Scientific instrument support system and method for mitigating spectral drift

Abstract

Scientific instrument support systems, related methods, computing devices, and computer readable media are disclosed herein. A method of mitigating distortion of an optical emission spectrum obtained from an optical emission spectrometer is provided. The method may include the steps of obtaining a spectrum recorded with a spectrometer and respective one or more condition parameters indicative of operating conditions at the time of recording the spectrum. The method may also include the step of providing a model configured to output one or more transformation parameters to be applied to a transformation of the obtained spectrum in response to the one or more condition parameters. Transforms may be applied to the obtained spectrum according to the obtained one or more transformation parameters to mitigate distortion of the spectrum due to differences between operating conditions and baseline operating conditions.

Description

[0001] The present disclosure is directed to mitigating distortions in optical emission spectra due to varying operating conditions of an optical spectrometer. Background Art

[0002] Scientific instruments for obtaining optical spectra may include a complex arrangement of movable parts, sensors, input ports and output ports, energy sources and consumable parts. Therefore, the obtained spectrum is sensitive to changes in the operating conditions of the scientific instrument (e.g., changes in temperature). For example, known methods for correcting optical spectra for temperature changes are disclosed in WO2021 / 018992A1 and GB2586046, which are incorporated herein by reference. These methods involve aligning the obtained spectrum with a reference spectrum using peaks identified in the obtained spectrum. An echelle dispersion grating can be used in optical emission spectrometry to achieve high dispersion of light. When used in combination with a second dispersive element such as a prism or another grating, a high-resolution 2D spectrum called an echellogram, also known as a full-frame or echellogram spectrum, can be obtained.

[0003] A method may include obtaining a spectrum recorded with a spectrometer and corresponding one or more condition parameters, which indicate the operating conditions of the optical emission spectrometer or the environment of the optical emission spectrometer when recording the spectrum with the spectrometer and / or at a previous time. For example, in some embodiments, the one or more condition parameters may include a parameter indicating temperature. The one or more condition parameters indicating temperature may include at least one temperature measurement result obtained, for example, from a temperature sensor attached to a mechanical structure of the optical emission spectrometer. In some embodiments, the one or more condition parameters may include a parameter related to or indicating temperature. The one or more condition parameters may include one or more of the following: a parameter indicating a heating current applied to a heating device for heating the optical system and / or stabilizing the temperature of the optical system; a parameter indicating an ambient temperature of the spectrometer; a parameter indicating the temperature of the optical system; a parameter indicating at least one temperature measurement obtained from a temperature sensor attached to a mechanical structure supporting one or more optical components of the optical system; a parameter indicating an RF power of a radio frequency (RF) generator used to generate a plasma used to obtain an emission spectrum; a parameter indicating a temperature of a power supply of the RF generator; a parameter indicating a voltage of the power supply of the RF generator; a parameter indicating a current of the power supply of the RF generator; a parameter indicating a voltage applied to a heating pad attached to the optical system; a parameter indicating a current at a control board of the scientific instrument; a parameter indicating a current at a printed circuit board (PCB) of a camera; a parameter indicating a current at a PCB of an optical part; a parameter indicating a temperature of one or more components in the PCB; a parameter indicating whether an additional gas option is installed; a parameter indicating whether the plasma is turned on; a parameter indicating an exhaust pressure of a plasma chamber for containing a plasma used to obtain an emission spectrum and a parameter indicating a temperature of an exhaust pipe. In some embodiments, the one or more condition parameters may include a condition change parameter, which indicates a change in a condition or a direction of change in a condition, which may be provided as part of the input data of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0004] Each embodiment will be easily understood by the following detailed description in conjunction with the accompanying drawings. To facilitate this description, the same reference numerals indicate the same structural elements. Each embodiment is illustrated in the various figures of the accompanying drawings by way of example and not limitation.

[0005] Figure 1 is a diagram of an example inductively coupled plasma optical emission spectrometer 1000 for obtaining an optical emission spectrum, according to some embodiments.

[0006] Figure 2 is a block diagram of an example scientific instrument support module for performing support operations according to various embodiments.

[0007] Figure 3 is an example full-frame spectrum of the intensity distribution captured by the photodetector.

[0008] Figure 4 is an example of an estimated warp profile for registering a shifted spectrum to a reference spectrum according to some embodiments.

[0009] Figure 5 is a flow chart of an example method of mitigating distortion of an optical emission spectrum according to various embodiments.

[0010] Figure 6 is a flow chart of an example method of obtaining training data according to various embodiments.

[0011] Figure 7 is a flowchart of an example method for training a machine learning model according to various embodiments.

[0012] Figure 8 Example input data for a machine learning model and example output data that the model is trained to provide are shown according to various embodiments.

[0013] Fig. 9A The prediction of lateral drift by a machine learning model in an example implementation is shown. Left: actual drift (grey) and predicted drift (black) Right: prediction error (difference). Fig. 9B The prediction of vertical drift by the machine learning model in an example implementation is shown. Left: actual drift (grey) and predicted drift (black) Right: prediction error (difference). Fig. 9C Shown is a comparison between actual lateral drift and predicted lateral drift in an exploratory test of a machine learning model trained using only one conditional parameter.

[0014] Fig.10 is an example of a graphical user interface that may be used to perform some or all of the supporting methods disclosed herein, according to various embodiments.

[0015] Fig.11 is a block diagram of an example computing device that may perform some or all of the scientific instrument support methods disclosed herein, according to various embodiments.

[0016] Fig.12 is a block diagram of an example scientific instrument support system in which some or all of the scientific instrument support methods disclosed herein may be performed, according to various embodiments. DETAILED DESCRIPTION

[0017] Scientific instrument support systems and related methods, computing devices, and computer-readable media are disclosed herein. In some aspects, a method for reducing the distortion of an optical emission spectrum obtained from an optical emission spectrometer is performed. The optical emission spectrometer includes an optical system for forming a spectrum, for example, the optical emission spectrometer can be an inductively coupled plasma emission spectrometer, which can be configured to record a mid-step emission spectrum of light emitted from a plasma. In some embodiments, the spectrometer can be a Raman spectrometer configured to record a spectrum of inelastic scattered photons; in some embodiments, the spectrometer can be an infrared spectrometer configured to record an interaction spectrum of infrared radiation interacting with an analyte. For the avoidance of doubt, it should be understood that references to "light" and "optical" refer to any portion of the electromagnetic spectrum, such as any one or more of the ultraviolet region, visible region, or infrared region of the electromagnetic spectrum.

[0018] A method may include obtaining a spectrum recorded with a spectrometer and corresponding one or more condition parameters, which indicate the operating conditions of the optical emission spectrometer or the environment of the optical emission spectrometer when recording the spectrum with the spectrometer and / or at a previous time. For example, in some embodiments, the one or more condition parameters may include a parameter indicating temperature. The one or more condition parameters indicating temperature may include at least one temperature measurement result obtained, for example, from a temperature sensor attached to a mechanical structure of the optical emission spectrometer. In some embodiments, the one or more condition parameters may include a parameter related to or indicating temperature. The one or more condition parameters may include one or more of the following: a parameter indicating a heating current applied to a heating device for heating the optical system and / or stabilizing the temperature of the optical system; a parameter indicating an ambient temperature of the spectrometer; a parameter indicating the temperature of the optical system; a parameter indicating at least one temperature measurement obtained from a temperature sensor attached to a mechanical structure supporting one or more optical components of the optical system; a parameter indicating an RF power of a radio frequency (RF) generator used to generate a plasma used to obtain an emission spectrum; a parameter indicating a temperature of a power supply of the RF generator; a parameter indicating a voltage of the power supply of the RF generator; a parameter indicating a current of the power supply of the RF generator; a parameter indicating a voltage applied to a heating pad attached to the optical system; a parameter indicating a current at a control board of the scientific instrument; a parameter indicating a current at a printed circuit board (PCB) of a camera; a parameter indicating a current at a PCB of an optical part; a parameter indicating a temperature of one or more components in the PCB; a parameter indicating whether an additional gas option is installed; a parameter indicating whether the plasma is turned on; a parameter indicating an exhaust pressure of a plasma chamber for containing a plasma used to obtain an emission spectrum and a parameter indicating a temperature of an exhaust pipe. In some embodiments, the one or more condition parameters may include a condition change parameter, which indicates a change in a condition or a direction of change in a condition, which may be provided as part of the input data of the model.

[0019] It should be understood that when referring to condition parameters indicating operating conditions at the time of recording, no particular timing accuracy is implied. That is, the condition parameters obtained at the time of recording affect or contain information about the state of the spectrometer at the time of recording, such as the mechanical configuration of the optical components affected by the ambient and optical system temperature, which affects the distortion in the recorded spectrum. Such a time "at the time of recording" may therefore be to some extent before or even after the actual precise time at which the spectrum was recorded, as long as the condition parameters at such a time affect or contain information about the distortion of the recorded spectrum relative to the baseline conditions.

[0020] A method may also include providing a machine learning model configured to output one or more transformation parameters of a transformation in response to the one or more conditional parameters, which will be applied to the obtained spectrum to reduce the distortion of the spectrum caused by the difference between the operating condition and the baseline operating condition. For example, in some embodiments, the one or more transformation parameters of the transformation may define all or part of the transformation to be applied to the obtained spectrum. For example, in some embodiments, the transformation includes at least one of the following: translation; rotation; scaling operation; centering operation; normalization operation; beveling operation; stretching operation; and deformation field. The one or more conditional parameters can be applied as input to the machine learning model, and one or more transformation parameters can be obtained as the output of the machine learning model. Such embodiments can reduce distortion under a wide range of operating conditions, and therefore, useful analysis results can be obtained under a wider range of operating conditions relative to conventional methods, and conventional methods require the spectrometer to operate under stable operating conditions within a constrained range to obtain useful analysis results. Typically, reaching the stable operating conditions within the constrained range requires running the spectrometer for a considerable period of time to allow the operating conditions to stabilize before performing any analysis, which wastes energy, time, and expensive consumables (such as argon used in plasma generation). By enabling the use of the spectrometer over a wider and varying range of operating conditions, this wasteful "start-up" period can be reduced or eliminated. Furthermore, while varying operating conditions might conventionally require analytical results to be "thrown away" (e.g., due to changes in the temperature of the spectrometer chamber), various embodiments of the embodiments disclosed herein allow valid and analytically useful spectra to be obtained during these varying conditions, thereby improving throughput and instrument availability. Additionally, because the transformation parameters are inferred based on the operating conditions, the described methods do not require fitting a transformation to each spectrum obtained, thereby allowing distortion mitigation to be performed in a computationally efficient manner.

