Quality inspection device, quality inspection method, and program

Abstract

To provide a quality inspection device, a quality inspection method and a program capable of further precisely inspecting the quality of a freeze-dried formulation or a powdery medicine.SOLUTION: A quality inspection device 1 inspects the quality of an inspection target sample which is a freeze-dried formulation or a powdery medicine filled in a transparent container B. The quality inspection device includes an illuminator 13 that illuminates the bottom surface of the transparent container B, a multispectral photo-detector (imaging unit 12) that captures an image of the side surface of the transparent container B, a calculator (controller 21) that calculates an absorption spectrum at a plurality of wavelengths based on a result of the photo-detection by the multispectral photo-detector, and a detector (controller 21) that detects a test target component contained in a test target sample based on the absorption spectrum calculated by the calculator.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a quality inspection device, a quality inspection method, and a program. [Background technology]

[0002] In the production of freeze-dried or powdered pharmaceuticals, if the preparation is not sufficiently dried and moisture remains, the moisture and active ingredient may undergo a hydrolysis reaction, which may affect the storage stability of the preparation. Therefore, it is required to manufacture the preparation so that the moisture content is sufficiently low.

[0003] There is a need to inspect the components of freeze-dried preparations or powdered pharmaceuticals, such as the water content, during their production. Patent Documents 1 to 3 disclose techniques for inspecting freeze-dried preparations or pharmaceuticals.

[0004] In the production of freeze-dried formulations or powdered pharmaceuticals, there are cases where the moisture content of some formulations is not as designed due to factors such as the increase in size of the formulation manufacturing equipment (freeze-drying chamber). Also, there may be variations in the moisture content between formulations. Therefore, there is a need for non-destructive 100% inspection during the production of such formulations. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Patent No. 3351910 [Patent Document 2] International Publication No. 2007 / 063840 Summary of the Invention [Problem to be solved by the invention]

[0006] However, conventionally, ingredient inspections of freeze-dried preparations or pharmaceuticals during manufacturing have generally been conducted by destructive sampling testing, which has raised concerns about whether all defective products have been detected.On the other hand, conventional visual inspections, which are non-destructive inspections, have made it difficult to directly measure the amount of ingredients or internal defects of freeze-dried preparations or pharmaceuticals. The production of freeze-dried or powdered pharmaceuticals requires rigorous quality inspection of the formulation.

[0007] Therefore, in order to solve the above problems, an object of the present invention is to provide a quality inspection device, a quality inspection method, and a program that can inspect the quality of freeze-dried preparations or powdered pharmaceuticals more precisely. [Means for solving the problem]

[0008] The quality inspection device according to claim 1 comprises: A quality inspection device for inspecting the quality of a sample to be inspected, which is a freeze-dried preparation or a powdered pharmaceutical product filled in a transparent container, an illumination unit that illuminates the bottom surface of the transparent container; a multispectral light receiving unit that captures an image of the side surface of the transparent container; a calculation unit that calculates an absorption spectrum at a plurality of wavelengths based on the light reception results by the multispectral light receiving unit; and a detection unit that detects a target component contained in the test sample based on the absorption spectrum calculated by the calculation unit.

[0009] The invention described in claim 2 is the quality inspection device described in claim 1, The detection unit detects the component to be inspected by comparing the absorption spectrum of a reference sample with the absorption spectrum of the inspection target sample.

[0010] The invention described in claim 3 is the quality inspection device described in claim 1, The detection unit calculates the amount of the component to be inspected by applying the absorption spectrum to a calibration model that has been constructed in advance.

[0011] The invention described in claim 4 is the quality inspection device described in claim 1, The component to be inspected is at least one of an active pharmaceutical ingredient, an excipient, and water.

[0012] The invention described in claim 5 is the quality inspection device described in claim 1, The detection unit compares the shapes of the absorption spectra of the plurality of test samples, and determines that the test sample corresponding to the absorption spectrum whose shape matches with other absorption spectra less than a predetermined threshold is abnormal.

[0013] The invention described in claim 6 is the quality inspection device according to any one of claims 2 to 5, the multispectral light receiving unit is a camera or a multichannel spectrometer, and captures images at multiple points in the vertical direction of the transparent container; The calculation unit calculates the absorption spectrum based on a relative amount of received light at one point relative to the amount of received light at the other point, out of the results of received light at the plurality of points.

[0014] The invention described in claim 7 is the quality inspection device according to any one of claims 2 to 5, a reference light receiving unit that measures the amount of light emitted from the light source of the illumination unit; The calculation unit calculates the absorption spectrum based on the amount of light emitted measured by the reference light receiving unit.

[0015] The invention described in claim 8 is the quality inspection device described in claim 1, The illumination range of the illumination unit that illuminates the bottom surface of the transparent container is adjustable.

[0016] The invention described in claim 9 is the quality inspection device described in claim 1, The multispectral light receiving unit is a hyperspectral camera.

