Raman spectrum imaging method for semisolid preparation

Abstract

The invention discloses a Raman spectrum imaging method for a semi-solid preparation, and relates to the technical field of Raman imaging.The method comprises the steps that an original sample (such as cream and gel plaster) of the semi-solid preparation is taken, the original form of the original sample is kept, grinding, dissolving or chemical modification is not conducted, the sample is placed on an inert slide glass of a two-dimensional translation table, and the sample is placed on a Raman spectrometer; the advantages of no mark, no sample preparation, high specificity and the like of the Raman spectrum technology are utilized, microscopic imaging and chemometrics analysis are combined, crystal form identification, component distribution visualization and quantitative analysis of the active pharmaceutical ingredient (API) in the semi-solid preparation are achieved, and the method can be applied to quality control, process optimization and consistency evaluation of the semi-solid preparations such as cream and gel, and has a wide application prospect. The method overcomes the defects that a traditional XRD method needs sample preparation and cannot realize online monitoring, and has the advantages of high resolution, real-time performance and full quantitative analysis capability.

Description

Technical Field

[0001] This invention relates to the field of Raman imaging technology, specifically to a Raman spectroscopy imaging method for semi-solid formulations. Background Technology

[0002] The quality of semi-solid preparations (such as creams and gels) is closely related to the crystal form and component distribution of APIs. Traditional analytical methods have obvious shortcomings: XRD requires sample processing (such as grinding), which may change the crystal form, and can only provide bulk phase information, not microscopic distribution analysis; fluorescent labeling methods are prone to interfering with biological activity and are complicated to operate. Raman spectroscopy can make up for the above shortcomings with its advantages of being label-free, non-destructive, and highly specific. However, existing Raman methods still need to be optimized for the full quantitative analysis and real-time monitoring of semi-solid preparations. The prior art document CN202411894915.1 discloses a "high-throughput analysis method for imaging the distribution of drug components in solid dosage forms". This method achieves high-throughput component distribution analysis of solid dosage forms through the combination of electron microscopy, energy dispersive spectroscopy and Raman spectroscopy. It can perform secondary Raman identification of substances containing characteristic elements and confirm the composition of the prescription by color marking the component distribution. It has advantages in the rapid batch analysis of solid dosage forms (such as tablets) and can obtain component distribution information of large samples in a short time. It has practical value for the quality control of drug tablets. However, this method has significant limitations when applied to the analysis of semi-solid preparations: First, it relies on electron microscopy sample preparation (such as cryopolishing), which can easily damage the droplet structure and gel network of semi-solid preparations, making it impossible to restore their original microscopic state and resulting in distortion of the component distribution and crystal form analysis results; Second, for excipients without characteristic elements and with similar morphologies (such as different matrix oils), it is necessary to rely on deep learning morphology recognition, which poses a risk of misjudgment and requires secondary Raman verification, increasing the complexity of the operation process and making it difficult to meet the core requirement of "non-destructive, in-situ analysis" of semi-solid preparations. Based on this, this solution proposes "a Raman spectroscopy imaging method for semi-solid formulations" to address the aforementioned problems. Summary of the Invention

[0003] The purpose of this invention is to provide a Raman spectroscopy imaging method for semi-solid formulations, in order to solve the problems mentioned in the background art, where existing market equipment relies on electron microscopy for sample preparation and deep learning is required for morphology recognition of excipients (such as different matrix oils) that lack characteristic elements and have similar morphologies.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a Raman spectroscopy imaging method for semi-solid preparations, comprising the following detailed operating steps: Step one: Take the semi-solid preparation as is (such as cream, gel patch), keep its original form, do not grind, dissolve or chemically modify, place the sample on the inert carrier of the two-dimensional translation table - the inert carrier is made of quartz material, which can avoid physical adsorption or chemical reaction with the sample; At the same time, use a flexible clamp made of silica gel to lightly press the edge of the sample. This material also does not interact with the sample, which can ensure that the sample surface is flat and has no structural damage, and the contact area between the carrier and the sample must be strictly free of impurities to provide a true and pure sample basis for subsequent analysis.

