AI-based method development and data acquisition assistant for spectrochemical analysis

Abstract

According to a first aspect of the present disclosure, the present disclosure describes a computer-implemented method for determining a particular analysis protocol for a sample, the sample being one of a plurality of sample types, and each sample type being associated with a corresponding particular analysis protocol. Performing the method comprises: obtaining a baseline spectrum of the sample using a baseline analysis scheme, the baseline analysis scheme being the same for the plurality of sample types; providing a machine learning model, such as a convolutional neural network, trained to output output data indicative of a particular analysis scheme for a sample in response to a spectrum of the sample; taking the obtained spectrum of the sample as the input of the machine learning model, and obtaining the output of the machine learning model; and determining a specific analysis protocol for the sample based on the output. This minimizes the burden on determining the correct analysis protocol for the sample.

Description

Technical Field

[0001] This disclosure relates to the availability of scientific instruments such as spectrometers, and in particular to facilitating the selection of analytical protocols for samples. Background Technology

[0002] There is a pervasive and ongoing need for systems and methods to reduce the level of analytical chemistry literacy required for the effective operation of scientific instruments and the analysis of samples. Typically, analyzing samples and operating the scientific instruments used to analyze them requires at least knowledge of the components that make up the sample and the appropriate analytical techniques for those components. This means that two barriers need to be overcome to effectively analyze samples, and both of these barriers require expertise in analytical chemistry.

[0003] In some cases, analytical chemists may possess basic knowledge of sample purity analysis techniques but be completely unaware of common impurities or bound compounds. In other cases, they may know the exact composition of a sample but lack the knowledge of which instruments and setups to use, or, for example, which measurement ranges to examine with those instruments to uncover additional information about the compound, such as its purity. Both of these barriers can incur time or financial costs due to the cost and availability of analytical chemists. Therefore, there is a need to minimize the burden of determining sample characteristics, particularly the burden of setting up instruments and specific analytical protocols. Summary of the Invention

[0004] According to a first aspect of this disclosure, a computer-implemented method for determining a specific analytical protocol for a sample, which is one of a variety of sample types, and each sample type is associated with a corresponding specific analytical protocol.

[0005] The computer-implemented method includes: obtaining a baseline spectrum of the sample using a baseline analysis protocol, which is the same for various sample types; providing a machine learning model, such as a convolutional neural network, trained to output data indicating a specific analytical protocol for the sample in response to the sample's spectrum; taking the obtained spectrum of the sample as input to the machine learning model and obtaining the model's output; and determining the specific analytical protocol for the sample based on the output. Advantageously, this minimizes the burden of determining the correct analytical protocol for a sample, allowing for the correct protocol to be followed even when the sample is unknown, so that it can be further studied. This method ensures the safe handling of potentially hazardous substances (e.g., heavy metals) in the presence of the sample. It should be understood that "spectrum" can refer to a spectrum obtained using any appropriate spectroscopic technique, such as: optical emission spectroscopy, such as inductively coupled plasma optical emission spectroscopy; mass spectrometry, such as inductively coupled plasma mass spectrometry; gas chromatography-mass spectrometry; liquid chromatography-mass spectrometry; nuclear magnetic resonance spectroscopy; or infrared spectroscopy. For example, a baseline spectrum can be an echelle grating spectrum obtained by optical emission spectroscopy; an intensity spectrum as a function of mass-to-charge ratio (m / z) obtained by mass spectrometry; or an intensity spectrum as a function of time obtained by gas chromatography or liquid chromatography. Advantageously, only a small amount of information about the sample is required to obtain the baseline spectrum because it is obtained through a baseline protocol that is identical for each of the multiple samples. Therefore, no sample-specific knowledge or information is needed to obtain the baseline spectrum.

[0006] In some examples, the method further includes recording further spectra of the sample using a determined specific analytical protocol.

[0007] In some examples, a particular analytical protocol includes one or more of the following: a sample acquisition method; one or more analytical techniques for analyzing the sample; one or more preferred settings of the analytical instrument for analyzing the sample; one or more standards for storing the sample; one or more analytical wavelength ranges for the functional groups of the sample; a concentration range for the preparation of external standards; one or more calibration factors; one or more injection settings, including peristaltic pump speed, tubing diameter and material, nebulizer type, nebulizer gas flow, auxiliary gas flow, cooling gas flow, and plasma power; and one or more optical system settings, including plasma observation direction (axial or radial), observation height (if radial observation is appropriate), exposure duration, subarray size, dilution settings for calibration curves, labeling / printing limits, and recommended number of repetitions per sample. In examples where the acquisition method includes one or more preferred settings of the analytical instrument for analyzing the sample, the method may further include adjusting one or more settings on one or more analytical instruments based on one or more preferred settings. Advantageously, this not only reduces the amount of analytical chemistry knowledge required to determine the analytical protocol for the sample but also reduces the amount of time required to perform the steps of the analytical protocol. It can also reduce errors, such as calibration setting errors or errors caused by the user not checking all analytically important species currently present in the sample. Reducing such errors avoids unnecessary increases in analysis time, and in addition, improves analytical accuracy and reduces waste.

[0008] In some examples, a method for training a machine learning model to output output data indicating a specific analytical protocol for a particular sample among a plurality of specific analytical protocols, each applicable to a corresponding sample type, in response to input of a sample's spectrum. The method includes: obtaining a training dataset comprising training data pairs, each training data pair including a spectrum obtained from the sample and an indication of a specific analytical protocol among the specific analytical protocols for that sample, the training data including corresponding spectra for each sample type obtained from samples of each sample type; and adjusting the parameters of the machine learning model in response to the corresponding spectra of the training data pairs to reduce the discrepancy between the indications of the training data pairs and the indications output by the machine learning model.

