Body fluid specimen quality testing methods, devices, transport devices, equipment and media

Abstract

The present application relates to the field of digital medicine and is applied to the field of smart medicine. It provides a method, device, transport device, equipment and medium for detecting the quality of body fluid specimens. The method includes: obtaining a test tube picture of the body fluid specimen; annotating the test tube picture according to preset recognition requirements to obtain an annotated picture; inputting the annotated picture into the recognition model for quality recognition and outputting the recognition result; obtaining the inspection purpose of the body fluid specimen, and configuring the specimen quality rules in the preset inspection knowledge graph according to the inspection purpose; making a quality judgment on the identification result according to the specimen quality rules; and determining whether the body fluid specimen meets the specimen quality requirements according to the result of the quality judgment. The body fluid specimen quality detection method, device, transport device, equipment and medium provided in the present application can quickly and accurately identify the quality of body fluid specimens.

Description

Technical Field

[0001] The present application relates to the technical field of digital medicine, and in particular to a method, device, transport device, equipment and medium for detecting the quality of a body fluid specimen. Background Art

[0002] At present, when third-party health testing institutions or regional health testing centers need to test body fluid specimens, due to the imperfection of equipment or technology, professional logistics specialists are required to go to various medical institutions or regional health testing centers to collect and transport body fluid specimens, conduct preliminary quality inspections, and confirm that the test items are consistent with the specimen storage tubes. For example, whole blood samples need to use purple anticoagulant tubes, serum samples need to use yellow procoagulant tubes, whole blood samples cannot coagulate, serum samples must coagulate, and other specimen requirements. Specimens that meet the requirements are then placed in refrigerators for transportation. When conducting preliminary quality inspections, professional logistics specialists need to have professional knowledge of inspections and be proficient in the specimen quality requirements for various inspection purposes. At the same time, logistics specialists rely on visual inspections and experience judgments, which have low accuracy and efficiency. Summary of the invention

[0003] The main purpose of this application is to provide a method, device, transport device, equipment and medium for testing the quality of body fluid specimens, aiming to solve the technical problem of low accuracy and efficiency of quality testing caused by logistics specialists relying on visual inspection and experience judgment.

[0004] To achieve the above objectives, the present application provides a method for detecting the quality of a body fluid sample, comprising the following steps:

[0005] Pictures of test tubes used to obtain body fluid specimens;

[0006] Annotate the test tube image according to preset recognition requirements to obtain an annotated image;

[0007] Input the annotated image into a recognition model for quality recognition, and output a recognition result; wherein the recognition model is trained based on the Yolo v3 model, and the recognition model is used to recognize one or more of the test tube color, specimen type, and specimen status;

[0008] Obtaining the inspection purpose of the body fluid specimen, and configuring specimen quality rules in a preset inspection knowledge graph according to the inspection purpose; and making quality judgments on the recognition results according to the specimen quality rules;

[0009] Determine whether the body fluid specimen meets specimen quality requirements based on the result of the quality judgment.

[0010] Furthermore, the step of inputting the annotated image into a recognition model for quality recognition and outputting the recognition result includes:

[0011] Convert the labeled image into a TFRecords file;

[0012] Input the TFRecords file into the recognition model for calculation;

[0013] The output of the recognition model is subjected to logistic calculation to output the recognition result.

[0014] Furthermore, the purpose of obtaining the body fluid specimen is to configure a specimen quality rule in a preset inspection knowledge graph according to the inspection purpose; and the step of judging the quality of the recognition result according to the specimen quality rule includes:

[0015] Identify the barcode in the test tube image to obtain the test purpose of the body fluid specimen;

[0016] Obtaining a test knowledge graph, obtaining one or more of the test tube color, specimen type, and specimen status corresponding to the test purpose in the test knowledge graph, and configuring corresponding specimen quality rules;

[0017] The identification result is compared with the sample quality rule to make a quality judgment.