[0005] Step two: Start the spontaneous Raman spectrum microscope system, use a continuous laser light source to generate excitation light, which is guided and focused by a mirror to irradiate the sample; The Raman scattered light generated by the sample is purified by a dichroic mirror (to filter out the excitation light) and a short-pass filter, and then collected by a detection lens to a detector, which synchronously collects the Raman scattered light generated by the sample and transmits it to the detector through the dichroic mirror and the short-pass filter.

[0006] Step three: Use a standard crystalline API sample as a calibration reference. This sample is a single crystalline solid that has been verified in previous studies and has stable characteristic spectra, which can reduce the interference of polymorphism. Place it on the sample stage, turn on the excitation light, and adjust the angle of the third beam splitter through the microscope to observe the coincidence of the reflected light spots until the two beams form a completely collinear spot on the sample surface without splitting or shifting at the edges. Then adjust the height of the first focusing lens to focus the light spot on the sample surface, and determine the best focusing position by observing the characteristic Raman signal intensity change of the calibration sample to ensure clear and undistorted signals and improve calibration accuracy and focusing accuracy.

[0007] Step four: Install a short-pass filter in front of the detector. The model of the filter should be selected according to the wavelength of the light source to ensure that only Raman scattered light passes through, effectively blocking incident laser light and other stray light, and reducing background noise. At the same time, adjust the angle of the filter so that the filter surface is perpendicular to the light path to enhance the filtering effect. Then set the scanning path of the two-dimensional translation table to ensure that it covers the sample area to be analyzed, and control the translation table to move point by point at the preset step size. Each time a point is moved, the detector is triggered to collect the Raman spectrum signal at that position and simultaneously record the position coordinates to provide a reference for subsequent composition distribution analysis.

[0008] Step five: Preprocess the collected spectral data. First, perform baseline correction to eliminate the influence of baseline drift on analysis. Then use the wavelet transform denoising algorithm to process the spectral curve. This algorithm can remove random noise while preserving key information such as wavelength, intensity and peak shape of characteristic peaks through multi-scale decomposition. Finally, subtract the fluorescence background interference to ensure the integrity of the effective signal and provide high-quality spectral data support for subsequent data analysis.

[0009] Step six: The pre-processed spectral data is analyzed by using the multivariate curve resolution method of chemical metrology. The method can reduce the spectral overlap interference of complex excipients and API in the semi-solid preparation by using the pre-set excipient pure substance spectrum as a constraint condition, and improve the accuracy of spectral separation of API and excipients. According to the wave number position and peak shape of the characteristic spectrum of the separated active ingredient (API) and each excipient, the types of API and excipients are identified, and the spatial distribution image of each component in the sample is generated. At the same time, the characteristic spectrum of the API is compared with the standard crystal form spectrum library, which contains the standard Raman spectrum of the raw material API and known crystal forms. The relative position and peak shape similarity of the characteristic peaks are focused on during the comparison, so as to accurately determine the crystal form type of the API.

[0010] Step seven: If it is necessary to analyze the change of crystal form with environmental conditions, the optical path and collection parameters are kept unchanged, and an environmental control device is built around the sample stage. The device surrounds the sample stage through a sealed cavity, and the temperature and humidity in the cavity can be adjusted through an external controller to ensure stable environmental parameters and isolate the influence of external environmental fluctuations. Then steps four to six are repeated at preset time intervals to record the changes of API crystal form and component distribution at different time points, and dynamic analysis results are formed to provide controllable experimental data for studying the stability of semi-solid preparations under different environments.

