Special medicine management method and system based on artificial intelligence

Abstract

The invention discloses a special medicine management method and system based on artificial intelligence, and relates to the technical field of medicine management. Acquiring receiving registration information, delivery information and delivery images of the special drugs; using registration information and using registration images of the special medicine are collected, an artificial intelligence detection model is constructed and compared with warehouse-out information and warehouse-out images of the special medicine, if wrong registration exists, registration information of an authorizer corresponding to the special medicine is obtained, and the authorizer is traced; if the special medicine has the unused registration, obtaining the unused registration information of the special medicine, and if the special medicine does not have the unused registration, carrying out artificial intelligence empty bottle detection and recovery on the used special medicine. According to the invention, the whole-course traceable management of the medicine can be realized, the source, the destination and the use condition of the special medicine can be effectively monitored and recorded, and the transparency and the reliability of medicine management are greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of drug management, and in particular to a special drug management method and system based on artificial intelligence. Background Art

[0002] Narcotic and psychotropic drugs are addictive and have the potential for abuse due to their special pharmacological effects. They need to be strictly managed to ensure their legal, safe and rational use and prevent them from flowing into illegal channels. These drugs include drugs and other substances listed in the National Catalogue of Narcotic Drugs and the Catalogue of Psychotropic Drugs, which have important uses in medical practice, such as analgesia and sedation. However, these drugs are addictive and long-term use may lead to physical and psychological dependence, posing a threat to personal health and social stability. Moreover, the abuse of these drugs may lead to serious health problems, including respiratory depression, heart problems, mental disorders, etc., and may even endanger life. In addition, these drugs are in high demand in the illegal market. If they are not properly managed, they are easily used in illegal transactions, causing serious social harm and illegal and criminal acts.

[0003] In hospital departments such as the anesthesiology department and operating room, the management of anesthetic and psychotropic drugs is particularly strict. According to the "Regulations on the Management of Narcotic Drugs and Psychotropic Drugs" and related management regulations, medical institutions must establish a sound management system, clarify the division of responsibilities, and ensure the procurement, storage, allocation, use and safe management of drugs. Specific measures include: special warehouse storage, prescription management, full process management and information management. Through these strict management systems and measures, the safe use of special drugs such as anesthetics and psychotropic drugs in the anesthesia department and operating room can be ensured, while preventing their abuse and illegal circulation.

[0004] Therefore, the present invention provides a special drug management method and system based on artificial intelligence, which not only improves the efficiency and accuracy of special drug management, but also ensures the safety and compliance of special drugs through multiple security measures, providing strong technical support for special drug supervision. Summary of the invention

[0005] The technical problem to be solved by the present invention is the refined full-process detection and management of the use of special drugs.

[0006] In order to solve the above technical problems, the technical solution provided by the present invention is: a special drug management method based on artificial intelligence, comprising the following steps: Step S1: When special drugs are put into storage, the storage information and storage images of the special drugs are collected, and the special drugs are stored in a storage device; Step S2: when receiving the special drug registration notification from the authorized person on the storage device, obtaining the authorized person's registration information, collecting the special drug's outbound information and outbound image, and updating the special drug's inventory status; Step S3: Collect the usage registration information and usage registration images of special drugs, build an artificial intelligence detection model, compare the usage registration information and usage registration images of special drugs with the outbound information and outbound images of special drugs, and determine whether there is an erroneous registration of special drugs. If there is an erroneous registration, obtain the registration information of the corresponding authorized person of the special drugs and trace the authorized person; Step S4: Determine whether there is any unused registration for special drugs. If there is any unused registration for special drugs, obtain the unused registration information of special drugs, and re-register the unused special drugs into the warehouse; Step S5: If there is no unused registration for special drugs, perform artificial intelligence empty bottle detection on the used special drugs, recycle all empty bottles of special drugs, archive all registration information of special drugs, and update the inventory status of special drugs.

