WO2019170060A1 - Image authentication-based medical image labeling method and device, user terminal, and computer readable storage medium - Google Patents

Abstract

An image authentication-based medical image labeling method, an image authentication-based medical image labeling device and a user terminal. The method comprises: obtaining a labeled medical image and an unlabeled medical image in a medical image library, and displaying the labeled medical image and the unlabeled medical image on a login interface as authentication conditions (S100); and acquiring, by a user at the login interface, labeling information of the unlabeled medical image according to target labeling content of the labeled medical image on the unlabeled medical image (S200). According to the method, when a user performs login authentication, not only authentication can be realized, but also labeling work for a large number of medical images can be realized, so that a library with a large quantity of labeled medical images can be obtained; a labeling time period is short, and labeling efficiency is high; labeling costs are greatly reduced, and a labeling accuracy rate is improved; and great convenience is brought to professional doctors required to label a large number of clinical medical images.

Description

Medical image annotation method, device, user terminal and computer readable storage medium based on image authentication

Cross-reference to related applications

The present application claims priority to Chinese Patent Application No. 2018101805400, entitled "Image Image-Based Medical Image Labeling Method, Apparatus and User Terminal", filed on March 5, 2018, the entire disclosure of which is hereby incorporated by reference. The content is incorporated herein by reference.

Technical field

The present application relates to the field of medical imaging technology, and more particularly to a medical image annotation method, apparatus, user terminal and computer readable storage medium based on image authentication.

Background technique

Medical imaging refers to the technology and process of obtaining internal tissue images in a non-invasive manner for the human body or a part of the human body for medical or medical research.

In the medical field, it is necessary to image and measure a large number of medical image images, which serves as the basis for medical decision-making and medical treatment. For example, in the fetal image, the skull edge needs to be accurately labeled to calculate the head diameter, which is used to determine whether it is well developed, and whether it is delivered or cesarean section. For example, in the treatment of radiation tumors, in order to accurately determine the tumor and surrounding normal organs. Position, the doctor needs to use the mouse to manually circle all relevant organs (tumors) in layers in the 3D image.

Image annotation plays a key role in precision medicine, and all image-based measurements and decisions need to be based on accurate image annotation. The existing method for labeling medical image images mainly requires a professional doctor to work on a medical image of a patient for several hours, which is long and inefficient, and the medical image is labeled by artificial intelligence technology. The hardware requirements are high and the accuracy is significantly lower than the manual labeling of the professional doctor. In short, the existing methods for labeling medical image images have long time, low efficiency, high labeling cost, and low accuracy through automatic identification of artificial intelligence, which brings about the demand for labeling clinically large number of medical image images by professional doctors. Huge inconvenience.

Summary of the invention

In view of this, the present application provides a medical image annotation method, apparatus, user terminal, and computer readable storage medium based on image authentication to solve the deficiencies of the prior art.

To solve the above problem, the present application provides a medical image annotation method based on image authentication, including:

Obtaining the labeled medical image image and the unlabeled medical image image in the medical image library, and displaying the labeled medical image image and the unlabeled medical image image as an authentication condition on the login interface;

The annotation information of the unlabeled medical image image is obtained by the user on the login interface according to the labeled content of the unlabeled medical image image according to the labeled medical image image.

Preferably, the labeled medical image image comprises a labeled medical image image for teaching and a medical image image of known labeled data;

Before the user obtains the annotation information of the unlabeled medical image image by using the target labeling content of the unlabeled medical image image according to the labeled medical image image by the user in the login interface, the method further includes:

Obtaining the verification annotation content input by the user on the medical data image of the known annotation data, and the target annotation content of the unlabeled medical image image, wherein the verification annotation content is used by the user according to the teaching Verification of the labeled content of the medical image image input by the labeled medical image;

Determining whether the similarity between the verified annotation content input by the user and the known annotation data according to the labeled medical image image of the teaching to the known annotation data reaches a preset similarity;

If yes, the target labeling content is authenticated and stored, so that the user can obtain the unlabeled medical image by using the target labeling content of the unlabeled medical image image according to the labeled medical image image by the user at the login interface. The annotation information of the image.