[0011] As a preferred technical solution of the present application, in step one, the inert carrier sheet is made of quartz material, and the flexible clamp is made of silica gel material to avoid physical adsorption or chemical reaction with the semi-solid preparation. By using the above technical solution, the inert carrier sheet is made of quartz material, and the flexible clamp is made of silica gel material. The quartz and silica gel materials can avoid physical adsorption or chemical reaction with the semi-solid preparation, which prevents the sample from changing the composition and structure due to adsorption or chemical reaction, ensures that the semi-solid preparation maintains the original form and chemical properties, and also avoids the introduction of impurities caused by material falling off, reduces background interference, and provides a real and reliable sample basis for subsequent spectral analysis.

[0012] As a preferred technical solution of the present application, in step two, the delay line is adjusted by a mechanical knob, and the clarity of the interference fringes is observed in real time to determine whether the time difference of the excitation light is in the best state. By using the above technical solution, the delay line is adjusted by a mechanical knob, and the optical path difference is judged by the clarity of the interference fringes. The mechanical knob has high adjustment precision and can accurately control the optical path difference to ensure that the excitation light time synchronization is in the best state. The clarity of the interference fringes provides an intuitive basis for adjustment, avoids subjective errors, improves the efficiency of optical path matching, and ensures the intensity and stability of the subsequent Raman signal.

[0013] As a preferred technical solution of the present application, in step three, the standard crystal form API sample is a single crystal form solid verified in advance, and the light spot morphology is observed by a microscope to ensure that the edges of the collinear light spots are not split or offset; By adopting the above technical solution, the standard crystal form API is a single crystal form solid, and the light spot collinear state is observed by a microscope, the characteristic spectrum of the single crystal form API is stable, and as a calibration reference, the calibration accuracy can be improved by reducing the polymorphic interference, and at the same time, the microscope can directly observe the light spot morphology and accurately judge the light beam collinear state, avoiding the signal scattering caused by light spot offset, and ensuring the accuracy of the focusing position.

[0014] As a preferred technical solution of the present application, in step four, the model of the short-pass filter is selected according to the wavelength of the light source to ensure that only the Raman scattered light is allowed to pass through, and the incident laser and other stray light are blocked; By adopting the above technical solution, the model of the short-pass filter is selected according to the wavelength of the light source, which can filter the incident laser, stray light and other non-Raman signals, significantly reduce the background noise, and at the same time, avoid the loss of effective signals caused by the mismatch between the filter and the wavelength, maximize the retention of Raman scattered light, and improve the signal-to-noise ratio of the spectral data.

[0015] As a preferred technical solution of the present application, in step five, the noise reduction algorithm is wavelet transform denoising, which retains the wave number and intensity information of the characteristic peak through multi-scale decomposition; By adopting the above technical solution, the noise reduction algorithm is wavelet transform denoising, compared with the traditional noise reduction algorithm, the wavelet transform can accurately retain the wave number, intensity and peak shape of the characteristic peak through multi-scale decomposition while removing random noise, avoid the loss of useful information, improve the smoothness of the spectral curve and the recognition degree of the characteristic peak, and provide more reliable data for subsequent analysis.

[0016] As a preferred technical solution of the present application, in step six, the multivariate curve resolution method uses the pre-set pure excipient spectrum as a constraint condition to improve the accuracy of API and excipient spectrum separation; By adopting the above technical solution, the multivariate curve resolution method uses the pure excipient spectrum as a constraint condition, the excipient components in the semi-solid preparation are complex, and the spectrum is easy to overlap with the API spectrum, so the constraint of the pure excipient spectrum can significantly reduce the cross interference and improve the separation accuracy of the API and excipient characteristic spectrum, ensuring the accuracy of component identification, and is especially suitable for multi-component system analysis; As a preferred technical solution of the present application, in step six, the standard crystal form spectrum library contains the standard Raman spectrum of the raw material API and known crystal forms, and the relative position and peak shape similarity of the characteristic peaks are focused on during comparison; By adopting the technical scheme, the standard crystal form spectrum library contains raw materials and known crystal forms, comparison is made on the relative positions and peak shapes of characteristic peaks, the coverage of crystal form comparison is expanded, misjudgment caused by incomplete standard library is reduced, comparison is made in combination with the relative positions and peak shapes of characteristic peaks, the specificity of crystal form identification can be significantly improved, the false positive rate is reduced, and similar crystal forms can be easily distinguished.

