CN112150451A

CN112150451A - Symmetry information detection method and device, computer equipment and storage medium

Info

Publication number: CN112150451A
Application number: CN202011049562.7A
Authority: CN
Inventors: 吴叶芬
Original assignee: Shanghai United Imaging Healthcare Co Ltd
Current assignee: Shanghai United Imaging Healthcare Co Ltd
Priority date: 2020-09-29
Filing date: 2020-09-29
Publication date: 2020-12-29

Abstract

The application relates to a symmetry information detection method, a device, a computer device and a storage medium, wherein medical image data of a target part is acquired and input into a preset network model to obtain a thermodynamic diagram capable of reflecting the symmetry of the medical image of the target part or the target part, and then the thermodynamic diagram is fitted to obtain the symmetry information of the medical image of the target part or the target part. The method is used for fitting according to the actual symmetry distribution condition of the target part or the medical image of the target part, and can accurately and efficiently determine the symmetry information of the target part or the medical image of the target part.

Description

Symmetry information detection method and device, computer equipment and storage medium

Technical Field

The present application relates to the field of medical technology, and in particular, to a method and an apparatus for detecting symmetric information, a computer device, and a storage medium.

Background

In the medical image scanning process, when a target part, such as a head, is scanned, due to the fact that the scanned target part is not correctly positioned, the scanned image is inclined, so that the corresponding obtained image of the sagittal plane and the frontal plane cannot well display the anatomical structure of the target part, and therefore, the symmetry of the target part needs to be determined before image interpretation or symmetry analysis.

In the conventional technology, the symmetry of the image is adjusted by, but not limited to, manual adjustment by a doctor or analysis adjustment by an algorithm. The manual adjustment by a doctor can seriously waste the reading time of the doctor, and when the analysis and adjustment are carried out by algorithms, each algorithm has certain limitation, for example, a plane equation is optimized by calculating the symmetry correlation coefficient of an original image and a flip image, but the adaptability to the data with pathological changes at a target part is not particularly good; or searching line by line to find the points on the statistical curve which is satisfied by the change of the pixel values of each line, and finally randomly sampling the points to fit a plane, but the method is only suitable for the image which satisfies the statistical curve.

Therefore, a method for efficiently and accurately determining the symmetry plane or the symmetry line of the scanned part image is lacking in the prior art.

Disclosure of Invention

In view of the above, it is desirable to provide a method, an apparatus, a computer device, and a storage medium for detecting symmetry information, which can efficiently and accurately determine the symmetry information of a scanned area image.

In a first aspect, an embodiment of the present application provides a method for detecting symmetric information, where the method includes:

acquiring medical image data of a target part;

inputting medical image data into a preset network model to obtain a thermodynamic diagram of the target part or the symmetric information of the medical image of the target part; the thermodynamic diagram reflects the target part or the symmetry distribution condition of the medical image of the target part;

and fitting the thermodynamic diagram to obtain the symmetry information of the target part or the medical image of the target part.

In one embodiment, the acquiring medical image data of the target region includes:

acquiring an original medical image of a target part;

and preprocessing the original medical image to obtain medical image data of the target part.

In one embodiment, the preprocessing includes one or more of adjusting the resolution of the original medical image, normalizing the gray value of each pixel point in the original medical image, and capturing the region of the target portion from the original medical image.

In one embodiment, before the inputting the medical image data into the preset network model, the method further includes:

acquiring training sample data; the training sample data comprises thermodynamic diagrams corresponding to the medical image data of each part and the medical image data of each part;

and training the initial network according to the training sample data to obtain a network model after the training is finished.

In one embodiment, the obtaining training sample data includes:

acquiring sample medical images of various parts, and performing incremental processing on each sample medical image to obtain an incrementally processed sample medical image;

acquiring a sample thermodynamic diagram corresponding to each sample medical image subjected to incremental processing;

and determining the sample medical image after each incremental processing and the corresponding sample thermodynamic diagram as training sample data.

