Drug quality data processing method and system based on visual assistance
Through visually assisted drug quality data processing methods, combined with quantitative imaging and multi-level microbial detection processes, the problems of single and low efficiency of traditional drug quality data collection methods are solved, and more efficient and accurate drug quality testing is achieved.
Patent Information
- Application Number
- CN202510023813.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-07
- Publication Date
- 2025-03-28
AI Technical Summary
Traditional drug quality data collection methods are single, and lack of systematic data integration testing, resulting in information silos and reducing the efficiency and accuracy of drug quality testing.
The visually assisted drug quality data processing method is adopted, and the accuracy and efficiency of drug quality detection is significantly improved through quantitative imaging errors and multi-level microbial detection processes, including initial appearance quality inspection, microbial limit re-examination and final dissolution effect inspection.
It improves the accuracy, efficiency and reliability of drug quality testing, reduces the testing cost and cycle, and ensures that the results of each testing link are rigorously evaluated.
Smart Images
Figure CN119715563A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of image analysis, and specifically relates to a drug quality data processing method and system based on vision assistance. Background Art
[0002] As the supervision of drug quality becomes increasingly strict, pharmaceutical companies are facing greater demands for drug quality and efficacy monitoring and must process more drug test data. In this context, pharmaceutical companies need to improve their ability to collect and process drug data to ensure the compliance and safety of drug quality testing. However, traditional drug quality data collection methods are often relatively simple and lack systematic data integration testing, resulting in different quality indicators being tested independently by different laboratories or departments, forming information islands. This phenomenon not only reduces data sharing and sharing, but also leads to a lack of comprehensiveness in the comprehensive evaluation of drug quality and an inability to truly reflect the logic of drug use.
[0003] In addition, existing drug quality data processing methods are also inefficient and inaccurate. For drugs that have failed certain quality indicators, other indicators still need to be tested repeatedly, which not only wastes manpower and laboratory resources, but also prolongs the overall testing cycle and reduces testing efficiency. Summary of the invention
[0004] In view of the shortcomings of the prior art, the present invention provides a drug quality data processing method and system based on vision assistance, which significantly improves the accuracy, efficiency and reliability of drug quality detection by quantifying imaging errors and multi-level microbial detection processes.
[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: The drug quality data processing method based on visual assistance includes the following steps: Obtaining a data set of preliminary inspection of the appearance quality of the drugs to be inspected, and obtaining preliminary inspection judgment results of the drugs to be inspected based on the data set of preliminary inspection of the appearance quality, wherein the preliminary inspection judgment results include unqualified appearance quality and qualified appearance quality; If the initial inspection result shows that the appearance quality is unqualified, the drug to be inspected will be marked as unqualified; If the initial inspection result shows that the appearance quality is qualified, the drug to be inspected will be marked as a drug to be re-inspected; Determine whether the drug to be retested is qualified in terms of microbial limit quality. If the microbial limit quality is unqualified, mark the drug to be retested as unqualified drug; if the microbial limit quality is qualified, mark the drug to be retested as drug to be final tested; Conduct dissolution effect test on the drug to be final inspected. If the dissolution effect test fails, the drug to be final inspected is marked as an unqualified drug. If the dissolution effect test passes, the drug to be final inspected is marked as a qualified drug.
[0006] Preferably, the process of obtaining the preliminary inspection result of the drug to be inspected based on the appearance quality preliminary inspection data set is: Perform high-resolution image scanning on the drugs to be inspected and collect images of the drugs to be inspected; Based on the imaging of the drugs to be inspected, the initial inspection data set of appearance quality is obtained, including the initial inspection imaging correction set and the appearance quality evaluation set. The initial inspection imaging correction set includes scanning illumination uniformity, reflective area ratio, and initial inspection exposure deviation time. The appearance quality evaluation set includes the drug surface defect ratio, the minimum circumscribed rectangle cumulative deviation area, the surface deviation color ratio, and the cumulative deviation perimeter of the drug contour; Obtaining a preliminary inspection imaging quality assessment value based on a preliminary inspection imaging correction set; Matching the initial inspection imaging correction index stored in the database based on the initial inspection imaging quality assessment value; Obtaining an appearance quality assessment value based on an appearance quality assessment set and a preliminary inspection imaging correction index; Obtain the appearance quality standard value stored in the database; Compare the appearance quality standard value and appearance quality evaluation value stored in the database to obtain the initial inspection result: If the appearance quality evaluation value is greater than or equal to the appearance quality standard value stored in the database, the initial inspection result is that the appearance quality is qualified; If the appearance quality evaluation value is less than the appearance quality standard value stored in the database, the initial inspection result is that the appearance quality is unqualified.
[0007] Preferably, the calculation formula for the appearance quality evaluation value is: ; Where y is the appearance quality assessment value, q is the proportion of drug surface defects, j is the cumulative deviation area of the minimum circumscribed rectangle, b is the proportion of surface deviation color, l is the cumulative deviation perimeter of the drug contour, is the initial imaging correction index, e is a natural constant, and lg represents the logarithm with base 10.
