Traditional Chinese medicine decoction piece dispensing and rechecking system based on artificial intelligence

Abstract

The invention discloses a traditional Chinese medicine decoction piece dispensing and rechecking system based on artificial intelligence, and relates to the technical field of artificial intelligence visual identification. According to the traditional Chinese medicine decoction piece dispensing and re-checking system, a hardware system composed of a high-definition camera shooting assembly, a high-precision electronic scale and a computer server is combined with a software system composed of a data acquisition module, an AI processing module, an automatic re-checking module and a communication module to jointly form the traditional Chinese medicine decoction piece dispensing and re-checking system, and the identity and working track of deploying personnel can be recorded through the posture recognition technology; the types and the weights of the prepared traditional Chinese medicine decoction pieces are recorded through image recognition and weighing, meanwhile, the prescription information and the collected information are compared through the automatic re-checking module, effective recording and full-automatic re-checking of the traditional Chinese medicine decoction piece dispensing process can be achieved, the traditional Chinese medicine decoction piece dispensing efficiency and accuracy are effectively improved, and dispensing errors are avoided.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence visual recognition technology, and in particular to an artificial intelligence-based Chinese medicine preparation review system. Background Art

[0002] In traditional technology, the review of Chinese medicine dispensing is done manually, the dispensing efficiency is difficult to improve, the review is difficult, and dispensing errors are prone to occur; and most of the existing automation technologies for dispensing Chinese herbal medicines are concentrated in the field of automation of dispensing Chinese herbal medicines, and the automation of dispensing Chinese herbal medicines is achieved through machine assistance, such as the automated dispensing systems for Chinese herbal medicines disclosed in Chinese invention patents CN201910805513.2 and CN201910011600.0, etc., and there are few studies on the automation of the review of Chinese herbal medicines.

[0003] In the prior art, the process of dispensing Chinese herbal medicines includes five aspects: prescription review, preparation of herbal medicines, review, dispensing, and explanation of medication. The current automation technology solution focuses on preparation of herbal medicines, which mainly solves the problems of slow manual preparation and low work efficiency in the preparation of herbal medicines. However, there is no better technical solution for the accuracy of review and preparation of Chinese herbal medicines. In addition, due to the differences in the preparation of Chinese herbal medicines during processing, the shapes of the herbal medicines vary greatly. In addition, due to automation reasons, if there is an error in the filling of the herbal medicine bucket or the machine takes out the bucket incorrectly, the accuracy of the herbal medicine preparation cannot be avoided, and it is difficult to ensure the accuracy of the Chinese herbal medicine dispensing process, which leads to dispensing errors. Therefore, the present invention proposes an artificial intelligence-based Chinese herbal medicine dispensing review system to solve the problems existing in the prior art. Summary of the Invention

[0004] In response to the above problems, the purpose of the present invention is to propose an artificial intelligence-based Chinese herbal medicine dispensing and review system, which uses artificial intelligence technology, especially image recognition and posture recognition technology, to achieve the tracking of the Chinese herbal medicine dispensing process and the automation of the dispensing review, thereby improving the accuracy of Chinese herbal medicine dispensing.

[0005] In order to achieve the purpose of the present invention, the present invention is implemented through the following technical solutions: an artificial intelligence-based Chinese medicine decoction piece dispensing and review system, including a hardware system and a software system, characterized in that: the hardware system includes a high-definition camera component, a high-precision electronic scale and a computer server, the software system includes a data acquisition module, an AI processing module, an automatic review module and a communication module, the data acquisition module collects Chinese medicine decoction pieces and the body information of the dispensing pharmacist through the high-definition camera component, and collects the weight data of Chinese medicine decoction pieces through the high-precision electronic scale, the AI ​​processing module and the automatic review module are both deployed on the computer server, the AI ​​processing module is used to identify the type of Chinese medicine decoction pieces and the identity of the dispensing pharmacist, the automatic review module is used to compare and review the Chinese medicine prescription information, the identified type of Chinese medicine decoction pieces and the weight data, and the communication module provides a data transmission path between the computer server and the high-definition camera component and the high-precision electronic scale through a wireless network.