[0009] In some examples, a method for obtaining training data for training a machine learning model is provided. The method includes: for each of a plurality of specific analysis schemes, obtaining a baseline spectrum for each of a plurality of samples of a type to which that specific analysis scheme is applicable, the spectrum being obtained using a baseline analysis scheme that is identical for all the plurality of sample types; and storing input and output data pairs for the machine learning model, each pair of input data including the obtained baseline spectrum, and each pair of output data including an indication of the analysis scheme for the corresponding sample. The samples for the training data are selected to cover the sample space that may be encountered when using the model.

[0010] According to another aspect of this disclosure, a system is provided that includes one or more processors and one or more memories storing computer-readable instructions thereon, the computer-readable instructions being configured to cause the one or more processors to perform any of the methods disclosed herein. In some examples, the system further includes a spectrometer for recording spectra.

[0011] According to another aspect of this disclosure, a computer program including instructions is provided that, when executed by a computer, causes the computer to perform any of the methods disclosed herein. Attached Figure Description

[0012] The disclosed embodiments will now be described with the aid of examples and reference to the accompanying drawings, in which: Figure 1 shows an example of an optical spectrometer 100 used to obtain the spectrum of a sample; Figure 2 is a flowchart of an example method 200 for obtaining training data for training a machine learning model, based on various examples; Figure 3 is a flowchart of an example method 300 for training machine learning models based on various examples to output output data indicating a specific analytical scheme for a sample in response to the spectral input of the sample; Figure 4 is a flowchart of an example method 400 for determining the analytical scheme of a sample based on various examples; Figure 5A shows an example of the spectrum of a sample containing aluminum oxide. Figure 5B shows an example of the spectrum of a sample containing silicon oxide.

[0013] Figure 6 is an example of a graphical user interface according to various embodiments that can be used to perform some or all of the supported methods disclosed herein.

[0014] Figure 7 is a block diagram of an example computing device that can perform some or all of the scientific instrument support methods disclosed herein, according to various embodiments.

[0015] Figure 8 is a block diagram of an example scientific instrument support system according to various embodiments, in which some or all of the scientific instrument support methods disclosed herein may be performed. Detailed Implementation

[0016] Referring to Figure 1, an example optical spectrometer 100 for obtaining a spectrum of a sample includes a sample introduction system 102, an optical system 104, and a detector 106. Although an optical spectrometer is used herein as an illustrative example, it should be understood that other types of spectrometers can be used to obtain other types of spectra to measure the properties of a sample within a given range. This disclosure relates to methods that can be performed using data obtained from one or more different types of spectrometers (i.e., they are not spectrometer-specific in themselves, as will be discussed in more detail with reference to Figures 2 through 4). For example, the optical spectrometer 100 may be an inductively coupled plasma emission spectrometer (ICP-OES), which operates by exciting atoms and ions to emit electromagnetic (EM) radiation at characteristic wavelengths of a particular element. Electrons emit a spectrum of EM radiation when they transition from a high-energy state to a low-energy state. In such a case, the sample introduction system 102 may include a plasma chamber connected to a radio frequency (RF) voltage source and a gas (e.g., argon) source. The voltage source is used to apply a voltage to a coil that induces a radio frequency (RF) electromagnetic field. Argon gas can be ionized by an electromagnetic field inside the plasma chamber to generate and sustain plasma within the chamber. The optical system 104 may include an echelle diffraction grating, prisms, and multiple focusing mirrors to selectively focus light from the plasma chamber and diffract it into multiple diffraction orders, thereby producing a high-resolution 2D spectrum, referred to as an echelle diffraction pattern, also known as a full frame when detected by detector 106. The optical spectrometer 100 may be part of a scientific instrument support system, such as the scientific instrument support system 800 described with reference to Figure 8, and thus may be coupled to computing devices 820, 830, and 840, which include a processing device 802, a storage device 804, and an interface device 806. The optical spectrometer 100 requires numerous setup adjustments for each sample. For example, for ICP-OES, settings include the temperature of the optical system, the spectral acquisition rate, and the RF power supplied to the plasma chamber. In addition to the spectrometer settings, the analytical protocol includes detailed information on auxiliary steps such as sample preparation procedures. As will be discussed in more detail with reference to Figures 2 through 4, many different types of spectrometers can be used in addition to ICP-OES. For example, inductively coupled plasma mass spectrometry (ICP-MS) or total ion chromatography (TIC for GC-MS) can be used.

[0017] Referring to Figures 2 through 4, methods for determining analytical protocols for samples and for training machine models to provide such determinations are described. These methods reduce the amount of analytical chemistry literacy required to efficiently operate chromatographic, mass spectrometric, and / or spectroscopic instruments, such as the optical spectrometer 100 described above with reference to Figure 1. To train the machine learning model to determine specific analytical protocols for samples, training data is obtained for training the machine learning model.

[0018] Figure 2 is a flowchart of an example method 200 for obtaining training data for training a machine learning model, based on various examples. The training dataset includes training data pairs, where each training data pair includes a baseline spectrum obtained from a sample and a label indicating a specific analytical scheme among a plurality of specific analytical schemes corresponding to the sample. For example, the label "Scheme 1" may correspond to a first specific analytical scheme, the label "Scheme 2" may correspond to a second specific analytical scheme, and so on.