[0008] Preferably, the process of determining whether the drug to be retested is of qualified microbial limit quality is: For the drugs to be retested, sample inoculation and colony counting are performed to obtain a colony statistics data set, including colony culture time, colony culture temperature, total colony count, and number of colony types; Comprehensively analyze the colony statistics data set to obtain the colony statistics results, and determine whether to perform pathogen screening based on the colony statistics results. The colony statistics results include colony counts exceeding the standard and colony counts not exceeding the standard; If the colony count result shows that the colony exceeds the standard, the microbial limit quality is unqualified and the drug to be retested is marked as unqualified drug; If the colony count result shows that the colony does not exceed the standard, the drug to be retested will be screened for pathogenic bacteria to obtain the retest judgment result: If the re-test result is that the drug contains pathogenic bacteria, the microbial limit quality is unqualified and the drug to be re-tested is marked as unqualified; If the re-inspection result shows that it does not contain pathogenic bacteria, the microbial limit quality is qualified and the drug to be re-inspected is marked as a drug to be final inspected.
[0009] Preferably, the process of comprehensively analyzing the colony statistics data set to obtain the colony statistics results is: Get colony statistics dataset; Based on the colony culture time and colony culture temperature, the total number of standard colonies and the number of standard colony types stored in the matching database; Obtaining a deviation rate of the total number of colonies based on the total number of colonies and the standard total number of colonies stored in a database; Obtaining a colony type quantity deviation rate based on the colony type quantity and the standard colony type quantity stored in a database; Obtain the total colony count deviation threshold and colony type count deviation threshold stored in the database; Determine whether the total colony count deviation rate is less than the total colony count deviation threshold: If not, the colony count result is that the colony exceeds the limit; If yes, determine whether the colony type quantity deviation rate is less than the colony type quantity deviation threshold: If yes, the colony count result is that the colony does not exceed the standard; If not, the colony count result is that the colony count exceeds the standard.
[0010] Preferably, the process of testing and judging the dissolution effect of the drug to be finally inspected is: Conduct dissolution effect testing on the drugs to be final tested and obtain a set of dissolution effect feedback graphs; The dissolution effect feedback graph group and the dissolution effect standard graph stored in the database are comprehensively compared to obtain the dissolution effect quality final inspection data set, including the final inspection imaging correction set and the dissolution characteristic data set. The final inspection imaging correction set includes scanning light uniformity, reflective area ratio, and final inspection exposure deviation time. The dissolution characteristic data set includes drug dissolution diffusion deviation, dissolution start time, undissolved deviation volume, and initial dissolution rate. Obtaining a dissolution effect quality evaluation value based on the dissolution effect quality final inspection data set, and obtaining a final inspection judgment result based on the dissolution effect quality evaluation value, wherein the final inspection judgment result includes whether the dissolution effect quality is unqualified or qualified; If the final inspection result is that the dissolution effect quality is unqualified, the drug with qualified microbial limit quality awaiting final inspection shall be marked as a drug with unqualified dissolution effect quality; If the final inspection result is that the dissolution effect quality is qualified, then the drug awaiting final inspection with the microbial limit quality qualified will be marked as a qualified drug.
[0011] Preferably, the calculation formula for the dissolution effect quality assessment value is: ; In the formula, r is the dissolution effect quality assessment value, k is the drug dissolution diffusion deviation, c is the start of dissolution time, n is the undissolved deviation volume, v is the initial dissolution rate, is the final inspection imaging correction index, e is a natural constant, lg represents the logarithm with base 10, and ln represents the logarithm with base e.
[0012] Preferably, the process of obtaining the final inspection judgment result based on the dissolution effect quality evaluation value is: Obtain the final quality inspection data set for dissolution effect; Calculate the final inspection imaging quality assessment value based on the final inspection imaging correction set; Matching the final inspection imaging correction index stored in the database based on the final inspection imaging quality assessment value; Obtain dissolution effect quality assessment value based on in vitro dissolution characteristic data set and final inspection imaging correction index; Obtaining the dissolution effect quality standard value stored in the database; If the dissolution effect quality evaluation value is greater than or equal to the dissolution effect quality standard value stored in the database, the final inspection judgment result is that the dissolution effect quality is qualified; If the dissolution effect quality assessment value is less than the dissolution effect quality standard value stored in the database, the final inspection judgment result is that the dissolution effect quality is unqualified.
[0013] Preferably, the process of comprehensively comparing the dissolution effect feedback graph group and the dissolution effect standard graph stored in the database to obtain the dissolution effect quality final inspection data set is: Obtaining a dissolution effect feedback graph group, which is a collection of dissolution state images of the drug at each standard interval time point within the detection time; Obtain the dissolution state imaging that first presents the dissolution state in the dissolution effect feedback graph group, mark the standard interval time point corresponding to the dissolution state imaging that first presents the dissolution state as the start dissolution time, and take the ratio of the drug dissolution volume from the standard interval time point corresponding to the start dissolution time to the next standard interval time point to the interval time between the two standard interval time points as the initial dissolution rate; The dissolution state imaging at the end of the detection time in the dissolution effect feedback graph group is obtained, and mapped in the dissolution effect standard graph stored in the database, the absolute value of the drug dissolution diffusion volume difference between the two images and the ratio of the standard dissolution diffusion volume in the dissolution effect standard graph are obtained as the drug dissolution diffusion deviation, and the absolute value of the volume difference of the remaining undissolved drug between the two images are obtained as the undissolved deviation volume.
