A method and system for identifying raw and auxiliary materials in preparations based on multi-device combination

Abstract

The present invention discloses a method and system for identifying raw materials and auxiliary materials in preparations based on the combined use of multiple devices. The method includes performing a cryo-polishing process on a cross section of a sample to be tested to obtain the cross section to be tested; observing the cross section to be tested using an electron microscope to obtain an electron microscope image of the cross section to be tested, and determining a region to be tested on the cross section to be tested; scanning the region to be tested using an energy spectrometer to obtain an energy spectrum image; judging whether the sample contains characteristic elements based on the electron microscope image and the energy spectrum image, and if so, identifying the characteristic elements based on the energy spectrum image; judging whether the sample contains components that match preset morphological features based on the electron microscope image and the energy spectrum image, and if so, analyzing and processing the electron microscope image using a pre-built convolutional neural network model to identify the components that match the preset morphological features. The present invention can more accurately and quickly identify the components in the sample to be analyzed.

Description

Technical Field

[0001] The present invention relates to the technical field of preparation analysis, and in particular to a method and system for identifying raw and auxiliary materials in preparations based on the combined use of multiple devices. Background Art

[0002] In the field of formulation analysis, traditional chromatography and mass spectrometry methods occupy an important position. Liquid chromatography (HPLC) and gas chromatography (GC) use the interaction between the sample and the stationary phase to achieve the separation of components. They can effectively process complex mixtures and help researchers quantitatively analyze the purity and concentration of drug ingredients. At the same time, mass spectrometry provides molecular weight and structural information by ionizing the sample and measuring the mass-to-charge ratio of the ions. Usually, these two techniques are used in combination (such as LC-MS or GC-MS) to accurately identify and quantify low-concentration components in complex matrices. However, chromatography and mass spectrometry usually require sample extraction or dissolution, which may cause damage to certain sensitive samples, and the analysis time is relatively long, so there are certain limitations in rapid and non-destructive analysis.

[0003] The disclosure of the above background technology content is only used to assist in understanding the inventive concept and technical solution of the present invention. It does not necessarily belong to the prior art of the present application, nor does it necessarily provide technical guidance. In the absence of clear evidence that the above content has been disclosed before the filing date of the present application, the above background technology should not be used to evaluate the novelty and creativity of the present application. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and monitoring system for identifying raw materials and auxiliary materials in preparations based on the combined use of multiple devices, which can achieve more accurate and rapid identification of components in samples to be analyzed.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0006] A method for identifying raw materials and auxiliary materials in a preparation based on the combined use of multiple devices, characterized by comprising the following steps:

[0007] Performing cryo-polishing on a cross section of the sample to be tested to obtain the cross section to be tested;

[0008] Observing the cross section to be tested using an electron microscope to obtain an electron microscope image of the cross section to be tested, and determining a region to be tested on the cross section to be tested;

[0009] Scanning the area to be tested with an energy spectrometer to obtain an energy spectrum image;

[0010] Determining whether the test sample contains characteristic elements based on the electron microscope image and the energy spectrum image, and if so, identifying the characteristic elements based on the energy spectrum image;

[0011] Based on the electron microscope image and the energy spectrum image, it is determined whether the test sample contains components that match the preset morphological features. If there are components that match the preset morphological features, the electron microscope image is analyzed and processed using a pre-built convolutional neural network model to identify the components that match the preset morphological features.

[0012] Furthermore, any one of the above technical solutions or a combination of multiple technical solutions may further include the following steps:

[0013] If the test sample contains neither characteristic elements nor components matching the preset morphological features, Raman spectroscopy is used to perform Raman spot collection on the test area to obtain a Raman spectrum.

[0014] The components of the test sample are obtained according to the Raman spectrum analysis.

[0015] Further, based on any one of the technical solutions or a combination of multiple technical solutions described above, in the process of analyzing the components of the test sample according to the Raman spectrum, if there is a component with similar characteristic peaks, it is determined to be the target component, and the presence of the component with similar characteristic peaks is determined by the following method: the difference between the peak values ​​of at least two characteristic peaks is within a preset peak difference threshold range;

[0016] The atomic force microscope analysis method is used to obtain the microstructure information of the area to be tested and the target component is determined in combination with the mechanical properties of the target component.

[0017] Furthermore, based on any one of the technical solutions or a combination of multiple technical solutions described above, it also includes using a pre-built deep learning model to analyze and process the energy spectrum image to identify the characteristic elements, and the deep learning model is configured to perform feature extraction, noise reduction and image enhancement on the energy spectrum image to identify and display the characteristic elements.

[0018] Furthermore, based on any one of the above technical solutions or a combination of multiple technical solutions, the electron microscope image is analyzed and processed using a pre-built convolutional neural network model to identify components that match preset morphological features, including:

[0019] Preprocessing the electron microscope image to obtain a processed electron microscope image, wherein the preprocessing includes one or more of denoising, normalization, and size adjustment;

[0020] The processed electron microscope image is input into the convolutional neural network model, and the convolutional neural network model predicts and segments the components matching the preset morphological features in the processed electron microscope image based on the morphological features of multiple polymer compounds that have been pre-labeled and learned to determine the components of each matching the preset morphological feature.

