CN111403006A

CN111403006A - Microorganism detection system and device

Info

Publication number: CN111403006A
Application number: CN202010492258.3A
Authority: CN
Inventors: 吴刚
Original assignee: Chengdu Yishitong Biotechnology Co ltd
Current assignee: Chengdu Yishitong Biotechnology Co ltd
Priority date: 2020-06-03
Filing date: 2020-06-03
Publication date: 2020-07-10
Anticipated expiration: 2040-06-03
Also published as: CN111403006B

Abstract

The application provides a microorganism detection system and a microorganism detection device, wherein the system comprises detection equipment, a user terminal and a server; the detection equipment is used for acquiring at least one image of a detection area on the kit and sending the at least one image to the bound user terminal; the user terminal is used for receiving at least one image and sending the at least one image to the server; the server is used for identifying each image in the at least one image by using the image identification model to obtain the identification result of each image, determining the detection result of the current detection of the detected user according to the identification result of the at least one image, and sending the detection result to the user terminal. The microorganism detection system provided by the embodiment can be used for wireless monitoring of microorganism screening results in a plurality of different scenes, people with symptom and asymptomatic infection can be found in time, the cost is low, the coverage range is wide, and the risk of cross infection is reduced because a user does not need to go to a hospital for examination.

Description

Microorganism detection system and device

Technical Field

The application relates to the technical field of data processing, in particular to a microorganism detection system and a microorganism detection device.

Background

The new coronavirus (2019-nCoV) is a new coronavirus strain which has extremely strong infectivity and can cause fever, cough, dyspnea and even death of an infected person. The incubation period of pneumonia caused by the new coronavirus infection is 1-14 days, and the infected person has obvious infectivity in the whole course of disease, so that early and timely diagnosis can effectively control the infection of the disease and is helpful for giving medical support to the patient in advance to improve the disease outcome. The current screening for new coronavirus pneumonia requires a comprehensive judgment of the results of clinical symptoms, blood routine, CT examination or nucleic acid examination. CT inspection and nucleic acid inspection all need independent place, expensive equipment and professional personnel, can't satisfy the demand of screening to crowd on a large scale at present, simultaneously, because the user need go to hospital's inspection, still have cross infection's risk.

Disclosure of Invention

An object of the embodiment of the application is to provide a microorganism detection system and device, can be used for screening the microorganism of crowd in a large scale, have better detection effect to new coronavirus, can reduce the risk of cross infection between the crowd simultaneously.

In a first aspect, an embodiment of the present application provides a microorganism detection system, which includes a detection device, a user terminal, and a server; the detection equipment is used for acquiring at least one image of a detection area on the kit and sending the at least one image to the bound user terminal; wherein the kit is located within the detection device; the user terminal is used for receiving the at least one image and sending the at least one image to the server; the server is used for identifying each image in the at least one image by using the image identification model to obtain the identification result of each image, determining the detection result of the current detection of the detected user according to the identification result of the at least one image, and sending the detection result to the user terminal.

In the scheme, the system can realize self-detection, self-isolation and automatic uploading of detection results of the crowd to the server at home with lower cost, and can meet the wireless monitoring requirements on the microorganism screening results in a plurality of different scenes, for example, for new coronavirus, more symptomatic people and asymptomatic infected people who cannot be hospitalized in time can be found as far as possible. Meanwhile, for the user, the risk of cross infection is greatly reduced because the user does not need to go to a hospital for examination.

In an alternative embodiment, the system further comprises a physician terminal; the doctor terminal is used for receiving the target image sent by the server, receiving the annotation of the doctor on the target image and sending the annotated target image to the server; the server is used for determining the detection result of the current detection of the detected user according to the identification result of the at least one image, and comprises the following steps: sending a target image of which the recognition result meets a preset condition in the at least one image to the doctor terminal; receiving the labeled target image returned by the doctor terminal, and updating the identification result of the target image in the at least one image by the label of the doctor to the target image; and determining the detection result of the current detection of the detected user according to the updated identification result of the at least one image.

After the server identifies the image of the user to be detected, a primary identification result is obtained, the image meeting the conditions is sent to the doctor terminal, the doctor rechecks the image through the doctor terminal, and the obtained detection result is double-guaranteed by the image identification model and the doctor rechecking, so that the efficiency is improved, and the result is accurate.

In an alternative embodiment, the identification result of each image identified by the server comprises a microorganism detection result and a confidence of the microorganism detection result; the preset conditions include: the confidence of the microorganism detection result of the target image is lower than a threshold value; and/or the microorganism detection result of the target image is positive.

The server can send the images which cannot be identified in the at least one image and the positive images to the doctor terminal so as to improve the accuracy of the detection result.

In an optional embodiment, the server is further configured to: and after receiving the labeled target image returned by the doctor terminal, optimizing the image recognition model by using the target image and the label of the doctor on the target image to obtain an optimized image recognition model, and taking the optimized image recognition model as the image recognition model used in the next detection.

The server can optimize the image recognition model by utilizing the marking of the doctor on the target image, the precision of the image recognition model is improved along with the increase of marked target images, and the obtained recognition result is more accurate.

In an optional implementation manner, the user terminal is further configured to: before the detection equipment collects at least one image of a detection area on the kit, receiving identity information input by a detected user, and carrying out living body face collection on the detected user to obtain a face image of the detected user; calling a public security interface to verify the identity information and the face image of the detected user; and after the verification is passed, completing the identity authentication of the detected user to obtain verified identity information.

In order to accurately correlate the data of each test with the user, the microorganism detection system can only provide detection service for the authenticated user, so that the tested user needs to be authenticated.

In an optional implementation manner, the user terminal sends the at least one image to the server, and also sends pre-stored user information of the user to be detected and the unique identifier detected this time to the server; wherein, the user information of the tested user comprises the verified identity information and the current position information of the tested user; and the server is also used for storing the at least one image in a correlation manner according to the user information of the detected user and the unique identification detected this time.