[0007] Furthermore, in step S1, collecting the storage information and storage images of special drugs includes: Collecting special medicines The storage information includes obtaining the storage name, storage time, storage batch, quantity and type of special drugs to generate a collection , ;in, Special drugs The name, time, batch, quantity and type of the goods entering the warehouse; Collecting special medicines The image in the library is recorded as , for special medicines Image library of Grayscale processing to generate special medicines Grayscale images of the database , for special medicines Grayscale images of the database Perform edge computing; For special medicines Grayscale images of the database Each pixel has: ; ; ; in, represents the Gaussian filter function; Indicates special medicines Grayscale images of the database The horizontal result after Gaussian filtering of each pixel; Indicates special medicines Grayscale images of the database The vertical result of each pixel after Gaussian filtering; For special medicines Grayscale images of the database The average of the horizontal and vertical results of each pixel after Gaussian filtering is calculated: ; ; in, Indicates special medicines The grayscale image of the first entry The average value of the horizontal results after Gaussian filtering of each pixel; Indicates special medicines Grayscale images of the database The vertical average of each pixel after Gaussian filtering; Get special medicines Grayscale images of the database The initial input vector of : ; in, Indicates special medicines The initial grayscale image The initial input vector of .

[0008] Further, in step S2, the registration information of the authorizer includes the name, work number and department of the authorizer; collecting special medicines Outbound information includes obtaining special medicines Generate a collection of outbound name, outbound time, outbound batch, quantity and type , ;in, Special drugs The name, time, batch, quantity and type of the shipment; Collect special medicines The outbound image is denoted as For special medicines Image from stock Grayscale processing to generate special medicines Grayscale image of the outbound , for special medicines The grayscale image of the warehouse is used for edge computing to obtain special medicines The initial outbound vector of: .

[0009] Furthermore, in step S3, constructing an artificial intelligence detection model includes: Collect special medicines The use registration information includes obtaining special medicines Generate a collection of registration name, registration time, registration batch, quantity and type , ;in, Special drugs The name, time, batch, quantity and type of the shipment; Collecting special medicines The use of the registered image is denoted by For special medicines Use of registered images Grayscale processing to generate special medicines Use of registered grayscale images , for special medicines Use registered grayscale images for edge computing to obtain special medicines Use registration vector ; When special medicines The use registration information set is the same as the outbound information set of special drugs, that is, When it is determined that there is no unused registration for the drug, the special drug Use of registered images and special medicines Compare with the outbound images; Calculation of special medicines Similarity: ; in, Indicates special medicines The similarity between the registered image and the out-of-stock image is used; Set up special medicine The similarity threshold ; When special medicines When the similarity does not exceed the threshold, the special drug is judged There is no erroneous registration; When special medicines When the similarity exceeds the threshold, it is determined that the special drug If there is an erroneous registration, obtain the registration information of the corresponding authorizer of the special drug and trace the authorizer.

[0010] Further, in step S4, when the special drug The use registration information set is different from the outbound information set of special drugs, that is, When determining special drugs There is unused registration; obtain the unused registration information of special drugs, re-register and put the unused special drugs into storage, archive all the registration information of special drugs, and update the inventory status of special drugs.

[0011] The present invention also provides a special drug management system based on artificial intelligence, including an inbound registration module, a requisition registration module, a usage registration module, an artificial intelligence detection model construction and analysis module, an information traceability module, and an intelligent storage update module; the output end of the inbound registration module is connected to the input end of the requisition registration module; the output end of the requisition registration module is connected to the input end of the usage registration module, and the output end of the usage registration module is connected to the input end of the artificial intelligence detection model construction and analysis module; the output end of the artificial intelligence detection model construction and analysis module is connected to the input end of the information traceability module; the output end of the information traceability module is connected to the input end of the intelligent storage update module; The inbound registration module is used to collect the inbound information and inbound images of special drugs; the requisition registration module is used to obtain the registration information of the authorizer and collect the outbound information and outbound images of special drugs; The usage registration module is used to collect the usage registration information and usage registration images of special drugs; The artificial intelligence detection model construction and analysis module is used to construct an artificial intelligence detection model, analyze whether there is incorrect registration of special drugs, and perform empty bottle detection on the used special drugs; The information traceability module is used to trace the registration information of the corresponding authorizer of the special drug when there is incorrect registration; The intelligent storage update module is used to obtain the unused registration information of special drugs, re-register and put the unused special drugs into storage, archive all the registration information of special drugs, and update the inventory status of special drugs in real time.