[0021] In some embodiments, the input data for the machine learning model may include a time series of one or more condition parameters at each time point in a plurality of time points. For example, the time span of the time series may be 15 minutes. The number of optical emission spectra obtained in the entire time span of the time series may vary, and it should be understood that any appropriate number of optical emission spectra may be obtained. For example, in some embodiments, an optical emission spectrum may be obtained at each time point in the plurality of time points. In an example where the time series is 15 minutes, an optical emission spectrum may be obtained every minute, and one or more condition parameters may be obtained every minute or at a higher frequency, and a mean, standard deviation, mean value, and slope may be provided at each time step. In some embodiments, an optical emission spectrum may be obtained only at the last time point or at any time point. In some embodiments, an optical emission spectrum is not obtained within the time span of the time series, but is obtained at a predetermined time after the last time point of the time series. The time span of the time series of one or more condition parameters may be defined before or including any time period of the last time point in the one or more time points at which one or more optical emission spectra are obtained. It should be understood that in such embodiments where the input data includes a time series, the output data may include or may not include a time series of transformation parameters. In some embodiments, the time span includes a time point until and including the time when an optical emission spectrum is obtained. For example, in some embodiments, the input data includes a time series of one or more conditional parameters at each time point in a plurality of time points from a predetermined amount of time (e.g., 15 minutes) before obtaining the optical emission spectrum until and including the time when the optical emission spectrum is obtained. In these embodiments, the output data may include one or more transformation parameters to be applied to the transformation of the optical emission spectrum. In some embodiments, the input data includes a time series of one or more conditional parameters within a predetermined time span (e.g., 15 minutes), wherein the optical emission spectrum is obtained at each time point or at a subset of the time point. In some embodiments, the time series can be used to predict the output of the transformation parameters as a single time point. In such embodiments, the one or more conditional parameters may also include one or more of the mean, standard deviation, maximum-minimum difference, or variation slope of the conditional parameters for each time interval before the image recording time. For example, [0-1 minute, 1-2 minutes, ... 14-15 minutes]. In some embodiments, the output data may include a time series of one or more transformation parameters at each time point or a subset of the plurality of time points. In these embodiments, the method may include applying a transformation according to one or more transformation parameters at a time point of the obtained output data instance to an optical emission spectrum obtained at the corresponding time point.A transformation according to corresponding transformation parameters may be applied to the corresponding spectrum at each time point in the time series.

[0022] Since the dependence of spectral distortion on operating conditions may exhibit a degree of hysteresis or historical dependence, using a time series of operating conditions as input may improve the accuracy of the predicted transformation parameters. This is because the model may account for historical dependence compared to a model that uses only instantaneous operating conditions (where temporal information is lost). This is true whether the output of the model includes transformation parameters at a single time point or at multiple time points. In some embodiments, as an alternative or in addition to using a time series as input, an element of historical dependence may be accounted for in the input data by including a parameter in the input data that indicates a change or direction of change in the operating condition.

[0023] In some embodiments, the spectrum can be an image, and the transformation can include operations that vary across multiple positions in the image, for example, in some embodiments, the operation can include applying a distortion field (also referred to as a warp field) to the image in order to register the image. Therefore, a wide range of local distortions can be captured. Compared with global affine transformations such as translation, rotation, and scaling, such local non-affine transformations can capture and mitigate a wider range of spectral distortions, and therefore enable more accurate identification of peaks and corresponding analytes. Specifically, local transformations can capture functional groups with different degrees of distortion in different parts of the spectrum, thereby increasing the possibility of correct identification. Although the full text uses examples related to the spectrum represented as a digital image, it should be understood that the spectrum can be represented in any suitable manner, for example, according to a set of identified intensity peaks or valleys and their corresponding positions, such as corresponding two-dimensional coordinates of intensity peaks in the case of mid-step spectrograms or other two-dimensional spectra.

[0024] The machine learning model can be any appropriate machine learning model. For example, a random forest or gradient boosted decision tree (GBDT) learning algorithm, both of which are models that include multiple decision trees and use "bagging" and "boosting" techniques, respectively. In one embodiment, the XGBoost algorithm can be used. The XGBoost algorithm is a known decision tree-based integrated machine learning algorithm that is similar to GBDT using gradient boosting. XGBoost is introduced in the following paper: "Chen, T. and Guestrin, C., August 2016, XGBoost: A scalable tree boosting system, in Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining (pp. 785-794), incorporated herein by reference". A machine learning model using the XGBoost algorithm can be trained to provide one or more transformation parameters to be applied to the transformation of the obtained spectrum in response to a time series of one or more conditional parameters.

[0025] In some embodiments, the machine learning model includes a feedforward neural network. In embodiments where the input data includes a time series, one or more conditional parameters at each of the multiple time points can be applied together as input to the feedforward neural network. In some embodiments, the machine learning model may include a recurrent neural network, such as an LSTM recurrent neural network. In some embodiments, the machine learning model may include a transformer model.

[0026] In one aspect of the present disclosure, a method for training a machine learning model includes obtaining a training data set including training data pairs. Each training data pair includes the values ​​of one or more conditional parameters for the corresponding operating conditions of an optical emission spectrometer or the environment of the optical emission spectrometer that are different from the baseline operating conditions, and the corresponding values ​​of one or more transformation parameters. It should be understood that the values ​​of the conditional parameters and / or transformation parameters may be scalar values ​​or non-scalar values ​​(e.g., sets, vectors, or other multidimensional values). The method includes the following steps: adjusting the parameters of the machine learning model to reduce the difference between the values ​​of one or more transformation parameters in the training data pair and the values ​​of one or more transformation parameters output by the machine learning model in response to the corresponding values ​​of one or more conditional parameters of the training data pair. For example, the one or more transformation parameters may be lateral (X direction) and vertical (Y direction) transformations, and the model may be trained to minimize the mean square error between the values ​​of the lateral (X direction) and vertical (Y direction) transformations of the training data pair and the values ​​of the lateral (X direction) and vertical (Y direction) transformations output by the model. In one embodiment, the model trained to minimize the mean square error may use the XGBoost architecture.

[0027] In some aspects of the present disclosure, a method for obtaining training data for training a machine learning model includes: recording multiple optical emission spectra of a reference analyte with an optical emission spectrometer for corresponding operating conditions; and storing input data for the machine learning model. The input data includes, for each recorded optical emission spectrum, one or more parameters indicating the corresponding operating conditions at the time and / or previous time of the recorded spectrum. For each recorded optical emission spectrum, one or more transformation parameters for a transformation for aligning the optical emission spectrum to a baseline optical emission spectrum using a transformation are generated by adjusting these parameters to align the optical emission spectrum to the baseline optical emission spectrum. The baseline optical emission spectrum is recorded for a set of baseline operating conditions (e.g., operating conditions recommended by the manufacturer of the spectrometer) for capturing data. The method includes storing output data for the machine learning model, the output data including: for each optical emission spectrum, one or more corresponding adjusted transformation parameters (as training targets) associated with the corresponding input data (as training data pairs) of each optical emission spectrum. It should be understood that the training data can be generated using a spectrometer that is the same as or different from the spectrometer used to record the baseline spectrum. Likewise, the training data may include spectra recorded using different corresponding spectrometers, which may help generalize the model across different spectrometers. Preferably, the spectrometer used to generate the training data is of the same type, such as the same brand and / or model, as the spectrometer whose spectrum will be adjusted by the training model, but in some embodiments, the spectrometer used to generate the training data is of a different brand and / or model than the spectrometer whose spectrum will be adjusted by the training model. In some embodiments, the same spectrometer is used to generate the training data. This may help generate a model specific to the spectrometer, which may make the accuracy of the spectrometer higher.

[0028] The scientific instrument support embodiments disclosed herein can achieve improved performance relative to conventional methods. For example, although spectral distortion due to temperature changes is a common problem, conventional methods use peaks that appear in both reference spectra and sample spectra, such as carbon peaks, to calculate distortion from the expected position of the peak. The determined drift can then be used to offset the identified position of the unknown peak in the same sample spectrum. In some embodiments of the present disclosure, the spectrum is an image, and adjusting one or more transformation parameters to align the spectrum may include comparing the intensity between the images. Such embodiments do not require identification of peaks. Since there is no need to correctly identify the peak, the resulting method can be more flexible and accurate due to considering the image as a whole (as opposed to only considering the identified peak). Thus, the embodiments disclosed herein provide improvements to scientific instrument technology (e.g., improvements in computer technology supporting such scientific instruments, and other improvements).

[0029] Aspects and embodiments disclosed herein can achieve more flexible transformations by capturing a wider range of distortions relative to conventional methods. For example, conventional methods rely on the identification of peaks that appear in both the reference spectrum and the sample spectrum, and thus become difficult to implement when no peak has a clearly identifiable position. In addition, these methods have many technical problems and limitations caused by the fact that the offset is applied linearly to the entire spectrum, which is particularly relevant for step or full-frame spectra.

[0030] Various aspects and embodiments disclosed herein, such as embodiments in which the adjustment of transformation parameters provided in a method for providing training data for a machine learning model can include comparing the intensity between two images, can improve conventional methods to achieve technical advantages of reducing distortion without the need to find peaks. Such technical advantages cannot be achieved by routine and conventional methods, and all users of systems including such embodiments can benefit from these advantages (for example, by helping users perform technical tasks, such as providing a wider range of temperatures over which spectra can be obtained because no identifiable peaks are required). Therefore, the technical features of the embodiments disclosed herein are clearly unconventional in the field of spectroscopy, as are the combinations of features of the embodiments disclosed herein. The computational and user interface features disclosed herein not only involve the collection and comparison of information, but also apply new analysis and techniques to change the operation of machine learning models used in the field. Therefore, the present disclosure introduces functions that neither conventional computing devices nor humans can perform.