By storing the image data detected each time in association with the user information and the unique identifier, confusion among different users and confusion among the data detected each time can be avoided.

In an optional implementation manner, the unique identifier of the current detection is an equipment identifier of the detection equipment or a reagent identifier of the kit, the equipment identifier of the detection equipment is used for uniquely characterizing the detection equipment, and the reagent identifier of the kit is used for uniquely characterizing the kit.

In an optional embodiment, the detection device is further configured to: before at least one image of a detection area on a kit is collected, sending identification information to the user terminal, wherein the identification information is an equipment identification of the detection equipment or a reagent identification of the kit, the equipment identification of the detection equipment is used for uniquely representing the detection equipment, and the reagent identification of the kit is used for uniquely representing the kit; the user terminal is also used for sending the identification information and the user information of the detected user to the server together so as to request the server for the unique identification detected this time; the server is used for generating the unique identification detected this time according to the identification information and the user information, and sending the unique identification detected this time to the user terminal.

In an optional implementation manner, the user terminal is further configured to: before the detection equipment collects at least one image of a detection area on a kit, a communication address of the detection equipment is obtained by scanning an identification code, and a binding request is sent to the detection equipment through the communication address; the detection device is further configured to establish a binding relationship with the user terminal based on the binding request.

In an alternative embodiment, the detection device is configured to collect at least one image of a detection area on a reagent cartridge and send the at least one image to a bound user terminal, and includes: after receiving the trigger of the tested user, starting timing; acquiring images of a detection area on the kit within a preset time length to obtain at least one image of the detection; and sending the at least one image to a user terminal which is in binding relation with the detection equipment through a wireless communication module.

In an optional embodiment, the server is further configured to: and when the detection result of the current detection of the detected user is positive, sending the user information of the detected user and the detection result to an epidemic situation prevention and control center together. Therefore, the system can provide timely and effective epidemic situation information for epidemic situation prevention and control.

In a second aspect, an embodiment of the present application provides a microorganism detection apparatus configured in a server, the apparatus including: the image receiving module is used for receiving at least one image which is sent by the user terminal and obtained by the detection; the at least one image is obtained by acquiring an image of a detection area on the kit by detection equipment bound with the user terminal and sending the acquired image to the user terminal; the image identification module is used for identifying each image in the at least one image by using the image identification model to obtain an identification result of each image; and the microorganism detection module is used for determining the detection result of the current detection of the detected user according to the identification result of the at least one image and sending the detection result to the user terminal.

In an alternative embodiment, the microbiological detection module comprises: the result screening unit is used for sending the target image of which the identification result meets the preset condition in the at least one image to the doctor terminal; the result receiving unit is used for receiving the labeled target image returned by the doctor terminal and updating the identification result of the target image in the at least one image by the label of the doctor to the target image; and the detection unit is used for determining the detection result of the current detection of the detected user according to the updated identification result of the at least one image.

In an alternative embodiment, the identification result of each image identified by the image identification module comprises a microorganism detection result and a confidence of the microorganism detection result; the preset conditions include: the confidence of the microorganism detection result of the target image is lower than a threshold value; and/or the microorganism detection result of the target image is positive.

In an alternative embodiment, the apparatus further comprises: and the model optimization module is used for optimizing the image recognition model by using the target image and the label of the doctor on the target image after the result receiving unit receives the labeled target image returned by the doctor terminal to obtain an optimized image recognition model, and the optimized image recognition model is used as the image recognition model used by the image recognition module in the next detection.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments of the present application will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and that those skilled in the art can also obtain other related drawings based on the drawings without inventive efforts.

Fig. 1 is a schematic structural diagram of a detection apparatus provided in an embodiment of the present application;

fig. 2 is another schematic structural diagram of a detection apparatus provided in an embodiment of the present application;

FIG. 3 is a schematic view of a microorganism detection system provided in an embodiment of the present application;

FIG. 4 is a schematic view of a detection process for each apparatus in the microbiological detection system;

FIG. 5 is a schematic view of a microorganism detection apparatus provided in an embodiment of the present application;

FIG. 6 is another schematic view of a microorganism detection apparatus provided in the embodiments of the present application.

Icon: 100-a detection device; 101-a housing; 102-a circuit board; 103-an image acquisition device; 104-a trigger device; 105-a blood drip; 106-a power supply module; 107-indicator light; 210-a server; 220-user terminal.

Detailed Description

The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application.

The microorganism rapid detection product (kit) developed based on the immune colloidal gold technology has the advantages of rapidness, simplicity, convenience, high stability, low price and self-detection, and can effectively meet the requirement of the current disease on the common screening of all people. But its use needs to be done in the medical institution and it is difficult to penetrate into a wider area of the population. Based on the kit product, the embodiments of the present application provide a microorganism detection system, which can effectively detect microorganisms of a user to be detected, it should be noted that the microorganisms detected by the present application may include but are not limited to viruses, bacteria, fungi and other microorganisms, any object detectable by the colloidal gold test paper in the kit should be included in the detection range of the present application, and the microorganisms detectable by the present application may be varied in many ways when the kit is replaced. For new coronavirus, the microorganism detection system can help a disease control center to screen more people with symptoms and people without symptoms and infection, and solve the problem of epidemic prevention and control.

The present embodiment provides a microorganism detection system, which can detect a new coronavirus that has been abused currently, but it should be understood that the new coronavirus is only one of the microorganism detection scenarios in the present embodiment, and does not mean that the microorganism detection system can only be used to detect the new coronavirus, and for other infectious diseases, microorganisms that cause diseases, and the like, the microorganism detection system in the present embodiment may also be used to perform detection.