[0017] As a preferred technical scheme of the present application, in step seven, the environmental control device surrounds the sample stage through a sealed cavity, and the temperature and humidity in the cavity are adjusted through an external controller to ensure the stability of the environmental parameters. By adopting the technical scheme, the environmental control device is a sealed cavity with an external controller to adjust the temperature and humidity, which can isolate the influence of external environmental fluctuations on the sample and ensure the consistency of the environmental conditions in dynamic analysis, and the external controller can accurately adjust the parameters to simulate different storage or use environments and provide a controllable experimental environment for the stability research of semi-solid preparations.

[0018] As a preferred technical scheme of the present application, the method further comprises a result visualization step of integrating the component distribution image, the crystal form identification result and the dynamic change trend into a visual report, wherein different components are marked with differentiated colors for facilitating intuitive observation of the distribution characteristics.

[0019] By adopting the technical scheme, the components are marked with differentiated colors to convert complex spectral data into intuitive images, which facilitates rapid identification of the spatial distribution characteristics of APIs and excipients, the visualization of the dynamic change trend can clearly present the evolution process of the crystal form with the environment, improves the readability and analysis efficiency of the results, and provides an intuitive basis for formulation optimization.

[0020] Compared with the prior art, the present application has the following beneficial effects: 1. Overcome the defects of traditional XRD methods: The present application does not need to perform grinding and other pretreatments on the sample, avoiding the change of crystal form caused by sample treatment; through high-resolution Raman imaging, the component distribution and crystal form analysis of micro regions can be realized, breaking through the limitation of XRD that can only provide bulk information; at the same time, online monitoring is supported, solving the problem that XRD cannot be analyzed online; 2. Avoid the shortcomings of fluorescent labeling method: The present application is based on the label-free characteristics of Raman spectrum, without the need for fluorescent labeling molecules, avoiding the interference of labeling on the activity of biological molecules, the reduction of signal-to-noise ratio caused by fluorescent bleaching, the restriction of phototoxicity on in vivo analysis, and the poor repeatability caused by complex operation, and is more suitable for the analysis requirements of semi-solid preparations in a non-destructive and in-situ manner; 3. Optimize the limitations of existing Raman methods: avoid the non-resonant background interference of coherent Raman by stimulated Raman technology, combined with wavelet transform denoising and short-pass filter to filter stray light, which significantly improves the signal-to-noise ratio; with the help of multivariate curve resolution method, the API and complex excipients are accurately separated, and the full quantitative analysis is realized by combining with the characteristic peak comparison, which solves the shortcomings of existing Raman methods in full quantitative and complex system analysis; at the same time, through the environmental control device, the real-time monitoring of dynamic process is realized, which makes up for the defects of traditional Raman in real-time; 4. Breakthrough the limitations of the comparative file: without relying on electron microscopy sample preparation, it avoids the destruction of the emulsion droplet structure and gel network of semi-solid preparations, and can restore the original microstate of the sample completely; through the high specificity of Raman spectrum, it can directly identify the excipients without characteristic elements and similar appearance, without the aid of deep learning identification, reducing the risk of misjudgment, simplifying the operation process, and perfectly adapting to the core needs of semi-solid preparation "non-destructive, in-situ analysis". BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 The figure is a schematic diagram of the Raman spectrum imaging light path block structure of the application. DETAILED DESCRIPTION

[0022] The technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, not all. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor belong to the scope of protection of the application.