[0018] Furthermore, the step of obtaining a test tube image of a body fluid specimen comprises:

[0019] obtaining a front view and a rear view of the body fluid sample;

[0020] Performing enhancement processing on the front view and the rear view;

[0021] Identifying the left edge of the body fluid sample in the front view after the enhancement process and the right edge of the body fluid sample in the rear view by using edge recognition technology;

[0022] The left edge and the right edge are spliced ​​together to obtain a test tube image of the body fluid specimen.

[0023] Furthermore, the step of labeling the test tube image according to the preset recognition requirements to obtain the labeled image includes:

[0024] Inputting the image into a first image annotation model, annotating the color of the test tube head of the body fluid specimen with a first preset graphic and displaying a corresponding annotation label, thereby obtaining a first image;

[0025] Inputting the first image into a second image annotation model, annotating the specimen type of the body fluid specimen with a second preset graphic and displaying a corresponding annotation label, thereby obtaining a second image;

[0026] The second image is input into the third image annotation model, the specimen state of the body fluid specimen is annotated with a third preset graphic and a corresponding annotation label is displayed to obtain an annotated image.

[0027] Furthermore, after the step of determining whether the body fluid specimen meets the specimen quality requirements according to the result of the quality judgment, the following steps are included:

[0028] If the body fluid specimen meets the specimen quality requirements, obtaining the required environmental temperature of the body fluid specimen;

[0029] The transport temperature of the body fluid specimen is controlled according to the ambient temperature.

[0030] The present application also provides a body fluid sample quality detection device, comprising:

[0031] An acquisition unit, used for acquiring a test tube image of a body fluid specimen;

[0032] A labeling unit, used to label the test tube image according to preset recognition requirements to obtain a labeled image;

[0033] A recognition unit, used for inputting the annotated image into a recognition model for quality recognition and outputting a recognition result; wherein the recognition model is trained based on the Yolo v3 model, and the recognition model is used for recognizing one or more of the test tube color, specimen type, and specimen status;

[0034] A configuration unit, used to obtain the inspection purpose of the body fluid specimen, configure a specimen quality rule in a preset inspection knowledge graph according to the inspection purpose; and perform quality judgment on the recognition result according to the specimen quality rule;

[0035] A determination unit is used to determine whether the body fluid sample meets the sample quality requirements according to the result of the quality judgment.

[0036] The present application also provides a body fluid specimen transport device, comprising:

[0037] A shell, a detection container, a camera module, and a detection module, wherein the detection container, the camera module, and the detection module are arranged on the inner side of the shell, the camera modules include at least two, the camera modules are respectively located on both sides of the detection container and are arranged opposite to each other, the camera module is communicatively connected with the detection module, and the detection module is used to implement the steps of the body fluid sample quality detection method described in any one of the above items.

[0038] The present application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any of the above methods when executing the computer program.

[0039] The present application also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any of the above-mentioned methods are implemented.

[0040] The body fluid specimen quality detection method, device, transportation device, equipment and medium provided in this application perform quality identification through a recognition model, and then automatically load the judgment rules for judgment based on a preset inspection knowledge graph. Compared with manual visual inspection and judging the quality of collected and transported specimens based on experience, it can make quick judgments and has higher quality inspection accuracy. At the same time, it has lower requirements for logistics specialists. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a schematic diagram of the steps of a method for detecting the quality of a body fluid sample in one embodiment of the present application;

[0042] Figure 2 This is a structural block diagram of a body fluid sample quality detection device in one embodiment of the present application;

[0043] Figure 3 This is a schematic diagram of the structure of a body fluid specimen transport device in one embodiment of the present application;

[0044] Figure 4 A schematic block diagram of the structure of a computer device according to an embodiment of the present application.

[0045] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0047] Reference Figure 1 An embodiment of the present application provides a method for detecting the quality of a body fluid sample, comprising the following steps:

[0048] Step S1, obtaining a test tube image of a body fluid specimen;

[0049] Step S2, annotating the test tube image according to preset recognition requirements to obtain an annotated image;

[0050] Step S3, inputting the annotated image into a recognition model for quality recognition, and outputting a recognition result; wherein the recognition model is trained based on the Yolo v3 model, and the recognition model is used to recognize one or more of the test tube color, specimen type, and specimen status;

[0051] Step S4, obtaining the inspection purpose of the body fluid specimen, configuring specimen quality rules in a preset inspection knowledge graph according to the inspection purpose; and performing quality judgment on the recognition result according to the specimen quality rules;

[0052] Step S5, determining whether the body fluid specimen meets specimen quality requirements according to the result of the quality judgment.