In order to construct the microorganism detection system, the application provides a special detection device, and fig. 1 to 2 show a structural schematic diagram of the detection device. Referring to fig. 1 and 2, the detecting apparatus 100 includes: a housing 101, a circuit board 102, an image capture device 103, and a trigger device 104. Wherein the detection device 100 is provided with a reagent box containing area for placing a reagent box with microorganism detection function, in an alternative embodiment, one end of the detection device 100 is provided with an opening through which the reagent box can be inserted into the detection device 100 and taken out of the detection device 100; in another alternative embodiment, a cover plate area is formed on the casing 101 of the detection device 100, and after a reagent cover is tightly closed with the cover plate area, a hollow reagent box containing area is formed in the detection device 100, and after the reagent cover is removed or opened, a reagent box can be put into the reagent box containing area, and the reagent box can be taken out from the reagent box containing area. The blood storage point is arranged on the kit, a detected user can realize detection by dripping blood to the blood storage point of the kit, the surface of the detection device 100 is provided with the blood dripping hole 105, the position of the blood dripping hole 105 is coincident with the position of the blood storage point of the kit after the detection device 100 is placed in the kit, so that after the kit is placed on the detection device 100, the detected user only needs to drip blood into the blood dripping hole 105 of the detection device 100, and the dripped blood can reach the blood storage point of the kit. It will be appreciated that the above-described blood collection site is also suitable for use in the detection of saliva or other body fluids, depending on which body fluid is required for the detection of the kit, e.g., a test strip for diabetes in the kit, and then the blood collection site is also suitable for the detection of urine, e.g., an antibody or nucleic acid amplification product extracted from a collected nasopharyngeal swab, etc.

The circuit board 102 is disposed inside the housing 101, and is configured to complete functions that the detection device 100 needs to implement in this embodiment, the circuit board 102 is provided with a wireless communication module, and the detection device 100 may communicate with a bound user terminal through the wireless communication module, for example, may communicate in a wireless manner such as bluetooth, mobile communication, near field communication NFC, WIFI, fifth generation mobile communication 5G, and the like. The image acquisition device 103 is arranged at a position on the detection equipment 100 where a detection area on the kit can be completely acquired, and in part of kit products, one or more detection lines are arranged on the detection area, so that whether the detected user is infected with corresponding microorganisms or not can be known by observing the color development conditions on the detection lines in the detection area. In one embodiment, the image capture device 103 may be disposed inside the housing 101 above and directly opposite the examination area. Image acquisition device 103 can be for the macro camera, for obtaining better formation of image effect, still is equipped with the light filling lamp in check out test set, and under dim environment, the light filling lamp lights to make image acquisition device 103 adapt to different collection environment. The triggering device 104 can receive a trigger of a user to start the detection process by the detection apparatus 100, wherein the triggering device 104 may be an "on" key of an entity, and when the "on" key is pressed by the detected user, the detection apparatus 100 starts the detection process, and in addition, the triggering device 104 may also be another type of touch device. For example, a touch area is set on the upper surface of the detection device 100, and when the detected user clicks or touches the touch area, the detection device 100 starts a detection process; the triggering device 104 may also be a touch display screen disposed on the upper surface of the detection apparatus 100, and when the detected user clicks a virtual "on" button on the touch display screen, the detection apparatus 100 starts the detection process, and meanwhile, the touch display screen may also be used to display the current progress of the detection process and other information.

Optionally, the detection apparatus 100 further includes: a power supply module 106, configured to supply power to each circuit module in the detection apparatus 100; the indicator light 107 is used for indicating the current state of the detection device 100, for example, after the detection device 100 is powered on, the indicator light 107 is turned on, and after the detection device 100 starts a detection process, the indicator light 107 is turned on.

The structure of the detection device shown in fig. 1 and fig. 2 is only an optional example, and the specific structure of the detection device can be flexibly developed according to the functions that the detection device needs to implement, and on the basis of fig. 1 and fig. 2, the structure of the detection device can be further changed, which is not described in detail in this embodiment.

Fig. 3 shows a schematic view of a microorganism detection system provided in an embodiment of the present application, and as shown in fig. 3, the microorganism detection system includes: detection device 100, user terminal 220, and server 210. There may be a plurality of user terminals 220 and a plurality of detection devices 100. The user terminal 220 is used for collecting user information of a detected user and binding the detection device 100, and the user terminal 220 can bind a plurality of detection devices 100; the detection device 100 is configured to collect image data of a kit located in the detection device, and send the collected image data to the bound user terminal 220; the user terminal 220 is used for sending the user information and the image data sent by the detection device 100 to the server 210; the server 210 is configured to receive data sent by the user terminal 220, recognize an image by using a trained image recognition model, and finally obtain a detection result of the detected user.

In a more specific embodiment, each device in the microbiological detection system is used to perform the detection process shown in FIG. 4. As shown in fig. 4, the detection process includes:

step 310: the detection equipment collects at least one image of a detection area on the kit and sends the detected at least one image to the bound user terminal.

The reagent kit is positioned in the detection equipment, and specifically, the reagent kit can be inserted into the detection equipment through an opening at one end of the detection equipment, or can be placed into the detection equipment through a cover plate area on a shell of the detection equipment. The kit has a microorganism detection function, and can be used for detecting new coronavirus or other infectious disease viruses, such as influenza, African swine fever and the like. Besides, the detection test paper in the kit can also be diabetes test paper, pregnancy test paper and the like.

Step 320: the user terminal receives the at least one image and sends the at least one image to the server.

Optionally, the user terminal may further send the pre-stored user information of the detected user to the server together with the at least one image.

Step 330: and the server identifies each image in the at least one image by using the image identification model to obtain the identification result of each image.

Step 340: the server determines the detection result of the current detection of the detected user according to the identification result of the at least one image and sends the detection result to the user terminal.

Step 310 and step 340 describe a process of detecting a certain detected user.

Optionally, the user terminal provides a service to the user to be tested through an APP, an applet, or a microservice, and before step 310, the user to be tested needs to complete a corresponding setting operation, which includes: registration and login of APP, identity information authentication, equipment binding detection and the like.

In one embodiment, the user under test first registers and logs in for APP, applet or microservice. Meanwhile, in order to provide accurate information for epidemic situation prevention and control and accurately associate data detected each time with a user, the microorganism detection system can only provide detection service for an authenticated user, and after the detected user completes registration and login, the user terminal is further used for executing the following steps:

1. and receiving identity information input by the tested user. The identity information comprises an identity card number and a name of the tested user.

