Focused linear model correction and linear model correction for multivariate calibration model maintenance

Abstract

The present application relates to focused linear model correction and linear model correction for multivariate calibration model maintenance. A device may obtain master beta coefficients of a master calibration model associated with a master instrument. The master beta coefficients may be at a grid of a target instrument. The device may perform constrained optimization of an objective function according to a set of constraints to determine a pair of transfer beta coefficients. The constrained optimization may be performed based on a pair of initial transfer beta coefficients, master beta coefficients, and a spectrum associated with a scout set. The device may determine a transfer beta coefficient based on the pair of transfer beta coefficients. The device may determine a final transfer beta coefficient based on a set of transfer beta coefficients including the transfer beta coefficient. The final transfer beta coefficient may be associated with generating a transfer calibration model corresponding to the master calibration model for use by the target instrument.

Description

[0001] This application is a divisional application of the application with the application date of July 10, 2019, the application number of 201910621447.3, and the invention title of "Focused Linear Model Calibration and Linear Model Calibration for Multivariate Calibration Model Maintenance". Background

[0002] Spectral instruments can be configured with calibration models for calibrating spectral measurements performed by the spectral instruments. Calibration models are typically generated based on reference values corresponding to known samples and spectra corresponding to the known samples measured by the spectral instruments.

[0003] Overview

[0004] According to some possible embodiments, a method may include: obtaining, by a device, main β (beta) coefficients of a main calibration model associated with a main instrument, where the main β coefficients are at a grid of a target instrument; performing, by the device, constrained optimization of an objective function according to a set of constraints to determine a pair of transfer β coefficients, where the constrained optimization is performed based on a pair of initial transfer β coefficients, the main β coefficients, and spectra associated with a scouting set; determining, by the device and based on the pair of transfer β coefficients, transfer β coefficients; and determining, by the device, final transfer β coefficients based on a set of transfer β coefficients including the transfer β coefficients, where the final transfer β coefficients are associated with generating a transfer calibration model corresponding to the main calibration model for use by the target instrument.

[0005] According to some possible embodiments, a method may include: determining, by a device, that a grid of a main instrument associated with a main calibration model does not match a grid of a target instrument for which a transfer calibration model corresponding to the main calibration model is to be generated; interpolating, by the device and based on determining that the grid of the main instrument does not match the grid of the target instrument, the β coefficients of the main calibration model to the grid of the target instrument; and determining, by the device, main β coefficients associated with generating the transfer calibration model based on a result of interpolating the β coefficients of the main calibration model to the grid of the target instrument.

[0006] According to some possible embodiments, a method may include: obtaining, by a device, a scouting set associated with an updated calibration model, where the scouting set includes spectra associated with a set of samples based on which the calibration model is to be updated; determining, by the device, β coefficients associated with the calibration model; determining, by the device and based on the β coefficients and using linear model correction (LMC) techniques, updated β coefficients associated with the updated calibration model; and updating, by the device, the calibration model based on the updated β coefficients.

[0007] Aspects of the present disclosure may be implemented in one or more of the following embodiments:

[0008] 1) A method, including:

[0009] Obtain the main β coefficients of the main calibration model associated with the main instrument, where the main β coefficients are located at the grid of the target instrument;

[0010] Perform constrained optimization of the objective function by the device according to a set of constraints to determine a pair of transfer β coefficients, where the constrained optimization is performed based on a pair of initial transfer β coefficients, the main β coefficients, and the spectra associated with the scout set;

[0011] Determine the transfer β coefficients by the device and based on the pair of transfer β coefficients; and

[0012] Determine the final transfer β coefficients by the device based on a set of transfer β coefficients including the transfer β coefficients, where the final transfer β coefficients are associated with generating a transfer calibration model corresponding to the main calibration model for use by the target instrument.

[0013] 2) The method according to 1), wherein the transfer calibration model is generated based on the final transfer β coefficients.

[0014] 3) The method according to 1), wherein the set of transfer β coefficients includes at least one other transfer β coefficient, and each other transfer β coefficient is determined based on a corresponding execution of the constrained optimization of the objective function according to a corresponding pair of initial transfer β coefficients.

[0015] 4) The method according to 1), wherein obtaining the main β coefficients includes:

[0016] Determine that the grid of the main instrument matches the grid of the target instrument; and

[0017] Identify the β coefficients of the main calibration model as the main β coefficients.

[0018] 5) The method according to 1), wherein obtaining the main β coefficients includes:

[0019] Determine that the grid of the main instrument does not match the grid of the target instrument;

[0020] Based on determining that the grid of the main instrument does not match the grid of the target instrument, interpolate the main calibration set to the grid of the target instrument to create interpolated calibration data;

[0021] Generate a regression model based on the interpolated calibration data; and

[0022] Determine the main β coefficients as the β coefficients of the regression model.

[0023] 6) The method according to 1), wherein obtaining the main β coefficients includes:

[0024] Determine that the grid of the master instrument does not match the grid of the target instrument;

[0025] Based on determining that the grid of the master instrument does not match the grid of the target instrument, interpolate the β coefficient of the master calibration model to the grid of the target instrument; and

[0026] Determine the master β coefficient based on the result of interpolating the β coefficient of the master calibration model to the grid of the target instrument.

[0027] 7) The method according to 6), wherein, based on determining that the master calibration set associated with the master calibration model is not available, interpolate the β coefficient of the master calibration model to the grid of the target instrument.

[0028] 8) The method according to 1), wherein the set of constraints includes correlation constraints associated with each of the master β coefficient and each of the pair of transfer β coefficients.

[0029] 9) The method according to 1), wherein the set of constraints includes slope constraints associated with each of the master β coefficient and each of the pair of transfer β coefficients.

[0030] 10) The method according to 1), wherein the set of constraints includes calibration range constraints of predicted values associated with the reconnaissance set.

[0031] 11) The method according to 1), further comprising generating the pair of initial transfer β coefficients based on at least one of the following:

[0032] Random generation of the pair of initial transfer β coefficients,

[0033] Applying a linear function associated with a random value to the master β coefficient, or

[0034] Adding a random value to the master β coefficient.

[0035] 12) A method, comprising:

[0036] Determine by the device that the grid of the master instrument associated with the master calibration model does not match the grid of the target instrument, and generate a transfer calibration model corresponding to the master calibration model for the target instrument;

[0037] Interpolate the β coefficient of the master calibration model to the grid of the target instrument by the device and based on determining that the grid of the master instrument does not match the grid of the target instrument; and

[0038] The device determines a master beta coefficient associated with generating the transfer calibration model based on the result of interpolating the beta coefficients of the master calibration model to the grid of the target instrument.

[0039] 13) The method according to 12), wherein, based on determining that the master calibration set associated with the master calibration model is unavailable, the beta coefficients of the master calibration model are interpolated to the grid of the target instrument.

[0040] 14) The method according to 12) further comprises:

[0041] Performing constrained optimization of an objective function according to a set of constraints to determine a pair of transfer beta coefficients, wherein the constrained optimization is performed based on a pair of initial transfer beta coefficients, the master beta coefficient, and the spectra associated with the scout set;

[0042] Determining a transfer beta coefficient based on the pair of transfer beta coefficients;

[0043] Determining a final transfer beta coefficient based on a set of transfer beta coefficients including the transfer beta coefficient; and

[0044] Generating the transfer calibration model based on the final transfer beta coefficient.

[0045] 15) The method according to 14), wherein the set of constraints includes:

[0046] Relevance constraints associated with each of the master beta coefficient and the transfer beta coefficients in the pair of transfer beta coefficients,

[0047] Slope constraints associated with each of the master beta coefficient and the transfer beta coefficients in the pair of transfer beta coefficients, and

[0048] Calibration range constraints of the predicted values associated with the scout set.

[0049] 16) The method according to 14) further comprises generating the pair of initial transfer beta coefficients based on at least one of the following:

[0050] Random generation of the pair of initial transfer beta coefficients,

[0051] Applying a linear function associated with a random value to the master beta coefficient, or

[0052] Adding a random value to the master beta coefficient.

[0053] 17) The method according to 12) further comprises:

[0054] Determining a transfer beta coefficient associated with generating the transfer calibration model based on the master beta coefficient and using linear model correction (LMC) technology; and

[0055] Generate the transfer calibration model based on the transfer β coefficient.