[0017] The quality inspection method according to claim 10 comprises: A quality inspection device for inspecting the quality of a sample to be inspected, which is a freeze-dried preparation or a powdered pharmaceutical product filled in a transparent container, an illumination unit that illuminates the bottom surface of the transparent container; a multispectral light receiving unit that captures an image of a side surface of the transparent container, a calculation step of calculating an absorption spectrum at a plurality of wavelengths based on the light reception results by the multispectral light receiving unit; and a detection step of detecting a target component contained in the test sample based on the absorption spectrum calculated in the calculation step.

[0018] The program according to claim 11 comprises: A quality inspection device for inspecting the quality of a sample to be inspected, which is a freeze-dried preparation or a powdered pharmaceutical product filled in a transparent container, an illumination unit that illuminates the bottom surface of the transparent container; a multispectral light receiving unit that captures an image of the side surface of the transparent container; a calculation unit that calculates an absorption spectrum at a plurality of wavelengths based on the light reception results by the multispectral light receiving unit; The detector functions as a detector that detects a target component contained in the test sample based on the absorption spectrum calculated by the calculator. [Effects of the Invention]

[0019] According to the present invention, the quality of freeze-dried preparations or powdered pharmaceuticals can be inspected more precisely. [Brief explanation of the drawings]

[0020] [Figure 1] 1 is a block diagram of a quality inspection device according to an embodiment of the present invention. [Figure 2] FIG. 2 is a perspective view of an inspection unit. [Figure 3] FIG. 2 is a diagram showing the configuration of an inspection unit. [Figure 4]FIG. 10 is a diagram illustrating an example of light scattering characteristics. [Figure 5] 10 is a flowchart illustrating an example of the flow of a quality inspection process. [Figure 6] 10 is a flowchart showing the flow of an absorption spectrum calculation process. [Figure 7] FIG. 10 is a diagram illustrating an example of an evaluation position. [Figure 8] FIG. 10 is a diagram showing an example of an absorption spectrum. [Figure 9] 10A and 10B are diagrams showing an example of matching a reference absorption spectrum with an absorption spectrum of a test sample; [Figure 10] FIG. 10 is a diagram showing an example of a comparison of absorption spectra. [Figure 11A] 10A and 10B are diagrams illustrating an example in which the illumination range of the illumination unit is enlarged. [Figure 11B] 10A and 10B are diagrams illustrating an example in which the illumination range of the illumination unit is reduced. DETAILED DESCRIPTION OF THE INVENTION

[0021] [Configuration example of quality inspection device 1] 1 is a block diagram of a quality inspection device 1 according to this embodiment. The quality inspection device 1 is a device for inspecting the quality of an inspection target sample, which is a freeze-dried preparation or a powdered pharmaceutical product. As shown in Fig. 1, the quality inspection device 1 includes an inspection unit 10 and a processing device 20. Fig. 2 is a perspective view of the inspection unit 10, and Fig. 3 is a diagram showing the configuration of the inspection unit 10. In Figs. 2 and 3, the X-axis direction and the Y-axis direction are two horizontal directions that are perpendicular to each other, and the vertical direction that is perpendicular to the X-axis and Y-axis is the Z-axis direction.

[0022] The inspection unit 10 includes a rotating section 11, an imaging section 12 as a multispectral light receiving section, an illumination section 13, and the like. The rotating unit 11 rotates a sample stage 111 on a horizontal plane (XY plane) by driving a motor or the like (not shown), on which a transparent container B filled with a sample S is placed. The sample S is a freeze-dried preparation or a powdered pharmaceutical product to be inspected by the quality inspection device 1. In this embodiment, the quality inspection device 1 is installed in a production line (in-line) that manufactures freeze-dried preparations or powdered pharmaceuticals. The inspection unit 10 includes a transport mechanism that transports the transparent container B to the sample stage 111. However, the quality inspection device 1 may be installed as a benchtop offline device in a laboratory, etc. In this case, the inspection unit 10 does not need to include a transport mechanism for transporting the transparent container B to the sample stage 111.

[0023] The imaging unit 12 is, for example, a hyperspectral camera or a multi-channel spectrometer. The imaging unit 12 is placed on the sample stage 111 and captures an image of the side surface of the transparent container B rotated by the rotation unit 11. In this embodiment, the image capturing unit 12 captures images of the sample S while rotating it once, so the spectroscopic principle of the hyperspectral camera serving as the image capturing unit 12 is optimally a push-broom type. However, the spectroscopic principle of the hyperspectral camera may be a wavelength scanning type such as a Fabry-Perot type, or a snapshot type. If there is no problem in limiting the wavelength range of the image capturing by the image capturing unit 12 to a predetermined range, the image capturing unit 12 may be a multispectral camera.