Preferably, before acquiring the verification annotation content input by the user to the medical data image of the known annotation data, and the labeling content of the target of the unlabeled medical image image, the method further includes:

Determining the labeled medical image image for teaching and the medical image image of the known annotation data, wherein the teaching content is displayed in the labeled medical image image, and the known annotation data medical image image is not displayed Know the data of the annotation;

Marking the known annotation data medical image image that does not display the mark according to the teaching medical target image, and marking the unlabeled medical image image, thereby obtaining the known Marking content of the data medical image image and target labeling content for the unlabeled medical image image.

Preferably, after the “determining whether the similarity of the verified annotation content of the known annotation data medical image image input reaches a preset similarity”, the method further includes:

If not, the authentication failure is prompted, and returning to “acquire the labeled medical image image and the unlabeled medical image image in the medical image library, and display the labeled medical image image and the unlabeled medical image image as the authentication condition. The steps in the login screen.

Preferably, the “acquiring the labeled medical image image and the unlabeled medical image image in the medical image library, and displaying the labeled medical image image and the unlabeled medical image image as the authentication condition on the login interface” includes :

In the medical image library, the plane positioning information of the center of gravity corresponding to the preset to-be-marked area in any labeled medical image is obtained according to the preset to-be-marked area;

Performing a screenshot on the marked medical image and the unlabeled medical image according to the plane positioning information, to obtain the labeled medical image image and the unlabeled medical image image;

The labeled medical image image and the unlabeled medical image image are displayed as authentication conditions on the registration interface.

Preferably, the plane positioning information comprises any one of the following: a spatial coordinate in the three-dimensional image, a transverse plane horizontal to a certain vertical axis in the three-dimensional image, and a certain range of a certain center of gravity of the two-dimensional plane.

Preferably, the “using the user to obtain the annotation information of the unlabeled medical image image according to the target annotation content of the unlabeled medical image image according to the labeled medical image image by the user on the login interface” includes:

Obtaining the similarity of each target label content corresponding to the unlabeled medical image image;

The target labeling content with the highest similarity is selected as the annotation information of the unlabeled medical image image.

Preferably, after the user uses the target labeling content of the unlabeled medical image image to obtain the labeling information of the unlabeled medical image image according to the labeled medical image image by the user in the login interface, :

Obtaining the annotation information of the plurality of different planes corresponding to the preset to-be-marked area;

And combining the annotation information of the plurality of different planes corresponding to the preset to-be-marked area to obtain three-dimensional annotation data corresponding to the preset to-be-marked area.

Preferably, before the “acquiring the labeled medical image image and the unlabeled medical image image in the medical image library”, the method further includes:

Receiving the verification request of the user, so as to obtain the labeled medical image image and the unlabeled medical image image in the medical image library according to the user request.

Preferably, the medical image library comprises: a first medical image sub-library and a second medical image sub-library;

Obtaining the labeled medical image image and the unlabeled medical image image in the medical image library includes:

When the user is authenticated, the labeled medical image image is obtained from the first medical image sub-library, and the unlabeled medical image image is obtained in the second medical image sub-library as the authentication for the user to log in to the website or other platform. Conditions are displayed on the login screen.

Preferably, the target annotation content may include at least one of the following: a planar annotation area of the planar image, coordinate information, relative position data, and range data, and the annotation information of the unlabeled medical image image may include at least one of the following: three-dimensional Data, spatial location information, spatial volume information, spatial relative location data.

In addition, in order to solve the above problem, the present application further provides a medical image annotation device based on image authentication, comprising: an acquisition module and an annotation module;

The acquiring module is configured to acquire the labeled medical image image and the unlabeled medical image image in the medical image library, and display the labeled medical image image and the unlabeled medical image image as an authentication condition on the login interface;

The labeling module is configured to acquire the labeling information of the unlabeled medical image image by using the target labeling content of the unlabeled medical image image according to the labeled medical image image by the user on the login interface.

In addition, in order to solve the above problems, the present application further provides a user terminal, including a memory and a processor, the memory configured to be based on an image authentication medical image annotation program, the processor running the image authentication based medical image annotation The program is to cause the user terminal to perform a medical image annotation method based on image authentication as described above.

Preferably, the user terminal is a mobile terminal device having a display function, and the user terminal comprises: a processor, a network interface, a user interface, a memory, and a communication bus.

