A medicine transportation management method, system and device and a storage medium

Abstract

The application relates to a medicine transportation management method, system and device and a storage medium, and belongs to the technical field of medicine management. The method comprises the following steps: determining the risk level of medicines; matching a corresponding detection period according to the risk level; in the detection period, an evaluation model is used to process image data and sensing data of the medicines to obtain result data, the image data is used to reflect the appearance shape of the medicines, and the sensing data is used to reflect the physical properties of the medicines and environmental parameters of the medicines. The application has the effect of improving the detection efficiency of toxic and narcotic medicines.

Description

Technical Field

[0001] This application relates to the field of drug management technology, and in particular to a method, system, device and storage medium for managing drug transportation. Background Technology

[0002] Medicines are directly related to human health and safety, and their production, distribution, and use are strictly controlled. Medicines are classified according to their control level, such as narcotic drugs, psychotropic drugs, toxic drugs for medical use, and radioactive drugs.

[0003] When transporting the aforementioned controlled substances, some of them are highly sensitive to environmental parameters, such as temperature, humidity, and pressure. Therefore, it is crucial to constantly monitor whether the environmental parameters meet the requirements for drug transport. However, the current common method for assessing whether the transport of controlled substances meets the requirements is still manual observation and recording. This method is not only inefficient but also lacks accuracy, potentially leading to the failure to detect abnormal controlled substances in a timely manner, resulting in cross-contamination between drugs. Summary of the Invention

[0004] To address the problem of low detection efficiency of controlled drugs, this application provides a drug transportation management method, system, device, and storage medium.

[0005] In a first aspect of this application, a method for managing drug transportation is provided. The method includes:

[0006] Determine the risk level of the drug;

[0007] Match the corresponding detection cycle according to the risk level;

[0008] During the detection cycle, the image data and sensor data of the drug are processed using an evaluation model to obtain result data. The image data is used to reflect the appearance and shape of the drug, and the sensor data is used to reflect the physical properties of the drug and the environmental parameters in which it is located.

[0009] By adopting the above technical solution, drugs are first classified into different risk levels, and then different testing cycles are established for drugs of different risk levels, thereby avoiding high-frequency testing of all drugs and reducing testing efficiency. Secondly, an evaluation model is used to process image data and sensor data to determine the condition of the drug. Compared with traditional manual recording, this testing method improves testing efficiency. In other words, this application has the effect of improving drug testing efficiency.

[0010] In one possible implementation: the image data and sensor data of the drug are processed using an evaluation model to obtain result data, including:

[0011] The state category of the drug is determined based on the image data and sensor data, and the state category is used to reflect the drug status obtained by the evaluation model.

[0012] The result data is obtained by comparing the state category and the standard category. The standard category corresponds to the state category, and the standard category is the image data and sensor data corresponding to the normal state of the drug.

[0013] In one possible implementation: the evaluation model includes a quantitative statistics sub-model, a leakage identification sub-model, and a degradation identification sub-model;

[0014] Determining the state category of the drug based on the image data and sensor data includes:

[0015] The image data and sensor data are processed using the aforementioned quantity statistics sub-model to obtain quantity categories, which are either normal quantity or abnormal quantity.

[0016] The image data and sensor data are processed using the leak identification sub-model to obtain a leak category, which is either a leak occurred or no leak occurred.

[0017] The image data and sensor data are processed using the aforementioned degradation identification sub-model to obtain a degradation category, which is either degradation occurred or no degradation occurred.

[0018] The set of quantity category, leakage category, and deterioration category constitutes the state category.

[0019] By adopting the above technical solution, this application uses a quantitative statistics sub-model, a leakage identification sub-model, and a deterioration identification sub-model to process image data and sensor data respectively, and detects the state category of the drug from multiple dimensions to ensure the accuracy of the drug status obtained subsequently.

[0020] In one possible implementation: the quantity statistics sub-model, the leakage identification sub-model, and the deterioration identification sub-model all employ deep learning models.