[0031] Therefore, the embodiments of the present disclosure can serve any of many technical purposes. These technical purposes include: controlling a specific technical system or process; determining how to control a machine based on measurement results; enhancement of analysis; reducing the amount of sensor data to be processed or providing faster processing of sensor data, the latter two technical purposes are at least partially attributed to the elimination of the need for peak identification to mitigate spectral distortion. In particular, the present disclosure provides a technical solution to the technical problem, including but not limited to providing a more flexible transformation that captures a wide range of local distortions, and eliminating the need to identify spectral peaks when analyzing spectra, and thus reducing the computational requirements for inference. Further technical solutions include providing an instrument that can be used under almost any operating conditions (such as temperature), and thus reducing the cost of expensive argon gas used in plasma, because a useful spectrum can be obtained before the operating conditions stabilize at the baseline operating conditions. Therefore, the embodiments disclosed herein provide improvements to analytical techniques (e.g., improvements in computer technology to support chemical analysis, and other improvements).

[0032] In the following detailed description, reference is made to the accompanying drawings, which form a part of the detailed description, wherein the same reference numerals designate the same parts throughout, and in the accompanying drawings, practicable embodiments are shown by way of example. It should be understood that other embodiments may be utilized and structural or logical changes may be made without departing from the scope of the present disclosure. Therefore, the following detailed description should not be considered to have a limiting meaning.

[0033] The various operations may be described as a plurality of discrete actions or operations in a manner that is most helpful for understanding the subject matter disclosed herein. However, the described order should not be interpreted as implying that these operations must rely on the order. Specifically, these operations may not be performed in the order presented. The described operations may be performed in an order different from the described embodiments. Various additional operations may be performed, and / or the described operations may be omitted in additional embodiments.

[0034] For purposes of this disclosure, the phrases "A and / or B" and "A or B" mean (A), (B), or (A and B). For purposes of this disclosure, the phrases "A, B, and / or C" and "A, B, or C" mean (A), (B), (C), (A and B), (A and C), (B and C), or (A, B, and C). Although some elements may be referred to in the singular (e.g., "processing device"), any appropriate element may be represented by multiple instances of the element, and vice versa. For example, a set of operations described as being performed by a processing device may be implemented by different ones of the operations being performed by different processing devices.

[0035] This specification uses the phrases "embodiment," "various embodiments," and "some embodiments," each of which may refer to one or more of the same or different embodiments. In addition, the terms "including," "comprising," "having," and the like as used with respect to the embodiments of the present disclosure are synonymous. When used to describe a range of sizes, the phrase "between X and Y" means a range that includes X and Y. As used herein, "device" may refer to any individual device, collection of devices, part of a device, or collection of parts of a device. The drawings are not necessarily drawn to scale.

[0036] Figure 1 1 is a diagram of an example of an inductively coupled plasma optical emission spectrometer (ICP-OES) 1000 or obtaining an optical emission spectrum according to various embodiments. The ICP-OES 1000 can be a component in a scientific support system 1200, for example, as described below with reference to Fig.12As described. An emission spectrometer operates by exciting atoms and ions to emit electromagnetic (EM) radiation at wavelengths characteristic of a particular element. As electrons transition from a high energy state to a lower energy state, a spectrum containing EM radiation of various frequencies is emitted. ICP-OES1000 may include a plasma chamber 1002. Plasma chamber 1002 may be connected to a radio frequency (RF) source and a gas (e.g., argon) source. Argon may be ionized within an oscillating RF field generated by an RF source to form and maintain a plasma within plasma chamber 1002. ICP-OES1000 may be based on a mid-step optical design to produce a two-dimensional optical emission spectrum.

[0037] Specifically, the ICP-OES may include an echelle diffraction grating 1014, a prism 1012, and a plurality of focusing mirrors 1004, 1006, 1010, 1016. These components together provide an optical system 1024. Light from the plasma chamber 1002 enters the ICP-OES 1000 and is selectively focused using a plurality of focusing mirrors (e.g., a first mirror 1004 and a second mirror 1006). The focused light may pass through an entrance slit 1008 and enter the prism 1012 using a mirror 1010. The prism 1012 may separate the light by wavelength. The echelle diffraction grating 1014 may diffract the separated light from the prism 1012 into a plurality of diffraction orders, producing a high-resolution 2D spectrum called an echelle spectrum diagram, also referred to as a full frame or echelle spectrum. After passing through these optical elements, the mirror 1016 may collect the spectrum and focus it onto a detector (e.g., a camera 1018). The optical system is housed in a housing 1026. One or more heating pads 1022 may be located on an outer surface of the housing 1026 to allow the temperature of the optical system 1024 to be controlled and maintained stable at a baseline operating temperature by heating the optical system 1024 and / or its environment within the housing 1026. The temperature sensor 1020 can indicate the temperature of the environment of, for example, an optical emission spectrometer. In one embodiment, the temperature sensor 1020 can indicate the temperature of an optical emission spectrometer or an optical system of an optical emission spectrometer including optical components. Advantageously, the temperature sensor 1020 can be disposed on a mechanical structure supporting one or more optical components to better capture temperature changes that affect the configuration of the optical components and, therefore, the alignment of the spectrum. A second temperature sensor can be positioned away from the optical system 1024, but in some embodiments, can still be located inside the instrument. The second temperature sensor can provide ambient temperature measurements. Reference above Figure 1The number, identity, and location of the components described are used illustratively, and it should be understood that other configurations and / or optical components may also be used. Although embodiments are described with reference to an inductively coupled plasma emission spectrometer, these embodiments are equally applicable to other types of spectra obtained using other types of spectrometers. For example, the methods described herein can be used for similar spectroscopic techniques, such as inductively coupled plasma mass spectrometry (ICP-MS) and atomic absorption spectroscopy (AAS), and can be applied to virtually any method of obtaining a spectrum, where the spectrum depends on the operating conditions of the spectrometer. Various embodiments include the use of spectroscopic techniques, such as Rayleigh spectroscopy, Raman spectroscopy, or mass spectrometry.

[0038] Figure 2 is a method for performing a spectrometer (such as a reference Figure 2 1 and 1 . A block diagram of a scientific instrument support module 2000 for performing support operations for a spectrometer described herein (e.g., a spectrometer) may be implemented by circuitry (e.g., including electrical and / or optical components) such as a programmed computing device. The logic of the scientific instrument support module 2000 may be included in a single computing device or may be distributed across multiple computing devices that communicate with each other as appropriate. Fig.11 The computing device 11000 of the present invention discusses examples of computing devices that can implement the scientific instrument support module 2000 individually or in combination, and the present invention is referred to herein. Fig.12 The scientific instrument support system 12000 discusses an example of a system of interconnected computing devices, wherein the scientific instrument support module 2000 may be implemented across one or more of these computing devices.

[0039] The scientific instrument support module 2000 may include data acquisition logic 2002, image registration logic 2004, training logic 2006, and inference logic 2008. As used herein, the term "logic component" may include a device for performing a set of operations associated with the logic component. For example, any of the logic elements included in the support module 2000 may be implemented by one or more computing devices, which are programmed with instructions to cause one or more processing devices of the computing device to perform an associated set of operations. In a specific embodiment, the logic element may include one or more non-transient computer-readable media, which have instructions on the one or more non-transient computer-readable media, which, when executed by one or more processing devices in one or more computing devices, cause the one or more computing devices to perform an associated set of operations. As used herein, the term "module" may refer to a collection of one or more logic elements, which together perform functions associated with the module. Different logic elements in the logic elements in the module may take the same form or may take different forms. For example, some logic components in the module may be implemented by a programmed general-purpose processing device, while other logic components in the module may be implemented by an application-specific integrated circuit (ASIC). As another example, different ones of the logic elements in a module may be associated with different sets of instructions executed by one or more processing devices. A module may not include all of the logic elements depicted in the associated figures; for example, when the module is to perform a subset of the operations discussed herein with reference to the module, the module may include a subset of the logic elements depicted in the associated figures. A particular arrangement of logic elements and modules is described, but it should be understood that the corresponding functions may be implemented and divided in many other ways.

[0040] Data acquisition logic 2002 may be configured to control ICP-OES 1000 to acquire data for use, for example, by training logic 2006 or inference logic 2008. Data acquisition logic 2002 may be configured to control ICP-OES 1000 to change operating conditions of ICP-OES 1000, for example, by controlling heating pad 1022.

[0041] For example, the data acquisition logic 2002 can control the operating conditions of the ICP-OES to correspond to specific predetermined baseline operating conditions for obtaining a baseline spectrum. For example, in order to obtain a baseline spectrum, the data acquisition logic 2002 can be configured to maintain the operating conditions of the optical components at a stable temperature of 38 degrees Celsius for a predetermined amount of time. For example, in one embodiment, the data acquisition logic 2002 can be configured to maintain the operating conditions of the optical components at a stable temperature for 30 minutes before obtaining the baseline spectrum. In some embodiments, multiple spectra can be acquired within a predetermined amount of time, such as at a rate of one spectrum per second. The data acquisition logic can compare one or more of the spectra acquired under the baseline operating conditions to confirm that there is no drift under the baseline operating conditions or to confirm that there is a negligible amount of drift (e.g., 0.5 pixels within 30 minutes) under the baseline operating conditions. After confirming that there is no drift or a negligible amount of drift, the data acquisition logic 2002 can use any one of the spectra acquired under the baseline operating conditions as the baseline spectrum. The data acquisition logic 2002 can be configured to change the operating conditions to multiple specific predetermined operating conditions for obtaining training data, as described below with reference to Figure 6 Detailed description. For example, the data acquisition logic 2002 can change the operating conditions, for example, by using the heating pad 1022 to cycle through multiple temperature settings while periodically acquiring spectra (e.g., one spectrum per second). The data acquisition logic can simultaneously or sequentially change the RF power supplied to the plasma chamber 1002 to change the operating conditions of the optical system, which can include setting the RF power to alternate between a high setting (e.g., 1500W) and a low setting (e.g., 1150W) within a predetermined time period (e.g., 2 minutes). Sequentially cycling the temperatures supplied by the heating pad 1022 and the RF generator can increase the thermal regulation of the optical system of the ICP-OES 1000, because plasma heat is one or the main factor that disrupts the thermal balance of the optical system.