[0053] In this embodiment, as described in the above step S1, the image of the body fluid sample is obtained, and specifically, it can be obtained through a body fluid sample transportation device.

[0054] As described in step S2 above, the test tube image is annotated according to the preset recognition requirements. Image annotation is to mark the bounding box including the target in the image (for example, use a rectangular box or an elliptical box to frame the area including the target); for the target recognition task of the image, image annotation is to mark the bounding box including the target in the sample image, and to set a label for the annotated bounding box, which is the category of the target in the bounding box (for example, test tube head color, barcode, specimen type, etc.). Specifically, the test tube head color, barcode, specimen type, specimen status, etc. can be marked. Specimen types such as whole blood, serum, plasma, tissue, urine, feces, pleural effusion, etc. can be annotated through tools such as Labelme, labelImg, Yolo mark, Vatic, etc. to obtain annotated images.

[0055] As described in step S3 above, the recognition model is trained based on the Yolo v3 model of the TensorFlow framework. TensorFlow is a computational process that flows from one end of the flow graph to the other end. TensorFlow is a system that transfers complex data structures to artificial intelligence neural networks for analysis and processing. The number of attempted prediction borders in Yolo v3 has increased by more than 10 times, and it is performed at different resolutions, so the mAP and the detection effect of small objects have been improved to a certain extent. The recognition model is obtained by training the Yolo v3 model. The new network for feature extraction used by Yolo v3 integrates the network Darknet-19 in Yolo v2 and the newly popular residual network (ResNet's residual structure). The residual network can ensure that the network structure can still converge in a very deep case, and the model can be trained. The deeper the network, the better the features expressed, and the better the classification + detection effects. The network makes extensive use of 3*3 and 1*1 convolutional layers connected in sequence, and adds shortcut connections, so its network structure is more complex, with 53 convolutional layers, but it can identify the quality of labeled images faster and more accurately. The recognition model identifies one or more of the test tube color, specimen type, and specimen status.

[0056] As described in the above steps S4-S5, the inspection knowledge graph is obtained, and the inspection knowledge graph includes the quality requirements corresponding to all body fluid specimens. For example, whole blood is required for reticulocyte detection, and serum is required for liver function detection. For reticulocyte detection, purple EDTA anticoagulant tubes are used to store whole blood specimens, while yellow separation gel coagulation tubes are used for liver function detection. Each body fluid specimen has its corresponding inspection purpose. The inspection purpose is obtained, and the specimen quality rules are configured in the inspection knowledge graph according to the inspection purpose. After the identification result of the body fluid specimen is obtained, the configured specimen quality rules and the identification result are used for quality judgment. For example, if the inspection purpose is reticulocyte detection, the corresponding specimen quality rules are whole blood and purple EDTA anticoagulant tubes. If the specimen type is identified as whole blood and the color of the test tube head is purple, the quality judgment result is normal, indicating that the body fluid specimen meets the quality requirements.

[0057] In this embodiment, quality identification is performed through the recognition model, and then the judgment rules are automatically loaded according to the preset inspection knowledge graph for judgment. Compared with manual visual inspection and judging the quality of collected and transported specimens based on experience, it can make quick judgments and the quality inspection accuracy is higher. At the same time, the requirements for logistics specialists are lower.

[0058] In one embodiment, the step S3 of inputting the annotated image into a recognition model for quality recognition and outputting the recognition result includes:

[0059] Step S31, converting the annotated image into a TFRecords file;

[0060] Step S32, inputting the TFRecords file into the recognition model for calculation;

[0061] Step S33, subjecting the output of the recognition model to logistic calculation and outputting the recognition result.