[0021] By adopting the above technical solution, the quantity statistics sub-model, leakage identification sub-model, and deterioration identification sub-model all use deep learning models, which can ensure the accuracy of the obtained drug status through the prior training of deep learning models.

[0022] In one possible implementation: determining the risk level of the drug includes:

[0023] Obtain target environmental parameters and actual environmental parameters. The target environmental parameters refer to the environmental requirements that need to be met during normal transportation of medicines, and the actual environmental parameters refer to the environmental parameters provided to the medicine by the medicine box.

[0024] A first risk value is calculated based on the target environmental parameters and the actual environmental parameters;

[0025] Obtain characteristic data of the drug, wherein the characteristic data is the physical properties of the drug;

[0026] A second risk value is calculated based on the aforementioned feature data;

[0027] The total risk value is obtained by calculating the difference between the first risk value and the second risk value.

[0028] The corresponding risk level is matched based on the total risk value.

[0029] By adopting the above technical solution, this application uses the drug's tolerance to the environment of the medicine box and the physical characteristics of the drug itself as the basis for classifying the risk level of the drug, thereby ensuring the accuracy of the risk level classification and providing data support for matching an accurate testing cycle.

[0030] In one possible implementation: the target environmental parameters include a target temperature range, a target humidity range, a target air pressure range, and a target vibration frequency; the actual environmental parameters include an actual temperature range, an actual humidity range, an actual air pressure range, and an actual vibration frequency.

[0031] The calculation of the first risk value based on the target environmental parameters and the actual environmental parameters includes:

[0032] The temperature risk value is calculated based on the target temperature range and the actual temperature range.

[0033] The humidity risk value is calculated based on the target humidity range and the actual humidity range.

[0034] The pressure risk value is calculated based on the target pressure range and the actual pressure range.

[0035] The vibration risk value is calculated based on the target vibration frequency and the actual vibration frequency.

[0036] The first risk value is obtained by summing the temperature risk value, humidity risk value, air pressure risk value, and vibration risk value.

[0037] By adopting the above technical solution, the first risk value is calculated by comprehensively considering multiple factors such as temperature, humidity, air pressure and vibration frequency, thereby ensuring the accuracy of the first risk value and thus providing a guarantee for the accuracy of the subsequent risk level classification.

[0038] In one possible implementation: the step of calculating the temperature risk value based on the target temperature range and the actual temperature range includes:

[0039] Calculate the difference between the center point of the target temperature range and the center point of the actual temperature range, and take the absolute value of the difference as the first difference;

[0040] The second difference is calculated based on the minimum value of the target temperature range and the minimum value of the actual temperature range and / or the maximum value of the target temperature range and the maximum value of the actual temperature range;

[0041] The temperature risk value is obtained by adding the first difference and the second difference.

[0042] By adopting the above technical solution, the higher the overlap between the target temperature range and the actual temperature range, and even the closer the center point of the target temperature range is to the center point of the actual temperature range, the smaller the obtained temperature risk value, thereby reflecting the drug's tolerance to the temperature provided by the medicine box.

[0043] In a second aspect of this application, a drug transport management system is provided. The system includes:

[0044] The data determination module is used to determine the risk level of a drug;

[0045] The data matching module is used to match the corresponding detection cycle according to the risk level;

[0046] The data generation module is used to process the image data and sensor data of the drug using an evaluation model during the detection cycle to obtain result data. The image data is used to reflect the appearance and shape of the drug, and the sensor data is used to reflect the physical properties of the drug and the environmental parameters in which it is located.

[0047] In a third aspect of this application, a drug transport management device is provided. The device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the program to implement any of the aforementioned drug transport management methods.

[0048] In a fourth aspect of this application, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements any of the above-described methods for managing drug transport.

