[go: up one dir, main page]

WO2024178272A1 - Détection électrochimique de concentrations d'ions pour une surveillance presque en temps réel - Google Patents

Détection électrochimique de concentrations d'ions pour une surveillance presque en temps réel Download PDF

Info

Publication number
WO2024178272A1
WO2024178272A1 PCT/US2024/016974 US2024016974W WO2024178272A1 WO 2024178272 A1 WO2024178272 A1 WO 2024178272A1 US 2024016974 W US2024016974 W US 2024016974W WO 2024178272 A1 WO2024178272 A1 WO 2024178272A1
Authority
WO
WIPO (PCT)
Prior art keywords
trained
processors
indication
regression model
current density
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/US2024/016974
Other languages
English (en)
Inventor
John O. DELANCEY
James A. ASHTON-MILLER
Mark A. Burns
Anna NELSON
Sanaz HABIBI
Alyssa SCHUBERT
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
University of Michigan System
University of Michigan Ann Arbor
Original Assignee
University of Michigan System
University of Michigan Ann Arbor
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by University of Michigan System, University of Michigan Ann Arbor filed Critical University of Michigan System
Publication of WO2024178272A1 publication Critical patent/WO2024178272A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N27/00—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means
    • G01N27/26—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating electrochemical variables; by using electrolysis or electrophoresis
    • G01N27/416—Systems
    • G01N27/4161—Systems measuring the voltage and using a constant current supply, e.g. chronopotentiometry

