Testis volume determination method and device, electronic equipment and storage medium

Abstract

The invention discloses a testis volume determination method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining a target three-dimensional magnetic resonance image of a testis to be measured, and obtaining a plurality of magnetic resonance image slices corresponding to the target three-dimensional magnetic resonance image; for each magnetic resonance image slice, inputting the magnetic resonance image slice and the adjacent magnetic resonance image slice into an image segmentation model, and obtaining a testis segmentation image slice corresponding to the magnetic resonance image slice based on an output result of the image segmentation model; and determining a testis three-dimensional segmented image corresponding to the target three-dimensional magnetic resonance image based on testis segmented image slices corresponding to the plurality of magnetic resonance image slices, and determining the volume of the testis to be measured based on the total number of pixel points contained in a testis region in the testis three-dimensional segmented image. By adopting the technical scheme, the testis volume can be automatically and accurately determined through the testis magnetic resonance image.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and particularly to a method, device, electronic device and storage medium for determining testicular volume. Background Art

[0002] Testicular volume measurement has important value in clinical practice, which is specifically reflected in the following aspects: reflecting sperm and hormone status: since the seminiferous tubules account for about 80 - 90% of the testicular mass, testicular volume can reflect sperm and hormone status. Monitoring testicular function and pathological conditions: in clinical practice, TV is a key parameter for monitoring testicular function and pathological conditions. Evaluating pubertal development status: the increase in testicular volume is the earliest sign of elevated gonadotropins during puberty. Therefore, testicular volume measurement is used to monitor testicular development and pubertal status. Related to normal sperm production: normal sperm production occurs only when the total testicular volume is close to or normal. The amount of testicular volume loss is related to the degree of spermatogenesis disorder. A key component of male infertility assessment: it has been proven that TV is related to semen analysis results. Therefore, testicular volume measurement is a key component of male infertility assessment. For patients with various diseases affecting testicular growth and fertility, accurate and individualized testicular volume measurement may improve the accuracy of diagnosis and treatment. In summary, testicular volume measurement is a key parameter for evaluating male fertility, testicular function and possible pathological conditions. For doctors, understanding and accurately measuring testicular volume is crucial to ensure proper diagnosis and treatment of patients.

[0003] Regarding related testicular volume measurement methods, one is to measure through physical tools such as calipers and measuring instruments. This measurement method is easily affected by nearby soft tissues such as the epididymis, scrotal skin and subcutaneous tissues, especially in the case of small testicles and hydrocele. Therefore, there is a problem of poor objectivity in the measured testicular volume. Another is to measure with the aid of ultrasound. This measurement method requires deriving the testicular volume through a formula. However, since the testis is an organ with elasticity and compressibility, its shape is neither uniform nor necessarily ellipsoidal. Therefore, this method of deriving testicular volume by formula has a high subjectivity, and there are differences in measurement results among different studies, formulas and examiners. It is considered only a rough estimate of testicular volume, and there is often a deviation from the true volume. This makes it difficult to establish normal testicular volume values and critical values for distinguishing pathological conditions. Summary of the Invention

[0004] The present invention provides a method, device, electronic device and storage medium for determining testicular volume to solve the technical problem of inaccurate testicular volume measurement.

[0005] According to one aspect of the present invention, there is provided a method for determining testicular volume, the method comprising:

[0006] Obtain a target three-dimensional magnetic resonance image of the testis to be measured, and obtain a plurality of magnetic resonance image slices corresponding to the target three-dimensional magnetic resonance image;

[0007] For each of the magnetic resonance image slices, input the magnetic resonance image slice and its adjacent magnetic resonance image slices into an image segmentation model, and obtain a testicular segmentation image slice corresponding to the magnetic resonance image slice based on the output result of the image segmentation model, wherein the image segmentation model includes a self-convolutional neural network, an encoding module based on an attention mechanism, and a decoding module;

[0008] Determine a testicular three-dimensional segmentation image corresponding to the target three-dimensional magnetic resonance image based on the testicular segmentation image slices corresponding to the plurality of magnetic resonance image slices, and determine the volume of the testis to be measured based on the total number of pixel points included in the testicular region in the testicular three-dimensional segmentation image.

[0009] According to another aspect of the present invention, there is provided a testicular volume determination device, the device includes:

[0010] A magnetic resonance image acquisition module, configured to obtain a target three-dimensional magnetic resonance image of the testis to be measured, and obtain a plurality of magnetic resonance image slices corresponding to the target three-dimensional magnetic resonance image;

[0011] A testicular image segmentation module, configured to, for each of the magnetic resonance image slices, input the magnetic resonance image slice and its adjacent magnetic resonance image slices into an image segmentation model, and obtain a testicular segmentation image slice corresponding to the magnetic resonance image slice based on the output result of the image segmentation model, wherein the image segmentation model includes a self-convolutional neural network, an encoding module based on an attention mechanism, and a decoding module;

[0012] A testicular volume determination module, configured to determine a testicular three-dimensional segmentation image corresponding to the target three-dimensional magnetic resonance image based on the testicular segmentation image slices corresponding to the plurality of magnetic resonance image slices, and determine the volume of the testis to be measured based on the total number of pixel points included in the testicular region in the testicular three-dimensional segmentation image.

[0013] According to another aspect of the present invention, there is provided an electronic device, the electronic device includes:

[0014] At least one processor; and

[0015] A memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores a computer program executable by the at least one processor. When executed by the at least one processor, the computer program enables the at least one processor to execute the testicular volume determination method according to any embodiment of the present invention.