2. And starting image scanning, and carrying out living body face acquisition on the detected user to obtain a face image of the detected user.

3. And calling a public security interface to verify the identity information and the face image of the detected user, comparing the identity information and the face image with the information in the public security system, if the comparison result is consistent, the verification is passed, and if the comparison result is inconsistent, the verification is not passed.

4. And after the verification is passed, completing the identity authentication of the detected user to obtain verified identity information. If the verification fails, the detected user can be refused to enter the next detection process.

In a specific embodiment, a user to be tested needs to open a location service, a user terminal may require the user to open the location service to continue operation, and in the detection process, the user terminal needs to record a location address of the user to be tested detected this time, and the location address and related information of the detection this time are sent to a server together. If the user terminal does not support the location service, the tested user can carry out detection by filling in the location address.

After the identity information authentication is completed, the user terminal needs to be bound with the detection device. Each detection device has a unique device identification, the device identification is used for uniquely representing the detection device, and the device identifications of different detection devices are different.

Optionally, before the detection starts, in order to implement the binding between the user terminal and the detection device, the user terminal is configured to obtain a communication address of the detection device by scanning an identification code on the detection device, and initiate a binding request to the detection device through the communication address, and the detection device is configured to establish a binding relationship with the user terminal based on the binding request. Specifically, the user terminal may obtain a communication address of the detection device by scanning an identification code, such as a barcode or a two-dimensional code, on the surface of the detection device, where the communication address may be a bluetooth address of the detection device or another communication address, and for example, taking a bluetooth address as an example, after obtaining the bluetooth address of the detection device, the user terminal establishes a bluetooth binding relationship with the detection device through the bluetooth address. After the detection equipment is powered on, the Bluetooth module is automatically started to wait for the binding of the user terminal. The user terminal may bind a plurality of detection devices.

In this embodiment, the identification code may be attached to the detection device, and the attachment position of the identification code may be any position of the upper surface, the lower surface, the side surface, and the like of the detection device; of course, the identification code may also be displayed through a display unit on the detection device, for example, after the detection device is powered on, the bluetooth module is turned on, and the corresponding barcode or two-dimensional code is displayed on the display unit at the same time.

After the user terminal binds the detection equipment, the detected user can perform related detection actions. Specifically, the user to be tested drops blood and diluent into the blood dropping hole on the detection device, the reagent kit is placed in the detection device at the moment, the position of the blood dropping hole coincides with the position of the blood storage point of the reagent kit, and the blood reaches the blood storage point on the reagent kit through the blood dropping hole, or the blood and the diluent can be firstly stored in the blood on the reagent kit and then the reagent kit is placed in the detection device. Then, the detected user can start the detection process through the trigger device on the detection device. It is understood that the kit located in the detection device may be a detection kit for a new coronavirus if it is required to detect whether the user to be detected is infected with the new coronavirus, or may be a detection kit for a corresponding virus if it is required to detect whether the user to be detected is infected with other infectious diseases virus, such as influenza, african swine fever, etc.

After the above operations are completed, the detection apparatus may perform the above step 310, and perform image acquisition on the detection area on the reagent cartridge through the image acquisition device. The detection process of the kit needs a period of time, and a corresponding time strategy is set in the detection equipment, for example, the detection time is set to be 15 minutes, which means that 15 minutes is needed for one detection.

Optionally, in step 310, the detection device specifically executes: after receiving the trigger of the tested user, starting timing; acquiring images of a detection area on the kit within a preset time length to obtain at least one image of the detection; and sending the at least one image detected this time to a user terminal which is in binding relation with the detection equipment through a wireless communication module. After the timing is finished, this detection is completed, and then the image data can be uploaded. In the image acquisition process, the detection device may acquire only one image of the detection area, for example, at a time point when the timing is finished or at a preset time point when the timing is about to be finished, the image is acquired, at this time, the detection line in the detection area gradually develops color, and the acquired image may represent the microorganism detection condition of the user to be detected; of course, the detection device may also collect multiple images of the detection area, for example, periodically or randomly collect images within a timing period to obtain multiple images of the detection, and if the images are collected once every minute, 15 images are obtained in total. Since the development process of the detection lines on the reagent cartridge is developed gradually with time, the detection lines in the partial images collected at the early stage may not be developed yet, and the detection lines in the partial images collected at the later stage are developed gradually. More accurate results can be obtained by collecting a plurality of images, and the fact that the whole detection result is invalid due to the fact that a certain image is an invalid image is avoided.

In the process that the detection equipment starts to detect, the user terminal and the detection equipment can keep smooth communication, the condition that data transmission fails or uploading fails is avoided, and at least one collected image is sent to the user terminal until the detection of the detection equipment is completed. After the image transmission is completed, the detection device may disconnect the communication with the user terminal. In another embodiment, the communication connection between the user terminal and the detection device may also be established after the detection of the detection device is completed, and then it is confirmed that the communication connection between the user terminal and the detection device is established when data transmission is required, so as to complete uploading of image data, thereby avoiding overlong continuous waiting time of the user. Optionally, the detection device communicates with the user terminal through a wireless communication module, for example, the detection device may communicate in a wireless manner such as bluetooth, mobile communication, near field communication NFC, WIFI, and the like.

Optionally, when the detection device sends the at least one detected image, the detection device sends the at least one detected image to the currently bound user terminal. When a new user terminal initiates a binding request to the detection device, the detection device establishes a binding relationship with the new user terminal, and removes the binding relationship with the originally bound user terminal. Of course, the detection device may also establish a binding relationship with a plurality of user terminals, maintain a binding relationship list in the detection device, and send the at least one image to a user terminal with the closest binding time in the case that the binding relationship with the plurality of user terminals exists.