Definitions

  • the present disclosure generally relates to electrochemical detection of ion concentrations in a fluid sample, and in particular, to calculating ion concentrations using a chronopotentiometric sensor.
  • Urinary chloride is a versatile, readily available biomarker for diagnosing and monitoring a wide variety of diseases involving kidney and circulatory health. As the most abundant extracellular anion, chloride accounts for 70% of the total anion concentration in the body and is responsible for regulating the extra/intracellular volume, maintaining acid/base equilibrium, and upholding muscular activity. As a mobile ion that is primarily regulated through kidney excretion, urinary chloride is physiologically dynamic. Timed spot and 24-hour sample collection are often needed to account for intra- and inter-day variations and to capture ion changes, which can aid in early disease detection and monitor organ health. Specifically, chloride concentration has been shown to be an indicator for prerenal acute kidney disease, acid-base disorders, dietary salt intake, and acute heart failure.
  • chloride concentrations ⁇ 53 mM
  • ICU intensive care unit
  • a chloride sensor is needed to quickly detect ion change.
  • a near realtime sensor could accelerate detection of disease, monitor organ health, and inform treatment decisions.
  • Optical chemosensors have been investigated for cost effective, near real-time chloride detection.
  • Optical chemosensors are typically paper-based sensors (e.g., dipsticks); due to their portability, they have been widely used in qualitative field studies.
  • optical chemosensors are typically single use, and often require adding reagents (such as fluorescent probes, quenching agents) to obtain results. Thus, they are not suitable for longterm near real-time monitoring of urinary chloride.
  • the following relates to systems and methods for electrochemical detection of ion concentrations in a fluid sample using a chronopotentiometric sensor.
  • the chronopotentiometric sensor in accordance with the disclosure may utilize screen-printed electrodes to quantify clinically relevant chloride concentration (5-250 mM).
  • the sensor may be reliably reused at chloride concentrations less than 50 mM.
  • the chronopotentiometric sensor disclosed herein may use simple conductive metal working and reference electrodes and carbon or metal counter electrodes with no surface modification or added reagents.
  • a potentiostat in accordance with the disclosure may be electrically coupled to the chronopotentiometric sensor.
  • the potentiostat may cause a specified current or current density to flow between the working electrode and the counter electrode of the chronopotentiometric sensor through a fluid sample over a time period.
  • the potentiostat may measure a plurality of voltage levels between the working electrode and a reference electrode of the chronopotentiometric sensor across the fluid sample over the time period.
  • a user device in accordance with the disclosure may be communicatively coupled to the potentiostat.
  • the user device may provide a specified current or current density to the potentiostat.
  • the user device may receive the measured plurality of voltage levels over time.
  • the user device may output a detected ion concentration.
  • the user device uses the Sand equation to calculate an ion concentration.
  • the user device calculates a transition time, which is an inflection point of the voltage levels over the time period.
  • the user device calculates the ion concentration level as a function of the current density and the square root of the transition time.
  • the user device uses a trained machine learning (ML) regression model to predict an ion concentration.
  • the ML regression model may receive the specified current density and voltage levels over time as input and predict the ion concentration.
  • the trained ML regression models may predict the ion concentration in a shorter time, e.g., 0.1 sec to 1 sec, than the occurrence of the transition time.
  • a method of sensing an ion concentration level in a fluid sample in contact with a chronopotentiometric sensor may be provided.
  • the method may include: (1) causing, by one or more processors during a time period, a potentiostat to generate an electric current having one or more current density levels flowing through the fluid sample from a working electrode to a counter electrode of the chronopotentiometric sensor; (2) receiving, by the one or more processors from the potentiostat, an indication of a plurality of resulting voltage levels across the fluid sample between the working electrode and a reference electrode of the chronopotentiometric sensor measured at a plurality of time intervals during the time period; (3) providing, by the one or more processors, an indication of the one or more current density levels and the indication of the plurality of resulting voltage levels to a trained machine learning (ML) regression model to cause the trained ML regression model to predict the ion concentration level in the fluid sample; (4) receiving, by the one or more processors, an indication of the ML
  • a method of sensing an ion concentration level in a fluid sample in contact with a chronopotentiometric sensor may be provided.
  • the method may include: (1) causing, by one or more processors during a time period, a potentiostat to generate an electric current having one or more current density levels flowing through the fluid sample from a working electrode to a counter electrode of the chronopotentiometric; (2) receiving, by the one or more processors from the potentiostat, an indication of a plurality of resulting voltage levels across the fluid sample between the working electrode and a reference electrode of the chronopotentiometric sensor measured at a plurality of time intervals during the time period; (3) calculating, by the one or more processors, an inflection point of the plurality of the resulting voltage levels over the time period, wherein a transition time corresponds to a time of an occurrence of the inflection point; (4) calculating, by the one or more processors, the ion concentration level as a function of the one or more
  • the methods and systems disclosed herein represent an improvement to an existing technology or technologies, specifically chronopotentiometric sensing.
  • Existing technologies do not exist for rapid and accurate ion sensing using reusable sensors without the application of surface reagents.
  • the methods and systems disclosed herein use a particular machine, a potentiostat, and a particular apparatus, a chronopotentiometric sensor.
  • the chronopotentiometric sensor comprises working and reference electrodes comprising conductive metal and a counter electrode comprising carbon or metal.
  • the methods and systems disclosed herein cause a particular transformation of a substance.
  • Application of a current between the working electrode and counter electrode of a sensor causes a chemical reaction of the ions with the working electrode.
  • chloride ions react with a silver working electrode to form silver chloride.
  • Figure 1 depicts a block diagram of an exemplary chronopotentiometric system in which methods and systems for sensing ion concentration are implemented, according to some aspects.
  • Figure 2A depicts a plan view of an exemplary chronopotentiometric sensor, according to some aspects.
  • Figure 2B depicts a block diagram of the electrodes of an exemplary chronopotentiometric sensor, according to some aspects.
  • Figure 3 depicts a block diagram of an exemplary chronopotentiometric application, according to some aspects.