[0019] A specific analytical protocol is a known protocol corresponding to a specific sample type (unlike a baseline protocol that is not sample-specific) and may involve sample-specific analytical techniques, storage settings, one or more scientific instruments (such as an optical spectrometer 100) used to analyze the sample, and corresponding instrument calibrations. More specifically, an analytical protocol may specify one or more of the following: the sample acquisition method, one or more analytical techniques used to analyze the sample, one or more preferred settings of the analytical instruments used to analyze the sample, one or more standards used to store the sample, one or more analytical wavelength ranges for the functional groups of the sample, concentration ranges for the preparation of external standards, and one or more calibration factors. For example, if the analytical protocol specifies that the sample analysis is performed on a mass spectrometer, the protocol may further include: one or more injection settings, including peristaltic pump speed, tubing diameter and material, nebulizer type, nebulizer gas flow, auxiliary gas flow, cooling gas flow, and plasma power; and one or more optical system settings, including plasma observation direction (axial or radial), observation height (if radial observation is appropriate), exposure duration, and subarray size. For example, if the analytical protocol details UV analysis, it may further include dilution settings for calibration curves, labeling / printing limits, or recommended number of repetitions per sample. The concentration range for external standard preparation can refer to the range used to determine the relationship between the intensity of emission lines observed in the ICP-OES spectrum and the concentration of the analyte in the sample. Labeling / printing limits can indicate regulatory requirements, guidelines, and detection limits. Limits can be quality control (QC) limits or recovery limits. QC limits require that the measured concentration of the QC standard fall within an acceptable range (e.g., ±5%) near a known value. Recovery limits require that recovery studies (adding a known amount of analyte to the sample) demonstrate that the recovery is within an acceptable range (e.g., 90% to 110%). Matrix spiked recovery requires that matrix spiked recovery studies demonstrate that the method can accurately measure analytes in complex sample matrices. LOD and LOQ refer to the lowest analyte concentration that can be distinguished from the assay background, and the lowest concentration at which the analyte can be quantified at defined levels of imprecision and accuracy, respectively. Stability limits indicate the instrument's stability over time, ensuring that machine drift and fluctuations are within acceptable limits. Repeatability indicates that repeated measurements of the same sample should have good accuracy (low relative standard deviation, typically less than 5%).

[0020] Obtaining training data may include recording baseline spectra on scientific instruments such as the optical spectrometer 100 described with reference to Figure 1, or retrieving baseline spectra from databases or other storage systems or devices such as those described below, for example, in Figure 8. Example baseline spectra are shown in Figures 5A and 5B, which respectively show example inductively coupled plasma emission spectra of aluminum oxide and silicon oxide samples.

[0021] Unlike specific analytical protocols, the baseline protocol used to obtain baseline spectra is the same for every sample across multiple sample types. It is defined in this way to generate spectra suitable for identifying a variety of different sample types. Throughout the training dataset, baseline spectra are recorded in a consistent manner and on the same type of scientific instrument (such as ICP-OES) according to the same predefined calibration. These baseline spectra recorded on the same type of scientific instrument are then correlated with the corresponding analytical protocols, and a machine learning model is trained using the baseline spectrum and analytical protocol pairs. The trained model can then be used to predict specific analytical protocols for use with corresponding new samples.

[0022] At step 202, a baseline spectrum is obtained for each of the multiple samples applicable to a given specific analytical protocol. For example, when there are 100 different specific analytical protocols labeled “Protocol 1” through “Protocol 100”, multiple samples are obtained for each of the 100 different specific analytical protocols, and a baseline spectrum is obtained for each sample. In some examples, this multiple may be around 100 to 1000, but it should be understood that this number is specific to the machine learning model, and in some cases, the model may allow fewer or require more samples to converge or meet the stopping condition, as described in more detail with reference to Figure 3.

[0023] In some examples, obtaining a baseline spectrum may require dissolving the sample in a solvent such as water or an organic compound. In such cases, additional spectra of the solvent can also be obtained (from a computing device or by recording the spectra). When obtaining the solvent spectrum, in some examples, peaks corresponding to the solvent are removed from the sample's spectrum in the solvent to obtain the baseline spectrum. This can be done by any suitable method, for example, by aligning the solvent's spectrum with the sample's spectrum in the solvent to determine equivalent peaks between the solvent's and the sample's spectra. These peaks can then be removed by subtracting the equivalent peaks from the spectrum of the unknown sample. Of course, if a baseline spectrum for a certain type of sample is obtained in this way (by subtracting solvent peaks from the spectrum), the same process will be used to obtain the baseline spectrum for that type of sample during inference. In some examples, solvent peak subtraction applies to baseline spectra for all sample types, regardless of sample type.

[0024] At step 204, input and output data pairs are stored, where the input data includes the obtained baseline spectra, and the output data for each pair includes a label for the analytical protocol of the corresponding sample (such as labels "Program 1", "Program 2", and "Program 3" corresponding to analytical protocols one, two, and three). The stored input and output data pairs provide training data for the machine learning model. For example, the dataset may include N data pairs corresponding to each of M different analytical protocols (i.e., a total of N x M data pairs).

[0025] Figure 3 is a flowchart of an example method 300 for training a machine learning model to output output data indicating a specific analytical protocol for a sample in response to the input of the sample's baseline spectrum. The output data indicates a specific analytical protocol among multiple specific analytical protocols, each applicable to a corresponding sample type, via corresponding labels.