[0042] The data acquisition logic 2002 may be configured to acquire and store spectral data and corresponding operating condition data. Figure 3 An example full frame spectrum or mid-step spectrum graph depicting the intensity distribution captured by the photodetector 1018 on the ICP-OES 1000 according to some embodiments. Multiple peaks are seen as white dots of varying intensity, such as Figure 3 As shown. Each of the multiple peaks may have a characteristic intensity and wavelength indicative of a particular atom or molecule. The intensity distribution captured by the photodetector 1018 may be captured and stored by the data acquisition logic 2002 in the form of a digital image, which includes a set of pixels having pixel values ​​reflecting the captured intensity distribution. The pixel value may be a grayscale scalar value or a vector value representing color information.

[0043] The data acquisition logic 2002 may be configured to perform checks after initializing the ICP-OES 1000, and may be further configured to apply known background correction techniques before acquiring spectra. The data acquisition logic 2002 may perform standard sample preparation related operations, standard data quality control operations, and any standard data acquisition operations, and may also control the introduction of samples into the spectrometer and the exposure time of the analytes to the plasma according to standard operating procedures based on the concentration of the analytes.

[0044] The image registration logic 2004 may be configured to register the acquired spectral image (also referred to as a spectral image) to generate transformation parameters for a transformation that aligns the acquired spectral image to the baseline spectral image. The image registration logic 2004 may be configured to retrieve an image of the baseline spectrum stored by the data acquisition logic 2002 and an image of the acquired spectrum stored by the data acquisition logic 2002 and perform image registration. In some embodiments, the image registration logic 2004 may be configured to perform image registration based on identified local points of the spectrum as landmarks (e.g., peaks of the spectrum). In these embodiments, when the corresponding transformation is applied to the acquired image, the transformation parameters are adjusted to minimize the positional error between the landmarks in the acquired spectrum and the baseline spectrum. In other embodiments, the image registration logic 2004 may be configured to perform a spectrum (such as a spectrum) based on, for example, comparing the image intensities of all pixels or a set of control points between the acquired spectrum and the baseline spectrum. Figure 3 Image registration of the entire image (full frame spectrum shown in ).

[0045] The image registration logic 2004 may thus determine the transformation parameters of the transformation to align the acquired optical emission spectrum to the baseline optical emission spectrum. The transformation may be an affine transformation, such as a combination of a translation operation and a rotation operation applied globally to the acquired image, or the transformation may be non-affine and apply a local transformation, for example, through a distortion field or a warp field known in the field of image registration. The image registration logic 2004 may perform image registration using any suitable known method. For example, in one embodiment, the image registration logic 2004 may include an adversarial similarity network [Adversarial Similarity Network for Evaluating Image Alignment in Deep Learning Based Registration Jingfan Fan, author Jingfan Fan, Xiaohuan Cao, ZhongXue, Pew-Thian Yap and Dinggang Shen, published in Med Image Comput Comput Assist Interv. 2018 September; 11070:739-746.doi:10.1007 / 978-3-030-00928-1_83, published online on September 26, 2018, PMID: 30627709; PMCID: PMC6322551, incorporated herein by reference], which is known to be used for estimating image warping / shifting for medical purposes.

[0046] Figure 4 An example of a warp field used to register an acquired spectrum to a baseline spectrum is depicted in FIG. The warp field defines a local transformation indicated by the arrows that aligns the local intensity distribution in the corresponding image and thus illustrates not only a specific type of transformation used in some embodiments, but also distortions that may occur due to varying operating conditions. Typically, such distortions occur due to varying operating conditions, which may be indicative of, or correlated to, the temperature of the optical system.

[0047] The training logic 2006 may be configured to perform the training process 7000 using the training data acquired by the data acquisition logic 2004 to generate a machine learning model that generates transformation parameters in response to input of parameters describing operating conditions of the optical emission spectrometer 1002, as described below with reference to Figure 7 Described in detail.

[0048] The inference logic 2008 may be configured to perform the inference process 5000 to generate transformation parameters and apply a corresponding transformation to the acquired spectrum to mitigate operating condition-related distortions in the acquired spectrum, as described below with reference to Figure 5Described in detail.

[0049] Figure 5 2 is a flow chart of an inference method 5000 for mitigating distortion of an optical emission spectrum according to various embodiments. The steps of the process may be performed, for example, by inference logic 2008. Although reference may be made to specific embodiments disclosed herein (e.g., Figure 2 The Scientific Instrument Support Module 2000 discussed in this article is referenced Fig.10 The graphical user interface (GUI) 10000 discussed in this article refers to Fig.11 The computing device 11000 discussed herein and / or referenced herein Fig.12 The operation of method 5000 is illustrated using the scientific instrument support system 12000 discussed above, but method 5000 can be used in any suitable setting to perform any suitable support operation. Figure 5 The operations are each illustrated once in a particular order, but the operations may be reordered and / or repeated as needed and appropriate (eg, the operations may be performed in parallel where appropriate).

[0050] At 5002, a first operation may be performed. The first operation may include obtaining an optical emission spectrum using a spectrometer, the spectrometer including an optical system for forming the optical emission spectrum. The spectrometer may be the spectrometer described above with reference to Figure 1 The spectrum can be, for example, a full-frame, such as Figure 3 As shown. The spectrum can be obtained when the spectrometer is operating under stable operating conditions (e.g., baseline operating conditions) or at other times. For example, the disclosed method enables the spectrum to be usefully obtained when the spectrometer is warmed up to reach the baseline operating condition. This is because the disclosed method allows the correction or at least mitigation of operating condition-related distortions, such as thermal distortions, when the spectrometer reaches the baseline operating condition after being turned on. Therefore, the spectrum can become available more quickly and the consumption of electricity and other consumables can be reduced.

[0051] At 5004, a second operation may be performed. The second operation may include obtaining one or more condition parameters, which indicate the operating conditions of the optical emission spectrometer or the environment of the optical emission spectrometer when recording the optical emission spectrum and / or at a previous time. For example, in some embodiments, the one or more condition parameters may indicate the operating conditions of the optical system, which may indicate the temperature of the optical system or be related to the temperature of the optical system. The condition parameters may include one or more of the optical component temperature, RF voltage, RF current, optical component heater voltage, and ambient temperature board back readings. In some embodiments, the one or more condition parameters may be a time series of one or more condition parameters at each time point in a plurality of time points (e.g., a plurality of time points up to and including 15 minutes of the time of obtaining the optical emission spectrum). In some embodiments, in the case of obtaining more than one condition parameter, these condition parameters may be obtained over different time spans. For example, the input data may include one or more condition parameters at each time point in a plurality of time points up to and including 15 minutes of the time of obtaining the optical emission spectrum and a time series of one or more other condition parameters at each time point in a plurality of time points up to and including 5 minutes of the time of obtaining the optical emission spectrum.

[0052] At 5006, a third operation may be performed. The third operation may include providing the obtained one or more conditional parameters as input to a machine learning model. For example, in some embodiments, a machine learning model using an XGBoost algorithm may be used. In some embodiments, the machine learning model includes a feedforward neural network. In embodiments where the input data includes a time series, one or more conditional parameters at each of a plurality of time points may be applied together as input to the feedforward neural network. In some embodiments, the machine learning model may include a recursive neural network, such as an LSTM unit. In some embodiments, the machine learning model may include a transformer model. For example, the machine learning model may have been based on the following reference Figure 7 Describe the process for training.

[0053] At 5008, a fourth operation may be performed. The fourth operation may be obtaining one or more transformation parameters as an output of the machine learning model.

[0054] At 5010, a fifth operation may be performed. The fifth operation may be applying a transformation according to one or more transformation parameters to the obtained optical emission spectrum to mitigate distortion of the optical emission spectrum. As discussed above, mitigating distortion due to changing operating conditions allows for more rapid collection of useful spectra without having to wait for operating conditions to stabilize.

[0055] Figure 66 is a flow chart of an example data collection method 6000 for obtaining training data for training a machine learning model used in the inference method 5000 according to various embodiments. Unlike a conventional operating mode (where the system is maintained at a stable operating condition within a constrained range prior to data collection), training data is obtained under a wide range of operating conditions. The steps of this process may be performed, for example, by the data acquisition logic 2002 and the image registration logic 2004. Although reference may be made to specific embodiments disclosed herein (e.g., references herein to Figure 2 The Scientific Instrument Support Module 2000 discussed in this article is referenced Fig.10 GUI10000 discussed in this article Fig.11 The computing device 11000 discussed herein and / or referenced herein Fig.12 The operation of the data collection method 6000 is illustrated using the scientific instrument support system 12000 discussed above, but the method 6000 can be used in any suitable setting to perform any suitable support operation. Figure 6 The operations are each illustrated once in a particular order, but the operations may be reordered and / or repeated as needed and appropriate (eg, different operations may be performed in parallel where appropriate).

[0056] At 6002, a first operation may be performed. The first operation may include providing a spectrometer including an optical system for forming an optical emission spectrum of a reference analyte. The spectrometer may be the reference analyte Figure 1 Description of the ICP-OES1000.

[0057] At 6004, a second operation may be performed. The second operation may include recording an optical emission spectrum of a reference analyte for an operating condition of the optical emission spectrometer or an environment of the optical emission spectrometer. For example, in one embodiment, the second operation may include recording an operating condition of an optical system of the optical emission spectrometer.