  • Figure 4 depicts a block diagram of exemplary machine learning regression model training, according to some aspects.
  • Figure 5 depicts a flow diagram of an exemplary method for predicting ion concentration in a fluid sample, according to some aspects.
  • Figure 6 depicts a flow diagram of an exemplary method for calculating ion concentration in a fluid sample, according to some aspects.
  • FIG. 1 depicts a block diagram of an exemplary chronopotentiometric system 100 in which methods and systems for sensing ion concentration may be performed, in accordance with various aspects discussed herein.
  • the chronopotentiometric system 100 includes one or more training servers 110, user devices 140, potentiostats 160, sensors 170, and networks 180.
  • the training servers 110 may perform ML training functionality as part of a cloud network or may otherwise communicate with other hardware or software components within one or more cloud computing environments to send, retrieve, or otherwise analyze data or information described herein.
  • the training servers 110 may include one or more processors 120, memories 122, network interface cards (NIC) 124, databases 126, and computing modules 130.
  • NIC network interface cards
  • the processor 120 may include one or more suitable processors (e.g., central processing units (CPUs) and/or graphics processing units (GPUs)).
  • the processor 120 may be connected to the memory 122 via a computer bus (not depicted) responsible for transmitting electronic data, data packets, or otherwise electronic signals to and from the processor 120 and memory 122 in order to implement or perform the machine-readable instructions, methods, processes, elements or limitations, as illustrated, depicted, or described for the various flowcharts, illustrations, diagrams, figures, and/or other disclosure herein.
  • the processor 120 may interface with the memory 122 via a computer bus to execute an operating system (OS) and/or computing instructions contained therein, and/or to access other services/aspects.
  • OS operating system
  • the processor 120 may interface with the memory 122 via the computer bus to create, read, update, delete, or otherwise access or interact with the data stored in the memory 122 and/or a database 126.
  • the memory 122 may include one or more forms of volatile and/or non-volatile, fixed and/or removable memory, such as read-only memory (ROM), electronic programmable read-only memory (EPROM), random access memory (RAM), erasable electronic programmable read-only memory (EEPROM), and/or other hard drives, flash memory, MicroSD cards, and others.
  • the memory 122 may store an operating system (OS) e.g., Microsoft Windows, Linux, UNIX, MacOS, etc.) capable of facilitating the functionalities, apps, methods, or other software as discussed herein.
  • OS operating system
  • the memory 122 may store a plurality of computing modules 130, implemented as respective sets of computer-executable instructions e.g., one or more source code libraries, ML training modules, input/output modules, etc.) as described herein.
  • computing modules 130 implemented as respective sets of computer-executable instructions e.g., one or more source code libraries, ML training modules, input/output modules, etc.
  • a computer program or computer based product, application, or code may be stored on a computer usable storage medium, or tangible, non- transitory computer-readable medium (e.g., standard random access memory (RAM), an optical disc, a universal serial bus (USB) drive, or the like) having such computer-readable program code or computer instructions embodied therein, wherein the computer-readable program code or computer instructions may be installed on or otherwise adapted to be executed by the processor(s) 120 (e.g., working in connection with the respective operating system in memory 122) to facilitate, implement, or perform the machine readable instructions, methods, processes, elements or limitations, as illustrated, depicted, or described for the various flowcharts, illustrations, diagrams, figures, and/or other disclosure herein.
  • a computer usable storage medium e.g., the model(s), such as ML regression models, or other computing instructions described herein
  • tangible, non- transitory computer-readable medium e.g., standard random access memory (RAM), an optical disc, a universal serial
  • the program code may be implemented in any desired program language, and may be implemented as machine code, assembly code, byte code, interpretable source code or the like (e.g., via Golang, Python, C, C++, C#, Objective-C, Java, Scala, ActionScript, JavaScript, HTML, CSS, XML, etc.).
  • the network interface card (NIC) 124 may include any suitable network interface controller(s), such as wired/wireless controllers (e.g., Ethernet controllers), and facilitate bidirectional/ multiplexed networking over the network 180 between the server 110 and other components of the environment 100 (e.g., the user device 140, etc.).
  • network interface controller such as wired/wireless controllers (e.g., Ethernet controllers), and facilitate bidirectional/ multiplexed networking over the network 180 between the server 110 and other components of the environment 100 (e.g., the user device 140, etc.).
  • the database 126 may be a relational database, such as Oracle, DB2, MySQL, a NoSQL based database, such as MongoDB, or another suitable database.
  • the database 126 may store training data and be used to train and/or test one or more ML regression models.
  • the computing modules 130 may include an ML training module (MLTM) 132.
  • the MLTM 132 may be included as a library or package executed on training server(s) 110.
  • libraries may include the TensorFlow based library, the PyTorch library, the HuggingFace library, and/or the scikit-learn Python library.
  • the MLTM 132 employs supervised learning, which involves identifying patterns in existing data to make predictions about subsequently received data.
  • the ML model is “trained” (e.g., via MLTM 132) using training data, which includes example inputs and associated example outputs.
  • the ML model may generate a predictive function which maps outputs to inputs and may utilize the predictive function to generate ML outputs based upon data inputs.
  • the exemplary inputs and exemplary outputs of the training data may include any of the data inputs or ML outputs described above.
  • a processing element may be trained by providing it with a large sample of data with known characteristics or features.
  • the computing modules 130 may include an input/output (I/O) module 134 comprising a set of computer-executable instructions implementing communication functions.
  • the I/O module 134 may include a communication component configured to communicate e.g., send and receive) data via one or more external/network port(s) to one or more networks or local terminals.
  • the training servers 110 may include a client-server platform technology such as ASP.NET, Java J2EE, Ruby on Rails, Node.js, a web service or online API, responsive for receiving and responding to electronic requests.
  • I/O module 134 may further include or implement an operator interface configured to present information to an administrator or operator and/or receive inputs from the administrator and/or operator.
  • An operator interface may provide a display screen.
  • the I/O module 134 may facilitate I/O components e.g., ports, capacitive or resistive touch sensitive input panels, keys, buttons, lights, LEDs), which may be directly accessible via, or attached to, servers 110.
  • an administrator or operator may access the training servers 110 via a user device 140 to review information, make changes, input training data, initiate training via the MLTM 132, and/or perform other functions.
