Method, device and system for identifying sagittal plane in head

Abstract

The invention relates to a method, a device and a system for identifying a sagittal plane in a head. According to the recognition method, at least two kinds of core tissues and / or bilaterally symmetrical tissues are selected from a head 3D medical image, the core tissues take points on the middle sagittal plane of the head as centroid points, and the bilaterally symmetrical tissues take points on the middle sagittal plane of the head as symmetrical points; segmenting the selected core organization and / or the bilaterally symmetrical organization based on a first learning network and generating a corresponding 3D segmentation mask; determining a centroid point of the 3D segmentation mask in each image layer based on the 3D segmentation mask of the core organization, and determining a symmetric point of the 3D segmentation mask in each image layer based on the 3D segmentation mask of the bilateral symmetry organization; and based on the centroid point / symmetric point cloud set of each tissue, performing multiple rounds of curved surface fitting by using a curved surface fitting algorithm to obtain a final head midsagittal plane. According to the method, the accurate and robust identification of the sagittal plane in the head can be realized under the condition of not excessively depending on image quality, single tissue segmentation quality and symmetry calculation possibly generating precision errors.

Description

Technical Field

[0001] The present application relates to the field of medical data processing, and in particular to a method, device and system for identifying the midsagittal plane of the head. Background Art

[0002] The midsagittal plane (MSP, also known as the median sagittal plane) is an ideal geometric plane that symmetrically divides the brain into two equal parts, left and right, through the human body. It is essential for determining the location, size, and extent of lesions. Midsagittal plane detection in medical images such as head CT has important application value in clinical medicine. In the diagnosis of brain tumors, the midsagittal plane can help doctors accurately locate tumors and evaluate their impact on surrounding brain tissue. In addition, the midsagittal plane is also very important for the evaluation of conditions such as hydrocephalus, brain trauma, and cerebrovascular lesions, and can provide important information about changes in the ventricular system and cerebral sulci. In emergency situations, such as stroke, midsagittal plane detection can quickly identify whether hemorrhage or ischemic lesions lead to more serious secondary changes, such as brain herniation, which is crucial for treatment decisions and also plays an indispensable role in the diagnosis and treatment planning of neurological diseases.

[0003] Among the existing mid-sagittal plane extraction methods, one type requires using binarization and expansion algorithms to remove the skull from the cranial medical image and extract a three-dimensional mask of the brain parenchyma, using a region growing algorithm to fill the cavities in the brain parenchyma and expand to the original skull boundary, and then using the middle slice of the brain as a reference to fit a plane ellipse, and finally determining the mid-sagittal plane based on the major axis coordinates and rotation angle of the ellipse. This method has high requirements on image quality and relies on the quality of a single brain parenchyma tissue segmentation, and the region growing algorithm is not very robust to segmentation, especially for data in pathological changes such as tumors and hemorrhage. Another type of mid-sagittal plane recognition method requires first using SVM (support vector machine) and Hough transform to detect the location of the longitudinal fissure, determine the initial mid-sagittal plane, and then select multiple undetermined planes around the initial mid-sagittal plane, and determine the final mid-sagittal plane based on the symmetry parameters and grayscale parameters of these undetermined planes. This type of method is also highly dependent on image quality and the accuracy of identifying the location of the brain fissure. For example, threshold segmentation of the SVM algorithm and Hough transform are sensitive to parameter selection and may need to be adjusted for different cases, which will also affect the data effect on complex cases.

[0004] It can be seen that in the prior art, there has not yet been found a method for recognizing the midsagittal plane of the head that is less dependent on image quality and single tissue segmentation accuracy, is not dependent on the recognition accuracy of the symmetry axis, and has high recognition accuracy for different data and strong generalization ability. Summary of the invention

[0005] The present application is provided to solve the above-mentioned defects existing in the prior art. A method, device and system for identifying the mid-sagittal plane of the head are needed, which can realize accurate and robust identification of the mid-sagittal plane of the head without over-reliance on image quality and single tissue segmentation quality, while avoiding precision errors that may be generated by reliance on symmetry calculations such as symmetry axes.