[0056] 18) The method according to 17), wherein a reference value for a scout set associated with using the LMC technique is predicted based on the main calibration model and the main transfer set.

[0057] 19) A method, comprising:

[0058] Obtain, by a device, a scout set associated with an updated calibration model, wherein the scout set includes spectra associated with a set of samples, and the calibration model will be updated based on the set of samples;

[0059] Determine, by the device, a β coefficient associated with the calibration model;

[0060] Determine, by the device and based on the β coefficient and using linear model correction (LMC) technique, an updated β coefficient associated with updating the calibration model; and

[0061] Update, by the device, the calibration model based on the updated β coefficient.

[0062] 20) The method according to 19), wherein the update of the calibration model is performed during operation of the device. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figures 1A - 1C is an overview diagram of an example embodiment described herein.

[0064] Figure 2 is a diagram of an example environment in which the systems and / or methods described herein may be implemented.

[0065] Figure 3 is Figure 2 a diagram of example components of one or more devices of

[0066] Figure 4 is a flowchart of an example process of a focused linear model correction technique as described herein, which is associated with determining a transfer β coefficient for generating a transfer calibration model.

[0067] Figures 5A - 5C is associated with Figure 4 an example diagram of a focused linear model correction technique of

[0068] Figure 6 is a flowchart of an example process for interpolating the β coefficient of a main calibration model to a grid of a target instrument to determine a main β coefficient for use in a focused linear model correction technique or a linear model correction technique as described herein.

[0069] Figures 7A - 7C and Figure 8A and Figure 8B are diagrams associated with interpolating the β coefficients of a master calibration model to the grid of a target instrument and using the LMC technique and the fLMC technique associated with performing calibration model transfer, respectively.

[0070] Figures 9A - 9D , Figure 10A , Figure 10B , Figure 11A and Figure 11B are diagrams showing example results associated with standardizing a calibration model across multiple instruments.

[0071] Figure 12 is a flowchart of an example process for model update using the linear model correction technique as described herein.

[0072] Figure 13A and Figure 13B are diagrams showing example results of performing calibration model update using the linear model correction technique.

[0073] Figure 14A and Figure 14B are diagrams showing example results associated with predicting reference values using a master calibration model and a master transfer set. DETAILED DESCRIPTION

[0074] The following detailed description of example embodiments refers to the accompanying drawings. The same reference numbers in different drawings may identify the same or similar elements.

[0075] Calibration model transfer and calibration model update are two important areas of multivariate calibration model maintenance for spectral applications such as those in the near-infrared (NIR) region.

[0076] In some cases, when using a multivariate calibration model developed on a first spectral instrument (or under one environmental condition) to predict calibration attributes of samples measured on a second spectral instrument (or by the first spectral instrument under different environmental conditions), the results are unacceptable. Additionally, even for the same spectral instrument, the signal may drift over time, meaning that existing calibration models will need to be updated. When updating a calibration model, to avoid the cumbersome and expensive task of recollecting data and recalibrating the existing calibration model, calibration transfer techniques can be implemented to transfer the calibration model from one condition to another, regardless of the source of the drift.

[0077] One requirement of general calibration model transfer techniques is to obtain a transfer dataset that includes spectra of the same set of samples collected by a first instrument (e.g., the master instrument from which the calibration model will be transferred, or a given instrument under original conditions) and a second instrument (e.g., the target instrument to which the calibration model will be transferred, or a given instrument under target conditions). In some cases, obtaining the transfer dataset is difficult or impossible. For example, when a calibration model for perishable materials needs to be transferred from a master instrument in one country to a target instrument in another country, it may be impossible to obtain the transfer dataset.

[0078] The Linear Model Correction (LMC) technique can solve this problem by only requiring a small number of spectra collected only by the target instrument. The set of spectra used by the LMC technique is called the scout set. However, in order for the LMC technique to work, the reference values of the scout set (e.g., the actual values measured in a chemical laboratory) are needed. In some cases, obtaining these reference values can be very time-consuming and / or expensive.

[0079] Some embodiments described herein provide a Focused LMC (fLMC) technique that can be used in association with performing calibration model transfer. Similar to the LMC technique, the fLMC technique only requires a scout set collected by the target instrument. However, different from the LMC technique, the fLMC technique does not require the reference values of the scout set. Therefore, using the fLMC technique in association with calibration model transfer reduces the cost, difficulty, and / or complexity of calibration model transfer (e.g., compared to the LMC technique and the above general calibration model transfer techniques).

[0080] In addition, calibration model transfer from a master instrument with relatively high spectral resolution and / or relatively wide wavelength range to a target instrument with relatively low spectral resolution and / or relatively narrow wavelength range is often encountered (e.g., when transferring a calibration model from a benchtop instrument to a portable instrument). For calibration model transfer in such cases, general calibration transfer techniques require a complete master calibration set (e.g., a set of spectra associated with a set of samples measured by the master instrument) in order to initiate the calibration model transfer process. Here, the spectra of the master calibration set are interpolated into the grid of the target instrument, and then an intermediate model is developed to transfer to the target instrument.

[0081] However, it is not always possible to access the master calibration set. Even when the master calibration set is accessible, in some cases, the master database may be large and / or may have a long maintenance history. Therefore, it may be difficult and / or time-consuming to obtain a clean master calibration set from the database.

[0082] Some embodiments described herein provide a process in which the fLMC technique or the LMC technique uses the beta coefficients of a primary calibration model in connection with performing calibration model transfer without requiring a primary calibration set. The use of beta coefficients (instead of a primary calibration set) reduces the cost, difficulty, and / or complexity of calibration model transfer.

[0083] In addition, when a calibration model developed on a primary instrument is to be deployed on multiple other instruments (e.g., multiple different target instruments) that may have inter-instrument variations, it may be difficult to perform calibration model transfer using conventional calibration model transfer techniques (e.g., when the target instruments are located at a great distance from the primary instrument). In some embodiments, to address this issue, the LMC technique or the fLMC technique can be configured on the multiple target instruments. When the primary calibration model is transferred to the target instruments, the user only needs to collect a scout set (e.g., spectra from several samples associated with a given application). The calibration model can use these spectra to perform automatic correction in combination with the LMC technique (e.g., when reference values are available) or in combination with the fLMC technique (e.g., regardless of whether reference values are available).

[0084] In addition, as described above, after a calibration model is deployed on a given instrument, calibration model updates may be required (e.g., due to changes in samples, measurement environment, etc.). A general technique for performing calibration model updates is to add new samples to an existing calibration set and then reconstruct the calibration model. However, this technique may require a large number of samples to fit the calibration model to the new samples or new conditions. In addition, this technique requires all calibration data to be available. Further, when the calibration database is large, especially when the spectral range is wide and the spectral resolution is high, reconstructing the calibration model may consume a large amount of time and / or resources (e.g., processor resources, battery power, etc.). Thus, it may be impossible to update the calibration model during the online operation of the instrument.

[0085] Some embodiments described herein provide techniques for using the LMC technique to perform calibration model updates. In essence, the LMC technique requires a relatively small number of samples to perform calibration model updates. In some embodiments, the update set (i.e., the scout set associated with performing calibration model updates) can include samples representing different conditions of future samples in order to make future predictions more accurate. In addition, calibration model updates using the LMC technique are relatively faster than the general update technique described above. For example, calibration model updates using the LMC program can be performed in a few seconds, making it possible to update the calibration model during online operation.

[0086] In addition, in some cases, transfer sets from both the master instrument and the target instrument may be available, while the reference values of the transfer sets may not be available. In such cases, as described herein, the transfer set from the target instrument can be used as the scout set and the reference value predicted by the master calibration model of the transfer set from the master instrument can be used as the reference value to perform the LMC technique. Thus, when spectral data is available, the LMC technique can be performed, otherwise the spectral data can be used to perform other conventional calibration transfer techniques.

[0087] Figures 1A - 1C is an illustration of an example embodiment described herein. Figure 1A and Figure 1B is an illustration of an example embodiment 100 associated with using a Focused Linear Model Correction (fLMC) correction technique, which is associated with generating a transfer calibration model corresponding to a master calibration model associated with a master instrument for configuration on a target instrument.