[0014] The drug quality data processing system based on visual assistance is used to implement the above method, including a drug appearance initial inspection module, a microbial limit re-inspection module and a dissolution effect final inspection module, wherein: The drug appearance initial inspection module is used to obtain the appearance quality initial inspection data set of the drug to be inspected, and obtain the initial inspection judgment result of the drug to be inspected based on the appearance quality initial inspection data set. The initial inspection judgment result includes unqualified appearance quality and qualified appearance quality; If the initial inspection result shows that the appearance quality is unqualified, the drug to be inspected will be marked as unqualified; If the initial inspection result shows that the appearance quality is qualified, the drug to be inspected will be marked as a drug to be re-inspected; The microbial limit re-test module is used to determine whether the microbial limit quality of the drug to be re-tested is qualified. If the microbial limit quality is unqualified, the drug to be re-tested is marked as an unqualified drug. If the microbial limit quality is qualified, the drug to be re-tested is marked as a drug to be final tested. The dissolution effect final inspection module is used to perform dissolution effect inspection on the drugs to be inspected. If the dissolution effect inspection fails, the drugs to be inspected are marked as unqualified drugs. If the dissolution effect inspection passes, the drugs to be inspected are marked as qualified drugs.
[0015] The present invention has the following beneficial effects: By quantifying the impact of imaging effects on drug detection quality assessment, the present invention can refine the imaging correction data set in different detection stages and environments of drug quality data collection, which not only improves the accuracy of data collection, but also makes the detection results more transparent and reliable, thereby effectively improving the overall level of drug quality detection.
[0016] The present invention adopts a multi-level detection process in the microbial limit detection. First, colony statistics are performed. If the results meet the standards, pathogen screening is performed. This strategy not only saves detection costs and shortens the overall detection cycle, but also ensures that the results of each link are rigorously evaluated, thereby improving the overall quality of the data and ensuring the reliability of the final test results. In addition, the visual analysis of the dissolution effect provides the dissolution state of the drug at different time points, enhances the understanding of the dissolution characteristics of the drug, and further improves the accuracy and reliability of drug quality data processing. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 It is a schematic diagram of the method steps of the present invention; Figure 2 It is a schematic diagram of system module connection of the present invention. DETAILED DESCRIPTION
[0018] The technical solutions in the embodiments of the present invention are described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.
[0019] like Figure 1 As shown, the method for processing drug quality data based on visual assistance includes the following steps: obtaining a data set of preliminary inspection of the appearance quality of the drugs to be inspected, and obtaining preliminary inspection judgment results of the drugs to be inspected based on the preliminary inspection data set of appearance quality, wherein the preliminary inspection judgment results include unqualified appearance quality and qualified appearance quality.
[0020] The specific analysis process is as follows: use a high-resolution image scanner to perform high-resolution image scanning on the drugs to be inspected to ensure that the subtle features of the drugs to be inspected can be captured, place the drugs to be inspected on a smooth and flat surface, use a fixed fixture to ensure that the drugs to be inspected do not move during the scanning process, obtain a high-resolution image of the drugs to be inspected, and mark it as imaging of the drugs to be inspected.
[0021] Based on the imaging of the drug to be inspected, the appearance quality initial inspection data set is obtained. The appearance quality initial inspection data set includes the initial inspection imaging correction set and the appearance quality assessment set. The initial inspection imaging correction set includes scanning illumination uniformity, reflective area ratio, and initial inspection exposure deviation time. Among them, scanning illumination uniformity refers to the uniformity of the light source irradiating the drug surface during image scanning; reflective area ratio refers to the proportion of the reflective area on the drug surface during the scanning process, and its value is the ratio of the area of the highlight spots in the imaging of the drug to be inspected to the total imaging area of the drug to be inspected; initial inspection exposure deviation time refers to the deviation between the exposure time and the set ideal exposure time during the initial inspection image scanning process.
[0022] The appearance quality assessment set includes the proportion of drug surface defects, the cumulative deviation area of the minimum circumscribed rectangle, the proportion of surface deviation colors, and the cumulative deviation perimeter of the drug contour. Among them, the proportion of drug surface defects refers to the ratio between the defect area found in the drug surface inspection and the total surface area of the drug; the cumulative deviation area of the minimum circumscribed rectangle refers to the area difference between the actual outer contour of the drug and the minimum rectangular frame enclosing the contour of the drug, which is used to evaluate the regularity of the drug shape; the proportion of surface deviation colors refers to the proportion of the color deviation area on the drug surface compared with the standard color; the cumulative deviation perimeter of the drug contour is to extract the drug contour using Canny edge detection or Sobel operator, extract the boundary of the drug through the contour detection algorithm, and describe the perimeter of the drug contour, and then calculate the perimeter difference between the actual contour of the drug and the theoretical standard contour, which is used to evaluate the integrity and regularity of the drug shape.
[0023] The initial inspection imaging quality assessment value is obtained based on the initial inspection imaging correction set, and its calculation formula is: ; In the formula, is the initial inspection image quality evaluation value, g is the scanning light uniformity, f is the reflective area ratio, is the initial inspection exposure deviation time, e is a natural constant, and ln represents the logarithm with e as the base.
[0024] Used to quantitatively describe the image quality of the corresponding drug imaging during the initial inspection process. The larger the g, the higher the image quality of the drug and the better the imaging effect. Uniform lighting can ensure that all parts of the drug surface can be uniformly exposed. A higher g can make the image details clearer and the colors more realistic. A lower f has fewer reflective areas, which means fewer highlight spots in the image, which can better show the true color and details of the drug, thereby improving Appropriate exposure time can capture the details of the medicine, ensure that the image brightness is appropriate and the details are clear, thereby improving .
[0025] The preliminary inspection imaging correction index stored in the database is matched based on the preliminary inspection imaging quality assessment value.
[0026] The initial inspection image quality assessment value is set as the initial inspection designated label, and the initial inspection designated label is compared with each initial inspection setting label stored in the database to determine the initial inspection setting label of the initial inspection designated label system, and obtain the initial inspection imaging correction index corresponding to the initial inspection setting label. By automating the data comparison, the evaluation efficiency is improved, and the time and errors of manual operation to obtain the initial inspection judgment results are reduced. In addition, the use of the appearance quality standard value stored in a fixed database for evaluation can ensure the consistency of the initial inspection judgment results made by different evaluators at different times. This repeatability helps to form a stable processing flow in the processing of drug quality data.
[0027] Obtain an appearance quality assessment value based on an appearance quality assessment set and an initial inspection imaging correction index; obtain an appearance quality standard value stored in a database; compare the appearance quality standard value stored in the database with the appearance quality assessment value to obtain an initial inspection judgment result: if the appearance quality assessment value is greater than or equal to the appearance quality standard value stored in the database, the initial inspection judgment result is that the appearance quality is qualified; if the appearance quality assessment value is less than the appearance quality standard value stored in the database, the initial inspection judgment result is that the appearance quality is unqualified.
[0028] The calculation formula for the appearance quality evaluation value is: ; Where y is the appearance quality assessment value, q is the proportion of drug surface defects, j is the cumulative deviation area of the minimum circumscribed rectangle, b is the proportion of surface deviation color, l is the cumulative deviation perimeter of the drug contour, It is the correction index of the initial inspection image, and lg represents the logarithm with base 10.
[0029] y is used to describe the initial inspection results of the drugs to be inspected, that is, the appearance quality of the drugs to be inspected. The larger y is, the better the appearance quality of the drugs to be inspected. A higher q means that there are obvious flaws in the appearance of the drugs, and the appearance quality of the drugs to be inspected is poor, which will directly reduce y. A smaller j means that the outline of the drug is more regular and meets the standard shape, which can improve y. A higher b means that there is a high proportion of color deviation areas, which will cause the appearance of the drug to be inconsistent with the standard, thereby reducing y. A smaller l means that the edges of the drug are clear and meet the expected standards, reflecting the integrity of the appearance of the drug, which helps to improve y.
[0030] Determine whether the drug to be retested meets the microbial limit quality standards. If the microbial limit quality standards are unqualified, mark the drug to be retested as an unqualified drug. If the microbial limit quality standards are qualified, mark the drug to be retested as a drug to be final inspected.
[0031] The specific analysis process is as follows: sample inoculation and colony counting are performed on the drugs to be retested to obtain a colony statistics data set, including colony culture time, colony culture temperature, total colony count, and number of colony types.
[0032] Comprehensively analyze the colony statistics data set to obtain the colony statistics results, and judge whether to perform pathogenic bacteria screening based on the colony statistics results. The colony statistics results include colony exceeding the standard and colony not exceeding the standard. If the colony statistics result is that the colony exceeds the standard, the drug to be re-tested is marked as an unqualified drug. If the colony statistics result is that the colony does not exceed the standard, pathogenic bacteria screening is performed on the drug to be re-tested to obtain the re-test judgment result: If the re-test judgment result is that pathogenic bacteria are contained, the microbial limit quality is unqualified, and the drug to be re-tested is marked as an unqualified drug; if the re-test judgment result is that pathogenic bacteria are contained, the microbial limit quality is qualified, and the drug to be re-tested is marked as a drug to be final inspected.
[0033] Sample inoculation and colony counting for drugs to be retested means selecting appropriate culture media (such as nutrient agar, malt agar, etc.) according to the characteristics of the drugs to be retested. The culture media should be able to effectively support the growth of the microorganisms to be tested. Use a sterile inoculation loop or inoculation stick to inoculate or streak on the surface of the culture media to ensure that the microorganisms are evenly distributed. Place the inoculated culture media in a constant temperature incubator and culture it at a specific culture temperature and culture time. After the culture is completed, open the incubator, observe the colony growth on the surface of the culture media, and count the target data.
[0034] Screening for pathogenic bacteria on drugs to be retested involves extracting the colonies that appear from the culture medium, using aseptic techniques to transfer different types of colonies to a slide for subsequent staining, applying Margolan stain, evenly smearing to cover the sample, rinsing the slide with a buffer solution or distilled water after a regular reaction time, gently removing unbound dye, observing the colored portion using a microscope, recording the condition of the colored portion, observing the colored portion, and determining its quantity and distribution.
[0035] The specific analysis process of comprehensively analyzing the colony statistics data set to obtain the colony statistics results is as follows: Obtain a colony statistics data set; match the standard total colony count and the standard colony type number stored in the database based on the colony culture time and the colony culture temperature; obtain the total colony count deviation rate based on the total colony count and the standard total colony count stored in the database; obtain the colony type number deviation rate based on the colony type number and the standard colony type number stored in the database; obtain the total colony count deviation threshold and the colony type number deviation threshold stored in the database; determine whether the total colony count deviation rate is less than the total colony count deviation threshold: if not, the colony statistics result is that the colonies exceed the standard; if so, determine whether the colony type number deviation rate is less than the colony type number deviation threshold: if so, the colony statistics result is that the colonies do not exceed the standard; if not, the colony statistics result is that the colonies exceed the standard.
[0036] The colony culture time and colony culture temperature can comprehensively describe the current colony culture mode, and the matched standard colony total number and standard colony type number are more in line with the current colony culture status. The total colony count deviation rate is the ratio of the absolute value of the difference between the total colony count and the standard colony total number to the standard colony total number. The colony type number deviation rate is the ratio of the absolute value of the difference between the colony type number and the standard colony type number to the standard colony type number.
[0037] Conduct dissolution effect test on the drug to be final inspected. If the dissolution effect test fails, the drug to be final inspected is marked as an unqualified drug. If the dissolution effect test passes, the drug to be final inspected is marked as a qualified drug.
[0038] The specific analysis process is as follows: conduct dissolution effect testing on the drugs to be final inspected and obtain a dissolution effect feedback graph group; comprehensively compare the dissolution effect feedback graph group and the dissolution effect standard graph stored in the database to obtain the dissolution effect quality final inspection data set, including a final inspection imaging correction set and a dissolution characteristic data set. The final inspection imaging correction set includes scanning light uniformity, reflective area ratio, and final inspection exposure deviation time.
[0039] Among them, the dissolution effect test for the drug to be final inspected and the dissolution effect feedback graph group obtained refer to the dissolution state imaging set at each standard interval time point within the detection time. The scanning light uniformity and the reflective area ratio contained in the initial inspection imaging correction set and the final inspection imaging correction set have the same meaning, but the properties of the initial inspection and final inspection imaging objects are different, so the exposure time standards are different, which are divided into initial inspection exposure deviation time and final inspection exposure deviation time. The final inspection exposure deviation time refers to the deviation between the exposure time and the set ideal exposure time during the final inspection image scanning process.
[0040] The dissolution characteristic data set includes drug dissolution diffusion deviation, dissolution start time, undissolved deviation volume, and initial dissolution rate. The dissolution effect quality evaluation value is obtained based on the dissolution effect quality final inspection data set. The final inspection judgment result is obtained based on the dissolution effect quality evaluation value. The final inspection judgment result includes dissolution effect quality unqualified and dissolution effect quality qualified. If the final inspection judgment result is that the dissolution effect quality is unqualified, the microbial limit qualified drug awaiting final inspection is marked as a dissolution effect quality unqualified drug; if the final inspection judgment result is that the dissolution effect quality is qualified, the microbial limit qualified drug awaiting final inspection is marked as a qualified drug.
[0041] The calculation formula of dissolution effect quality assessment value is: ; In the formula, r is the dissolution effect quality assessment value, k is the drug dissolution diffusion deviation, c is the start of dissolution time, n is the undissolved deviation volume, v is the initial dissolution rate, It is the final inspection imaging correction index.
[0042] r is used to describe the final result of the drug to be tested, that is, to evaluate whether the dissolution effect quality represented by the dissolution effect level of the drug to be tested is qualified. The larger the r, the better the dissolution effect of the drug to be tested. The shorter the c, the shorter the time for the drug to start releasing and dissolving, and the faster the dissolution process. The higher the v, the faster the drug is released in the initial stage and can be dissolved more effectively. The smaller the k, the closer the dissolution process of the drug is to the standard situation, and the better the dissolution effect. The smaller the n, the better the dissolution effect of the drug and the less undissolved drug.
[0043] The final inspection judgment result is obtained based on the dissolution effect quality assessment value. The specific analysis process is: obtaining the dissolution effect quality final inspection data set; calculating the final inspection imaging quality assessment value based on the final inspection imaging correction set; matching the final inspection imaging correction index stored in the database based on the final inspection imaging quality assessment value.
[0044] The final inspection imaging quality assessment value is set as the final inspection designated label, the final inspection designated label is compared with each final inspection setting label stored in the database, the final inspection setting label that is the same as the final inspection designated label is determined, and the final inspection imaging correction index corresponding to the final inspection setting label is obtained.
[0045] Obtain a dissolution effect quality evaluation value based on the external dissolution characteristic data set and the final inspection imaging correction index; obtain the dissolution effect quality standard value stored in the database; compare the dissolution effect quality standard value stored in the database with the dissolution effect quality evaluation value to obtain the final inspection judgment result: if the dissolution effect quality evaluation value is greater than or equal to the dissolution effect quality standard value stored in the database, the final inspection judgment result is that the dissolution effect quality is qualified; if the dissolution effect quality evaluation value is less than the dissolution effect quality standard value stored in the database, the final inspection judgment result is that the dissolution effect quality is unqualified.
[0046] The calculation formula for the final inspection imaging quality evaluation value is: ; In the formula, is the final inspection imaging quality assessment value, It is the final inspection exposure deviation time.
[0047] Used to quantitatively describe the image quality of pharmaceutical imaging during final inspection. The larger the value, the higher the image quality of the drug and the better the imaging effect. This may cause the image to be overexposed or too dark, resulting in loss of image details and significantly reducing .
[0048] In the visual-assisted drug quality data processing, the error caused by imaging effects on the actual drug inspection quality assessment is quantified, and different imaging correction data sets are distinguished in different inspections and data collection environments for drug quality data collection. Since the properties of the imaging objects for initial inspection and final inspection are different, the corresponding exposure time standards are different. The influence of the properties of the imaging objects on data acquisition is more comprehensively considered, and the parameters are refined into two categories: initial inspection exposure deviation time and final inspection exposure deviation time. This can more clearly explain the influence of the characteristics of different inspection stages on the drug quality inspection results, making the inspection results more transparent and reliable.
[0049] The dissolution effect quality final inspection data set is obtained, and the specific analysis process is as follows: obtain a dissolution effect feedback graph group, which is a set of dissolution state imaging of the drug at each standard interval time point within the detection time; take the standard interval time point corresponding to the dissolution state imaging that first presents the dissolution state in the dissolution effect feedback graph group as the start dissolution time, and take the ratio of the drug dissolution volume from the standard interval time point corresponding to the start dissolution time to the next standard interval time point to the interval time between the two standard interval time points as the initial dissolution rate; obtain the dissolution state imaging at the end of the detection time in the dissolution effect feedback graph group, and map it in the dissolution effect standard graph stored in the database, obtain the absolute value of the drug dissolution diffusion volume difference between the two images and the standard dissolution diffusion volume in the dissolution effect standard graph as the drug dissolution diffusion deviation, and obtain the absolute value of the volume difference of the remaining undissolved drug between the two images as the undissolved deviation volume.
[0050] The dissolution effect quality final inspection data set is obtained by comprehensively comparing the dissolution effect feedback graph group and the dissolution effect standard graph, which provides an intuitive representation of the dissolution process and can clearly show the dissolution status of the drug at different time points. This visual information is easier to understand and analyze than simple numerical data. The dissolution effect feedback graph group contains imaging of the drug dissolution status at multiple time points, dynamically monitors the drug dissolution process, and more comprehensively reflects the dissolution characteristics of the drug. This allows subsequent analysis to take into account the impact of time series and improve the accuracy and reliability of drug quality data processing.
[0051] like Figure 2 As shown, a drug quality data processing system based on visual assistance is used to implement the above method, including a drug appearance initial inspection module, a microbial limit re-inspection module and a dissolution effect final inspection module, wherein: The drug appearance initial inspection module is used to obtain the appearance quality initial inspection data set of the drug to be inspected, and obtain the initial inspection judgment result of the drug to be inspected based on the appearance quality initial inspection data set. The initial inspection judgment result includes unqualified appearance quality and qualified appearance quality; If the initial inspection result shows that the appearance quality is unqualified, the drug to be inspected will be marked as unqualified; If the initial inspection result shows that the appearance quality is qualified, the drug to be inspected will be marked as a drug to be re-inspected; The microbial limit re-test module is used to determine whether the microbial limit quality of the drug to be re-tested is qualified. If the microbial limit quality is unqualified, the drug to be re-tested is marked as an unqualified drug. If the microbial limit quality is qualified, the drug to be re-tested is marked as a drug to be final tested. The dissolution effect final inspection module is used to perform dissolution effect inspection on the drugs to be inspected. If the dissolution effect inspection fails, the drugs to be inspected are marked as unqualified drugs. If the dissolution effect inspection passes, the drugs to be inspected are marked as qualified drugs.
[0052] A specific calculation example is: The scanning illumination uniformity is 0.85, the reflective area ratio is 0.02, the initial inspection exposure deviation time is 0.1 second, and the initial inspection imaging quality assessment value is calculated to be 1.45. The initial inspection imaging correction index stored in the database based on the initial inspection imaging quality assessment value is 1.05.
[0053] The proportion of drug surface defects is 0.02, the cumulative deviation area of the minimum circumscribed rectangle is 0.03 square centimeters, the surface deviation color ratio is 0.01, the cumulative deviation circumference of the drug contour is 0.015, the initial inspection imaging correction index is 1.05, and the calculated appearance quality assessment value is 0.499. The appearance quality standard value stored in the database is 0.4. The initial inspection judgment result is obtained by comparing the appearance quality standard value and the appearance quality assessment value stored in the database. If the appearance quality assessment value is greater than the appearance quality standard value stored in the database, the initial inspection judgment result is that the appearance quality is qualified and it is marked as a drug for re-inspection.
[0054] The colony culture time is 48 hours, the colony culture temperature is 37 degrees Celsius, the total colony count is 26, the number of colony types is 19, the total number of standard colonies stored in the matching database is 24, the number of standard colony types stored in the database is 18, and the total colony count deviation rate obtained based on the total colony count and the standard total colony count stored in the database is 7.69%; the colony type number deviation rate obtained based on the colony type number and the standard colony type number stored in the database is 5.26%, the total colony count deviation threshold stored in the database is 10%, and the colony type number deviation threshold stored in the database is 8%. The total colony count deviation rate is less than the total colony count deviation threshold, and the process is to determine whether the colony type number deviation rate is less than the colony type number deviation threshold. The judgment result is that the colony type number deviation rate is less than the colony type number deviation threshold. The colony statistics result is that the colonies do not exceed the standard, and the pathogen screening is performed.
[0055] The number of pathogenic bacterial colony types is 1, the total proportion of the colored part is 0.02, the color development reaction time is 0.25 hours, the calculated microbial limit quality assessment value is 0.746, the microbial limit quality standard value stored in the database is 1.5, the reference number of pathogenic bacterial colony types is 5, the microbial limit quality assessment value is less than the microbial limit quality standard value stored in the database, and the judgment is made whether the number of pathogenic bacterial colony types is less than the reference number of pathogenic bacterial colony types. The judgment result is that the number of pathogenic bacterial colony types is less than the reference number of pathogenic bacterial colony types. The re-inspection judgment result is that the microbial limit quality is qualified and it is marked as a drug awaiting final inspection.
