the computing device can obtain process measurement data and test data from the process measurement system 310 and the test system 320, respectively, and then develop a measurement parameter calibration factor model using the received data by or at the computing device (e.g., the modeling system 330 is implemented at the computing device). In other embodiments, the computing device can instead receive a measurement parameter calibration factor model from the modeling system 330 for storage and / or maintenance at the database 904 for subsequent deployment to the electronic devices 906.

[0072] The computing device generally represents a server or other remote device configured to receive data or other information from the electronic device 906, store or otherwise manage data in the database 904, and analyze or otherwise monitor data received from the electronic device 906 and / or stored in the database 904. In practice, the computing device can reside at a physically distinct and / or separate location from the electronic device 906, such as, for example, at a facility owned and / or operated by or otherwise affiliated with a manufacturer of one or more medical devices used in conjunction with the patient data management system 900. For illustrative purposes, but without limitation, the computing device can alternatively be referred to herein as a server, remote server 902, or variations thereof. The server 902 generally includes a processing system and a data storage element (or memory) capable of storing programming instructions for execution by the processing system that, when read and executed, cause the processing system to create, generate, or otherwise facilitate application programs or software modules configured to perform or otherwise support the processes, tasks, operations, and / or functions described herein. Depending on the embodiment, the processing system can be implemented using any suitable processing system and / or device, such as one or more processors, central processing units (CPUs), controllers, microprocessors, microcontrollers, processing cores configured to support the operations of the processing system described herein, and / or other hardware computing resources. Similarly, the data storage element or memory can be implemented as random access memory (RAM), read only memory (ROM), flash memory, magnetic or optical mass storage, or any other suitable non-transitory short or long term data storage, or other computer readable medium, and / or any suitable combination thereof.

[0073] In example embodiments, the database 904 is used to store or otherwise maintain historical patient data 910 for a plurality of different patients. For example, as described above, the database 904 can store or otherwise maintain reference blood glucose measurements (e.g., finger stick or metered blood glucose values) for different patients in association with contemporaneous or currently calibrated measurement parameters output by respective sensing devices 104, 200 associated with the respective patients at or about the time of the respective blood glucose measurements. Further, the patient data 910 can maintain personal information associated with the different patients, including the age, gender, height, weight, body mass index (BMI), demographic data, and / or other parameters characterizing the respective patients. In one or more embodiments, the database 904 is also used to store or otherwise maintain modeling data 920 that can be uploaded to the server 902 and / or determined by the server, such as, for example, process measurement data, test data, calibration models, and the like.

[0074] In one or more embodiments, the server 902 determines a sensor performance model for a particular type or configuration of sensing element 202 and / or sensing device 104, 200 using historical patient data 910 stored in the database 904, in a similar manner as described above in the context of Figure 7 Thereafter, the server 902 can store or otherwise maintain the sensor performance model in the database 904, and subsequently provide the sensor performance model to instances of the particular type or configuration of sensing element 202 and / or sensing device 104, 200. For example, upon initialization of the sensing device 104, 200, 906, the control module 204 can be configured to download or otherwise obtain the appropriate sensor performance model from the remote server 902 via the network 908. Thereafter, the control module 204 can utilize the sensor performance model in conjunction with the calibration factor stored locally in the memory 206 to determine calibrated glucose measurements for the current patient without the need for finger stick measurements or other calibration procedures. In other embodiments, the sensor performance model can be provided to the infusion device 102, 906 or another electronic device 106, 108, 906 in the infusion system 100 that is configured to receive calibrated measurement parameters from the sensing device 104, 200. In such embodiments, the infusion device 102, 906 or other electronic device 106, 108, 906 can utilize the obtained sensor performance model to determine calibrated glucose measurements using the calibrated measurement parameters provided by the sensing device 104, 200 without the need for finger stick measurements or other calibration procedures.

[0075] With the subject matter described herein, individual sensing elements can be individually calibrated prior to deployment in a manner that accounts for process variation using measurement data obtained from the substrate without the need for separate testing or calibration steps after fabrication. Further, the calibrated measurement parameters can be used in conjunction with individual patient data to determine calibrated measurements of patient physiological conditions without the need for the patient to perform calibration steps (e.g., obtain a finger stick measurement, etc.). Incorporating time or other time variables in the sensor performance model can also account for or compensate for the variability or aging of interstitial glucose sensing elements over time during their respective lifetimes.