[0088] For Figure 1A and Figure 1B the purposes of the example embodiment 100, the master calibration model configured on the master instrument will be transferred to the target instrument. In other words, a transfer calibration model corresponding to the master calibration model configured on the master instrument will be generated for use by the target instrument. The example embodiment 100 describes the use of the fLMC technique associated with generating the transfer calibration model.

[0089] As shown by reference numeral 105 in Figure 1A a modeling device (e.g., a device associated with generating the transfer calibration model) can obtain the master β - coefficients of the master calibration model at the grid of the target instrument.

[0090] The master β - coefficients can include a set of coefficients associated with the master calibration model. For example, the master β - coefficients can include a vector of regression coefficients associated with a Partial Least Squares (PLS) regression calibration model configured on the master instrument.

[0091] As described above, the master β - coefficients are located on the grid of the target instrument. The grid of the target instrument is a parameter of the target instrument defined by the spectral resolution and wavelength range of the target instrument. Similarly, the grid of the master instrument is a parameter of the master instrument defined by the spectral resolution and wavelength range of the master instrument. In some embodiments, the grid of the master instrument can be different from the grid of the target instrument (e.g., when the master instrument has a relatively higher spectral resolution and / or a wider wavelength range compared to the target instrument). Alternatively, the grid of the master instrument can match the grid of the target instrument (e.g., when the spectral resolution and wavelength range of the master instrument match the spectral resolution and wavelength range of the target instrument within a threshold amount).

[0092] In some embodiments, the way the modeling device obtains the main β coefficient can be based on whether the grid of the main instrument matches the grid of the target instrument.

[0093] For example, the modeling device can determine whether the grid of the main instrument matches the grid of the target instrument (e.g., based on information provided by the main instrument and / or the target instrument, based on information stored or accessible by the modeling device). In some embodiments, if the modeling device determines that the grid of the main instrument matches the grid of the target instrument, the modeling device can identify the β coefficient of the main calibration model as the main β coefficient. In other words, when the grid of the main instrument matches the grid of the target instrument, the modeling device can directly use the β coefficient of the main instrument as the main β coefficient (e.g., because the β coefficient of the main calibration model is already on the grid of the target instrument). In this case, the β coefficient of the main calibration model can be used as the main β coefficient regardless of whether the main calibration set associated with the main calibration model is available.

[0094] In some embodiments, if the modeling device determines that the grid of the main instrument does not match the grid of the target instrument, the main instrument can obtain the main β coefficient based on the main calibration set associated with the main calibration model. For example, if the grid of the main instrument does not match the grid of the target instrument, the modeling device can interpolate the main calibration set to the grid of the target instrument to create interpolated calibration data (i.e., the spectrum of the main calibration set interpolated to the grid of the target instrument). Here, the modeling device can generate a regression model (e.g., a PLS model, a principal component regression (PCR) model, etc.) based on the interpolated calibration data, and can determine the main β coefficient as the β coefficient of the regression model. In some embodiments, when the main calibration set is available, the modeling device can obtain the main β coefficient in this way. For example, the modeling device can determine that the main calibration set is available (e.g., accessible, not exceeding a threshold size or complexity level), and can proceed as described above.

[0095] In some embodiments, if the modeling device determines that the grid of the main instrument does not match the grid of the target instrument, the main instrument can obtain the main β coefficient based on the β coefficient of the main calibration model, an example of which is shown in Figure 1C as follows.

[0096] Figure 1CFIG. 150 is an illustration of an example implementation 150 associated with interpolating the β coefficients of a master calibration model to a grid of a target instrument to obtain master β coefficients. As shown by reference numeral 155, the modeling device may determine that the grid of the master instrument does not match the grid of the target instrument. As shown by reference numeral 160, based on this determination, the modeling device may interpolate the β coefficients of the master calibration model to the grid of the target instrument. As shown by reference numeral 165, the result of interpolating the β coefficients of the master calibration model to the grid of the target instrument may be used as the master β coefficients. In some implementations, when the master calibration set is not available, the modeling device may obtain the master β coefficients in this manner. For example, the modeling device may determine that the master calibration set is not available (e.g., inaccessible, exceeds a threshold size or complexity level), and may proceed as described above.

[0097] In some implementations, the modeling device may determine the master β coefficients by correlatively using the fLMC technique (as described in connection with example implementation 100) transmitted for the calibration model based on interpolating the β coefficients of the master calibration model to the grid of the target instrument. Additionally or alternatively, the modeling device may determine the master β coefficients by correlatively using the LMC technique based on interpolating the β coefficients of the master calibration model to the grid of the target instrument. In other words, the interpolation of the β coefficients of the master calibration model to the grid of the target instrument may be used correlatively with performing the fLMC technique or the LMC technique transmitted for the calibration model.

[0098] Returning to the fLMC technique associated with example implementation 100, in some implementations, the modeling device may determine the final transmitted β coefficients based on a set of transmitted β coefficients. The final transmitted β coefficients are the β coefficients used to generate the transmitted calibration model. In some implementations, as described below, the modeling device may determine each of the set of transmitted β coefficients based on a corresponding iteration of the constrained optimization of an objective function.

[0099] In some implementations, for each iteration, the modeling device may perform the constrained optimization of the following objective function:

[0100]

[0101] with the following constraints:

[0102] corr(b transA ,b 主 )≥r (1)

[0103] corr(b transB ,b 主 )≥r (2)

[0104] slope(b transA ,b 主 )≥r (3)

[0105] slope(b transB ,b 主 )≥r (4)

[0106] minY cal <<X 侦察 b transA <<maxY cal (5)

[0107] minY cal <<X 侦察 b transB <<maxY cal (6)

[0108] where X 侦察 is a reconnaissance set (e.g., the spectrum of the reconnaissance set measured by a target instrument), b transA and b transB are a pair of transfer β coefficients associated with a given iteration, b 主 is the main β coefficient, r is a constraint threshold, and minY cal and maxY cal define a calibration range associated with the target instrument.

[0109] In some embodiments, the constraint threshold r (e.g., the relevant constraints and / or slope constraints as described in the above equations) can be optimized using a validation set. In this case, a set of constraint thresholds r can be used iteratively, and the best r (e.g., determined based on the root mean square error of prediction (RMSEP) of the validation set) can be used in association with determining the resulting transfer β coefficients. In some embodiments, this constraint threshold optimization can be used in association with the fLMC technique or the LMC technique.

[0110] To establish an objective function, the concept of reproducibility is introduced. Assuming that each of a pair of transfer β coefficients b transA and b transB is capable of fitting the reconnaissance set, the difference in the predicted values of the reconnaissance set using b transA and b transB should be small. Thus, the objective function is used to minimize the squared difference in the predicted values of the reconnaissance set using b transA and b transB By using this concept of reproducibility, the need for a reference value of the reconnaissance set is eliminated. In other words, due to this concept of reproducibility, the fLMC technique does not require a reference value of the reconnaissance set (unlike the LMC technique).

[0111] To obtain meaningful results, the minimization of the objective function needs to be carried out under a set of constraints. For example, the set of constraints can include those related to the main β coefficient (b 主 ) and each of a pair of transfer β coefficients (btransA and b transB ) associated constraints, the pair of transfer β coefficients are associated with a given iteration of the constrained optimization of the objective function. According to this associated constraint, b transA and b 主 the correlation between and b transB and b 主 the correlation between should satisfy a threshold (e.g., indicated by equations (1) and (2) respectively, where, for example, the value r can be greater than or equal to 0.95).

[0112] As another example, the set of constraints can include slope constraints associated with each of the main β coefficient and a pair of transfer β coefficients, the pair of transfer β coefficients being associated with a given iteration of the constrained optimization of the objective function. According to this slope constraint, b transA and b 主 the slope between and b transB and b 主 the slope between should satisfy a threshold (e.g., indicated by equations (3) and (4) respectively, where, for example, the value r can be greater than or equal to 0.95).