In one embodiment, the performing incremental processing on each sample medical image includes:

performing incremental processing on each sample medical image in a preset incremental mode; the incremental mode at least comprises one of data clipping, data rotation, Gaussian noise increasing and data translation.

In one embodiment, the medical image data comprises two-dimensional medical image data or three-dimensional volume data.

In a second aspect, an embodiment of the present application provides a symmetry information detection apparatus, including:

an acquisition module for acquiring medical image data of a target part;

the processing module is used for inputting medical image data into a preset network model to obtain a target part or a thermodynamic diagram of the symmetry of the medical image of the target part; the thermodynamic diagram reflects the symmetry distribution of the target part or the medical image of the target part;

and the fitting module is used for fitting the thermodynamic diagram to obtain the symmetry information of the target part or the medical image of the target part.

In a third aspect, an embodiment of the present application provides a computer device, including a memory and a processor, where the memory stores a computer program, and the processor implements any one of the method steps in the above first aspect when executing the computer program.

In a fourth aspect, embodiments of the present application provide a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the method steps in any one of the above first aspect embodiments.

The embodiment of the application provides a symmetry information detection method, a device, computer equipment and a storage medium, wherein medical image data of a target part is acquired and input into a preset network model to obtain thermodynamic diagrams capable of reflecting the symmetry of the medical image of the target part or the target part, and then the thermodynamic diagrams are fitted to obtain the symmetry information of the medical image of the target part or the target part. According to the method, a preset network model is adopted to learn a thermodynamic diagram of the target part or the symmetric information of the medical image of the target part from the medical image data, namely, the symmetric distribution condition of the target part is known in advance through the thermodynamic diagram, and then plane fitting is carried out based on the symmetric distribution condition, so that fitting is carried out based on the actual symmetric distribution condition of the medical image of the target part or the target part, and the symmetric information of the target part can be accurately and efficiently determined; and the network model is trained before use, and the thermodynamic diagram of the medical image of the target part or the target part can be quickly and accurately obtained by inputting the medical image data into the preset network model, so that the efficiency and the accuracy of determining the symmetrical information of the medical image of the target part or the target part are further improved.

Drawings

FIG. 1 is a diagram of an exemplary embodiment of a symmetric information detection method;

FIG. 2 is a flow chart illustrating a method for detecting symmetric information according to an embodiment;

FIG. 2a is a schematic illustration of a thermodynamic diagram in one embodiment;

FIG. 2b is a diagram illustrating a detection result of symmetry information in one embodiment;

FIG. 2c is a diagram illustrating a detection result of symmetry information in another embodiment;

FIG. 3 is a flow chart illustrating a method for detecting symmetric information according to another embodiment;

FIG. 4 is a flow chart illustrating a method for detecting symmetric information according to another embodiment;

FIG. 5 is a flowchart illustrating a method for detecting symmetric information according to another embodiment;

FIG. 5a is a diagram illustrating a data clipping method according to another embodiment;

FIG. 5b is a diagram illustrating a data clipping method according to another embodiment;

FIG. 6 is a flow diagram of a method for symmetry information detection in one embodiment;

fig. 7 is a schematic structural diagram of a symmetric information detection apparatus in an embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments.

The symmetry information detection method provided by the present application can be applied to the application environment shown in fig. 1, and the computer device in fig. 1 includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing the relevant data of the symmetry information detection. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a symmetry information detection method.

The symmetry adjustment of the image plane in the scanning process by the related technology is analyzed based on the traditional algorithm, including but not limited to optimizing a plane equation by calculating the symmetry correlation coefficient of the original image and the flip image; or searching for points on a statistical curve meeting the change of the pixel values of each line row by row, and finally randomly sampling the points to fit a plane; or sampling points near the sagittal plane in the image, manually extracting features, performing regression by using a random forest, and finding a value with higher response of the sagittal plane so as to fit a final plane; the symmetry information can also be analyzed by segmenting specific structures; however, the method for calculating the symmetry correlation coefficient of the original image and the flip image to optimize the plane equation is not particularly good for the data with pathological changes, and the method for searching for the point on the statistical curve which satisfies the change of the pixel value of each line row by row is only suitable for the image which satisfies the statistical curve. Segmentation schemes may be more limited in certain scenarios, for example, if head symmetry information is obtained by segmenting the eyes, but the craniocerebral data does not necessarily have eyes.