[0049] In summary, this application includes the following beneficial technical effect:

[0050] First, based on the drug's tolerance to the packaging environment and its own physical properties, the risk level of the drugs is classified, and different testing cycles are established for drugs of different risk levels, thereby avoiding high-frequency testing of all drugs and reducing testing efficiency. Second, a quantitative statistical sub-model, a leakage identification sub-model, and a deterioration identification sub-model are used to process image data and sensor data respectively, detecting the drug's state category from multiple dimensions. Finally, the drug's condition is determined by comparing the detected state category with the standard category. That is, combining multiple models to detect and identify the drug's condition ensures the accuracy of the obtained drug condition. Moreover, the detection and identification by the model is more efficient than traditional manual recording. Therefore, this application has the effect of improving the efficiency of drug detection. Attached Figure Description

[0051] Figure 1 This is a schematic diagram of an exemplary operating environment of an embodiment of this application.

[0052] Figure 2 This is a flowchart of a drug transportation management method according to an embodiment of this application.

[0053] Figure 3 This is a block diagram of a drug transport management system according to an embodiment of this application.

[0054] Explanation of reference numerals in the attached drawings: 1. Intelligent robot; 2. Component module; 3. Control device; 31. Data determination module; 32. Data matching module; 33. Data generation module; 4. Display device. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0056] Figure 1 A schematic diagram of an exemplary operating environment in which embodiments of this application can be implemented is shown. The operating environment includes a transfer system and a management system. The transfer system is used to transport medicines, while the management system is used to assist the transfer system in transporting medicines, specifically by monitoring the condition of the medicines to promptly detect abnormal medicines or to provide early warnings of medicines that are about to become abnormal.

[0057] The transport system includes an intelligent robot 1 and multiple modules mounted on the intelligent robot 1, each module also referred to as component module 2. The intelligent robot 1 has multiple medicine boxes, each used to store different types of medicines. In this example, the medicine boxes store controlled substances, hereinafter referred to as medicines. Each of the multiple modules on the intelligent robot 1 contains identical components, such as a miniature camera, a light source, and sensor components (not shown in the figure). Each module is independently located within a medicine box. Specifically, the miniature camera is fixedly mounted on the top of the medicine box, and the light source is also located on the top of the medicine box to provide supplementary lighting for the miniature camera when it photographs the medicines inside. The sensor components consist of various types of sensors, installed at different locations within the medicine box. For example, when the sensor components consist of a temperature and humidity sensor, a vibration sensor, a gravity sensor, and a colorimetric sensor, the temperature and humidity sensor, the vibration sensor, and the colorimetric sensor can be located on the top of the medicine box to measure the temperature and humidity of the space formed by the medicine box, the vibration frequency of the medicines, and the color of the medicines, respectively. The gravity sensor is located at the bottom of the medicine box to measure the weight of the medicines. In other words, the sensors in the sensor assembly select their appropriate installation locations based on their respective detection functions.

[0058] Meanwhile, a radio frequency identification (RFID) electronic tag is also installed on the outer wall of the medicine box. Before being put into use, this electronic tag has been bound to the medicine stored in the medicine box, meaning that the RFID electronic tag can identify the medicine. The RFID electronic tag is communicatively connected to a miniature camera and a sensor assembly, respectively, for wirelessly transmitting image data collected by the miniature camera and sensor data collected by the sensor assembly. The sensor data is a collection of data collected by various types of sensors, such as parameters including the temperature, humidity, and air pressure of the space formed by the medicine box, as well as the vibration frequency, color, and weight of the medicine.

[0059] It should be noted that the intelligent robot 1 used above is an existing robot commonly used for transporting medicines. The shape and model of the intelligent robot 1 used are not limited here, as long as it can achieve the function of transporting medicines. This application, based on the transport function of the intelligent robot 1, adaptively deploys miniature sensors for acquiring images of medicines and sensor components for acquiring sensor data on the intelligent robot 1. In practical applications, other methods can also be used to acquire image data and sensor data of medicines; this example does not impose any limitations.