[0021] Furthermore, based on any one of the above technical solutions or a combination of multiple technical solutions, the convolutional neural network model is pre-constructed in the following manner:

[0022] Obtaining a learning sample set, the learning sample set comprising a plurality of learning samples, each learning sample comprising one or more electron microscope image samples corresponding to a known component, the known component having regular morphological features, and each electron microscope image sample being annotated with its corresponding known component;

[0023] A basic neural network model is determined, the learning sample set is input into the basic neural network model, and the basic neural network model is trained using a preset loss function to obtain the convolutional neural network model.

[0024] Furthermore, based on any one of the above technical solutions or a combination of multiple technical solutions, predicting and segmenting the components matching the preset morphological features in the processed electron microscope image to determine the components matching the preset morphological features includes:

[0025] Determining the substance corresponding to each component matching the preset morphological features;

[0026] Segmented images corresponding to the components matching the preset morphological features are obtained, and each segmented image is configured to display a component matching the preset morphological features.

[0027] Furthermore, any one of the above technical solutions or a combination of multiple technical solutions further includes:

[0028] The electron microscope image of the cross section to be tested obtained by observing the cross section to be tested with an electron microscope is configured as a first electron microscope image;

[0029] Determining whether the test sample has morphological features based on the electron microscope image and the energy spectrum image, and if so, analyzing and processing the electron microscope image using a pre-built convolutional neural network model to identify components matching the preset morphological features, including the following steps:

[0030] Observing the area to be tested using an electron microscope to obtain a second electron microscope image of the area to be tested, wherein the resolution of the second electron microscope image is greater than that of the first electron microscope image;

[0031] The pre-built convolutional neural network model is used to analyze and process the second electron microscope image to identify components that match the preset morphological features.

[0032] Furthermore, based on any one of the above technical solutions or a combination of multiple technical solutions, for different components in the test sample having matching morphological features, after analyzing and processing the electron microscope image using a pre-built convolutional neural network model to identify components matching the preset morphological features, the method further includes:

[0033] The energy spectrum image is further analyzed and processed using a pre-built deep learning model to identify components with similar morphological features.

[0034] Furthermore, based on any one of the above technical solutions or a combination of multiple technical solutions, whether the test sample has a component matching the preset morphological characteristics is determined by the following method:

[0035] Predetermining a plurality of components with regular morphological features to construct a morphological feature component library, wherein the morphological feature component library includes a plurality of components and a reference morphological feature corresponding to each component, wherein the reference morphological feature is one or more;

[0036] The electron microscope image and the energy spectrum image are analyzed and compared with the morphological feature component library. If the morphological features of the components to be analyzed in the electron microscope image and / or the energy spectrum image match the reference morphological features, it is determined whether the test sample contains components that match the preset morphological features.

[0037] Furthermore, based on any one of the above-mentioned technical solutions or a combination of multiple technical solutions, the method for identifying raw materials and auxiliary materials in preparations based on the combination of multiple devices is applicable to analysis objects including at least: preparations, polymer materials and semiconductor materials; and / or,

[0038] The following steps are also included:

[0039] If the test sample contains neither characteristic elements nor components matching the preset morphological features, the to-be-tested area is analyzed using X-ray diffraction analysis to obtain the components of the test sample.

[0040] According to another aspect of the present invention, a system for identifying raw materials and auxiliary materials in preparations based on the combined use of multiple devices is provided, including an electron microscope, an energy dispersive spectrometer, and a processor;

[0041] The system for identifying raw materials and excipients in preparations is configured to identify raw materials and excipients in preparations in the following ways:

[0042] A cross section of the sample to be tested is subjected to cryo-polishing treatment in advance to obtain the cross section to be tested;

[0043] Observing the cross section to be tested using the electron microscope to obtain an electron microscope image of the cross section to be tested, and determining a region to be tested on the cross section to be tested;

[0044] Scanning the area to be tested using the energy spectrometer to obtain an energy spectrum image;

[0045] The processor determines whether the test sample contains characteristic elements based on the electron microscope image and the energy spectrum image, and if so, identifies the characteristic elements based on the energy spectrum image;

[0046] The processor determines whether the test sample contains components that match preset morphological features based on the electron microscope image and the energy spectrum image. If there are components that match the preset morphological features, the electron microscope image is analyzed and processed using a pre-built convolutional neural network model to identify the components that match the preset morphological features.

[0047] Furthermore, any one of the above technical solutions or a combination of multiple technical solutions further includes a Raman spectrometer;

[0048] The system for identifying raw materials and excipients in preparations is further configured to identify raw materials and excipients in preparations in the following manner:

[0049] If the test sample contains neither characteristic elements nor components matching the preset morphological features, Raman spectroscopy is used to perform Raman spot collection on the test area to obtain a Raman spectrum.

[0050] The processor obtains the components of the test sample according to the Raman spectrum analysis.

[0051] Furthermore, any one of the above technical solutions or a combination of multiple technical solutions further includes an atomic force microscope;

[0052] The system for identifying raw materials and excipients in preparations is further configured to identify raw materials and excipients in preparations in the following manner:

[0053] In the process of analyzing the components of the test sample according to the Raman spectrum, if there is a component with similar characteristic peaks, it is determined to be the target component, and the presence of the component with similar characteristic peaks is determined by the following method: the difference between the peak values ​​of at least two characteristic peaks is within a preset peak difference threshold range;

[0054] The atomic force microscope analysis method is used to obtain the microstructure information of the area to be tested and the target component is determined in combination with the mechanical properties of the target component.