In step 320, the user terminal receives at least one image of the current detection sent by the detection device, and sends the user information of the user to be detected and the at least one image to the server. The user information of the tested user comprises verified identity information obtained after identity authentication and current position information of the tested user. And after receiving the at least one image, the user terminal sends the user information of the detected user, the at least one image obtained by the detection and the unique identifier of the detection to the server. The server receives the following information corresponding to the detection: and then, the server stores the at least one image in an associated manner according to the user information of the tested user and the unique identifier detected by the tested user at this time.

For example, the account of the user "zhang san" includes a unique identifier, image data and a corresponding detection result of each detection, and the user can view a history detection record through the user terminal.

In a specific embodiment, the unique identifier detected this time may be obtained as follows:

1. taking the equipment identification of the detection equipment as the only identification of the detection

After the user terminal scans the identification code on the detection equipment, the binding relationship is established with the detection equipment, and the user terminal simultaneously obtains the equipment identification of the detection equipment, wherein the equipment identification is used for uniquely representing the detection equipment. After the detection device starts a detection flow, the detection device which carries out detection at present sends a device identification of the detection device to the user terminal, after the detection is finished, the user terminal receives at least one image sent by the detection device, and the user terminal takes the obtained device identification as a unique identification of the detection, and sends the unique identification and the at least one image to the server. In this case, the detection device will be used as a single-use device.

2. The reagent mark of the kit is used as the only mark for the detection

Each reagent kit is provided with a reagent mark which is used for uniquely representing the reagent kit, and the reagent marks of different reagent kits are different. The method comprises the steps that a detection device obtains a reagent identification of a kit located in the detection device, the detection device which carries out detection at present sends the reagent identification to a user terminal after the detection device starts a detection flow, the user terminal receives at least one image sent by the detection device after the detection is finished, and the user terminal sends the obtained reagent identification to a server together with the at least one image which is detected at this time as a unique identification of the detection.

In a specific embodiment, a radio frequency identification RFID tag may be set on the reagent kit, the reagent identifier corresponding to the reagent kit is stored in the RFID tag, and after the reagent kit is placed in the detection device, an RFID reader in the detection device may read the RFID tag on the reagent kit, so as to obtain the reagent identifier corresponding to the reagent kit.

3. The detection equipment locally generates the unique identification of the detection

The detection equipment acquires the equipment identification of the detection equipment, and locally generates an identification according to the equipment identification and the detection related information, and the identification is used as the unique identification of the detection. For example, the identifier is generated according to the device identifier and the current detection time, and one generation rule is as follows: and assuming that the equipment identifier is AABBCC and the current detection time is D year, E month, F month, G hour, H second, generating a corresponding identifier which is AABBCCDEFGH. Or, generating an identifier according to the device identifier and the detection times of the detection device, where one generation rule is: if the device identifier is AABBCC and the current detection is the 2 nd detection of the detection device, a corresponding identifier is generated as AABBCC-2. And after the detection equipment starts a detection flow, sending the locally generated unique identifier to the user terminal.

It will be appreciated that the detection device may also generate the identifier locally from the reagent identifier and detection-related information of the kit, in the same manner as described above.

4. The user terminal locally generates the unique identification of the detection

After the detection device starts a detection flow, the current detection device sends a device identifier of the current detection device or a reagent identifier of the kit to the user terminal, and the user terminal locally generates an identifier according to the device identifier and relevant detection information, or locally generates an identifier according to the reagent identifier and relevant detection information, and uses the generated identifier as a unique identifier of the current detection. The generation method is the same as the method 3, and is not described herein.

It can be understood that the user terminal may also generate the unique identifier for the current detection according to the device identifier and the user information of the user to be detected, or the reagent identifier and the user information of the user to be detected.

5. The server generates the unique identification of the detection

The detection equipment receives the trigger of the user, starts detection, and the detection equipment currently performing detection sends identification information to the user terminal, wherein the identification information can be the equipment identification of the detection equipment or the reagent identification of the used kit. And the user terminal receives the identification information sent by the detection equipment, and sends the identification information and the user information of the detected user to the server together so as to request the server for the unique identification detected this time. And after receiving the user information and the identification information, the server generates the unique identification of the current detection according to a certain algorithm rule and transmits the generated unique identification of the current detection to the user terminal. The user terminal obtains the unique identifier corresponding to the detection.

The user terminal associates and uploads the detection data through the unique identifier detected this time, so as to achieve the purpose of associating the data. And sending the own equipment identification to the user terminal once each time the detection equipment performs detection, namely each time the trigger of the user is received, so as to obtain the unique identification of the current detection.

In step 330, the server identifies each image in the at least one detected image by using the image identification model to obtain an identification result of each image, i.e. at least one identification result. The image recognition model in this embodiment is a model based on a neural network, and is obtained by training a pre-constructed target neural network.

In order to obtain the image recognition model, a target neural network is constructed in advance, and the target neural network is trained. Before training, a certain number of training samples are obtained, a training sample set is constructed, each training sample comprises a training image and a label corresponding to the training image, and the label comprises a positive label and a negative label. In a specific embodiment, 300 training images are acquired, including 200 images with negative results and 100 images with positive results. After obtaining a number of training samples, the training samples may be processed as follows:

first, image preprocessing

First, Histogram Equalization (Histogram Equalization) may be performed. The server adjusts the contrast of the image by using the image histogram, and the local contrast is enhanced while the overall contrast is not influenced. This step is mainly to eliminate the error caused by the difference in mean contrast between different images.

Second, Gaussian filtering (Gauss filter) may be performed. The server carries out smoothing processing on the image through Gaussian filtering, and the influence of noise on the result is reduced.

Furthermore, edge detection can be performed. The server extracts the image edge (such as Sobel, Canny and other operators) by using an edge detection operator, and superposes the image edge and the original image to increase the high-frequency characteristics of the image.

Then, image enhancement may be performed. The server performs operations such as rotation, mirroring and scaling on the obtained training images to obtain more training images, so that the trained model has better robustness.