[0024] The imaging unit 12 captures an image of the range of one frame for one full rotation of the sample S. The imaging unit 12 separates the wavelength of light into a plurality of wavelength bands and captures the image. Specifically, the imaging unit 12 generates a data cube in which two-dimensional planar images of the Y-axis and Z-axis of the sample S, which is the image target, are stacked in layers for each of the separated wavelength regions. As a result, the imaging unit 12 measures the intensity and light scattering characteristics of the transmitted diffused light generated by irradiating the sample S with illumination light from the illumination unit 13 in all directions of the sample S. The imaging unit 12 outputs the measurement results of the intensity and light scattering characteristics of the transmitted diffused light to the processing device 20. Figure 4 shows the details of the light scattering characteristics obtained by capturing images of one complete rotation of sample S. In the example shown in Figure 4, the horizontal axis represents pixels (px) in the Z-axis direction within one frame, and the vertical axis represents the intensity of the measured transmitted diffused light. As shown in Figure 4, the light scattering characteristics peak at a specific position (Area 0). On the positive Z-axis side of Area 0, the light scattering characteristics gradually attenuate as the light moves in the positive Z-axis direction due to the scattering and absorption characteristics of the sample. The light scattering characteristics vary depending on the formulation type and manufacturing process of sample S, etc.

[0025] 2 and 3, the inspection unit 10 is provided with one imaging unit 12, but the inspection unit 10 may be provided with two or more imaging units 12. In this case, multiple images of the samples S can be taken simultaneously by the multiple imaging units 12. This makes it possible to improve the takt time and inspection speed in the quality inspection process for inspecting the quality of the samples S. The imaging wavelength range of the imaging unit 12 preferably includes a near-infrared region, which has relatively high transmittance and can acquire characteristic absorption peaks of components such as active ingredients, excipients, and water contained in the sample S. In this embodiment, the imaging wavelength range of the imaging unit 12 is 900 to 1700 nm.

[0026] The illumination unit 13 includes an illumination light 131, a light source 132, etc., and irradiates the bottom surface of the transparent container B in which the sample S is filled with illumination light. The illumination light 131 is configured by bundling multiple optical fibers, and guides illumination light emitted from a light source 132 to the tip side (bottom side of the transparent container B) of the illumination light 131. The illumination light 131 irradiates illumination light from the tip surface toward the bottom side of the transparent container B. The light source 132 is preferably configured as a halogen lamp that has a broad wavelength range and a relatively high light intensity, but may also be configured as an LED (Light Emitting Diode).

[0027] The processing device 20 is, for example, a personal computer, etc. The processing device 20 is connected to the rotating unit 11 and the imaging unit 12 of the inspection unit 10 via wiring (not shown), and controls the operations of the rotating unit 11 and the imaging unit 12. The processing device 20 may also be connected to the lighting unit 13 via wiring, and control the operation of the lighting unit 13. The processing device 20 includes a control unit 21 , a storage unit 22 , a communication unit 23 , an operation unit 24 , and a display unit 25 .

[0028] The control unit 21 includes, for example, a processor such as a CPU (Central Processing Unit) or an MPU (Micro Processing Unit) and a memory such as a RAM (Random Access Memory). The control unit 21 executes a program 22a stored in a memory such as the RAM, a storage unit 22, etc., to realize various processes including quality inspection processes for the sample S.

[0029] The storage unit 22 includes any storage module, such as a hard disk drive (HDD), a solid state drive (SSD), a read only memory (ROM), and a RAM. The storage unit 22 stores, for example, system programs, application programs, and various data. Specifically, the storage unit 22 stores a program 22a for executing quality inspection processing of the sample S, etc.

[0030] The communication unit 23 includes, for example, a communication module including a NIC (Network Interface Card), a receiver, and a transmitter, etc. The communication unit 23 communicates various information and data with external devices connected via a network such as the Internet.

[0031] The operation unit 24 includes, for example, a mouse, a keyboard, switches, buttons, etc. The operation unit 24 may be, for example, a touch panel integrally combined with the display unit 25, or an interface that accepts voice input. The operation unit 24 accepts instructions corresponding to various input operations from the user, converts the accepted instructions into operation signals, and outputs the operation signals to the control unit 21.

[0032] The display unit 25 is, for example, a liquid crystal display, an organic EL (Electro Luminescence) display, etc. The display unit 25 performs display based on display data output from the control unit 21.

[0033] By configuring the quality inspection device 1 as described above, it is possible to easily incorporate the quality inspection device 1 into an existing visual inspection device. For example, there are space limitations when incorporating a diffuse reflectance measurement benchtop spectroscopic analyzer in-line. On the other hand, by using the imaging unit 12 and the lighting unit 13 as a replacement for the machine vision camera of the existing visual inspection device, it is possible to easily incorporate the quality inspection device 1 into the existing visual inspection device.

[0034] [Quality Inspection Device 1 Operation] Next, the operation of the quality inspection device 1 according to this embodiment will be described. Fig. 5 is a flowchart showing an example of the flow of a quality inspection process in which the quality inspection device 1 inspects the quality of the sample S.