[0055] The beneficial effects brought about by the technical solution provided by the present invention are as follows:

[0056] a. By selecting electron microscopy analysis, energy spectrum analysis, or using a convolutional neural network model to analyze components with characteristic morphologies based on the different components in the sample to be tested, the present invention can significantly shorten the analysis time of each finished sample and obtain more accurate, larger-scale, and higher-resolution analysis results in a shorter time;

[0057] b. The present invention further combines energy spectrum analysis with analysis of different components in the sample with matching / similar / close morphological features to effectively distinguish different components with similar morphological features, thereby avoiding errors or omissions in the analysis of components with similar morphological features;

[0058] c. The method for identifying raw and auxiliary materials in preparations based on the combined use of multiple devices proposed in the present invention is not only suitable for identifying raw and auxiliary materials in preparations, but can also be used to identify polymer materials and semiconductor materials. It can be applied to various industries such as polymer materials and semiconductors and has a wide range of uses. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments recorded in this application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0060] Figure 1 A general flow chart of a method for identifying raw materials and excipients in a preparation provided by an exemplary embodiment of the present invention;

[0061] Figure 2 A partial flow chart of a method for identifying raw materials and excipients in a preparation provided by an exemplary embodiment of the present invention;

[0062] Figure 3 A first electron microscope image of a sample to be tested provided in accordance with an exemplary embodiment of the present invention;

[0063] Figure 4 A second electron microscope image of a sample to be tested provided in accordance with an exemplary embodiment of the present invention;

[0064] Figure 5 An energy spectrum image of the S element in a sample to be tested provided by an exemplary embodiment of the present invention;

[0065] Figure 6 An energy spectrum image of the Mg element in a sample to be tested provided by an exemplary embodiment of the present invention;

[0066] Figure 7An image of the distribution of S elements in a sample to be tested provided by an exemplary embodiment of the present invention;

[0067] Figure 8 An image of the Mg element distribution results in a sample to be tested provided by an exemplary embodiment of the present invention;

[0068] Figure 9 Reference images of the topographical features of corn starch in a formulation provided for an exemplary embodiment of the present invention;

[0069] Figure 10 Reference images of the morphological features of low-substituted hydroxypropyl cellulose in a formulation provided for an exemplary embodiment of the present invention;

[0070] Figure 11 A reference image of the morphological characteristics of lactose in a formulation provided for an exemplary embodiment of the present invention;

[0071] Figure 12 A segmented image of corn starch in a breprazol sample provided as an exemplary embodiment of the present invention;

[0072] Figure 13 A segmented image of low-substituted hydroxypropyl cellulose in a breprazol sample provided as an exemplary embodiment of the present invention;

[0073] Figure 14 A segmented image of lactose in a breprazol sample provided as an exemplary embodiment of the present invention;

[0074] Figure 15 A schematic diagram of a candidate region of a microcrystal to be Raman confirmed, provided for an exemplary embodiment of the present invention;

[0075] Figure 16 A Raman spectrum of a substance at a C element collection point provided by an exemplary embodiment of the present invention;

[0076] Figure 17 A diagram showing analysis results of a brepizole sample provided by an exemplary embodiment of the present invention;

[0077] Figure 18 A comparison chart of the analysis results of the present method and the Raman method for the same sample provided as an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0078] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0079] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, apparatus, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0080] In one embodiment of the present invention, a method for identifying raw materials and auxiliary materials in a preparation based on the combination of multiple devices is provided. Figure 1 , the method comprises the following steps:

[0081] Performing cryo-polishing on a cross section of the sample to be tested to obtain the cross section to be tested;

[0082] Observing the cross section to be tested using an electron microscope to obtain an electron microscope image of the cross section to be tested, and determining a region to be tested on the cross section to be tested;

[0083] Scanning the area to be tested with an energy spectrometer to obtain an energy spectrum image;

[0084] Determining whether the test sample contains characteristic elements based on the electron microscope image and the energy spectrum image, and if so, identifying the characteristic elements based on the energy spectrum image;

[0085] Based on the electron microscope image and the energy spectrum image, it is determined whether the test sample contains components that match the preset morphological features. If there are components that match the preset morphological features, the electron microscope image is analyzed and processed using a pre-built convolutional neural network model to identify the components that match the preset morphological features.

[0086] There are multiple ways to determine whether the test sample contains components that match the preset morphological features. One way is to use manual judgment based on experience; another way is to use automatic judgment through the following methods:

[0087] Predetermining a plurality of components with regular morphological features to construct a morphological feature component library, wherein the morphological feature component library includes a plurality of components and a reference morphological feature corresponding to each component, wherein the reference morphological feature is one or more;

[0088] The electron microscope image and the energy spectrum image are analyzed and compared with the morphological feature component library. If the morphological features of the components to be analyzed in the electron microscope image and / or the energy spectrum image match the reference morphological features, it is determined whether the test sample contains components that match the preset morphological features.