[0062] In this embodiment, as described in step S31 above, the annotated image is converted into a TFRecords file. The TFRecords file is actually a binary file. The TFRecords file contains a tf.train.Example protocol memory block (the protocol memory block contains the field Features). The data is filled into the Example protocol memory block, the protocol memory block is serialized into a string, and written to the TFRecords file through tf.python_io.TFRecordWriter.

[0063] As described in the above steps S32-S33, the TFRecords file is input into the recognition model for calculation, and the output of the recognition model is subjected to logistic calculation. Specifically, it can be calculated by the sigmoid function. The expression of the sigmoid function is: , where x is the output of the recognition model. After the sigmoid function is calculated, a real number between 0 and 1 is obtained, and the recognition result is determined based on the real number. For example, when recognizing the color of a test tube head, the colors of the test tube head include purple and yellow. After the sigmoid function is calculated, the probability that the color of the test tube head belongs to purple is 0, and the probability that it belongs to yellow is 0.8, which indicates that the color of the test tube head belongs to yellow.

[0064] In one embodiment, the inspection purpose of obtaining the body fluid specimen is to configure a specimen quality rule in a preset inspection knowledge graph according to the inspection purpose; and step S4 of judging the quality of the recognition result according to the specimen quality rule includes:

[0065] Step S41, identifying the barcode in the test tube image to obtain the test purpose of the body fluid sample;

[0066] Step S42, obtaining a test knowledge graph, obtaining one or more of the test tube color, specimen type, and specimen status corresponding to the test purpose in the test knowledge graph, and configuring corresponding specimen quality rules;

[0067] Step S43, performing quality judgment on the recognition result and the sample quality rule.

[0068] In this embodiment, as described in the above step S41, a corresponding barcode will be affixed to each test tube after the body fluid specimen is injected. The barcode corresponds to the inspection purpose of each body fluid specimen, such as reticulocyte detection, liver function test, etc.; when taking a picture of the body fluid specimen, the barcode on the test tube will also be displayed on the picture, and machine vision technology can be used to identify the barcode on the picture to obtain the purpose of the specimen inspection.

[0069] As described in the above steps S42-S43, each inspection purpose has its corresponding quality requirements. For example, whole blood is required for reticulocyte detection, and purple EDTA anticoagulant tubes are used to store whole blood specimens. The tube head color, specimen type, and specimen status in the identification result are compared with the tube head color, specimen type, and specimen status in the specimen quality rules. If all comparisons are correct, it means that the storage container is correct and meets the inspection specimen requirements. By inspecting the knowledge graph, the quality requirements of each specimen can be obtained, and the judgment rules corresponding to each body fluid specimen can be determined, which can be applied to the quality inspection of the body fluid specimen.

[0070] In one embodiment, the step S1 of obtaining a test tube image of a body fluid specimen comprises:

[0071] Step S11, obtaining a front view and a rear view of the body fluid sample;

[0072] Step S12, performing enhancement processing on the front view and the rear view;

[0073] Step S13, identifying the left edge of the body fluid sample in the front view after the enhancement process and the right edge of the body fluid sample in the rear view by using edge recognition technology;

[0074] Step S14, splicing the left edge and the right edge to obtain a test tube image of the body fluid specimen.

[0075] In this embodiment, as described in step S11 above, the body fluid specimen is placed in a test tube, which is a cylinder. A camera device is provided on the front and rear sides of the test tube to obtain a front view and a rear view. The two cameras are at the same distance from the test tube and at the same height.

[0076] As described in the above step S12, due to reasons such as light, the obtained front view and rear view may not be clear. By enhancing the front view and rear view, the clarity of the body fluid specimens in the front view and rear view is improved. Specifically, the front view and rear view can be subjected to corrosion processing and expansion processing in turn.

[0077] As described in step S13 above, the left edge of the body fluid specimen in the front view and the right edge of the body fluid specimen in the rear view are identified by edge recognition technology. In another embodiment, the right edge of the body fluid specimen in the front view and the left edge of the body fluid specimen in the rear view can be identified by edge detection technology. Specifically, edge recognition can be performed by Canny, Sobel operator or Prewitt operator.