[0006] According to a first scheme of the present application, a method for identifying a midsagittal plane of a head is provided, the method comprising: obtaining a 3D medical image of the head; normalizing and preprocessing the obtained 3D medical image of the head for size changes; selecting at least two core tissues and / or bilaterally symmetrical tissues from the preprocessed 3D medical image of the head as segmentation objects, wherein the core tissue has a point on the midsagittal plane of the head as a centroid point, and the bilaterally symmetrical tissue has a point on the midsagittal plane of the head as a symmetric point; based on a first learning network, segmenting the selected core tissues / laterally symmetrical tissues to generate 3D segmentation masks of each core tissue / laterally symmetrical tissue; determining the centroid point of each core tissue in each image layer based on the 3D segmentation mask of each core tissue, determining the symmetric point of each bilaterally symmetrical tissue in each image layer based on the 3D segmentation mask of each bilaterally symmetrical tissue, and taking the set of the centroid point of each core tissue / the symmetric point of the bilaterally symmetrical tissue as the point cloud set of the corresponding tissue; and performing multiple rounds of surface fitting using a surface fitting algorithm based on the point cloud set of each tissue, so as to screen out the midsagittal plane of the head obtained based on the fitting of some centroid points / symmetric points.

[0007] According to a second solution of the present application, a device for identifying the midsagittal plane of the head is provided, comprising an interface and a processor, wherein the interface is configured to receive a 3D medical image of the head; and the processor is configured to execute the method for identifying the midsagittal plane of the head as described in the various embodiments of the present application.

[0008] According to a third solution of the present application, a system for identifying the midsagittal plane of the head is provided. The system includes a head 3D medical imaging device and a device for identifying the midsagittal plane of the head as described in the various embodiments of the present application.

[0009] The methods, devices and systems for identifying the midsagittal plane of the head provided in various embodiments of the present application, based on fully exploring the essential characteristics of the midsagittal plane, utilize the important characteristic that the centroid point of the core tissue and the symmetry points of the bilaterally symmetrical tissue should theoretically be on the midsagittal plane or very close to the midsagittal plane. First, part or all of these two types of tissues are segmented, and then the centroid points / symmetry points of these tissues are determined. A large number of centroid points / symmetry points contain rich information associated with the midsagittal plane of the head. Using these points, through multiple rounds of surface fitting, the midsagittal plane of the head is fitted by using the best subset of centroid points / symmetry points. The above-mentioned recognition method does not rely on the segmentation of a single tissue, nor does it require pre-identification of the symmetry axis / symmetry plane used as a reference. Therefore, it overcomes the shortcomings of the prior art that over-rely on image quality and single tissue segmentation quality. Moreover, even if some tissues have pathological changes that cause the corresponding centroid points / symmetry points to deviate greatly from the optimal midsagittal plane of the head, multiple rounds of surface fitting can be used to minimize the impact of these deviations. Therefore, the recognition method of the present application can have high recognition accuracy and strong robustness for different recognition objects.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0011] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0013] Figure 1 A schematic flow chart of a method for identifying the midsagittal plane of the head according to an embodiment of the present application is shown.

[0014] Figure 2 A schematic diagram showing a specific process of fitting the midsagittal plane of the head based on a point cloud set of various tissues according to an embodiment of the present application is shown.

[0015] Figure 3 A schematic diagram showing some components of a device for identifying the midsagittal plane of the head according to an embodiment of the present application is shown.

[0016] Figure 4 A schematic diagram showing the composition of a system for identifying the midsagittal plane of the head according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the embodiment of the present application clearer, the technical solution of the embodiment of the present application will be clearly and completely described in conjunction with the drawings of the embodiment of the present application. Obviously, the described embodiment is a part of the embodiment of the present application, not all of the embodiments. Based on the described embodiment of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0018] Unless otherwise defined, the technical terms or scientific terms used in this application shall have the common meanings understood by persons with ordinary skills in the field to which this application belongs. "Include" or "comprising" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, but do not exclude other elements or objects.