[0113] As another example, the set of constraints can include calibration range constraints for the predicted values associated with the reconnaissance set. According to this calibration constraint, the values of the reconnaissance set predicted using b transA (i.e., X 侦察 b transA ) and the values of the reconnaissance set predicted by b transB (i.e., X 侦察 b transB ) should be within the calibration range (e.g., indicated by equations (5) and (6) respectively), or within a range close to the reference value of the reconnaissance set.

[0114] To initiate a given iteration of the above constrained optimization process, initial values of b transA and b transB are required (i.e., b transA0 and b transB0 respectively). In some embodiments, the modeling device can generate a pair of initial transfer β coefficients based on the random generation of a pair of initial transfer β coefficients. Additionally or alternatively, the modeling device can generate a pair of initial transfer β coefficients based on applying a linear function associated with a random value to the main β coefficient (e.g., b transA0 , b transB0 = m × b 主 + n, where m and n are random numbers). Additionally or alternatively, the modeling device can generate a pair of initial transfer β coefficients based on adding a random value to the main β coefficient (e.g., b transA0 , b transB0 = b 主+n, where n is a random number) to generate an initial pair of transfer β coefficients.

[0115] For a given iteration of the constrained optimization, the modeling device can generate an initial pair of transfer β coefficients (e.g., b transAi0 and b transBi0 for iteration i, and b transAk0 and b transBk0 for iteration k), and can perform the constrained optimization of the objective function to determine a pair of transfer β coefficients (e.g., b transAi and b transBi for iteration i, and b transAk and b transBk for iteration k). Then, the modeling device can determine the transfer β coefficients (e.g., b transi for iteration i and b transk for iteration k) based on the pair of transfer β coefficients. For example, as shown by reference numeral 110 regarding iteration i, the modeling device can generate b transAi0 and b transBi0 , perform the constrained optimization of the objective function to determine b transAi and b transBi , and determine the transfer β coefficient associated with iteration i (b transAi and b transBi ) based on the pair of transfer β coefficients (e.g., based on the average of b transi ). As another example, as shown by reference numeral 115 regarding iteration k, the modeling device can generate b transAk0 and b transBk0 , perform the constrained optimization of the objective function to determine b transAk and b transBk , and determine the transfer β coefficient associated with iteration k (b transAk and b transBk ) based on the pair of transfer β coefficients (e.g., based on the average of b transk ). Here, b transi and b transk are included in a set of transfer β coefficients, and the modeling device can determine the final transfer β coefficient (b trans ) based on this set of transfer β coefficients.

[0116] In some embodiments, the modeling device can be configured to perform multiple (e.g., 5 times, 20 times, 100 times, etc.) iterations of the constrained optimization of the objective function (e.g., to avoid deviation results based on the random nature of a pair of initial transfer β coefficients).

[0117] As Figure 1B shown by reference numeral 120 intrans )。 For example, the modeling device can determine the final transfer β coefficient to be equal to the average value, median value, mode, etc. of the set of transfer β coefficients (e.g., b transi to b transk ).

[0118] As shown by reference numeral 125, the modeling device can generate a transfer calibration model based on the final transfer β coefficient. For example, the modeling device can generate a regression model (e.g., PLS model, PCR model, etc.) based on the final transfer β coefficient. As shown by reference numeral 130, the modeling device can provide the transfer calibration model to the target instrument (e.g., such that the target instrument can use the transfer calibration model). In this way, the modeling device can be configured to use the fLMC technique, which allows the modeling device to generate a transfer calibration model using the spectra associated with the scout set without the need for reference values of the scout set.

[0119] As described above, Figures 1A - 1C is provided only as an example. Other examples are possible and can be different from the example described with respect to Figures 1A - 1C description.

[0120] Figure 2 is a diagram of an example environment 200 in which the systems and / or methods described herein can be implemented. As Figure 2 shown, the environment 200 can include a primary instrument 205, a target instrument 210, a modeling device 215, and a network 220. The devices of the environment 200 can be interconnected via a wired connection, a wireless connection, or a combination of wired and wireless connections.

[0121] The primary instrument 205 includes a device configured with a primary calibration model and capable of performing spectral measurements on a sample. For example, the primary instrument 205 can include a benchtop (i.e., non-handheld) spectrometer device that performs spectroscopy (e.g., vibrational spectroscopy such as near-infrared (NIR) spectroscopy, mid-infrared spectroscopy (mid-IR), Raman spectroscopy, etc.). In some embodiments, the primary instrument 205 can be capable of obtaining spectral measurement results at a higher resolution than the spectral measurement results obtained by the target instrument 210 (i.e., the primary instrument 205 can be a high-resolution device while the target instrument 210 can be a low-resolution device). For example, the primary instrument 205 can be capable of obtaining spectral measurement results on 400 channels, while the target instrument 210 may be capable of obtaining spectral measurement results on 125 channels. In some embodiments, the primary instrument 205 can be configured with a primary calibration model for calibrating the spectral measurement results obtained by the primary instrument 205. In some embodiments, the primary instrument 205 can receive information and / or transmit information to another device (such as the modeling device 215) in the environment 200.

[0122] The target instrument 210 includes a device capable of performing spectral measurements on a sample based on a target calibration model, where the target calibration model can be generated based on information associated with a master calibration model associated with the master instrument 205 as described herein. For example, the target instrument 210 can include a mobile spectrometer device or a handheld spectrometer device that performs spectroscopy. In some embodiments, the target instrument 210 can be capable of obtaining spectral measurement results at a lower resolution than the spectral measurement results obtained by the master instrument 205. In some embodiments, the target instrument 210 can receive information from and / or transmit information to another device (such as the modeling device 215) in the environment 200.

[0123] The modeling device 215 includes a device capable of performing operations associated with the following actions: transferring the master calibration model from the master instrument 205 to the target instrument 210 (i.e., generating a transfer calibration model corresponding to the master calibration model) and / or updating the calibration model configured on a given instrument (such as the master instrument 205 or the target instrument 210) as described herein. For example, the modeling device 215 can include a server, a group of servers, a computer, a cloud computing device, etc. In some embodiments, the modeling device 215 can receive information from and / or transmit information to another device (such as the master instrument 205 and / or the target instrument 210) in the environment 200. In some embodiments, the modeling device 215 and the master instrument 205 can be implemented within a single device. Alternatively, in some embodiments, the modeling device 215 and the target instrument 210 can be implemented within a single device.

[0124] The network 220 includes one or more wired and / or wireless networks. For example, the network 220 can include a cellular network (such as a New Radio (NR / 5G) network, a Long Term Evolution (LTE) network, a 3G network, a Code Division Multiple Access (CDMA) network, etc.), a Public Land Mobile Network (PLMN), a Local Area Network (LAN), a Wide Area Network (WAN), a Metropolitan Area Network (MAN), a telephone network (such as a Public Switched Telephone Network (PSTN)), a private network, an ad hoc network, an intranet, the Internet, a fiber-based network, a cloud computing network, etc. and / or a combination of these or other types of networks.

[0125] Figure 2 The number and arrangement of the devices and networks shown are provided as an example. In fact, compared to Figure 2 the devices and networks shown, there can be additional devices and / or networks, fewer devices and / or networks, different devices and / or networks, or differently arranged devices and / or networks.

[0126] Furthermore, Figure 2 two or more of the devices shown can be implemented within a single device, or Figure 2The single device shown can be implemented as multiple distributed devices. For example, although the master instrument 205 and the modeling device 215 are described as two separate devices, the master instrument 205 and the modeling device 215 can be implemented within a single device. As another example, the target instrument 210 and the modeling device 215 can be implemented within a single device. Additionally or alternatively, a set of devices (e.g., one or more devices) of the environment 200 can perform one or more functions described as being performed by another set of devices of the environment 200.

[0127] Figure 3 is a diagram of example components of a device 300. The device 300 can correspond to the master instrument 205, the target instrument 210, and / or the modeling device 215. In some embodiments, the master instrument 205, the target instrument 210, and / or the modeling device 215 can include one or more devices 300 and / or one or more components of the device 300. As Figure 3 shown, the device 300 can include a bus 310, a processor 320, a memory 330, a storage component 340, an input component 350, an output component 360, and a communication interface 370.