[0078] As described in step S14, the identified left edge is spliced ​​to the right edge to form a composite image as the image of the body fluid specimen. In this embodiment, since the test tube is cylindrical, the front view and the back view of the body fluid specimen are combined into a composite image as the test tube image of the body fluid specimen, so that the test tube image of the body fluid specimen can be obtained simply and quickly.

[0079] In one embodiment, the step S2 of labeling the test tube image according to the preset recognition requirements to obtain the labeled image includes:

[0080] Step S21, inputting the image into a first image annotation model, annotating the color of the test tube head of the body fluid specimen with a first preset pattern and displaying a corresponding annotation label, thereby obtaining a first image;

[0081] Step S22, inputting the first image into a second image annotation model, annotating the specimen type of the body fluid specimen with a second preset graphic and displaying a corresponding annotation label, thereby obtaining a second image;

[0082] Step S23, inputting the second image into the third image annotation model, annotating the specimen state of the body fluid specimen with a third preset graphic and displaying a corresponding annotation label to obtain an annotated image.

[0083] In this embodiment, three positions are marked in the annotated image, namely, the color of the test tube head, the type of specimen, and the state of the specimen. For example, a part of the test tube head area is selected with a triangular frame in the image, an area that can represent the type of specimen is selected with a rectangular frame, and an area that can represent the state of the specimen is selected with a circular frame. Specifically, the first image annotation model, the second image annotation model, and the third image annotation model can be trained based on Labelme, labelImg, Yolo mark, Vatic, etc., and when subsequently input into the recognition model for recognition, the corresponding graphic area is directly recognized.

[0084] In one embodiment, after step S5 of determining whether the body fluid specimen meets the specimen quality requirements according to the result of the quality judgment, the following steps are included:

[0085] Step S6, if the body fluid sample meets the sample quality requirements; obtaining the required environmental temperature of the body fluid sample;

[0086] Step S7, controlling the transportation temperature of the body fluid specimen according to the ambient temperature.

[0087] In this embodiment, when the body fluid specimen meets the corresponding specimen quality requirements, it needs to be transported, and the ambient temperature required during the transportation of the specimen body fluid is obtained. The transportation temperature in the transportation device is automatically controlled according to the ambient temperature to ensure that the specimen transportation environment meets the inspection requirements.

[0088] The body fluid specimen quality detection method, device, transport device, equipment and medium provided in this application can be applied in the field of smart medical care, accelerate the construction of digital medical care, and thus promote the construction of smart cities.

[0089] The body fluid specimen quality detection method provided in this application can be used in the field of blockchain, and the test tube images are stored in the blockchain network. Blockchain is a new application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the blockchain underlying platform, platform product service layer and application service layer.

[0090] The underlying blockchain platform can include processing modules such as user management, basic services, smart contracts, and operation monitoring. Among them, the user management module is responsible for the identity information management of all blockchain participants, including maintaining public and private key generation (account management), key management, and the maintenance of the correspondence between the user's real identity and the blockchain address (authority management), etc., and, under authorization, supervises and audits the transactions of certain real identities and provides risk control rule configuration (risk control audit); the basic service module is deployed on all blockchain node devices to verify the validity of business requests, and records valid requests to storage after consensus is reached. For a new business request, the basic service first adapts the interface for parsing and authentication (interface adaptation), and then encrypts the business information through the consensus algorithm (consensus management). The smart contract module is responsible for the registration and issuance of contracts, as well as contract triggering and contract execution. Developers can define the contract logic in a programming language and publish it to the blockchain (contract registration). According to the logic of the contract terms, the key or other events are called to trigger the execution and complete the contract logic. It also provides the function of contract upgrade and cancellation. The operation monitoring module is mainly responsible for the deployment, configuration modification, contract setting, cloud adaptation and real-time status visualization output of the product during the product release process, such as alarm, network monitoring, node equipment health monitoring, etc.