[0019] The words "first", "second" and similar words used in this application do not indicate any order, quantity or importance, but are only used to distinguish. Words such as "include" or "comprise" and similar words mean that the elements before the word include the elements listed after the word, and do not exclude the possibility of also covering other elements. The execution order of the various steps in the method described in this application in conjunction with the accompanying drawings is not intended to be limiting. As long as it does not affect the logical relationship between the various steps, several steps can be integrated into a single step, a single step can be decomposed into multiple steps, and the execution order of the various steps can be changed according to specific needs.

[0020] It should also be understood that the term "and / or" in this application is only a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this application generally indicates that the associated objects before and after are in an "or" relationship.

[0021] In order to keep the following description of the embodiments of the present application clear and concise, the present application omits detailed descriptions of known functions and known components.

[0022] According to an embodiment of the present application, a method for identifying the midsagittal plane of the head is provided. Figure 1 A schematic flow chart of a method for identifying the midsagittal plane of the head according to an embodiment of the present application is shown.

[0023] like Figure 1As shown, first, in step 101, a processor may acquire a head 3D medical image. In some embodiments, the head 3D medical image may be, for example, a head plain scan CT image, a head enhanced CT scan image, or a head magnetic resonance image (MRI), and the head 3D medical image may include multiple image layers.

[0024] In step 102, it is usually necessary to normalize and pre-process the acquired 3D medical image of the head. In order to make the original medical images of different specifications better adapt to the subsequent tissue segmentation and centroid / symmetry point recognition and other processing, it is necessary to make the image grayscale and image size (including image layer spacing, etc.) of the acquired 3D medical image of the head within a certain range through normalization and size change pre-processing, wherein the normalization algorithm can adopt Z-Score normalization method, minimum maximum normalization method, and mean normalization method, etc., and the present application does not limit the specific method. As an example only, for example, the image size can be changed to a uniform 512*512, and the image layer spacing can be transformed to no more than 1.5mm by linear interpolation and other methods, etc. The specific size can be determined according to the requirements of the subsequent processing flow for the input image, and the present application does not limit this.

[0025] In step 103, no less than two core tissues and / or bilaterally symmetrical tissues are selected from the preprocessed head 3D medical image as segmentation objects, wherein the core tissue has a point on the midsagittal plane of the head as a centroid point, and the bilaterally symmetrical tissue has a point on the midsagittal plane of the head as a symmetry point, and further based on the first learning network, the selected core tissues / laterally symmetrical tissues are segmented to generate 3D segmentation masks for each core tissue / laterally symmetrical tissue.

[0026] The core tissues may include, for example, brain parenchyma, brainstem, comb, sella turcica center, bottom of the inferior colliculus, internal occipital protuberance, nasal bone, superior sagittal sinus, inferior sagittal sinus, etc.; the bilaterally symmetrical tissues may include, for example, left eyeball and right eyeball, left lens and right lens, left optic nerve and right optic nerve, left and right maxillary sinus, optic chiasm, left and right intracranial segments of internal carotid artery, etc., wherein the intracranial portion of internal carotid artery mainly includes C2 segment to C7 segment. It is worth noting that the above core tissues / laterally symmetrical tissues may have an inclusion relationship. For example, clinically, the brain parenchyma includes the brainstem, which is located in a smaller layer of the lower half of the brain. However, from the perspective of imaging, the brainstem region usually has a tissue boundary and can be segmented separately. Therefore, when it is inconvenient to completely segment the brain parenchyma, segmenting the brainstem separately as a core tissue can also be used for fitting the midsagittal plane.