In addition, in order to solve the above problems, the present application further provides a computer readable storage medium on which a medical image annotation program based on image authentication is stored, and the medical image annotation program based on image authentication is processed. The image recognition method based on image authentication as described above is implemented when the device is executed.

The invention provides a medical image annotation method, device, user terminal and computer readable storage medium based on image authentication. The method provided by the present application displays the labeled medical image image and the unlabeled medical image image on the login interface, and uses the user to mark the content of the unlabeled medical image image, thereby obtaining the annotation information of the unlabeled medical image image. Thereby, the online authentication method is implemented to realize large-scale collaboration of all login users, and then the medical image images are marked. Enable users to perform login authentication, not only to perform online authentication, but also to achieve a large number of medical image annotation work, thereby obtaining a large number of labeled medical image libraries, marking time period, high efficiency, through each login user The certification process carries out the labeling of medical images, which greatly reduces the cost of labeling, and improves the labeling accuracy rate by manual labeling, which brings great convenience for professional doctors to label the clinical medical image.

DRAWINGS

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings to be used in the embodiments will be briefly described below. It should be understood that the following drawings show only certain embodiments of the present application, and therefore It should be seen as a limitation on the scope, and those skilled in the art can obtain other related drawings according to these drawings without any creative work.

1 is a schematic structural diagram of a hardware operating environment involved in an embodiment of a method for marking medical images based on image authentication according to an embodiment of the present invention;

2 is a schematic flow chart of a first embodiment of a medical image annotation method based on image authentication according to the present application;

3 is a schematic flow chart of a second embodiment of a medical image annotation method based on image authentication according to the present application;

4 is a schematic flow chart of a third embodiment of a medical image annotation method based on image authentication according to the present application;

FIG. 5 is a schematic flowchart diagram of a fourth embodiment of a medical image annotation method based on image authentication according to the present application; FIG.

6 is a schematic flowchart of a fifth embodiment of a medical image annotation method based on image authentication according to the present application;

FIG. 7 is a schematic diagram of functional modules of the medical image annotation device of the present application.

The implementation, functional features and advantages of the present application will be further described with reference to the accompanying drawings.

Detailed ways

The embodiments of the present application are described in detail below, wherein the same or similar reference numerals indicate the same or similar elements or elements having the same or similar functions.

Moreover, the terms "first" and "second" are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, features defining "first" and "second" may include one or more of the features either explicitly or implicitly. In the description of the present application, the meaning of "a plurality" is two or more unless specifically and specifically defined otherwise.

In the present application, the terms "installation", "connected", "connected", "fixed" and the like shall be understood broadly, and may be either a fixed connection or a detachable connection, unless otherwise explicitly stated and defined. , or integrated; can be mechanical connection, or can be electrical connection; can be directly connected, or can be indirectly connected through an intermediate medium, can be the internal communication of two elements or the interaction of two elements. For those skilled in the art, the specific meanings of the above terms in the present application can be understood on a case-by-case basis.

It is understood that the specific embodiments described herein are merely illustrative of the application and are not intended to be limiting.

As shown in FIG. 1 , FIG. 1 is a schematic structural diagram of a hardware operating environment of a user terminal according to an embodiment of the present application.

The user terminal in the embodiment of the present application may be a PC, or may be a mobile terminal device having a display function, such as a smart phone, a tablet computer, an e-book reader, an MP3 player, an MP4 player, a portable computer, or the like.

As shown in FIG. 1, the user terminal may include a processor 1001, such as a CPU (Central Processing Unit), a network interface 1004, a user interface 1003, a memory 1005, and a communication bus 1002. Among them, the communication bus 1002 is configured to implement connection communication between these components. The user interface 1003 can include a display screen, an input unit such as a keyboard, a remote controller, and the optional user interface 1003 can also include a standard wired interface, a wireless interface. The network interface 1004 can optionally include a standard wired interface, a wireless interface (such as a WI-FI interface). The memory 1005 may be a high speed RAM (Random Access Memory) memory or a stable memory such as a disk memory. The memory 1005 can also optionally be a storage device independent of the aforementioned processor 1001.