[0060] The management system includes a control device 3 and a display device 4. The control device 3 is one of the electronic devices in the server and terminal equipment, and the display device 4 is a display screen. Both can be installed on the intelligent robot 1 or in an external computer room. The control device 3 acquires image data and sensor data corresponding to each medicine through RFID electronic tags, and evaluates the condition of the medicines in the intelligent robot 1 based on the image data and sensor data. If the evaluation result of the control device 3 is that the medicine is abnormal, the display device 4 will show the medicine box containing the abnormal medicine, so that the management personnel can intuitively see the location of the medicine box and take timely management actions.

[0061] It should be noted that, Figure 1 The operating environment shown is illustrative only and is not intended to limit the application or use of the embodiments of this application. For example, the operating environment may include multiple intelligent robots 1 and multiple control devices 3. During the transfer of medicines, the condition of the medicines inside each intelligent robot 1 is monitored sequentially by multiple control devices 3. That is, when the distance between the intelligent robot 1 and the control device 3 is less than a preset distance, image data and sensor data are transmitted to the corresponding control device 3 through RFID electronic tags.

[0062] Figure 2 A flowchart of a drug transfer management method according to an embodiment of this application is shown, which is applied to the control device 3 described above. Specifically, the main flow of the drug transfer management method is described below.

[0063] Step S1: Determine the risk level of the drug.

[0064] Before transporting the medicines, they are classified into different risk levels based on their tolerance to the environment of the medicine box and the physical characteristics of the medicines themselves. The specific classification process is shown in steps S11-S16.

[0065] Step S11: Obtain the target environment parameters and the actual environment parameters.

[0066] Target environmental parameters are the environmental requirements that must be met during normal drug transportation. They represent the minimum / maximum temperature, minimum / maximum humidity, minimum / maximum air pressure, and maximum vibration frequency at the critical point between the drug's abnormal and normal state. Excessively low or high temperatures can cause the drug to transition from a normal to an abnormal state; similarly, excessively low or high humidity, air pressure, and vibration frequency can also lead to this transition. Therefore, target environmental parameters include target temperature range, target humidity range, target air pressure range, and target vibration frequency. The target temperature range is the interval between the minimum and maximum temperature values; the target humidity range is the interval between the minimum and maximum humidity values; the target air pressure range is the interval between the minimum and maximum air pressure values; and the target vibration frequency is the maximum vibration frequency. In practical applications, target environmental parameters can include more types of ranges, which are not limited here.

[0067] The correspondence between actual environmental parameters and target environmental parameters refers to the actual temperature range, actual humidity range, actual air pressure range, and actual vibration frequency that the medicine box can provide to the medicine.

[0068] Step S12: Calculate the first risk value based on the target environmental parameters and the actual environmental parameters.

[0069] First, the temperature risk value is calculated based on the target temperature range and the actual temperature range. Specifically, the difference between the center point of the target temperature range and the center point of the actual temperature range is calculated, and the absolute value of this difference is taken as the first difference value. Then, it is determined whether the target temperature range falls within the actual temperature range. If so, the difference between the minimum value of the target temperature range and the minimum value of the actual temperature range is calculated, and the absolute value of this difference is taken as the first comparison value. The difference between the maximum value of the target temperature range and the maximum value of the actual temperature range is also calculated, and the absolute value of this difference is taken as the second comparison value. The maximum value is selected from the first and second comparison values ​​and is used as the second difference value. Finally, the first and second differences are added together to obtain the first risk value. If not, the second difference value is calculated. The calculation of the second difference value has two cases: First, when the minimum value of the target temperature range is less than the minimum value of the actual temperature range, the second difference is the difference between their minimum values, and the absolute value is taken. Second, when the maximum value of the target temperature range is greater than the maximum value of the actual temperature range, the second difference is the difference between their maximum values, and the absolute value is taken. Finally, the first difference and the second difference are added together to obtain the temperature risk value.

[0070] It should be noted that, since the selected intelligent robot 1 has the ability to transport medicines, the actual environmental parameters provided by the intelligent robot 1 to the medicines have at least a certain overlap with the target environmental parameters. For example, at least one end of the target temperature range falls within the actual temperature range.