[0091] See also Figure 2 , an embodiment of the present application further provides a body fluid sample quality detection device, comprising:

[0092] A first acquisition unit 10 is used to acquire a test tube image of a body fluid specimen;

[0093] A labeling unit 20, used to label the test tube image according to preset recognition requirements to obtain a labeled image;

[0094] The recognition unit 30 is used to input the annotated image into a recognition model for quality recognition and output a recognition result; wherein the recognition model is trained based on the Yolo v3 model, and the recognition model is used to recognize one or more of the test tube color, specimen type, and specimen status;

[0095] A configuration unit 40 is used to obtain the inspection purpose of the body fluid specimen, configure a specimen quality rule in a preset inspection knowledge graph according to the inspection purpose; and perform a quality judgment on the recognition result according to the specimen quality rule;

[0096] The determination unit 50 is used to determine whether the body fluid sample meets the sample quality requirement according to the result of the quality judgment.

[0097] In one embodiment, the identification unit 30 includes:

[0098] A conversion subunit, used to convert the annotated image into a TFRecords file;

[0099] A calculation subunit, used for inputting the TFRecords file into the recognition model for calculation;

[0100] The output subunit is used to perform logistic calculation on the output of the recognition model and output the recognition result.

[0101] In one embodiment, the configuration unit 40 includes:

[0102] An identification subunit, used to identify the barcode in the test tube image to obtain the body fluid specimen for testing purposes;

[0103] A configuration subunit is used to obtain a test knowledge graph, obtain one or more of the test tube color, specimen type, and specimen status corresponding to the test purpose in the test knowledge graph, and configure a corresponding specimen quality rule;

[0104] The judgment subunit is used to perform quality judgment on the recognition result and the specimen quality rule.

[0105] In one embodiment, the first acquisition unit 10 includes:

[0106] An acquisition subunit, used for acquiring a front view and a rear view of the body fluid sample;

[0107] An enhancement subunit, used for enhancing the front view and the rear view;

[0108] an identification subunit, configured to identify the left edge of the body fluid sample in the front view after the enhancement processing and the right edge of the body fluid sample in the rear view by using an edge recognition technology;

[0109] The splicing subunit is used to splice the left edge and the right edge to obtain a test tube image of the body fluid specimen.

[0110] In one embodiment, the labeling unit 20 includes:

[0111] A first annotation subunit is used to input the picture into a first picture annotation model, annotate the color of the test tube head of the body fluid specimen with a first preset graphic and display a corresponding annotation label to obtain a first picture;

[0112] a second annotation subunit, configured to input the first image into a second image annotation model, annotate the specimen type of the body fluid specimen with a second preset graphic and display a corresponding annotation label, so as to obtain a second image;

[0113] The third annotation subunit is used to input the second image into a third image annotation model, annotate the specimen state of the body fluid specimen with a third preset graphic and display a corresponding annotation label to obtain an annotated image.

[0114] In one embodiment, the body fluid sample quality detection device further includes:

[0115] If the body fluid sample meets the sample quality requirements, obtaining the environmental temperature required for the body fluid sample;

[0116] The transport temperature of the body fluid specimen is controlled according to the ambient temperature.

[0117] In this embodiment, the specific implementation of each of the above-mentioned units and sub-units may refer to the description in the above-mentioned method embodiment, and will not be repeated here.

[0118] See also Figure 3 , an embodiment of the present application further provides a body fluid specimen transport device, comprising:

[0119] A shell 1, a detection container 2, a camera module 3, and a detection module 4, wherein the detection container 2, the camera module 3, and the detection module 4 are arranged on the inner side of the shell 1, the camera module 3 includes at least two, the camera modules 3 are respectively located on both sides of the detection container 2 and are arranged opposite to each other, the camera module 3 is communicatively connected with the detection module 4, and the detection module 4 is used to implement the steps of the above-mentioned body fluid sample quality detection method.