[0017] According to another aspect of the present invention, there is provided a computer-readable storage medium storing computer instructions for implementing the testicular volume determination method according to any embodiment of the present invention when executed by a processor.

[0018] The technical solution of the embodiment of the present invention first obtains a target three-dimensional magnetic resonance image of the testis to be measured and obtains a plurality of magnetic resonance image slices corresponding to the target three-dimensional magnetic resonance image, so as to fully obtain relevant information of the testis to be measured. Then, for each magnetic resonance image slice, the magnetic resonance image slice and its adjacent magnetic resonance image slices are input into an image segmentation model, and a testicular segmentation image slice corresponding to the magnetic resonance image slice is obtained based on the output result of the image segmentation model, which can realize automatic and intelligent image segmentation. Since the image segmentation model includes a self-convolution neural network, an encoding module based on an attention mechanism, and a decoding module, it can effectively obtain the global image information of the target three-dimensional magnetic resonance image, thereby more accurately segmenting the testicular region. Based on the testicular segmentation image slices corresponding to the plurality of magnetic resonance image slices, a testicular three-dimensional segmentation image corresponding to the target three-dimensional magnetic resonance image is determined, and the volume of the testis to be measured is determined based on the total number of pixel points included in the testicular region in the testicular three-dimensional segmentation image. The determination method is simple and convenient, solves the technical problem of inaccurate measurement of testicular volume, and achieves the beneficial effect of automatically and accurately determining the testicular volume without the need to rely on physical measurement tools and without manual intervention.

[0019] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0021] Figure 1 is a schematic flowchart of a testicular volume determination method according to Embodiment 1 of the present invention;

[0022] Figure 2 It is a schematic flowchart of a method for determining testicular volume according to Embodiment 2 of the present invention;

[0023] Figure 3 It is a schematic diagram comparing the effects before and after separating the testicular region in a method for determining testicular volume according to Embodiment 2 of the present invention in the case of testicular adhesion;

[0024] Figure 4 It is a schematic structural diagram of a device for determining testicular volume according to Embodiment 3 of the present invention;

[0025] Figure 5 It is a schematic structural diagram of an electronic device for implementing the method for determining testicular volume of the embodiments of the present invention. Detailed implementation manners

[0026] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0027] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described here can be implemented in an order other than those illustrated or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0028] It should be noted that the modifications of "one" and "multiple" mentioned in this disclosure are illustrative rather than restrictive. Those skilled in the art should understand that unless clearly specified otherwise in the context, it should be understood as "one or more".

[0029] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only for illustrative purposes and are not used to limit the scope of these messages or information.

[0030] It is understandable that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, usage scopes, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to users and the authorization of users should be obtained through appropriate means in accordance with relevant laws and regulations.

[0031] For example, when responding to an active request from a user, a prompt message is sent to the user to clearly prompt the user that the operation requested by the user will require obtaining and using the user's personal information. Thus, the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, an application program, a server, or a storage medium that performs the operations of the technical solutions of the present disclosure according to the prompt message.

[0032] As an optional but non-limiting implementation manner, the manner of sending a prompt message to the user in response to receiving an active request from the user may be, for example, in the form of a pop-up window, and the prompt message may be presented in text in the pop-up window. In addition, the pop-up window may also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0033] It is understandable that the above process of notifying and obtaining user authorization is only illustrative and does not limit the implementation manner of the present disclosure, and other manners that meet relevant laws and regulations can also be applied to the implementation manner of the present disclosure.

[0034] It is understandable that the data involved in the present technical solution (including but not limited to the data itself, the acquisition or use of the data) should comply with the requirements of corresponding laws, regulations and related provisions.

[0035] Embodiment 1

[0036] Figure 1 The flowchart of a method for determining testicular volume provided in Embodiment 1 of the present invention is shown. This embodiment is applicable to the situation of measuring the volume of the testis. This method can be executed by a testicular volume determination device, and the testicular volume determination device can be implemented in the form of hardware and / or software. Optionally, it is implemented through an electronic device, and the electronic device can be a mobile terminal, a PC terminal or a server, etc.

[0037] As Figure 1 shown, this method may specifically include:

[0038] S110. Obtain a target three-dimensional magnetic resonance image of the testis to be measured, and obtain a plurality of magnetic resonance image slices corresponding to the target three-dimensional magnetic resonance image.

[0039] Among them, the testis to be measured can be understood as the testis whose volume is to be determined. The target three-dimensional magnetic resonance image can be understood as a three-dimensional magnetic resonance image to be divided into a plurality of magnetic resonance image slices for testicular region segmentation.

[0040] In an embodiment of the present invention, obtaining a target three-dimensional magnetic resonance image of a testis to be measured may specifically include: obtaining an original three-dimensional magnetic resonance image of the testis to be measured, and performing normalization processing on the original three-dimensional magnetic resonance image to obtain a target three-dimensional magnetic resonance image. Among them, there are various ways of normalization processing. For example, normalization processing based on the maximum value and the minimum value, normalization processing based on the mean value and the standard deviation, or normalization processing based on the logarithmic function, etc.

[0041] Exemplarily, the maximum-minimum normalization processing may be performed on the original three-dimensional magnetic resonance image based on the following formula:

[0042]

[0043] Where x′ represents the pixel value of a pixel point in the original three-dimensional magnetic resonance image after normalization processing, x represents the original value of the pixel point in the original three-dimensional magnetic resonance image, x min represents the minimum value among the pixel values of all pixel points in the original three-dimensional magnetic resonance image, x max represents the minimum value among the pixel values of all pixel points in the original three-dimensional magnetic resonance image.