[0071] Secondly, the humidity risk value is calculated based on the target humidity range and the actual humidity range, and the air pressure risk value is calculated based on the target air pressure range and the actual air pressure range. The specific calculation process can be compared with the process of calculating the temperature risk value mentioned above, and will not be repeated here.

[0072] In addition, the vibration risk value is calculated based on the target vibration frequency and the actual vibration frequency. Specifically, the difference between the target vibration frequency and the actual vibration frequency is calculated. Usually, the target vibration frequency is less than the actual vibration frequency, so the difference between the two is often negative, which also eliminates the influence of the vibration risk value on the subsequent calculation of the first risk value.

[0073] Finally, the sum of the temperature risk value, humidity risk value, air pressure risk value, and vibration risk value is calculated to obtain the first risk value.

[0074] Step S13: Obtain the characteristic data of the drug.

[0075] The characteristic data of pharmaceuticals include parameters such as the physical state, shape, and hardness of the drug. The physical state of a drug includes three states: gaseous, liquid, and solid. The shape of the drug refers to the shape of the packaging box, primarily reflecting the contact area between the box and the medicine. The hardness of the drug refers to the hardness of the packaging box, which is related to the material of the packaging box.

[0076] Step S14: Calculate the second risk value based on the feature data.

[0077] First, a morphology score is obtained based on the drug's physical state. This example pre-defines the morphology score for each state, with the morphology score for gaseous state < the morphology score for liquid state < the morphology score for solid state. Therefore, after determining the drug's physical state, the corresponding morphology score can be obtained based on the drug's physical state matching.

[0078] Then, the contact score is calculated based on the shape of the medicine. Since the larger the contact area between the packaging box and the medicine box, the greater the friction between them, and the more stable the packaging box is during transportation, the contact area between the packaging box and the medicine box is calculated and used as the contact score.

[0079] Similarly, this example pre-sets corresponding hardness scores for packaging boxes with different hardness levels. After obtaining the hardness of the packaging box, the corresponding hardness score can be obtained based on the hardness of the packaging box.

[0080] Finally, the sum of the morphology score, contact score, and hardness score is calculated to obtain the second risk value.

[0081] Step S15: Calculate the difference between the first risk value and the second risk value to obtain the total risk value.

[0082] Step S16: Match the corresponding risk level based on the total risk value. Similarly, this example pre-sets risk levels corresponding to different total risk values, and the higher the total risk value, the higher the corresponding risk level. A higher risk level also indicates that drugs with high risk levels need to be given special attention during transportation.

[0083] Step S2: Match the corresponding detection cycle according to the risk level.

[0084] Similarly, this example also pre-sets the detection cycle corresponding to the risk level. The higher the risk level of the drug, the more detection cycles there are in the same time period, that is, the shorter the time distance between two adjacent detection cycles. In fact, when the risk level of the drug reaches level one, the time distance between two adjacent detection cycles is 0, that is, to keep real-time monitoring of drugs belonging to the level one risk level.

[0085] Step S3: During the detection cycle, the image data and sensor data of the drug are processed using an evaluation model to obtain result data. The image data is used to reflect the appearance and shape of the drug, and the sensor data is used to reflect the physical properties of the drug and the environmental parameters in which it is located.

[0086] Based on the detection cycle obtained in step S2, in each detection cycle, the RFID automatically uploads the image data and sensor data of the medicine to the control device 3. The control device 3 is pre-trained with an evaluation model, which is used to process the image data and sensor data to obtain result data on the condition of the medicine.