[0056] The scanning illumination uniformity is 0.85, the reflective area ratio is 0.02, the final inspection exposure deviation time is 0.2 seconds, and the final inspection imaging quality assessment value is calculated to be 1.31. The final inspection imaging correction index stored in the database based on the final inspection imaging quality assessment value is 1.1.
[0057] The drug dissolution diffusion deviation is 0.02, the starting dissolution time is 2 minutes, the undissolved deviation volume is 0.01 cubic centimeters, the initial dissolution rate is 0.05 cubic centimeters per second, the final inspection imaging correction index is 1.1, and the calculated dissolution effect quality assessment value is 4.829. The dissolution effect quality standard value stored in the database is 3. The dissolution effect quality assessment value is greater than the dissolution effect quality standard value stored in the database. The final inspection judgment result is that the dissolution effect quality is qualified, and the drug to be final inspected is marked as a qualified drug.
Claims
1. A drug quality data processing method based on visual assistance, characterized in that the steps include: Obtaining a data set of preliminary inspection of the appearance quality of the drugs to be inspected, and obtaining preliminary inspection judgment results of the drugs to be inspected based on the data set of preliminary inspection of the appearance quality, wherein the preliminary inspection judgment results include unqualified appearance quality and qualified appearance quality; If the initial inspection result shows that the appearance quality is unqualified, the drug to be inspected will be marked as unqualified; If the initial inspection result shows that the appearance quality is qualified, the drug to be inspected will be marked as a drug to be re-inspected; Determine whether the drug to be retested is qualified in terms of microbial limit quality. If the microbial limit quality is unqualified, mark the drug to be retested as unqualified drug; if the microbial limit quality is qualified, mark the drug to be retested as drug to be final tested; Conduct dissolution effect test on the drug to be final inspected. If the dissolution effect test fails, the drug to be final inspected is marked as an unqualified drug. If the dissolution effect test passes, the drug to be final inspected is marked as a qualified drug.
2. The visually assisted drug quality data processing method according to claim 1, characterized in that: The process of obtaining the initial inspection results of the drugs to be inspected based on the appearance quality initial inspection data set is as follows: Perform high-resolution image scanning on the drugs to be inspected and collect images of the drugs to be inspected; Based on the imaging of the drugs to be inspected, the initial inspection data set of appearance quality is obtained, including the initial inspection imaging correction set and the appearance quality evaluation set. The initial inspection imaging correction set includes scanning illumination uniformity, reflective area ratio, and initial inspection exposure deviation time. The appearance quality evaluation set includes the drug surface defect ratio, the minimum circumscribed rectangle cumulative deviation area, the surface deviation color ratio, and the cumulative deviation perimeter of the drug contour; Obtaining a preliminary inspection imaging quality assessment value based on a preliminary inspection imaging correction set; Matching the initial inspection imaging correction index stored in the database based on the initial inspection imaging quality assessment value; Obtaining an appearance quality assessment value based on an appearance quality assessment set and a preliminary inspection imaging correction index; Obtain the appearance quality standard value stored in the database; Compare the appearance quality standard value and appearance quality evaluation value stored in the database to obtain the initial inspection result: If the appearance quality evaluation value is greater than or equal to the appearance quality standard value stored in the database, the initial inspection result is that the appearance quality is qualified; If the appearance quality evaluation value is less than the appearance quality standard value stored in the database, the initial inspection result is that the appearance quality is unqualified.
3. The visually assisted drug quality data processing method according to claim 2, characterized in that: The calculation formula for the appearance quality evaluation value is: ； Where y is the appearance quality assessment value, q is the proportion of drug surface defects, j is the cumulative deviation area of the minimum circumscribed rectangle, b is the proportion of surface deviation color, l is the cumulative deviation perimeter of the drug contour, is the initial imaging correction index, e is a natural constant, and lg represents the logarithm with base 10.
4. The visually assisted drug quality data processing method according to claim 1, characterized in that: The process for determining whether the drug to be retested meets the microbial limit quality is as follows: For the drugs to be retested, sample inoculation and colony counting are performed to obtain a colony statistics data set, including colony culture time, colony culture temperature, total colony count, and number of colony types; Comprehensively analyze the colony statistics data set to obtain the colony statistics results, and determine whether to perform pathogen screening based on the colony statistics results. The colony statistics results include colony counts exceeding the standard and colony counts not exceeding the standard; If the colony count result shows that the colony exceeds the standard, the microbial limit quality is unqualified and the drug to be retested is marked as unqualified drug; If the colony count result shows that the colony does not exceed the standard, the drug to be retested will be screened for pathogenic bacteria to obtain the retest judgment result: If the re-test result is that the drug contains pathogenic bacteria, the microbial limit quality is unqualified and the drug to be re-tested is marked as unqualified; If the re-inspection result shows that it does not contain pathogenic bacteria, the microbial limit quality is qualified and the drug to be re-inspected is marked as a drug to be final inspected.
5. The visually assisted drug quality data processing method according to claim 4, characterized in that: The process of comprehensively analyzing the colony statistics data set to obtain the colony statistics results is as follows: Get colony statistics dataset; Based on the colony culture time and colony culture temperature, the total number of standard colonies and the number of standard colony types stored in the matching database; Obtaining a deviation rate of the total number of colonies based on the total number of colonies and the standard total number of colonies stored in a database; Obtaining a colony type quantity deviation rate based on the colony type quantity and the standard colony type quantity stored in a database; Obtain the total colony count deviation threshold and colony type count deviation threshold stored in the database; Determine whether the total colony count deviation rate is less than the total colony count deviation threshold: If not, the colony count result is that the colony exceeds the limit; If yes, then determine whether the colony type quantity deviation rate is less than the colony type quantity deviation threshold: If yes, the colony count result is that the colony does not exceed the standard; If not, the colony count result is that the colony count exceeds the standard.