[0044] Optionally, obtaining a plurality of magnetic resonance image slices corresponding to the target three-dimensional magnetic resonance image includes: determining a spatial coordinate system corresponding to the target three-dimensional magnetic resonance image, and along the axis direction of the spatial coordinate system, obtaining a plurality of magnetic resonance image slices corresponding to the target three-dimensional magnetic resonance image based on the slice spacing corresponding to the axis.

[0045] Among them, the axis direction may be that the spatial coordinate system includes the x, y axes or the z axis. The spatial coordinate system can be understood as the coordinate system used when collecting the magnetic resonance scan data of the testis to be measured. Generally, specifically, the z axis of the spatial coordinate system is the direction parallel to the main magnetic field B0 direction, and its positive direction points to one end of the examination table. The direction of the z axis is consistent with the long axis of the supine position of the scanned object. After the z axis direction is determined, x and y are orthogonal (perpendicular) to the z axis. Specifically, the x axis direction is parallel to the coronal plane, that is, the left and right of the scanned object, and the left side is set as the positive direction. The y axis direction is parallel to the sagittal plane, that is, the front and back of the scanned object, and the front is set as the positive direction.

[0046] S120. For each of the magnetic resonance image slices, input the magnetic resonance image slice and its adjacent magnetic resonance image slice into an image segmentation model, and obtain a testicular segmentation image slice corresponding to the magnetic resonance image slice based on the output result of the image segmentation model.

[0047] Among them, the image segmentation model includes a self-convolutional neural network, an encoding module based on an attention mechanism, and a decoding module. Exemplarily, the image segmentation model may include a TransUnet network model, which adopts a self-attention mechanism to encode tokenized image patches of feature maps from a Convolutional Neural Network (CNN) into an input sequence for extracting global context. Then, to make up for the loss brought by Transformer feature encoding, TransUNet adopts a hybrid CNN-Transformer architecture. The decoding module upsamples the encoded features and then combines them with the feature maps output from each layer of different high-resolution CNNs that have not been processed from the encoding path to achieve precise positioning of the testicular region.

[0048] It can be understood that before inputting the magnetic resonance image slice and its adjacent magnetic resonance image slices into the image segmentation model, it further includes: obtaining a training image group composed of an odd number of adjacent sample image slices in the arrangement position, training the model to be trained based on the training image group and the expected segmentation result corresponding to the training image group to obtain an image segmentation model. Among them, the model to be trained includes a self-convolutional neural network, an encoding module based on an attention mechanism, and a decoding module. The sample image slice can be understood as multiple magnetic resonance image slices corresponding to the three-dimensional magnetic resonance image of the testis used as a training sample. The expected segmentation result corresponding to the training image group is the expected segmentation result corresponding to the sample image slice located in the middle of the odd number of sample image slices in the arrangement position.

[0049] Specifically, input the sample image slice and the sample image slice adjacent to the sample image slice into the model to be trained to obtain a model output result. Furthermore, calculate the loss between the model output result and the expected output result, and adjust the model parameters of the model to be trained based on the loss to obtain the image segmentation model. Among them, the expected segmentation result is the testicular segmentation result expected to be output by the image segmentation model corresponding to the magnetic resonance image slice.

[0050] Optionally, perform data augmentation processing on the sample image slices to enhance the generalization ability of the model. Among them, the data augmentation processing may include, but is not limited to, one or more of random rotation, random translation, random brightness change, local brightness transformation, etc.

[0051] Optionally, the loss function adopted by the image segmentation model during training is constructed based on a binary cross-entropy loss function and a set similarity metric function, and the loss calculation weight corresponding to each pixel point in the binary cross-entropy loss function is associated with the loss value of the pixel point in the current iteration.

[0052] Exemplarily, the loss function adopted by the image segmentation model during training can be obtained by summing or weighted summing the binary cross-entropy loss function and the set similarity metric function. Specifically, the loss calculation weight corresponding to each pixel point in the binary cross-entropy loss function is positively correlated with the loss value of the pixel point in the current iteration. That is, the larger the loss value of the pixel point in the current iteration, the larger the loss calculation weight corresponding to each pixel point in the binary cross-entropy loss function.

[0053] Among them, the binary cross-entropy loss function can calculate the loss for each pixel point in the magnetic resonance image slice to determine whether the pixel point is a positive sample (a pixel point belonging to the testicular region) or a negative sample (a pixel point located in other regions outside the testis). Exemplarily, the expression of the binary cross-entropy loss function can be shown as follows:

[0054]

[0055] Among them, BCE_Loss represents the binary cross-entropy loss value corresponding to a single pixel point in the magnetic resonance image slice, p(x) represents the segmentation result of this pixel point in the magnetic resonance image slice actually output by the image segmentation model in the current iteration; y represents the segmentation result of this pixel point in the magnetic resonance image slice expected to be output by the image segmentation model in the current iteration.

[0056] When the loss value of a certain pixel point in the magnetic resonance image slice is less than the preset first loss threshold t1, it indicates that the model to be trained already has a great ability to predict this pixel point correctly. At this time, the model to be trained does not need to focus on this pixel point; when the loss value of a certain pixel is greater than the first loss threshold t1 and less than the preset second loss threshold t2, it indicates that there is still room for optimization in the segmentation of this pixel point, and the basic weight 1 is given; when the loss value of a certain pixel point is greater than the second loss threshold t2, it indicates that the prediction of the pixel point is completely wrong and a larger weight, such as 2, is required. Specifically, it can be expressed as:

[0057]

[0058] It should be noted that the specific values of the first loss threshold t1, the second loss threshold t2, and W(BCE_Loss) in each range can be set according to actual needs. The above are only examples of values rather than limitations.