[0026] The machine learning model can be any suitable architecture used to process spectral data. For many scientific instruments described herein, such as the ICP-OES illustrated with reference to Figure 1, the obtained spectra are in the form of image data, as shown in Figure 5, for example. In such cases, the machine learning model can be one of any suitable known architectures of a convolutional neural network (CNN) with convolutional and pooling layers, where the output layer is adapted for image classification. For example, Res-Net (Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun: “Deep Residual Learning for Image Recognition” arXiv preprint arXiv:1512.03385, 2015) is a convolutional neural network, which is trained according to method 300 in some embodiments. In some embodiments, the spectrum is not in the form of image data, but rather in the form of, for example, a list of spectral peak coordinates, a graph of intensity versus mass-to-charge ratio (m / z), or a graph of time versus mass-to-charge ratio (m / z) with peak positions on it, and can typically be one-dimensional or two-dimensional. In these embodiments, the disclosed methods remain applicable, but machine learning models suitable for processing the data in question are used, making the disclosed methods applicable to any type of spectral data in the appropriate corresponding embodiment.

[0027] At step 302, the results of a previous application of method 200 are obtained online using method 200 or retrieved from a digital storage device, to obtain a training dataset including training data pairs.

[0028] At step 304, in response to the corresponding spectra of the training data pairs, the parameters of the machine learning model are adjusted to reduce the discrepancy between the labels of the training data pairs and the labels output by the machine learning model. For example, the discrepancy can be represented by a loss function that represents the difference between the label and the prediction. Any suitable loss function can be used, such as cross-entropy or focus loss. Any suitable optimization algorithm can be used to tune the parameters, such as stochastic gradient descent, such as Adam.

[0029] Step 304 can be performed until a stopping condition is met. For example, step 304 can be performed iteratively for a predetermined number of iterations (in batches or epochs), or until a threshold difference is reached (e.g., represented by a threshold of the magnitude of the loss function), or until both conditions are met simultaneously. For completeness, although batch size and epochs are common terms in the art, they are described briefly here. In this context, the batch size of the model refers to the number of data pairs to be processed before tuning the model's parameters. The number of epochs defines the number of times the dataset is fully traversed. Of course, it should be understood that the stopping condition of step 304 depends on the machine learning model, and therefore any appropriate stopping condition corresponding to the machine learning model discussed can be used.

[0030] The trained machine learning model can be used to determine a specific analytical protocol for a sample according to method 400 as described in Figure 4. Method 400 provides the user with any information needed to analyze, store, or react the sample, regardless of whether the user knows the components in the sample.

[0031] At step 402, a baseline spectrum of the sample is obtained using a baseline analysis protocol. As discussed above with reference to method 200 for obtaining training data, the baseline analysis protocol is a set of predetermined steps and instrument calibrations to be performed to obtain a baseline spectrum using predefined scientific instruments.

[0032] At step 404, a machine learning model is provided. The machine learning model has been trained to output output data indicating a sample-specific analytical protocol in response to the sample's spectrum. For example, the machine learning model can be trained using method 300 described above with reference to Figure 3.

[0033] At step 406, the obtained baseline spectrum of the sample is used as input to the machine learning model, and the output of the machine learning model is obtained. Based on the baseline spectrum of the sample, the output provides an indication of the analytical protocol to be used with the sample. This indication can be a label indicating the analytical protocol to be used. For example, the labels “Program 1,” “Program 2,” and “Program 3” may correspond to different corresponding analytical protocols that have been associated (via the baseline spectrum) with sodium salts containing different halide components (fluorine, chlorine, bromine, etc.). In another example, the classifications “Program 4,” “Program 5,” and “Program 6” may correspond to different analytical protocols that are associated with different concentrations of the same compound that require different dilutions for further analysis or different procedures for storage. Of course, these are just illustrative examples, and the indication of the corresponding analytical protocol can actually correspond to any analytical protocol for different types of samples, which are associated with different types of samples by the corresponding baseline spectrum on which the machine learning model is trained.

[0034] At step 408, a specific analytical protocol for the sample is obtained based on the output. For example, a label identifying one of multiple analytical protocols is used to search for data records corresponding to that analytical protocol, for example, by means of a dictionary or lookup table or by any other suitable method, such as any form of database access or query. In this way, the label identifying the analytical protocol can be used to retrieve sample-specific analytical techniques, storage settings, one or more scientific instruments (such as an optical spectrometer 100 for analyzing the sample), and corresponding instrument calibrations corresponding to the specific analytical protocol label. In some examples, the protocol may be an analytical protocol from a scientific institution or journal, such that the specific analytical protocol includes methods for sample collection, analysis, and storage according to industry regulations or standards.

[0035] In examples where a particular analytical scheme includes one or more preferred settings of analytical instruments for analyzing samples, in some embodiments, the method further includes adjusting one or more settings of one or more analytical instruments based on one or more preferred settings. Using a tag identifying a particular analytical scheme among a plurality of specific analytical schemes, the corresponding calibration instructions are obtained from the data record of the retrieved analytical scheme. The calibration instructions can be tagged with an automated tag indicating that, upon retrieval, the corresponding calibration instructions can be provided from computing devices 820, 830, 840 to one or more scientific instruments 810 via one or more communication paths 808 as shown in FIG. 8. Scientific instrument 810, user local computing device 820, service local computing device 830, and remote computing device 840 can communicate with other components of scientific instrument support system 800 via communication path 808 (as described in more detail with reference to FIG. 8). The scientific instrument control area 606 coupled to the instrument (described in more detail with reference to FIG. 6) can then control the calibration settings on the scientific instrument.

