CN117351250A

CN117351250A - Method and device for automatically screening medical image pictures

Info

Publication number: CN117351250A
Application number: CN202210738271.1A
Authority: CN
Inventors: 李明宙; 徐辉雄; 郭乐杭; 伯小皖
Original assignee: Beijing Yiyin Artificial Intelligence Technology Co ltd
Current assignee: Beijing Yiyin Artificial Intelligence Technology Co ltd
Priority date: 2022-06-27
Filing date: 2022-06-27
Publication date: 2024-01-05

Abstract

The disclosure discloses a method and a device for automatically screening medical image pictures, relates to the field of medical image equipment, and overcomes the defect that the screened medical image pictures are not optimal for diagnosis due to human factors in traditional operation, overcomes the defect that medical staff leak diagnosis rate due to human factors caused by overload work, and reduces tedious and repeated medical image diagnosis analysis work of the medical staff. The method comprises the following steps: acquiring a medical image picture set, wherein each medical image picture in the medical image picture set comprises focus image information; inputting a medical image picture set into a classification model, and determining a target category and category score of a focus in each medical image picture in the medical image picture set based on the classification model, wherein the category score is used for representing the probability that the focus belongs to the target category; and screening the target medical image pictures in the medical image picture set according to the category scores of each medical image picture in the medical image picture set.

Description

Method and device for automatically screening medical image pictures

Technical Field

The disclosure relates to the field of medical imaging equipment, in particular to a method and a device for automatically screening medical image pictures.

Background

Along with the acceleration of the aging trend of society and the rising incidence of serious diseases, the clinical requirements on the quality and coverage fields of medical examination are higher. The medical imaging equipment becomes the first-choice disease examination means of clinical medicine by the characteristics of noninvasive property, wide applicability, rapid diagnosis, high accuracy and the like. And due to factors such as uneven distribution of medical resources, shortage of related talents and the like, the medical institutions need intelligent expansion of equipment so as to improve diagnosis efficiency and accuracy.

At present, the screening of medical image pictures is completely manually screened, so that the time consumption is long, the subjectivity is strong, the missed diagnosis rate is high, and a great deal of manpower and material resources of image doctors are consumed to do repeated work.

Disclosure of Invention

In order to solve the above problems in the prior art, the present disclosure provides a method and apparatus for automatically screening medical image pictures.

A first aspect of the present disclosure provides a method for automatically screening medical image pictures, applied to medical imaging equipment, the method comprising: acquiring a medical image picture set, wherein each medical image picture in the medical image picture set comprises focus image information; inputting the medical image picture set into a classification model, and determining a target category and category score of a focus in each medical image picture in the medical image picture set based on the classification model, wherein the category score is used for representing the probability that the focus belongs to the target category; and screening target medical image pictures in the medical image picture set according to the category scores of each medical image picture in the medical image picture set.

In an embodiment, the determining the target category and the category score of the focus in each of the medical image pictures in the medical image picture set based on the classification model includes: based on the classification model, focus reminding information is displayed in the medical image picture, and the focus reminding information is used for representing the position and the range of a focus in the medical image picture; and determining the target category and category score of the focus indicated by the focus reminding information.

In another embodiment, the determining the target class and class score of the lesion in each of the medical image pictures in the medical image picture set comprises: and scoring the target category of the focus based on the S-shaped growth curve, and generating the category score of the focus.

In yet another embodiment, one or more focus reminding information and target categories and category scores of focuses indicated by the one or more focus reminding information are displayed in the medical image picture.

In yet another embodiment, the classification model is trained in the following manner: acquiring a training set, wherein the training set comprises a plurality of medical image pictures marked with focus categories and focus attributes; and training a focus detection model by using the training set to obtain the classification model, wherein the focus detection model is used for detecting whether focus information is contained in the medical image picture.

A second aspect of the present disclosure provides an apparatus for automatically screening medical image pictures, applied to medical imaging devices, the apparatus comprising: the acquisition module is used for acquiring a medical image picture set, and each medical image picture in the medical image picture set comprises focus image information; the input module is used for inputting the medical image picture set into a classification model, and determining the target category and category score of the focus in each medical image picture in the medical image picture set based on the classification model, wherein the category score is used for representing the probability that the focus belongs to the target category; and the screening module is used for screening the target medical image pictures in the medical image picture set according to the category scores of each medical image picture in the medical image picture set.