[0076] Performance-based manufacturing control for manufacturing calibration

[0077] Figure 10 FIG. 6 illustrates a system 600 suitable for use with the above-described embodiments of the sensor performance model 500, in accordance with one or more embodiments. The system 600 includes a substrate 602, a sensing element 604, a sensing device 606, and a server 608. The substrate 602 can be a glucose sensor, such as the sensor 102 described above in the context of FIG. 1. The sensing element 604 can be a glucose sensing element, such as the sensing element 202 described above in the context of FIG. 2. The sensing device 606 can be a glucose sensing device, such as the sensing device 104 described above in the context of FIG. 3. The server 608 can be a server, such as the server 902 described above in the context of FIG. 9. Figures 5 to 8the process described above in the context of the exemplary embodiment of the performance testing process 1000. In this regard, the performance testing process 1000 reduces process variation by effectively filtering out or otherwise excluding instances of sensing elements that represent critical cases (or process criticals) and that exhibit deviations in their respective output measurement signals relative to the possible distribution of output measurement signals for that sensing element. The various tasks performed in connection with the testing process 1000 can be performed by hardware, firmware, software executed by processing circuitry, or any combination thereof. For illustrative purposes, the following description may Figures 1 to 3 refer to elements mentioned above in connection with the Figure 10 exemplary embodiment of the process model development process 500. It should be appreciated that the testing process 1000 can include any number of additional or alternative tasks, the tasks need not be performed in the illustrated order and / or the tasks can be performed concurrently, and / or the testing process 1000 can be incorporated into a more comprehensive procedure or process having additional functionality not described in detail herein. In addition, a person of ordinary skill in the art will appreciate that one or more of the tasks shown and described in the context of the testing process 1000 could be omitted, along with any additional tasks not described in detail above, without departing from the overall functionality of the testing process 1000.

[0078] Similar to the process model development process 500, the illustrated testing process 1000 receives or otherwise obtains process measurements from different regions of different substrates on which different instances of sensing elements were processed, and receives or otherwise obtains measurement signal outputs from the different instances of sensing elements processed on the substrates (tasks 1002, 1004). As described above, the process measurement system 310 can analyze the PCM regions 306 on the substrate 302 to obtain one or more measurements of the physical characteristics of the respective PCM regions 306. The measurements of the physical characteristics of the PCM regions 306 are provided to the modeling system 330, which maintains an association between the respective measurements and the respective locations of the respective PCM regions 306 on the substrate 302. The testing system 320 then tests or otherwise analyzes the different sensing elements 304 processed on the substrate 302 using one or more known reference inputs to obtain, for each sensing element 304, a reference measurement output generated or otherwise provided by the respective sensing element 304 in response to the reference input. For example, each sensing element 304 can be exposed to a known reference glucose concentration to obtain a corresponding output current measurement, EIS value, etc. The process measurements are assigned to each of the sensing elements, and an association between the assigned process measurements and the reference measurement outputs for each sensing element is maintained in a similar manner as described above (e.g., tasks 506, 508) (tasks 1006, 1008).

[0079] Figure 10 Still referring to, the test process 1000 generates or otherwise determines a predictive model based on the relationship between the reference measurement outputs of the different sensing elements and the process measurements, in a manner similar to that described above (e.g., task 510), that predicts a characteristic of the output measurement signal from an instance of the sensing element as a function of the process measurements (task 1010). In this regard, the modeling system 330 can utilize machine learning or artificial intelligence techniques to: determine, in response to known reference glucose concentrations, which combinations of process measurement parameters have a correlation or predictive quality to the output current measurements; and then determine a corresponding formula, function, or model for calculating the magnitude, frequency, or other characteristic of the output current generated by the sensing element based on that set of input process measurement variables. Thus, the model is able to characterize or map particular combinations of one or more process measurement parameters to the output measurement signal.