Based on this, embodiments of the present application provide a symmetry information detection method, apparatus, computer device, and storage medium, which can efficiently and accurately determine scan location image symmetry information. The following describes in detail the technical solutions of the present application and how the technical solutions of the present application solve the above technical problems by embodiments and with reference to the drawings. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments. It should be noted that in the symmetry information detection method provided in the present application, the execution main body of fig. 2 to fig. 6 is a computer device, wherein the execution main body may also be a symmetry information detection apparatus, and the apparatus may be implemented as part or all of the computer device by software, hardware, or a combination of software and hardware.

In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. The following describes the symmetry information detection method provided by the present application with specific examples.

As shown in fig. 2, in one embodiment, there is provided a symmetry information detection method, including the steps of:

s101, medical image data of a target part is obtained.

The medical image data includes, but is not limited to, a Computed Tomography (CT) image, a Magnetic Resonance Imaging (MRI) image, and the like, and optionally, the medical image data includes a two-dimensional medical image or three-dimensional volume data, for example, a two-dimensional tomographic image which may be an image of a shoulder or a head or other parts, or three-dimensional volume data, which is not limited in this embodiment. The target region refers to a region that needs to be scanned currently, for example, data of head scanning, and may also be a head image of a region such as a cranium, a paranasal sinus, an eye socket, an inner ear, and the like, which is not limited in this application.

The acquired medical image data is a medical image to be input into the network model, and may be raw medical image data acquired by an imaging device, or may be an image stored in a storage medium such as a workstation or a PACS (medical image archiving and communication system).

For example, the computer device may obtain the medical image data of the target portion by sending a request to the medical image device, and then the medical image device collects an image of the target portion and sends the image back to the computer device; or the computer equipment receives the image input by the third-party equipment through the connecting interface; further alternatively, the image information may be downloaded from a network or a medical image database, and the like, which is not limited in this embodiment of the present application.

S102, inputting medical image data into a preset network model to obtain a thermodynamic diagram of the target part or the symmetric information of the medical image of the target part; the thermodynamic diagram reflects the distribution of the target region or the symmetry of the medical image of the target region.

After medical image data of a target portion is acquired, the medical image data of the target portion is input into a preset network model, the network model is a model which is trained in advance and used for extracting a thermodynamic diagram from a medical image, for example, the network model may be a neural network model trained based on a deep learning technique, or may be an algorithm model trained based on a mathematical algorithm, and the network model trained by the deep learning technique may be any one of a regression, segmentation or positioning network model, and the model is not limited in this embodiment.

After the medical image data of the target portion is input into the preset network model, the obtained output result is a thermodynamic diagram of the symmetry information of the target portion or the medical image of the target portion, where the thermodynamic diagram refers to a diagram that can reflect the symmetry distribution of the target portion or the medical image of the target portion, please refer to fig. 2a, the left side is an example of the medical image, the right side is a corresponding thermodynamic diagram, a commonly used thermodynamic diagram representation manner includes but is not limited to a mask, the symmetry information of the medical image of the target portion or the target portion can be accurately obtained through the thermodynamic diagram, and subsequent operations such as symmetry adjustment can be performed according to the obtained symmetry information.

And S103, fitting the thermodynamic diagram to obtain the symmetry information of the target part or the medical image of the target part.

Fitting the thermodynamic diagram, for example, performing a plane fitting, commonly used plane fitting methods include, but are not limited to, least squares linear fitting, SVD or Principal Component Analysis (PCA) computing Principal planes, and the like.