In an embodiment, the determining module is specifically configured to display focus reminding information in the medical image picture based on the classification model, where the focus reminding information is used to characterize a position and a range of a focus in the medical image picture; and determining the target category and category score of the focus indicated by the focus reminding information.

In another embodiment, the determining module is specifically further configured to score a target category of the focus based on an S-type growth curve, and generate a category score of the focus.

A third aspect of the present disclosure provides an apparatus for automatically screening medical image pictures, comprising: a memory for storing instructions; a processor for invoking instructions stored in a memory to perform a method as described in the first aspect and any embodiment thereof.

A fourth aspect of the present disclosure provides a computer-readable storage medium comprising computer program instructions which, when read by a computer, perform the method of the first aspect and any of its embodiments described above.

The technical scheme provided by the embodiment of the disclosure at least can comprise the following beneficial effects: the medical image picture set containing focus image information is input into the classification model, the target category and category score of focus contained in each medical image picture in the medical image picture set can be determined, the higher the analogy score is, the larger the probability that the focus belongs to the target category is indicated and the more suitable for diagnosis, therefore, the target medical image picture is screened from the medical image picture set according to the category score of each medical image picture, the defect that the screened medical image picture is not the most suitable medical image picture for diagnosis due to human factors in the traditional operation is avoided, the missing diagnosis rate of the human factors caused by overload work of medical staff is compensated, and the fussy and repeated medical image diagnosis and analysis work of the medical staff is lightened.

Drawings

The above, as well as additional purposes, features, and advantages of embodiments of the present invention will become apparent in the following detailed written description and claims upon reference to the accompanying drawings. Several embodiments of the present invention are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which:

FIG. 1 is a schematic diagram of a system provided by an embodiment of the present disclosure;

FIG. 2 is a flowchart of a method for automatically screening medical image pictures according to an embodiment of the present disclosure;

FIG. 3 is a flow chart of a method for training a classification model according to an embodiment of the present disclosure;

fig. 4 is a schematic diagram of an apparatus for automatically screening medical image pictures according to an embodiment of the disclosure;

fig. 5 is a schematic structural diagram of a medical imaging device according to an embodiment of the present disclosure;

fig. 6 is a block diagram of an apparatus for automatically screening medical image pictures according to an embodiment of the present disclosure.

In the drawings, the same or corresponding reference numerals indicate the same or corresponding parts.

Detailed Description

The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are presented merely to enable those skilled in the art to better understand and practice the invention and are not intended to limit the scope of the invention in any way.

It should be noted that, although the terms "first", "second", etc. are used herein to describe various modules, steps, data, etc. of the embodiments of the present invention, the terms "first", "second", etc. are merely used to distinguish between the various modules, steps, data, etc. and do not denote a particular order or importance. Indeed, the expressions "first", "second", etc. may be used entirely interchangeably.

Based on the above problems, the disclosure provides a method for automatically screening medical image pictures, which avoids the defect that the screened medical image pictures are not optimal for diagnosis due to human factors in the traditional operation, makes up for the missing rate of the human factors caused by overload work of medical staff, and reduces the tedious and repeated medical image diagnosis analysis work of the medical staff.

FIG. 1 illustrates an exemplary system architecture to which embodiments of the present disclosure may be applied.

As shown in fig. 1, the system architecture may include terminal devices 11, 12, a network 13, and a server 14. The network 13 is a medium used to provide a communication link between the terminal devices 11, 12 and the server 14. The network 13 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.

The user 15 may interact with the server 14 via the network 13 using the terminal devices 11, 12 to receive or send messages or the like. The terminal devices 11, 12 may be various medical imaging devices including, but not limited to, electronic computed tomography (Computed Tomography, CT), magnetic resonance imaging (Magnetic Resonance Imaging, MRI), computed radiography (ComputedRadiography, CR) or digital radiography, medical ultrasound diagnostic apparatus, etc. imaging devices. The server 14 may be a server providing various services. The server can store, analyze and the like the received data and feed back the processing result to the terminal equipment.

It should be noted that, the method for automatically screening medical image pictures provided in the embodiments of the present application may be executed by the processor in the terminal device 11, 12, or may be executed by the server 14, where the terminal device 11, 12 interacts with the server 14 through communication to obtain the screening result. The specific application hardware scenario of the method for screening medical image pictures is not strictly limited in the disclosure.