[0058] At 6006, a third operation may be performed. The third operation may include storing one or more parameters indicative of an operating condition of the optical system under which the optical emission spectrum was recorded as input data for the machine learning model.

[0059] At 6008, a fourth operation may be performed. The fourth operation may include adjusting one or more transformation parameters of the transformation to align the optical emission spectrum to the baseline optical emission spectrum using the transformation, as described above with reference to the image registration logic 2004. When the spectrometer is in a stable baseline operating condition, the baseline optical emission spectrum may be recorded for the baseline operating condition, for example, by the data acquisition logic 2002. For example, in some embodiments, the baseline optical emission spectrum may be recorded on the same spectrometer as the optical emission spectrum to be aligned. In some embodiments, the baseline optical emission spectrum may be an industry standard baseline spectrum obtained from a third party.

[0060] At 6010, a fifth operation may be performed. The fifth operation may include storing the adjusted one or more transformation parameters associated with the corresponding input data of the optical emission spectrum (as a training data pair) as output data of the machine learning model.

[0061] After 6010, method 6000 may loop back to 6004 and perform operations 6004 to 6010 under different corresponding operating conditions for each loop, thereby recording training data pairs for different operating conditions. For example, multiple spectra for multiple corresponding operating conditions may be obtained using data acquisition logic 2002. Alternatively, operations 6004 to 6008 may be performed as a series of loops, each time under different corresponding operating conditions, and then step 6010 is performed in batches for all spectra to form training data pairs. In either case, these loops may all be performed on the same spectrometer, or these loops may be performed using different spectrometers, such as obtaining training data pairs for each operating condition in a plurality of operating conditions from each of a plurality of spectrometers of the same type (e.g., the same brand and model).

[0062] Figure 7 5000 is a flowchart of an example training method for training a machine learning model used in inference method 5000 according to various embodiments. Any suitable machine learning model can be used and trained according to various embodiments, for example, a machine learning model using the XGBoost algorithm can be used. The steps of the process can be performed, for example, by training logic 2006. Although reference may be made to specific embodiments disclosed herein (e.g., reference may be made to Figure 2 The Scientific Instrument Support Module 2000 discussed in this article is referenced Fig.10 GUI 10000 discussed, this article references Fig.11 The computing device 11000 discussed herein and / or referenced herein Fig.12 The operation of method 5000 is illustrated using the scientific instrument support system 12000 discussed above, but method 5000 can be used in any suitable setting to perform any suitable support operation. Figure 7The operations are each illustrated once in a particular order, but the operations may be reordered and / or repeated as needed and appropriate (eg, different operations may be performed in parallel where appropriate).

[0063] At 7002, a first operation may be performed. The first operation may include obtaining a training data set having training data pairs of input data and output data, for example, as described above with reference to Figure 6 and data acquisition method 6000 as described.

[0064] At 7004, a second operation may be performed. The second operation may include applying the training input to the machine learning model to generate an output of the machine learning model.

[0065] At 7006, a third operation may be performed. The third operation may include adjusting the parameters of the machine learning model to reduce the difference between the generated output and the training output of the training pair. Operations 7002 to 7006 are performed on all training data pairs in the training data set, and the parameters may be adjusted in multiple passes and rounds of the training data set, such as until a stopping criterion indicating a satisfactory output is met. For example, the stopping criterion may be a test error between the actual output and the target output evaluated on a test data set separate from the training data set. In some embodiments, the test error is evaluated using n-fold, such as 5-fold cross validation.

[0066] Figure 8 An example of input data 8002 provided to a machine learning model 8004 and an example of output data 8006 provided by the model being trained are shown according to various embodiments. The input data 8002 can be obtained according to the data acquisition method 6000 in some embodiments, for example, using the data acquisition logic 2002. The output data 8004 can be obtained, for example, using the image registration logic 2004. Figure 8 The number and types of parameters depicted in are purely illustrative, and any appropriate number or type of parameters may be included as input data 8002 and output data 8006.

[0067] Input data 8002 indicates one or more parameters of the operating conditions of the optical system of the spectrometer used to record the spectrum. For example, in some embodiments, the one or more parameters may indicate the temperature of the optical system or be related to the temperature of the optical system. Input data 8002 may include, for example, a sample index indicating the corresponding full frame, an optical temperature back reading indicating the temperature of the optical system at the time of acquisition, and an ambient temperature measurement result obtained by a sensor placed on an instrument control board away from the optical box but still inside the instrument. In some embodiments, the sample index may be used only for data management and may not be input to the model. Input data may also include an RF power supply (voltage), which together with the RF power supply current gives a direct estimate of the power supplied to the plasma and its equivalent temperature. The plasma temperature strongly affects the temperature of the optical system and can therefore provide further useful information to the machine learning model. Input data may also include an optics heater voltage indicating a voltage value set by a thermal stabilization routine, which evaluates the absolute value and change rate of the optical temperature back reading to control the temperature of the optical system and thus provides an indication of the rate of change of the temperature of the optical system. Input data may also include an acquisition time, which contains information about the rate of temperature change and temperature information. In some embodiments, the acquisition time and temperature information can be used to calculate the rate of change of the temperature of the optical system as a further input to the machine learning model. In some embodiments, the input data includes a time series of condition parameters, i.e., several samples of the condition parameters at corresponding different acquisition times applied to the input, as described below.

[0068] The machine learning model 8004 can be any suitable model trained to provide output data 8006 in response to input data 8002. In order to process the input data as a time series, in some embodiments, the machine learning model may include a recursive neural network or a transformer model. In some embodiments, the machine learning model may include an XGBoost algorithm. In such embodiments, the model may be trained to provide one or more transformation parameters to be applied to the transformation of the obtained spectrum in response to the time series of one or more conditional parameters. In some embodiments, the machine learning model may include a feedforward neural network. In some embodiments, the feedforward neural network is suitable for processing time series input data by providing an input unit of the conditional parameters at each time step of the time series. In such embodiments, the sample number and / or timestamp may not be included in the input data because the identity of each input unit indicates the corresponding time step (and parameter).

[0069] Output data 8006 indicates one or more transformation parameters of a transformation defined according to the transformation parameters for aligning the optical emission spectrum to the baseline optical emission spectrum, as described above. In embodiments where the input to the machine learning model includes a time series of conditional parameters, the output includes a corresponding time series of transformation parameters for the corresponding transformation of the corresponding spectrum at each time step.

[0070] The scientific instrument support methods disclosed herein may include interaction with a human user (e.g., via the following reference Fig.12 Fig.12 These interactions may include providing information to the user (e.g., about scientific instruments such as Fig.12 information about the operation of a scientific instrument 12010), information about a sample being analyzed or other test or measurement performed by the scientific instrument, information retrieved from a local or remote database, or other information) or providing the user with the option of inputting commands (e.g., controlling a scientific instrument such as Fig.12 In some embodiments, these interactions may be performed through a graphical user interface (GUI) that includes a display device (e.g., a graphical user interface such as a display device ... Fig.11 A visual display on a display device 11010 discussed herein) that provides output to a user and / or prompts a user to provide input (e.g., via a display device 11010 discussed herein). Fig.11 Other I / O devices 11012 discussed herein include one or more input devices such as a keyboard, mouse, trackpad, or touch screen). The scientific instrument support system disclosed herein may include any suitable GUI for interacting with a user.

[0071] Fig.10 An example GUI 10000 that can be used to perform some or all of the supporting methods disclosed herein is depicted according to various embodiments. As described above, the GUI 10000 can be provided in a scientific instrument support system (e.g., as described herein with reference to Fig.12 A computing device (e.g., a scientific instrument support system 12000 as discussed herein) Fig.11 The computing device 11000 discussed herein) may include a display device (eg, Fig.11 The user may use any suitable input device (e.g., the display device 11010 discussed herein) to input the content of the display device 11010. Fig.11The GUI 10000 may include any of the input devices included in the other I / O devices 11012 discussed) and input techniques (e.g., cursor movement, motion capture, facial recognition, gesture detection, voice recognition, button actuation, etc.) to interact with the GUI 10000. The GUI 10000 may include a data display area 10002, a data analysis area 10004, a scientific instrument control area 10006, and a settings area 10008. Fig.10 The specific number and arrangement of regions depicted in the GUI 10000 are exemplary only, and any number and arrangement of regions (including any desired features) may be included in the GUI 10000. The data display area 10002 may display data generated by a scientific instrument (e.g., a device described herein). Fig.12 Fig.12 For example, the data display area 10002 may display one or more optical emission spectra obtained from an optical emission spectrometer, such as Figure 3 The data display area 10002 may further display one or more operating conditions of the optical system of the spectrometer 1000. The data analysis area 10004 may display the results of the data analysis (eg, the results of analyzing the data shown in the data display area 10002 and / or other data).

[0072] For example, the data analysis region 10004 can display one or more transformation parameters of a transformation to align the optical emission spectrum to a baseline optical emission spectrum using the transformation. In some embodiments, the data display region 10002 and the data analysis region 10004 can be combined in the GUI 10000 (e.g., to include data output from a scientific instrument and some analysis of the data in a common graph or region). The scientific instrument control region 10006 may include a user interface that allows the user to control the scientific instrument (e.g., as described herein with reference to FIG. 1 ). Fig.12 Fig.12 For example, the scientific instrument control region 10006 may include temperature control and RF control. The settings region 10008 may include options that allow the user to control the features and functions of the GUI 10000 (and / or other GUIs) and / or perform common computing operations with respect to the data display region 10002 and the data analysis region 10004 (e.g., saving data to a storage device (such as the one described herein). Fig.11 The settings area 10008 may include the option to save the spectrum as it is acquired or the option to set a certain temperature or RF power supply voltage.