[0128] The bus 310 includes components that permit communication among the components of the device 300. The processor 320 is implemented in hardware, firmware, or a combination of hardware and software. The processor 320 takes the form of a central processing unit (CPU), a graphics processing unit (GPU), an accelerated processing unit (APU), a microprocessor, a microcontroller, a field-programmable gate array (FPGA), an application-specific integrated circuit (ASIC), or another type of processing component. In some embodiments, the processor 320 includes one or more processors that can be programmed to perform functions. The memory 330 includes random access memory (RAM), read-only memory (ROM), and / or another type of dynamic or static storage device (e.g., flash memory, magnetic memory, and / or optical memory) that stores information and / or instructions for use by the processor 320.

[0129] The storage component 340 stores information and / or software related to the operation and use of the device 300. For example, the storage component 340 can include a hard disk (e.g., a magnetic disk, an optical disk, a magneto-optical disk, and / or a solid state disk), a compact disc (CD), a digital versatile disc (DVD), a floppy disk, a cartridge, a tape, and / or another type of non-transitory computer-readable medium along with corresponding drives.

[0130] The input component 350 includes components that allow the device 300 to receive information, for example, via user input (such as a touch screen display, keyboard, keypad, mouse, button, switch, and / or microphone). Additionally or alternatively, the input component 350 may include sensors for sensing information (such as a Global Positioning System (GPS) component, accelerometer, gyroscope, and / or actuator). The output component 360 includes components that provide output information from the device 300 (such as a display, speaker, and / or one or more light emitting diodes (LEDs)).

[0131] The communication interface 370 includes transceiver-like components (such as a transceiver and / or separate receiver and transmitter) that enable the device 300 to communicate with other devices, for example, via a wired connection, wireless connection, or a combination of wired and wireless connections. The communication interface 370 may allow the device 300 to receive information from another device and / or provide information to another device. For example, the communication interface 370 may include an Ethernet interface, optical interface, coaxial interface, infrared interface, radio frequency (RF) interface, Universal Serial Bus (USB) interface, Wi-Fi interface, cellular network interface, etc.

[0132] The device 300 may perform one or more of the processes described herein. The device 300 may execute these processes based on software instructions stored by a non-transitory computer-readable medium (such as the memory 330 and / or the storage component 340) and executed by the processor 320. A computer-readable medium is defined herein as a non-transitory memory device. A memory device includes storage space within a single physical storage device or storage space spread across multiple physical storage devices.

[0133] The software instructions may be read into the memory 330 and / or the storage component 340 from another computer-readable medium or from another device via the communication interface 370. When executed, the software instructions stored in the memory 330 and / or the storage component 340 may cause the processor 320 to perform one or more of the processes described herein. Additionally or alternatively, hardwired circuitry may be used instead of or in combination with the software instructions to perform one or more of the processes described herein. Accordingly, the embodiments described herein are not limited to any particular combination of hardware circuitry and software.

[0134] Figure 3 The number and arrangement of the components shown are provided as an example. In fact, compared to Figure 3 the components shown, the device 300 may include additional components, fewer components, different components, or components arranged differently. Additionally or alternatively, a set of components of the device 300 (such as one or more components) may perform one or more functions described as being performed by another set of components of the device 300.

[0135] Figure 4 is a flowchart of an example process 400 of a focused linear model correction (fLMC) technique associated with determining transfer beta coefficients for generating a transfer calibration model as described herein. In some embodiments, Figure 4 one or more process blocks of may be performed by the modeling device 215. In some embodiments, Figure 4 one or more process blocks of may be performed by another device or group of devices (such as the master instrument 205 and / or the target instrument 210) separate from or including the modeling device 215.

[0136] As Figure 4 shown, process 400 may include obtaining the master beta coefficients of the master calibration model associated with the master instrument, where the master beta coefficients are at the grid of the target instrument (block 410). For example, the modeling device 215 may obtain the master beta coefficients of the master calibration model associated with the master instrument 205, where as described above, the master beta coefficients are at the grid of the target instrument 210.

[0137] As Figure 4 further shown in, process 400 may include performing constrained optimization of an objective function according to a set of constraints to determine a pair of transfer beta coefficients, where the constrained optimization is performed based on a pair of initial transfer beta coefficients, the master beta coefficients, and the spectra associated with the scout set (block 420). For example, the modeling device 215 may perform constrained optimization of the objective function according to a set of constraints to determine a pair of transfer beta coefficients, where as described above, the constrained optimization is performed based on a pair of initial transfer beta coefficients, the master beta coefficients, and the spectra associated with the scout set.

[0138] As Figure 4 further shown in, process 400 may include determining transfer beta coefficients based on the pair of transfer beta coefficients (block 430). For example, as described above, the modeling device 215 may determine transfer beta coefficients based on the pair of transfer beta coefficients.

[0139] As Figure 4 further shown in, process 400 may include determining final transfer beta coefficients based on a set of transfer beta coefficients including the transfer beta coefficients, where the final transfer beta coefficients are associated with generating a transfer calibration model corresponding to the master calibration model for use by the target instrument (block 440). For example, the modeling device 215 may determine final transfer beta coefficients based on a set of transfer beta coefficients including the transfer beta coefficients, where the final transfer beta coefficients are associated with generating a transfer calibration model corresponding to the master calibration model for use by the target instrument 210.

[0140] Process 400 may include additional embodiments, such as any individual embodiment or any combination of embodiments described below and / or in combination with one or more other processes described elsewhere herein.

[0141] In some embodiments, the modeling device 215 and / or the target instrument 210 may generate a transfer calibration model based on the final transfer beta coefficients.

[0142] In some embodiments, a set of transfer beta coefficients includes at least one other transfer beta coefficient, each other transfer beta coefficient being determined based on a respective execution optimized for a constraint on an objective function according to a respective pair of initial transfer beta coefficients.

[0143] In some embodiments, when obtaining the master beta coefficients, the modeling device 215 may determine that the grid of the master instrument 205 matches the grid of the target instrument 210 and identify the beta coefficients of the master calibration model as the master beta coefficients.

[0144] In some embodiments, when obtaining the master beta coefficients, the modeling device 215 may determine that the grid of the master instrument 205 does not match the grid of the target instrument 210; based on determining that the grid of the master instrument 205 does not match the grid of the target instrument 210, interpolate the master calibration set to the grid of the target instrument 210 to create interpolated calibration data; generate a regression model based on the interpolated calibration data; and determine the master beta coefficients as the beta coefficients of the regression model.

[0145] In some embodiments, when obtaining the master beta coefficients, the modeling device 215 may determine that the grid of the master instrument 205 does not match the grid of the target instrument 210; based on determining that the grid of the master instrument 205 does not match the grid of the target instrument 210, interpolate the beta coefficients of the master calibration model to the grid of the target instrument 210; and determine the master beta coefficients based on the result of interpolating the beta coefficients of the master calibration model to the grid of the target instrument 210. In some embodiments, based on a determination that the master calibration set associated with the master calibration model is not available, the beta coefficients of the master calibration model are interpolated to the grid of the target instrument 210.

[0146] In some embodiments, in addition to the calibration range constraints on the predicted values associated with the scout set, a set of constraints includes correlation constraints associated with each of the master beta coefficients and a pair of transfer beta coefficients, and / or slope constraints associated with each of the master beta coefficients and a pair of transfer beta coefficients.

[0147] In some embodiments, the modeling device 215 may generate a pair of initial transfer beta coefficients based on a random generation of a pair of initial transfer beta coefficients, applying a linear function associated with a random value to the master beta coefficients, and / or adding a random value to the master beta coefficients.

[0148] While Figure 4illustrates an example block of process 400, but in some embodiments, compared to the blocks depicted in Figure 4 process 400 may include additional blocks, fewer blocks, different blocks, or blocks arranged differently. Additionally or alternatively, two or more blocks of process 400 may be executed in parallel.

[0149] To illustrate the effectiveness of the fLMC technique, a PLS regression model of the Brix of sugarcane was transferred from a bench-top FOSS NIR master instrument to a portable MicroNIR target instrument. Figures 5A - 5C is a diagram associated with the results of the transfer of this example calibration model using the fLMC technique.