The models referred to hereinafter may be provided in the terminal devices 11, 12 or in the server 14. In some embodiments, the model may be trained in the server 14, and the trained model may be stored in the server 14 for use in screening medical image pictures.

It should be understood that the number of terminal devices, networks and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

Fig. 2 is a flowchart illustrating an automatic screening of medical image pictures according to an exemplary embodiment, as shown in fig. 2, including the following steps.

In step S1, a medical image picture set is acquired, and each medical image picture in the medical image picture set includes focus image information.

The medical image photo set may be a medical image such as an electronic computed tomography (Computed Tomography, CT), a magnetic resonance imaging (Magnetic Resonance Imaging, MRI), a computed radiography (ComputedRadiography, CR) or a digital radiography (Digital radiography, DR), a medical ultrasonic diagnostic apparatus, and the like, which is not specifically limited in this disclosure. The medical image picture set can be directly extracted from corresponding instruments such as CT, MRI, medical ultrasonic diagnostic equipment and the like, and can also be obtained from a database of a hospital. In addition, the content of the medical image itself can be adjusted according to the actual application scene requirement, for example, the medical image can be a transmission image in a brain examination scene of neurosurgery or a transmission image in a lung examination scene of thoracic surgery.

In step S2, the medical image picture set is input to a classification model, and a target class and class score of the focus in each medical image picture in the medical image picture set are determined based on the classification model.

Wherein the category score is used to characterize the probability that the lesion belongs to the target category.

It will be appreciated that the higher the class score, the higher the probability that the lesion belongs to the target class, and the more suitable the medical image picture is for diagnosis.

Alternatively, there may be multiple categories of lesions, such as nodules, calcifications, bumps, etc., in a single medical image, and the target category and category score for each lesion may be determined based on the classification model.

It should be understood that the lesion categories corresponding to each location are not exactly the same, and exemplary thyroid lesion categories include nodules, tumors, etc., and breast lesion categories include calcifications, masses, tumors, etc.

In step S3, the target medical image picture is screened from the medical image picture set according to the category score of each medical image picture in the medical image picture set.

Optionally, sorting according to the category scores of the medical image pictures from high to low, selecting the medical image picture with the highest category score as the target medical image picture, or selecting the N medical image pictures with the highest category scores as the target medical image picture according to the score sequence.

It should be noted that, if the medical image picture input to the classification model does not include the lesion image information, the classification model outputs that the target class of the lesion is none, and the class score is 0.

According to the embodiment of the disclosure, the medical image picture set containing focus image information is input into the classification model, so that the target category and category score of focus contained in each medical image picture in the medical image picture set can be determined, the higher the analogy score is, the greater the probability that the focus belongs to the target category is, and the more suitable for diagnosis is indicated, therefore, the target medical image picture is screened from the medical image picture set according to the category score of each medical image picture, the defect that the screened medical image picture is not the most suitable medical image picture for diagnosis due to human factors in the traditional operation is avoided, the missing diagnosis rate of human factors caused by overload work of medical staff is made up, and the fussy and repeated medical image diagnosis analysis work of the medical staff is lightened.

In some embodiments, focus alert information is displayed in a medical image picture based on the classification model, and a target class and class score of a focus indicated by the focus alert information are determined.

The focus reminding information is used for representing the position and the range of a focus in the medical image picture. Optionally, the focus reminding information may include a first reminding information and a second reminding information, where the first reminding information is used for representing a focus and a general position and a general range of the focus in the medical image picture, and the second reminding information is used for representing a specific position and a specific range of the focus in the medical image picture.

The first reminding information can be represented by regular figures such as squares or rectangles, and the second reminding information can be represented by irregular polygons.

Further, after determining the specific position and specific range of the focus in the medical image picture, determining the target category of the focus based on the classification model, and scoring the target category of the focus based on an S-shaped growth curve (Sigmoid function), so as to generate the category score of the focus.

Optionally, when a medical image picture contains a plurality of focuses, the target category of each focus is scored based on the S-shaped growth curve, and the category score of each focus is generated.

Wherein the Sigmoid function is a nonlinear function of neurons. Is widely applied to the neural network.

The neural network is learned based on a set of samples that include inputs and outputs (denoted herein as desired outputs), with how many components of the inputs and outputs there are as many input and output neurons corresponding thereto. Initially the weights (Weight) and thresholds (Threshold) of the neural network are arbitrarily given, and learning is to gradually adjust the weights and thresholds so that the actual and desired outputs of the network coincide.