[0027] In some embodiments, only at least two core tissues can be selected as segmentation objects, or at least two left-right symmetrical tissues can be selected as segmentation objects. Both types of tissues can also be taken into account, and the total number of tissues used as segmentation objects is guaranteed to be no less than two. There is no restriction on the specific number of tissues. When the computing conditions can support and the real-time recognition is not too high, a larger number of tissues can bring higher recognition accuracy of the midsagittal plane of the head. In other embodiments, when selecting tissues as segmentation objects, the centroid points of the core tissues, or the symmetric points of the left-right symmetrical tissues, have different depths in the front-back direction of the midsagittal plane of the head as much as possible, so that when performing surface fitting, these centroid points / symmetric points can be relatively dispersed, and are not concentratedly distributed at positions approximately at the same depth, affecting the accuracy of surface fitting.

[0028] Preferably, when the image quality of the 3D medical image of the head allows, the brain parenchyma should be taken as the segmentation object, because from the perspective of anatomical structure, there is no other left-right symmetrical tissue above the eyes. Therefore, including the brain parenchyma in the segmentation object can provide more available surface fitting point information for the part above the eyes in the subsequent midsagittal plane fitting process, thereby better ensuring the accuracy of midsagittal plane fitting.

[0029] In step 104, the centroid point of each core tissue in each image layer is determined based on the 3D segmentation mask of each core tissue, and the symmetry point of each bilaterally symmetric tissue in each image layer is determined based on the 3D segmentation mask of each bilaterally symmetric tissue, and the set of the centroid point of each core tissue / the symmetry point of the bilaterally symmetric tissue is used as the point cloud set of the corresponding tissue.

[0030] In some embodiments, for example, existing tissue segmentation algorithms such as deep learning technology based on convolutional neural networks, Unet or ResNet and other algorithm frameworks can be used to segment each tissue to obtain the centroid points or symmetry points of each tissue in different image layers. This application does not impose any restrictions on this.

[0031] In step 105, based on the point cloud set of each tissue, a surface fitting algorithm is used to perform multiple rounds of surface fitting, so as to screen out the head midsagittal plane obtained based on the fitting of some centroid points / symmetric points. Among them, the surface fitting algorithm can use, for example, the least squares method, or a deformation algorithm such as the weighted least squares method, the total least squares method, the robust least squares method, or other applicable polynomial fitting, spline function fitting, neural network fitting method, etc., which is not limited in this application. In each round of surface fitting, for example, a subset of different centroid points / symmetric points can be selected from the point cloud set of each tissue to obtain different fitting surfaces, so as to select the optimal surface as the final head midsagittal plane.

[0032] As mentioned above, the centroid points of each core tissue and the symmetry points of each left-right symmetrical tissue should be located on the midsagittal plane of the head or very close to it from a priori perspective. Therefore, the set of these points contains sufficiently rich information about the midsagittal plane of the head. By utilizing the surface fitting algorithm and performing multiple rounds of fitting operations, defects such as unclear images, non-standard shapes due to lesions in tissues, and further loss of accuracy caused by segmentation can be eliminated to a certain extent. This ensures that the midsagittal plane in head medical images can always be identified robustly and accurately.

[0033] Figure 2 A schematic diagram showing a specific process of fitting the midsagittal plane of the head based on a point cloud set of various tissues according to an embodiment of the present application is shown.

[0034] like Figure 2 As shown, in step 201, based on the point cloud set of each tissue, multiple rounds of surface fitting are performed using a surface fitting algorithm, and steps 2011 to 2013 are executed sequentially in each round of surface fitting.

[0035] In step 2011, a subset of centroid points / symmetric points is obtained by randomly sampling the point cloud sets of each tissue according to the first sampling ratio of this round.