In this embodiment, the processor 1001 is configured to perform the processing steps of the following modules:

Obtaining a module, acquiring the labeled medical image image and the unlabeled medical image image in the medical image library, and displaying the labeled medical image image and the unlabeled medical image image as an authentication condition on the login interface;

The labeling module acquires the labeling information of the unlabeled medical image image by using the target labeling content of the unlabeled medical image image according to the labeled medical image image by the user on the login interface.

Optionally, the user terminal may further include a camera, an RF (Radio Frequency) circuit, a sensor, an audio circuit, a WiFi module, and the like. In addition, the mobile terminal can also be configured with other sensors such as a gyroscope, a barometer, a hygrometer, a thermometer, an infrared sensor, and the like, and details are not described herein again.

It will be understood by those skilled in the art that the user terminal shown in FIG. 1 does not constitute a limitation to the terminal, and may include more or less components than those illustrated, or combine some components, or different component arrangements.

As shown in FIG. 1, a memory 1005 as a computer readable storage medium may include an operating system, a data interface control program, a network connection program, and a medical image tagging program based on image authentication.

The invention provides a medical image annotation method, device and user terminal based on image authentication. The method enables the user to perform not only on-line authentication but also the labeling work for a large number of medical images while performing login authentication, thereby obtaining a large number of labeled medical image libraries, marking time periods and high efficiency. The labeling of medical images is carried out by the authentication process of each logged-in user, which greatly reduces the cost of labeling, and the manual labeling improves the accuracy of labeling, which brings great convenience for professional doctors to label the clinical medical image images. .

Example 1:

Referring to FIG. 2, a first embodiment of the present application provides a medical image annotation method based on image authentication, including:

Step S100: Obtain an labeled medical image image and an unlabeled medical image image in the medical image library, and display the labeled medical image image and the unlabeled medical image image as an authentication condition on the login interface.

In the above, the medical image library is an image library containing a large number of medical images of the same series or the same lesion, and may include, for example, angiography, computed tomography, mammography, positron emission tomography, magnetic resonance imaging, and ultrasound examination. As a result of the data, the image may be a two-dimensional planar image or a three-dimensional stereoscopic visible image.

The medical image library may include an annotated medical image sub-library (ie, a first medical image sub-library) and an unlabeled medical image sub-library (ie, a second medical image sub-library), wherein the labeled medical image sub-library includes Medical image images have been marked, and unlabeled medical image sub-libraries contain unlabeled medical image images.

In the above, when the user is authenticated, the labeled medical image image is obtained from the medical image sub-library, and the unlabeled medical image image is obtained in the unlabeled medical image sub-library as the authentication condition for the user to log in to the website or other platform. , displayed on the login screen.

In the above, the login interface may be a login interface in the website, or may be a login interface of different clients or APPs.

Step S200: The user uses the target labeling content of the unlabeled medical image image according to the labeled medical image image on the login interface to obtain the annotation information of the unlabeled medical image image.

In the above, the medical image image has been marked as a labeling prompt for the user's learning and whether the labeling of the user is accurate, thereby realizing the labeling of the unlabeled medical image image, obtaining the target labeling content, and then obtaining the content by the target labeling content. Label the annotation information of the medical image.

The target labeling content may include, but is not limited to, a plane labeling area of the plane image, coordinate information, relative position data, range data, and the like. The labeling information of the unlabeled medical image image may include, but is not limited to, three-dimensional data, spatial position information, Information on spatial volume information, spatial relative position data, and the like.

The method provided in this embodiment displays the labeled medical image and the unlabeled medical image on the login interface, and uses the user to mark the content of the unlabeled medical image, thereby obtaining the annotation information of the unlabeled medical image. The online authentication method is used to realize the large-scale collaboration of all login users, and then the medical image images are marked. Enable users to perform login authentication, not only to perform online authentication, but also to achieve a large number of medical image annotation work, thereby obtaining a large number of labeled medical image libraries, marking time period, high efficiency, through each login user The certification process carries out the labeling of medical images, which greatly reduces the cost of labeling, and improves the labeling accuracy rate by manual labeling, which brings great convenience for professional doctors to label the clinical medical image.