[0023] Please refer to Figure 1 The technical solutions of the application include the following detailed operation steps: Step 1: Take the original semi-solid preparation such as cream, gel and paste, keep its original shape, do not grind, dissolve or chemically modify, place the sample on the inert carrier of the two-dimensional translation stage, use a flexible clamp that does not interact with the sample to press the edge of the sample, ensure that the sample surface is flat and has no structural damage, and there is no impurity pollution in the contact area between the carrier and the sample; Step 2: Start the spontaneous Raman spectrum microscope system, use a continuous laser light source to generate excitation light, guide it by a reflecting mirror and focus it by a focusing lens to irradiate the sample; the Raman scattered light generated by the sample is purified by a dichroic mirror (to filter out the excitation light) and a short-pass filter, and then converged to a detector by a detection lens, the Raman scattered light generated by the sample is collected synchronously, and then transmitted to the detector by a dichroic mirror and a short-pass filter; Step three: take the standard crystal form API sample as the calibration reference, place it on the sample stage, turn on the excitation light, adjust the angle of the third beam splitter by observing the degree of coincidence of the reflected light spots until the two beams of light form completely collinear light spots on the sample surface, adjust the height of the first focusing lens so that the light spots are focused on the sample surface, determine the optimal focusing position by observing the characteristic Raman signal intensity changes of the calibration sample, and ensure that the signal is clear and free of interference; Step four: install a short-pass filter in front of the detector, adjust its angle so that the filter surface is perpendicular to the optical path, filter out non-Raman scattered light, set the scanning path of the two-dimensional translation stage to cover the sample area to be analyzed, control the translation stage to move point by point at the preset step size, trigger the detector to collect the Raman spectrum signal at each position, and record the coordinates of the position simultaneously; Step five: perform baseline correction on the collected spectral data to eliminate baseline drift effects, use noise reduction algorithms to process the spectral curve, retain characteristic peak information while removing random noise, subtract the fluorescence background interference to ensure the integrity of the effective signal; Step six: use chemometrics multivariate curve resolution methods to analyze the preprocessed spectral data, separate the characteristic spectra of the active ingredient API and each excipient in the semi-solid preparation, identify the types of API and excipients according to the wave number position and peak shape of the characteristic spectra, and generate a spatial distribution image of each component in the sample. Compare the characteristic spectrum of API with the standard crystal form spectrum library, and determine the crystal form type of API by matching the characteristic peaks. Step seven: If you need to analyze the changes in crystal form under environmental conditions, keep the optical path and collection parameters unchanged, build an environmental control device around the sample stage to adjust the temperature, humidity, and other conditions. Repeat steps four to six at the preset time intervals to record the changes in API crystal form and ingredient distribution at different time points, and form a dynamic analysis result.