  • the user device(s) 140 may include any suitable device and include one or more desktop computers, laptop computers, server computers, smartphones, tablets, and/or other electronic or electrical component.
  • the user device 140 may include a memory and a processor for, respectively, storing and executing one or more modules.
  • the memory may include one or more suitable storage media such as a magnetic storage device, a solid-state drive, random access memory (RAM), etc.
  • the user device 140 may access services or other components of the environment 100 via the network 180.
  • the user device 140 may include a hardware port 142.
  • the hardware port 142 may be an Ethernet, USB, USB-C, Lightning, or any other suitable port.
  • the hardware port 142 may enable the user device 140 to be communicatively connected to the potentiostat 160 via a cable.
  • the user device 140 may include one or more wireless interfaces 144.
  • the wireless interface 144 may enable communication via cellular, ZigBee, Bluetooth, IEEE 802.11 , or other suitable wireless network technologies.
  • the user device 140 may communicate with the potentiostat 160 and/or the network 180 via the wireless interface 144.
  • the user device 140 may include one or more applications 148.
  • the application 148 may include computer-executable instructions for allowing a user to run a chronopotentiometric test on a sample and receive as output the measured ion concentration in the sample.
  • the application 140 may include a web application, smartphone application, standalone executable, or any other suitable type of application.
  • the potentiostat 160 may include circuits, hardware, software, and/or firmware for generating a specified electric current and measuring a resulting voltage.
  • the potentiostat 160 may measure the resulting voltage at a sampling rate of 1 nsec, 10 nsec, 100 nsec, or 200 nsec.
  • the potentiostat 160 may be configured to receive commands from the user device 140 and transmit data to the user device 140.
  • the potentiostat 160 may include working, counter, and reference terminals.
  • One example of a potentiostat 160 is the Biologic Science Instruments SP- 200.
  • the sensor 170 may include a chronopotentiometric sensor.
  • the sensor 170 may include a working, a counter, and a reference electrode.
  • the working, counter, and reference electrodes of the sensor 170 may be configured to be electrically connected to the working, counter, and reference terminals, respectively, of the potentiostat 160.
  • the sensor 170 may include a temperature probe, a pH probe, and/or other measurement probes.
  • the network 180 may be a single communication network or may include multiple communication networks of one or more types (e.g., one or more wired and/or wireless local area networks (LANs), and/or one or more wired and/or wireless wide area networks (WANs) such as the Internet).
  • the network 180 may enable bidirectional communication between the server 110 and the user device 140, and/or between other computing devices/ instances, for example.
  • Figure 2A depicts a plan view of the sensor 170, in accordance with various aspects discussed herein.
  • One example of the sensor 170 is the DS-C013 from Metrohm Dropsens.
  • the sensor 170 may be configured to be immersed in a fluid sample.
  • the sensor 170 may include a substrate 202.
  • the substrate 202 may include a non-conductive, water-resistant material, such as plastic, fiberglass, and/or epoxy.
  • the sensor 170 may include a plurality of electrodes, such as a working electrode 210, a counter electrode 220, and a reference electrode 230.
  • the working electrode 210 and the reference electrode 230 may include a conductive metal, such as silver or platinum.
  • the counter electrode 220 may include carbon or metal.
  • the working electrode 210, the counter electrode 220, and the reference electrode 230 may be spaced apart from each other by gaps such that a fluid would fill the gaps when the sensor 170 is immersed in a fluid sample.
  • the sensor 170 may include a cover 240.
  • the cover 240 may include a non- conductive, water-resistant material, such as plastic, fiberglass, and/or epoxy.
  • the cover 240 may be configured to prevent the fluid from contacting and causing undesired electrical connections between components beneath the cover 240.
  • the sensor 170 may include a plurality of contacts, such as a working contact 250A, a counter contact 250B, and a reference contact 250C.
  • the working contact 250A, the counter contact 250B, and the reference contact 250C may include an electrically conductive metal, such as copper or aluminum.
  • the working contact 250A, the counter contact 250B, and the reference contact 250C may be electrically connected to the working electrode 210, the counter electrode 220, and the reference electrode 230, respectively.
  • the electrical connections between the contacts 250A-250C and the electrodes 210-230 may include leads or wires beneath the cover 240.
  • the working contact 250A, the counter contact 250B, and the reference contact 250C may be configured to be removably coupled with the working, counter, and reference terminals of the potentiostat 160.
  • Figure 2B depicts a block diagram of the sensor 170, in accordance with various aspects discussed herein.
  • a current source 260 such as the potentiostat 160, may be electrically connected to the working electrode 210 and the counter electrode 220.
  • the current source 260 may apply a current / flowing through the fluid between the working electrode 210 and a counter electrode 220.
  • the applied current / may be constant, decrease, or increase over time.
  • the applied current / may induce a faradaic reaction on the surface of the working electrode 210.
  • the silver working electrode 210 will react with chloride ions in the fluid to form a solid silver chloride deposit, as shown in Equation (1): Ag(s) + Cl (aq) AgCI(s) + e- (1 )
  • the chloride ions When the mass transport of the chloride ions from the bulk solution to the working electrode surface is slower than the faradaic reaction, the chloride ions will start to locally deplete. The local depletion creates a chloride ion concentration gradient between the working electrode 210 and the reference electrode 230.
  • the concentration gradient is derived from the migrative and diffusive terms of the Nernst-Planck equation. The derivation of the concentration gradient assumes the convective terms are negligible, as the system is in stagnant operating conditions.
  • the voltage V may be measured by the potentiostat 160 over time at a specified sampling rate.
  • FIG. 3 depicts a block diagram of the application 148 on the user device 140, in accordance with various aspects discussed herein.
  • the application 148 may include a user interface module 310, a potentiostat input/output module 320, a control module 330, a data collection and processing module 340, a transition time calculator module 350, and/or a ML operation module 360.
  • the user interface module 310 may include computer-readable instructions for receiving input from and providing output to a user.
  • the user interface module 310 may provide a graphical user interface (GUI) on a screen or display of the user device 140.
  • GUI graphical user interface
  • the user interface module 310 may receive as input a potentiostat model selection, a sensor selection, an ion selection, and/or one or more environmental parameter values, such as temperature, pH, and/or known concentration of other ions, e.g., sulfate.
  • the user interface module 310 may output a calculated ion concentration.