[0080] For example, for a plurality of different instances of the interstitial sensing element, each instance of the interstitial sensing element can be exposed to one or more reference glucose concentrations to obtain a corresponding reference output measurement (e.g., a reference value of the output current signal) associated with the reference glucose concentration. Further, representative process measurement values for a plurality of different process measurement variables (e.g., GOx thickness, GOx activity, GLM thickness, WE platinum dummy impedance, CE platinum dummy impedance, HSA concentration, etc.) can be obtained or otherwise assigned to each instance of the interstitial sensing element as described above. A formula can then be determined based on the relationship between the reference output measurements and the different process measurement variable values associated with the different instances of the interstitial sensing element using machine learning, artificial intelligence, or other regression techniques, for calculating a predicted or expected value of the output measurement as a function of particular combinations of the process measurement variables.

[0081] Still referring to Figure 10 After developing a predictive model that calculates the measurement output generated by the sensing element as a function of the input process measurement variables, the test process 1000 continues by calculating or otherwise generating a simulated distribution of the output measurement across a range of the input process measurement variables (task 1012). In this regard, the test process 1000 calculates or otherwise determines estimated output measurements for various combinations of values of the process measurement variables input to the predictive model. Thus, by independently varying the values of the process measurement variables input to the predictive model within their specified ranges (e.g., as specified by the process or other specifications), the predictive model can be utilized to extrapolate or interpolate the signal characteristics of the sensing element across the range of process possibilities.

[0082] For example, if there is a predictive model for calculating an output current measurement as a function of working electrode platinum virtual impedance, GOx activity, GOx thickness, HSA concentration, and GLM thickness, the test procedure 1000 calculates corresponding estimated output current values for different combinations of values from working electrode platinum virtual impedance, GOx activity, GOx thickness, HSA concentration, and GLM thickness within respective potential ranges for working electrode platinum virtual impedance, GOx activity, GOx thickness, HSA concentration, and GLM thickness. In this regard, a first estimated output current distribution (alternatively referred to herein as a low-side simulation distribution) can be calculated according to the process for a combination of high-amplitude working electrode platinum virtual impedance distribution, low-potential GOx activity distribution, low-potential GOx thickness distribution, low-potential HSA concentration distribution, and high-potential GLM thickness distribution, and another estimated output current distribution (alternatively referred to herein as a high-side simulation distribution) can be calculated for a combination of low-amplitude working electrode platinum virtual impedance, high-potential GOx activity distribution, high-potential GOx thickness distribution, high-potential HSA concentration distribution, and low-potential GLM thickness distribution. The respective input variables can be varied individually and independently (e.g., using Monte Carlo techniques) near respective ends of the design ranges of the respective input variables to obtain a desired number of input combinations (e.g., 10,000 combinations), which are then input or otherwise provided to the predictive model to obtain a corresponding number of simulated outputs (e.g., 10,000 output samples) at respective ends of the expected output range. In this way, simulated or estimated output current values across an entire potential value range within a two-dimensional variable space defined by the input variables of working electrode virtual impedance, GOx activity, GOx thickness, HSA concentration, and GLM thickness are obtained with the predictive model. The estimated output current values represent an expected distribution of output current measurements across the input variable space of the predictive model, which corresponds to a subset of process measurements (or biological, chemical, electrical, and / or physical characteristics) that are predictive or correlated to the output current measurements.

[0083] In example embodiments, the test process 1000 identifies or otherwise determines a boundary of a normal operating region or a critical performance threshold of the measurement output generated by the sensing element based on the simulated distribution of the measurement output derived using the predictive model (task 1014). In this regard, the boundary value or threshold represents a critical case (or process critical) delineating or otherwise defining a normal operating range of the measurement output in response to the known reference input. The critical performance threshold or boundary value can be identified or otherwise determined based on a statistical analysis of the simulated distribution of the measurement output. In this regard, it should be noted that there are any number of different statistical techniques that can be used to characterize a distribution of values to limit a normal operating region within the distribution, and the subject matter described herein is not limited to any particular implementation. In example embodiments, the test system 320 stores or otherwise maintains a critical threshold value defining a normal operating region associated with the reference input value that is used to subsequently test the output of the sensing element in response to the reference input.