[0012] The advantages of the present invention compared with the prior art are as follows: 1. Improve and enhance the security of the automated system, providing multiple security guarantees. Through the intelligent system, strict verification and acceptance can be carried out for key steps such as the inbound, requisition, and usage of special drugs. The special drug management method based on artificial intelligence image recognition can intelligently analyze the requisition information and usage information of special drugs, quickly and accurately screen out the problematic special drugs and authorizer information, and ensure the accuracy and compliance of each link.

[0013] 2. It can realize the whole-process traceable management of drugs. From the source, whereabouts to the usage situation of special drugs, effective monitoring and recording can be carried out, greatly improving the transparency and reliability of drug management.

[0014] 3. On this basis, seamless connection and real-time updating of special drug information can be achieved, thereby improving work efficiency and reducing the possibility of human errors. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flow chart of a special drug management method based on artificial intelligence of the present invention. DETAILED DESCRIPTION

[0016] The following is a further detailed description of an artificial intelligence-based special drug management method and system of the present invention in conjunction with the accompanying drawings.

[0017] Combined with Figure 1 , the present invention is introduced in detail.

[0018] A special drug management method based on artificial intelligence, comprising the following steps: Step S1: When special drugs are put into storage, the storage information and storage images of the special drugs are collected, and the special drugs are stored in a storage device; Step S2: when receiving the special drug registration notification from the authorized person on the storage device, obtaining the authorized person's registration information, collecting the special drug's outbound information and outbound image, and updating the special drug's inventory status; Step S3: Collect the usage registration information and usage registration images of special drugs, build an artificial intelligence detection model, compare the usage registration information and usage registration images of special drugs with the outbound information and outbound images of special drugs, and determine whether there is an erroneous registration of special drugs. If there is an erroneous registration, obtain the registration information of the corresponding authorizer of the special drugs, and trace the authorizer; Step S4: Determine whether there is any unused registration for special drugs. If there is any unused registration for special drugs, obtain the unused registration information of special drugs, and re-register the unused special drugs into the warehouse; Step S5: If there is no unused registration for special drugs, perform artificial intelligence empty bottle detection on the used special drugs, recycle all empty bottles of special drugs, archive all registration information of special drugs, and update the inventory status of special drugs.

[0019] Furthermore, in step S1, collecting the storage information and storage images of special drugs includes: Collecting special medicines The storage information includes obtaining the storage name, storage time, storage batch, quantity and type of special drugs to generate a collection , ;in, Special drugs The name, time, batch, quantity and type of the goods entering the warehouse; Collecting special medicines The image in the library is recorded as , for special medicines Image library of Grayscale processing to generate special medicines Grayscale images of the database , for special medicines Grayscale images of the database Perform edge computing; For special medicines Grayscale images of the database Each pixel has: ; ; ; in, represents the Gaussian filter function; Indicates special medicines Grayscale images of the database The horizontal result after Gaussian filtering of each pixel; Indicates special medicines Grayscale images of the database The vertical result of each pixel after Gaussian filtering; For special medicines Grayscale images of the database The average of the horizontal and vertical results of each pixel after Gaussian filtering is calculated: ; ; in, Indicates special medicines The grayscale image of the first entry The average value of the horizontal results after Gaussian filtering of each pixel; Indicates special medicines Grayscale images of the database The vertical average of each pixel after Gaussian filtering; Get special medicines Grayscale images of the database The initial input vector of : ; in, Indicates special medicines The initial grayscale image The initial input vector of .

[0020] In the above technical solution, first of all, the storage information and images of all special drugs are obtained to provide a digital foundation for the automated drug management system; secondly, the contours of all special drugs are extracted by performing image grayscale processing and edge computing to ensure the clarity and accuracy of the image; finally, the pixel coordinates of the grayscale image of the special drugs are calculated to reflect the changes of each pixel point and obtain the feature vector of the grayscale image.