[0089] Preferably, the analysis method for components having matching preset morphological features (also referred to as components having morphological features) comprises the following steps:

[0090] Preprocessing the electron microscope image to obtain a processed electron microscope image, wherein the preprocessing includes one or more of denoising, normalization, and size adjustment;

[0091] The processed electron microscope image is input into the convolutional neural network model. The convolutional neural network model predicts and segments the components matching the preset morphological features in the processed electron microscope image based on the morphological features of multiple polymer compounds that have been pre-labeled and learned to determine the components of each matching the preset morphological feature, including: determining the substance corresponding to each component matching the preset morphological feature; obtaining the segmented image corresponding to each component matching the preset morphological feature, each segmented image being configured to display a component matching the preset morphological feature.

[0092] In one embodiment of the present invention, the convolutional neural network model is pre-built in the following manner:

[0093] Obtaining a learning sample set, the learning sample set comprising a plurality of learning samples, each learning sample comprising one or more electron microscope image samples corresponding to a known component, the known component having regular morphological features, and each electron microscope image sample being annotated with its corresponding known component;

[0094] A basic neural network model is determined, the learning sample set is input into the basic neural network model, and the basic neural network model is trained using a preset loss function to obtain the convolutional neural network model.

[0095] Regarding the different components in the test sample whose morphological features match / similar / close, it should be noted that, in the present application, different components whose morphological features match / similar / close refer to two different components with similar but not completely identical morphological features. Specifically, the shapes of the morphological profiles of the two components are the same, but the sizes are different, or the morphological sizes of the two components are close (the size difference is within a certain range), and the morphological profiles of the two components are not much different. Specifically, for example, if the degree of matching of the morphological profiles of the two components is greater than a certain first preset value (for example, 80% or 90%), it is judged that the morphological features of the two components are matched / similar / close. At this time, if only a pre-built convolutional neural network model is used to analyze and process the electron microscope image to identify components that match the preset morphological features, errors may occur or components may be missed.

[0096] Therefore, in one embodiment of the present invention, for different components in the test sample having matching / similar / close morphological features, after analyzing and processing the electron microscope image using a pre-built convolutional neural network model to identify components matching the preset morphological features, the following steps are further included:

[0097] The energy spectrum image is further analyzed and processed using a pre-built deep learning model to identify components with similar morphological features. Thus, by observing the elemental composition of different regions through energy spectrum analysis (EDS), it is possible to efficiently and accurately distinguish components with similar morphological features in the sample. For example, some particles may be very similar in morphology, but their constituent elements may be different, or the distribution of elements in different regions may be different. By comparing the distribution of these elements, subtle differences between substances can be discovered, helping to determine whether they contain other unique elements or whether there are different distribution patterns of elements, thereby further confirming the type and structure of the compound.

[0098] In the combined electron microscopy and energy spectrum analysis process, the sample is imaged using an electron microscope to obtain high-resolution images to observe its microstructure. During this process, sample preparation and imaging conditions are optimized to obtain clear and detailed images. Next, the acquired electron microscopy images are de-noised and enhanced to improve overall image quality. This step includes the use of techniques such as filtering and histogram equalization to enhance the sample's visual features and improve analysis accuracy. Subsequently, the sample is analyzed using an energy spectrum analyzer to obtain elemental distribution information. This analysis determines the relative concentration of each element and generates a corresponding elemental map for a better understanding of the sample's composition. During this stage, the electron microscopy image is registered with the energy spectrum data to analyze the distribution of elements within the microstructure. Accurate registration of the two data sets in the same coordinate system is crucial for effective joint analysis. Finally, the electron microscopy image and energy spectrum data are combined to generate a detailed analysis report summarizing the sample's microstructure and elemental composition. This report provides an important basis for subsequent research and further explores the sample's characteristics and applications.

[0099] In this embodiment, the method further includes the following steps:

[0100] If the test sample contains neither characteristic elements nor components matching the preset morphological features, Raman spectroscopy is used to perform Raman spot collection on the test area to obtain a Raman spectrum.

[0101] The components of the test sample are obtained according to the Raman spectrum analysis.

[0102] In the absence of obvious morphological features and characteristic elements, Raman spectroscopy is an effective non-destructive testing technology. By performing Raman spot collection on the target area, the Raman spectrum of the area is obtained. Each compound has unique molecular vibration characteristics, which are manifested as specific Raman scattering peaks. These characteristic peaks can be used to distinguish different compounds. By comparing the obtained spectrum with the established Raman database, it is possible to accurately determine which compound the sample belongs to. This method is particularly suitable for analyzing complex samples containing multiple substances and can provide high-resolution molecular information. However, it is time-consuming to use Raman spectroscopy to analyze sample components. If all components of the test sample are analyzed using Raman spectroscopy, it will take a long time. See Figure 18 In a specific embodiment of the present invention, the method provided by the present invention is based on the method for identifying raw materials and auxiliary materials in preparations using multiple devices, such as Figure 18 As shown on the left, the combined analysis method of electron microscopy + energy spectrum + Raman spectroscopy can determine the components of the sample to be tested in 2 hours, compared with the existing Raman method which takes 13 hours to analyze the components of the sample to be tested (as shown in Figure 2). Figure 18This solution greatly shortens the analysis time, and the method obtains 10mm through 2 hours of test analysis. 2 The analysis results of the test range +300nm resolution are 5mm compared to the existing single Raman method. 2 The analysis results of the test range + 11μm resolution show that this method not only greatly shortens the analysis time, but also can obtain analysis results with a larger test range and higher resolution.