[0084] For example, a statistical average output current value can be calculated or otherwise determined based on the simulated distribution of the output current values, where the critical threshold output current value is determined based on a standard deviation, variance, or other statistical measure of the simulated distribution of the output current values relative to the average output current value. For example, an upper threshold or boundary value to be associated with a given reference input stimulus can be determined by adding three times the standard deviation of the high-side simulated distribution to the average output current value of the high-side simulated distribution in response to that reference input, and a lower threshold or boundary value can be determined by subtracting three times the standard deviation of the low-side simulated distribution from the average output current value of the low-side simulated distribution. Thus, when a subsequent instance of the sensing element generates an output current value that is not within three standard deviations of the average value of the high-side or low-side simulated distribution in response to that reference input, the instance of the sensing element can be discarded even if all other measured parameters or characteristics of the instance of the sensing element are within the desired range. In other embodiments, the threshold or boundary value for acceptance or rejection can be different from the critical value derived from the simulated distribution, such as by adding or subtracting some offset from the critical value. For example, an upper retention threshold can be determined by adding one and a half times the standard deviation of the high-side simulated distribution to the upper critical value, the upper retention threshold being equal to the average output current value of the high-side simulated distribution plus three standard deviations of the high-side simulated distribution, such that any subsequent instance of the sensing element generating an output current value greater than four and a half standard deviations of the average value of the high-side simulated distribution in response to the reference input is discarded, while output current values less than this retention threshold are retained. Thus, in such embodiments, a subsequent instance of the sensing element can generate an output current value from the simulated distribution that is not within the critical boundary but is still retained so long as the output current value is sufficiently close to the critical boundary value (e.g., within one and a half standard deviations), and all other measured parameters or characteristics of the instance of the sensing element are within the desired range.

[0085] In an exemplary implementation, test process 1000 verifies or otherwise validates the performance of sensing elements after processing using model-derived normal operating range performance thresholds, and filters or otherwise excludes unqualified sensing elements prior to calibration and subsequent deployment (task 1016). In this regard, test process 1000 determines whether to accept or discard a sensing element when it responds to one or more output measurements of a known reference input that exceed the corresponding normal operating range of that output measurement. When an instance of a sensing element generates an output measurement in response to a known stimulus that is greater than or less than a corresponding critical threshold defining the upper or lower boundary of the normal operating region, the instance of the sensing element may be discarded or otherwise rejected (thereby reducing yield) without regard to... Figure 6 The sensor initialization process 600 performs calibration or other initialization. In this respect, even if the process measurements are within acceptable limits, the sensing element or substrate may be rejected. Conversely, when the output measurement is within a critical performance threshold defining the normal operating range, execution... Figure 6 The sensor initialization process 600 determines the calibration factor of the sensing element.

[0086] For example, continuing the above example, if the output current measurement generated by a particular sensing element in response to a reference glucose concentration is greater than or less than a critical threshold derived from a simulated distribution of output current measurements ranging from the transpotential working electrode virtual impedance, GOx activity, GOx thickness, HSA concentration, and GLM thickness values, the sensing element may be discarded or otherwise rejected by the test system 320, even if the working electrode virtual impedance, GOx activity, GOx thickness, HSA concentration, and GLM thickness measurements of the sensing element are all within acceptable ranges. Conversely, if the output current measurement generated by the sensing element in response to a reference glucose concentration is within a critical performance threshold derived from a simulated distribution of output current measurements ranging from the transpotential working electrode virtual impedance, GOx activity, GOx thickness, HSA concentration, and GLM thickness values, the sensing element, according to... Figure 6 The sensor initialization process 600 involves calibration and deployment. In this regard, measurements of the working electrode virtual impedance, GOx activity, GOx thickness, HSA concentration, and GLM thickness of the sensing element can affect the calibration factors associated with the sensing element, as described above.

[0087] By controlling manufacturing variations using the test process 1000, the effects of process excursions or process variations on the sensing elements can be mitigated by ensuring that the sensing elements are calibrated and deployed with functionality that is within the normal or expected operating range of the measurement constraints during the fabrication process. In this way, the performance of the sensing elements can be verified or otherwise validated in addition to the physical, biological, chemical, and electrical characteristics being verified or validated prior to performing the manufacturing calibration and subsequent deployment. By filtering or otherwise removing process excursions or other potential performance outliers from the data sets employed by the fabrication model development process 500 and / or the performance model development process 700, Figure 6 the accuracy and reliability of the sensor initialization process 600 is improved. Moreover, in some embodiments, the test process 1000 can be utilized to filter or otherwise remove sensing elements from the data sets employed by the fabrication model development process 500 and / or the performance model development process 700, thereby improving the accuracy and reliability of the resulting models employed by the sensor initialization process 600 and / or the measurement process 800.