[0021] Further, in step S2, the registration information of the authorizer includes the name, work number and department of the authorizer; collecting special medicines Outbound information includes obtaining special medicines Generate a collection of outbound name, outbound time, outbound batch, quantity and type , ;in, Special drugs The name, time, batch, quantity and type of the shipment; Collecting special medicines The outbound image is denoted as For special medicines Image from stock Grayscale processing to generate special medicines Grayscale image of the outbound , for special medicines The grayscale image of the warehouse is used for edge computing to obtain special medicines The initial outbound vector of: .

[0022] In the above technical solution, the drug collection information is collected and verified using the same method as the special drug warehousing information, which serves as one of the input data of the artificial intelligence detection model.

[0023] Furthermore, in step S3, constructing an artificial intelligence detection model includes: Collecting special medicines The use registration information includes obtaining special medicines Generate a collection of registration name, registration time, registration batch, quantity and type , ;in, Special drugs The name, time, batch, quantity and type of the shipment; Collecting special medicines The use of the registered image is denoted by For special medicines Use of registered images Grayscale processing to generate special medicines Use of registered grayscale images , for special medicines Use registered grayscale images for edge computing to obtain special medicines Use registration vector ; When special medicines The use registration information set is the same as the outbound information set of special drugs, that is, When it is determined that there is no unused registration for the drug, the special drug Use of registered images and special medicines Compare with the outbound images; Calculation of special medicines Similarity: ; in, Indicates special medicines The similarity between the registered image and the out-of-stock image is used; Set up special medicine The similarity threshold ; When special medicines When the similarity does not exceed the threshold, the special drug is judged There is no erroneous registration; When special medicines When the similarity exceeds the threshold, it is determined that the special drug If there is an erroneous registration, obtain the registration information of the corresponding authorizer of the special drug and trace the authorizer.

[0024] In the above technical solution, the image recognition algorithm and the cosine similarity algorithm are combined to compare the outbound images and usage images of special drugs. The closer the cosine value is to 1, the more similar the two images are. This can be used to determine whether there is an erroneous registration of the drug because it does not meet the usage standards.

[0025] Further, in step S4, when the special drug The use registration information set is different from the outbound information set of special drugs, that is, When determining special drugs There are unused registrations; obtain the unused registration information of special drugs, re-register the unused special drugs into the warehouse, archive all registration information of special drugs, and update the inventory status of special drugs.

[0026] In the above technical solution, in step S5, the image processing method in the artificial intelligence detection model can be used to perform grayscale processing on special medicines, and the pixels in the grayscale image can be set to 0 or 255 to make the image present an obvious black and white effect. By setting a suitable threshold, the empty bottles of special medicines can be distinguished.

[0027] An artificial intelligence-based special drug management system adopting the above-mentioned artificial intelligence-based special drug management method includes a storage registration module, a collection registration module, a use registration module, an artificial intelligence detection model construction and analysis module, an information tracing module and an intelligent storage update module; the output end of the storage registration module is connected to the input end of the collection registration module; the output end of the collection registration module is connected to the input end of the use registration module, and the output end of the use registration module is connected to the input end of the artificial intelligence detection model construction and analysis module; the output end of the artificial intelligence detection model construction and analysis module is connected to the input end of the information tracing module; the output end of the information tracing module is connected to the input end of the intelligent storage update module; The entry registration module is used to collect entry information and images of special drugs; the withdrawal registration module is used to obtain the registration information of the authorized person and collect the exit information and images of special drugs; The usage registration module is used to collect usage registration information and usage registration images of special drugs; The artificial intelligence detection model building and analysis module is used to build an artificial intelligence detection model to analyze whether there is an erroneous registration of special drugs and to perform empty bottle detection on special drugs after use; The information tracing module is used to trace the registration information of the corresponding authorized person of the special drug when there is an erroneous registration; The intelligent storage update module is used to obtain unused registration information of special drugs, re-register unused special drugs into the warehouse, archive all registration information of special drugs, and update the inventory status of special drugs in real time.