[0087] The evaluation model includes a quantitative statistics sub-model, a leakage identification sub-model, and a degradation identification sub-model. These three sub-models are set up in parallel and are used to independently process image data and sensor data. The quantity statistics sub-model, leakage detection sub-model, and spoilage detection sub-model all employ deep learning models. These sub-models can be selected from recurrent neural network (RNN) and convolutional neural network (CNN) models. The working processes of RNN and CNN models are similar. When counting the number of medicines: feature information such as the edges, shape, and texture of the medicines is extracted from the image. Bounding boxes are generated based on this feature information, with each bounding box containing one medicine. The number of bounding boxes is then counted to obtain the total number of medicines. Environmental parameters, specifically the weight of the medicines, are also extracted from the sensor data to verify whether the quantity of medicines has changed. When identifying whether medicines have spoiled, feature information such as the shape, size, and color of the medicines is extracted from the image. Environmental parameters, such as the temperature, humidity, air pressure, and vibration frequency of the space formed by the medicine box, are also extracted from the sensor data to determine whether these parameters exceed the tolerance range of the medicines. Finally, the shape, size, color, and environmental parameters of the medicines are combined to assess whether the medicines have spoiled. The aforementioned leakage identification sub-model can be selected from graph n-neural network (GNN) and graph convolutional network (GCN) models. The working process of GNN and GCN models is similar. Both are used to extract feature information from images to obtain the appearance shape and size of the drug. They also extract the physical properties from the sensor data, which refer to the weight and state of the drug. They determine whether the weight and state of the drug have changed, and comprehensively assess whether the drug has leaked by combining the appearance shape, size, weight change and state change of the drug.

[0088] Since the quantitative statistics sub-model, leakage detection sub-model, and deterioration detection sub-model all use deep learning models, the training process of the three sub-models is similar. The difference lies in the training sample data. The following takes the training of the quantitative statistics sub-model as an example: First, obtain sample training images of various types of drugs and sensor data (drug weight) matching each sample training image. Then, extract the image features of each sample training image, and fuse the image features with the sensor data to obtain the associated sample input data. Then, use the associated sample input data as the input data of the deep learning model to obtain the prediction result output by the model. Finally, based on the prediction result and the corresponding real result of the sample, construct a loss function and iteratively train the deep learning model. After training, the deep learning model outputs the quantitative statistics sub-model.

[0089] Therefore, after obtaining the image data and sensor data corresponding to the drug in step S2, the image data and sensor data are processed by the quantity statistics sub-model, the leakage identification sub-model, and the deterioration identification sub-model, respectively. The quantity statistics sub-model determines whether the quantity of the drug has changed, thus obtaining the quantity category; the leakage identification sub-model determines whether the drug has leaked, thus obtaining the leakage category; and the deterioration identification sub-model determines whether the drug has deteriorated, thus obtaining the deterioration category. The quantity category, leakage category, and deterioration category are then combined to form the state category, that is, the state categories are abnormal quantity / normal quantity, leakage occurred / no leakage occurred, and deterioration occurred / no deterioration occurred.

[0090] Based on the obtained status category, the corresponding standard category is retrieved. For example, if the status category is normal quantity, leakage, or deterioration, then image and sensor data of no leakage and no deterioration are retrieved. By comparing: image data of leakage in the status category with image data of no leakage in the standard category, sensor data of leakage in the status category with sensor data of no leakage in the standard category, image data of deterioration in the status category with image data of no deterioration in the standard category, and sensor data of deterioration in the status category with sensor data of no deterioration in the standard category, it is finally determined whether leakage or deterioration has occurred, thus obtaining the specific condition of the drug, i.e., the result data.

[0091] It should be noted that if the result data contains any of the following situations: abnormal quantity, leakage, or deterioration, the control device 3 will mark the specific medicine box with a special identifier on the display device 4. For example, the display device 4 will show a model of the 3D intelligent robot 1 and highlight the medicine box containing the abnormal medicine in red, so that the management personnel can accurately locate the abnormal medicine.