[0006] A further improvement is that the high-definition camera assembly includes a panoramic camera arranged in the pharmacy and a high-definition camera set on a temporary storage table. The panoramic camera is used to track and record the work trajectory of the pharmacist, and the high-definition camera is used to collect local high-definition images of Chinese herbal medicines.

[0007] A further improvement is that the high-precision electronic scale is placed on a temporary storage table in the pharmacy and is linked to the high-precision scanner to synchronously trigger image acquisition and weighing operations after the Chinese herbal medicines are placed on the temporary storage table.

[0008] Further improvements are: the AI ​​processing module includes a medicinal piece recognition submodule and a posture recognition submodule. The medicinal piece recognition submodule recognizes the appearance characteristics of medicinal pieces based on a convolutional neural network and outputs the type of medicine. The posture recognition submodule verifies the identity of the dispensing pharmacist through human key point detection technology.

[0009] A further improvement is that the execution strategy of the automatic review module is:

[0010] S1. Compare the types of Chinese herbal medicines identified by the AI ​​processing module with the list of prescribed medicinal materials item by item;

[0011] S2. Compare the weighing data of Chinese herbal medicine pieces with the prescribed dosage;

[0012] S3. If the comparison is consistent, a review pass instruction is generated; otherwise, an error detail is returned.

[0013] A further improvement is that it also includes a task scheduling module, which responds to the patient's check-in operation, automatically allocates the prescription to the designated dispensing pharmacist terminal, and binds the pharmacist's identity information to the prescription task.

[0014] A further improvement is that the task scheduling module activates the AI ​​processing module when the preparation begins, and confirms the operation authority by comparing the physical characteristics of the preparing pharmacist with the bound pharmacist identity database in real time.

[0015] A further improvement is that the computer server is equipped with a storage database to store related data including prescription information, preparation process video and trajectory log, medicinal material identification results and weight data, review results and error records.

[0016] A review method for a Chinese herbal medicine preparation review system based on artificial intelligence, comprising the following steps:

[0017] Step 1: The doctor writes a prescription for Chinese herbal medicine slices. After the patient signs in at the check-in machine, the system's computer server extracts the patient's prescription information, assigns the prescription task to the dispensing pharmacist, and temporarily stores the prescription information;

[0018] Step 2: The dispensing pharmacist accepts the dispensing task and starts dispensing. The dispensing pharmacist's dispensing trajectory is recorded through the pharmacy's panoramic camera, and the dispensing pharmacist's body posture is recognized to confirm the dispensing pharmacist's identity;

[0019] Step 3: After the pharmacist prepares the Chinese herbal medicine slices, he or she places them on a temporary storage table equipped with a high-precision camera and a high-precision electronic scale, and records the weight and type of the slices;

[0020] Step 4: After the prescription is fully prepared, the automatic review module automatically compares the types and weights of the medicinal materials in the prescription. If the review is successful, it will enter the drug dispensing stage. If the review is unsuccessful, an error message will be prompted to notify the dispensing personnel to make corrections.

[0021] The beneficial effects of the present invention are as follows: the present invention forms a Chinese herbal medicine dispensing and review system through a hardware system composed of a high-definition camera component, a high-precision electronic scale and a computer server, combined with a software system composed of a data acquisition module, an AI processing module, an automatic review module and a communication module. The system can record the identity and work trajectory of the dispensing personnel through posture recognition technology, and record the type and weight of the prepared Chinese herbal medicines through image recognition and weighing. At the same time, the automatic review module is used to compare prescription information and collected information, which can realize the effective recording and fully automatic review of the Chinese herbal medicine dispensing process, effectively improving the efficiency and accuracy of Chinese herbal medicine dispensing and avoiding dispensing errors. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a structural diagram of the artificial intelligence-based Chinese herbal medicine preparation review system of the present invention;

[0023] Figure 2 This is a network structure diagram of the artificial intelligence-based Chinese herbal medicine preparation review system of the present invention;

[0024] Figure 3 It is a topological diagram of the artificial intelligence-based Chinese medicine preparation review system of the present invention;

[0025] Figure 4 The present invention is a schematic flow chart of a review method of an artificial intelligence-based Chinese herbal medicine dispensing review system. DETAILED DESCRIPTION

[0026] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0027] Traditional Chinese medicine (TCM) is a natural medicine and preparation used to prevent and treat diseases under the guidance of traditional Chinese medicine theory, including herbal medicines, animal medicines, mineral medicines, etc. TCM dispensing is an important professional technology in TCM. It refers to the operation process of standardizing the dosage and compatibility requirements of various TCM decoction pieces according to the prescription of TCM issued by the physician. Its core is to ensure the safety and effectiveness of the medication.