[0036] In other examples, calibration settings may be provided to the user on a GUI 600 (Figure 6) coupled to the spectrometer 810. In such examples, the user can approve the calibration settings via the scientific instrument control area 606 before the calibration settings on the instrument 810 are automatically updated.

[0037] In the detailed example, an experiment was conducted to determine specific analytical protocols for aluminum, silicon, and boron. Aluminum, silicon, and boron have different properties; for example, they exhibit peaks at different wavelengths and have different boiling points, and therefore have specific protocols that must be followed to obtain useful information about the purity, functional groups, etc., of the sample.

[0038] The training dataset was obtained using samples in the form of acidified elements of one of the following: aluminum; silicon; and boron. Additionally, samples containing blank solutions of no these elements were also used for the training dataset. The purpose of the blank solutions in this experiment was primarily to study the accuracy of the model, but the blank solutions could also be used to subtract solvent peaks, as described above with reference to step 202.

[0039] A total of 353 samples were used in the experiment, including 102 aluminum oxide samples, 81 silicon oxide samples, 106 boron oxide samples, and 64 blank samples. Baseline spectra were obtained for each of the 353 samples according to the same baseline protocol, which detailed the predefined calibration of ICP-OES. Examples of baseline spectra can be seen in Figures 5A and 5B, which show the baseline spectra of aluminum-containing and silicon-containing samples, respectively.

[0040] Each of the 353 baseline spectra was labeled with a tag indicating a specific analytical protocol among several specific analytical protocols corresponding to the sample, in order to obtain training data as discussed above with reference to Figure 2, which associates aluminum oxide, silicon oxide, and boron oxide samples with the corresponding analytical protocol for each type of sample.

[0041] The training data was divided into a training set containing 213 samples (approximately 60% of the data), a validation set containing 70 samples (approximately 20% of the data), and a test set containing 70 samples (approximately 20% of the data).

[0042] The model was then tested using the remaining 70 test samples. For each test sample, after obtaining a label indicating the analytical protocol as the model's output, the analytical protocol was retrieved from the database and provided to the user, for example, by displaying it on a monitor. The analytical protocol contains information needed to analyze, store, or react the sample, or information provided to the control area of ​​the scientific instrument to trigger automatic updates of the calibration settings on the scientific instrument.

[0043] Figure 6 depicts an example GUI 600 according to various embodiments, which can be used to perform some or all of the support methods disclosed herein. As described above, GUI 600 can be located on a computing device (e.g., referenced herein to the scientific instrument support system 800 discussed with reference to Figure 8) of a scientific instrument support system. Figure 7 The GUI 600 can be interacted with by a user on a display device (e.g., the display device 710 discussed herein with reference to FIG. 7) and by any suitable input device (e.g., any input device included in the other I / O devices 712 discussed herein with reference to FIG. 7) and input technology (e.g., cursor movement, motion capture, facial recognition, gesture detection, speech recognition, button driving, etc.).

[0044] GUI 600 may include a data display area 602, a data analysis area 604, a scientific instrument control area 606, and a settings area 608. The specific number and arrangement of areas depicted in Figure 6 are illustrative only, and GUI 600 may include any number and arrangement of areas, including any desired features.

[0045] Data display area 602 can display data generated by a scientific instrument (e.g., scientific instrument 810 discussed herein with reference to Figure 8). For example, data display area 602 can display calibration settings, such as temperature or wavelength.

[0046] Data analysis area 604 can display the results of data analysis (e.g., the results of analyzing the data shown in data display area 602 and / or other data). For example, data analysis area 604 can display wavelength readings or timing. In some embodiments, data display area 602 and data analysis area 604 can be combined in GUI 600 (e.g., to include data output from scientific instruments and some analysis of the data in a public graphic or area).

[0047] The scientific instrument control area 606 may include options that allow a user to control a scientific instrument (e.g., scientific instrument 810 discussed herein with reference to FIG8). For example, the scientific instrument control area 606 may include calibration control.

[0048] Settings area 608 may include options that allow users to control the features and functions of GUI 600 (and / or other GUIs) and / or perform common computational operations regarding data display area 602 and data analysis area 604 (e.g., storing data on a storage device such as storage device 704 discussed herein with reference to FIG. 7, sending data to another user, tagging data, etc.). For example, settings area 608 may include options to retrieve stored spectra (e.g., spectra of known solvents).

[0049] Figure 7 is a block diagram of a computing device 700 according to various embodiments, capable of performing some or all of the methods disclosed herein.

[0050] The computing device 700 of Figure 7 is illustrated as having multiple components, but any or more of these components may be omitted or duplicated to suit the application and setup. In some embodiments, some or all of the components included in the computing device 700 may be attached to one or more motherboards and encapsulated in a housing (e.g., including plastic, metal, and / or other materials). In some embodiments, some of these components may be fabricated onto a single system-on-a-chip (SoC) (e.g., the SoC may include one or more processing devices 702 and one or more storage devices 704). Furthermore, in various embodiments, the computing device 700 may not include one or more of the components shown in Figure 7, but 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 700 may not include display device 710, but may include display device interface circuitry (e.g., connectors and driver circuitry) to which display device 710 may be coupled.