[0103] Electron microscopy can provide nanometer-level spatial resolution, allowing researchers to deeply observe the microstructure and particle morphology of samples, thereby effectively understanding the distribution of particles in pharmaceutical preparations and their interactions. At the same time, energy spectrum technology can obtain elemental composition information of samples in real time, and fully grasp the chemical composition and distribution of materials. The method for identifying raw and auxiliary materials in preparations based on the combined use of multiple devices proposed in the present invention, combined with the analysis method of electron microscopy, energy spectrum and Raman spectroscopy, has shown significant advantages in preparation research. Compared with traditional chromatography and mass spectrometry methods, the combined method proposed in the present invention not only reduces the complexity of sample processing, but also allows non-destructive analysis, especially the application of Raman spectroscopy, which enables researchers to obtain results quickly without damaging the sample. This rapidity is particularly suitable for online monitoring and quality control. The combined use of these technologies can more comprehensively reveal the characteristics of the preparation and provide a richer perspective for in-depth analysis.

[0104] The shortcomings of existing methods also lie in their single-mindedness and limitations. Traditional analytical methods typically rely on a single technique, such as chemical analysis, spectroscopy, or morphological analysis, but these methods often fail to comprehensively and accurately capture all the information in complex samples. For example, some substances may lack both distinct elemental and morphological characteristics, and existing methods are often insufficient to extract valuable data from such substances. Existing methods typically select analytical methods based on known material properties, lacking flexibility and being unable to automatically adjust analytical strategies based on the sample's specific characteristics. This rigid approach often fails to provide the most appropriate analytical path for different substance types. In particular, for substances lacking both characteristic elemental and morphological features, existing technologies may not provide effective analytical tools, resulting in inaccurate or incomplete results. Existing methods are inefficient when processing complex samples, and their accuracy is often unsatisfactory. Traditional image processing or simple chemical analysis techniques, especially for complex substances such as high-molecular-weight hydrocarbons and microcrystalline cellulose, are prone to misjudgment and prolonged analysis times, compromising data reliability and efficiency. However, in the analysis of multi-component samples, existing methods find it difficult to accurately distinguish and quantify each component, especially when there are overlapping characteristics between substances, and the extraction and identification of information often presents great difficulties.

[0105] The method for identifying raw and auxiliary materials in preparations based on the combined use of multiple devices proposed in the present invention can effectively solve the problems of existing methods such as reliance on a single analytical technology, lack of flexibility and accuracy, and can more efficiently and accurately extract key component information of the samples to be analyzed when processing complex or multi-component samples.

[0106] In one embodiment of the present invention, see Figure 2 In the process of analyzing the components of the test sample according to the Raman spectrum, if there is a component with similar characteristic peaks, it is determined to be the target component, and the presence of the component with similar characteristic peaks is determined by the following method: the difference between the peak values ​​of at least two characteristic peaks is within a preset peak difference threshold range;

[0107] The atomic force microscope analysis method is used to obtain the microstructure information of the area to be tested and the target component is determined in combination with the mechanical properties of the target component.

[0108] Among them, atomic force microscopy (AFM) scans the surface morphology of the sample and obtains its microstructural information. Although this method is mainly used for surface morphology analysis, in some cases, combined with its mechanical properties (such as nanomechanics, friction, etc.) analysis, it can be used to distinguish different compounds. By comparing the mechanical responses of different substances, the material composition of the sample can be inferred. Although this method cannot directly provide all chemical composition information, it can help further screen compounds in some cases and provide clues for subsequent analysis.

[0109] It should be noted that if the test sample contains neither characteristic elements nor components that match the preset morphological features, in addition to using X-ray diffraction analysis to analyze the area to be tested to obtain the components of the test sample, X-ray diffraction analysis can also be used to analyze the area to be tested to obtain the components of the test sample. X-ray diffraction (XRD) is a powerful material analysis technology, mainly used to analyze the crystal structure of substances. The diffraction pattern of each compound or crystal has its own unique fingerprint characteristics. By measuring the diffraction intensity of the sample at different angles, the crystal structure information of the compound can be obtained. Although XRD is more suitable for crystalline materials, it can provide information such as the lattice constant and interplanar spacing of the substance to help identify the type of substance. By comparing the known diffraction patterns in the database, the chemical composition of the sample can be inferred, and compound identification can be achieved.

[0110] In a specific embodiment of the present invention, a method for rapidly analyzing the composition and distribution of breprazol is provided, which specifically comprises the following steps:

[0111] (1) Take a brepazine test sample and perform cryo-polishing on the test sample along the largest cross section using a soft ion beam cryo-polishing instrument. Then observe the cross section of the test sample after polishing using an electron microscope to obtain a complete first electron microscope image of the cross section, such as Figure 3 As shown, a test area is determined in the cross section as Figure 3 The middle rectangle shows the area to be tested.

[0112] (2) The formula ingredients of the brepizole sample are brepizole, magnesium stearate, lactose, low-substituted hydroxypropyl cellulose, corn starch, and microcrystalline cellulose.