The symmetry information includes a symmetry plane or a symmetry line, that is, a symmetry plane or a symmetry line of the target portion or the medical image of the target portion obtained after the planar fitting of the thermodynamic diagram, and the symmetry information (the symmetry plane or the symmetry line) may be used to adjust the symmetry of the target portion in the medical image data. Wherein a plane of symmetry is obtained in case of a three-dimensional medical image and a line of symmetry is obtained in case of a two-dimensional medical image. Please refer to fig. 2b and fig. 2c, which illustrate a test result detected by using the symmetry information detection method provided in the embodiment of the present application, wherein fig. 2b is a result diagram of an example of a symmetry plane of a CT image of a soft tissue window head detected by using the symmetry information detection method provided in the embodiment of the present application; fig. 2c is a result diagram of an example of a symmetry plane detected by the symmetry information detection method provided in the embodiment of the present application in the bone window head CT image.

In the method for detecting symmetric information provided in this embodiment, medical image data of a target portion is acquired, the medical image data is input into a preset network model, a thermodynamic diagram capable of reflecting symmetry of the medical image of the target portion or the target portion is obtained, and then the thermodynamic diagram is fitted to obtain symmetric information of the medical image of the target portion or the target portion. In the method, a preset network model is adopted to learn the thermodynamic diagram of the target part or the medical image of the target part from the medical image data, namely, the symmetry distribution condition of the target part is known in advance through the thermodynamic diagram, and then fitting is carried out based on the symmetry distribution condition, so that the fitting is carried out based on the actual symmetry distribution condition of the medical image of the target part or the target part, and the symmetry information of the medical image of the target part or the target part can be accurately and efficiently determined; and the network model is trained before use, and the thermodynamic diagram of the medical image of the target part or the target part can be quickly and accurately obtained by inputting the medical image data into the preset network model, so that the efficiency and the accuracy of determining the symmetrical information of the medical image of the target part or the target part are further improved.

On the basis of the above embodiment, an embodiment of a symmetry information detection method is also provided, and the embodiment mainly relates to a specific process of acquiring medical image data of a target region by a computer device, as shown in fig. 3, the embodiment includes the following steps:

s201, acquiring an original medical image of a target part.

The embodiment takes a mode of acquiring medical image data of a target part through medical equipment as an example for explanation; the original medical image of the target region is referred to as a non-cropped and non-processed image acquired at the beginning of the scan, i.e., the acquired original medical image is based on the range that the device can scan, and may include other regions besides the target region, for example, when scanning head data, the target region is a sinus, and the scanned original medical image includes not only the sinus but also an orbit, which may be a part or a whole of the orbit, and is not limited thereto.

S202, preprocessing the original medical image to obtain medical image data of the target part.

After the original medical image of the target part is acquired, the original medical image is preprocessed, and then the medical image of the target part can be obtained. The preprocessing is to remove unnecessary regions in the original image, or to process the size, proportion, resolution, etc. of the original medical image to meet the requirements of the network model, so that the thermodynamic diagram of the target part can be determined effectively by using the network model, and the situation that the original medical image cannot be successfully input into the network model due to the unsatisfactory condition is avoided.

Optionally, the preprocessing that may be performed includes one or more of adjusting the resolution of the original medical image, normalizing the gray value of each pixel point in the original medical image, and capturing the region of the target portion from the original medical image.

Where the pre-processing is used to process the raw medical image to conform to the requirements of the regression network model input data. For example, the original medical image may be adjusted to a uniform resolution: [1.796mm,1.796mm,5mm ], the grey values of the original medical image are normalized: the value in the gray scale range of the original image-1024,2000 is stretched to [0,1 ]. For another example, assuming that the target region is the entire region of the head, the following extra regions of the head may be cropped to reduce interference: calculating the position of the top layer of the head of a CT image or a maximum intensity projection (mip) image of the CT in the sagittal direction, then intercepting the image according to the physical length of the head, abandoning a slice layer below the head, and only reserving the head area and the like.

The image obtained after the original medical image is preprocessed in the above way can be used as the medical image data of the target part to be input into the network model, so that the medical image data of the target part can accurately obtain the thermodynamic diagram of the target part or the medical image of the target part through the network model.