[0073] Example Implementation

[0074] In the example implementation described below, training data was obtained from two different types of optical emission spectrometers: Model Thermo Scientific TM iCAP TM PRO ICP-OES and iCAP TM PRO X ICP-OES. For each optical emission spectrometer, a baseline spectrum of the following composition of reference analytes was obtained under baseline operating conditions: [Al: 1 mg / L, Ba: 0.2 mg / L, Ca: 0.2 mg / L, Cu: 1 mg / L, K: 5 mg / L, Mn: 1 mg / L, Ni: 5 mg / L, Pm: 10 mg / L, Zn: 0.2 mg / L, Mg: 0.2 mg / L, Others: 0.2% HNO3 mg / L]. The baseline operating conditions were maintained for 30 minutes before obtaining the baseline spectrum. The baseline operating conditions for obtaining the baseline spectrum included: an optical system temperature of 38 degrees Celsius and an RF power of 1150 W.

[0075] Condition parameters indicative of operating conditions were collected at a sampling frequency of 0.1 Hz.

[0076] The condition parameters include:

[0077] · Parameter indicating whether the user has pressed a button to start the plasma being generated (0, 1);

[0078] a parameter indicative of the temperature of the optical system (in degrees Celsius) obtained from a temperature sensor attached to a mechanical structure supporting one or more optical components of the optical system;

[0079] Parameter indicating the temperature change of the optical system (in degrees Celsius);

[0080] A parameter indicating the temperature (in degrees Celsius) surrounding the scientific instrument obtained from a temperature sensor attached to a mechanical structure remote from the optical system;

[0081] A parameter indicative of the temperature on the power supply connected to the RF generator of the plasma chamber 1002 (in degrees Celsius);

[0082] A parameter indicating the temperature of the printed circuit board that controls the heating pad of the optical system (in degrees Celsius);

[0083] A parameter indicating the voltage on the power supply of the RF generator connected to the plasma chamber 1002 (V);

[0084] A parameter indicating the current on the power supply of the RF generator connected to the plasma chamber 1002 (A);

[0085] A parameter indicating the voltage applied to the optical system heating pad (V);

[0086] · Parameter indicating the pressure in the exhaust pipe for pumping the discharged plasma gas (mBar);

[0087] A parameter indicating the temperature of the exhaust pipe used to pump the discharged plasma gas (in degrees Celsius);

[0088] Parameter indicating the discharge operation (1 / s);

[0089] · Parameter indicating the current at the control board of the spectrometer (A);

[0090] ·Parameter indicating the current at the camera's PCB (A);

[0091] · Parameter indicating the current at the circuit board of the optical system (including heating pad, shutter, slit) (A);

[0092] Training data is obtained by modifying the operating conditions of the optical emission spectrometer when obtaining the spectrum. Training data is obtained from the corresponding optical emission spectrometer within 10 hours of opening 15 optical emission spectrometers. Test data and verification data are obtained from different optical emission spectrometers to provide an unbiased evaluation of model performance. Verification data is obtained from another 3 optical emission spectrometers, and test data is obtained from another 4 optical emission spectrometers. By setting the RF power source connected to the plasma chamber 1002 to high power or low power, a spectrum is obtained under high power conditions and low power conditions. This is achieved using two methods. The first method is achieved by running the entire experiment under high power (1500W) or low power (1150W) settings. The second method is achieved by setting the RF power source connected to the plasma chamber 1002 to alternate between 1500W and 1150W settings. The cycle rate between the RF power is different between the training samples, and includes, for example, alternating between high power and low power every 2 minutes. Spectra are obtained at a sampling frequency of 1 spectrum / minute. The one or more transformation parameters are provided to the machine learning model. In the example implementation, the XGBoost model is used, however, it should be understood that any appropriate machine learning model, such as a recursive neural network or a transformer model, can be used. One or more transformation parameters of each transformation are provided to the XGBoost model as the target variable for training the XGBoost model. For each transformation, the input variables are included in the corresponding time series of the above-mentioned conditional parameters at each time point of 15 minutes until and including the time of obtaining the spectrum corresponding to the transformation. By training the model to minimize the mean square error, the XGBoost model is fitted to the training data using the input variables and the output variables.

[0093] The hyperparameters of the XGBoost model are adjusted to maximize the model performance on the validation dataset. During hyperparameter tuning, models are selected based on their performance on the validation dataset, specifically the model with the lowest mean absolute error in predictions between 0.5 and 10 hours after the instrument start time. In the algorithm file ( https: / / xgboost.readthedocs.io / en / stable / parameter.html ) is described in detail in XGBoost model hyperparameters. Fig. 9A and Fig. 9B The performance of the trained model on the test data of three optical emission spectrometers (iCAPPRO60465 (test), iCAPPRO60474 (test) and iCAPPRO60468 (test)) is shown. Fig. 9A Shown is the prediction of lateral drift on the 3 instruments from which the test dataset was obtained by the XGBoost model. Left: actual drift (grey) and predicted drift (black) Right: prediction error (difference). Fig. 9B Shown are the predictions of vertical drift on the 3 instruments from which the test data sets were obtained by the XGBoost model. Left: Actual drift (grey) and predicted drift (black) Right: Prediction error (difference). The low frequency waves showing the largest changes in drift are due to temperature cycling of the heating pads, which is consistent with the understanding that the drift is due to thermal deformation of components of the optical system. The high frequency waves showing smaller drift changes are due to rapid (e.g. every 1 to 2 minutes) changes in the RF supplied to the plasma.

[0094] Exploratory testing of the dataset

[0095] The separate condition parameters used to train the model allow the model to predict different degrees of drift. To explore this, spectra were acquired at a sampling frequency of 1 spectrum / minute, and condition parameters indicative of operating conditions were collected at a sampling frequency of 1 spectrum every 6 seconds, and each of these condition parameters was used individually to train a separate machine learning model. The mean squared error for each trained model was calculated using the following formula: Where n is the number of spectra (the number of data points). For each condition parameter. One or more models are trained, each model using data collected at different amounts of time before obtaining the spectrum. Table 1 shows the results of this exploratory test. As noted above, the mean square error indicates the deviation between the predicted drift and the true drift predicted. In the exploratory test captured in Table 1, the drift in the x direction (lateral drift) is used to obtain the mean square error. The rolling history window indicates the amount of time that the mean square error of the model is the lowest before obtaining the spectrum for each variable. For the temperature of the optical system, the temperature change of the optical system, the current at the optical system circuit board, and the voltage on the heating pad, the best fit using individual parameters is obtained. This is consistent with the understanding that drift may be caused by thermal deformation of components of the optical system. Another driving factor for temperature changes in the optical system is the heat emitted by the plasma located near the optical system. The temperature at the RF power supply close to and therefore related to the plasma temperature provides an intermediate fit as a separate parameter. It has been found that other parameters are poor individual predictors of spectral distortion.

[0096]

[0097] Table 1: Results of exploratory testing showing the mean squared error of models predicting spectral drift trained using each conditional parameter alone

[0098] Fig. 9C Shown is a comparison between the actual lateral drift (first row) and the predicted lateral drift (subsequent rows) resulting from exploratory testing by training a machine learning model using only one conditional parameter and then predicting the drift in two different spectrometers (iCAPPRO60614 and iCAPPRO60485). Fig. 9C It is shown that the drift predicted by the model trained using only temperature or only current at the optical system circuit board is closer to the true drift than the exhaust pressure (the pressure in the exhaust pipe used to pump the exhausted plasma gas) or the ambient temperature (the temperature around the scientific instrument). Fig. 9A and Fig. 9B When comparing, Fig. 9C It also shows that the current at the optics circuit board performs well in predicting the drift caused by fluctuations in the RF supplied to the plasma. This is consistent with the proximity of the plasma to the optics, which results in a correlation between the current supplied to the heater pads and the RF supplied to the plasma to keep the optics at a stable temperature.

[0099] Computing Infrastructure

[0100] As noted above, the scientific instrument support module 2000 may be implemented by one or more computing devices. Fig.11is a block diagram of a computing device 11000 that can perform some or all of the scientific instrument support methods disclosed herein, according to various embodiments. In some embodiments, the scientific instrument support module 2000 can be implemented by a single computing device 11000 or by multiple computing devices 11000. In addition, as discussed below, the computing device 1100 (or multiple computing devices 11000) implementing the scientific instrument support module 1000 can be Fig.12 Part of one or more of the scientific instrument 12010, the user local computing device 12020, the service local computing device 12010, or the remote computing device 12040.

[0101] Fig.11 The computing device 11000 is illustrated as having multiple components, but any one or more of these components may be omitted or repeated depending on suitability for the application and setting. In some embodiments, some or all of the components included in the computing device 11000 may be attached to one or more motherboards and encapsulated in a housing (e.g., comprising plastic, metal, and / or other materials). In some embodiments, some of these components may be manufactured on a single system on a chip (SoC) (e.g., the SoC may include one or more processing devices 11002 and one or more storage devices 11004). In some embodiments, one or more of these components may be remote from the computing device 11000. Additionally, in various embodiments, the computing device 11000 may not include Fig.11 One or more of the illustrated components may include interface circuitry (not shown) for coupling to one or more components using any suitable interface (e.g., a universal serial bus (USB) interface, a high-definition multimedia interface (HDMI) interface, a controller area network (CAN) interface, a serial peripheral interface (SPI) interface, an Ethernet interface, a wireless interface, or any other suitable interface). For example, computing device 11000 may not include display device 11010, but may include display device interface circuitry (e.g., a connector and driver circuitry) to which display device 11010 may be coupled.

[0102] The computing device 11000 may include a processing device 11002 (e.g., one or more processing devices). As used herein, the term "processing device" may refer to any device or portion of a device that processes electronic data from registers and / or memory to convert the electronic data into other electronic data that can be stored in registers and / or memory. The processing device 11002 may include one or more digital signal processors (DSPs), application specific integrated circuits (ASICs), central processing units (CPUs), graphics processing units (GPUs), cryptographic processors (specialized processors that execute cryptographic algorithms in hardware), server processors, or any other suitable processing devices.