  • the potentiostat input/output module 320 may include computer-readable instructions for sending commands to and receiving measurements from the potentiostat 160.
  • the potentiostat input/output module 320 may include one or more application programming interfaces (APIs) for communicating with one or more potentiostat models.
  • APIs application programming interfaces
  • the potentiostat input/output module 320 may transmit, via the hardware port 142 and/or the wireless interface 144, a specified current level and test duration, to the potentiostat 160.
  • the potentiostat input/output module 320 may receive a plurality of voltage measurements from the potentiostat 160.
  • the control module 330 may include computer-readable instructions for selecting the specified current level and duration for the test.
  • the control module 330 may select the current level based upon the sensor or ion selection provided by the user.
  • the current level may be a current magnitude (in Amps) or current density (in Amps per square meter) and may be steady state, rising, or falling over a time period.
  • the current density may be 60 A/m 2 , 120 A/m 2 , 240 A/m 2 , 480 A/m 2 , or 960 A/m 2 , for example.
  • the test duration may be 5 sec to 10 sec.
  • the test duration may be 0.1 sec to 1 sec.
  • the data collection and processing module 340 may include computer-readable instructions for collecting output from the potentiostat 160 during a test.
  • the data collection and processing module 340 may collect the plurality of voltages and corresponding timestamps measured by potentiostat 160 at the specified time intervals.
  • the data collection and processing module 340 may collect additional environmental data, such as sample temperature and pH, from the potentiostat 160 or the sensor 170.
  • the data collection and processing module 340 may process and perform calculations on the collected data to generate additional data, such as the first and second derivatives of voltage versus time, the initial values of the derivatives, spline interpolations of voltage versus time, rolling average of the voltage over different time windows, and/or the initial values of the rolling average.
  • t cl - is the transference number of chloride, or the fraction of chloride ions per total ions in solution.
  • the chloride ion transference number can be calculated (Equation (3)) for chloride (Cl“) and other ion species (k) present in the electrolyte, where z is the charge of the ion, A is the limiting molar conductivity, and C*is the bulk ion concentration. In the presence of many ions that are not chloride (e.g., high conductivity of supporting electrolyte), t cr is negligible.
  • the bulk chloride ion concentration directly relates to the transition time and can be used as calculated ion concentration. In some embodiments, the transition time occurs between 0.1 sec and 6 sec after the current is applied.
  • the application 148 may include an ML operation module (MLOM) 360.
  • the MLOM 360 may be included as a library or package, such as the TensorFlow based library, the PyTorch library, the HuggingFace library, and/or the scikit- learn Python library.
  • the MLOM 360 may include computer-readable instructions implementing ML loading, configuration, initialization and/or operation functionality.
  • the MLOM 360 may include instructions for storing trained ML regression models (e.g., in the database 126).
  • the MLOM 360 may provide one or more of the applied current densities, voltage vs. time, first derivative of the voltage vs. time, second derivative of the voltage vs.
  • the input provided by the MLOM 360 may include all or a subset of the features included in the training data.
  • the MLOM 360 may receive the predicted ion concentration output by the trained ML regression model.
  • FIG. 4 illustrates an exemplary ML environment 400 for ML training and validation, in accordance with various aspects discussed herein.
  • the ML training and validation may be performed by the MLTM 132 or by any other suitable code or software.
  • MLTM 132 may access database 126 or any other data source for data suitable to generate one or more ML regression models appropriate to receive and/or process chronopotentiometric data.
  • the chronopotentiometric data may be sample laboratory data used to fit the parameters (weights) of an ML regression model with the goal of training it by example.
  • the chronopotentiometric data may be split into a training data set and a validation data set.
  • the trained ML regression model may be incorporated into a user application and distributed.
  • the one or more untrained ML regression models 410 may include one or more ensemble ML regression techniques, including random forest, adaptive boosting (AdaBoost), and/or extreme gradient boosting (XGBoost).
  • the one or more untrained ML regression models 410 may be configured with a set of initial hyperparameters 420.
  • the set of initial hyperparameters 420 may include specified values for the number of decision trees, number of features sampled per split, maximum tree depth, minimum sample split, maximum terminal nodes, minimum samples per leaf, maximum samples per tree, and/or maximum features per tree.
  • the number of decision trees may be 50 to 1 ,000
  • the number of features sampled per split may be 2 to 12
  • the minimum samples per leaf may be 5 to 20.
  • one or more chronopotentiometric data collections may be split into a training dataset 430 and a validation dataset 450.
  • the training dataset 430 and the validation dataset 450 may include data from a plurality of experiments using fluids with known ion concentrations.
  • the training dataset 430 and the validation dataset 450 may include a plurality of features, such as known ion concentration values, the applied current densities, voltage vs. time, first derivative of the voltage vs. time, second derivative of the voltage vs. time, the initial values of the derivatives, spline interpolations of voltage versus time, rolling average of the voltage over different time windows, the initial values of the rolling average, temperature, pH, sulfate concentration, and any other suitable parameters.
  • the MLTM 132 may retrieve the training dataset 430 from the database 126 and provide the training dataset 430 to the untrained ML regression models 410 in a training step.
  • the training may cause trained ML regression models 440 to be generated from the untrained ML regression models 410.
  • the trained ML regression models 440 may include one or more decision trees that predict an ion concentration based upon input data having a plurality of features.
  • the trained ML regression models 440 may assign different parameters to different features such that some features are weighted more heavily than others.
  • the MLTM 132 may retrieve the validation dataset 450 from the database 126 and provide the validation dataset 450 to the trained ML regression models 450 in a validation step.
  • the MLTM 132 may withhold the known ion concentration values when providing the validation dataset 450 to the trained ML regression models 450.
  • the validation may cause the trained ML regression models 440 to generate predicted ion concentrations.
  • the MLTM 132 may calculate a prediction error 460 by comparing the predicted ion concentration to the known ion concentration value.
  • the MLTM 132 may use the prediction error 460 to tune the trained ML regression models 440 by adjusting one or more parameters of the trained ML regression models 440 to minimize prediction error 460.
  • the trained ML regression models 440 may be tuned with a set of tuning hyperparameters 470.