[0051] Computing device 700 may include processing device 702 (e.g., one or more processing devices). Herein, the term "processing device" can refer to any device or part of a device that processes electronic data from registers and / or memory to convert said electronic data into other electronic data that can be stored in registers and / or memory. Processing device 702 may include one or more digital signal processors (DSPs), application-specific integrated circuits (ASICs), central processing units (CPUs), graphics processing units (GPUs), cryptographic processors (dedicated processors that execute cryptographic algorithms within hardware), server processors, or any other suitable processing device.

[0052] Computing device 700 may include storage device 704 (e.g., one or more storage devices). Storage device 704 may include one or more memory devices, such as random access memory (RAM) (e.g., static RAM (SRAM) devices, magnetic RAM (MRAM) devices, dynamic RAM (DRAM) devices, resistive RAM (RRAM) devices, or conductive bridged RAM (CBRAM) devices), hard disk drive-based memory devices, solid-state memory devices, network drives, cloud drives, or any combination of memory devices. In some embodiments, storage device 704 may include memory sharing a die with processing device 702. In such embodiments, the memory may be used as cache memory and may include, for example, embedded dynamic random access memory (eDRAM) or spin-transfer torque magnetic random access memory (STT-MRAM). In some embodiments, storage device 704 may include a non-transitory computer-readable medium having instructions thereon that, when executed by one or more processing devices (e.g., processing device 702), cause computing device 700 to perform any suitable method or portion thereof of the methods disclosed herein.

[0053] Computing device 700 may include interface device 706 (e.g., one or more interface devices 706). Interface device 706 may include one or more communication chips, connectors, and / or other hardware and software to manage communication between computing device 700 and other computing devices. For example, interface device 706 may include circuitry for managing wireless communication used to transfer data with computing device 700. The term "wireless" and its derivatives can be used to describe circuits, devices, systems, methods, techniques, communication channels, etc., that can transmit data through a non-solid medium using modulated electromagnetic radiation. This 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 interface device 706 for managing wireless communications may implement any of a number 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 interface device 4006 for managing wireless communications may operate according to 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 networks. In some embodiments, the circuitry included in the interface device 4006 for managing wireless communications may operate according to Enhanced Data Rate GSM Evolution (EDGE), GSM EDGE 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 4006 for managing wireless communications may operate according to Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Digital Enhanced Cordless Telecommunications (DECT), Evolved Data Optimization (EV-DO) and its derivatives, as well as any other wireless protocol designated as 3G, 4G, 5G, and higher. In some embodiments, the interface device 706 may include one or more antennas (e.g., one or more antenna arrays) to receive and / or transmit wireless communications.

[0054] In some embodiments, interface device 706 may include circuitry for managing wired communications, such as electrical, optical, or any other suitable communication protocol. For example, interface device 706 may include circuitry supporting communications based on Ethernet technology. In some embodiments, interface device 706 may support both wireless and wired communications, and / or may support multiple wired communication protocols and / or multiple wireless communication protocols. For example, a first set of circuitry for interface device 706 may be dedicated to short-range wireless communications such as Wi-Fi or Bluetooth, while a second set of circuitry for interface device 706 may be dedicated to long-range wireless communications such as Global Positioning System (GPS), EDGE, GPRS, CDMA, WiMAX, LTE, EV-DO, etc. In some embodiments, a first set of circuitry for interface device 4006 may be dedicated to wireless communications, while a second set of circuitry for interface device 4006 may be dedicated to wired communications.

[0055] The computing device 700 may include a battery / power circuit 708. The battery / power circuit 708 may include one or more energy storage devices (e.g., batteries or capacitors) and / or circuitry for coupling components of the computing device 700 to a power source (e.g., AC line power) that is separate from the computing device 700.

[0056] Computing device 700 may include display device 710 (e.g., multiple display devices). Display device 710 may include any visual indicator, such as a head-up display, computer monitor, projector, touch screen display, liquid crystal display (LCD), light-emitting diode display, or flat panel display.

[0057] The computing device 700 may include other input / output (I / O) devices 712. Other I / O devices 712 may include, for example, one or more audio output devices (e.g., speakers, headphones, earphones, alarm clocks, 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 4000, as 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, styluses, trackballs, or touchpads), barcode readers, quick-response (QR) code readers, or radio frequency identification (RFID) readers.

[0058] The computing device 700 may have any suitable form factor for its application and setup, such as a handheld or mobile computing device (e.g., a cellular phone, smartphone, mobile internet device, tablet computer, laptop computer, netbook computer, ultrabook computer, personal digital assistant (PDA), ultra-mobile personal computer, etc.), a desktop computing device, a server computing device, or other networked computing components.

[0059] 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. Figure 8 is a block diagram of an example scientific instrument support system 800 according to various embodiments, in which some or all of the scientific instrument support methods disclosed herein may be performed. The scientific instrument support modules and methods disclosed herein (e.g., scientific instrument 100 of Figure 1 and method 400 of Figure 4) may be implemented by one or more of the scientific instrument 810, user local computing device 820, service local computing device 830, or remote computing device 840 of the scientific instrument support system 800.

[0060] Any of the scientific instrument 810, the user local computing device 820, the service local computing device 830, or the remote computing device 840 may include any of the embodiments of the computing device 700 discussed herein with reference to FIG. 7, and any of the scientific instrument 810, the user local computing device 820, the service local computing device 830, or the remote computing device 840 may take the form of any suitable embodiment of the computing device 700 discussed herein with reference to FIG. 7.

