Apparatus and methods for analyte sensor mismatch correction
Abstract
Apparatus and methods for response modeling and correction of signals associated with a parameter sensor. In one exemplary embodiment, the parameter sensor is configured to measure a physiologic analyte of a living being (e.g., blood glucose), and the apparatus and methods employ a mathematical transformation of two or more sensing elements (electrodes) of the sensor in order to compensate for temporal response differences or “mismatch.” This compensation enables the calculated blood analyte level, which results from processing of the signals of the two or more sensing electrodes, to be more accurate than calculations made without such compensation. In one variant, the parameter signals are generated, and compensation processing conducted, autonomously via a common implanted sensor platform.
Claims
exact text as granted — not AI-modified1 .- 21 . (canceled)
22 . A computerized apparatus configured for use with an implantable sensor having at least a first sensing element and a second sensing element, the computerized apparatus comprising:
a data communication interface, the data communication interface configured to receive at least first data related to signals from the first sensing element and second data related to signals from the second sensing element; processor apparatus in data communication with the data communication interface; and data storage apparatus in data communication with the processor apparatus, the data storage apparatus having at least one computer program stored thereon, the at least one computer program comprising a plurality of instructions which are configured to, when executed on the processor apparatus, cause the computerized apparatus to:
cause algorithmic generation of a temporal mismatch correction model using at least a machine learning algorithm, the generation of the temporal mismatch correction model based at least in part on the received first data and the received second data;
cause application of the generated temporal mismatch correction model to at least one of the first data or the second data to generate transformed data, the temporal mismatch correction model being a function of (i) at least one temporal parameter associated with the signals from the first sensing element (ii) at least one temporal parameter associated with the signals from the second sensing element, and (iii) one or more of a delay correction value or a lag correction value, the generated transformed data having a target response characteristic over time which is common to both the first data and the second data; and
utilize at least the transformed data in determination of a physiologic parameter value.
23 . The computerized apparatus of claim 22 , wherein:
the first sensing element comprises a glucose-modulated sensing element, and the second sensing element comprises a background oxygen sensing element, the glucose-modulated sensing element and the background oxygen sensing element comprising a differential pair of sensing elements; the transformed data comprises data corrected for of one or more of delay time or lag time between responses of the differential pair of sensing elements due to one or more differing material configurations of the glucose-modulated sensing element and the background oxygen sensing element; and the utilization of at least the transformed data in determination of the physiologic parameter value comprises utilization of at least the transformed data in calculation of a blood glucose concentration value based on an oxygen concentration detected at the glucose-modulated sensing element relative to a background oxygen concentration detected at the background oxygen sensing element.
24 . The computerized apparatus of claim 22 , wherein:
the implantable sensor further comprises a third sensing element, and the data communication interface is further configured to receive data related to signals from the third sensing element; and the plurality of instructions are further configured to, when executed on the processor apparatus, cause the computerized apparatus to apply the temporal mismatch correction model on the data related to signals from the third sensing element to generate additional transformed data.
25 . The computerized apparatus of claim 24 , wherein:
the first sensing element comprises an analyte-modulated sensing element, the second sensing element comprises a first background sensing element, and the third sensing element comprises a second background sensing element, the analyte-modulated sensing element, the first background sensing element, and the third background sensing element comprising a differential group of sensing elements; and the application of the temporal mismatch correction model on the at least one of the first data or the second data to generate the transformed data comprises application of the temporal mismatch correction model on the second data, each of the transformed data and the additional transformed data comprising background species data corrected for one or more of delay time or lag time related to a response characteristic of the analyte-modulated sensing element.
26 . The computerized apparatus of claim 24 , wherein:
the first sensing element comprises an analyte-modulated sensing element, and the second sensing element comprises a background sensing element, the analyte-modulated sensing element and the background sensing element comprising a differential pair of sensing elements configured for analyte detection; the third sensing element comprises a non-analyte sensing element; the transformed data comprises data corrected for of one or more of delay time or lag time between responses of the differential pair of sensing elements due to one or more differing material compositions of the analyte-modulated sensing element and the background sensing element; the additional transformed data comprises data corrected for one or more of delay time or lag time between a response of the non-analyte sensing element and at least one element of the differential pair of sensing elements configured for analyte detection; and the utilization of at least the transformed data in determination of the physiologic parameter value comprises utilization of at least the transformed data and the additional transformed data in determination of a blood analyte concentration value.
27 . The computerized apparatus of claim 22 , wherein the plurality of instructions are further configured to, when executed on the processor apparatus, cause the computerized apparatus to:
determine that one or more criteria are met for automatic evaluation of the temporal mismatch correction model; collect a series of the first data over a specified period of time; collect a series of the second data over a specified period of time; evaluate each of the series of first data and the series of second data with each of a plurality of candidate delay correction values and each of a plurality of candidate lag correction values; identify one or more of the plurality of candidate delay correction values and one or more of the plurality of candidate lag correction values which minimize a pre-defined error cost function; and cause generation of an updated temporal mismatch correction model based at least in part on the identification.