In some embodiments, one or more lesion reminder information and a target category and category score of the lesion indicated by the one or more lesion reminder information are displayed in the medical image picture.

Optionally, a score threshold is preset, and only category scores greater than the score threshold are displayed in the medical image picture.

Optionally, the target category and the category score of the focus are displayed in the medical image picture by means of an annotation frame. Specifically, the frame can be made on the outer side of the position of the focus, more specifically, a rectangular frame can be made, the category and the category score of the focus are reflected through the color of the rectangular frame, and different colors correspond to different categories. Or, the brightness of the focus area is enhanced to show the position of the focus, the category score is displayed through the change of specific brightness conditions, and the target category of the focus is shown in a text prompt mode. A step of

Further, when a plurality of focus reminding information is displayed in the medical image picture, when the target medical image picture is screened, target image information is screened according to target categories of all focuses. For example, three existing medical image pictures are provided, wherein the first medical image picture comprises a focus, the target category of the focus is a nodule, and the category score is 85; the second medical image picture comprises two focuses, the target categories are a nodule and a tumor respectively, the category score of the nodule is 90, and the category score of the tumor is 60; the third medical image picture comprises two focuses, the target categories are nodules and tumors, the category score of the nodules is 20, and the category score of the tumors is 70. If only the medical image picture with the highest category score is used as the target medical image picture, the second medical image picture and the third medical image picture are both target medical image pictures, the second medical image picture is the target medical image picture of the nodule category, and the third medical image picture is the target medical image picture of the tumor category.

The following describes the training process of the classification model, which, as shown in fig. 3, includes the steps of:

in step S21, a training set is acquired.

The training set comprises a plurality of medical image pictures marked with focus categories and focus attributes;

in step S22, the focus detection model is trained by the training set to obtain a classification model.

The focus detection model is used for detecting whether focus information is contained in the medical image picture.

Optionally, in the training process, the focus position included in the medical image picture is determined first, the coordinates of the focus position are obtained, and the focus type corresponding to the medical image picture is corresponding to the coordinates of the focus position and the focus attribute.

Optionally, the focus detection model may adopt a convolutional neural network architecture, a fully connected neural network architecture, etc., and specific types and architectures thereof may be adjusted according to requirements of actual application scenarios, and the specific types and architectures of the focus detection model are not strictly limited in the present application.

Taking a convolutional neural network model as an example, the classification model in the embodiment of the disclosure is based on a convolutional neural network, where the convolutional neural network has a weight sharing characteristic, and the weight sharing refers to a convolutional kernel, and the same features of different positions of image data can be extracted through the operation of one convolutional kernel, in other words, the same targets of different positions in one image data, and the features of the same targets are basically the same. It will be appreciated that only a portion of the features can be obtained using one convolution kernel, and that features of the picture can be extracted by setting a multi-kernel convolution, each of which learns a different feature. In the image classification, the function of the convolution layer is to extract and analyze low-level features as high-level features, wherein the low-level features are basic features such as textures and edges, and the high-level features such as the shapes of faces and objects can better represent the properties of samples, and the process is the layering property of the convolution neural network.

Further, after the focus detection model is trained to obtain the classification model, a test set can be set to test the classification model, and if the error of the test result is not in the allowable range, the focus detection model is retrained until the error of the test result is in the allowable range.

Optionally, after determining the category of the lesion in the medical image picture by using the classification model, the classification model may be retrained by using the medical image picture of the target category of the determined lesion.

According to the method and the device for obtaining the target category and the category score of the focus, the target category and the category score of the focus can be obtained through training the classification model, the defect that the screened medical image picture is not the most suitable medical image picture for diagnosis due to human factors in the traditional operation is avoided, the missing rate of the human factors caused by overload of medical staff is overcome, and the tedious and repeated medical image diagnosis and analysis work of the medical staff is reduced.

Based on the same conception, the embodiment of the disclosure also provides a device for automatically screening medical image pictures.