[0035] (Quality inspection processing) The user blocks light and captures an image using the imaging unit 12 of the quality inspection device 1. This causes the imaging unit 12 to perform a dark measurement and output dark data, which is the measurement result, to the processing device 20 (step S1). Next, the user photographs a sample that serves as a reference in the quality inspection (hereinafter referred to as the reference sample) using the imaging unit 12. As a result, the imaging unit 12 measures the reference sample and outputs the measurement data of the reference sample to the processing device 20 (step S2). During imaging, the imaging unit 12 captures an image based on the signal intensity acquired through the optical path to the imaging unit 12, where absorption and scattering occur within the sample due to illumination light irradiated from the illumination unit 13. Therefore, when measuring a reference sample, signals from an object whose absorption and scattering characteristics are known may be acquired in advance and used as reference values. The reference sample may be a standard white plate such as PTFE (polytetrafluoroethylene), or a predetermined sample similar to the sample to be measured. The measurement of the reference sample in step S2 includes calibration of the imaging unit 12 when it is started up.

[0036] Next, the user uses the imaging unit 12 to capture an image of the sample S, which is the sample to be inspected and placed on the sample stage 111, for one full rotation. As a result, the imaging unit 12 measures the intensity and light scattering characteristics of the transmitted diffused light for one full rotation of the sample to be inspected, and outputs the measurement data to the processing device 20 (step S3). In step S3, the imaging unit 12 may measure only one frame of the test sample, or any range of the test sample, instead of measuring the entire rotation of the test sample. In this case, the intensity of the transmitted diffused light can be measured from the measurement results for one frame, making it possible to calculate the absorption spectrum from the measurement data. If a crack has occurred in the test sample, the intensity of the transmitted diffused light and the light scattering characteristics may not be measurable at the location of the crack. Therefore, it is preferable to exclude the location of the crack from the target range of the detection process in step S6, which will be described later. Next, the control unit 21 of the processing device 20 acquires the measurement data output from the imaging unit 12. Next, the control unit 21 executes the absorption spectrum calculation process shown in Fig. 6 based on the acquired measurement data of the test sample (step S4).

[0037] (Absorption spectrum calculation process) The control unit 21 executes a dark correction process (step S11). Specifically, the control unit 21 subtracts the dark data measured in step S1 from the acquired measurement data of the inspection target sample. This makes it possible to remove noise due to dark current in the imaging unit 12. Next, the control unit 21 calculates the diffuse transmittance of the inspection target sample based on the measurement data after the dark correction process in step S11 (step S12). Two methods for calculating the diffuse transmittance of the inspection target sample will be described below.

[0038] (Method 1 for calculating diffuse transmittance) In the first method of calculating the diffuse transmittance, in step S3, the user uses the imaging unit 12 to capture images of the test sample at two or more measurement points in the Z-axis direction (the vertical direction of the transparent container B). The control unit 21 performs the following calibration process on the light scattering characteristics, which are measurement data measured at two or more measurement points. Specifically, the control unit 21 sets px, which indicates a peak, as in Area 0 shown in FIG. 4, as the calibration point. The control unit 21 divides the intensity of the transmitted diffused light at other measurement points by the intensity of the transmitted diffused light at the calibration point. The other measurement points are, for example, Areas 1, 2, and 3 shown in FIG. 4. Alternatively, the control unit 21 may divide the intensity of the transmitted diffused light at other measurement points by the average value of the intensity of the transmitted diffused light in a predetermined range including the calibration point. The control unit 21 performs this calibration process for each frame captured by the imaging unit 12. The measurement data after the calibration process is data indicating the diffuse transmittance of the sample to be inspected. By performing the calibration process, the control unit 21 normalizes the overall diffuse transmittance by setting the average value of the diffuse transmittance in a predetermined range including the calibration point to 1. That is, the control unit 21 calculates the absorption spectrum of the sample to be inspected in step S13 described below based on the relative amount of received light of one transmitted diffused light relative to the intensity (amount of received light) of the other transmitted diffused light, out of the measurement data (light reception results by the imaging unit 12) at multiple measurement points.

[0039] By performing the above calibration process, the influence of the light source 132 on the diffuse transmittance of the test sample can be removed. As a result, in the detection process in step S6 described below, evaluation can be performed focusing only on the absorption of illumination light by the test sample between the calibration point (reference position) and other measurement points (evaluation positions). Figure 7 shows the reference position A0 and the positions of evaluation positions A1 to A3. As shown in Figure 7, the distance from reference position A0 to evaluation position A1 is Δd1, the distance from reference position A0 to evaluation position A2 is Δd2, and the distance from reference position A0 to evaluation position A3 is Δd3. In this case, the diffuse transmittance Tn at a predetermined evaluation position An can be calculated using the following formula (1) (Beer-Lambert law):

number

[0040] As shown in the above formula (1), the influence (I0) of the light source 132 on the diffuse transmittance Tn of the test sample can be removed. This allows evaluation to be performed in the detection process of step S6, which will be described later, focusing on the influence of the components and concentrations of the test sample. In the detection process of step S6, evaluation can be performed in the detection process of step S6, focusing on a predetermined evaluation position An on the test sample. As described above, when calculating the diffuse transmittance of the inspection sample by Method 1, the measurement of the reference sample in step S2 may be omitted.

[0041] (Diffuse transmittance calculation method 2) In the diffuse transmittance calculation method 2, the control unit 21 calculates the diffuse transmittance Tn at a predetermined evaluation position An using the measurement data of the reference sample measured in step S2 as W in the above formula (1). In this case, the detection process in step S6, which will be described later, can be performed based on the reference sample, making it possible to detect abnormalities in the sample to be inspected. In the above formula (1), measurement data of a predetermined sample or measurement data at another evaluation position on the sample to be inspected may be used as W.