In this embodiment, the target neural network may extract image features using a convolutional neural network, and then perform end-to-end classification based on the extracted features. In a particular embodiment, the image recognition model uses weighted cross-entropy loss as a classification loss for the neural network. One design of the weighted cross-entropy penalty E is as follows:

wherein the content of the first and second substances,

for cross entropy, the true result of sample k in the training sample set is

The prediction result obtained by predicting the sample k through the neural network is

The cross entropy can be used to measure the prediction results

And true results

The degree of difference therebetween. Detailed calculation of cross entropy can be referred to prior art.

As coefficients, coefficients multiplied when calculating for different samples

Different from, onIn the formula, i class is a few class samples (usually positive class samples), and in order to increase the cost of misclassifying the i-th class, the i-th class samples in the training sample set are multiplied by a coefficient larger than 1 in terms of cross entropy, so that the loss E increases if the class of the i-th class samples is misclassified. With the collection of images, the negative samples in the obtained training samples are far larger than the positive samples; the problem of neural network error rate caused by sample imbalance can be effectively solved through the setting of the loss function, and the model precision is effectively improved.

In this embodiment, based on the current microbial detection technology, a corresponding target neural network is constructed, for example, a target neural network is constructed by fusing neural network models of FPN (Feature pyramids), Resnet (residual neural network), and Shufflenet; the FPN is used for ensuring that the network can extract target features with different sizes; resnet, which adds cross-layer connection, can make the network deeper and more likely to extract high-dimensional features; shuffle, channel rearrangement brought by packet convolution, mainly aims to enhance the robustness of the network; the full convolution layer replaces the full connection layer at the top layer of the network and can process images with different resolutions after being zoomed. In the field of image recognition, the related technology of a neural network is mature, a technician can design a target neural network based on the related technology, and the embodiment does not give much description to the target neural network.

In step 330, the server identifies and obtains an identification result of each image by using the image identification model, wherein the identification result includes a microorganism detection result, and the microorganism detection result includes: null results, positive results, negative results. Taking the detection of the new coronavirus as an example, the kit in this embodiment may be an antigen detection kit, an antibody detection kit, or a nucleic acid detection kit. Further, taking the antibody detection kit as an example, the antibody detection kit has three detection lines, which are a C line (control line), a T1 line (IGM antibody detection line), and a T2 line (IGG antibody detection line), respectively, wherein the IGM antibody is a marker of recent infection, and the IGG antibody is a marker of past infection. If the C line is discolored, the detection is valid, if the T1 line is discolored, the IGM is positive, and if the T2 line is discolored, the IGG is positive. Of course, the kit may also be a kit product with one detection line, with two detection lines, or with another number of detection lines.

In one embodiment, the image recognition model analyzes color development information of the detection lines in the image to obtain corresponding microorganism detection results, which includes: (1) ineffective results (i.e., no discoloration of the C-line, possibly due to blood drop failure or diluent addition failure, etc.); (2) positive results, including IGM positive and IGG positive, IGM positive and IGG negative, IGM negative and IGG positive; (3) negative results, including IGM negative and IGG negative. For different microorganism detection scenes, when detection kits of different microorganisms are adopted, the image recognition model can analyze images in different manners, for example, microorganism detection results can be obtained by analyzing whether a specific detection line in a plurality of detection lines changes color or analyzing the color development degree and the color development area of each detection line.

In step 340, after obtaining at least one identification result of the current detection, the server determines the detection result of the current detection of the detected user according to the at least one identification result, and sends the detection result to the user terminal, so that the detected user can obtain the detection result of the current detection through the user terminal. The server can generate a corresponding detection report according to the user information of the detected user and the detection result of the detection, the detection report is sent to the user terminal, and the user terminal displays the detection report to the detected user through the APP, the small program or the micro service. In addition, the server records the detection result of each detection of each detected user, and the information can be sent to a specified medical service platform or an epidemic situation prevention and control center. In a specific embodiment, the server is configured to send the user information of the detected user and the detection result to the epidemic prevention and control center together when the detection result of the current detection of the detected user is positive, and the epidemic prevention and control center may perform further investigation on the detected user according to the user information of the detected user.

On the basis of the above embodiment, the microorganism detection system further includes: and the doctor terminal can be used for judging the image by the doctor terminal. In this embodiment, the doctor is a qualified professional doctor. The doctor needs to check and park after recording, and the doctor terminal is used for receiving and displaying the target image sent by the server, receiving the annotation of the doctor on the target image and sending the annotated target image to the server.

Specifically, the target image sent to the doctor terminal by the server is an image of which the identification result meets the preset condition in the at least one detected image. In step 330, the recognition result of each image recognized by the server includes a microorganism detection result and a confidence level of the microorganism detection result, wherein the confidence level is used for representing the confidence level of the microorganism detection result. One preset condition includes: the confidence of the microorganism detection result of the target image is lower than a threshold value; and/or the microorganism detection result of the target image is positive.

And after the server obtains the identification result of at least one image, selecting an image with a confidence coefficient of a microorganism detection result lower than a threshold value and/or a microorganism detection result positive from the at least one identification result as a target image, and sending the target image to the doctor terminal. A doctor examines and marks a target image through a doctor terminal, then the marked target image is sent back to a server, the server receives the marked target image returned by the doctor terminal, the identification result of the target image in at least one image is updated according to the mark of the doctor on the target image, and then the detection result of the current detection of the detected user is determined according to the updated identification result of the at least one image.

In an example, when at least one identification result in the at least one identification result of the current detection is positive, the detection result of the current detection of the user to be detected is determined to be positive.

Further, the server is further configured to optimize the adopted image recognition model by using the target image and the annotation of the doctor on the target image after receiving the annotated target image returned by the doctor terminal, so as to obtain an optimized image recognition model. And the server takes the optimized image recognition model as an image recognition model used in the next detection. Along with the increase of the marked target images, the precision of the image recognition model is improved, and the obtained recognition result is more accurate.