[0059] Finally, multiplying BCE_Loss by the adaptive weight W(BCE_Loss) can obtain the weighted loss value of each pixel point, and the final loss value is:

[0060] AW_BCE = mean(W * BCE_loss)

[0061] The set similarity metric function can specifically be the Dice loss, which can be used to calculate the similarity between two samples and can alleviate the negative impact brought by the area imbalance between the foreground (testis region) and the background (regions other than the testis region) in the output result of the image segmentation model and the expected output result. The foreground-background imbalance means that most regions in the magnetic resonance image slice do not contain the testis, and only a small part of the regions contain the testis. Dice Loss training pays more attention to the excavation of the testis region, that is, ensuring a low probability of predicting testis pixels incorrectly. Its expression can be specifically as follows:

[0062]

[0063] Among them, DiceLoss represents the set similarity metric loss, X represents the number of pixels that predict non-testis region pixels as testis region pixels, and Y represents the number of pixels that predict testis region pixels as non-testis region pixels.

[0064] The AW_BCE loss function combined with DiceLoss is used as the final loss function, and its expression is as follows:

[0065] loss = α * AW_BCE + β * DiceLoss

[0066] Among them, α and β are the weights of AW_BCE and DiceLoss respectively, and are hyperparameters greater than 0, and their specific values can be set according to requirements.

[0067] Adopting the above technical solution can enable the image segmentation model to have a global receptive field and can automatically weight different magnetic resonance image slices, making the model more robust.

[0068] Specifically, inputting the magnetic resonance image slice and its adjacent magnetic resonance image slices into the image segmentation model includes: respectively obtaining a preset number of magnetic resonance image slices located on both sides of the magnetic resonance image slice and adjacent to the magnetic resonance image slice in the arrangement direction corresponding to the plurality of magnetic resonance image slices as the associated image slices corresponding to the magnetic resonance image slice; inputting the magnetic resonance image slice and the associated image slices corresponding to the magnetic resonance image slice into the image segmentation model.

[0069] Exemplarily, three consecutive magnetic resonance image slices can be randomly extracted on the z-axis plane corresponding to the target three-dimensional magnetic resonance image. The model outputs the corresponding label as the segmentation result corresponding to the middle magnetic resonance image slice, that is, the image segmentation model outputs the result of the middle magnetic resonance image slice. Specifically, three consecutive magnetic resonance image slices are sequentially extracted from top to bottom along the z-axis and input into the model for testicular region segmentation. The image segmentation model outputs the segmentation result as the corresponding result of the middle magnetic resonance image slice. After all the magnetic resonance image slices are traversed, the prediction results are integrated into the layout corresponding to the original magnetic resonance image slices to obtain a three-dimensional segmentation result.

[0070] S130. Determine a three-dimensional testicular segmentation image corresponding to the target three-dimensional magnetic resonance image based on the testicular segmentation image slices corresponding to multiple magnetic resonance image slices, and determine the volume of the testis to be measured based on the total number of pixel points included in the testicular region in the three-dimensional testicular segmentation image.

[0071] As can be seen from the foregoing, the target three-dimensional magnetic resonance image is composed of multiple magnetic resonance image slices. By segmenting each magnetic resonance image slice, a testicular segmentation image slice corresponding to each magnetic resonance image slice is obtained. Furthermore, a three-dimensional testicular segmentation image corresponding to the target three-dimensional magnetic resonance image can be determined based on the testicular segmentation image slices.

[0072] Optionally, the testicular segmentation image slices corresponding to multiple magnetic resonance image slices are spliced according to the arrangement order of the magnetic resonance image slices in the target three-dimensional magnetic resonance image to obtain a three-dimensional testicular segmentation image corresponding to the target three-dimensional magnetic resonance image; or, the testicular segmentation image slices are reprocessed to remove scattered false positive pixel points (pixel points misidentified as testicular pixel points) in the testicular segmentation image slices. For example, the connected regions composed of testicular pixel points in the testicular segmentation image slices can be marked, and the connected regions with an area smaller than a preset threshold are removed, etc. Among them, removing the connected regions with an area smaller than the preset threshold can be to modify the pixel values of each pixel point in the connected regions with an area smaller than the preset threshold to a preset value, etc. The preset value can specifically be a value different from the pixel values corresponding to the pixel points in the testicular region, such as 0.

[0073] Specifically, the volume of the testis to be measured can be determined based on the total number of pixel points included in the testis region in the three-dimensional segmentation image of the testis. That is, based on the total number of pixel points included in the testis region in the three-dimensional segmentation image of the testis and the actual lengths of each pixel point in each axis direction of the spatial coordinate system corresponding to the target three-dimensional magnetic resonance image, the volume of the testis to be measured is determined. Suppose the total number of pixel points included in the testis region of a certain three-dimensional segmentation image of the testis is N, and the slice spacings of the three axial planes of the spatial coordinate system corresponding to the target three-dimensional magnetic resonance image are Sx, Sy, and Sz respectively. Then the volume of this testis is N * Sx * Sy * Sz.