[0024] In step one, the inert slide is made of quartz, and the flexible clamp is made of silicone, which can avoid physical adsorption or chemical reaction with the semi-solid preparation. The quartz and silicone materials can effectively avoid adverse interactions with the semi-solid preparation, prevent the sample from changing its composition and structure due to adsorption or chemical reaction, ensure that the semi-solid preparation maintains its original form and chemical properties, and at the same time avoid the introduction of impurities caused by material shedding, reduce background interference, and lay a solid foundation for subsequent spectral analysis. In step two, the delay line is adjusted by a mechanical knob, and the clarity of the interference fringes is observed in real time to determine whether the time difference of the excitation light is in the optimal state. The mechanical knob adjustment of the delay line can achieve high-precision control, accurately control the optical path difference to ensure that the time synchronization of the excitation light is optimal, and the clarity of the interference fringes provides an intuitive basis for adjustment, which helps to avoid subjective errors, improves the efficiency of optical path matching, and thus ensures the intensity and stability of the subsequent Raman signal. In step three, the standard crystal form API sample is a single crystal form solid verified in advance, and the light spot morphology is observed by microscope to ensure that the light spot edge is not split or offset after collimation, the characteristic spectrum of the single crystal form API has good stability, and as a calibration reference, it can reduce the interference of polymorphism and improve the calibration accuracy; By observing the light spot morphology directly through the microscope, the collimation state of the light beam can be accurately judged, and the signal scattering caused by the light spot offset can be avoided, ensuring the accuracy of the focusing position; In step four, the model of the short-pass filter is selected according to the wavelength of the light source to ensure that only the Raman scattered light is allowed to pass through, and the incident laser and other stray light are blocked. By selecting a short-pass filter that matches the wavelength of the light source, non-Raman signals such as incident laser and stray light can be targeted filtered, significantly reducing background noise, while avoiding the loss of effective signals caused by mismatch between the filter and the wavelength, maximizing the retention of Raman scattered light, and improving the signal-to-noise ratio of the spectral data; In step five, the noise reduction algorithm is wavelet transform denoising, which retains the wave number and intensity information of the characteristic peaks through multi-scale decomposition. Compared with traditional denoising algorithms, wavelet transform can accurately retain the wave number, intensity, and peak shape of the characteristic peaks through multi-scale decomposition while removing random noise, avoiding the loss of useful information, improving the smoothness of the spectral curve and the recognition of the characteristic peaks, and providing more reliable data support for subsequent analysis; In step six, the multivariate curve resolution method uses pre-set pure excipient spectrum as a constraint condition to improve the accuracy of API and excipient spectrum separation. For the problem of complex excipient composition in semi-solid preparations and the overlapping of their spectra with API spectra, using pure excipient spectrum as a constraint can significantly reduce cross interference and improve the separation accuracy of API and excipient characteristic spectra, ensuring the accuracy of component identification, especially suitable for analysis of multi-component systems; In step six, the standard crystal form spectrum library contains the standard Raman spectra of raw materials API and known crystal forms. When comparing, the relative position and peak shape similarity of the characteristic peaks are focused on. The standard crystal form spectrum library covers raw materials and known crystal forms, which can expand the coverage of crystal form comparison and reduce false judgments caused by incomplete standard library. Combined with the relative position and peak shape of the characteristic peaks, the specificity of crystal form identification can be significantly improved, and the false positive rate can be reduced, which is convenient for distinguishing similar crystal forms; In step seven, the environmental control device surrounds the sample stage by a sealed cavity, and the temperature and humidity in the cavity are adjusted by an external controller to ensure stable environmental parameters. The sealed cavity can isolate the influence of external environmental fluctuations on the sample, ensuring the consistency of environmental conditions in dynamic analysis. The external controller can accurately adjust the parameters to simulate different storage or use environments, providing a controllable experimental environment for studying the stability of semi-solid preparations; The method further comprises a result visualization step: integrating the component distribution image, the crystal form identification result and the dynamic change trend into a visual report, wherein different components are marked with differentiated colors, so as to intuitively observe the distribution characteristics; through the differentiated color marking of the component distribution, complex spectral data can be converted into intuitive images, facilitating the rapid identification of the spatial distribution characteristics of the API and the excipient; the visualization of the dynamic change trend can clearly present the evolution process of the crystal form with the environment, improve the readability and analysis efficiency of the result, and provide an intuitive basis for the optimization of the preparation.

[0025] Working principle: a Raman spectral imaging method for semi-solid preparations uses, based on the Raman scattering characteristics, combined with precise optical regulation, signal acquisition and data analysis technology, to realize the analysis of the component distribution and crystal form characteristics of semi-solid preparations, the specific process is as follows: Firstly, the original form of the semi-solid preparation such as cream, gel and paste is kept as it is without grinding, dissolving and other treatments, a quartz inert carrier sheet and a silica gel flexible clamp are used to fix the sample, so as to avoid physical adsorption or chemical reaction between the sample and the carrier, ensure that the original components and structure of the sample are not disturbed, and provide a true material basis for subsequent analysis; Secondly, the pump light, depletion light and probe light are generated by the femtosecond optical comb light source system, the optical path is adjusted by using beam splitters, mirrors and delay lines, and the optical axes of the three beams are precisely converged at the same point on the sample surface, wherein the delay line adjusts the optical path difference by a mechanical knob, and the beam time synchronization is judged by the interference fringe definition; a single crystal form API is used as a calibration reference, the light spot shape is observed by a microscope, the collinearity and accurate focusing of the light beams are ensured, and the light beams are effectively interacted with the sample to excite Raman scattering signals; Then, a short-pass filter suitable for the wavelength of the light source is configured in front of the detector to filter out non-Raman signals such as incident laser light and stray light, only Raman scattering light is retained, a two-dimensional translation stage is controlled to move the sample point by point according to the preset path, the detector is triggered synchronously to collect the Raman spectrum signals of each point and record the coordinates, and the whole analysis area of the sample is scanned; Then, the collected spectral data is preprocessed: the baseline correction is used to eliminate the drift influence, the wavelet transform denoising algorithm is used to remove random noise and retain the details of the characteristic peaks, the fluorescence background interference is deducted, and the purity of the spectral data is improved, then the multivariate curve resolution method is used, combined with the excipient pure substance spectrum constraint, the characteristic spectrum of the active ingredient API and the excipient is separated, the component type is identified according to the wave number and peak shape of the characteristic peak, and a spatial distribution image is generated; the API characteristic spectrum is compared with the standard spectrum library containing raw materials and known crystal forms, and the crystal form type is determined by the relative position and peak shape similarity of the characteristic peaks; Finally, if the dynamic changes of the crystal form with the environment need to be analyzed, the temperature and humidity of the sample stage are stabilized by sealing the cavity and connecting the controller, the light path and the acquisition parameters are kept unchanged, the scanning and analysis process is repeated at the preset time interval, the changes of the crystal form and the composition distribution at different time points are recorded, and the dynamic analysis results are formed, so that the precise Raman spectrum imaging analysis of the semi-solid preparation is realized.

[0026] Thus, a series of work, which is not described in detail in the specification, belongs to the prior art known to those skilled in the art.

[0027] Although the embodiments of the present application have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made thereto without departing from the principles and spirit of the present application, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A Raman spectral imaging method for a semi-solid preparation, characterized by, The detailed operation steps include the following: Step 1: Take the semi-solid preparation as it is (such as cream, gel patch), keep its original form, do not grind, dissolve or chemically modify, place the sample on the inert carrier of the two-dimensional translation table - the inert carrier is made of quartz; At the same time, use a flexible clamp made of silica gel to lightly press the edge of the sample. This material also does not interact with the sample, and the contact area between the carrier and the sample must be strictly free of impurities. Step 2: Start the spontaneous Raman spectrum microscope system, use a continuous laser light source to generate excitation light, which is guided and focused by a mirror to irradiate the sample; the Raman scattered light generated by the sample is purified by a dichroic mirror (to filter out the excitation light) and a short-pass filter, and then collected by a detection lens and a detector; the Raman scattered light generated by the sample is transmitted to the detector through the dichroic mirror and the short-pass filter. Step 3: Use a standard crystalline API sample as a calibration reference. This sample is a single crystalline solid that has been previously verified. Place it on the sample stage, turn on the excitation light, and adjust the angle of the third beam splitter by observing the coincidence of the reflected light spots until the two light spots form a completely collinear spot on the sample surface without splitting or shifting at the edges. Then adjust the height of the first focusing lens to focus the light spot on the sample surface, and determine the best focusing position by observing the characteristic Raman signal intensity change of the calibration sample to ensure clear and undistorted signals. Step 4: Install a short-pass filter in front of the detector, and select the filter type according to the wavelength of the light source; adjust the angle of the filter so that the filter surface is perpendicular to the light path; then set the scanning path of the two-dimensional translation table to ensure that it covers the sample area to be analyzed, and control the translation table to move point by point at the preset step size. Each time a point is moved, the detector is triggered to collect the Raman spectrum signal at that position, and the position coordinates are recorded simultaneously to provide a reference for subsequent component distribution analysis. Step 5: Preprocess the collected spectrum data, first perform baseline correction, then use wavelet transform denoising algorithm to process the spectrum curve, and finally subtract the fluorescence background interference. Step 6: Use multivariate curve resolution methods of chemometrics to analyze the preprocessed spectrum data - this method uses pre-set pure material spectrum of excipients as constraint condition; according to the wave number position and peak shape of the characteristic spectrum of the separated active ingredient (API) and each excipient, the types of API and excipients are identified, and the spatial distribution image of each component in the sample is generated; At the same time, the characteristic spectrum of API is compared with the standard crystal spectrum library, which contains the standard Raman spectrum of raw API and known crystal forms. During comparison, focus on the relative position and peak shape similarity of characteristic peaks to accurately determine the crystal type of API. Step seven: If the change of crystal form with environmental conditions needs to be analyzed, keep the optical path and collection parameters unchanged, and build an environmental control device around the sample stage. The device surrounds the sample stage through a sealed cavity, and the temperature and humidity in the cavity can be adjusted through an external controller to isolate the influence of external environmental fluctuations. Then repeat steps four to six at preset time intervals to record the changes in API crystal form and composition distribution at different time points, forming dynamic analysis results and providing controllable experimental data for studying the stability of semi-solid preparations under different environments.

2. The Raman spectral imaging method for a semi-solid preparation according to claim 1, characterized by, In step one, the inert slide is made of quartz, and the flexible clamp is made of silicone to avoid physical adsorption or chemical reaction with the semi-solid preparation.

3. The Raman spectral imaging method for a semi-solid preparation according to claim 2, characterized by, In step two, the delay line is adjusted by a mechanical knob. The clarity of the interference fringes is observed in real time to determine whether the time difference of the excitation light is in the optimal state.

4. The Raman spectral imaging method for a semi-solid preparation according to claim 3, characterized by, In step three, the standard crystal form API sample is a single crystal form solid verified in the previous stage. The spot morphology is observed under a microscope to ensure that the edges of the collimated spot are not split or shifted.

5. The method for Raman spectral imaging of a semi-solid preparation according to claim 4, characterized in that, In step four, the type of short-pass filter is selected according to the wavelength of the light source to ensure that only Raman scattered light is allowed to pass through, and incident laser light and other stray light are blocked.

6. The method for Raman spectral imaging of a semi-solid preparation according to claim 5, characterized in that, In step five, the noise reduction algorithm is wavelet transform denoising, which preserves the wavenumber and intensity information of the characteristic peaks through multi-scale decomposition.

7. The method for Raman spectral imaging of a semi-solid preparation according to claim 6, characterized in that, In step six, the multivariate curve resolution method uses pre-set auxiliary material pure substance spectra as constraint conditions to improve the accuracy of API and auxiliary material spectral separation.

8. The Raman spectral imaging method for a semi-solid preparation according to claim 4, characterized by, In step six, the standard crystal form spectrum library contains the standard Raman spectra of raw material API and known crystal forms. When comparing, focus on the relative position and peak shape similarity of the characteristic peaks.

9. The method for Raman spectral imaging of a semi-solid preparation according to claim 5, wherein, In step seven, the environmental control device surrounds the sample stage through a sealed cavity, and the temperature and humidity in the cavity are adjusted through an external controller to ensure stable environmental parameters.

10. The method for Raman spectral imaging of a semi-solid preparation according to claim 6, wherein The method also includes a result visualization step: integrate the composition distribution image, crystal form identification results, and dynamic change trend into a visual report, where different components are marked with differentiated colors for intuitive observation of the distribution characteristics.

Images