It can be understood that, in order to achieve the above functions, the apparatus for screening medical image pictures provided in the embodiments of the present disclosure includes a hardware structure and/or a software module that perform each function. The disclosed embodiments may be implemented in hardware or a combination of hardware and computer software, in combination with the various example elements and algorithm steps disclosed in the embodiments of the disclosure. Whether a function is implemented as hardware or computer software driven hardware depends upon the particular application and design constraints imposed on the solution. Those skilled in the art may implement the described functionality using different approaches for each particular application, but such implementation is not to be considered as beyond the scope of the embodiments of the present disclosure.

Fig. 4 is a block diagram of an apparatus for automatically screening medical image pictures according to an exemplary embodiment. Referring to fig. 4, the apparatus includes an acquisition module 401, an input module 402, and a screening module 403.

An acquisition module 401, configured to acquire a medical image picture set, where each medical image picture in the medical image picture set includes focus image information;

the input module 402 inputs the medical image picture set to the classification model, and determines a target class and class score of the focus in each medical image picture in the medical image picture set based on the classification model, wherein the class score is used for representing the probability that the focus belongs to the target class;

the screening module 403 is configured to screen the target medical image picture in the medical image picture set according to the category score of each medical image picture in the medical image picture set.

In an embodiment, the obtaining module 401 is specifically configured to display focus reminding information in the medical image picture based on the classification model, where the focus reminding information is used to characterize a position and a range of a focus in the medical image picture; and determining the target category and category score of the focus indicated by the focus reminding information.

In another embodiment, the obtaining module 401 is specifically further configured to score a target category of the lesion based on the S-type growth curve, and generate a category score of the lesion.

In yet another embodiment, one or more lesion reminding information and a target category and category score of the lesion indicated by the one or more lesion reminding information are displayed in the medical image picture.

In yet another embodiment, the classification model is trained in the following manner: acquiring a training set, wherein the training set comprises a plurality of medical image pictures marked with focus categories and focus attributes; training focus detection models by using a training set to obtain classification models, wherein the focus detection models are used for detecting whether focus information is contained in medical image pictures.

Fig. 5 is a block diagram illustrating an apparatus 100 for automatically screening medical image pictures according to an exemplary embodiment. For example, the apparatus 100 may be a medical imaging device.

The apparatus 100 may specifically include: processor 101, communication device 102, display 103, power supply 104, memory 105, audio circuitry 106, multimedia device 107, sensor 108, and peripheral interface 109. Those skilled in the art will appreciate that the hardware configuration shown in fig. 5 is not limiting of the medical imaging device, and that the medical imaging device may include more or fewer components than shown, or may combine certain components, or a different arrangement of components.

The processor 101 is a control center of the medical imaging device, connects various parts of the cellular phone using various interfaces and lines, performs various functions of the medical imaging device and processes data by running or executing an application program (hereinafter, may be abbreviated as App) stored in the memory 105, and calling data stored in the memory 105. In some embodiments, the processor 101 may include one or more processing units.

The communication device 102 may be used for receiving and transmitting wireless signals during a messaging or conversation. Specifically, the communication device 102 may receive downlink data of the base station and then process the downlink data for the processor 101; in addition, data relating to uplink is transmitted to the base station. Typically, the radio frequency circuitry includes, but is not limited to, an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, and the like. In addition, the communication apparatus 102 may also communicate with other devices through wireless communication. The wireless communication may use any communication standard or protocol including, but not limited to, global system for mobile communications, general packet radio service, code division multiple access, wideband code division multiple access, long term evolution, email, short message service, and the like.

The display 103 may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensor may sense not only the boundary of a touch or slide action, but also the duration and pressure associated with the touch or slide operation.

The power supply 104 is logically connected to the processor 101 through a power management chip, so as to realize functions of managing charging, discharging, power consumption management, and the like.

The memory 105 is used to store application programs and data, and the processor 101 executes various functions and data processing of the mobile phone by executing the application programs and data stored in the memory 105. The memory 105 mainly includes a storage program area and a storage data area, wherein the storage program area can store an operating system, an application program (such as a sound playing function, an image playing function, etc.) required for at least one function; the storage data area may store data (such as audio data, phonebooks, etc.) created according to the use of the handset. In addition, memory 105 may include high-speed random access memory, and may also include non-volatile memory, such as magnetic disk storage devices, flash memory devices, or other volatile solid-state storage devices, and the like. The memory 105 may store various operating systems, for example, iOS operating systems developed by apple corporation, android operating systems developed by google corporation, and the like.

The audio circuit 106 is configured to output and/or input audio signals.