[0120] In this embodiment, the camera modules 3 are arranged opposite to each other, located on both sides of the detection container 2, and connected to the inner side of the shell 1. The test tube is cylindrical, and the two camera modules 3 are arranged opposite to each other. The pictures taken by the two camera modules 3 are spliced ​​to obtain the test tube picture of the body fluid specimen. The camera module 3 is connected to the detection module 4 in communication, and the taken pictures can be transmitted to the detection module 4 for quality identification. The detection container 2 is used to place the test tube. Multiple detection containers 2 can be set to place multiple test tubes. When multiple detection containers 2 are set, the detection containers 2 are arranged in a row, and the camera module 3 is located at the midpoint, so that all detection containers 2 can be located within the camera range of the camera module 3. The body fluid specimen transportation device provided in this embodiment can not only transport body fluid specimens, but also perform quality inspection. The body fluid specimen is directly placed in the detection container 2 for quality inspection. After the quality inspection is qualified, it is directly placed in the device for transportation. Furthermore, an indicator light 5 may be provided on the housing 1 of the device. When the quality requirements are met, the indicator light 5 is green, and when they are not met, the indicator light 5 is red. The logistics specialist can clearly know whether the body fluid specimens in the device meet the quality requirements. When the transport device can hold multiple body fluid specimens, an indicator light 5 may be provided on the detection container 2 of each body fluid specimen. The transport specialist can know whether the batch of body fluid specimens meets the quality requirements through the indicator light 5 on the housing 1. If they do not meet the requirements, the transport device is opened and the specific indicator light 5 is used to know which body fluid specimen does not meet the quality requirements. Furthermore, the device also includes a temperature control module, which is in communication with the detection module and is used to control the temperature inside the device according to the ambient temperature required by the body fluid of the specimen.

[0121] Reference Figure 4 In an embodiment of the present application, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store test tube image data, etc. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for detecting the quality of a body fluid specimen is implemented.

[0122] Those skilled in the art will understand that Figure 4 The structure shown in is merely a block diagram of a portion of the structure related to the present application solution and does not constitute a limitation on the computer device to which the present application solution is applied.

[0123] An embodiment of the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, a method for detecting the quality of a body fluid sample is implemented.

[0124] In summary, the body fluid specimen quality detection method, device, transport device, equipment and medium provided in the embodiments of the present application include: obtaining a test tube picture of the body fluid specimen; annotating the test tube picture according to preset recognition requirements to obtain an annotated picture; inputting the annotated picture into a recognition model for quality recognition, and outputting a recognition result; wherein the recognition model is trained based on the Yolo v3 model, and the recognition model is used to identify one or more of the test tube color, specimen type, and specimen status; obtaining the inspection purpose of the body fluid specimen, and configuring the specimen quality rules in the preset inspection knowledge graph according to the inspection purpose; performing quality judgment on the recognition result according to the specimen quality rules; and determining whether the body fluid specimen meets the specimen quality requirements according to the result of the quality judgment. The body fluid specimen quality detection method, device, transportation device, equipment and medium provided in this application perform quality identification through a recognition model, and then automatically load the judgment rules for judgment based on a preset inspection knowledge graph. Compared with manual visual inspection and judging the quality of collected and transported specimens based on experience, it can make quick judgments and has higher quality inspection accuracy. At the same time, it has lower requirements for logistics specialists.

[0125] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0126] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0127] The above description is only a preferred embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly used in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for detecting the quality of a body fluid specimen, characterized in that: The following steps are involved: Pictures of test tubes used to obtain body fluid specimens; The test tube image is annotated according to preset identification requirements to obtain an annotated image; the annotation includes test tube color, specimen type, specimen status, and a barcode; the barcode corresponds to the inspection purpose of each body fluid specimen, and the inspection purpose has a corresponding quality requirement; Input the annotated image into a recognition model for quality recognition, and output a recognition result; wherein the recognition model is trained based on the Yolo v3 model, and the recognition model is used to recognize one or more of the test tube color, specimen type, and specimen status; Obtaining the inspection purpose of the body fluid specimen, and configuring specimen quality rules in a preset inspection knowledge graph according to the inspection purpose; and making quality judgments on the recognition results according to the specimen quality rules; Determining whether the body fluid specimen meets specimen quality requirements according to the result of the quality judgment; The purpose of obtaining the body fluid specimen is to configure a specimen quality rule in a preset inspection knowledge graph according to the inspection purpose; and the step of judging the quality of the recognition result according to the specimen quality rule includes: Identify the barcode in the test tube image to obtain the test purpose of the body fluid specimen; Obtaining a test knowledge graph, obtaining one or more of the test tube color, specimen type, and specimen status corresponding to the test purpose in the test knowledge graph, and configuring corresponding specimen quality rules; The identification result is compared with the sample quality rule to make a quality judgment.