[0150] A total of 1712 FOSS spectra were used to build the master calibration model. These spectra were first interpolated to the MicroNIR grid. The interpolated calibration data were used to build an intermediate master calibration model, and the resulting β coefficients were used as b 主 . The MicroNIR instrument collected 126 spectra, and 15 of these spectra were randomly selected as a scout set for performing fLMC. The remaining 111 spectra were used as an external validation set to validate the transferred calibration model. Using the FOSS validation set from the same 111 samples, the predictive performance of the transferred calibration model was compared with that of the master calibration model.

[0151] As Figure 5A shown, without performing the transfer of the calibration model, when using the intermediate master calibration model to predict the MicroNIR validation set, the root mean square error of prediction (RMSEP) was high. However, as Figure 5B shown, when using the fLMC technique for the transfer of the calibration model, the RMSEP decreased significantly. The RMSEP using the original FOSS model for the FOSS validation set was also calculated and used as a benchmark for evaluating the performance of the transferred calibration model. In Figure 5C it can be seen that the residuals between the predicted Brix values and the laboratory Brix values for the validation set through the transferred calibration model remained within approximately ±2RMSEP of the original FOSS master calibration model, indicating that the transferred MicroNIR calibration was within the original bounds of the FOSS calibration with a confidence of approximately 95%. These results show that the performance of the transferred calibration model generated using the fLMC technique was close to that of the original FOSS master calibration model (e.g., having a relatively wide wavelength range and relatively high spectral resolution).

[0152] In addition, for comparison, the same FOSS primary calibration model was transferred to MicroNIR using the mean difference correction (MDC) technique and the piecewise direct standardization (PDS) technique (two typical techniques for calibrating model transfer). To apply these two techniques, a transfer set consisting of 15 spectra from the primary instrument and the target instrument was used. These spectra were from the same samples used in the scout set when using the fLMC technique. Using the transfer calibration models with MDC and PDS, the RMSEP for the same validation set was 1.80 and 0.72, respectively. Therefore, in this case, the fLMC technique performed better than the MDC technique and worse than the PDS technique. However, different from the PDS technique, the fLMC technique does not require a transfer set on the primary instrument, thus making the cost and / or complexity of the fLMC technique relatively low while achieving similar performance.

[0153] As described above, Figures 5A - 5C is provided only as an example. Other examples are possible and can be different from the example regarding Figures 5A - 5C described.

[0154] Figure 6 is a flowchart of an example process 600 for interpolating the β coefficients of the primary calibration model to the grid of the target instrument to determine the primary β coefficients for use with the fLMC technique or the LMC technique.

[0155] In some embodiments, Figure 6 one or more process blocks of

[0156] In some embodiments, Figure 6 one or more process blocks of

[0157] As Figure 6 shown, process 600 may include determining that the grid of the primary instrument associated with the primary calibration model does not match the grid of the target instrument for which a transfer calibration model corresponding to the primary calibration model is to be generated (block 610). For example, as described above, the modeling device 215 may determine that the grid of the primary instrument 205 associated with the primary calibration model does not match the grid of the target instrument 210 for which a transfer calibration model corresponding to the primary calibration model is to be generated.

[0158] As Figure 6As further shown in, process 600 may include interpolating the β coefficients of the primary calibration model to the grid of the target instrument based on determining that the grid of the primary instrument does not match the grid of the target instrument (block 620). For example, as described above, the modeling device 215 may interpolate the β coefficients of the primary calibration model to the grid of the target instrument 210 based on determining that the grid of the primary instrument 205 does not match the grid of the target instrument 210.

[0159] As Figure 6 As further shown in, process 600 may include determining the primary β coefficients associated with generating the transfer calibration model based on the result of interpolating the β coefficients of the primary calibration model to the grid of the target instrument (block 630). For example, as described above, the modeling device 215 may determine the primary β coefficients associated with generating the transfer calibration model based on the result of interpolating the β coefficients of the primary calibration model to the grid of the target instrument 210.

[0160] Process 600 may include additional implementations, such as any single implementation or any combination of implementations described below and / or in combination with one or more other processes described elsewhere herein.

[0161] In some embodiments, based on the determination that the primary calibration set associated with the primary calibration model is not available, the β coefficients of the primary calibration model are interpolated to the grid of the target instrument 210.

[0162] In some embodiments, the modeling device 215 may perform constrained optimization of the objective function according to a set of constraints to determine a pair of transfer β coefficients, where the constrained optimization is performed based on a pair of initial transfer β coefficients, the primary β coefficients, and the spectra associated with the scout set. Here, the modeling device may determine the transfer β coefficients based on the pair of transfer β coefficients; the final transfer β coefficients may be determined based on a set of transfer β coefficients including the transfer β coefficients. In other words, in some embodiments, the modeling device 215 may use the fLMC technique to determine the final transfer β coefficients. In some embodiments, the set of constraints includes correlation constraints associated with each of the primary β coefficients and the pair of transfer β coefficients, slope constraints associated with each of the primary β coefficients and the pair of transfer β coefficients, and calibration range constraints of the predicted values associated with the scout set. In some embodiments, the modeling device 215 may generate a pair of initial transfer β coefficients based on the random generation of a pair of initial transfer β coefficients, applying a linear function associated with a random value to the primary β coefficients, or adding a random value to the primary β coefficients.

[0163] In some embodiments, the modeling device 215 may determine transfer beta coefficients associated with generating a transfer calibration model based on the primary beta coefficients and using linear model calibration (LMC) techniques. In other words, in some embodiments, the modeling device 215 may use LMC techniques to determine the final transfer beta coefficients. In some embodiments, reference values for a scout set associated with using LMC techniques are predicted based on a primary calibration model and a primary transfer set.

[0164] Although Figure 6 example blocks of process 600 are shown, in some embodiments, process 600 may include additional blocks, fewer blocks, different blocks, or differently arranged blocks compared to those depicted Figure 6 therein. Additionally or alternatively, two or more blocks of process 600 may be executed in parallel.

[0165] In some embodiments, as described above, the beta coefficients of the primary calibration model may be interpolated into the grid of the target instrument 210 and used as the primary beta coefficients. For example, in some embodiments, such techniques may be used in combination with LMC techniques or fLMC techniques. Figures 7A - 7C And Figure 8A and Figure 8B are diagrams associated with interpolating the beta coefficients of the primary calibration model into the grid of the target instrument and using LMC techniques and fLMC techniques respectively associated with performing calibration model transfer.

[0166] Using the same dataset as described above with respect to Figures 5A - 5C the LMC technique is performed using the result of interpolating the beta coefficients of the primary calibration model into the grid of the target instrument as the primary beta coefficients, and the results are shown in Figures 7A - 7C . Here, since there is no available primary calibration set, it is not possible to construct an intermediate primary calibration model. As Figure 7A shown, when directly using the interpolated beta coefficients to predict the validation set on the target instrument, the resulting RMSEP is high. As Figure 7B shown, when using the interpolated beta coefficients as the primary beta coefficients and performing the LMC technique, the RMSEP is significantly reduced. Additionally, as Figure 7C shown, the residuals between the predicted Brix values and the laboratory Brix values for the validation set through the transfer calibration model remain within ±2RMSEP of the original FOSS primary calibration model, with limited exceptions.

[0167] Furthermore, using the same dataset as described above with respect to Figures 5A - 5C the fLMC technique is performed using the result of interpolating the beta coefficients of the primary calibration model into the grid of the target instrument as the primary beta coefficients, and the results are shown in Figure 8A and Figure 8BAmong them. Although there is a slight performance degradation compared to using LMC technology, the RMSEP is significantly reduced compared to the case where (as shown in Figure 5A ) the calibration model transfer is not performed. As shown in Figure 8A , the RMSEP is quite low, where the normalized RMSEP is 7.7% (normalized to the average Brix value of the validation set). In addition, as shown in Figure 8B , most of the residuals between the predicted Brix values and the laboratory Brix values for the validation set by transferring the calibration model remain within ±2RMSEP of the original FOSS main calibration model. It should be noted that the fLMC technology is the only technology that can be used in the following situations: the main calibration set is not available, the grids between the main instrument and the target instrument are different, only the reconnaissance set collected by the target instrument is used for transfer, and there is no reference value for the reconnaissance set. In some embodiments, the performance of transferring the calibration model can be further improved with the update of the calibration model.