[0028] Chinese medicine dispensing review is the process in which qualified professionals (usually practicing Chinese medicine pharmacists or experienced Chinese medicine practitioners) recheck and inspect the prepared Chinese medicine after the Chinese medicine prescription is prepared in a Chinese medicine pharmacy or drugstore. This is a key quality control link to ensure that Chinese medicine dispensing is accurate, safe and effective, and is crucial to ensuring the safety of patients' medication.

[0029] In recent years, with the rapid development of information technology, artificial intelligence (AI) has become increasingly widely used in the medical field. Advances in image and body recognition technologies, in particular, have made it possible to automate and intelligently process the dispensing of traditional Chinese medicine (TCM) pieces. The application prospects of AI technology are particularly promising in the field of TCM dispensing and review.

[0030] Example 1

[0031] according to Figure 1 、 Figure 2 、 Figure 3As shown, this embodiment provides a Chinese herbal medicine dispensing and review system based on artificial intelligence, which is composed of a hardware system and a software system, wherein the hardware system includes a high-definition camera component for collecting images of Chinese herbal medicine pieces and body images of dispensing pharmacists, a high-precision electronic scale for collecting weight data of Chinese herbal medicine pieces, and a computer server for identifying the collected images of Chinese herbal medicine pieces and body images of dispensing pharmacists. The high-precision electronic scale of this embodiment is selected from Mettler Toledo XS205DU (range 220g, d=0.01mg) and is equipped with an RS485 interface and Modbus protocol;

[0032] The software system consists of a data acquisition module for controlling the hardware system to collect data, an AI processing module for identifying the types of Chinese herbal medicines and the identity of the dispensing pharmacist, an automatic review module for comparing and reviewing Chinese medicine prescription information, and a communication module for providing wireless data transmission paths for each module of the system. The data acquisition module collects the body information of Chinese herbal medicines and dispensing pharmacists through high-definition camera components, and collects the weight data of Chinese herbal medicines through high-precision electronic scales. The AI ​​processing module is deployed on the GPU cluster (NVIDIA T4 and above) of the computer server, supporting parallel processing of more than 10 video streams. The automatic review modules are all deployed on the computer server, used to compare and review Chinese herbal medicine prescription information, the identified types of Chinese herbal medicines and weight data. The communication module uses dual-band Wi-Fi6 (802.11ax) and 5G redundant backup, with a transmission delay of <100ms, and provides a data transmission path between the computer server and the high-definition camera components and high-precision electronic scales through a wireless network.

[0033] The high-definition camera assembly of this embodiment includes a panoramic camera and a high-definition camera. The panoramic camera is selected from AxisQ1656-DLE (180° field of view) and is installed in the center of the pharmacy ceiling to track and record the work trajectory of the pharmacist. The high-definition camera is set 50 cm above the temporary storage table and is equipped with a ring LED fill light (color temperature 5500K, color rendering index Ra>95) to ensure the shooting effect. It is used to collect local high-definition images of Chinese herbal medicines. The high-precision electronic scale is placed on the temporary storage table in the pharmacy and is linked to the high-definition camera. When the high-precision electronic scale detects a weight change ≥0.1g and remains stable for 500ms, it sends a GPIO signal to start the high-definition camera and trigger the image acquisition operation.