[0092] In summary, the implementation principle of the drug transportation management method of this application embodiment is as follows: First, based on the drug's tolerance to the environment of the medicine box and the physical characteristics of the drug itself, the risk level of the drug is classified, and different detection cycles are formulated for drugs with different risk levels, thereby avoiding high-frequency detection of all drugs and reducing detection efficiency. Second, a quantitative statistical sub-model, a leakage identification sub-model, and a deterioration identification sub-model are used to process image data and sensor data respectively, and the state category of the drug is detected from multiple dimensions. Finally, the condition of the drug is determined by comparing the detected state category with the standard category. That is, multiple models are combined to detect and identify the condition of the drug, ensuring the accuracy of the obtained drug condition. Moreover, the detection and identification by the model is more efficient than traditional manual recording. Therefore, this application has the effect of improving the efficiency of drug detection.

[0093] Figure 3 A block diagram of a drug transport management system according to an embodiment of this application is shown. The system is applied in a control device 3, specifically an application program within the control device 3. The drug transport management system includes a data determination module 31, a data matching module 32, and a data generation module 33.

[0094] Data determination module 31 is used to determine the risk level of a drug.

[0095] The data matching module 32 is used to match the corresponding detection cycle according to the risk level.

[0096] The data generation module 33 is used to process the image data and sensor data of the drug using an evaluation model during the detection cycle to obtain result data. The image data is used to reflect the appearance and shape of the drug, and the sensor data is used to reflect the physical properties of the drug and the environmental parameters in which it is located.

[0097] The modules described in the embodiments of this application can be implemented in software or hardware. The described modules can also be housed in a processor; for example, a processor may be described as including a data determination module 31, a data matching module 32, and a data generation module 33. The names of these modules do not necessarily limit the module itself; for example, the data determination module 31 may also be described as "a module for determining the risk level of a drug."

[0098] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the described module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0099] To better execute the procedures described above, this application also provides a drug transport management device, which includes a memory and a processor.

[0100] The memory can be used to store instructions, programs, code, code sets, or instruction sets. The memory may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for at least one function, and instructions for implementing the aforementioned drug transportation management method, etc.; the data storage area may store data involved in the aforementioned drug transportation management method, etc.

[0101] A processor may include one or more processing cores. The processor executes instructions, programs, code sets, or instruction sets stored in memory, and calls data stored in memory to perform various functions and process data as described in this application. The processor may be at least one of a specific application-specific integrated circuit, a digital signal processor, a digital signal processing device, a programmable logic device, a field-programmable gate array, a central processing unit, a controller, a microcontroller, and a microprocessor. It is understood that, for different devices, the electronic devices used to implement the above-described processor functions may also be other types, and the embodiments of this application do not specifically limit the specific implementation.

[0102] This application also provides a computer-readable storage medium, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and other media capable of storing program code. This computer-readable storage medium stores a computer program that can be loaded by a processor and executed as described above for the management method of drug transportation.

[0103] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the foregoing disclosed concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A method of managing the transfer of pharmaceutical products, characterized in that, The method comprises the following steps: determining the risk level of the medicine; matching a corresponding detection period according to the risk level; processing image data and sensor data of the medicine in the detection period by using an evaluation model to obtain result data, the image data being used to reflect the appearance shape of the medicine, and the sensor data being used to reflect the physical properties of the medicine and the environmental parameters thereof, the processing of the image data and the sensor data by using the evaluation model to obtain the result data comprising: determining the state category of the medicine according to the image data and the sensor data, the state category being used to reflect the medicine condition evaluated by the evaluation model; comparing the state category with a standard category to obtain result data, the standard category corresponding to the state category, and the standard category being corresponding image data and sensor data when the medicine condition is normal, the evaluation model comprising a quantity statistical sub-model, a leakage identification sub-model and a deterioration identification sub-model; the determination of the state category of the medicine according to the image data and the sensor data comprising: processing the image data and the sensor data by using the quantity statistical sub-model to obtain a quantity category, the quantity category being one of quantity normal and quantity abnormal; processing the image data and the sensor data by using the leakage identification sub-model to obtain a leakage category, the leakage category being one of leakage occurrence and no leakage occurrence; processing the image data and the sensor data by using the deterioration identification sub-model to obtain a deterioration category, the deterioration category being one of deterioration occurrence and no deterioration occurrence; the set of the quantity category, the leakage category and the deterioration category being the state category.