28 . A sensor apparatus, the sensor apparatus comprising:
a first detector; a second detector; interface apparatus configured to receive signals from each of the first detector and the second detector; processor apparatus in data communication with the interface apparatus; and storage apparatus in data communication with the processor apparatus, the storage apparatus having at least one computer program stored thereon, the at least one computer program comprising a plurality of instructions which are configured to, when executed by the processor apparatus, cause the sensor apparatus to:
determine data indicative of a signal response of the first detector;
determine data indicative of a signal response of the second detector, the data indicative of the signal response of the second detector having at least one temporal response characteristic different than that of the first detector;
utilize a temporal mismatch correction algorithm on the data indicative of the determined response of the second detector to generate data indicative of a transformed response of the second detector, the temporal mismatch correction algorithm based at least on (i) respective differential time constants associated with the signal response of the first detector and the signal response of the second detector, and (ii) both first and second electrical current values obtained from at least the first detector and the second detector, respectively during autonomous in vivo operation of the implantable sensor apparatus; and
calculate a physiologic parameter value based at least in part on (i) a ratiometric value derived from data generated from the temporal mismatch correction algorithm, and (ii) at least one of the data indicative of the determined signal response of the first detector, or the data indicative of the transformed response of the second detector.
29 . The sensor apparatus of claim 28 , wherein:
(i) the sensor apparatus comprises an implantable analyte sensor, (ii) the first detector comprises an analyte-modulated sensing element having a first membrane structure associated with a first electrode face, and (iii) the second detector comprises a background sensing element having a second membrane structure associated with a second electrode face, the analyte-modulated sensing element and the background sensing element comprising a differential pair of sensing elements configured for determination of data indicative of a blood analyte concentration value; and the pre-determined temporal mismatch correction algorithm is configured to characterize one or more of lag time or delay time between signal responses of the differential pair of sensing elements due to one or more material differences of the first membrane structure and the second membrane structure.
30 . The sensor apparatus of claim 29 , further comprising a third detector, the third detector comprising a non-analyte sensing element associated with the differential pair of sensing elements and being heterogeneous in at least one aspect of construction from either the first detector and second detector, the interface apparatus further configured to receive signals from the third detector; and
wherein the plurality of instructions are further configured to, when executed by the processor apparatus, cause the sensor apparatus to:
determine data indicative of a signal response of the third detector, the signal response of the third detector having at least one temporal response characteristic different than that of the first detector; and
utilize the pre-determined temporal mismatch correction algorithm on the data indicative of the determined response of the third detector to generate data indicative of a transformed response of the third detector.
31 . The sensor apparatus of claim 30 , wherein the calculation of the physiologic parameter value comprises in vivo processing via utilization of a sensor operational model to calculate the data indicative of the blood analyte concentration value based at least in part on (i) the data indicative of the determined signal response of the first detector, (ii) the data indicative of the transformed response of the second detector, and (iii) the data indicative of the transformed response of the third detector.
32 . A sensor apparatus, the sensor apparatus comprising:
a first detector; a second detector, the second detector having at least one temporal response characteristic different than a temporal response characteristic of the first detector; interface apparatus configured to receive signals from each of the first detector and the second detector; processor apparatus in data communication with the interface apparatus; and storage apparatus in data communication with the processor apparatus, the storage apparatus having at least one computer program stored thereon, the at least one computer program comprising a plurality of instructions which are configured to, when executed by the processor apparatus, cause the sensor apparatus to:
operate the sensor apparatus in a training mode wherein one or more time constant parameters associated with respective ones of the first and second detectors are dynamically determined in vivo;
generate a temporal mismatch correction algorithm based at least on the dynamically determined one or more time constant parameters;
utilize the generated temporal mismatch correction algorithm to generate data indicative of a transformed response of the second detector; and
calculate a physiologic parameter value based at least in part on the data indicative of the transformed response of the second detector.
33 . The sensor apparatus of claim 32 , wherein:
the sensor apparatus comprises a plurality of first detectors and a plurality of second detectors; and the generation of the temporal mismatch correction algorithm based at least on the dynamically determined one or more time constant parameters comprises generation of a plurality of temporal mismatch correction algorithms specific to individual respective ones of at least the plurality of first detector or the plurality of second detectors.
34 . A sensor apparatus, the sensor apparatus comprising:
a first detector configured to detect a first analyte; a second detector configured to detect a background analyte; interface apparatus configured to receive signals from each of the first detector and the second detector; processor apparatus in data communication with the interface apparatus; and storage apparatus in data communication with the processor apparatus, the storage apparatus having at least one computer program stored thereon, the at least one computer program comprising a plurality of instructions which are configured to, when executed by the processor apparatus, cause the sensor apparatus to:
intrinsically determine first data indicative of a signal response of the first detector;
intrinsically determine second data indicative of a signal response of the second detector, the second detector having at least one temporal response characteristic different than that of the first detector;
utilize a temporal mismatch correction algorithm to generate data indicative of a transformed response of the second detector, the temporal mismatch correction algorithm based at least on a ratio of concentration of the first analyte to concentration of the background analyte, the concentration of the first analyte determined using the intrinsically determined first data, and the concentration of the background analyte determined using the intrinsically determined second data; and
calculate a physiologic parameter value based at least in part on the data indicative of the transformed response of the second detector.Join the waitlist — get patent alerts
Track US2022163509A1 — get alerts on status changes and closely related new filings.
We store only your email — no account needed. See our privacy policy.