[0034] The AI ​​processing module of this embodiment consists of a decoction piece recognition submodule for identifying Chinese herbal medicines and a posture recognition submodule for identifying the pharmacist who prepares the medicine. The decoction piece recognition submodule adopts the improved ResNeXt-101 (64×4d) architecture, performs transfer learning based on ImageNet pre-training, and integrates an attention mechanism module (CBAM) to focus on extracting the texture (GLCM features) and cross-sectional features of the decoction pieces. It identifies the appearance features of Chinese herbal medicines based on the convolutional neural network (CNN) and outputs the type of medicine. The posture recognition submodule verifies the identity of the pharmacist who prepares the medicine through human key point detection technology, implements 17 key point detection based on the AlphaPose framework, constructs a dynamic gait feature library, and uses Triplet Loss to train the deep metric network to ensure that the cross-view recognition accuracy is greater than 98.7%.

[0035] The strategy for the automatic review module in this embodiment to perform review is:

[0036] S1. Compare the types of Chinese herbal medicines identified by the AI ​​processing module with the list of prescribed medicinal materials item by item;

[0037] S2. Compare the weight data of the Chinese herbal medicine slices weighed by the current high-precision electronic scale with the prescribed dosage;

[0038] S3. Determine the comparison structure. If the comparison is consistent, generate a review pass instruction; otherwise, return error details.

[0039] The Chinese herbal medicine dispensing and review system of this embodiment also includes a task scheduling module, which is used to respond to the patient's check-in operation at the check-in machine, automatically allocate the prescription to the designated dispensing pharmacist terminal, and bind the dispensing pharmacist's identity information to the prescription task. The task scheduling module activates the posture recognition submodule in the AI ​​processing module at the beginning of the dispensing, and confirms the operation authority by comparing the dispensing pharmacist's body characteristics with the bound pharmacist identity database in real time.

[0040] The computer server of this embodiment is equipped with a storage database for storing the following related data:

[0041] Prescription information (including name of medicinal material, dosage, and number of doses);

[0042] Video and track log of the deployment process;

[0043] Medicinal pieces identification results and weight data;

[0044] Review results and error records.

[0045] Example 2

[0046] according to Figure 4 As shown, this embodiment provides a review method of a Chinese herbal medicine preparation review system based on artificial intelligence, comprising the following steps:

[0047] Step 1: The doctor writes a prescription for Chinese herbal medicine slices. After the patient signs in at the check-in machine, the system's computer server extracts the patient's prescription information, assigns the prescription task to the dispensing pharmacist, and temporarily stores the prescription information;

[0048] Step 2: The dispensing pharmacist accepts the dispensing task and starts dispensing. The dispensing pharmacist's dispensing trajectory is recorded through the pharmacy's panoramic camera, and the dispensing pharmacist's body posture is recognized to confirm the dispensing pharmacist's identity;

[0049] Step 3: After the pharmacist prepares the Chinese herbal medicine slices, he or she places them on a temporary storage table equipped with a high-precision camera and a high-precision electronic scale, and records the weight and type of the slices;

[0050] Step 4: After all prescriptions are prepared, the automatic review module automatically compares the types and weights of the medicinal materials in the prescription. If the review is successful, the medicine will be dispensed. If the review is unsuccessful, an error message will be prompted to notify the dispensing staff to make corrections.

[0051] This example uses the BERT-BiLSTM-CRF model to digitally analyze traditional Chinese medicine prescriptions during the review process to accurately extract medicinal material aliases (e.g., "Chuanxiong" → "Chuanxiong"). A three-level weight tolerance strategy is implemented during weight review:

[0052] Expensive and fine medicinal materials (>5 yuan / g): ±0.5%

[0053] Conventional medicinal materials: ±2%

[0054] Leafy herbs: ±5%;

[0055] The error information in this embodiment includes level 1 error, level 2 error, and level 3 error, as follows:

[0056] Level 1 error: Variety mismatch / incompatibility (immediate termination of the process);

[0057] Level 2 error: dose deviation > 10% (requires double review);

[0058] Level 3 error: packaging defect (record release).