The network model generally has a function of performing targeted training on a use scene thereof to obtain a network model that better conforms to the effect of each scheme, and therefore, before the thermodynamic diagram of the target part is obtained through the preset network model, training sample data needs to be collected in advance, and the network model is obtained through training according to the training sample data. Based on this, the training process of the network model will be specifically described below.

As shown in FIG. 4, in one embodiment, the training process of the network model includes:

s301, acquiring training sample data; the training sample data includes a thermodynamic diagram in which each part medical image data corresponds to each part medical image data.

In practical application, besides training sample data, some training verification data can be prepared, namely, the training sample data is firstly adopted to train the network model, and then the training verification data is input into the trained network model to verify the performance of the network model. The training sample data and the training verification data are data of different batches, and the training verification data need to be data which does not participate in training the training network model.

The training sample data comprises medical image data of all parts and thermodynamic diagrams corresponding to the medical image data of all parts; the medical image data of each part, such as the medical image data of the eye frame part, the medical image data of the skull part, the medical image data of the inner ear and the like, the more parts are better when the sample data is acquired, the more comprehensive and complete the training sample data can be made, and therefore the applicability of the network model is improved.

In addition, it is necessary to acquire a thermodynamic diagram corresponding to medical image data of each region together with medical image data of each region, each training sample data includes a pair of medical image data of a region and a thermodynamic diagram corresponding to the medical image data of the region, and the thermodynamic diagram reflects medical image symmetry information of a target region or a target region, for example, in the case of craniocerebral data, the thermodynamic diagram shows an anatomical position where symmetry is relatively obvious, such as a sickle brain.

And S302, training the initial network according to the training sample data, and obtaining a network model after the training is finished.

After training sample data is obtained, the training sample data is input into an initial network, the initial network learns the mapping relation between the medical image data of each part in the training sample data and the thermodynamic diagrams corresponding to the medical image data, the learning is repeated, and the learning direction of the initial network can be adjusted according to the value of a preset loss function in the learning process until the training is completed to obtain a network model.

After the network model is obtained after training, the network model can be verified by adopting the above mentioned training verification data, that is, the medical image data of each part in the training verification data is input into the network model to verify the effect of the network, and a model with better verification effect under a certain iteration number can be selected through the verification effect, or a network model with better verification effect is selected from a plurality of different training network models.

In this embodiment, a network model that can directly obtain the thermodynamic diagrams of the respective parts based on the medical image data of the respective parts is obtained by obtaining training sample data and training, so that the thermodynamic diagrams of the target part or the medical image of the target part can be quickly and accurately obtained, and the efficiency and accuracy of determining the symmetry plane or the symmetry line of the medical image of the target part or the target part are improved.

For the network model, the more types of the acquired training sample data, the more accurate the thermodynamic diagram determined by the trained network model.

As shown in fig. 5, in an embodiment, the acquiring training sample data includes:

s401, obtaining sample medical images of various parts, and performing incremental processing on each sample medical image to obtain the sample medical image after the incremental processing.

In practical application, sufficient data cannot be collected for training, so that abundant sample medical images can be obtained by performing incremental processing on sample medical images of all parts in order to enrich training sample data.

The sample medical image refers to a medical image obtained by preprocessing an original medical image of each part, and the specific process can be referred to as the process shown in fig. 3.

Optionally, the performing incremental processing on each sample medical image includes: performing incremental processing on each sample medical image in a preset incremental mode; the incremental mode at least comprises one of data clipping, data rotation, Gaussian noise increasing and data translation.