[0103] The computing device 11000 may include a storage device 11004 (e.g., one or more storage devices). The storage device 11004 may include one or more memory devices, such as a random access memory (RAM) (e.g., a static RAM (SRAM) device, a magnetic RAM (MRAM) device, a dynamic RAM (DRAM) device, a resistive RAM (RRAM) device, or a conductive bridge RAM (CBRAM) device), a hard drive-based memory device, a solid-state memory device, a network drive, a cloud drive, or any combination of memory devices. In some embodiments, the storage device 11004 may include a memory that shares a die with the processing device 11002. In such an embodiment, the memory may be used as a cache memory and may include, for example, an embedded dynamic random access memory (eDRAM) or a spin transfer torque magnetic random access memory (STT-MRAM). In some embodiments, the storage device 11004 may include a non-transitory computer-readable medium having instructions thereon that, when executed by one or more processing devices (e.g., processing device 11002), cause the computing device 11000 to perform any appropriate method or portion of the methods disclosed herein.

[0104] The computing device 11000 may include an interface device 11006 (e.g., one or more interface devices 11006). The interface device 11006 may include one or more communication chips, connectors, and / or other hardware and software to manage communications between the computing device 11000 and other computing devices. For example, the interface device 11006 may include circuits for managing wireless communications for transmitting data to and from the computing device 11000. The term "wireless" and its derivatives may be used to describe circuits, devices, systems, methods, techniques, communication channels, etc. that can communicate data through a non-solid medium using modulated electromagnetic radiation. The term does not imply that the associated device does not contain any wires, although in some embodiments it may not contain any wires. The circuitry included in the interface device 11006 for managing wireless communications may implement any of a plurality of wireless standards or protocols, including but not limited to Institute of Electrical and Electronics Engineers (IEEE) standards, including Wi-Fi (IEEE 802.11 series), IEEE 802.16 standards (e.g., IEEE 802.16-2005 amendments), Long Term Evolution (LTE) projects, and any amendments, updates, and / or revisions (e.g., Advanced LTE projects, Ultra Mobile Broadband (UMB) projects (also known as "3GPP2"), etc.). In some embodiments, the circuitry included in the interface device 11006 for managing wireless communications may operate according to a Global System for Mobile Communications (GSM), General Packet Radio Service (GPRS), Universal Mobile Telecommunications System (UMTS), High Speed ​​Packet Access (HSPA), Evolved HSPA (E-HSPA), or LTE network. In some embodiments, the circuitry included in the interface device 11006 for managing wireless communications may operate according to Enhanced Data GSM Evolution (EDGE), GSM Evolution Radio Access Network (GERAN), Universal Terrestrial Radio Access Network (UTRAN), or Evolved UTRAN (E-UTRAN). In some embodiments, the circuitry included in the interface device 11006 for managing wireless communications may operate according to Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Digital Enhanced Cordless Telecommunications (DECT), Evolution Data Optimized (EV-DO) and their derivatives, as well as any other wireless protocols designated as 3G, 4G, 5G and higher. In some embodiments, the interface device 11006 may include one or more antennas (e.g., one or more antenna arrays) to receive and / or send wireless communications.

[0105] In some embodiments, interface device 11006 may include a circuit for managing wired communication, such as an electrical communication protocol, an optical communication protocol, or any other suitable communication protocol. For example, interface device 11006 may include a circuit supporting communication according to Ethernet technology. In some embodiments, interface device 11006 may support both wireless and wired communication, and / or may support multiple wired communication protocols and / or multiple wireless communication protocols. For example, a first group of circuits of interface device 11006 may be dedicated to shorter distance wireless communication such as Wi-Fi or Bluetooth, and a second group of circuits of interface device 11006 may be dedicated to longer distance wireless communication such as global positioning system (GPS), EDGE, GPRS, CDMA, WiMAX, LTE, EV-DO, etc. In some embodiments, a first group of circuits of interface device 11006 may be dedicated to wireless communication, and a second group of circuits of interface device 11006 may be dedicated to wired communication.

[0106] Computing device 11000 may include battery / power circuitry 11008. Battery / power circuitry 11008 may include one or more energy storage devices (e.g., batteries or capacitors) and / or circuitry for coupling components of computing device 11000 to an energy source separate from computing device 11000 (e.g., AC line power).

[0107] Computing device 11000 may include a display device 11010 (eg, multiple display devices). Display device 11010 may include any visual indicator, such as a heads-up display, a computer monitor, a projector, a touch screen display, a liquid crystal display (LCD), a light emitting diode display, or a flat panel display.

[0108] The computing device 11000 may include other input / output (I / O) devices 11012. The other I / O devices 11012 may include, for example, one or more audio output devices (e.g., speakers, headphones, earbuds, alarms, etc.), one or more audio input devices (e.g., microphones or microphone arrays), positioning devices (e.g., GPS devices that communicate with satellite-based systems to receive the location of the computing device 11000 as is known in the art), audio codecs, video codecs, printers, sensors (e.g., thermocouples or other temperature sensors, humidity sensors, pressure sensors, vibration sensors, accelerometers, gyroscopes, etc.), image capture devices such as cameras, keyboards, cursor control devices such as mice, stylus, trackballs, or touchpads, barcode readers, quick response (QR) code readers, or radio frequency identification (RFID) readers.

[0109] The computing device 11000 may have any suitable form factor suitable for its applications and settings, such as a handheld or mobile computing device (e.g., a cellular phone, a smart phone, a mobile Internet device, a tablet computer, a laptop computer, a netbook computer, an ultrabook computer, a personal digital assistant (PDA), an ultra-mobile personal computer, etc.), a desktop computing device, or a server computing device or other networked computing component.

[0110] One or more computing devices implementing any of the scientific instrument support modules or methods disclosed herein may be part of a scientific instrument support system. Fig.12 12000 is a block diagram of an example scientific instrument support system 12000 in which some or all of the scientific instrument support methods disclosed herein may be performed, according to various embodiments. The scientific instrument support modules and methods disclosed herein (e.g., Figure 2 The Scientific Instrument Support Module 1000 and Figure 5 Method 5000, Figure 6 Method 6000 and Figure 7 Method 7000) can be implemented by one or more of the scientific instrument 12010 of the scientific instrument support system 12000, the user local computing device 12020, the service local computing device 12030 or the remote computing device 12040.

[0111] Any of the scientific instrument 12010, the user local computing device 12020, the service local computing device 12030, or the remote computing device 12040 may include the references herein Fig.11 Any of the embodiments of the computing device 11000 discussed herein, and any of the scientific instrument 12010, the user local computing device 12020, the service local computing device 12030, or the remote computing device 12040 may adopt the embodiments described herein. Fig.11 Any suitable form of implementation of the computing device 11000 discussed herein.

[0112] The scientific instrument 12010, the user local computing device 12020, the service local computing device 12030, or the remote computing device 12040 may each include a processing device 12002, a storage device 12004, and an interface device 12006. The processing device 12002 may take any suitable form, including any suitable form described herein. Fig.11 The processing device 11002 discussed above may be in the form of any processing device, and the processing device 12002 included in different devices of the scientific instrument 12010, the user local computing device 12020, the service local computing device 12030, or the remote computing device 12040 may take the same form or different forms. The storage device 12004 may take any suitable form, including those described herein. Fig.11The storage device 11004 discussed above may be in the form of any storage device, and the storage device 12004 included in different devices of the scientific instrument 12010, the user local computing device 12020, the service local computing device 12030, or the remote computing device 12040 may take the same form or different forms. The interface device 12006 may take any suitable form, including those described herein. Fig.11 The form of any of the interface devices 11006 discussed, and the interface devices 12006 included in different devices of the scientific instrument 12010, the user local computing device 12020, the service local computing device 12030, or the remote computing device 12040 may take the same form or different forms.

[0113] The scientific instrument 12010, the user local computing device 12020, the service local computing device 12030, and the remote computing device 12040 can communicate with other elements of the scientific instrument support system 12000 via a communication path 12008. The communication path 12008 can communicatively couple the interface device 12006 (as shown) of different elements of the scientific instrument support system 12000, and can be a wired or wireless communication path (e.g., according to the present reference Fig.11 any communication technology discussed in the communication technologies of the interface device 11006 of the computing device 11000). Fig.12 The specific scientific instrument support system 12000 depicted in FIG. 1 includes a communication path between each pair of devices in the scientific instrument 12010, the user local computing device 12020, the service local computing device 12030, and the remote computing device 12040, but this specific implementation of "fully connected" is merely illustrative, and in various embodiments, various communication paths in the communication paths 12008 may not exist. For example, in some embodiments, the service local computing device 12030 may not have a direct communication path 12008 between its interface device 12006 and the interface device 12006 of the scientific instrument 12010, but may communicate with the scientific instrument 12010 via the communication path 12008 between the service local computing device 12030 and the user local computing device 12020 and the communication path 12008 between the user local computing device 12020 and the scientific instrument 12010.

[0114] Scientific instrument 12010 may include any suitable scientific instrument, such as an optical emission spectrometer, for example Figure 1 An inductively coupled plasma optical emission spectrometer 1000 is shown.

[0115] The user-local computing device 12020 can be a computing device local to a user of the scientific instrument 12010 (e.g., according to any of the embodiments of the computing device 11000 discussed herein). In some embodiments, the user-local computing device 12020 can also be local to the scientific instrument 12010, but this is not necessarily the case; for example, a user-local computing device 12020 in a user's home or office can be remote from the scientific instrument 12010, but in communication with the scientific instrument so that the user can use the user-local computing device 12020 to control and / or access data from the scientific instrument 12010. In some embodiments, the user-local computing device 12020 can be a laptop, a smart phone, or a tablet device. In some embodiments, the user-local computing device 12020 can be a portable computing device.

[0116] The service local computing device 12030 can be a computing device local to the entity serving the scientific instrument 12010 (e.g., according to any of the embodiments of the computing device 11000 discussed herein). For example, the service local computing device 12030 can be a device local to the manufacturer of the scientific instrument 12010 or to a third-party service company. In some embodiments, the service local computing device 12030 can communicate with the scientific instrument 12010, the user local computing device 12020, and / or the remote computing device 12040 (e.g., via a direct communication path 12008 or via multiple "indirect" communication paths 12008, as described above) to receive data regarding the operation of the scientific instrument 12010, the user local computing device 12020, and / or the remote computing device 12040 (e.g., self-test results of the scientific instrument 12010, calibration coefficients used by the scientific instrument 12010, measurements of sensors associated with the scientific instrument 12010, etc.). In some embodiments, the service local computing device 12030 may communicate with the scientific instrument 12010, the user local computing device 12020 and / or the remote computing device 12040 (e.g., via a direct communication path 12008 or via multiple "indirect" communication paths 12008, as described above) to send data to the scientific instrument 12010, the user local computing device 12020 and / or the remote computing device 12040 (e.g., thereby updating programmed instructions (such as firmware) in the scientific instrument 12010 to initiate execution of a test or calibration sequence in the scientific instrument 12010, updating programmed instructions (such as software) in the user local computing device 12020 or the remote computing device 12040, etc.). A user of the scientific instrument 12010 may utilize the scientific instrument 12010 or the user's local computing device 12020 to communicate with the service local computing device 12030 to report problems with the scientific instrument 12010 or the user's local computing device 12020, thereby requesting a technician visit to improve the operation of the scientific instrument 12010, order consumables or replacement parts associated with the scientific instrument 12010, or for other purposes.

[0117] Remote computing device 12040 can be a computing device remote from scientific instrument 12010 and / or user local computing device 12020 (e.g., according to any of the embodiments of computing device 11000 discussed herein). In some embodiments, remote computing device 12040 can be included in a data center or other large-scale server environment. In some embodiments, remote computing device 12040 can include a network attached storage device (e.g., as part of storage device 12004). Remote computing device 12040 can store data generated by scientific instrument 12010, perform analysis of data generated by scientific instrument 12010 (e.g., according to programmed instructions), facilitate communication between user local computing device 12020 and scientific instrument 12010, and / or facilitate communication between service local computing device 12030 and scientific instrument 12010.

[0118] In some embodiments, Fig.12 One or more of the elements of the scientific instrument support system 12000 illustrated in the example may not be present. In addition, in some embodiments, Fig.12 Multiple of the various elements of the elements of the scientific instrument support system 12000 may be present. For example, the scientific instrument support system 12000 may include multiple user local computing devices 12020 (e.g., different user local computing devices 12020 associated with different users or located in different locations). As another example, the scientific instrument support system 12000 may include multiple scientific instruments 12010, all of which communicate with the service local computing device 12030 and / or the remote computing device 12040; in such an embodiment, the service local computing device 12030 may monitor these multiple scientific instruments 12010, and the service local computing device 12030 may cause updates or other information to be "broadcasted" to multiple scientific instruments 12010 simultaneously. Different scientific instruments of the scientific instruments 12010 in the scientific instrument support system 12000 may be close to each other (e.g., in the same room) or far away from each other (e.g., on different floors of a building, in different buildings, in different cities, etc.). In some embodiments, the scientific instrument 12010 may be connected to an Internet of Things (IoT) stack that allows the scientific instrument 12010 to be commanded and controlled through web-based applications, virtual or augmented reality applications, mobile applications, and / or desktop applications. Any of these applications may be accessed by a user operating a user-local computing device 12020 that communicates with the scientific instrument 12010 through an intermediary remote computing device 12040. In some embodiments, the scientific instrument 12010 may be sold by a manufacturer along with one or more associated user-local computing devices 12020 as part of a local scientific instrument computing unit 12012.

[0119] In some embodiments, different scientific instruments in the scientific instruments 12010 included in the scientific instrument support system 12000 can be different types of scientific instruments 12010; for example, one scientific instrument 12010 can be an optical emission spectrometer. In some such embodiments, the remote computing device 12040 and / or the user's local computing device 12020 can combine data from different types of scientific instruments 12010 included in the scientific instrument support system 12000.

Claims

1. A method of reducing distortion of an optical emission spectrum obtained from an optical emission spectrometer, wherein the optical emission spectrometer includes an optical system for forming the spectrum, the method comprising: obtaining a spectrum recorded with the optical emission spectrometer and input data, the input data comprising one or more condition parameters indicative of operating conditions of the optical emission spectrometer or an environment of the optical emission spectrometer when the spectrum was recorded with the spectrometer; providing a machine learning model configured to output, in response to the input data, output data including one or more transformation parameters of a transformation to be applied to the acquired spectrum to mitigate distortion of the spectrum due to a difference between the operating condition and a baseline operating condition; Applying the input data as input to the machine learning model and obtaining one or more transformation parameters as output of the machine learning model; as well as The transform according to the obtained one or more transform parameters is applied to the obtained spectrum to mitigate distortion of the spectrum caused by a difference between the operating condition and the baseline operating condition. 2 . The method of claim 1 , wherein the spectrum comprises sets of intensity values ​​at corresponding two-dimensional locations, and the transform comprises an operation that varies across locations. The method of claim 2 , wherein the operating comprises applying a deformation field to the optical emission spectrum.

4. A method according to any preceding claim, wherein the one or more condition parameters comprises a parameter indicative of temperature.

5. The method according to any preceding claim, wherein the one or more condition parameters are indicative of an operating condition of the optical system of the optical emission spectrometer.

6. The method of claim 4 or 5, wherein the one or more condition parameters include at least one temperature measurement obtained from a temperature sensor attached to a mechanical structure of the optical emission spectrometer.

7. The method of claim 6, wherein the mechanical structure of the optical emission spectrometer supports one or more optical components of the optical system.

8. A method according to any preceding claim, wherein the one or more condition parameters include one or more condition change parameters indicating a change in the condition or a direction of change in the condition.

9. A method according to any preceding claim, wherein the input data comprises a time series of one or more condition parameters at each of a plurality of time points.

10. A method according to claim 9, wherein the output data comprises a corresponding time series of one or more transformation parameters of the transformation at each of the plurality of time points, and the method comprises applying the transformation according to the one or more transformation parameters at the time point of the instance of the output data obtained to the optical emission spectrum obtained at the corresponding time point.

11. A method according to any preceding claim, wherein the machine learning model comprises an ensemble machine learning algorithm based on a decision tree.

12. A method according to any preceding claim, wherein the transformation comprises at least one of: a translation; a rotation; a scaling operation; a skewing operation; a stretching operation; and a deformation field.

13. A method according to any preceding claim, wherein the one or more condition parameters include one or more of the following: a parameter indicative of a heating current applied to a heating device for heating the optical system and stabilizing a temperature of the optical system; a parameter indicative of the ambient temperature of the spectrometer; a parameter indicative of a temperature of the optical system; a parameter indicative of at least one temperature measurement obtained from a temperature sensor attached to a mechanical structure supporting one or more optical components of the optical system; a parameter indicative of an RF power of an RF generator used to generate a plasma used to obtain the emission spectrum; and A parameter indicative of an exhaust pressure of a plasma chamber containing a plasma used to obtain said emission spectrum.

14. A method according to any preceding claim, wherein the optical emission spectrometer is a plasma emission spectrometer configured to record an emission spectrum of light emitted from a plasma.

15. A method according to any preceding claim, wherein the spectrum is a mid-step spectrum.

16. One or more non-transitory computer readable media comprising instructions thereon which, when executed by one or more processing devices in a scientific instrument support apparatus, cause the scientific instrument support apparatus to perform a method according to any preceding claim.

17. A scientific instrument system comprising an optical emission spectrometer for recording an optical emission spectrum and a scientific instrument support device configured to implement the method of any preceding claim to mitigate distortion of the optical spectrum.

18. The scientific instrument system of claim 17, wherein the optical emission spectrometer is an inductively coupled plasma optical emission spectrometer.

19. A method of training a machine learning model to output one or more transformation parameters in response to one or more conditional parameters, wherein the one or more condition parameters indicate operating conditions of the optical emission spectrometer or an environment of the optical emission spectrometer when the optical emission spectrum of the reference analyte is recorded using the optical emission spectrometer, and wherein the one or more transformation parameters of the transformation are to be applied to a corresponding recorded optical emission spectrum to reduce distortion of the spectrum relative to a baseline optical emission spectrum recorded using the optical emission spectrometer and the optical system under baseline operating conditions, The method comprises: obtaining a training data set comprising training data pairs, each training data pair comprising values ​​of the one or more condition parameters for a corresponding operating condition of the optical system different from the baseline operating condition, and corresponding values ​​of the one or more transformation parameters; as well as Adjust parameters of the machine learning model to reduce differences between values ​​of the one or more transformation parameters in the training data pair and values ​​of the one or more transformation parameters output by the machine learning model in response to corresponding values ​​of the one or more conditional parameters in the training data pair.

20. A method for obtaining training data for training a machine learning model, the method comprising: recording a plurality of optical emission spectra of a reference analyte with the optical emission spectrometer for corresponding operating conditions of the optical emission spectrometer; storing input data for the machine learning model, the input data comprising, for each recorded optical emission spectrum, one or more parameters indicative of the corresponding operating condition; for each recorded optical emission spectrum, adjusting one or more transformation parameters of a transformation to register the optical emission spectrum to a baseline optical emission spectrum of the reference analyte using the transformation, wherein the baseline optical emission spectrum was recorded for baseline operating conditions; as well as Output data for the machine learning model is stored, the output data comprising, for each optical emission spectrum, corresponding adjusted one or more transformation parameters associated with corresponding input data of each optical emission spectrum as a training data pair.

21. The method of claim 20, wherein the optical emission spectrum and the baseline optical emission spectrum each comprise respective sets of intensity values ​​at respective two-dimensional locations, and the transforming comprises an operation that varies across a plurality of locations.

22. The method of claim 21, wherein the operating comprises applying a distortion field to the optical emission spectrum.

23. The method of claim 21, wherein the optical emission spectrum and the baseline optical emission spectrum are corresponding images, and adjusting the one or more transformation parameters comprises comparing corresponding image intensities between the corresponding images.

24. The method of claim 23, wherein adjusting the one or more transformation parameters comprises using an adversarial similarity network.

25. A machine learning model trained using the method of claim 19.

26. A training data set for a machine learning model obtained using the method according to claims 20 to 24.

Images