[0113] In formulating subsequent treatment methods, this application will conduct the following analysis based on the above prescription:

[0114] For brepazole and magnesium stearate with characteristic elements, characteristic element analysis was used to extract effective information;

[0115] For high molecular weight hydrocarbon compounds such as lactose, low-substituted hydroxypropyl cellulose, and corn starch that have morphological features but lack characteristic elements, deep learning networks will be used for processing;

[0116] For microcrystalline cellulose that has neither morphological features nor characteristic elements, Raman spectroscopy is used for analysis to obtain more comprehensive material information. Existing methods may not select the most appropriate analysis method based on the different characteristics of the substance (such as whether it has characteristic elements, morphological features, etc.). Our method is more flexible and efficient by accurately classifying and selecting appropriate analysis techniques (elemental analysis, deep learning, Raman spectroscopy) for different characteristics. For substances that have no characteristic elements but have morphological characteristics, existing methods may rely on traditional image processing or chemical analysis techniques, while our method uses deep learning networks to learn morphological features from data, which is more intelligent and accurate. For substances that have no morphological features and characteristic elements (such as microcrystalline cellulose), existing methods may be difficult to analyze effectively, and this method innovatively uses Raman spectroscopy to supplement the analysis, which can obtain more comprehensive information.

[0117] (3) Among them, brepazole and magnesium stearate have the characteristic elements S and Mg.

[0118] The characteristic elements generally refer to chemical elements that are unique to certain substances or ingredients (substances that are not found in other ingredients), which can be used as markers for identification and analysis. In drug or material analysis, characteristic elements are the key to distinguishing or identifying active ingredients (such as the main ingredients of drugs) or excipients (such as additives, fillers). For example, if a certain active ingredient or excipient in a prescription contains iron (Fe), while other ingredients do not contain iron, then iron can be regarded as the characteristic element of this active ingredient. By detecting these characteristic elements, we can accurately locate or identify the different ingredients in the drug.

[0119] Under an electron microscope, an energy dispersive spectrometer used in conjunction with the electron microscope is used to scan the area to be tested to obtain a second electron microscope image of the area. Figure 4 As shown, the resolution of the second electron microscope image is greater than that of the first electron microscope image, and the distribution of all characteristic elements in the area is obtained, including Figure 5 S elements in the selected region shown, and Figure 6 Mg element in the selected area shown.

[0120] The energy spectrum image is analyzed and processed using a pre-built deep learning model to identify the characteristic elements, that is, features are extracted from its neighborhood (for example, pixels within a local window) through a multi-layer convolutional neural network, and features are extracted from the energy spectrum image. De-noising is performed by utilizing the structural information of the image itself, thereby effectively removing noise from the input image. Then, image enhancement technology is used to generate the following image: Figure 7 The S element distribution diagram shown and Figure 8 Mg element distribution diagram shown.

[0121] (4) Among them, the carbon, hydrogen and oxygen polymer compounds lactose, low-substituted hydroxypropyl cellulose and corn starch have morphological characteristics but lack characteristic elements. The sample is imaged using a high-resolution electron microscope to obtain a high-precision image of the sample surface morphology, namely the second electron microscope image, such as Figure 2 shown.

[0122] Preprocess the acquired second electron microscope image, including denoising, normalization and resizing. Load the pre-trained model, which has learned the morphological characteristics of hydrocarbon polymer compounds based on the labeled data. Figures 9 to 11 These are the reference images of the morphological characteristics of corn starch, low-substituted hydroxypropyl cellulose, and lactose in the preparations.

[0123] The pre-processed second electron microscope image is input into the trained model for prediction and segmentation, and the prediction and segmentation results are output in the form of data and images. Figures 12 to 14 ,in, Figure 12 is the segmentation image of corn starch, Figure 13is the segmentation image of low-substituted hydroxypropyl cellulose. Figure 14 This is the segmentation image of lactose.

[0124] (5) For microcrystalline cellulose that has neither morphological features nor characteristic elements, Raman spectroscopy is used for analysis. Within the area intercepted in step (3), at the same coordinates, a confocal micro-Raman imaging system is used to detect the Raman spectrum of the material at the collection point to obtain the Raman spectrum of the area. Each compound has unique molecular vibration characteristics, which are manifested as specific Raman scattering peaks. These characteristic peaks can be used to distinguish different compounds. By comparing the obtained spectrum with the established Raman database, it is possible to accurately determine which compound the sample belongs to. Figure 16 is the C element collection point ( Figure 15 ) is a Raman spectrum of a substance corresponding to microcrystalline cellulose.

[0125] (6) If different compounds or components contain the same characteristic element, the concentration and distribution of the element can be evaluated in detail by energy spectrum analysis (such as energy dispersive X-ray spectroscopy (EDS)). Energy spectrum analysis can provide information on the local distribution and concentration of elements. By observing the distribution of elements in different regions, the role of the element in the sample can be inferred. For example, active ingredients usually have a higher concentration in pharmaceutical preparations and may be concentrated in a specific area, while excipients may show a more uniform distribution. By analyzing the concentration differences and distribution patterns of elements, different components can be effectively distinguished and the chemical composition of the sample can be further confirmed.

[0126] (7) Figure 17 This is the final analysis result diagram, where the colors of different areas represent different active ingredients. This diagram can be used to determine the content, particle size and other data of different active ingredients.

[0127] The method for identifying raw and auxiliary materials in preparations based on the combined use of multiple devices provided by the present invention is not only suitable for identifying raw and auxiliary materials in preparations, but can also be used to identify polymer materials and semiconductor materials. It can be applied to various industries such as polymer materials and semiconductors and has a wide range of uses.