[0168] As described above, Figures 7A - 7C and Figure 8A and Figure 8B are provided only as examples. Other examples are possible and can be different from the examples described with respect to 7A- Figure 7C and Figure 8A and Figure 8B .

[0169] In some embodiments, the techniques described herein can be used to standardize calibration models among multiple instruments. As described above, instruments or devices of the same type typically encounter inter-instrument differences. Therefore, when a calibration model is developed on one instrument but needs to be deployed on multiple (e.g., hundreds, millions, etc.) instruments, the inter-instrument differences may result in inconsistent performance. For this problem, it may be impractical to use general methods to perform calibration model transfer, especially when the instruments are located at different positions. To solve this problem, the LMC technology and the fLMC technology can be configured on the instruments. Here, when the main calibration model is transferred to the target instrument, only the spectra from a small number of samples need to be collected. The calibration model can be automatically corrected using the LMC technology (e.g., when the reference value of the reconnaissance set is available) or using the fLMC technology (e.g., regardless of whether the reference value of the reconnaissance set is available).

[0170] Figures 9A - 9D , Figure 10A , Figure 10B , Figure 11A and Figure 11B are diagrams showing example results associated with standardizing calibration models among multiple instruments. In connection with Figures 9A - 9D , Figure 10A , Figure 10B , Figure 11A and Figure 11BIn the associated example, the raw data from the MicroNIR device was calibrated in two different ways (Data A and Data B) to simulate between-instrument variations. 759 spectra from 38 mixture samples were used to build a calibration model to predict caffeine content. 200 spectra from another 10 mixture samples were used as a validation set. As Figure 9A and Figure 9B shown, when using calibration model A to predict validation data A, or when using calibration model B to predict validation data B, the performance was similar. However, as Figure 9C shown, when using calibration model A to predict validation data B, the performance deteriorated. As Figure 9D shown, many of the residuals between the predicted values and the laboratory values for validation set B exceeded the ±2RMSEP benchmark for using model A to predict validation A.

[0171] Ten samples with three replicate spectra were randomly selected from calibration set B as a scout set to perform the LMC technique and the fLMC technique. As Figure 10A shown, using the LMC technique, the RMSEP was significantly reduced. As Figure 10B shown, all of the prediction residuals were within the ±2RMSEP benchmark for using model A to predict validation A. As Figure 11A shown, using the fLMC technique, the RMSEP was similarly reduced. As Figure 11B shown, most of the prediction residuals were within the ±2RMSEP benchmark for using model A to predict validation A. Therefore, it is effective to use the LMC technique or the fLMC technique to correct the between-instrument differences in model performance.

[0172] In fact, the LMC technique and the fLMC technique are effective when using as few as 8 samples as a scout set. It is worth noting that Figure 10A 、 Figure 10B 、 Figure 11A and Figure 11B The results in are examples of medium performance. The scout samples were randomly selected to simulate a real test scenario on the user side. The final performance results are affected by which samples are used as the scout set. Also, the fLMC technique performs less well than the LMC technique. However, when there are no reference values available for the scout set, the fLMC technique is the only technique available for calibration model transfer.

[0173] As described above, Figures 9A - 9D 、 Figure 10A 、 Figure 10B 、 Figure 11A and Figure 11B are provided only as examples. Other examples are possible and can be different from those regarding Figures 9A - 9D 、 Figure 10A 、 Figure 10B 、 Figure 11A andFigure 11B The described example.

[0174] As described above, in some cases, by using updated samples as the scout set for the LMC technique, the LMC technique can be applied to calibrate model updates. It should be noted that this does not require the use of all calibration data (e.g., as required by general model update techniques that add updated samples to the calibration set and recalibrate the model), and it takes a relatively small amount of time, such that the calibration model update can be performed during the online operation of the instrument (e.g., the main instrument 205, the target instrument 210).

[0175] Figure 12 is a flowchart of an example process 1200 for performing a calibration model update using the LMC technique. In some embodiments, Figure 12 one or more process blocks of can be performed by the modeling device 215. In some embodiments, Figure 12 one or more process blocks of can be performed by another device or a group of devices (such as the main instrument 205 and / or the target instrument 210) that is separate from or includes the modeling device 215.

[0176] As Figure 12 shown, process 1200 can include obtaining a scout set associated with updating the calibration model, where the scout set includes spectra associated with a set of samples based on which the calibration model will be updated (block 1210). For example, the modeling device 215 can obtain a scout set associated with updating the calibration model, where the scout set includes spectra associated with a set of samples based on which the calibration model will be updated.

[0177] As Figure 12 further shown, process 1200 can determine the β coefficients associated with the calibration model (block 1220). For example, the modeling device 215 can determine the β coefficients associated with the calibration model.

[0178] As Figure 12 further shown, process 1200 can include determining updated β coefficients associated with the updated calibration model based on the β coefficients and using the LMC technique (block 1230). For example, the modeling device 215 can determine updated β coefficients associated with the updated calibration model based on the β coefficients and using the LMC technique.

[0179] As Figure 12 further shown, process 1200 can include updating the calibration model based on the updated β coefficients (block 1240). For example, the modeling device 215 can update the calibration model based on the updated β coefficients (e.g., such that the updated calibration model uses the updated β coefficients associated with performing the calibration).

[0180] Process 1200 may include additional embodiments, such as any single embodiment or any combination of embodiments described below and / or in combination with one or more other processes described elsewhere herein.

[0181] In some embodiments, the update of the calibration model is performed during the operation of the instrument (e.g., the master instrument 205, the target instrument 210) without taking the device offline.

[0182] Although Figure 12 example blocks of process 1200 are shown, in some embodiments, process 1200 may include additional blocks, fewer blocks, different blocks, or differently arranged blocks compared to those depicted in Figure 12 In addition or alternatively, two or more blocks of process 1200 may be executed in parallel.

[0183] Figure 13A And Figure 13B are diagrams showing example results of performing the update of the calibration model using the linear model correction technique.

[0184] To update the Brix model of sugarcane (described above in connection with Figure 7B ), 30 additional MicroNIR spectra are used as the update set. Here, the LMC technique is applied to update the calibration model. Figure 13A is a diagram showing the prediction results associated with this update for the same validation set used in Figure 7B . As shown, the model performance is improved, where as Figure 13B shown, the RMSEP is reduced and the prediction residuals are reduced.

[0185] As indicated above, Figure 13A and Figure 13B are provided only as examples. Other examples are possible and may be different from the examples described with respect to Figure 13A and Figure 13B .

[0186] As described above, the LMC technique requires reference values for the scout set. When the transfer sets from both the master instrument 205 and the target instrument 210 are available, but the reference values for these samples are not available, the master calibration model and the master transfer set can be used to predict the reference values in order to make the LMC technique available.

[0187] Figure 14A and Figure 14B are diagrams showing example results associated with predicting the reference values using the master calibration model and the master transfer set. As Figure 14A shown, using the same data set as those associated with Figures 5A - 5C , when using the true reference values, the RMSEP is 0.44. As Figure 14BAs shown, when using the reference values predicted using the primary calibration model and the primary transfer set, the RMSEP is 0.80. Although the performance of the LMC technique using the predicted reference values is not as good as that of the LMC technique using the true reference values, as described above, this performance is improved compared to using the MDC technique or the fLMC technique and is similar to using the PDS technique.

[0188] As indicated above, Figure 14A and Figure 14B are provided only as examples. Other examples are possible and can be different from the examples described with respect to Figure 14A and Figure 14B described.

[0189] Some embodiments described herein provide a focused LMC (fLMC) technique that can be used in association with performing calibration model transfer. Similar to the LMC technique, the fLMC technique only requires a scout set collected by the target instrument. However, different from the LMC technique, the fLMC technique does not require reference values for the scout set. Thus, using the fLMC technique in association with calibration model transfer reduces the cost, difficulty, and / or complexity of calibration model transfer (e.g., compared to the LMC technique and the general calibration model transfer techniques described above).

[0190] Some embodiments described herein provide a process in which the fLMC technique or the LMC technique uses the β coefficients of the primary calibration model associated with performing calibration model transfer without requiring a primary calibration set.

[0191] Some embodiments described herein provide a process for model update using the LMC technique.