The preset incremental mode includes multiple modes, for example, data clipping, data rotation, gaussian noise increase, data translation, and the like, where, taking the paranasal sinus as an example, the specific process of data clipping includes: after the collected sample medical images are cleaned, a sagittal plane Mip image is generated, data containing a nose are selected from the sagittal plane image, then a cutting frame is marked on the sagittal plane according to a scanning frame of the nasal sinus scanning, and the data are cut according to the marking frame, wherein the cutting mode can be shown in fig. 5a, wherein (I) in fig. 5a is an illustration that the head sagittal plane Mip does not contain the nose, fig. 5a is an illustration that (II) is an illustration of cutting of the nasal sinus scanning, fig. 5a is an illustration that (III) is another illustration of cutting mode of the nasal sinus scanning, please refer to fig. 5b, fig. 5b is an illustration that (I) in fig. 5b is an original image that the head contains different visual angles of the nasal sinuses, and fig. 5b is an illustration that (II) and (I) in fig. 5b are cut after the cutting of the images in the up-down one-to-one correspondence; the simulation of sinus data by various cropping methods is shown in fig. 5a and 5 b. Similarly, sample medical images of the orbital scan and also of the inner ear scan may be augmented in a similar manner.

The specific process of data rotation comprises the following steps: given that during the scan the patient may have some setup errors resulting in the data being tilted, and considering that the head data is symmetric, the data is randomly flipped left and right and rotated by an increment of a certain angle, e.g. (0 ° -30 °) left and right, in 3D space. Thus, the scanned data is rotated by a certain angle or turned left and right to form different samples. Gaussian noise is added to the medical images of the samples of all parts; or data translation is carried out, and the image is translated to a certain degree up, down, left and right to increase the type of data, considering that some data are not completely shot in the scanning visual field.

S402, acquiring a sample thermodynamic diagram corresponding to each increment processed sample medical image.

And S403, determining the sample medical image after each incremental processing and the corresponding sample thermodynamic diagram as training sample data.

After the sample medical images are subjected to increment processing in the various increment modes, the sample medical images after the increment processing and the sample thermodynamic diagrams corresponding to the sample medical image data after the increment processing are obtained, and the sample medical image data after the increment processing and the corresponding sample thermodynamic diagrams are determined as training sample data.

In this embodiment, sample medical images of a plurality of parts are acquired, the sample medical images are subjected to incremental processing to obtain incrementally processed sample medical images, and then the incrementally processed sample medical images and corresponding sample thermodynamic diagrams are determined as training sample data.

In an embodiment, there is further provided a symmetry information detection method, as shown in fig. 6, the embodiment includes the following steps:

s1, acquiring a sample medical image, and performing incremental processing on the image; execution of S2;

s2, acquiring thermodynamic diagrams corresponding to the images after the incremental processing, wherein the images after the incremental processing and the corresponding thermodynamic diagrams are training data, and executing S3;

s3, training the initial network by using the training data, and obtaining a network model after the training is finished; execution of S4;

s4, acquiring a medical image of the target part; execution of S5;

s5, preprocessing the medical image; execution of S6;

s6, inputting the preprocessed image into the trained network model to obtain a thermodynamic diagram of the symmetric information of the medical image of the target part or the target image; the thermodynamic diagram reflects the symmetry distribution of the target part or the medical image of the target image; execution of S7;

s7, the thermodynamic diagram is fitted to obtain the symmetry information of the target region or the medical image of the target image.

The implementation principle and technical effect of each step in the symmetric information detection method provided in this embodiment are similar to those in the previous embodiments of the symmetric information detection method, and are not described herein again. The implementation manner of each step in the embodiment of fig. 6 is only an example, and is not limited to this, and the order of each step may be adjusted in practical application as long as the purpose of each step can be achieved.

It should be understood that although the various steps in the flow charts of fig. 2-6 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2-6 may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed in turn or alternately with other steps or at least some of the other steps.

In one embodiment, as shown in fig. 7, there is provided a symmetry information detecting apparatus including: an acquisition module 10, a processing module 11, a fitting module 12, wherein,

an acquisition module 10 for acquiring medical image data of a target site;

the processing module 11 is configured to input medical image data into a preset network model to obtain a thermodynamic diagram of the target portion or the symmetric information of the medical image of the target portion; the thermodynamic diagram reflects the symmetry distribution of the target part or the medical image of the target part;

and the fitting module 12 is configured to fit the thermodynamic diagram to obtain symmetry information of the target portion or the medical image of the target portion.