[0074] The technical solution of the embodiment of the present invention first obtains the target three-dimensional magnetic resonance image of the testis to be measured and obtains a plurality of magnetic resonance image slices corresponding to the target three-dimensional magnetic resonance image, so as to fully obtain the relevant information of the testis to be measured. Then, for each magnetic resonance image slice, the magnetic resonance image slice and its adjacent magnetic resonance image slices are input into the image segmentation model, and the testis segmentation image slice corresponding to the magnetic resonance image slice is obtained based on the output result of the image segmentation model, which can realize automatic and intelligent image segmentation. Since the image segmentation model includes a self-convolution neural network, an encoding module based on an attention mechanism, and a decoding module, it can effectively obtain the global image information of the target three-dimensional magnetic resonance image, thereby more accurately segmenting the testis region. The three-dimensional segmentation image of the testis corresponding to the target three-dimensional magnetic resonance image is determined based on the testis segmentation image slices corresponding to the plurality of magnetic resonance image slices, and the volume of the testis to be measured is determined based on the total number of pixel points included in the testis region in the three-dimensional segmentation image of the testis. The determination method is simple and convenient, solves the technical problem of inaccurate measurement of the testis volume, and achieves the beneficial effect of automatically and accurately determining the testis volume without relying on physical measurement tools and without manual intervention.

[0075] Embodiment 2

[0076] Figure 2Schematic flowchart of a method for determining testicular volume provided in the second embodiment of the present invention. Based on the above embodiment, this embodiment further refines the method for generating a three-dimensional testicular segmentation image from magnetic resonance image slices. Optionally, determining a three-dimensional testicular segmentation image corresponding to the target three-dimensional magnetic resonance image based on testicular segmentation image slices corresponding to multiple magnetic resonance image slices includes: constructing an initial three-dimensional segmentation image corresponding to the target three-dimensional magnetic resonance image based on multiple magnetic resonance image slices, and determining a testicular region in the initial three-dimensional segmentation image; in the case where the number of testicular regions is one, segmenting the testicular region based on the shortest distance between each pixel point in the testicular region and the region boundary of the testicular region to obtain a three-dimensional testicular segmentation image. For specific implementation manners, reference may be made to the description of this embodiment. Technical features that are the same as or similar to those in the foregoing embodiments will not be described herein again.

[0077] As Figure 2 shown, the method may specifically include:

[0078] S210. Obtain a target three-dimensional magnetic resonance image of the testis to be measured, and obtain multiple magnetic resonance image slices corresponding to the target three-dimensional magnetic resonance image.

[0079] S220. For each magnetic resonance image slice, input the magnetic resonance image slice and its adjacent magnetic resonance image slices into an image segmentation model, and obtain a testicular segmentation image slice corresponding to the magnetic resonance image slice based on the output result of the image segmentation model.

[0080] S230. Construct an initial three-dimensional segmentation image corresponding to the target three-dimensional magnetic resonance image based on multiple magnetic resonance image slices, and determine a testicular region in the initial three-dimensional segmentation image.

[0081] As mentioned above, the target three-dimensional magnetic resonance image is composed of multiple magnetic resonance image slices. By segmenting each magnetic resonance image slice, a testicular segmentation image slice corresponding to each magnetic resonance image slice is obtained. Therefore, an initial three-dimensional segmentation image corresponding to the target three-dimensional magnetic resonance image can be constructed from multiple testicular segmentation image slices. Specifically, according to the arrangement order corresponding to multiple magnetic resonance image slices, splice the testicular segmentation image slices corresponding to the magnetic resonance image slices to obtain an initial three-dimensional segmentation image corresponding to the target three-dimensional magnetic resonance image.

[0082] Optionally, determining the testicular region in the initial three-dimensional segmentation image includes: performing binarization processing on multiple pixel points in the initial three-dimensional segmentation image based on the output result of the image segmentation model and a preset pixel segmentation threshold to obtain a binarized three-dimensional segmentation image; determining the connected regions included in the binarized three-dimensional segmentation image, and determining the testicular region in the initial three-dimensional segmentation image based on the volume of the connected regions.

[0083] Among them, the output result of the image segmentation model can specifically be the probability that each pixel point in the initial three-dimensional segmentation image belongs to the testicular region. The preset pixel segmentation threshold can be understood as a probability threshold preset for distinguishing testicular pixel points and non-testicular pixel points.

[0084] Specifically, performing binarization processing on multiple pixel points in the initial three-dimensional segmentation image based on the output result of the image segmentation model and a preset pixel segmentation threshold to obtain a binarized three-dimensional segmentation image may include: taking the pixel points in the output result of the image segmentation model in the initial three-dimensional segmentation image that reach the preset pixel segmentation threshold as testicular pixel points and assigning a first value (e.g., 1); taking the pixel points in the output result of the image segmentation model in the initial three-dimensional segmentation image that do not reach the preset pixel segmentation threshold as non-testicular pixel points and assigning a second value (e.g., 0), thereby obtaining a binarized three-dimensional segmentation image.

[0085] It should be noted that there are many ways to determine the connected regions included in the binarized three-dimensional segmentation image. Existing connected region labeling methods can be used, such as the depth-first search algorithm, etc. No specific limitation is imposed on which connected region labeling method to use here.

[0086] Optionally, determining the testicular region in the initial three-dimensional segmentation image based on the volume of the connected regions includes: in the case where the volume of the connected region is greater than a preset volume threshold, taking the connected region as a candidate region; furthermore, determining the testicular region in the initial three-dimensional segmentation image according to the number of candidate regions and the theoretical number of testicles to be measured. Generally, the theoretical number of testicles to be measured is two, but in some special scenarios (such as, lesions or gene mutations, etc.), there may also be cases where the theoretical number of testicles to be measured is one or three.