The multimedia device 107 may include a WIFI device, a bluetooth component, etc. The Wi-Fi device is used for providing network access conforming to Wi-Fi related standard protocols for the mobile phone, the mobile phone can be accessed to the Wi-Fi access point through the Wi-Fi device, so that a user can be helped to send and receive e-mails, browse webpages, access streaming media and the like, and wireless broadband Internet access is provided for the user. In other embodiments, the Wi-Fi device may also act as a Wi-Fi wireless access point, and may provide Wi-Fi network access for other terminals.

The medical imaging device may also include at least one sensor 108, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor, wherein the ambient light sensor may adjust the brightness of the display 103 according to the brightness of ambient light, and the proximity sensor may turn off the power of the display when the medical imaging device moves to the ear. As one of the motion sensors, the accelerometer sensor can detect the acceleration in all directions (generally three axes), and can detect the gravity and direction when stationary, and can be used for applications of recognizing the gesture of a mobile phone (such as horizontal and vertical screen switching, related games, magnetometer gesture calibration), vibration recognition related functions (such as pedometer and knocking), and the like; other sensors such as gyroscopes, barometers, hygrometers, thermometers, infrared sensors, etc. that may also be configured with the handset are not described in detail herein.

A peripheral interface 109 for providing various interfaces to external input/output devices (e.g., keyboard, mouse, external display, external memory, user identification module card, etc.). For example, the communication function is realized by connecting a Universal Serial Bus (USB) interface with a mouse or a display, connecting a metal contact on a card slot of a subscriber identity module with a subscriber identity module card (SIM) card provided by a telecom operator, and connecting the interface with other terminals through an interface of a Wi-Fi device, an interface of a Near Field Communication (NFC) device, an interface of a Bluetooth module and the like. Peripheral interface 109 may be used to couple the external input/output peripherals described above to processor 101 and memory 105.

Although not shown in fig. 5, the medical imaging device may further include a camera, a flash, a micro-projection device, a Near Field Communication (NFC) device, etc., which will not be described herein.

Fig. 6 is a block diagram illustrating an apparatus 200 for automatically screening medical image pictures according to an exemplary embodiment. For example, the apparatus 200 may be provided as a server. Referring to fig. 6, the apparatus 200 includes: memory 201, processor 202, input/Output (I/O) interface 203. Wherein the memory 201 is used for storing instructions. The processor 202 is configured to invoke the instructions stored in the memory 201 to execute the method for automatically screening medical image pictures according to the embodiment of the present disclosure. Wherein the processor 202 is coupled to the memory 201, the I/O interface 203, respectively, such as via a bus system and/or other form of connection mechanism (not shown). The memory 201 may be used to store programs and data, including programs of the footprint track display method involved in the embodiments of the present disclosure, and the processor 202 performs various functional applications of the apparatus 200 and data processing by running the programs stored in the memory 201.

The processor 202 in the disclosed embodiments may be implemented in at least one hardware form of a digital signal processor (Digital Signal Processing, DSP), field-programmable gate array (Field-Programmable Gate Array, FPGA), programmable logic array (Programmable Logic Array, PLA), the processor 202 may be one or a combination of several of a central processing unit (Central Processing Unit, CPU) or other form of processing unit having data processing and/or instruction execution capabilities.

The memory 201 in embodiments of the present disclosure may include one or more computer program products that may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. The volatile memory may include, for example, random access memory (Random Access Memory, RAM) and/or cache memory (cache), etc. The nonvolatile Memory may include, for example, a Read-Only Memory (ROM), a Flash Memory (Flash Memory), a Hard Disk (HDD), a Solid State Drive (SSD), or the like.

In the embodiment of the present disclosure, the I/O interface 203 may be used to receive input instructions (e.g., numeric or character information, and generate key signal inputs related to user settings and function control of the apparatus 200, etc.), and may also output various information (e.g., images or sounds, etc.) to the outside. The I/O interface 203 in embodiments of the present disclosure may include one or more of a physical keyboard, function keys (e.g., volume control keys, switch keys, etc.), a mouse, joystick, trackball, microphone, speaker, touch panel, etc.

The embodiments of the present disclosure also provide a computer-readable storage medium, where the computer-readable storage medium includes computer-executable instructions that, when executed on a computer, cause the computer to perform the method for automatically screening medical image pictures provided in the above embodiments.