  • the set of tuning hyperparameters 470 may include specified values for the number of decision trees, number of features sampled per split, maximum tree depth, minimum sample split, maximum terminal nodes, minimum samples per leaf, maximum samples per tree, and/or maximum features per tree.
  • Figure 5 depicts a flow diagram of an exemplary method 500 for sensing or predicting ion concentration in a fluid sample, in accordance with various aspects discussed herein.
  • One or more steps of the method 500 may be implemented as a set of instructions stored on a computer-readable memory and executable on one or more processors.
  • the method 500 of Figure 5 may be implemented via a system, such as the training server 1 10, the user device 140, the potentiostat 160, and/or the chronopotentiometric sensor 170.
  • the method 500 may operate in conjunction with the scenarios and/or environments illustrated in Figures 1 - 4 and/or in other environments.
  • the fluid sample comprises urine.
  • the method 500 may include training an ML regression model with a training dataset to generate a trained ML regression model.
  • the method 500 may include validating the trained ML regression model with a validation dataset.
  • the training and the validation may be performed by the MLTM 132 or other appropriate software.
  • the training dataset and the validation dataset comprise a plurality of applied current density levels, a plurality of resulting voltage levels, and a plurality of indications of the ion concentration levels.
  • the method 500 may include at block 510 causing a potentiostat, such as the potentiostat 160, to generate an electric current during a time period.
  • the electric current may flow from a working electrode to a counter electrode of a chronopotentiometric sensor, such as the sensor 170, through the fluid sample.
  • the electric current may have one or more current density levels.
  • the current density levels may increase or decrease during the time period.
  • the current density levels range between 60 A/m 2 and 960 A/m 2 , including one or more of 60 A/m 2 , 120 A/m 2 , 240 A/m 2 , 480 A/m 2 , or 960 A/m 2 .
  • the working electrode may comprise conductive metal, such as silver or platinum, and the counter electrode may comprise carbon or metal.
  • the time period may be between 0.1 sec and 1 .0 sec.
  • the method 500 may include at block 520 receiving from the potentiostat an indication of a plurality of resulting voltage levels.
  • the potentiostat may measure the plurality of resulting voltage levels and a plurality of time intervals during the time period. A duration between the plurality of time intervals may be between 1 psec and 200 nsec.
  • the plurality of resulting voltage levels may be measured between the working electrode and a reference electrode of the chronopotentiometric sensor across the fluid sample.
  • the reference electrode may comprise a conductive metal, such as silver or platinum.
  • the working electrode and the reference electrode may comprise silver.
  • the method 500 may include measuring a temperature of the fluid sample using a temperature probe.
  • the method 500 may include measuring a pH of the fluid using a pH probe.
  • the method 500 may include calculating a conductivity level of the fluid sample by causing the potentiostat to apply a DC or AC voltage between the working and reference electrodes and measuring the resulting current level.
  • the method 500 may include calculating a transference number of the fluid sample by causing the potentiostat to apply an AC voltage between the working and reference electrodes and measuring the resulting current level.
  • the method 500 may include calculating the first derivative of the voltage vs. time, second derivative of the voltage vs. time, the initial values of the derivatives, spline interpolations of voltage versus time, rolling average of the voltage over different time windows, and/or the initial values of the rolling average.
  • the method 500 may include at block 530 providing an indication of the one or more current density levels and the indication of the plurality of resulting voltage levels to a trained machine learning (ML) regression model to cause the trained ML regression model to predict the ion concentration level in the fluid sample.
  • the method 500 may include providing an indication of the temperature, pH, the conductivity level, the transference number, first derivative of the voltage vs. time, second derivative of the voltage vs. time, the initial values of the derivatives, spline interpolations of voltage versus time, rolling average of the voltage over different time windows, and/or the initial values of the rolling average of the fluid sample to the trained ML regression model.
  • the trained ML regression model may comprise a trained adaptive boosting regression model, a trained extreme gradient boosting regression model, and/or a trained random forest regression model.
  • the trained random forest regression model may comprise between 50 and 1 ,000 decision trees, 2 to 12 features sampled per split, and 5 to 20 samples per leaf.
  • the method 500 may include at block 540 receiving an indication of the ion concentration level from the trained ML regression model.
  • the ion concentration level may be a chloride concentration level.
  • the chloride concentration level may be between 5 mM and 250 mM.
  • the method 500 may include at block 550 outputting the indication of the ion concentration level to a user.
  • the method 500 may include determining whether a chloride concentration of a urine sample is less than 53 mM. Responsive to determining that the chloride concentration is less than 53 mM, the method 500 may include administering a diuretic to a patient associated with the urine sample.
  • Figure 6 depicts a flow diagram of an exemplary method 600 for sensing or calculating an ion concentration in a fluid sample, in accordance with various aspects discussed herein.
  • One or more steps of the method 600 may be implemented as a set of instructions stored on a computer-readable memory and executable on one or more processors.
  • the method 600 of Figure 6 may be implemented via a system, such as the user device 140, the potentiostat 160, and/or the chronopotentiometric sensor 170.
  • the method 600 may operate in conjunction with the scenarios and/or environments illustrated in Figures 1 - 5 and/or in other environments.
  • the fluid sample comprises urine.
  • the method 600 may include at block 610 causing a potentiostat, such as the potentiostat 160, to generate an electric current during a time period.
  • the electric current may flow from a working electrode to a counter electrode of a chronopotentiometric sensor, such as the sensor 170, through the fluid sample.
  • the electric current may have one or more current density levels.
  • the current density levels may increase or decrease during the time period.
  • the current density levels range between 60 A/m 2 and 960 A/m 2 , including one or more of 60 A/m 2 , 120 A/m 2 , 240 A/m 2 , 480 A/m 2 , or 960 A/m 2 .
  • the working electrode may comprise conductive metal, such as silver or platinum, and the counter electrode may comprise carbon or metal.
  • the time period may be between 5 sec and 10 sec.
  • the method 600 may include at block 620 receiving from the potentiostat an indication of a plurality of resulting voltage levels.
  • the potentiostat may measure the plurality of resulting voltage levels and a plurality of time intervals during the time period. A duration between the plurality of time intervals may be between 1 psec and 200 nsec.