In one embodiment, the obtaining module 10 includes:

the acquisition unit is used for acquiring an original medical image of a target part;

and the preprocessing unit is used for preprocessing the original medical image to obtain the medical image data of the target part.

In one embodiment, the apparatus further comprises:

the training sample acquisition module is used for acquiring training sample data; the training sample data comprises thermodynamic diagrams corresponding to the medical image data of each part and the medical image data of each part;

and the training module is used for training the initial network according to the training sample data and obtaining a network model after the training is finished.

In one embodiment, the training sample acquiring module includes:

the increment processing unit is used for acquiring sample medical images of various parts and carrying out increment processing on each sample medical image to obtain an increment processed sample medical image;

the acquisition unit is used for acquiring a sample thermodynamic diagram corresponding to each increment-processed sample medical image;

and the determining unit is used for determining the sample medical images after the incremental processing and the corresponding sample thermodynamic diagrams as training sample data.

In an embodiment, the increment processing unit is specifically configured to perform increment processing on each sample medical image in a preset increment manner; the incremental mode at least comprises one of data clipping, data rotation, Gaussian noise increasing and data translation.

In an embodiment, the medical image data comprises two-dimensional medical image data or three-dimensional volume data.

For the specific definition of the symmetry information detection apparatus, reference may be made to the above definition of the symmetry information detection method, which is not described herein again. The modules in the symmetry information detection apparatus can be implemented in whole or in part by software, hardware, and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, which may be a terminal, the internal structure of which may be as described above in fig. 1. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless communication can be realized through WIFI, an operator network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a symmetry information detection method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.

Those skilled in the art will appreciate that the above-described architecture shown in fig. 1 is merely a block diagram of some of the structures associated with the present solution, and does not constitute a limitation on the computing devices to which the present solution applies, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:

acquiring medical image data of a target part;

The implementation principle and technical effect of the computer device provided by the above embodiment are similar to those of the above method embodiment, and are not described herein again.

In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:

acquiring medical image data of a target part;

The implementation principle and technical effect of the computer-readable storage medium provided by the above embodiments are similar to those of the above method embodiments, and are not described herein again.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include at least one of non-volatile and volatile memory. Non-volatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical storage, or the like. Volatile Memory can include Random Access Memory (RAM) or external cache Memory. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others.

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. A method for detecting symmetry information, the method comprising:

acquiring medical image data of a target part;

inputting the medical image data into a preset network model to obtain a thermodynamic diagram of the target part or the symmetric information of the medical image of the target part; the thermodynamic diagram reflects a symmetry distribution of the target part or a medical image of the target part;

2. The method of claim 1, wherein the acquiring medical image data of the target site comprises:

acquiring an original medical image of the target part;

3. The method of claim 2, wherein the pre-processing comprises one or more of adjusting a resolution of the original medical image, normalizing a gray value of each pixel point in the original medical image, and extracting a region of the target portion from the original medical image.

4. The method according to any one of claims 1-3, wherein prior to inputting the medical image data into a preset network model, the method further comprises:

acquiring training sample data; the training sample data comprises medical image data of all parts and thermodynamic diagrams corresponding to the medical image data of all parts;

and training the initial network according to the training sample data, and obtaining the network model after training.

5. The method of claim 4, wherein the obtaining training sample data comprises:

acquiring sample medical images of multiple parts, and performing incremental processing on each sample medical image to obtain an incrementally processed sample medical image;

and determining the incremental processed sample medical images and the corresponding sample thermodynamic diagrams as the training sample data.

6. The method of claim 5, wherein said incrementally processing each of said sample medical images comprises:

7. A method according to any of claims 1-3, wherein the medical image data comprises two-dimensional medical image data or three-dimensional volume data.

8. A symmetry information detecting apparatus, characterized in that the apparatus comprises:

an acquisition module for acquiring medical image data of a target part;

the processing module is used for inputting the medical image data into a preset network model to obtain a thermodynamic diagram of the target part or the symmetric information of the medical image of the target part; the thermodynamic diagram reflects a symmetry distribution of the target part or a medical image of the target part;

9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 7.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.