[0087] Optionally, when the number of the candidate regions is less than or equal to two, the candidate regions are used as the testicular regions in the initial three-dimensional segmentation image. It can be understood that when the theoretical number of the testes to be measured is two, the detected candidate regions can be regarded as the testicular regions in the initial three-dimensional segmentation image. In the case of there being one testicular region, it can be determined that when the image segmentation model performs image segmentation on the testes to be measured, due to testicular adhesion, two testicular regions are recognized as one testicular region. At this time, the testicular region can be further segmented by means of distance transformation.

[0088] Optionally, when the number of the candidate regions is less than or equal to two, the two candidate regions with the largest volumes (i.e., the candidate region with the largest volume and the candidate region with the second largest volume) can be selected as the testicular regions in the initial three-dimensional segmentation image; or, multiple candidate regions are displayed, and in response to a region selection trigger operation, the selected candidate regions are used as the testicular regions in the initial three-dimensional segmentation image, etc.

[0089] S240. When the number of the testicular regions is one, the testicular region is segmented based on the shortest distance between each pixel point in the testicular region and the regional boundary of the testicular region, so as to obtain a three-dimensional testicular segmentation image.

[0090] It can be understood that in the initial three-dimensional segmentation image, there is a testicular region, that is, there is an initial boundary of the testicular region, that is, the regional boundary. As mentioned above, when the number of the testicular regions is one, it can be considered that when the image segmentation model performs image segmentation on the testes to be measured, testicular adhesion occurs, as shown in the left figure in Figure 3 Therefore, the distance between each pixel point in the testicular region and the regional boundary of the testicular region can be used to re-segment the testicular region to accurately segment two testicular regions.

[0091] Optionally, the segmenting the testicular region based on the shortest distance between each pixel point in the testicular region and the regional boundary of the testicular region to obtain a three-dimensional testicular segmentation image includes: respectively determining the shortest distance between each pixel point in the testicular region and the regional boundary of the testicular region; by gradually adjusting a distance threshold, segmenting the testicular region into two sub-regions based on the shortest distance and the distance threshold, so as to obtain a three-dimensional testicular segmentation image.

[0092] In the embodiments of the present invention, the slice intervals corresponding to the respective coordinate axes of the spatial coordinate system corresponding to the target three-dimensional magnetic resonance image may be the same or different. Exemplarily, when the slice intervals corresponding to the respective coordinate axes of the spatial coordinate system corresponding to the target three-dimensional magnetic resonance image are different, the original three-dimensional magnetic resonance image corresponding to the target three-dimensional magnetic resonance image can be resampled to the same slice interval. Wherein, the slice interval corresponding to each coordinate axis can be understood as the actual length of each pixel point in the direction of each coordinate axis of the spatial coordinate system. When the slice intervals corresponding to the respective coordinate axes of the spatial coordinate system corresponding to the target three-dimensional magnetic resonance image are the same, a distance transform is performed according to the prediction result, that is, the shortest distance from each pixel point in the testicular region of the initial three-dimensional segmentation image to the regional boundary of the testicular region is calculated; further, the distance threshold T is incremented from 1 until the testicular region is exactly divided into two connected domains. At this time, the testicular region is binarized with the current distance threshold t as the threshold to obtain two sub-regions, denoted as a and b; then, the distances from other testicular pixels to a and b are calculated respectively. If the pixel is closest to region a, it means that the pixel belongs to the sub-region where a is located; otherwise, it belongs to the sub-region where b is located. Thus, the adhered testicular regions are separated to obtain two final testicular regions, as shown in Figure 3 the right figure in

[0093] S230. Determine the volume of the testis to be measured based on the total number of pixel points included in the testicular region in the testicular three-dimensional segmentation image.

[0094] The technical solution of the embodiments of the present invention can initially construct an initial three-dimensional segmentation image according to the output result of the model by constructing an initial three-dimensional segmentation image corresponding to the target three-dimensional magnetic resonance image based on a plurality of the magnetic resonance image slices and determining the testicular region in the initial three-dimensional segmentation image, and then initially determine a three-dimensional testicular region; then, when the number of the testicular regions is one, it indicates that the testicular region is adhered. At this time, the testicular region is re-segmented based on the shortest distance between each pixel point in the testicular region and the regional boundary of the testicular region to obtain a testicular three-dimensional segmentation image, realizing the reprocessing of the adhered testicular region, thereby realizing the accurate segmentation of the testicular region.

[0095] Embodiment III

[0096] Figure 4 is a schematic structural diagram of a testis volume determination device provided in Embodiment III of the present invention. As shown in Figure 4As shown in the figure, the testicular volume determination device includes: a magnetic resonance image acquisition module 410, a testicular image segmentation module 420, and a testicular volume determination module 430. Among them, the magnetic resonance image acquisition module 410 is used to acquire a target three-dimensional magnetic resonance image of the testis to be measured, and acquire a plurality of magnetic resonance image slices corresponding to the target three-dimensional magnetic resonance image; the testicular image segmentation module 420 is used for each of the magnetic resonance image slices, input the magnetic resonance image slice and its adjacent magnetic resonance image slices into an image segmentation model, and obtain a testicular segmentation image slice corresponding to the magnetic resonance image slice based on the output result of the image segmentation model, where the image segmentation model includes a self-convolutional neural network, an encoding module based on an attention mechanism, and a decoding module; the testicular volume determination module 430 is used to determine a testicular three-dimensional segmentation image corresponding to the target three-dimensional magnetic resonance image based on the testicular segmentation image slices corresponding to the plurality of magnetic resonance image slices, and determine the volume of the testis to be measured based on the total number of pixel points included in the testicular region in the testicular three-dimensional segmentation image.