Example 2:

Referring to FIG. 3, a second embodiment of the present application provides a medical image annotation method based on image authentication. Based on the first embodiment shown in FIG. 2, the labeled medical image includes a labeled medical image and a medical image. It is known to label data medical image images;

The step S200, before the user obtains the annotation information of the unlabeled medical image image by using the target labeling content of the unlabeled medical image image according to the labeled medical image image by the user, further includes:

Step S300, acquiring verification annotation content input by the user to the medical data image of the known annotation data, and labeling content of the target of the unlabeled medical image image;

In the above, the massive image authentication technology adopted in this embodiment is the reCAPTCHA system. It is understood that CMU (Carnegie Mellon University) designed a powerful system called reCAPTCHA to let their computers go to humans for help. The specific method is: the text scan image that is not recognized by the OCR software is transmitted to the major websites of the world to replace the original verification code picture; after the users of those websites correctly recognize the text, the answer will be transmitted back to the CMU.

In the above, the labeled medical image image includes a labeled medical image image for teaching and a medical image image of known labeled data. That is, in the medical image library, two known medical image images of known labeled data are found, one of which serves as a labeled medical image image for teaching and the other as a medical image image of known labeled data. For example, in the medical image library, the hippocampus structure image includes 100 medical image images of known annotation data, wherein each hippocampus structure is marked. There are also 9900 three-dimensional brain images in which the hippocampus is not marked.

Step S400, determining whether the similarity between the verification annotation content input by the user and the known annotation data input by the user according to the labeled medical image image for teaching reaches a preset similarity;

In the above, when the client or the client authenticates, it is to circle different medical images. On the client side, we adopt a fast two-dimensional image interactive segmentation algorithm to assist the user to quickly circle his/her circle. The area of the painting. It does not require the user to drag the mouse to draw. Analogous to the reCAPTCHA system, the entire process can be completed in seconds.

Optionally, before acquiring the verification annotation content input by the user to the medical data image of the known annotation data, and the labeling content of the target of the unlabeled medical image image, the method further includes the following steps:

Determining the labeled medical image image for teaching and the medical image image of the known annotation data, wherein the teaching content is displayed in the labeled medical image image, and the known annotation data medical image image is not displayed Knowing the annotation data; and marking the known annotation data medical image image that does not display the mark according to the teaching medical image image, and marking the unlabeled medical image image, thereby obtaining the user's Marking content of the medical data image of the known labeled data and a content of the target for the unlabeled medical image image.

In the above process, in the process of performing authentication, that is, the process in which the user needs to log in, the system provides two labeled medical image images, and prompts the user to perform circle painting positioning, one of which is used as a medical medical image for teaching, and the other as Know the data of the medical image. The teaching uses the labeled medical image to display the marked content, while the other known labeled medical image does not display the known labeled data; the user does not display the marked according to the labeled medical image. Knowing the labeling data medical image image for marking, and also marking the unlabeled medical image image, so that the system obtains the user's marked content of the medical image image of the known labeled data and the target labeling content for the unlabeled medical image image, and then It is known that the similarity between the marked content of the labeled data medical image image and the known annotation data reaches a preset similarity.

Step S500, if yes, authenticating and storing the target annotation content, so as to obtain the target by using the user to mark the target of the unlabeled medical image image according to the labeled medical image image by the user on the login interface. Label the annotation information of the medical image.

In the above, if the similarity reaches the preset similarity, the user is judged to be a manual operation, a non-program operation, and the third sub-image of the user circle is the target annotation content of the unlabeled medical image image.

Step S600, if not, prompting the authentication failure, and returning to “acquire the labeled medical image image and the unlabeled medical image image in the medical image library, and use the labeled medical image image and the unlabeled medical image image as The authentication conditions are displayed on the login screen.

In the above, if the similarity does not meet the preset similarity, the authentication failure occurs, and the system will re-send the medical image image to the user for re-authentication until the authentication succeeds or a certain number of failures is reached.

Example 3:

Referring to FIG. 4, a third embodiment of the present application provides a medical image annotation method based on image authentication. Based on the first embodiment shown in FIG. 2, the step S100 “acquires the labeled medical image image in the medical image library. And displaying the medical image image and displaying the labeled medical image image and the unlabeled medical image image as authentication conditions on the login interface" includes:

Step S110, in the medical image library, acquiring, according to the preset to-be-marked area, plane positioning information of a center of gravity corresponding to the area to be marked in any of the labeled medical images;

In the above, in the medical image library, the plane positioning information of the center of gravity needs to be acquired according to the preset area to be marked, and the plane positioning information may be a space coordinate in the three-dimensional image, or may be a vertical direction in the three-dimensional image. A transverse plane of the axis, or a certain range of a certain center of gravity of a two-dimensional plane. In this embodiment, the medical image is a three-dimensional medical image. For example, the area to be marked is a medical image of the hippocampus, and the center of gravity of the hippocampus is automatically selected, and then a random plane passing through the center of gravity is generated.