[0042] Alternatively, the quality inspection device 1 may include a reference light-receiving unit (not shown) that measures the amount of light emitted by the light source 132 of the illumination unit 13. In this case, in Method 2 for calculating the diffuse transmittance, the control unit 21 calculates the diffuse transmittance Tn at a predetermined evaluation position An using the amount of light received by the light source 132 measured by the reference light-receiving unit as W in the above formula (1). That is, the control unit 21 calculates the absorption spectrum of the inspection target sample in step S13, which will be described later, based on the amount of light received by the light source 132 measured by the reference light-receiving unit.

[0043] Next, the control unit 21 calculates the absorbance based on the diffuse transmittance of the test sample calculated in step S12. The control unit 21 calculates the absorbance at each wavelength of the transmitted diffused light to calculate an absorption spectrum (step S13), and ends the absorption spectrum calculation process. That is, the control unit 21 calculates the absorption spectrum at multiple wavelengths based on the measurement data from the imaging unit 12 (the light reception results from the multispectral light receiving unit). The control unit 21 functions as a calculation unit. Step S13 is a calculation step. Below, two methods for calculating the absorbance of the test sample will be described.

[0044] (Method 1 for calculating absorbance) In the absorbance calculation method 1, the control unit 21 calculates the absorbance Abs at a predetermined evaluation position An for the test sample using the following formula (2): n Calculate.

number

[0045] As shown in the above formula (2), the absorbance Abs of the test sample n This allows the effect of the average scattering distance to be removed, making it possible to perform evaluation focusing on a predetermined evaluation position An on the inspection target sample in the detection process in step S6, which will be described later. An example of an absorption spectrum calculated based on the diffuse transmittance is shown in Figure 8. In the example shown in Figure 8, the horizontal axis represents the measured wavelength of the transmitted diffuse light, and the vertical axis represents the calculated absorbance.

[0046] (Method 2 for calculating absorbance) In the absorbance calculation method 2, the control unit 21 calculates the absorbance based on the light scattering characteristics of each wavelength of the transmitted diffused light. In the above formula (1), εcΔd n is unknown in the measurement data acquired in step S3. However, the light attenuation characteristics of the diffuse transmittance Tn can be acquired from the measurement data acquired in step S3. Therefore, the control unit 21 estimates εc by fitting the light attenuation characteristics of the diffuse transmittance Tn with an exponential function. This makes it possible to calculate the unit absorption spectrum Abs. Alternatively, the control unit 21 may n εc in the above equation (2) may be estimated by linearly approximating the attenuation characteristics of

[0047] Alternatively, the control unit 21 may estimate the scattering coefficient and the absorption coefficient using an arbitrary model such as a light diffusion theory based on the measurement approach of spatially resolved spectroscopy (SRS). When the light diffusion theory is used, the control unit 21 acquires a plurality of spatially different spectral information (intensity of transmitted diffused light) by the SRS method. Next, the control unit 21 substitutes the light attenuation characteristics of the spectral information into the light diffusion equation and solves the simultaneous equations to obtain the scattering coefficient μ s ′ and absorption coefficient μ a This calculates the scattering coefficient μ s ' and absorption coefficient μ a Since the content of the component in the test sample can be calculated, the content of the component in the test sample can be precisely detected in the detection process in step S6 described below. According to the principle of multiple scattering of light, the scattering direction is randomized at a certain depth Z0 in the scattering medium, which generates a virtual light source. The light diffusion equation R(ρ) at the isodiametric distance (average scattering distance) ρ is shown in the following equation (3). The effective attenuation coefficient μ eff is shown in the following formula (4).

number

[0048] The control unit 21 calculates the scattering coefficient μ s ′ and absorption coefficient μ a By calculating the above, fitting to the diffusion theory model shown in the following formulas (5) to (9) is performed.

number

[0049] Next, the control unit 21 performs preprocessing on the absorption spectrum calculated in step S4 (step S5), such as noise processing, normalization, correction processing (baseline correction processing, etc.), SNV (Standard Normal Variate) processing, smoothing by the Savitzky-Golay method, differentiation processing, etc.

[0050] Next, the control unit 21 executes a detection process to detect components to be inspected contained in the test sample based on the absorption spectrum after preprocessing in step S5 (step S6). The control unit 21 functions as a detection unit. Step S6 is a detection step. The components to be inspected include active pharmaceutical ingredients, excipients, water, etc. When the test sample has locally different components, the control unit 21 may perform the following process in step S6. Specifically, the control unit 21 generates spectroscopic imaging data from the measurement data of the test sample acquired in step S3. The control unit 21 then executes a detection process using the generated spectroscopic imaging data. Three examples for detecting a target component contained in a target sample will be described below.

[0051] (Detection Processing Example 1) In Example 1 of the detection process, the control unit 21 detects the test component by determining whether the test sample contains the test component. For example, the control unit 21 determines whether the test sample contains a desired active pharmaceutical ingredient (API) and excipients. Alternatively, the control unit 21 determines whether the test sample contains an unintended component. Specifically, the control unit 21 compares the reference absorption spectrum with the absorption spectrum of the test sample after the pretreatment in step S5. The control unit 21 performs the above-mentioned determination in Example 1 of the detection process based on the coincidence rate between the reference absorption spectrum and the absorption spectrum of the test sample after the pretreatment in step S5. The reference absorption spectrum is the absorption spectrum of components of multiple reference samples according to the type of preparation. FIG. 9 shows an example of comparing the reference absorption spectrum with the absorption spectrum of the test sample after the pre-processing of step S5. Graph (i) in FIG. 9 is the absorption spectrum of the test sample after the pre-processing of step S5. Graphs (ii) and (iii) in FIG. 9 are graphs of the reference absorption spectrum stored in, for example, the memory unit 22. In the example shown in FIG. 9, the agreement rate between graph (i) and graph (ii) is less than a predetermined threshold, and the agreement rate between graph (i) and graph (iii) is equal to or greater than the predetermined threshold. In this case, the control unit 21 determines that the test sample does not contain the test target component contained in the reference sample corresponding to graph (ii). The control unit 21 determines that the test sample contains the test target component contained in the reference sample corresponding to graph (iii). This makes it possible to detect contamination in the manufacturing process of the test sample, unintended component mutations in the test sample, etc., thereby contributing to the production of highly safe samples. By carrying out the above-mentioned determination in Example 1 of the detection process on a preparation of an unknown component as the test sample, the product to which the test sample corresponds can be identified.

[0052] (Detection Process Example 2) In Example 2 of the detection process, the control unit 21 constructs a calibration curve for the amounts of components (active ingredient amounts, moisture amounts, etc.) contained in the test sample. The control unit 21 quantifies the components contained in the test sample based on the calibration curve. The following describes the process for constructing a machine learning model as a calibration curve for the amount of water contained in a test sample. First, the user prepares a plurality of samples with different moisture contents as correct value samples. Next, the user photographs the prepared plurality of samples using the imaging unit 12. Next, the control unit 21 performs a moisture quantification analysis by chemical analysis (Karl Fischer method) on the measurement data of the multiple samples acquired from the imaging unit 12. Next, the control unit 21 constructs a calibration model by regression analysis such as PLS (Partial Least Squares Regression) using the data set obtained by the moisture content analysis (or data in a predetermined region of the data set). At this time, the control unit 21 may construct the calibration model using principal component regression, deep learning, or the like. When constructing the calibration model, the control unit 21 may remove noise by cleansing the data set obtained by the moisture content analysis. Next, the user photographs the test sample using the imaging unit 12. Next, the control unit 21 performs steps S4 and S5 of the quality inspection process described above to calculate and correct an absorption spectrum from the measurement data of the test sample acquired from the imaging unit 12. Next, the control unit 21 applies the absorption spectrum of the test sample to the calibration model constructed above to calculate the amount of the test target component contained in the test sample (performs quantitative analysis). At this time, the control unit 21 may not only output one quantitative analysis result for each test sample, but may also perform quantitative analysis on each frame photographed by the imaging unit 12. In this way, the control unit 21 may generate a distribution of the component amounts of the components contained in the test sample.

[0053] (Detection Process Example 3) In the third embodiment of the detection process, the control unit 21 detects an abnormal sample. A method for detecting an abnormal sample will be described below. First, the user prepares a plurality of specimens to be inspected, and then the user photographs the prepared specimens to be inspected using the imaging unit 12. Next, the control unit 21 performs steps S4 and S5 of the quality inspection process described above to calculate and correct absorption spectra from the measurement data of the multiple test samples acquired from the imaging unit 12. Next, the control unit 21 compares the absorption spectra and detects outlier absorption spectra. An outlier absorption spectrum is an absorption spectrum whose shape matches other absorption spectra less than a predetermined threshold. Next, the control unit 21 determines that the test sample corresponding to the absorption spectrum detected as an outlier is an abnormal sample. At this time, the control unit 21 may detect outliers by statistical processing, or may detect outliers using deep learning or the like.

[0054] Fig. 10 shows an example of a comparison of multiple absorption spectra (Abs1 to Abs4). In the example shown in Fig. 10, absorption spectra Abs2 to Abs4 have approximately the same shape and overlap. The shape of absorption spectrum Abs1 matches the shapes of absorption spectra Abs2 to Abs4 to a degree lower than a predetermined threshold. Therefore, in this case, the control unit 21 determines that the sample corresponding to absorption spectrum Abs1 is an abnormal sample.

[0055] Alternatively, in the third embodiment of the analysis process, the control unit 21 may detect abnormal locations within the same sample. A method for detecting abnormal locations within the same sample will be described below. First, the user uses the imaging unit 12 to take images at multiple measurement positions within the same sample. Next, the control unit 21 performs steps S4 and S5 of the quality inspection process to calculate and correct absorption spectra from the measurement data at each of the measurement positions acquired from the imaging unit 12. Next, the control unit 21 compares the absorption spectra and detects any outlier absorption spectra. Next, the control unit 21 determines that the measurement position corresponding to the absorption spectrum detected as an outlier is an abnormal position.