2. The method for detecting the quality of a body fluid specimen according to claim 1, characterized in that: The step of inputting the annotated image into a recognition model for quality recognition and outputting the recognition result comprises: Convert the labeled image into a TFRecords file; Input the TFRecords file into the recognition model for calculation; The output of the recognition model is subjected to logistic calculation to output the recognition result.

3. The method for detecting the quality of a body fluid specimen according to claim 1, characterized in that: The step of obtaining a test tube image of a body fluid specimen comprises: obtaining a front view and a rear view of the body fluid sample; Performing enhancement processing on the front view and the rear view; Identifying the left edge of the body fluid sample in the front view after the enhancement process and the right edge of the body fluid sample in the rear view by using edge recognition technology; The left edge and the right edge are spliced ​​together to obtain a test tube image of the body fluid specimen.

4. The method for detecting the quality of a body fluid specimen according to claim 1, characterized in that: The step of labeling the test tube image according to the preset recognition requirements to obtain the labeled image includes: Inputting the image into a first image annotation model, annotating the color of the test tube head of the body fluid specimen with a first preset graphic and displaying a corresponding annotation label, thereby obtaining a first image; Inputting the first image into a second image annotation model, annotating the specimen type of the body fluid specimen with a second preset graphic and displaying a corresponding annotation label, thereby obtaining a second image; The second image is input into the third image annotation model, the specimen state of the body fluid specimen is annotated with a third preset graphic and a corresponding annotation label is displayed to obtain an annotated image.

5. The method for detecting the quality of body fluid specimens according to claim 1, characterized in that: After the step of determining whether the body fluid specimen meets the specimen quality requirements according to the result of the quality judgment, the method further comprises: If the body fluid sample meets the sample quality requirements, obtaining the environmental temperature required for the body fluid sample; The transport temperature of the body fluid specimen is controlled according to the ambient temperature.

6. A body fluid sample quality detection device, characterized in that: include: An acquisition unit, used for acquiring a test tube image of a body fluid specimen; A labeling unit, used to label the test tube image according to preset recognition requirements to obtain a labeled image; The label includes the color of the test tube, the type of specimen, the state of the specimen, and a barcode; the barcode corresponds to the inspection purpose of each of the body fluid specimens, and the inspection purpose has a corresponding quality requirement; A recognition unit, used for inputting the annotated image into a recognition model for quality recognition and outputting a recognition result; wherein the recognition model is trained based on the Yolo v3 model, and the recognition model is used for recognizing one or more of the test tube color, specimen type, and specimen status; A configuration unit, used to obtain the inspection purpose of the body fluid specimen, configure a specimen quality rule in a preset inspection knowledge graph according to the inspection purpose; and perform a quality judgment on the recognition result according to the specimen quality rule; A determination unit, used to determine whether the body fluid specimen meets the specimen quality requirements according to the result of the quality judgment; The configuration unit comprises: An identification subunit, used to identify the barcode in the test tube image to obtain the body fluid specimen for testing purposes; A configuration subunit is used to obtain a test knowledge graph, obtain one or more of the test tube color, specimen type, and specimen status corresponding to the test purpose in the test knowledge graph, and configure a corresponding specimen quality rule; The judgment subunit is used to perform quality judgment on the recognition result and the specimen quality rule.

7. A body fluid specimen transport device, characterized in that: include: A shell, a detection container, a camera module, and a detection module, wherein the detection container, the camera module, and the detection module are arranged on the inner side of the shell, the camera modules include at least two, the camera modules are respectively located on both sides of the detection container and are arranged opposite to each other, the camera module is communicatively connected with the detection module, and the detection module is used to implement the steps of the body fluid specimen quality detection method described in any one of claims 1 to 5.

8. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the body fluid sample quality detection method described in any one of claims 1 to 5 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the body fluid sample quality detection method according to any one of claims 1 to 5 are implemented.

Images