[0192] The foregoing disclosure provides illustration and description, but is not intended to be exhaustive or to limit the embodiments to the precise forms disclosed. Modifications and variations are possible or can be obtained from the practice of the embodiments in light of the above disclosure.

[0193] As used herein, the term component is defined to be broadly construed as hardware, firmware, and / or a combination of hardware and software.

[0194] Some embodiments are described herein in connection with thresholds. As used herein, meeting a threshold can refer to a value that is greater than the threshold, more than the threshold, higher than the threshold, greater than or equal to the threshold, less than the threshold, fewer than the threshold, lower than the threshold, less than or equal to the threshold, equal to the threshold, etc.

[0195] It will be apparent that the systems and / or methods described herein can be implemented in different forms of hardware, firmware, or a combination of firmware and software. The actual specific control hardware or software code used to implement these systems and / or methods is not a limitation on the implementation. Thus, when describing the operation and behavior of the systems and / or methods herein without reference to specific software code, it should be understood that the software and hardware can be designed to implement the systems and / or methods based on the description herein.

[0196] Although specific combinations of features are recited in the claims and / or disclosed in the specification, these combinations are not intended to limit the disclosure of possible implementations. In fact, many of these features can be combined in ways not specifically recited in the claims and / or not disclosed in the specification. Although each dependent claim listed below may directly depend on only one claim, the disclosure of possible implementations includes each dependent claim in combination with every other claim in the claim set.

[0197] Any element, act, or instruction used herein should not be construed as critical or essential unless specifically described as such. Additionally, as used herein, the articles "a" and "an" are intended to include one or more items and may be used interchangeably with "one or more." Further, as used herein, the term "set" is intended to include one or more items (e.g., related items, unrelated items, combinations of related and unrelated items, etc.) and may be used interchangeably with "one or more." The term "one" or similar language is used when only one item is intended to be indicated. Additionally, as used herein, the terms "has," "have," "having," etc. are intended to be open-ended terms. Further, unless otherwise expressly stated, the phrase "based on" is intended to mean "at least partially based on."

Claims

1. A method for generating a transfer calibration model, comprising: determining, by one or more devices, whether a grid of a master instrument matches a grid of a target instrument; identifying, by the one or more devices, master beta coefficients based on determining whether a grid of the master instrument matches a grid of the target instrument; generating, by the one or more devices and using a focused linear model correction (fLMC) technique, a transfer calibration model based on the principal beta coefficients; and providing the transfer calibration model to the target instrument by the one or more devices, The master instrument is a device configured to perform spectral measurement, and the target instrument is another device configured to perform spectral measurement.

2. The method according to claim 1, wherein: Generating the transfer calibration model includes: generating a pair of initial transfer beta coefficients based on the main beta coefficients; and Using the fLMC technique, the transfer calibration model is generated based on the pair of initial transfer beta coefficients.

3. The method according to claim 2, wherein: Generating the pair of initial transfer beta coefficients includes: The pair of initial transfer beta coefficients is generated by one or more of the following: Apply a linear function associated with a random value to the principal beta coefficients, or Add random values ​​to the main beta coefficients.

4. The method according to claim 1, wherein: Determining whether the grid of the master instrument matches the grid of the target instrument includes: A determination is made whether a grid of the host instrument matches a grid of the target instrument based on information provided by one or more of the host instrument or the target instrument.

5. The method according to claim 1, wherein: Determining whether the grid of the master instrument matches the grid of the target instrument includes: determining that the grid of the master instrument matches the grid of the target instrument, and Wherein identifying the principal beta coefficients comprises: Based on determining that the grid of the master instrument matches the grid of the target instrument, beta coefficients of a master calibration model are identified as the master beta coefficients.

6. The method according to claim 1, wherein: The transfer calibration model is generated without using reference values ​​from a scout set.

7. The method according to claim 1, wherein: When the grid of the master instrument matches the grid of the target instrument, the beta coefficients of the master calibration model are identified as the master beta coefficients, regardless of whether a master calibration set associated with the master calibration model is available.

8. An apparatus for generating a transfer calibration model, comprising: one or more memories; as well as one or more processors communicatively coupled to the one or more memories, the one or more processors configured to: determining whether the grid of the master instrument matches the grid of the target instrument; identifying master beta coefficients based on determining whether a grid of the master instrument matches a grid of the target instrument; generating a transfer calibration model based on the principal beta coefficients using a focused linear model correction (fLMC) technique; and providing the transfer calibration model to the target instrument, The master instrument is a device configured to perform spectral measurement, and the target instrument is another device configured to perform spectral measurement.

9. The device according to claim 8, wherein: The one or more processors, when generating the transfer calibration model, are configured to: generating a pair of initial transfer beta coefficients based on the main beta coefficients; as well as Using the fLMC technique, the transfer calibration model is generated based on the pair of initial transfer beta coefficients.

10. The device according to claim 9, wherein: The one or more processors, when generating the pair of initial transfer beta coefficients, are configured to: The pair of initial transfer beta coefficients is generated by one or more of the following: Apply a linear function associated with a random value to the principal beta coefficients, or Add random values ​​to the main beta coefficients.

11. The device according to claim 8, wherein: The one or more processors, when determining whether the grid of the master instrument matches the grid of the target instrument, are configured to: A determination is made whether a grid of the host instrument matches a grid of the target instrument based on information provided by one or more of the host instrument or the target instrument.

12. The device according to claim 8, wherein: The one or more processors, when determining whether the grid of the master instrument matches the grid of the target instrument, are configured to: determining that a grid of the master instrument matches a grid of the target instrument; and wherein the one or more processors, when identifying the principal beta coefficients, are configured to: Based on determining that the grid of the master instrument matches the grid of the target instrument, beta coefficients of a master calibration model are identified as the master beta coefficients.

13. The apparatus according to claim 8, wherein: The transfer calibration model is generated without using reference values ​​from a scout set.

14. The apparatus according to claim 8, wherein: When the grid of the master instrument matches the grid of the target instrument, the beta coefficients of the master calibration model are identified as the master beta coefficients, regardless of whether a master calibration set associated with the master calibration model is available.

15. A non-transitory computer readable medium storing a set of instructions, the set of instructions comprising: One or more instructions that, when executed by one or more processors of one or more devices, cause the one or more devices to: determining whether the grid of the master instrument matches the grid of the target instrument; identifying master beta coefficients based on determining whether a grid of the master instrument matches a grid of the target instrument; generating a transfer calibration model based on the principal beta coefficients using a focused linear model correction (fLMC) technique; and providing the transfer calibration model to the target instrument, The master instrument is a device configured to perform spectral measurement, and the target instrument is another device configured to perform spectral measurement.

16. The non-transitory computer readable medium of claim 15, wherein: The one or more instructions that cause the one or more devices to generate the transfer calibration model cause the one or more devices to: generating a pair of initial transfer beta coefficients based on the main beta coefficients; as well as Using the fLMC technique, the transfer calibration model is generated based on the pair of initial transfer beta coefficients.

17. The non-transitory computer readable medium of claim 16, wherein: The one or more instructions that cause the one or more devices to generate the pair of initial transfer beta coefficients cause the one or more devices to: The pair of initial transfer beta coefficients is generated by one or more of the following: Apply a linear function associated with a random value to the principal beta coefficients, or Add random values ​​to the main beta coefficients.

18. The non-transitory computer readable medium of claim 15, wherein: The one or more instructions that cause the one or more devices to determine whether the grid of the master instrument matches the grid of the target instrument cause the one or more devices to: A determination is made whether a grid of the host instrument matches a grid of the target instrument based on information provided by one or more of the host instrument or the target instrument.

19. The non-transitory computer readable medium of claim 15, wherein: The one or more instructions that cause the one or more devices to determine whether the grid of the master instrument matches the grid of the target instrument cause the one or more devices to: determining that a grid of the master instrument matches a grid of the target instrument; and wherein the one or more instructions that cause the one or more devices to identify the principal beta coefficients cause the one or more devices to: Based on determining that the grid of the master instrument matches the grid of the target instrument, beta coefficients of a master calibration model are identified as the master beta coefficients.

20. The non-transitory computer readable medium of claim 15, wherein: The transfer calibration model is generated without using reference values ​​from a scout set.

Images