[0088] For the sake of brevity, conventional techniques related to glucose sensing and / or monitoring, sampling, filtering, calibration, closed loop glucose control, and other functional aspects of the present subject matter can not be described in detail herein. Moreover, certain terms can also be used herein for the sake of brevity, and therefore, are not intended to be limiting, unless otherwise indicated. For example, terms such as "first", "second", and other such numerical terms referring to structure do not imply an order or sequence unless clearly indicated by the context. The foregoing description can also refer to elements or nodes or structures that are "connected" or "coupled" together. As used herein, unless otherwise expressly stated, "coupled" means that the element / nodes / structure are directly or indirectly joined together, and are not necessarily mechanically connected directly together.

[0089] While at least one exemplary embodiment has been presented in the foregoing detailed description of the application, it should be appreciated that a vast number of variations exist. It should also be appreciated that the exemplary embodiment or embodiments described herein are not intended to limit the scope, applicability or configuration of the claimed subject matter in any way. For example, the claimed subject matter is not necessarily limited to the infusion devices and related systems described herein. Furthermore, the foregoing detailed description will provide those skilled in the art with a convenient road map for implementing the described one or more embodiments. It should be understood that various changes can be made to the function and arrangement of elements without departing from the scope of the claimed subject matter, including those changes that are known or obvious to those skilled in the art as of the filing date of this patent application. Thus, although the above description has been made for the exemplary embodiments or other limitations, it is not intended to limit the claimed subject matter in any way but is intended to serve as a route to a broadest permissible scope of the claimed subject matter.

Claims

1. A method of manufacturing an interstitial glucose sensing element capable of providing an electrical signal affected by physiological conditions in a user's body, the method comprising: obtaining process measurement data from one or more substrates having the interstitial glucose sensing element; obtaining reference output measurement data from the interstitial glucose sensing element in response to a reference stimulus; determining a predictive model for a measured output of the interstitial glucose sensing element as a function of a process measurement variable based on a relationship between the process measurement data and the reference output measurement data; generating a simulated output measurement distribution across a range of the process measurement variable using the predictive model; identifying one or more performance thresholds for the measured output based on the simulated output measurement distribution; obtaining output measurement data from the interstitial glucose sensing element in response to the reference stimulus; and verifying that the output measurement data satisfies the one or more performance thresholds prior to calibrating the interstitial glucose sensing element.

2. The method of claim 1, further comprising, after verifying that the output measurement data satisfies the one or more performance thresholds: determining calibration data associated with the interstitial glucose sensing element based on assigned process measurement data for the interstitial glucose sensing element using a calibration model associated with the interstitial glucose sensing element, the calibration data for converting the electrical signal to a calibrated measurement parameter; and storing the calibration data in a data storage element associated with the interstitial glucose sensing element.

3. The method of claim 2, wherein determining the calibration data comprises calculating the calibration data based on the assigned process measurement data by inputting the assigned process measurement data to the calibration model.

4. The method of claim 1, wherein the process measurement data comprises representative process measurement data for the interstitial glucose sensing element and the reference output measurement data comprises a reference output measurement from the interstitial glucose sensing element in response to the reference stimulus, wherein determining the predictive model comprises determining a formula for calculating output measurements based on a relationship between the reference output measurement and the representative process measurement data for the interstitial glucose sensing element.

5. The method of claim 4, wherein generating the simulated output measurement distribution comprises calculating a plurality of output measurement values corresponding to a plurality of combinations of input process measurement values using the formula, wherein each combination of input process measurement values is within a range of the process measurement variable.

6. The method of claim 5, wherein identifying the one or more performance thresholds comprises identifying an upper boundary and a lower boundary of a normal operating range based on a statistical analysis of the plurality of output measurement values.