[0059] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the foregoing embodiments. The foregoing embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. An artificial intelligence-based Chinese medicine preparation review system, comprising a hardware system and a software system, characterized in that: The hardware system includes a high-definition camera component, a high-precision electronic scale and a computer server. The software system includes a data acquisition module, an AI processing module, an automatic review module and a communication module. The data acquisition module collects the body information of Chinese herbal medicines and the dispensing pharmacist through the high-definition camera component, and collects the weight data of Chinese herbal medicines through the high-precision electronic scale. The AI ​​processing module and the automatic review module are both deployed on the computer server. The AI ​​processing module is used to identify the type of Chinese herbal medicines and the identity of the dispensing pharmacist. The automatic review module is used to compare and review the Chinese medicine prescription information, the identified type of Chinese herbal medicines and the weight data. The communication module provides a data transmission path between the computer server and the high-definition camera component and the high-precision electronic scale through a wireless network.

2. The artificial intelligence-based Chinese medicine preparation review system according to claim 1, characterized in that: The high-definition camera assembly includes a panoramic camera arranged in the pharmacy and a high-definition camera set on a temporary storage table. The panoramic camera is used to track and record the work trajectory of the pharmacist, and the high-definition camera is used to collect local high-definition images of Chinese herbal medicine pieces.

3. The artificial intelligence-based Chinese medicine preparation review system according to claim 2, characterized in that: The high-precision electronic scale is placed on a temporary storage table in the pharmacy and is linked to the high-precision scanner to synchronously trigger image acquisition and weighing operations after the Chinese herbal medicine pieces are placed on the temporary storage table.

4. The artificial intelligence-based Chinese medicine preparation review system according to claim 1, characterized in that: The AI ​​processing module includes a medicinal piece recognition submodule and a posture recognition submodule. The medicinal piece recognition submodule identifies the appearance features of medicinal pieces based on a convolutional neural network and outputs the type of medicine. The posture recognition submodule verifies the identity of the dispensing pharmacist through human key point detection technology.

5. The artificial intelligence-based Chinese medicine preparation review system according to claim 1, characterized in that: The execution strategy of the automatic review module is: S1. Compare the types of Chinese herbal medicines identified by the AI ​​processing module with the list of prescribed medicinal materials item by item; S2. Compare the weighing data of Chinese herbal medicine pieces with the prescribed dosage; S3. If the comparison is consistent, a review pass instruction is generated; otherwise, an error detail is returned.

6. The artificial intelligence-based Chinese medicine preparation review system according to claim 1, characterized in that: It also includes a task scheduling module, which responds to the patient's check-in operation, automatically allocates the prescription to the designated dispensing pharmacist terminal, and binds the pharmacist's identity information to the prescription task.

7. The artificial intelligence-based Chinese herbal medicine preparation review system according to claim 6, characterized in that: The task scheduling module activates the AI ​​processing module when the preparation begins, and confirms the operation authority by comparing the physical characteristics of the preparing pharmacist with the bound pharmacist identity database in real time.

8. The artificial intelligence-based Chinese medicine preparation review system according to claim 1, characterized in that: The computer server is equipped with a storage database that stores related data including prescription information, preparation process video and trajectory log, medicinal material identification results and weight data, review results and error records.

9. A review method for an artificial intelligence-based Chinese herbal medicine preparation review system according to any one of claims 1 to 9, characterized in that: The following steps are involved: Step 1: The doctor writes a prescription for Chinese herbal medicine slices. After the patient signs in at the check-in machine, the system's computer server extracts the patient's prescription information, assigns the prescription task to the dispensing pharmacist, and temporarily stores the prescription information; Step 2: The dispensing pharmacist accepts the dispensing task and starts dispensing. The dispensing pharmacist's dispensing trajectory is recorded through the pharmacy's panoramic camera, and the dispensing pharmacist's body posture is recognized to confirm the dispensing pharmacist's identity; Step 3: After the pharmacist prepares the Chinese herbal medicine slices, he or she places them on a temporary storage table equipped with a high-precision camera and a high-precision electronic scale, and records the weight and type of the slices; Step 4: After the prescription is fully prepared, the automatic review module automatically compares the types and weights of the medicinal materials in the prescription. If the review is successful, it will enter the drug dispensing stage. If the review is unsuccessful, an error message will be prompted to notify the dispensing personnel to make corrections.

Images