[0097] The technical solution of the embodiment of the present invention first obtains a target three-dimensional magnetic resonance image of the testis to be measured through the magnetic resonance image acquisition module, and obtains a plurality of magnetic resonance image slices corresponding to the target three-dimensional magnetic resonance image, which can fully obtain the relevant information of the testis to be measured; then, through the testicular image segmentation module, for each of the magnetic resonance image slices, input the magnetic resonance image slice and its adjacent magnetic resonance image slices into the image segmentation model, and obtain a testicular segmentation image slice corresponding to the magnetic resonance image slice based on the output result of the image segmentation model, which can realize automatic and intelligent image segmentation. Since the image segmentation model includes a self-convolutional neural network, an encoding module based on an attention mechanism, and a decoding module, it can effectively obtain the global image information of the target three-dimensional magnetic resonance image, so as to more accurately segment the testicular region; finally, through the testicular volume determination module, determine a testicular three-dimensional segmentation image corresponding to the target three-dimensional magnetic resonance image based on the testicular segmentation image slices corresponding to the plurality of magnetic resonance image slices, and determine the volume of the testis to be measured based on the total number of pixel points included in the testicular region in the testicular three-dimensional segmentation image. The determination method is simple and convenient, solves the technical problem of inaccurate measurement of testicular volume, and achieves the beneficial effect of automatically and accurately determining the testicular volume without the need to rely on physical measurement tools and without manual intervention.

[0098] On the basis of the technical solutions of the embodiments of the present invention, optionally, the loss function adopted by the image segmentation model during training is constructed based on a binary cross-entropy loss function and a set similarity metric function, and the loss calculation weight corresponding to each pixel point in the binary cross-entropy loss function is associated with the loss value of the pixel point in the current iteration.

[0099] Based on the technical solutions of the embodiments of the present invention, optionally, the testicular image segmentation module includes an associated image determination unit and an image segmentation unit. Among them, the associated image determination unit is configured to respectively obtain a preset number of the magnetic resonance image slices located on both sides of and adjacent to the magnetic resonance image slice in the arrangement direction corresponding to the plurality of magnetic resonance image slices as the associated image slices corresponding to the magnetic resonance image slice; the image segmentation unit is configured to input the magnetic resonance image slice and the associated image slices corresponding to the magnetic resonance image slice into an image segmentation model.

[0100] Based on the technical solutions of the embodiments of the present invention, optionally, the testicular volume determination module includes a preliminary testicular region determination unit and a reprocessing unit for the testicular region. Among them, the preliminary testicular region determination unit is configured to construct an initial three-dimensional segmentation image corresponding to the target three-dimensional magnetic resonance image based on the plurality of magnetic resonance image slices and determine the testicular region in the initial three-dimensional segmentation image; the reprocessing unit for the testicular region is configured to, when the number of the testicular regions is one, segment the testicular region based on the shortest distance between each pixel point in the testicular region and the region boundary of the testicular region to obtain a three-dimensional testicular segmentation image.

[0101] Based on the technical solutions of the embodiments of the present invention, optionally, the preliminary testicular region determination unit includes a binarization processing subunit and a testicular region determination subunit. Among them, the binarization processing subunit is configured to perform binarization processing on a plurality of pixel points in the initial three-dimensional segmentation image based on the output result of the image segmentation model and a preset pixel segmentation threshold to obtain a binarized three-dimensional segmentation image; the testicular region determination subunit is configured to determine the testicular region in the initial three-dimensional segmentation image based on the connected domains included in the binarized three-dimensional segmentation image and the volume of the connected domains.

[0102] Based on the technical solutions of the embodiments of the present invention, optionally, the testicular region determination subunit is specifically configured to: when the volume of the connected domain is greater than a preset volume threshold, use the connected domain as a candidate region; and when the number of the candidate regions is less than or equal to two, use the candidate region as the testicular region in the initial three-dimensional segmentation image.

[0103] Based on the technical solutions of the embodiments of the present invention, optionally, the reprocessing unit for the testicular region is configured to: respectively determine the shortest distance between each pixel point in the testicular region and the region boundary of the testicular region; and segment the testicular region into two sub-regions based on the shortest distance and a distance threshold by gradually adjusting the distance threshold to obtain a three-dimensional testicular segmentation image.

[0104] The testicular volume determination device provided by the embodiments of the present invention can execute the testicular volume determination method provided by any embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the testicular volume determination method.

[0105] Embodiment 4

[0106] Figure 5 FIG. shows a schematic structural diagram of an electronic device 10 that can be used to implement the embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as, for example, a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, for example, a personal digital processor, a cellular phone, a smart phone, a wearable device (such as a helmet, glasses, a watch, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present invention described herein and / or claimed.

[0107] As Figure 5 shown, the electronic device 10 includes at least one processor 11, and a memory communicatively connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc. Among them, the memory stores a computer program executable by the at least one processor, and the processor 11 can execute various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. The input / output (I / O) interface 15 is also connected to the bus 14.

[0108] A plurality of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0109] The processor 11 may be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the testicular volume determination method.

[0110] In some embodiments, the testicular volume determination method may be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed onto the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the testicular volume determination method described above may be executed. Alternatively, in other embodiments, the processor 11 may be configured to execute the testicular volume determination method by any other suitable means (e.g., by means of firmware).