Step S120, performing a screenshot on the marked medical image and the unmarked medical image according to the plane positioning information, to obtain the labeled medical image image and the unlabeled medical image image;

In step S130, the labeled medical image image and the unlabeled medical image image are displayed as an authentication condition on the login interface.

In the above, the marked and unmarked medical images are screenshotd according to the plane positioning information, thereby obtaining the labeled medical image image and the unlabeled medical image image. For example, when the client sends a verification request, randomly select two images from the 100 marked (verified by the imaging doctor) images, calling them A and B. One of the 9900 unlabeled images is randomly selected, called C. In A and B, because the hippocampus has been marked, we take the center of gravity of the hippocampus. Then, a random plane passing through this center of gravity is generated, and three three-dimensional images A, B, and C are intercepted by this plane. Obtain three two-dimensional images a, b, c. In a and b, the two-dimensional section of the hippocampus can be obtained by intercepting the three-dimensional annotation of A and B, and then the labeled medical image and the unlabeled medical image are used as authentication conditions to authenticate the user.

Example 4:

Referring to FIG. 5, a fourth embodiment of the present application provides a medical image annotation method based on image authentication. Based on the second embodiment shown in FIG. 3, the step S200 “uses the user according to the login interface according to the Labeling the medical image image with the target labeling content of the unlabeled medical image image to obtain the labeling information of the unlabeled medical image image includes:

Step S210, acquiring the similarity of each target label content corresponding to the unlabeled medical image image;

Step S220, selecting the target labeling content with the highest similarity as the labeling information of the unlabeled medical image image.

For different users, the system can send cross sections of the same or similar positions. By authenticating different users, multiple target annotation contents of the same or similar positions are obtained, and then the weight of the content is marked according to the similarity as the selection target, that is, the more accurate the user is not labeled with the medical image image, he The decision-making power of the circled painting result without the medical image image is greater than those of the user who does not have the high accuracy of the unlabeled medical image image (the user with low accuracy may reach the standard of the preset similarity).

Example 5:

Referring to FIG. 6 , a fifth embodiment of the present application provides a medical image annotation method based on image authentication. Based on the first embodiment shown in FIG. 2 , the step S200 “uses the user according to the login interface according to the After the labeling the medical image image to the target labeling of the unlabeled medical image image to obtain the labeling information of the unlabeled medical image image, the method further includes:

Step S700: Obtain the labeling information of multiple different planes corresponding to the preset to-be-marked area;

Step S800, synthesizing the annotation information of the plurality of different planes corresponding to the preset to-be-marked area, and obtaining three-dimensional annotation data corresponding to the preset to-be-marked area.

In the above, when it is determined that the input is manually operated, and the user saves the circle of the medical image image, the preset area of the three-dimensional image, such as the hippocampus area, is marked on the specific plane. For different users, the system can issue different sections of the area to be marked. The boundaries of different users in different sections of the circle, such as the boundary of the hippocampus, will be merged into a three-dimensional annotation in the background of the system.

Before the “acquiring the labeled medical image image and the unlabeled medical image image in the medical image library”, the method further includes:

Step S900, receiving the verification request of the user, so as to obtain the labeled medical image image and the unlabeled medical image image in the medical image library according to the user request.

In the above, when the user performs login for different websites or clients, an authentication request is made through the website, thereby performing an operation of further acquiring the labeled medical image and the unlabeled medical image in the medical image library according to the verification request.

In addition, the present application further provides a medical image annotation device based on image authentication, comprising: an acquisition module 10 and an annotation module 20;

The acquiring module 10 is configured to acquire the labeled medical image image and the unlabeled medical image image in the medical image library, and display the labeled medical image image and the unlabeled medical image image as an authentication condition on the login interface. ;

The labeling module 20 is configured to acquire the labeling information of the unlabeled medical image image by using the target labeling content of the unlabeled medical image image according to the labeled medical image image by the user on the login interface.

In addition, the present application further provides a user terminal, including a memory and a processor, the memory configured to be based on an image authentication medical image annotation program, the processor running the image authentication based medical image annotation program to enable the The user terminal performs a medical image annotation method based on image authentication as described above.

In addition, the present application further provides a computer readable storage medium, where the computer image readable storage medium stores a medical image annotation program based on image authentication, and the image authentication based medical image annotation program is implemented by a processor, such as The medical image annotation method based on image authentication described above.

It is to be understood that the term "comprises", "comprising", or any other variants thereof, is intended to encompass a non-exclusive inclusion, such that a process, method, article, or It also includes other elements that are not explicitly listed, or elements that are inherent to such a process, method, item, or system. An element defined by the phrase "comprising a ..." does not exclude the presence of additional equivalent elements in a process, method, article, or system that includes the element, without further limitation.

The serial numbers of the embodiments of the present application are merely for the description, and do not represent the advantages and disadvantages of the embodiments.

Through the description of the above embodiments, those skilled in the art can clearly understand that the foregoing embodiment method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be through hardware, but in many cases, the former is better. Implementation. Based on such understanding, the technical solution of the present application, which is essential or contributes to the prior art, may be embodied in the form of a software product stored in a storage medium (such as ROM/RAM as described above). , a disk, an optical disk, including a number of instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the methods described in the various embodiments of the present application.

The above is only a preferred embodiment of the present application, and is not intended to limit the scope of the patent application, and the equivalent structure or equivalent process transformations made by the specification and the drawings of the present application, or directly or indirectly applied to other related technical fields. The same is included in the scope of patent protection of this application.

Industrial applicability

The image authentication method, the device, the user terminal, and the computer readable storage medium are provided by the embodiment of the present application. The method provided in this embodiment enables the user to perform not only online authentication but also online authentication. For the labeling work of a large number of medical images, a large number of labeled medical image libraries are obtained, which mark time periods and high efficiency, and the medical images are marked by the authentication process of each login user, which greatly reduces the labeling cost, and Manual labeling improves the accuracy of labeling, which brings great convenience for professional doctors to label the clinical medical image.

Claims (15)