[0028] The specific implementation process of the artificial intelligence-based special drug management method and system of the present invention is as follows:

[0029] Collecting special medicines The storage information includes obtaining the storage name, storage time, storage batch, quantity and type of special drugs to generate a collection , ;in, Special drugs The name, time, batch, quantity and type of the goods entering the warehouse; Collecting special medicines The image in the library is recorded as , for special medicines Image library of Grayscale processing to generate special medicines Grayscale images of the database , for special medicines Grayscale images of the database Perform edge computing; For special medicines Grayscale images of the database Each pixel has: ; ; ; in, represents the Gaussian filter function; Indicates special medicines Grayscale images of the database The horizontal result after Gaussian filtering of each pixel; Indicates special medicines Grayscale images of the database The vertical result of each pixel after Gaussian filtering; For special medicines Grayscale images of the database The average of the horizontal and vertical results of each pixel after Gaussian filtering is calculated: ; ; in, Indicates special medicines The grayscale image of the first entry The average value of the horizontal results after Gaussian filtering of each pixel; Indicates special medicines Grayscale images of the database The vertical average of each pixel after Gaussian filtering; Get special medicines Grayscale images of the database The initial input vector of : ; in, Indicates special medicines The initial grayscale image The initial input vector of .

[0030] The registration information of the authorized person includes the name, work number and department of the authorized person; Collection of special medicines Outbound information includes obtaining special medicines Generate a collection of outbound name, outbound time, outbound batch, quantity and type , ;in, Special drugs The name, time, batch, quantity and type of the shipment; Collecting special medicines The outbound image is denoted as For special medicines Image from stock Grayscale processing to generate special medicines Grayscale image of the outbound , for special medicines The grayscale image of the warehouse is used for edge computing to obtain special medicines The initial outbound vector of: .

[0031] Collecting special medicines The use registration information includes obtaining special medicines Generate a collection of registration name, registration time, registration batch, quantity and type , ;in, Special drugs The name, time, batch, quantity and type of the shipment; Collecting special medicines The use of the registered image is denoted by For special medicines Use of registered images Grayscale processing to generate special medicines Use of registered grayscale images , for special medicines Use registered grayscale images for edge computing to obtain special medicines Use registration vector ; When special medicines The use registration information set is the same as the outbound information set of special drugs, that is, When it is determined that there is no unused registration for the drug, the special drug Use of registered images and special medicines Compare with the outbound images; Calculation of special medicines Similarity: ; in, Indicates special medicines The similarity between the registered image and the out-of-stock image is used; Set up special medicine The similarity threshold ; Because special medicines The similarity exceeds the threshold, and the special drug is determined If there is an erroneous registration, obtain the registration information of the corresponding authorizer of the special drug and trace the authorizer.

[0032] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. A special drug management method based on artificial intelligence, characterized by: The following steps are involved: Step S1: When special drugs are put into storage, the storage information and storage images of the special drugs are collected, and the special drugs are stored in a storage device; Step S2: when receiving the special drug registration notification from the authorized person on the storage device, obtaining the authorized person's registration information, collecting the special drug's outbound information and outbound image, and updating the special drug's inventory status; Step S3: Collect the usage registration information and usage registration images of special drugs, build an artificial intelligence detection model, compare the usage registration information and usage registration images of special drugs with the outbound information and outbound images of special drugs, and determine whether there is an erroneous registration of special drugs. If there is an erroneous registration, obtain the registration information of the corresponding authorized person of the special drugs and trace the authorized person; Step S4: Determine whether there is any unused registration for special drugs. If there is any unused registration for special drugs, obtain the unused registration information of special drugs, and re-register the unused special drugs into the warehouse; Step S5: If there is no unused registration for special drugs, perform artificial intelligence empty bottle detection on the used special drugs, recycle all empty bottles of special drugs, archive all registration information of special drugs, and update the inventory status of special drugs.