  • the plurality of resulting voltage levels may be measured between the working electrode and a reference electrode of the chronopotentiometric sensor across the fluid sample.
  • the reference electrode may comprise conductive metal, such as platinum or silver.
  • the working electrode and the reference electrode may comprise silver.
  • the method 600 may include calculating a transference number of the fluid sample by causing the potentiostat to apply an AC voltage between the working and reference electrodes and measuring the resulting current level.
  • the transference number may calculated by applying Equation (3).
  • the method 600 may include at block 630 calculating an inflection point of the plurality of the resulting voltage levels over the time period, where a transition time corresponds to a time of an occurrence of the inflection point.
  • the method 600 may include at block 640 calculating the ion concentration level as a function of the one or more current density levels and the square root of the transition time. Calculating the ion concentration level may comprise applying
  • Equation (2) Calculating the ion concentration level may comprise applying the following equation:
  • F Faraday’s constant
  • D is the mean chloride ion diffusion coefficient
  • j is the current density
  • C* is the bulk chloride concentration.
  • the ion concentration level may be a chloride concentration level.
  • the chloride concentration level may be between 5 mM and 250 mM.
  • the method 600 may include at block 650 outputting the indication of the ion concentration level to a user.
  • the method 600 may include determining whether a chloride concentration of a urine sample is less than 53 mM. Responsive to determining that the chloride concentration is less than 53 mM, the method 600 may include administering a diuretic to a patient associated with the urine sample.
  • routines, subroutines, applications, or instructions may constitute either software (e.g., code embodied on a non-transitory, machine-readable medium) or hardware.
  • routines, etc. are tangible units capable of performing certain operations and may be configured or arranged in a certain manner.
  • one or more computer systems e.g., a standalone, client or server computer system
  • one or more hardware modules of a computer system e.g., a processor or a group of processors
  • software e.g., an application or application portion
  • a hardware module may be implemented mechanically or electronically.
  • a hardware module may comprise dedicated circuitry or logic that may be permanently configured (e.g., as a special-purpose processor, such as a field programmable gate array (FPGA) or an application-specific integrated circuit (ASIC)) to perform certain operations.
  • a hardware module may also comprise programmable logic or circuitry (e.g., as encompassed within a general-purpose processor or other programmable processor) that may be temporarily configured by software to perform certain operations. It will be appreciated that the decision to implement a hardware module mechanically, in dedicated and permanently configured circuitry, or in temporarily configured circuitry (e.g., configured by software) may be driven by cost and time considerations.
  • the term “hardware module” should be understood to encompass a tangible entity, be that an entity that is physically constructed, permanently configured (e.g., hardwired), or temporarily configured (e.g., programmed) to operate in a certain manner or to perform certain operations described herein.
  • hardware modules are temporarily configured (e.g., programmed)
  • each of the hardware modules need not be configured or instantiated at any one instance in time.
  • the hardware modules comprise a general-purpose processor configured using software
  • the general-purpose processor may be configured as respective different hardware modules at different times.
  • Software may accordingly configure a processor, for example, to constitute a particular hardware module at one instance of time and to constitute a different hardware module at a different instance of time.
  • Hardware modules may provide information to, and receive information from, other hardware modules. Accordingly, the described hardware modules may be regarded as being communicatively coupled. Where multiple of such hardware modules exist contemporaneously, communications may be achieved through signal transmission (e.g., over appropriate circuits and buses) that connect the hardware modules. In embodiments in which multiple hardware modules are configured or instantiated at different times, communications between such hardware modules may be achieved, for example, through the storage and retrieval of information in memory structures to which the multiple hardware modules have access. For example, one hardware module may perform an operation and store the output of that operation in a memory device to which it may be communicatively coupled. A further hardware module may then, at a later time, access the memory device to retrieve and process the stored output. Hardware modules may also initiate communications with input or output devices, and may operate on a resource (e.g., a collection of information).
  • a resource e.g., a collection of information
  • processors may be temporarily configured (e.g., by software) or permanently configured to perform the relevant operations. Whether temporarily or permanently configured, such processors may constitute processor- implemented modules that operate to perform one or more operations or functions.
  • the modules referred to herein may, in some example embodiments, comprise processor- implemented modules.
  • the methods or routines described herein may be at least partially processor-implemented. For example, at least some of the operations of a method may be performed by one or more processors or processor-implemented hardware modules. The performance of certain of the operations may be distributed among the one or more processors, not only residing within a single machine, but deployed across a number of machines.
  • the processor or processors may be located in a single location (e.g., within a home environment, an office environment, or as a server farm), while in other embodiments the processors may be distributed across a number of locations.
  • the performance of certain of the operations may be distributed among the one or more processors, not only residing within a single machine, but deployed across a number of machines.
  • the one or more processors or processor- implemented modules may be located in a single geographic location (e.g., within a home environment, an office environment, or a server farm). In other example embodiments, the one or more processors or processor-implemented modules may be distributed across a number of geographic locations.
  • any reference to “one embodiment” or “an embodiment” means that a particular element, feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment.
  • the appearances of the phrase “in one embodiment” in various places in the specification are not necessarily all referring to the same embodiment.
  • the terms “comprises,” “comprising,” “may include,” “including,” “has,” “having” or any other variation thereof, are intended to cover a non-exclusive inclusion.
  • a process, method, article, or apparatus that comprises a list of elements is not necessarily limited to only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
  • “or” refers to an inclusive or and not to an exclusive or. For example, a condition A or B is satisfied by any one of the following: A is true (or present) and B is false (or not present), A is false (or not present) and B is true (or present), and both A and B are true (or present).