In actual scene application, the microorganism detection system can meet the requirement of screening of large-scale people and provide assistance for doctors as a tool for primary screening of microorganism infection, wherein a server can identify detection images of all detected users to obtain a primary identification result, the images which cannot be identified and positive images are sent to a doctor terminal, rechecking is carried out by the doctors, the doctors can also carry out spot check on negative images in the images regularly through the doctor terminal, and the detection result is doubly guaranteed by an image identification model and the doctor rechecking, so that the detection efficiency is improved, and the result is accurate.

The microorganism detection system in the embodiment can be used for wireless monitoring requirements of a plurality of different scenes (home, unit and hospital) on microorganism screening results, so that more people with symptoms and people without symptom infection can be found as far as possible. In order to improve the current situation of the current epidemic situation in time, the microorganism detection system can be firstly applied to the screening work of the new coronavirus pneumonia, and can be applied to the repeated production and rework, for example, to a plurality of scenes such as airports, station security inspection, families and the like. Furthermore, the APP client on the user terminal can guide the user in epidemic prevention, eliminate the tension and anxiety emotions of the user, and cut off the way of secondary virus propagation; meanwhile, the system can be associated with relevant diagnosis and treatment and management organizations of communities, regions and countries, so that the data can be shared in time, the system is purposeful, medical resources can be reasonably distributed, and the current disease control situations of unclear crowd infection rate, difficult control of infection ways, difficult collection of epidemic disease data and the like can be comprehensively and effectively solved. Meanwhile, for the user, the risk of cross infection is greatly reduced because the user does not need to go to a hospital for examination.

Optionally, in an early stage of deployment and application of the microbial detection system, due to the lack of enough training images, the problem that the obtained image recognition model is not high enough in recognition accuracy exists, in this case, a server in the system can also directly send at least one image obtained by the detection to a doctor terminal, and the doctor terminal directly performs judgment, so that the detection result of the detection of the detected user at this time is obtained. After enough training images are collected, the image recognition model gradually becomes mature, and can be formally put into the screening of microorganisms.

In an optional embodiment, the disease screening cloud platform is constructed based on cloud computing technology, the disease screening cloud platform is deployed in a server, and a network operating environment of the disease screening cloud platform can adopt a fifth-generation mobile communication 5G technology. The server completes the steps based on the disease screening cloud platform. The disease screening cloud platform effectively sends the detection result of the detected user to the user terminal and the medical institution through the 5G network, and the user terminal can provide data visualization service for the user and display relevant information of current detection.

On the premise of not changing the system architecture, the embodiment can also deploy an image recognition model for screening other diseases, and integrates the disease screening cloud platform into a large platform integrating the screening and monitoring capabilities of multiple diseases by matching with image data acquired by detection equipment, so that the public health epidemic prevention capability of China is improved by continuously supervising common acute and chronic infectious diseases. In the microorganism detection system provided by this embodiment, only the kit placed in the detection device needs to be replaced, and the image recognition model for disease screening is added to the server, so that simple and rapid adaptive upgrade can be performed for different epidemics, and the microorganism detection system can be applied to detection and prevention and control of epidemics such as influenza and african swine fever.

In an optional embodiment, the disease screening cloud platform can also utilize big data sediment of prevention and control of different types of epidemic diseases in the early stage to extract key factors of different types of diseases, and perform key factor matching on propagation and prevention and control of new types of epidemic diseases in the future, so that prevention and control deduction is realized, and reference is provided for planning timely and effective targeted prevention and control measures of epidemic situations of public health departments.

To sum up, the embodiment of the application constructs a low-cost and wide-coverage microbial detection system based on the microbial detection kit, the detection equipment, the user terminal and the server, provides remote disease detection screening service for users, achieves early disease screening and early diagnosis, and provides timely epidemic situation information for the epidemic situation prevention and control center. The system can realize self-detection, self-isolation and automatic uploading of detection results of the crowd to the server at home with lower cost, and simultaneously, suspected cases with positive identification results can be automatically isolated at home as early as possible, so that the propagation path is cut off, a large amount of social resources are saved, the danger brought to detection personnel by short-distance screening and detection is eliminated, and early discovery, early reporting and early isolation are realized.

Based on the same inventive concept, the embodiments of the present application further provide a microorganism detection apparatus, which is configured on a server, as shown in fig. 5, the apparatus includes:

an image receiving module 410, configured to receive at least one image obtained by the current detection sent by the user terminal; the at least one image is obtained by acquiring an image of a detection area on the kit by detection equipment bound with the user terminal and sending the acquired image to the user terminal;

the image recognition module 420 is configured to recognize each image in the at least one image by using an image recognition model to obtain a recognition result of each image;

and the microorganism detection module 430 is configured to determine a detection result of the current detection of the detected user according to the identification result of the at least one image, and send the detection result to the user terminal.

Alternatively, referring to fig. 6, the microorganism detection module 430 includes:

the result screening unit 4301 is configured to send a target image, of the at least one image, of which an identification result meets a preset condition, to the doctor terminal;

a result receiving unit 4302, configured to receive the labeled target image returned by the doctor terminal, and update an identification result of the target image in the at least one image with the label of the doctor on the target image;

and the detection unit 4303 is configured to determine a detection result of the current detection of the detected user according to the updated identification result of the at least one image.

Optionally, the recognition result of each image recognized by the image recognition module includes a microorganism detection result and a confidence of the microorganism detection result; the preset conditions include: the confidence of the microorganism detection result of the target image is lower than a threshold value; and/or the microorganism detection result of the target image is positive.

Optionally, the apparatus further comprises: a model optimization module to: and after the result receiving unit receives the labeled target image returned by the doctor terminal, optimizing the image recognition model by using the target image and the label of the doctor on the target image to obtain an optimized image recognition model, and taking the optimized image recognition model as the image recognition model used by the image recognition module in the next detection.

The microorganism detection device provided above corresponds to the server in the previous system embodiment, and can execute each step executed by the server, and the basic principles and the generated technical effects of the two are the same.