A medical image annotation method based on image authentication, comprising: Obtaining the labeled medical image image and the unlabeled medical image image in the medical image library, and displaying the labeled medical image image and the unlabeled medical image image as an authentication condition on the login interface; The annotation information of the unlabeled medical image image is obtained by the user on the login interface according to the labeled content of the unlabeled medical image image according to the labeled medical image image. The medical image annotation method based on image authentication according to claim 1, wherein the labeled medical image image comprises a labeled medical image image for teaching and a medical image image with known annotation data; Before the user obtains the annotation information of the unlabeled medical image image by using the target labeling content of the unlabeled medical image image according to the labeled medical image image by the user in the login interface, the method further includes: Obtaining the verification annotation content input by the user on the medical data image of the known annotation data, and the target annotation content of the unlabeled medical image image, wherein the verification annotation content is used by the user according to the teaching Verification of the labeled content of the medical image image input by the labeled medical image; Determining whether the similarity between the verified annotation content input by the user and the known annotation data according to the labeled medical image image of the teaching to the known annotation data reaches a preset similarity; If yes, the target labeling content is authenticated and stored, so that the user can obtain the unlabeled medical image by using the target labeling content of the unlabeled medical image image according to the labeled medical image image by the user at the login interface. The annotation information of the image. The medical image annotation method based on image authentication according to claim 2, wherein the verification annotation content input by the user to the medical data image of the known annotation data is acquired, and the unlabeled medical image image is Before the target is labeled, the method further includes: Determining the labeled medical image image for teaching and the medical image image of the known annotation data, wherein the teaching content is displayed in the labeled medical image image, and the known annotation data medical image image is not displayed Know the data of the annotation; Marking the known annotation data medical image image that does not display the mark according to the teaching medical target image, and marking the unlabeled medical image image, thereby obtaining the known Marking content of the data medical image image and target labeling content for the unlabeled medical image image. The medical image annotation method based on image authentication according to claim 2 or 3, wherein the determining whether the similarity of the verification annotation content input by the medical data image of the known annotation data reaches a preset similarity After that, it also includes: If not, the authentication failure is prompted, and returning to “acquire the labeled medical image image and the unlabeled medical image image in the medical image library, and display the labeled medical image image and the unlabeled medical image image as the authentication condition. The steps in the login screen. The medical image annotation method based on image authentication according to any one of claims 1 to 4, wherein the “acquiring the labeled medical image image and the unlabeled medical image image in the medical image library, and The labeled medical image image and the unlabeled medical image image are displayed as authentication conditions on the login interface" including: In the medical image library, the plane positioning information of the center of gravity corresponding to the preset to-be-marked area in any labeled medical image is obtained according to the preset to-be-marked area; Performing a screenshot on the marked medical image and the unlabeled medical image according to the plane positioning information, to obtain the labeled medical image image and the unlabeled medical image image; The labeled medical image image and the unlabeled medical image image are displayed on the login interface as authentication conditions. The medical image annotation method based on image authentication according to claim 5, wherein the plane positioning information comprises any one of the following: a spatial coordinate in the three-dimensional image, and a horizontal horizontal horizontal direction in the three-dimensional image. A certain range of the center of gravity of a plane, a two-dimensional plane. The medical image annotation method based on image authentication according to any one of claims 2 to 4, wherein the user uses the unlabeled medical image according to the labeled medical image image on the login interface by the user. The target labeling content of the image to obtain the labeling information of the unlabeled medical image image includes: Obtaining the similarity of each target label content corresponding to the unlabeled medical image image; The target labeling content with the highest similarity is selected as the annotation information of the unlabeled medical image image. The medical image annotation method based on image authentication according to claim 7, wherein the user uses the user to mark the target of the unlabeled medical image image according to the labeled medical image image on the login interface. After obtaining the annotation information of the unlabeled medical image image, the method further includes: Obtaining the annotation information of the plurality of different planes corresponding to the preset to-be-marked area; And combining the annotation information of the plurality of different planes corresponding to the preset to-be-marked area to obtain three-dimensional annotation data corresponding to the preset to-be-marked area. The medical image annotation method based on image authentication according to any one of claims 1 to 8, wherein before the “acquiring the labeled medical image image and the unlabeled medical image image in the medical image library”, the method further includes : Receiving the verification request of the user, so as to obtain the labeled medical image image and the unlabeled medical image image in the medical image library according to the user request. The medical image annotation method based on image authentication according to any one of claims 1 to 9, wherein the medical image library comprises: a first medical image sub-bank and a second medical image sub-library; Obtaining the labeled medical image image and the unlabeled medical image image in the medical image library includes: When the user is authenticated, the labeled medical image image is obtained from the first medical image sub-library, and the unlabeled medical image image is obtained in the second medical image sub-library as the authentication for the user to log in to the website or other platform. Conditions are displayed on the login screen. The medical image annotation method based on image authentication according to any one of claims 1 to 10, wherein the target annotation content may include at least one of the following: a planar annotation area of a planar image, coordinate information, and relative position data. And the range data, the label information of the unlabeled medical image image may include at least one of the following: three-dimensional data, spatial position information, spatial volume information, and spatial relative position data. A medical image annotation device based on image authentication, comprising: an acquisition module and an annotation module; The acquiring module is configured to acquire the labeled medical image image and the unlabeled medical image image in the medical image library, and display the labeled medical image image and the unlabeled medical image image as an authentication condition on the login interface; The labeling module is configured to acquire the labeling information of the unlabeled medical image image by using the target labeling content of the unlabeled medical image image according to the labeled medical image image by the user on the login interface. A user terminal, comprising: a memory and a processor configured to be based on an image authentication medical image annotation program, the processor running the image authentication based medical image annotation program to cause the user terminal A medical image annotation method based on image authentication according to any one of claims 1-11. The user terminal according to claim 13, wherein the user terminal is a mobile terminal device having a display function, and the user terminal comprises: a processor, a network interface, a user interface, a memory, and a communication bus. A computer readable storage medium, wherein the computer readable storage medium stores a medical image tagging program based on image authentication, and the image authentication based medical image tagging program is executed by a processor to implement the claims The medical image annotation method based on image authentication according to any one of 1-11.

Images