Landscapes

  • Chemical & Material Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Chemical Kinetics & Catalysis (AREA)
  • Electrochemistry (AREA)
  • Molecular Biology (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Investigating Or Analyzing Materials By The Use Of Electric Means (AREA)

Abstract

Les procédés et les systèmes de l'invention concernent la détection électrochimique de concentrations d'ions dans un échantillon de fluide. Un procédé selon l'invention consiste à amener un potentiostat à générer un courant électrique, à recevoir une indication d'une pluralité de niveaux de tension à partir du potentiostat, à fournir une indication du courant électrique et de la pluralité de niveaux de tension à un modèle de régression d'apprentissage automatique entraîné (ML), à recevoir une indication d'un niveau de concentration d'ions à partir du modèle de régression ML entraîné, et à délivrer en sortie l'indication du niveau de concentration d'ions à un utilisateur.
PCT/US2024/016974 2023-02-24 2024-02-23 Détection électrochimique de concentrations d'ions pour une surveillance presque en temps réel Ceased WO2024178272A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US202363448200P 2023-02-24 2023-02-24
US63/448,200 2023-02-24

Publications (1)

Publication Number Publication Date
WO2024178272A1 true WO2024178272A1 (fr) 2024-08-29

Family

ID=92501520

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2024/016974 Ceased WO2024178272A1 (fr) 2023-02-24 2024-02-23 Détection électrochimique de concentrations d'ions pour une surveillance presque en temps réel

Country Status (1)

Country Link
WO (1) WO2024178272A1 (fr)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070017821A1 (en) * 2003-07-09 2007-01-25 Erik Bakker Reversible electrochemical sensors for polyions
US20200400596A1 (en) * 2018-03-12 2020-12-24 Consejo Superior De Investigaciones Científicas Device and method for sensing the conductivity of a fluid

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070017821A1 (en) * 2003-07-09 2007-01-25 Erik Bakker Reversible electrochemical sensors for polyions
US20200400596A1 (en) * 2018-03-12 2020-12-24 Consejo Superior De Investigaciones Científicas Device and method for sensing the conductivity of a fluid

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
GABRIELI GIANMARCO, HU RUI, MATSUMOTO KEIJI, TEMIZ YUKSEL, BISSIG SACHA, COX AARON, HELLER RALPH, LÓPEZ ANTONIO, BARROSO JORGE, KA: "Combining an Integrated Sensor Array with Machine Learning for the Simultaneous Quantification of Multiple Cations in Aqueous Mixtures", ANALYTICAL CHEMISTRY, AMERICAN CHEMICAL SOCIETY, US, vol. 93, no. 50, 21 December 2021 (2021-12-21), US , pages 16853 - 16861, XP093207126, ISSN: 0003-2700, DOI: 10.1021/acs.analchem.1c03709 *

Similar Documents

Publication Publication Date Title
Vlasov et al. Nonspecific sensor arrays (" electronic tongue") for chemical analysis of liquids (IUPAC Technical Report)
CN102445476B (zh) 操作具有至少一个探针的测量装置的方法
CA2843157C (fr) Extrapolation de donnees de capteur interpolees pour augmenter un debit echantillon
KR102605753B1 (ko) 인공지능 기반의 혈당 예측 시스템 및 방법
JP7003138B2 (ja) 少なくとも１つの検体の濃度を決定するための方法及びデバイス
US11175281B2 (en) Well plate cover with embedded electronic sensors for monitoring cell metabolism
CN107209174A (zh) 通过交流电阻抗相角检测以确定电化学红细胞比容的系统和方法
CA3060910A1 (fr) Systeme et procede de mesure d'analyte
JP6228639B2 (ja) 試料処理量を高めるための内挿センサデータの外挿
WO2019115687A1 (fr) Procédé et dispositif de détermination d'informations sur une résistance série équivalente
CN107636452B (zh) 改进的生物传感器系统分析物测量
Kumar et al. Identification of fractional order model for a voltammetric E-tongue system
WO2017153997A1 (fr) Désagrégation d'utilisation d'appareil à partir de données de compteur électrique
CN103168230B (zh) 测量电容的方法和用途
CN112305051A (zh) 智能浓度测量系统、方法及智能浓度分析模块
WO2024178272A1 (fr) Détection électrochimique de concentrations d'ions pour une surveillance presque en temps réel
US20040108223A1 (en) Simplified signal processing method for voltammetry
WO2004053476A1 (fr) Procede simplifie de traitement des signaux applique en voltametrie
Deshpande et al. Development of an insilico model of eccrine sweat using molecular modelling techniques
US20230168219A1 (en) Electrochemical-sensing apparatus and method therefor
Wongkittisuksa et al. Development and application of a real-time capacitive sensor
US20230288387A1 (en) Device and method for determining a concentration of a target gas in a mobile application
CN111207856B (zh) 测量温度的准确度获取方法、装置、设备及可读存储介质
CN118016201B (zh) 一种便携式组网水质检测方法及系统
Kumar et al. Development of electronic interface for sensing applications with voltammetric electronic tongue

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 24761031

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE