Clinical traditional Chinese medicine closed-loop prevention and control system

Abstract

The invention relates to a clinical traditional Chinese medicine closed-loop prevention and control system, and belongs to the technical field of medical information processing, the distribution state of traditional Chinese medicine can be checked in time through the closed-loop prevention and control system, traditional Chinese medicine which is not distributed in place can be tracked, and a terminal database is controlled to perform automatic prescription checking, so that the accuracy and efficiency of prescription checking are effectively improved; by setting the visual recognition area, comparing a non-contact image of the AI recognition system with a traditional Chinese medicine pattern database and precisely checking a pharmacist dispensing result, through the function, the pharmacist dispensing error rate is reduced, intelligent auditing and whole-process closed-loop management are achieved, the problems existing in traditional Chinese medicine management are effectively solved, and the traditional Chinese medicine management efficiency is improved. The quality and efficiency of traditional Chinese medicine management are remarkably improved, and a powerful guarantee is provided for medication safety of patients.

Description

Technical Field

[0001] The present invention belongs to the technical field of medical information processing, and specifically relates to a clinical closed-loop prevention and control system for traditional Chinese medicine. Background Art

[0002] At present, there are significant problems in multiple links of clinical Chinese medicine management, which seriously affect the safety and effectiveness of Chinese medicine use. The specific manifestations are as follows: 1. Prescription review level: The traditional manual prescription review method, faced with numerous prescriptions for a dose of Chinese medicine, is difficult to accurately identify dosage errors and incompatibilities, posing a major medication safety risk. 2. Logistics tracking level: Nurses are unable to check whether Chinese medicine has been delivered to the clinic, resulting in a lack of effective monitoring and management of the flow status of Chinese medicine. 3. Pharmacy dispensing level: Pharmacists often make mistakes when grabbing Chinese medicine, and it is difficult to detect problems by relying solely on manual verification, which affects the accuracy of Chinese medicine preparation. In view of this, this solution was created. Summary of the Invention

[0003] In view of the deficiencies in the existing technology, the technical problem to be solved by the present invention is to provide a clinical closed-loop prevention and control system for traditional Chinese medicine, which improves the quality and efficiency of traditional Chinese medicine management and provides a strong guarantee for the safety of patients' medication.

[0004] To solve the above technical problems, the technical solution adopted by the present invention is: a clinical Chinese medicine closed-loop prevention and control system, comprising the following steps: S1. The doctor prescribes Chinese medicine in the hospital information system, and the prescription information is synchronized to the closed-loop prevention and control cloud in real time; S2. Manual verification once, and submit to the pharmacy if passed; S3: A second manual verification is performed in the pharmacy, and an automatic verification and prescription review is performed through the control terminal database. The verification results of the two are compared, and after the comparison is correct, the prescription is sorted and the medicine is dispensed; S4. The pharmacist dispenses the medicine and places it in the visual recognition area. The visual recognition area is photographed and the AI ​​recognition system compares the captured image with the images in the Chinese medicine image database to automatically identify whether the name, specification, and quantity of the Chinese medicine are consistent with the prescription. S5. Perform a third manual check based on the automatic recognition results and information; S6. After checking and confirming that everything is correct, the delivery person will deliver the Chinese medicine to the clinical department. During the delivery process, the delivery time and delivery person information will be recorded by scanning the code; S7. When clinical nurses receive Chinese medicine, they scan the code to verify the information of the Chinese medicine, accept the medicine after confirming that it is correct, and record the receipt time and receiving person information; S8. Dispense Chinese medicine to patients according to doctor's orders and inform them of the usage and precautions.

[0005] Furthermore, the visual recognition area in step S4 includes a large volume recognition area and a small volume recognition area. The specific recognition steps of step S4 are as follows: a plurality of Chinese medicines are divided into a group according to their body shape, and each time a type of Chinese medicine is placed in the large volume recognition area and the small volume recognition area. The camera shoots the Chinese medicine in the large volume recognition area and the small volume recognition area, and the color of the Chinese medicine, the actual length and width of the Chinese medicine, the size and shape of the crown, the number of roots and the length of the roots are collected through the image acquisition system. The AI ​​recognition system compares the Chinese medicine with the Chinese medicine in the Chinese medicine image database to identify the name of the Chinese medicine in the large volume recognition area and the small volume recognition area.

[0006] Furthermore, a ring-shaped LED light source is used for illumination during the camera shooting process, and the sheet-shaped Chinese medicine is laid flat without overlapping; for the whole Chinese medicine, after taking the photo, the crown and root areas are cut out for denoising, and then compared with the Chinese medicine in the Chinese medicine image database.

[0007] Furthermore, when the step S4 determines that the Chinese medicine is correct, a 10-digit code is generated for the prescription and affixed to the medicine box, the first digit of which is 1 or 2, where 1 represents male and 2 represents female; the second digit of which is 1 or 2, where 1 represents adult and 2 represents minor, and the last eight digits of the code are the Chinese medicine number.

[0008] Furthermore, when operating or transferring the Chinese medicine in steps S4 to S8, the code on the medicine box needs to be scanned, and the Chinese medicine information, scanning time and scanning location can be obtained after scanning.

[0009] Furthermore, after manually verifying the automatic recognition result in step S5, if it is confirmed to be correct, the result needs to be confirmed and submitted to the AI ​​recognition system for model training; If the manual confirmation result is wrong, the error area needs to be pointed out manually, corrected, and submitted to the AI ​​recognition system for model training.

[0010] Furthermore, the closed-loop prevention and control system also includes step S9; S9, observing the patient's reaction after oral administration of traditional Chinese medicine.

[0011] Furthermore, in step S9, after obtaining the patient's medication status and reaction information after medication, it needs to be recorded in the closed-loop prevention and control cloud.

[0012] Furthermore, in step S5, when the AI ​​recognition system makes an error in recognition, it generates a comparison library of easily confused medicinal materials through manual feedback of error information and correction information, regularly counts the medicinal material combinations that the AI ​​recognition system recognizes incorrectly, and automatically generates a feature difference comparison table.

[0013] Furthermore, in step S4, a separate option for prescription composition integrity verification is added. Selecting this option will only verify the integrity of the composition in the traditional Chinese medicine prescription.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. Effectiveness of Chinese medicine distribution management: Through the closed-loop prevention and control system, the distribution status of Chinese medicine can be checked in time, and the Chinese medicine that has not been delivered can be tracked. The existing missed delivery situation can be avoided. The closed-loop prevention and control method of this solution can track and retrieve the missed Chinese medicine in time. At the same time, the existing Chinese medicine distribution has delayed delivery. The monitoring and optimization of the distribution process through the closed-loop prevention and control system will help shorten the delivery time and improve the delivery efficiency.

[0015] 2. Prescription Review Effectiveness: Automated prescription review using the control terminal database has effectively improved the accuracy and efficiency of prescription review. Prescription errors are avoided, and dosage errors and incompatibilities are promptly identified and corrected, reducing medication safety risks at the source.

[0016] 3. Effectiveness of pharmacy dispensing: By setting up a visual recognition area and comparing the non-contact images of the AI ​​recognition system with the traditional Chinese medicine pattern database, the pharmacist's dispensing results are accurately verified. This function reduces the error rate of pharmacists in dispensing medicines, and promptly discovers and corrects problems such as wrong or missed traditional Chinese medicines due to human negligence, greatly improving the accuracy of traditional Chinese medicine preparation and providing more reliable technical guarantees for patients' medication safety.

[0017] 4. The clinical closed-loop prevention and control system for traditional Chinese medicine effectively solves the problems existing in traditional Chinese medicine management through intelligent review and full-process closed-loop management, significantly improves the quality and efficiency of Chinese medicine management, and provides strong guarantees for patients' medication safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a partial cross-sectional view of the crown and root of ginseng in Comparative Example 1 of the present invention; Figure 2 This is a partial cross-sectional schematic diagram of the crown and root of Phytolacca root in Comparative Example 1 of the present invention; Figure 3 The present invention is a flow chart of a clinical closed-loop prevention and control system of traditional Chinese medicine. DETAILED DESCRIPTION

[0019] In order to make the above features and advantages of the present invention more obvious and easy to understand, embodiments are given below with reference to the accompanying drawings for detailed description.

[0020] like Figure 1-Figure 3 As shown, this embodiment provides a clinical Chinese medicine closed-loop prevention and control system, including the following steps: S1. The doctor prescribes Chinese medicine in the hospital information system, and the prescription information is synchronized to the closed-loop prevention and control cloud in real time.

[0021] S2. Manually check once and submit to the pharmacy if passed.

[0022] S3. A second manual verification is carried out in the pharmacy, and automatic verification and prescription review are carried out through the control terminal database. The two verification results are compared, and after the comparison is correct, the number is arranged and the medicine is dispensed.

[0023] S4. The pharmacist dispenses the medicine and places it in the visual recognition area. Photos are taken in the visual recognition area and compared with the pictures in the Chinese medicine image database through the AI ​​recognition system to automatically identify whether the name, specifications and quantity of the Chinese medicine are consistent with the prescription. The camera in the AI ​​recognition system uses an industrial-grade CCD or CMOS camera.

[0024] S5. Perform a third manual verification based on the automatic recognition results and information.

[0025] S6. After checking that everything is correct, the delivery person will deliver the Chinese medicine to the clinical department. During the delivery process, the delivery time and delivery person information will be recorded by scanning the code.

[0026] S7. When clinical nurses receive Chinese medicine, they scan the code to verify the information of the medicine. After confirming that it is correct, they accept the medicine and record the receiving time and receiving person information.

[0027] S8. Dispense Chinese medicine to patients according to doctor's orders and inform them of the usage and precautions.

[0028] S9. Observe the patient's reaction after taking Chinese medicine orally.

[0029] In this embodiment, when medicinal materials are put into storage, traceability information of the medicinal materials needs to be entered and synchronized to the closed-loop control cloud. Suppliers are required to provide electronic credentials such as the origin, harvesting time, processing technology (such as parameters for wine-roasting and honey-roasting), batch number, etc. of the medicinal materials. By scanning the code, they are synchronized to the closed-loop control cloud and associated with the traditional Chinese medicine image database (such as marking "Manufacturer: XXX", "Origin: Yunnan", "Processing technology: Yellow wine-roasting", "Batch number: xxxxx" and other feature tags), which facilitates the traceability of subsequent traditional Chinese medicine issues.

[0030] The closed-loop prevention and control cloud in step S1 is the information terminal of the hospital's internal system.

[0031] In step S2, dedicated personnel will conduct an initial review of the prescription to see if there are any errors in the combination or dosage (this needs to be verified in conjunction with the patient's medical information). If an error is found, the prescription application will be rejected (with the doubts or errors stated), and the doctor who issued the prescription will be reminded through the closed-loop control cloud to revise or explain it and resubmit it.

[0032] In step S3, the pharmacist checks the prescription again. If the check fails, it will be sent back for resubmission as in step S2. After the pharmacist passes the review, the prescription will be checked again according to the control terminal database (existing computer prescription review). If the check fails, it will be sent back for resubmission as in step S2.

[0033] After the Chinese medicine pharmacist finishes picking the medicine in step S4, the Chinese medicine is classified according to type and volume. The large volume is placed in the large volume recognition area of ​​the visual recognition area (such as ginseng, Chinese rhizome of smilax china, etc.), and the small volume is placed in the small volume recognition area of ​​the visual recognition area (such as asarum, clematis root, etc.). The large volume recognition area and the small volume recognition area are separated by a red line. A pressure sensor, which is a commonly used electronic scale, is set below the large volume recognition area and the small volume recognition area. The pressure sensor is used to sense the weight of the Chinese medicine. When placing, be careful not to press the red line, and each time the Chinese medicine placed in the large volume recognition area and the small volume recognition area is the same. The camera shoots the Chinese medicine in the large volume recognition area and the small volume recognition area, and the image acquisition system collects the color of the Chinese medicine, the actual length and width of the Chinese medicine, the size and shape of the crown, the number of roots and the length of the root hairs. The AI ​​recognition system compares the Chinese medicine with the Chinese medicine in the Chinese medicine image database to identify the name of the Chinese medicine in the large volume recognition area and the small volume recognition area.

[0034] In step S4, some doctors prescribe prescriptions that add or subtract Chinese herbs from a fixed formula. A separate option, "Prescription Composition Integrity Verification," has been added to the automated prescription review process (in the AI ​​recognition system) in step S4. Selecting this option only verifies the integrity of the prescription composition, without identifying the type and name of the medicinal ingredients. This significantly shortens the automated review process when using fixed prescriptions. For example, the system has a built-in database of classic prescriptions (a collection of herbs from fixed prescriptions in a database of traditional Chinese medicine images, such as the standard composition of "Ma Huang Tang" and "Liu Wei Di Huang Wan"). When a prescription is a fixed prescription, the AI ​​recognition system verifies that the medicinal ingredients in the visual recognition area are complete and include all required ingredients, preventing omission of key ingredients. For custom prescriptions that are not fixed formulas, the system can indicate a "risk of missing core medicinal ingredients" (e.g., a prescription that uses "clearing heat and detoxifying" as a treatment principle but does not include commonly used herbs such as honeysuckle and forsythia) to assist with manual verification and judgment in step S5.

[0035] Preferably, a ring-shaped LED light source is used for illumination (to remove shadows) during camera shooting, and sheet-shaped Chinese medicines are laid flat without overlapping; for whole-type Chinese medicines, after taking pictures, the crown and root areas are cut out for denoising, and then compared with the Chinese medicines in the Chinese medicine image database.

[0036] Preferably, when the AI ​​recognition system in step S4 compares the photographed pictures of medicinal materials with the pictures in the medicinal material graphic database, in addition to the above-mentioned recognition features, a processing feature verification is also provided. This option needs to be checked separately (such as the epidermal texture and color changes of honey-roasted and vinegar-processed Chinese medicines) to ensure the accuracy of the medicinal materials used in the preparation of the medicines.

[0037] Comparative Example 1: Comparison of ginseng and pokeweed roots. The skin of ginseng and pokeweed roots are both yellow-brown, and their actual length and width are similar. Figure 1-Figure 2 As shown, the camera takes a picture, and the AI ​​recognition system captures the crown of the target Chinese medicine: the crown of ginseng has a rhizome, and the crown of the Chinese pokeweed root has no rhizome. The camera takes a picture, and the AI ​​recognition system captures the root part of the target Chinese medicine: the fibrous roots of ginseng are slender and dense, while the fibrous roots of Chinese pokeweed are few and sparse. By comparing the screenshots of the above two areas, the AI ​​recognition system can accurately know the name of the target Chinese medicine type and the observable horizontal lines on the trunk part, such as ginseng: there are irregular longitudinal wrinkles and fine horizontal ring lines, and the longitudinal wrinkles of the Chinese pokeweed root are deep but the horizontal lines are shallow and intermittent.

[0038] Comparative Example 2: Comparison of Asarum and Clematis. Target objects, such as Asarum or Clematis, are placed in the small volume recognition zone. These Chinese medicinal herbs are small and often form clumps. Simply pour the entire pile into the small volume recognition zone and the camera will take a picture. Asarum: Often curled into clumps, the rhizome is irregularly cylindrical with short branches, a gray-brown, rough surface, and ring-shaped nodes. Internodes are 0.2-0.3 cm long, and the branches have bowl-shaped stem scars at the tips. The roots are slender, densely distributed at the nodes, with a gray-yellow surface that is smooth or longitudinally wrinkled, and possesses fibrous roots and root scars. Clematis: The rhizome is columnar, with a light brownish-yellow surface. The stem base remains at the upper end, while numerous fine roots are borne on the lower side. The roots are slender, slightly curved, dark brown in color, with fine longitudinal striations. Some have detached bark, revealing yellowish-white xylem. The AI ​​recognition system can distinguish Chinese medicinal herbs based on their width, color, and root shape.

[0039] It is worth noting that some medicinal materials are sliced. For the sliced ​​medicinal materials, they need to be spread flat in the large volume identification area or the small volume identification area, and the type and name of the medicinal material can be judged by observing the cross-sectional shape, cross-sectional color and texture within the cross-sectional area.

[0040] Optimally, for powdered and sliced ​​medicinal materials, a "powder / particle identification zone" and a "slice identification zone" can be added to the visual recognition area based on the existing recognition system. For powdered medicinal materials (such as pollen pollen and talcum powder), "laser particle size analysis + color spectrum comparison" is used to assist AI recognition. For stacked slices (such as white peony root and gastrodia elata), 3D scanning technology is used to reconstruct the three-dimensional shape and extract characteristics such as slice thickness and texture distribution to avoid misjudgment due to overlap.

[0041] In step S5, the pharmacist will conduct another verification based on the recognition report of the AI ​​recognition system. If it is confirmed to be correct, the result must be confirmed and submitted to the AI ​​recognition system for model training. If the manual confirmation result is wrong, the error area must be manually pointed out and corrected, and then submitted to the AI ​​recognition system for model training.

[0042] When the AI ​​recognition system makes an error, a "comparison library of easily confused medicinal materials" is generated through the feedback of error information and correction information: regular statistics are kept on the medicinal material combinations with high AI recognition error rates (such as the comparison of ginseng and rhizome of Phytolacca root mentioned above), and a "feature difference comparison table" is automatically generated (marking key differences such as the presence or absence of rhizome, density of roots, etc.), which is pushed to the pharmacist workstation to assist in manual verification and focus on key points.

[0043] Preferably, in step S4, when it is determined that the Chinese medicine is correct, a 10-digit code is generated for the prescription and affixed to the medicine bag or box, the first digit of which is 1 or 2, where 1 represents male and 2 represents female; the second digit of which is 1 or 2, where 1 represents adults and 2 represents minors, and the last eight digits of the code are the Chinese medicine number, the first two digits of the last eight digits are the batch information for traceability of the medicinal materials, the middle four digits are the date, and the last two digits are the Chinese medicine code, so as to facilitate rapid positioning when there is a problem with the Chinese medicine.

[0044] Preferably, the background colors of the 10-digit code sticker include red, blue and white. The 10-digit code sticker with a red background represents the medicinal materials contained with toxicity, which is convenient for the delivery personnel to identify and control the dosage of the medicine. The 10-digit code sticker with a blue background represents the medicinal materials that need to be decocted first / then added later. The white background represents the medicinal materials that have no special requirements for normal decoction.

[0045] When operating or transferring the Chinese medicine in steps S4 to S8, the code on the medicine box needs to be scanned. After scanning, the Chinese medicine information, scanning time and scanning location can be obtained. The scanner can be a common QR code scanner.

[0046] Preferably, an abnormal alarm mechanism is added to the scanning process of S4-S8: if the scan fails, the location / time is abnormal, or it does not match the expected process (such as no clinical scan for a long time after the pharmacy scan), the closed-loop control cloud should have a real-time alarm to notify the relevant responsible persons.

[0047] In step S6, when the medicine delivery person picks up the medicine, the first QR code scan is performed (scanner in the pharmacy) to record the time, place and delivery person of picking up the medicine (the person has a QR code surface on his body, and scans the QR code on the medicine bag or medicine box and then scans himself in the following steps); in step S7, after arriving at the medicine delivery location, the medicine delivery person performs a second QR code scan (scanner in the clinical concentration area) to record the arrival time, place and delivery person. After the clinical nurse receives the traditional Chinese medicine, the third QR code scan is performed to record the arrival time, place and receiving person.

[0048] In step S8, the medicine delivery person, holding a mobile scanner, goes to the patient's area. If the patient is an inpatient, they scan the QR code on the medicine bag or box, then the QR code on the bed. Once the pairing is correct, the medicine is dispensed and instructions for use and precautions are provided. If the pairing is incorrect, the medicine delivery person will be notified. If the patient is not an inpatient, they scan the QR code on the medicine bag or box, then the QR code on the user's payment slip. Once the pairing is correct, the medicine is dispensed and instructions for use and precautions are provided. If the pairing is incorrect, the medicine delivery person will be notified.

[0049] After the code is scanned and paired correctly for the distribution of Chinese medicine in step S8, multimedia guidance is triggered: after the patient or nurse scans the code correctly, the system will automatically push a "medication guidance video" (such as the steps for decocting Chinese medicine, methods for processing special medicinal materials) and a "taboo reminder animation" (such as avoiding eating radish when taking prescriptions containing ginseng) in addition to displaying text instructions.

[0050] In step S9, after obtaining the patient's (for hospitalized patients) medication status and reaction information after medication, it needs to be recorded in the closed-loop prevention and control cloud. The closed-loop prevention and control cloud can automatically associate the traceability information of the medicinal materials used in the prescription (such as common problems of a batch of medicinal materials), and assist in quickly locating whether the adverse reaction is caused by the quality of the source of the medicinal materials.

[0051] Preferably, in step S9, a rapid feedback channel for medication reactions is established: inpatients are equipped with a "medication reaction entry terminal", and outpatients can upload adverse reaction symptoms through the WeChat applet. The system automatically classifies the feedback (such as mild discomfort, severe adverse reaction) and pushes it to the attending physician in real time, forming a rapid response closed loop of "medication-feedback-intervention".

[0052] This clinical closed-loop control system for traditional Chinese medicine effectively solves the problems existing in traditional Chinese medicine management through intelligent auditing and full-process closed-loop management, significantly improves the quality and efficiency of Chinese medicine management, and provides strong protection for patients' medication safety.

[0053] In addition to improving quality and safety, this implementation also reduces manual time spent on prescription review and medication dispensing (by 30%), reduces losses due to mismatched and missing TCM medicines (by 3%), and shortens average delivery time by 30 minutes. This reduces the risk of potential medical disputes or readmissions due to medication errors.