[0111] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuitry, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special-purpose or general-purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0112] The computer program for implementing the method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to the processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowchart and / or block diagram are implemented. The computer program can be executed entirely on the machine, partially on the machine, as a stand-alone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0113] In the context of the present invention, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0114] To provide for interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can also be used to provide for interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic, speech, or tactile input).

[0115] The systems and techniques described herein can be implemented in a computing system that includes backend components (such as, for example, a data server), or a computing system that includes middleware components (such as, for example, an application server), or a computing system that includes frontend components (such as, for example, a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (such as, for example, a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), blockchain network, and the Internet.

[0116] A computing system may include a client and a server. The client and the server are generally far from each other and usually interact via a communication network. The client-server relationship is created by computer programs that run on respective computers and have a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system, solving the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0117] It should be understood that various forms of the processes shown above can be used, with steps reordered, added, or deleted. For example, the steps recited in the present invention can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is made herein.

[0118] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for determining testicular volume, characterized in that, Including: Obtaining a target three-dimensional magnetic resonance image of a testis to be measured, and obtaining a plurality of magnetic resonance image slices corresponding to the target three-dimensional magnetic resonance image; For each of the magnetic resonance image slices, inputting the magnetic resonance image slice and its adjacent magnetic resonance image slices into an image segmentation model, and obtaining a testicular segmentation image slice corresponding to the magnetic resonance image slice based on the output result of the image segmentation model, wherein the image segmentation model includes a self-convolutional neural network, an encoding module based on an attention mechanism, and a decoding module; Determining a testicular three-dimensional segmentation image corresponding to the target three-dimensional magnetic resonance image based on the testicular segmentation image slices corresponding to the plurality of magnetic resonance image slices, and determining the volume of the testis to be measured based on the total number of pixel points included in the testicular region in the testicular three-dimensional segmentation image.

2. The method according to claim 1, wherein The loss function adopted by the image segmentation model during training is constructed based on a binary cross-entropy loss function and a set similarity metric function, and the loss calculation weight corresponding to each pixel point in the binary cross-entropy loss function is associated with the loss value of the pixel point in the current iteration.

3. The method according to claim 1, wherein The step of inputting the magnetic resonance image slice and its adjacent magnetic resonance image slices into the image segmentation model includes: On the arrangement direction corresponding to the plurality of magnetic resonance image slices, respectively obtaining a preset number of the magnetic resonance image slices located on both sides of the magnetic resonance image slice and adjacent to the magnetic resonance image slice as associated image slices corresponding to the magnetic resonance image slice; Inputting the magnetic resonance image slice and the associated image slices corresponding to the magnetic resonance image slice into the image segmentation model.

4. The method according to claim 1, characterized in that, The step of determining a testicular three-dimensional segmentation image corresponding to the target three-dimensional magnetic resonance image based on the testicular segmentation image slices corresponding to the plurality of magnetic resonance image slices includes: Constructing an initial three-dimensional segmentation image corresponding to the target three-dimensional magnetic resonance image based on the plurality of magnetic resonance image slices, and determining a testicular region in the initial three-dimensional segmentation image; In the case where the number of the testicular regions is one, segmenting the testicular region based on the shortest distance between each pixel point in the testicular region and the regional boundary of the testicular region to obtain a testicular three-dimensional segmentation image.

5. The method according to claim 4, wherein The step of determining the testicular region in the initial three-dimensional segmentation image includes: Performing binarization processing on a plurality of pixel points in the initial three-dimensional segmentation image based on the output result of the image segmentation model and a preset pixel segmentation threshold to obtain a binarized three-dimensional segmentation image; Determining the connected domains included in the binarized three-dimensional segmentation image, and determining the testicular region in the initial three-dimensional segmentation image based on the volume of the connected domains.

6. The method according to claim 5, characterized in that, The step of determining the testicular region in the initial three-dimensional segmentation image based on the volume of the connected domains includes: In the case where the volume of the connected domain is greater than a preset volume threshold, taking the connected domain as a candidate region; In the case where the number of the candidate regions is less than or equal to two, taking the candidate region as the testicular region in the initial three-dimensional segmentation image.

7. The method according to claim 4, wherein Segmenting the testicular region based on the shortest distance between each pixel point in the testicular region and the regional boundary of the testicular region to obtain a three-dimensional testicular segmentation image, including: Determining the shortest distance between each pixel point in the testicular region and the regional boundary of the testicular region respectively; By gradually adjusting the distance threshold, segmenting the testicular region into two sub-regions based on the shortest distance and the distance threshold to obtain a three-dimensional testicular segmentation image.

8. A testicular volume determination device, characterized in that, Including: A magnetic resonance image acquisition module, configured to acquire a target three-dimensional magnetic resonance image of a testis to be measured, and acquire a plurality of magnetic resonance image slices corresponding to the target three-dimensional magnetic resonance image; A testicular image segmentation module, configured to, for each of the magnetic resonance image slices, input the magnetic resonance image slice and its adjacent magnetic resonance image slices into an image segmentation model, and obtain a testicular segmentation image slice corresponding to the magnetic resonance image slice based on the output result of the image segmentation model, wherein the image segmentation model includes a self-convolutional neural network, an encoding module based on an attention mechanism, and a decoding module; A testicular volume determination module, configured to determine a three-dimensional testicular segmentation image corresponding to the target three-dimensional magnetic resonance image based on the testicular segmentation image slices corresponding to the plurality of magnetic resonance image slices, and determine the volume of the testis to be measured based on the total number of pixel points included in the testicular region in the three-dimensional testicular segmentation image.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the testicular volume determination method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing a processor to implement the testicular volume determination method according to any one of claims 1-7 when executed.

Images