It will be appreciated that although operations are depicted in the drawings in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. In certain circumstances, multitasking and parallel processing may be advantageous.

The methods and apparatus of embodiments of the present invention may be implemented using standard programming techniques with various method steps being performed using rule-based logic or other logic. It should also be noted that the words "apparatus" and "module" as used herein and in the claims are intended to include implementations using one or more lines of software code and/or hardware implementations and/or equipment for receiving inputs.

Any of the steps, operations, or procedures described herein may be performed or implemented using one or more hardware or software modules alone or in combination with other devices. In one embodiment, the software modules are implemented using a computer program product comprising a computer readable medium containing computer program code capable of being executed by a computer processor for performing any or all of the described steps, operations, or programs.

The foregoing description of the implementations of the invention has been presented for purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise form disclosed, and modifications and variations are possible in light of the above teachings or may be acquired from practice of the invention. The embodiments were chosen and described in order to explain the principles of the invention and its practical application to enable one skilled in the art to utilize the invention in various embodiments and with various modifications as are suited to the particular use contemplated.

Claims

1. A method for automatically screening medical image pictures, which is applied to medical image equipment, the method comprising:

acquiring a medical image picture set, wherein each medical image picture in the medical image picture set comprises focus image information;

inputting the medical image picture set into a classification model, and determining a target category and category score of a focus in each medical image picture in the medical image picture set based on the classification model, wherein the category score is used for representing the probability that the focus belongs to the target category;

and screening target medical image pictures in the medical image picture set according to the category scores of each medical image picture in the medical image picture set.

2. The method of claim 1, wherein determining a target category and category score for a lesion in each of the medical image pictures in the set of medical image pictures based on the classification model comprises:

based on the classification model, focus reminding information is displayed in the medical image picture, and the focus reminding information is used for representing the position and the range of a focus in the medical image picture;

and determining the target category and category score of the focus indicated by the focus reminding information.

3. The method of claim 1 or 2, wherein the determining the target category and category score for a lesion in each of the set of medical image pictures is:

and scoring the target category of the focus based on the S-shaped growth curve, and generating the category score of the focus.

4. The method of claim 3, wherein the medical image picture displays one or more lesion reminder information and a target category and category score of the lesion indicated by the one or more lesion reminder information.

5. The method of claim 1, wherein the classification model is trained by:

acquiring a training set, wherein the training set comprises a plurality of medical image pictures marked with focus categories and focus attributes;

and training a focus detection model by using the training set to obtain the classification model, wherein the focus detection model is used for detecting whether focus information is contained in the medical image picture.

6. An apparatus for automatically screening medical image pictures, which is applied to medical image equipment, the apparatus comprising:

the acquisition module is used for acquiring a medical image picture set, and each medical image picture in the medical image picture set comprises focus image information;

the input module is used for inputting the medical image picture set into a classification model, and determining the target category and category score of the focus in each medical image picture in the medical image picture set based on the classification model, wherein the category score is used for representing the probability that the focus belongs to the target category;

and the screening module is used for screening the target medical image pictures in the medical image picture set according to the category scores of each medical image picture in the medical image picture set.

7. The apparatus of claim 6, wherein the device comprises a plurality of sensors,

the acquisition module is specifically configured to display focus reminding information in the medical image picture based on the classification model, where the focus reminding information is used to characterize a position and a range of a focus in the medical image picture; and determining the target category and category score of the focus indicated by the focus reminding information.

8. The apparatus according to claim 6 or 7, wherein,

the acquisition module is specifically further configured to score a target category of the focus based on the S-type growth curve, and generate a category score of the focus.

9. The apparatus of claim 8, wherein the medical image picture displays one or more lesion reminder information and a target category and category score for the lesion indicated by the one or more lesion reminder information.

10. The apparatus of claim 6, wherein the classification model is trained by: acquiring a training set, wherein the training set comprises a plurality of medical image pictures marked with focus categories and focus attributes; and training a focus detection model by using the training set to obtain the classification model, wherein the focus detection model is used for detecting whether focus information is contained in the medical image picture.

11. An apparatus for automatically screening medical image pictures, said apparatus comprising: a memory for storing instructions;

a processor for invoking the memory stored instructions to perform the method of any of claims 1 to 5.

12. A computer readable storage medium, characterized in that the computer program readable storage medium comprises computer program instructions which, when read by a computer, perform the method of any of claims 1 to 5.