[0128] In one embodiment of the present invention, a system for identifying raw materials and auxiliary materials in preparations based on the combined use of multiple devices is provided, including an electron microscope, an energy spectrometer, a Raman spectrometer, an atomic force microscope and a processor. The system can correlate the electron microscope, energy spectrometer, Raman spectrometer, atomic force microscope and other data to analyze the sample to be analyzed, and identify the components of the sample to be analyzed more accurately and quickly.

[0129] The system for identifying raw materials and excipients in preparations is configured to identify raw materials and excipients in preparations in the following ways:

[0130] A cross section of the sample to be tested is subjected to cryo-polishing treatment in advance to obtain the cross section to be tested;

[0131] Observing the cross section to be tested using the electron microscope to obtain an electron microscope image of the cross section to be tested, and determining a region to be tested on the cross section to be tested;

[0132] Scanning the area to be tested using the energy spectrometer to obtain an energy spectrum image;

[0133] The processor determines whether the test sample contains characteristic elements based on the electron microscope image and the energy spectrum image, and if so, identifies the characteristic elements based on the energy spectrum image;

[0134] The processor determines whether the test sample contains components that match preset morphological features based on the electron microscope image and the energy spectrum image. If there are components that match the preset morphological features, the electron microscope image is analyzed and processed using a pre-built convolutional neural network model to identify the components that match the preset morphological features.

[0135] If the test sample contains neither characteristic elements nor components matching the preset morphological features, Raman spectroscopy is used to perform Raman dot acquisition on the test area to obtain a Raman spectrum map; the processor obtains the composition of the test sample based on the Raman spectrum map.

[0136] In the process of analyzing the components of the test sample according to the Raman spectrum, if there is a component with similar characteristic peaks, it is determined to be the target component, and the presence of the component with similar characteristic peaks is determined by the following method: the difference between the peak values ​​of at least two characteristic peaks is within a preset peak difference threshold range;

[0137] The atomic force microscope analysis method is used to obtain the microstructure information of the area to be tested and the target component is determined in combination with the mechanical properties of the target component.

[0138] In another embodiment of the present invention, a system for identifying raw materials and auxiliary materials in preparations based on the combined use of multiple devices is provided, including an electron microscope, an energy spectrometer, an X-ray diffraction analysis system, an atomic force microscope and a processor. In this embodiment, an X-ray diffraction analysis system is used instead of a Raman spectrometer, and the electron microscope, energy spectrometer, diffraction pattern, atomic force microscope and other data can be correlated to analyze the sample to be analyzed, so as to more accurately and quickly identify the components of the sample to be analyzed.

[0139] It should be noted that the above-mentioned embodiment of the system for identifying raw and auxiliary materials in preparations based on the combination of multiple devices and the embodiment of the method for identifying raw and auxiliary materials in preparations based on the combination of multiple devices are based on the same inventive concept, and all the contents of the embodiment of the method for identifying raw and auxiliary materials in preparations based on the combination of multiple devices are incorporated into the embodiment of the system for identifying raw and auxiliary materials in preparations based on the combination of multiple devices by reference.

[0140] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0141] The above is only a specific implementation method of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for identifying raw materials and auxiliary materials in preparations based on the combination of multiple devices, characterized in that: The following steps are involved: Performing cryo-polishing on a cross section of the sample to be tested to obtain the cross section to be tested; Observing the cross section to be tested using an electron microscope to obtain a first electron microscope image of the cross section to be tested, and determining a region to be tested on the cross section to be tested; Scanning the area to be tested with an energy spectrometer to obtain an energy spectrum image; Determining whether the test sample contains characteristic elements based on the first electron microscope image and the energy spectrum image, and if so, identifying the characteristic elements based on the energy spectrum image; Determining whether the test sample contains a component matching a preset morphological feature based on the first electron microscope image and the energy spectrum image; if so, observing the area to be tested using an electron microscope to obtain a second electron microscope image of the area to be tested, wherein the resolution of the second electron microscope image is greater than that of the first electron microscope image; and analyzing and processing the second electron microscope image using a pre-constructed convolutional neural network model to identify the component matching the preset morphological feature; For different components in the test sample having matching morphological features, after analyzing and processing the second electron microscope image using a pre-built convolutional neural network model to identify components matching preset morphological features, further analyzing and processing the energy spectrum image using a pre-built deep learning model to identify components with similar morphological features; If the test sample contains neither characteristic elements nor components matching the preset morphological features, Raman spectroscopy is used to perform Raman spot collection on the test area to obtain a Raman spectrum map, and the composition of the test sample is obtained based on the Raman spectrum map.

2. The method for identifying raw materials and auxiliary materials in preparations based on multi-device combination according to claim 1, characterized in that: In the process of analyzing the components of the test sample according to the Raman spectrum, if there is a component with similar characteristic peaks, it is determined to be a target component, and the presence of the component with similar characteristic peaks is determined in the following manner: the difference between the peak values ​​of at least two characteristic peaks is within a preset peak difference threshold range; The atomic force microscope analysis method is used to obtain the microstructure information of the area to be tested and the target component is determined in combination with the mechanical properties of the target component.