In the embodiments provided in the present application, it should be understood that the disclosed apparatus may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of each unit is only one logical function division, and there may be other division ways in actual implementation. Furthermore, the functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

It should be noted that the functions, if implemented in the form of software functional modules and sold or used as independent products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

In this document, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions.

The above description is only an example of the present application and is not intended to limit the scope of the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A microorganism detection system is characterized by comprising detection equipment, a user terminal and a server;

the detection equipment is used for acquiring at least one image of a detection area on the kit and sending the at least one image to the bound user terminal; wherein the kit is located within the detection device;

the user terminal is used for receiving the at least one image and sending the at least one image to the server;

the server is used for identifying each image in the at least one image by using the image identification model to obtain the identification result of each image, determining the detection result of the current detection of the detected user according to the identification result of the at least one image, and sending the detection result to the user terminal.

2. The system of claim 1, further comprising a physician terminal; the doctor terminal is used for receiving the target image sent by the server, receiving the annotation of the doctor on the target image and sending the annotated target image to the server;

the server is used for determining the detection result of the current detection of the detected user according to the identification result of the at least one image, and comprises the following steps: sending a target image of which the recognition result meets a preset condition in the at least one image to the doctor terminal; receiving the labeled target image returned by the doctor terminal, and updating the identification result of the target image in the at least one image by the label of the doctor to the target image; and determining the detection result of the current detection of the detected user according to the updated identification result of the at least one image.

3. The system according to claim 2, wherein the recognition result of each image recognized by the server comprises a microorganism detection result and a confidence level of the microorganism detection result; the preset conditions include: the confidence of the microorganism detection result of the target image is lower than a threshold value; and/or the microorganism detection result of the target image is positive.

4. The system of claim 2, wherein the server is further configured to: and after receiving the labeled target image returned by the doctor terminal, optimizing the image recognition model by using the target image and the label of the doctor on the target image to obtain an optimized image recognition model, and taking the optimized image recognition model as the image recognition model used in the next detection.

5. The system of claim 1, wherein the user terminal is further configured to: before the detection equipment collects at least one image of a detection area on the kit, receiving identity information input by a detected user, and carrying out living body face collection on the detected user to obtain a face image of the detected user; calling a public security interface to verify the identity information and the face image of the detected user; and after the verification is passed, completing the identity authentication of the detected user to obtain verified identity information.

6. The system according to claim 5, wherein the user terminal sends the at least one image to the server, and simultaneously sends the pre-stored user information of the user to be tested and the unique identifier of the current test to the server; wherein, the user information of the tested user comprises the verified identity information and the current position information of the tested user;

and the server is also used for storing the at least one image in a correlation manner according to the user information of the detected user and the unique identification detected this time.

7. The system according to claim 6, wherein the unique identifier of the current detection is an equipment identifier of the detection equipment or a reagent identifier of the kit, the equipment identifier of the detection equipment is used for uniquely characterizing the detection equipment, and the reagent identifier of the kit is used for uniquely characterizing the kit.

8. The system of claim 6, wherein the detection device is further configured to: before at least one image of a detection area on a kit is collected, sending identification information to the user terminal, wherein the identification information is an equipment identification of the detection equipment or a reagent identification of the kit, the equipment identification of the detection equipment is used for uniquely representing the detection equipment, and the reagent identification of the kit is used for uniquely representing the kit;

the user terminal is also used for sending the identification information and the user information of the detected user to the server together so as to request the server for the unique identification detected this time;

the server is used for generating the unique identification detected this time according to the identification information and the user information, and sending the unique identification detected this time to the user terminal.

9. The system of claim 1, wherein the user terminal is further configured to: before the detection equipment collects at least one image of a detection area on a kit, a communication address of the detection equipment is obtained by scanning an identification code, and a binding request is sent to the detection equipment through the communication address;

the detection device is further configured to establish a binding relationship with the user terminal based on the binding request.

10. The system of claim 9, wherein the detection device is configured to capture at least one image of a detection area on a reagent cartridge and transmit the at least one image to a bound user terminal, and comprises: after receiving the trigger of the tested user, starting timing; acquiring images of a detection area on the kit within a preset time length to obtain at least one image of the detection; and sending the at least one image to a user terminal which is in binding relation with the detection equipment through a wireless communication module.

11. The system of claim 6, wherein the server is further configured to: and when the detection result of the current detection of the detected user is positive, the user information of the detected user and the detection result are sent to an epidemic situation prevention and control center together.

12. A microorganism detection apparatus, configured to be installed in a server, the apparatus comprising:

the image receiving module is used for receiving at least one image which is sent by the user terminal and obtained by the detection; the at least one image is obtained by acquiring an image of a detection area on the kit by detection equipment bound with the user terminal and sending the acquired image to the user terminal;

the image identification module is used for identifying each image in the at least one image by using the image identification model to obtain an identification result of each image;

and the microorganism detection module is used for determining the detection result of the current detection of the detected user according to the identification result of the at least one image and sending the detection result to the user terminal.

13. The apparatus of claim 12, wherein the microorganism detection module comprises:

the result screening unit is used for sending the target image of which the identification result meets the preset condition in the at least one image to the doctor terminal;

the result receiving unit is used for receiving the labeled target image returned by the doctor terminal and updating the identification result of the target image in the at least one image by the label of the doctor to the target image;

and the detection unit is used for determining the detection result of the current detection of the detected user according to the updated identification result of the at least one image.

14. The apparatus according to claim 13, wherein the recognition result of each image recognized by the image recognition module comprises a microorganism detection result and a confidence of the microorganism detection result; the preset conditions include: the confidence of the microorganism detection result of the target image is lower than a threshold value; and/or the microorganism detection result of the target image is positive.

15. The apparatus of claim 13, further comprising:

and the model optimization module is used for optimizing the image recognition model by using the target image and the label of the doctor on the target image after the result receiving unit receives the labeled target image returned by the doctor terminal to obtain an optimized image recognition model, and the optimized image recognition model is used as the image recognition model used by the image recognition module in the next detection.