7. The method of claim 4, wherein the output measurement comprises an output current measurement, an electrochemical impedance spectroscopy value, a counter electrode voltage, or a combination thereof. ​ 8. The method of claim 7, wherein the reference stimulus comprises a reference glucose concentration.

9. The method of claim 8, wherein determining the formula comprises performing a linear regression or an artificial neural network technique to arrive at a linear formula for calculating the output measurement as a function of glucose oxidase thickness, glucose oxidase activity, glucose limiting membrane thickness, working electrode platinum virtual impedance, counter electrode platinum virtual impedance, human serum albumin concentration, or a combination thereof.

10. The method of claim 1, wherein obtaining the reference output measurement data comprises obtaining the reference output measurement data from the interstitial glucose sensing element in response to exposing the interstitial glucose sensing element to a reference glucose concentration.

11. The method of claim 10, wherein the process measurement data comprises representative process measurement data for the interstitial glucose sensing element, and the reference output measurement data comprises a reference output measurement from the interstitial glucose sensing element in response to the reference glucose concentration, and wherein determining the predictive model comprises determining a formula for calculating an output measurement based on a relationship between the reference output measurement and the representative process measurement data for the interstitial glucose sensing element.

12. The method of claim 11, wherein determining the formula comprises performing a linear regression or an artificial neural network technique to arrive at a linear formula for calculating the output measurement as a function of glucose oxidase thickness, glucose oxidase activity, glucose limiting membrane thickness, working electrode platinum virtual impedance, counter electrode platinum virtual impedance, and human serum albumin concentration, or a combination thereof.

13. The method of claim 11, wherein the output measurement comprises an output current measurement, an electrochemical impedance spectroscopy value, a counter electrode voltage, or a combination thereof.

14. The method of claim 1, wherein the process measurement data comprises glucose oxidase thickness, glucose oxidase activity, glucose limiting membrane thickness, working electrode platinum virtual impedance, counter electrode platinum virtual impedance, human serum albumin concentration, or a combination thereof.

15. A non-transitory computer readable medium having stored thereon computer executable instructions that, when executed by a processing system, cause the processing system to perform the method of claim 1.

16. A method of calibrating an interstitial glucose sensing element capable of providing an electrical signal affected by physiological conditions in a user’s body, the method comprising: obtaining a performance threshold associated with a response of the interstitial glucose sensing element to a reference stimulus, wherein the performance threshold is derived using a predictive model for a measured output of the interstitial glucose sensing element as a function of process measurement variables, and wherein the predictive model is determined based on process measurement data obtained from one or more substrates having the interstitial glucose sensing element; obtaining a current value of the measurement output from the interstitial glucose sensing element in response to the reference stimulus; and after verifying that the current value is within the performance threshold range: obtaining process measurement data associated with the interstitial glucose sensing element; obtaining a calibration model associated with the interstitial glucose sensing element; determining calibration data associated with the interstitial glucose sensing element based on the process measurement data using the calibration model, the calibration data used to convert the electrical signal to a calibrated measurement parameter; and storing the calibration data in a data storage element associated with the interstitial glucose sensing element.

17. The method of claim 16, further comprising determining the predictive model based on a relationship between reference output measurement data of the interstitial glucose sensing element in response to the reference stimulus and representative process measurement data, the representative process measurement data including measured values of the process measurement variable of the interstitial glucose sensing element.

18. The method of claim 17, wherein the reference stimulus includes a reference glucose concentration.

19. The method of claim 18, wherein the process measurement variable includes at least one of glucose oxidase thickness, glucose oxidase activity, glucose limiting membrane thickness, working electrode platinum virtual impedance, counter electrode platinum virtual impedance, or human serum albumin concentration.

20. A test system, the test system comprising: a data storage element to hold a performance threshold associated with a response of an interstitial glucose sensing element to a reference stimulus, wherein the performance threshold is derived using a predictive model for a measurement output of the interstitial glucose sensing element as a function of a process measurement variable, and wherein the predictive model is determined based on process measurement data obtained from one or more substrates having the interstitial glucose sensing element; and a processing system coupled to the data storage element to obtain output measurement data from the interstitial glucose sensing element in response to the reference stimulus, and to verify that the output measurement data satisfies the performance threshold prior to calibrating the interstitial glucose sensing element according to a calibration model. ​ ​

Images