2. The method of managing the transfer of pharmaceutical products according to claim 1, wherein, The quantity statistical sub-model, the leakage identification sub-model and the deterioration identification sub-model all use a deep learning model.

3. The method of managing the transfer of pharmaceutical products according to claim 1, wherein, The determination of the risk level of the medicine comprising: obtaining target environmental parameters and actual environmental parameters, the target environmental parameters being environmental requirements that need to be met when the medicine is normally transported, and the actual environmental parameters being environmental parameters provided by a medicine box for the medicine; calculating a first risk value according to the target environmental parameters and the actual environmental parameters; obtaining characteristic data of the medicine, the characteristic data being physical properties of the medicine; calculating a second risk value according to the characteristic data; calculating a total risk value by subtracting the second risk value from the first risk value; and matching a corresponding risk level according to the total risk value.

4. The method of managing the transfer of pharmaceutical products according to claim 3, wherein, The target environmental parameters comprise a target temperature range, a target humidity range, a target air pressure range and a target vibration frequency, and the actual environmental parameters comprise an actual temperature range, an actual humidity range, an actual air pressure range and an actual vibration frequency. The calculation of the first risk value according to the target environmental parameters and the actual environmental parameters comprising: calculating a temperature risk value according to the target temperature range and the actual temperature range; calculating a humidity risk value according to the target humidity range and the actual humidity range; calculating an air pressure risk value according to the target air pressure range and the actual air pressure range; and calculating a vibration risk value according to the target vibration frequency and the actual vibration frequency. Sum the temperature risk value, humidity risk value, air pressure risk value and vibration risk value to obtain the first risk value.

5. The method of managing the transfer of pharmaceutical products according to claim 4, wherein, The temperature risk value is calculated according to the target temperature range and the actual temperature range, including: Calculate the difference between the center point of the target temperature range and the center point of the actual temperature range, and take the absolute value of the difference as the first difference value; According to the minimum value of the target temperature range and the minimum value of the actual temperature range and / or the maximum value of the target temperature range and the maximum value of the actual temperature range, a second difference value is calculated; The first difference value and the second difference value are added to obtain the temperature risk value.

6. A management system for the transport of pharmaceutical products, characterized in that, Including: The data determination module (31) is used to determine the risk level of the medicine; The data matching module (32) is used to match the corresponding detection period according to the risk level; The data generation module (33) is used to process the image data and sensor data of the medicine in the detection period by using an evaluation model to obtain result data, the image data is used to reflect the appearance shape of the medicine, and the sensor data is used to reflect the physical properties of the medicine and the environmental parameters thereof, and the evaluation model is used to process the image data and sensor data of the medicine to obtain result data, including: According to the image data and sensor data, the state category of the medicine is determined, and the state category is used to reflect the medicine condition evaluated by the evaluation model; Comparing the state category and the standard category to obtain result data, the standard category corresponds to the state category, and the standard category is the image data and sensor data when the medicine condition is normal, and the evaluation model includes a quantity statistical sub-model, a leakage identification sub-model and a deterioration identification sub-model; According to the image data and sensor data, the state category of the medicine is determined, including: The quantity statistical sub-model is used to process the image data and sensor data to obtain a quantity category, and the quantity category is one of quantity normal and quantity abnormal; The leakage identification sub-model is used to process the image data and sensor data to obtain a leakage category, and the leakage category is one of leakage and no leakage; The deterioration identification sub-model is used to process the image data and sensor data to obtain a deterioration category, and the deterioration category is one of deterioration and no deterioration; The set of the quantity category, the leakage category and the deterioration category is the state category.

7. An apparatus, comprising: Including a memory and a processor, the memory has a computer program stored thereon, and the processor implements the method of any one of claims 1-5 when executing the program.

8. A computer-readable storage medium, characterized in that, A computer program is stored thereon, and the program is executed by a processor to implement the method of any one of claims 1-5.

Images