[0061] Scientific instrument 810, user local computing device 820, service local computing device 830, or remote computing device 840 may each include a processing device 802, a storage device 804, and an interface device 806. The processing device 802 may take any suitable form, including any of the processing devices 802 discussed herein with reference to FIG. 4, and the processing devices 802 included in different devices of scientific instrument 810, user local computing device 820, service local computing device 830, or remote computing device 840 may take the same or different forms. The storage device 804 may take any suitable form, including any of the storage devices 804 discussed herein with reference to FIG. 7, and the storage devices 804 included in different devices of scientific instrument 810, user local computing device 820, service local computing device 830, or remote computing device 840 may take the same or different forms. Interface device 806 may take any suitable form, including any of the interface devices 806 discussed herein with reference to FIG4, and the interface devices 806 included in different devices such as scientific instrument 810, user local computing device 820, service local computing device 830, or remote computing device 840 may take the same or different forms.

[0062] Scientific instrument 810, user local computing device 820, service local computing device 830, and remote computing device 840 can communicate with other elements of scientific instrument support system 800 via communication path 808. As shown, communication path 808 can communicatively couple interface devices 806 of different elements in scientific instrument support system 800 and can be a wired or wireless communication path (e.g., any communication technology discussed herein with reference to interface device 706 of computing device 700 in Figure D). The particular scientific instrument support system 800 depicted in Figure 8 includes communication paths between each pair of devices in scientific instrument 810, user local computing device 820, service local computing device 830, and remote computing device 840; however, this “fully connected” implementation is merely illustrative, and various communication paths 808 may not exist in various embodiments. For example, in some embodiments, the serving local computing device 830 may not have a direct communication path 808 between its interface device 806 and the interface device 806 of the scientific instrument 810, but may instead communicate with the scientific instrument 810 via a communication path 808 between the serving local computing device 830 and the user local computing device 820 and a communication path 808 between the user local computing device 820 and the scientific instrument 810.

[0063] Scientific instrument 810 may include any suitable scientific instrument, such as inductively coupled plasma optical emission spectrometer (ICP-OES), inductively coupled plasma mass spectrometer (ICP-MS), or total ion chromatograph for gas chromatography-mass spectrometry (TIC for GC-MS).

[0064] User local computing device 820 may be a computing device that is local to the user of scientific instrument 810 (e.g., any embodiment of the computing device 700 discussed herein). In some embodiments, user local computing device 820 may also be located locally to scientific instrument 810, but this is not necessarily the case; for example, user local computing device 820 located in the user's home or office may be remote from scientific instrument 810 but communicate with said scientific instrument, allowing the user to use user local computing device 820 to control and / or access data from scientific instrument 810. In some embodiments, user local computing device 820 may be a laptop, smartphone, or tablet device. In some embodiments, user local computing device 820 may be a portable computing device. In some embodiments, user local computing device 820 may perform the methods described herein with reference to Figures 2, 3, and 4.

[0065] The servicing local computing device 830 can be a computing device that is local to the entity serving the scientific instrument 810 (e.g., any embodiment of the computing device 4000 discussed herein). For example, the servicing local computing device 830 can be a local device of the manufacturer of the scientific instrument 810 or a third-party service company. In some embodiments, the servicing local computing device 830 can communicate with the scientific instrument 810, the user local computing device 820, and / or the remote computing device 840 (e.g., via a direct communication path 808 or via multiple “indirect” communication paths 808, as discussed above) to receive data regarding the operation of the scientific instrument 810, the user local computing device 820, and / or the remote computing device 840 (e.g., self-test results of the scientific instrument 810, calibration coefficients used by the scientific instrument 810, measurement results of sensors associated with the scientific instrument 810, etc.). In some embodiments, the service local computing device 830 may communicate with scientific instrument 810, user local computing device 820, and / or remote computing device 840 (e.g., via direct communication path 808 or via multiple “indirect” communication paths 808, as discussed above) to transmit data to scientific instrument 810, user local computing device 820, and / or remote computing device 840 (e.g., to update programming instructions (such as firmware) in scientific instrument 810, initiate the execution of test or calibration sequences in scientific instrument 810, update programming instructions (such as software) in user local computing device 820 or remote computing device 840, etc.). A user of scientific instrument 810 may use scientific instrument 810 or user local computing device 820 to communicate with service local computing device 830 to report problems with scientific instrument 810 or user local computing device 820, request on-site technician visits to improve the operation of scientific instrument 810, order consumables or replacement parts associated with scientific instrument 810, or for other purposes.

[0066] Remote computing device 840 may be a computing device located remotely from scientific instrument 810 and / or remotely from user local computing device 820 (e.g., any embodiment of the computing device 4000 discussed herein). In some embodiments, remote computing device 840 may be included in a data center or other large-scale server environment. In some embodiments, remote computing device 840 may include network-attached storage (e.g., as part of storage device 804). Remote computing device 840 may store data generated by scientific instrument 810, perform analysis on data generated by scientific instrument 810 (e.g., according to programming instructions), facilitate communication between user local computing device 820 and scientific instrument 810, and / or facilitate communication between service local computing device 830 and scientific instrument 810.