2. The special drug management method based on artificial intelligence according to claim 1, characterized in that: In step S1, collecting the storage information and storage images of special drugs includes: Collect special medicines The storage information includes obtaining the storage name, storage time, storage batch, quantity and type of special drugs to generate a collection , ;in, Special drugs The name, time, batch, quantity and type of the goods entering the warehouse; Collect special medicines The image in the library is recorded as , for special medicines Image library of Grayscale processing to generate special medicines Grayscale images of the database , for special medicines Grayscale images of the database Perform edge computing; For special medicines Grayscale images of the database Each pixel has: ； ； ； in, represents the Gaussian filter function; Indicates special medicines Grayscale images of the database The horizontal result after Gaussian filtering of each pixel; Indicates special medicines Grayscale images of the database The vertical result of each pixel after Gaussian filtering; For special medicines Grayscale images of the database The average of the horizontal and vertical results of each pixel after Gaussian filtering is calculated: ； ； in, Indicates special medicines The grayscale image of the first entry The average value of the horizontal results after Gaussian filtering of each pixel; Indicates special medicines Grayscale images of the database The vertical average of each pixel after Gaussian filtering; Get special medicines Grayscale images of the database The initial input vector of : ； in, Indicates special medicines The initial grayscale image The initial input vector of .

3. The special drug management method based on artificial intelligence according to claim 1 is characterized in that: In step S2, the registration information of the authorizer includes the name, work number and department of the authorizer; collect special medicines Outbound information includes obtaining special medicines Generate a collection of outbound name, outbound time, outbound batch, quantity and type , ;in, Special drugs The name, time, batch, quantity and type of the shipment; Collect special medicines The outbound image is denoted as For special medicines Image from stock Grayscale processing to generate special medicines Grayscale image of the outbound , for special medicines The grayscale image of the warehouse is used for edge computing to obtain special medicines The initial outbound vector of: .

4. The special drug management method based on artificial intelligence according to claim 1, characterized in that: In step S3, the building of an artificial intelligence detection model includes: Collect special medicines The use registration information includes obtaining special medicines Generate a collection of registration name, registration time, registration batch, quantity and type , ;in, Special drugs The name, time, batch, quantity and type of the shipment; Collect special medicines The use of the registered image is denoted by For special medicines Use of registered images Grayscale processing to generate special medicines Use of registered grayscale images , for special medicines Use registered grayscale images for edge computing to obtain special medicines Use registration vector ; When special medicines The use registration information set is the same as the outbound information set of special drugs, that is, When it is determined that there is no unused registration for the drug, the special drug Use of registered images and special medicines Compare with the outbound images; Calculation of special medicines Similarity: ； in, Indicates special medicines The similarity between the registered image and the out-of-stock image is used; Set up special medicine The similarity threshold ; When special medicines When the similarity does not exceed the threshold, the special drug is judged There is no erroneous registration; When special medicines When the similarity exceeds the threshold, it is determined that the special drug If there is an erroneous registration, obtain the registration information of the corresponding authorizer of the special drug and trace the authorizer.

5. The special drug management method based on artificial intelligence according to claim 1 is characterized in that: In step S4, when the special drug The use registration information set is different from the outbound information set of special drugs, that is, When determining special drugs There are unused registrations; Obtain unused registration information of special drugs, re-register unused special drugs into the warehouse, archive all registration information of special drugs, and update the inventory status of special drugs.

6. An artificial intelligence-based special drug management system using the artificial intelligence-based special drug management method according to any one of claims 1 to 5, characterized in that: include: Warehouse registration module, collection registration module, use registration module, artificial intelligence detection model building and analysis module, information tracing module and intelligent storage update module; The entry registration module is used to collect entry information and images of special drugs; the withdrawal registration module is used to obtain the registration information of the authorized person and collect the exit information and images of special drugs; The output end of the storage registration module is connected to the input end of the use registration module; the output end of the use registration module is connected to the input end of the use registration module, and the output end of the use registration module is connected to the input end of the artificial intelligence detection model construction and analysis module; the output end of the artificial intelligence detection model construction and analysis module is connected to the input end of the information tracing module; the output end of the information tracing module is connected to the input end of the intelligent storage update module; The usage registration module is used to collect usage registration information and usage registration images of special drugs; The artificial intelligence detection model building and analysis module is used to build an artificial intelligence detection model to analyze whether there is an erroneous registration of special drugs and to perform empty bottle detection on special drugs after use; The information tracing module is used to trace the registration information of the corresponding authorized person of the special drug when there is an erroneous registration; The intelligent storage update module is used to obtain unused registration information of special drugs, re-register unused special drugs into the warehouse, archive all registration information of special drugs, and update the inventory status of special drugs in real time.

Images