[0054] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which shall fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A clinical Chinese medicine closed-loop prevention and control system, characterized by: The following steps are involved: S1. The doctor prescribes Chinese medicine in the hospital information system, and the prescription information is synchronized to the closed-loop prevention and control cloud in real time; S2. Manual verification once, and submit to the pharmacy if passed; S3: A second manual verification is performed in the pharmacy, and an automatic verification and prescription review is performed through the control terminal database. The verification results of the two are compared, and after the comparison is correct, the prescription is sorted and the medicine is dispensed; S4. The pharmacist dispenses the medicine and places it in the visual recognition area. The visual recognition area is photographed and the AI ​​recognition system compares the captured image with the images in the Chinese medicine image database to automatically identify whether the name, specification, and quantity of the Chinese medicine are consistent with the prescription. S5. Perform a third manual check based on the automatic recognition results and information; S6. After checking and confirming that everything is correct, the delivery person will deliver the Chinese medicine to the clinical department. During the delivery process, the delivery time and delivery person information will be recorded by scanning the code; S7. When clinical nurses receive Chinese medicine, they scan the code to verify the information of the Chinese medicine, accept the medicine after confirming that it is correct, and record the receipt time and receiving person information; S8. Dispense Chinese medicine to patients according to doctor's orders and inform them of the usage and precautions.

2. A clinical Chinese medicine closed-loop prevention and control system according to claim 1, characterized in that: The visual recognition area in step S4 includes a large volume recognition area and a small volume recognition area. The specific recognition steps of step S4 are as follows: a plurality of Chinese medicines are divided into a group according to their body shape, and each time a type of Chinese medicine is placed in the large volume recognition area and the small volume recognition area, the camera shoots the Chinese medicine in the large volume recognition area and the small volume recognition area, and the color of the Chinese medicine, the actual length and width of the Chinese medicine, the size and shape of the crown, the number of roots and the length of the roots are collected through the image acquisition system, and the AI ​​recognition system compares the Chinese medicine with the Chinese medicine in the Chinese medicine image database to identify the name of the Chinese medicine in the large volume recognition area and the small volume recognition area.

3. A clinical Chinese medicine closed-loop prevention and control system according to claim 2, characterized in that: During the camera shooting process, a ring-shaped LED light source is used for illumination, and the sheet-shaped Chinese medicine is laid flat without overlapping; for the whole Chinese medicine, after taking the photo, the crown and root areas are cut out for denoising, and then compared with the Chinese medicine in the Chinese medicine image database.

4. A clinical Chinese medicine closed-loop prevention and control system according to claim 1, characterized in that: When the step S4 determines that the Chinese medicine is correct, a 10-digit code is generated for the prescription and affixed to the medicine box. The first digit is 1 or 2, where 1 represents male and 2 represents female; the second digit is 1 or 2, where 1 represents adult and 2 represents minor. The last eight digits are the Chinese medicine number.

5. A clinical Chinese medicine closed-loop prevention and control system according to claim 4, characterized in that: When operating or transferring the Chinese medicine in steps S4 to S8, the code on the medicine box needs to be scanned, and the Chinese medicine information, scanning time and scanning location can be obtained after scanning.

6. A clinical Chinese medicine closed-loop prevention and control system according to claim 1, characterized in that: After manually verifying the automatic recognition results in step S5, if they are correct, the results need to be confirmed and submitted to the AI ​​recognition system for model training; If the manual confirmation result is wrong, the error area needs to be pointed out manually, corrected, and submitted to the AI ​​recognition system for model training.

7. A clinical Chinese medicine closed-loop prevention and control system according to claim 1, characterized in that: The closed-loop prevention and control system also includes step S9: observing the patient's reaction after taking the Chinese medicine orally.

8. A clinical Chinese medicine closed-loop prevention and control system according to claim 7, characterized in that: In step S9, after obtaining the patient's medication status and reaction information after medication, it needs to be recorded in the closed-loop prevention and control cloud.

9. A clinical Chinese medicine closed-loop prevention and control system according to claim 6, characterized in that: In step S5, when the AI ​​recognition system makes an error, it generates a comparison library of easily confused medicinal materials through manual feedback of error information and correction information, regularly counts the medicinal material combinations that the AI ​​recognition system incorrectly recognizes, and automatically generates a feature difference comparison table.

10. A clinical Chinese medicine closed-loop prevention and control system according to claim 2, characterized in that: In step S4, a new option for prescription composition integrity verification is added. Selecting this option will only verify the integrity of the composition of the traditional Chinese medicine prescription.

Images