[0067] In some embodiments, one or more of the elements of the scientific instrument support system 800 shown in FIG. E may be absent. Furthermore, in some embodiments, multiple elements of the various elements of the scientific instrument support system 800 of FIG. E may be present. For example, the scientific instrument support system 800 may include multiple user local computing devices 820 (e.g., different user local computing devices 820 associated with different users or located in different locations). In another example, the scientific instrument support system 800 may include multiple scientific instruments 810, all of which communicate with a serving local computing device 830 and / or a remote computing device 840; in such embodiments, the serving local computing device 830 may monitor multiple scientific instruments 810, and the serving local computing device 830 may cause updates, or other information may be simultaneously “broadcast” to multiple scientific instruments 810. Different scientific instruments among the scientific instruments 810 in the scientific instrument support system 800 may be close to each other (e.g., in the same room) or far from each other (e.g., on different floors of a building, in different buildings, in different cities, etc.). In some embodiments, scientific instrument 810 may be connected to an Internet of Things (IoT) stack that allows command and control of scientific instrument 810 via web-based applications, virtual or augmented reality applications, mobile applications, and / or desktop applications. Any of these applications may be accessible to a user operating a user-local computing device 820 that communicates with scientific instrument 810 via an intermediate remote computing device 840. In some embodiments, scientific instrument 810 may be sold by a manufacturer along with one or more associated user-local computing devices 820 as part of a local scientific instrument computing unit 812.

[0068] In some embodiments, the different scientific instruments included in the scientific instruments 810 of the scientific instrument support system 800 may be different types of scientific instruments 810. In some such embodiments, the remote computing device 840 and / or the user local computing device 820 may combine data from the different types of scientific instruments 810 included in the scientific instrument support system 800.

Claims

1. A computer-implemented method for determining a specific analytical protocol for a sample, said sample being one of a plurality of sample types, and each sample type being associated with a corresponding specific analytical protocol, said method comprising: The baseline spectrum of the sample is obtained using a baseline analysis protocol, which is the same for all sample types. Provide a machine learning model that is trained to output output data indicating a specific analytical protocol for the sample in response to the sample's spectrum; The obtained spectrum of the sample is used as the input to the machine learning model, and the output of the machine learning model is obtained. as well as Based on the output, a specific analytical protocol for the sample is determined.

2. The method of claim 1, wherein the baseline spectrum is a baseline spectrum obtained using one of the following techniques: Optical emission spectroscopy, such as inductively coupled plasma optical emission spectroscopy; Mass spectrometry, such as inductively coupled plasma mass spectrometry; Gas chromatography-mass spectrometry; or Liquid chromatography-mass spectrometry, The machine learning model is trained based on the spectrum obtained using the same technique.

3. The method according to claim 1 or 2, wherein the method further comprises: The determined specific analytical protocol is used to record further spectra of the sample.

4. The method according to any of the preceding claims, wherein obtaining the baseline spectrum of the sample comprises obtaining the spectrum of the sample in the solvent and removing peaks corresponding to the solvent from the spectrum.

5. The method according to any of the preceding claims, wherein the machine learning model comprises a convolutional neural network.

6. The method according to any of the preceding claims, wherein the specific analytical scheme comprises: The method for collecting the sample.

7. The method according to any preceding claim, wherein the specific analytical scheme for the sample comprises one or more of the following: One or more analytical techniques are used to analyze the sample; One or more preferred configurations of the analytical instruments used to analyze the sample; One or more wavelength ranges of the functional groups of the sample; Concentration range for external standard preparation; One or more calibration factors; One or more sample introduction settings, including peristaltic pump speed, tubing diameter and material, nebulizer type, nebulizer gas flow, auxiliary gas flow, cooling gas flow and plasma power; One or more optical system settings, including plasma observation direction (axial or radial), observation height (if radial observation is appropriate), exposure duration, and subarray size; Dilution settings for the calibration curve; Marking / printing limits; and Recommended number of replicates for each sample.

8. The method of claim 7, wherein the analytical scheme includes one or more preferred configurations of an analytical instrument for analyzing the sample, the method further comprising: Adjust one or more settings on one or more analytical instruments based on one or more of the preferred settings.

9. The method according to any preceding claim, wherein the analytical scheme comprises: One or more standards for storing the sample.

10. A method for training a machine learning model to output output data indicating a particular analytical protocol for a sample among a plurality of particular analytical protocols, each particular analytical protocol being suitable for a corresponding sample type, the method comprising: Obtain a training dataset comprising training data pairs, each training data pair comprising a spectrum obtained from a sample and an indication of a specific analytical scheme for the sample, the training data comprising the corresponding spectrum of each sample type obtained from samples of each sample type; as well as In response to the corresponding spectrum of the training data pair, the parameters of the machine learning model are adjusted to reduce the difference between the indication of the training data pair and the indication output by the machine learning model.

11. A method for obtaining training data for training a machine learning model, the method comprising: For each of a plurality of specific analytical protocols, a baseline spectrum is obtained for each of a plurality of samples of the type to which the specific analytical protocol is applicable, the spectrum being obtained using a baseline analytical protocol that is the same for all of the plurality of sample types; The machine learning model stores input and output data pairs, where each input pair includes the obtained baseline spectrum and each output pair includes an indication of the analytical protocol for the corresponding sample.

12. One or more non-transitory computer-readable media having instructions thereon, which, when executed by one or more processing devices of a scientific instrument support system, cause the scientific instrument support device to perform the method according to any of the preceding claims.

13. A scientific instrument support system, the scientific instrument support system comprising: One or more processors; One or more memories thereon store computer-readable instructions configured to cause the one or more processors to perform operations including the method according to any one of claims 1 to 11.

14. A scientific instrument comprising the scientific support system of claim 13, and further comprising a spectrometer for recording spectra.

Images