[0056] In the above quality inspection process, by measuring the transmitted and diffused light using the imaging unit 12, it is possible to measure information about the interior of the sample, rather than measuring only components on or near the surface of the sample as in the case of reflected light measurement. This allows for more precise inspection of the quality of the sample. The amount of moisture in freeze-dried preparations or powdered pharmaceuticals is very small. Therefore, in the above quality inspection process, by measuring the transmitted and diffused light, which has a longer optical path length, it is possible to more precisely detect the amount of moisture, even if it is very small. In the quality inspection process, information about internal defects in the sample can be measured by measuring the light scattering characteristics using the imaging unit 12, so that internal defects in the sample can be detected by the quality inspection process. Since the light scattering characteristics depend on the scattering characteristics of the sample and the absorption characteristics of the components contained in the sample, the components of the sample can be detected by the quality inspection process.

[0057] [others] During measurements in steps S2 and S3 of the quality inspection process, the illumination range of the illumination unit 13 illuminating the bottom surface of the transparent container B may be adjusted. In this embodiment, the illumination range can be adjusted to the entire bottom surface of the transparent container B or only the area near the center of the transparent container B. Specifically, as shown in FIG. 11A, the illumination range can be increased by increasing the distance from the tip of the illumination light 131 to the bottom surface of the transparent container B. On the other hand, as shown in FIG. 11B, the illumination range can be reduced by reducing the distance from the tip of the illumination light 131 to the bottom surface of the transparent container B. Alternatively, the illumination range can be increased by guiding the illumination light through all of the optical fibers of the illumination light 131 and irradiating it onto the bottom surface of the transparent container B. On the other hand, the illumination range can be reduced by guiding the illumination light through only the optical fibers located in the center of the optical fibers of the illumination light 131 and irradiating it onto the bottom surface of the transparent container B. The length of the optical path F of the illumination light passing through the sample S can be adjusted by adjusting the illumination range of the illumination unit 13 that illuminates the bottom surface of the transparent container B. This makes it possible to set the length of the optical path F appropriate for the absorbance of the sample S. For example, if the absorption of the illumination light by the sample S is significantly small, the amount of illumination light absorbed by the sample S can be increased by increasing the length of the optical path F of the illumination light.

[0058] [effect] The quality inspection device 1 of this embodiment is a quality inspection device that inspects the quality of an inspection target sample, which is a freeze-dried preparation or a powdered pharmaceutical product filled in a transparent container B. The quality inspection device 1 includes an illumination unit 13 that illuminates the bottom surface of the transparent container B. The quality inspection device 1 includes a multispectral light receiving unit (imaging unit 12) that captures an image of the side surface of the transparent container B. The quality inspection device 1 includes a calculation unit (control unit 21) that calculates absorption spectra at multiple wavelengths based on the light reception results from the multispectral light receiving unit. The quality inspection device 1 also includes a detection unit (control unit 21) that detects the inspection target component contained in the inspection target sample based on the absorption spectrum calculated by the calculation unit. Therefore, it is possible to detect the target component contained in the test sample without destroying the test sample. As a result, it is possible to inspect all freeze-dried preparations or pharmaceuticals during their production. In other words, it is possible to inspect the quality of freeze-dried preparations or powdered pharmaceuticals more precisely.

[0059] In the quality inspection device 1 of this embodiment, the detection section (control section 21) detects the components to be inspected by comparing the absorption spectrum of the reference sample with the absorption spectrum of the inspection target sample. This makes it possible to detect contamination during the manufacturing process of the test sample, unintended component mutations in the test sample, etc., contributing to the production of highly safe samples. It is also possible to identify products that correspond to formulations with unknown components as test samples.

[0060] In the quality inspection device 1 of this embodiment, the detection unit (control unit 21) calculates the amount of the component to be inspected by applying the absorption spectrum of the inspection target sample to a calibration model constructed in advance. This allows quantitative detection of the components to be tested, allowing for more precise testing of the quality of the test sample.

[0061] In the quality inspection device 1 of this embodiment, the components to be inspected are at least one of an active pharmaceutical ingredient, an excipient, and water. This makes it possible to detect at least one of the active pharmaceutical ingredient, excipient, and water contained in the test sample, thereby enabling more precise testing of the quality of the test sample.

[0062] In the quality inspection device 1 of this embodiment, the detection unit (control unit 21) compares the shapes of the absorption spectra of multiple test samples and determines that the test sample corresponding to the absorption spectrum whose shape matches with other absorption spectra less than a predetermined threshold is abnormal. This allows the test sample to be easily inspected without constructing a calibration curve or preparing absorption spectra of multiple reference samples in advance.