6. The visually assisted drug quality data processing method according to claim 1, characterized in that: The process of testing and judging the dissolution effect of the final inspection drug is as follows: Conduct dissolution effect testing on the drugs to be final tested and obtain a set of dissolution effect feedback graphs; The dissolution effect feedback graph group and the dissolution effect standard graph stored in the database are comprehensively compared to obtain the dissolution effect quality final inspection data set, including the final inspection imaging correction set and the dissolution characteristic data set. The final inspection imaging correction set includes scanning light uniformity, reflective area ratio, and final inspection exposure deviation time. The dissolution characteristic data set includes drug dissolution diffusion deviation, dissolution start time, undissolved deviation volume, and initial dissolution rate. Obtaining a dissolution effect quality evaluation value based on the dissolution effect quality final inspection data set, and obtaining a final inspection judgment result based on the dissolution effect quality evaluation value, wherein the final inspection judgment result includes whether the dissolution effect quality is unqualified or qualified; If the final inspection result is that the dissolution effect quality is unqualified, the drug that is qualified in microbial limit quality and awaiting final inspection is marked as a drug that is unqualified in dissolution effect quality; If the final inspection result is that the dissolution effect quality is qualified, then the drug awaiting final inspection with the microbial limit quality qualified will be marked as a qualified drug.
7. The visually assisted drug quality data processing method according to claim 6, characterized in that: The calculation formula of dissolution effect quality assessment value is: ； In the formula, r is the dissolution effect quality assessment value, k is the drug dissolution diffusion deviation, c is the start of dissolution time, n is the undissolved deviation volume, v is the initial dissolution rate, is the final inspection imaging correction index, e is a natural constant, lg represents the logarithm with base 10, and ln represents the logarithm with base e.
8. The visually assisted drug quality data processing method according to claim 6, characterized in that: The process of obtaining the final inspection judgment result based on the dissolution effect quality assessment value is as follows: Obtain the final quality inspection data set for dissolution effect; Calculate the final inspection imaging quality assessment value based on the final inspection imaging correction set; Matching the final inspection imaging correction index stored in the database based on the final inspection imaging quality assessment value; Obtain dissolution effect quality assessment value based on in vitro dissolution characteristic data set and final inspection imaging correction index; Obtaining the dissolution effect quality standard value stored in the database; If the dissolution effect quality evaluation value is greater than or equal to the dissolution effect quality standard value stored in the database, the final inspection judgment result is that the dissolution effect quality is qualified; If the dissolution effect quality assessment value is less than the dissolution effect quality standard value stored in the database, the final inspection judgment result is that the dissolution effect quality is unqualified.
9. The visually assisted drug quality data processing method according to claim 7, characterized in that: The process of comprehensively comparing the dissolution effect feedback graph group and the dissolution effect standard graph stored in the database to obtain the dissolution effect quality final inspection data set is as follows: Obtaining a dissolution effect feedback graph group, which is a collection of dissolution state images of the drug at each standard interval time point within the detection time; Obtain the dissolution state imaging that first presents the dissolution state in the dissolution effect feedback graph group, mark the standard interval time point corresponding to the dissolution state imaging that first presents the dissolution state as the start dissolution time, and take the ratio of the drug dissolution volume from the standard interval time point corresponding to the start dissolution time to the next standard interval time point to the interval time between the two standard interval time points as the initial dissolution rate; The dissolution state imaging at the end of the detection time in the dissolution effect feedback graph group is obtained, and mapped in the dissolution effect standard graph stored in the database, the absolute value of the drug dissolution diffusion volume difference between the two images and the ratio of the standard dissolution diffusion volume in the dissolution effect standard graph are obtained as the drug dissolution diffusion deviation, and the absolute value of the volume difference of the remaining undissolved drug between the two images are obtained as the undissolved deviation volume.
10. A drug quality data processing system based on visual assistance, used to implement the method according to any one of claims 1 to 9, characterized in that: It includes the drug appearance initial inspection module, microbial limit re-inspection module and dissolution effect final inspection module, among which: The drug appearance initial inspection module is used to obtain the appearance quality initial inspection data set of the drug to be inspected, and obtain the initial inspection judgment result of the drug to be inspected based on the appearance quality initial inspection data set. The initial inspection judgment result includes unqualified appearance quality and qualified appearance quality; If the initial inspection result shows that the appearance quality is unqualified, the drug to be inspected will be marked as unqualified; If the initial inspection result shows that the appearance quality is qualified, the drug to be inspected will be marked as a drug to be re-inspected; The microbial limit re-test module is used to determine whether the microbial limit quality of the drug to be re-tested is qualified. If the microbial limit quality is unqualified, the drug to be re-tested is marked as an unqualified drug. If the microbial limit quality is qualified, the drug to be re-tested is marked as a drug to be final tested. The dissolution effect final inspection module is used to perform dissolution effect inspection on the drugs to be inspected. If the dissolution effect inspection fails, the drugs to be inspected are marked as unqualified drugs. If the dissolution effect inspection passes, the drugs to be inspected are marked as qualified drugs.