3. The method for identifying raw materials and auxiliary materials in preparations based on multi-device combination according to claim 1, characterized in that: It also includes using a pre-built deep learning model to analyze and process the energy spectrum image to identify the characteristic elements, and the deep learning model is configured to perform feature extraction, noise reduction and image enhancement on the energy spectrum image to identify and display the characteristic elements.

4. The method for identifying raw materials and auxiliary materials in preparations based on multi-device combination according to claim 1, characterized in that: Analyzing and processing the second electron microscope image using a pre-built convolutional neural network model to identify components matching preset morphological features, including: Preprocessing the second electron microscope image to obtain a processed second electron microscope image, wherein the preprocessing includes one or more of denoising, normalization, and size adjustment; The processed second electron microscope image is input into the convolutional neural network model, and the convolutional neural network model predicts and segments the components matching the preset morphological features in the processed second electron microscope image based on the morphological features of multiple polymer compounds that have been pre-labeled and learned to determine the components of each matching the preset morphological feature.

5. The method for identifying raw materials and auxiliary materials in preparations based on multi-device combination according to claim 4, characterized in that: The convolutional neural network model is pre-built in the following way: Obtaining a learning sample set, the learning sample set comprising a plurality of learning samples, each learning sample comprising one or more electron microscope image samples corresponding to a known component, the known component having regular morphological features, and each electron microscope image sample being annotated with its corresponding known component; A basic neural network model is determined, the learning sample set is input into the basic neural network model, and the basic neural network model is trained using a preset loss function to obtain the convolutional neural network model.

6. The method for identifying raw materials and auxiliary materials in preparations based on multi-device combination according to claim 4, characterized in that: Predicting and segmenting the components matching the preset topographical features in the processed second electron microscope image to determine the components matching the preset topographical features, including: Determining the substance corresponding to each component matching the preset morphological features; Segmented images corresponding to the components matching the preset morphological features are obtained, and each segmented image is configured to display a component matching the preset morphological features.

7. The method for identifying raw materials and auxiliary materials in preparations based on multi-device combination according to claim 1, characterized in that: Whether the test sample has components matching the preset morphological features is determined by the following method: Predetermining a plurality of components with regular morphological features to construct a morphological feature component library, wherein the morphological feature component library includes a plurality of components and a reference morphological feature corresponding to each component, wherein the reference morphological feature is one or more; The first electron microscope image and the energy spectrum image are analyzed and compared with the morphological feature component library. If the morphological features of the components to be analyzed in the first electron microscope image and / or the energy spectrum image match the reference morphological features, it is determined that the test sample contains components that match the preset morphological features.

8. The method for identifying raw materials and auxiliary materials in preparations based on multi-device combination according to claim 1, characterized in that: The analysis objects applicable to the identification method of raw materials and auxiliary materials in preparations based on the combination of multiple devices include at least: preparations, polymer materials and semiconductor materials; and / or, The following steps are also included: If the test sample contains neither characteristic elements nor components matching the preset morphological features, the to-be-tested area is analyzed using X-ray diffraction analysis to obtain the components of the test sample.

9. A system for identifying raw materials and auxiliary materials in preparations based on the combination of multiple devices, characterized in that: Includes electron microscope, energy spectrometer, Raman spectrometer and processor; The system for identifying raw materials and excipients in preparations is configured to identify raw materials and excipients in preparations in the following ways: A cross section of the sample to be tested is subjected to cryo-polishing treatment in advance to obtain the cross section to be tested; Observing the cross section to be tested using the electron microscope to obtain a first electron microscope image of the cross section to be tested, and determining a region to be tested on the cross section to be tested; Scanning the area to be tested using the energy spectrometer to obtain an energy spectrum image; The processor determines whether the test sample contains characteristic elements based on the first electron microscope image and the energy spectrum image, and if so, identifies the characteristic elements based on the energy spectrum image; The processor determines whether the test sample contains a component matching a preset morphological feature based on the first electron microscope image and the energy spectrum image; if so, observes the area to be tested using an electron microscope to obtain a second electron microscope image of the area to be tested, wherein the resolution of the second electron microscope image is greater than that of the first electron microscope image; and analyzes and processes the second electron microscope image using a pre-constructed convolutional neural network model to identify the component matching the preset morphological feature; For different components in the test sample having matching morphological features, after analyzing and processing the second electron microscope image using a pre-built convolutional neural network model to identify components matching preset morphological features, further analyzing and processing the energy spectrum image using a pre-built deep learning model to identify components with similar morphological features; If the test sample contains neither characteristic elements nor components matching the preset morphological features, a Raman spectrometer is used to perform Raman spot collection on the test area to obtain a Raman spectrum map, and the processor obtains the composition of the test sample based on the Raman spectrum map.

10. The system for identifying raw materials and auxiliary materials in preparations based on multi-device combination according to claim 9, characterized in that: It also includes atomic force microscopy; The system for identifying raw materials and excipients in preparations is further configured to identify raw materials and excipients in preparations in the following manner: In the process of analyzing the components of the test sample according to the Raman spectrum, if there is a component with similar characteristic peaks, it is determined to be a target component, and the presence of the component with similar characteristic peaks is determined in the following manner: the difference between the peak values ​​of at least two characteristic peaks is within a preset peak difference threshold range; The atomic force microscope analysis method is used to obtain the microstructure information of the area to be tested and the target component is determined in combination with the mechanical properties of the target component.

Images