[0063] In the quality inspection device 1 of this embodiment, the multispectral light receiving unit (imaging unit 12) is a camera or a multichannel spectroscope, and captures images at multiple points in the vertical direction of the transparent container B. The calculation unit (control unit 21) calculates the absorption spectrum based on the relative amount of received light at one point relative to the amount of received light at the other point, out of the light receiving results at the multiple points. This makes it possible to eliminate the influence of the light source 132 on the diffuse transmittance of the test sample, thereby enabling evaluation in the detection process of the test target component to be performed focusing only on the absorption of illumination light by the test sample between the calibration point (reference position) and other measurement points (evaluation positions).

[0064] The quality inspection device 1 of this embodiment includes a reference light-receiving unit that measures the amount of light emitted from the light source 132 of the illumination unit 13. The calculation unit (control unit 21) calculates the absorption spectrum based on the amount of light emitted measured by the reference light-receiving unit. This makes it possible to inspect the inspection target sample without measuring the inspection target sample at a plurality of measurement points.

[0065] In the quality inspection device 1 of this embodiment, the illumination range in which the illumination unit 13 illuminates the bottom surface of the transparent container B is adjustable. This allows the length of the optical path F to be set to a value suitable for the absorbance of the sample S.

[0066] In the quality inspection device 1 of this embodiment, the multispectral light receiving unit (imaging unit 12) is a hyperspectral camera. This allows the hyperspectral camera to capture images of objects at multiple wavelengths with high resolution, enabling more precise inspection of the quality of freeze-dried preparations or powdered pharmaceuticals based on the measurement data obtained through the imaging.

[0067] While the preferred embodiments of the present disclosure have been described in detail above with reference to the accompanying drawings, the technical scope of the present disclosure is not limited to such examples. Furthermore, various modifications and improvements made by those skilled in the art will naturally fall within the scope of the technical ideas described in the claims. [Explanation of symbols]

[0068] 1. Quality inspection equipment 10 Inspection Unit 11 Rotating part 12 Imaging unit 13 Lighting Department 131 Lighting 132 Light source 20 Processing equipment 21 control unit (calculation unit, detection unit) 22 Memory section 22a Program 23 Communications Department 24 Operation Department 25 Display unit B Transparent container

Claims

1. A quality inspection device for inspecting the quality of a sample to be inspected, which is a freeze-dried preparation or a powdered pharmaceutical product filled in a transparent container, an illumination unit that illuminates the bottom surface of the transparent container; a multispectral light receiving unit that captures an image of the side surface of the transparent container; a calculation unit that calculates an absorption spectrum at a plurality of wavelengths based on the light reception results by the multispectral light receiving unit; a detection unit that detects a target component contained in the test sample based on the absorption spectrum calculated by the calculation unit.

2. 2. The quality inspection device according to claim 1, wherein the detection unit detects the target component by comparing an absorption spectrum of a reference sample with an absorption spectrum of the target sample.

3. The quality inspection device according to claim 1 , wherein the detection unit calculates the amount of the target component by applying the absorption spectrum to a calibration model that has been constructed in advance.

4. 2. The quality inspection device according to claim 1, wherein the component to be inspected is at least one of an active pharmaceutical ingredient, an excipient, and water.

5. The quality inspection device according to claim 1, wherein the detection unit compares the shapes of the absorption spectra of the plurality of test samples and determines that the test sample corresponding to an absorption spectrum whose shape matches with other absorption spectra less than a predetermined threshold value is abnormal.

6. the multispectral light receiving unit is a camera or a multichannel spectrometer, and captures images at multiple points in the vertical direction of the transparent container; The quality inspection device according to claim 2 , wherein the calculation unit calculates the absorption spectrum based on a relative amount of received light at one point relative to the amount of received light at the other point, among the light reception results at the plurality of points.

7. a reference light receiving unit that measures the amount of light emitted from the light source of the illumination unit; The quality inspection device according to claim 2 , wherein the calculation unit calculates the absorption spectrum based on the amount of light emitted measured by the reference light receiving unit.

8. The quality inspection device according to claim 1 , wherein the illumination range of the bottom surface of the transparent container illuminated by the illumination unit is adjustable.

9. The quality inspection device according to claim 1 , wherein the multispectral light receiving unit is a hyperspectral camera.

10. A quality inspection device for inspecting the quality of a sample to be inspected, which is a freeze-dried preparation or a powdered pharmaceutical product filled in a transparent container, an illumination unit that illuminates the bottom surface of the transparent container; a multispectral light receiving unit that captures an image of a side surface of the transparent container, a calculation step of calculating an absorption spectrum at a plurality of wavelengths based on the light reception results by the multispectral light receiving unit; a detection step of detecting a target component contained in the test sample based on the absorption spectrum calculated in the calculation step.

11. A quality inspection device for inspecting the quality of a sample to be inspected, which is a freeze-dried preparation or a powdered pharmaceutical product filled in a transparent container, an illumination unit that illuminates the bottom surface of the transparent container; a multispectral light receiving unit that captures an image of the side surface of the transparent container; a calculation unit that calculates an absorption spectrum at a plurality of wavelengths based on the light reception results by the multispectral light receiving unit; a program that causes the detector to function as a detector that detects a target component contained in the test sample based on the absorption spectrum calculated by the calculator;

Images