[0036] In the embodiments of the present application, different sampling methods can be used for the subset of centroid points / symmetric points as needed. As an example only, the first sampling ratio of the first round of surface fitting can be determined as follows: based on the number of centroid points / symmetric points required for each round of surface fitting, the number of centroid points / symmetric points to be sampled is evenly distributed to each tissue, that is, when the total number of tissues is n and the first sampling ratio is expressed as a vector containing n elements, the first sampling ratio R1 can be expressed in the form of the following formula (1): R1=(r1, r2, …, r n )=(1 / n, 1 / n, …, 1 / n) (1) Among them, r1, r2, …, r n They respectively represent the first ratio of the number of centroid points / symmetry points to be sampled in the point cloud set of the corresponding tissue to the number of centroid points / symmetry points required for this round of surface fitting. Then, random sampling can be performed in the corresponding point cloud set according to the first ratio to obtain a subset of the centroid points / symmetry points.

[0037] In some other embodiments, the first sampling ratio of the first round of surface fitting can also be determined in the following manner: first, according to the number of centroid points / symmetric points contained in the point cloud set of each tissue, a first ratio between the number of centroid points / symmetric points and the total number of centroid points / symmetric points of all tissues is determined, and each first ratio is used as the ratio value of each corresponding tissue in the first sampling ratio R1, that is, the first sampling ratio R1 is calculated according to the following formula (2): R1=(r1, r2, …, r n ) = (c1 / c, c2 / c, …, c n / c) (2) Where c1, c2, …, c n Respectively represent the number of centroid points / symmetric points contained in the point cloud set of each tissue, and c represents the total number of centroid points / symmetric points of all tissues. According to the above formula (2), it can not only ensure that the subset of centroid points / symmetric points used for surface fitting contains the points of each segmented tissue, but also the number of sampling points of each tissue is proportional to the number of points in its point cloud set. In general, since the number of points in the point cloud set is related to the number of image layers, more points in the point cloud set often means that the tissue is more widely distributed in the vertical range. Therefore, sampling more points in the point cloud set will be conducive to a wider and more uniform coverage of different image layers by the sampling points in the subset of centroid points / symmetric points, thereby helping to improve the accuracy of surface fitting.

[0038] In some other embodiments, especially when the segmentation quality of each tissue in the 3D medical image of the head is uneven due to shooting or other reasons, the first sampling ratio of the first round of surface fitting can also be determined as follows: assigning a corresponding ratio weight to each tissue based on its segmentation quality, which is used to weight the first ratio of the sampling of the corresponding tissue, that is, the first sampling ratio R1 is calculated according to the following formula (3): R1=(w1*r1, w2*r2, …, w n *r n )(3) Among them, r1, r2, …, r n The first ratio corresponding to the tissue sampling can be uniformly 1 / n, 1 / n, …, 1 / n obtained according to the method of formula (1), or c1 / c, c2 / c, …, c obtained according to the method of formula (2). n / c, or the first ratio may be determined by other methods, which is not limited in the present application. nIndicates the proportional weight of each tissue. According to the above formula (3), the segmentation quality of a specific tissue can be further considered, and the number of sampling points in a tissue with a higher segmentation quality is relatively larger, thereby ensuring a higher accuracy of the surface fitting.

[0039] In other embodiments, the intermediate candidate surfaces obtained by each round of fitting are presented to users such as doctors in an intuitive manner, for example, they can be superimposed and displayed on the 3D medical image of the head together with the segmentation of each tissue, so as to facilitate the doctor to compare and observe. In particular, when the doctor user, for example, determines based on his experience that the intermediate candidate surfaces obtained by the previous round of fitting have obvious bias, the first sampling ratio and other parameters used in the current round of surface fitting can be manually intervened. As an example only, for example, when the user determines through observation that a core tissue or bilaterally symmetrical tissue in the 3D medical image of the head has a congenital deformity or a large range of lesions, so that its centroid point / symmetry point is obviously not on the midsagittal plane, the first sampling ratio to be used in the current round of surface fitting can be manually modified, for example, the sampling ratio value (i.e., the first ratio) of the deformed / lesioned tissue can be set to a very small value or even set to 0 to avoid the points of the tissue from having an adverse effect on the surface fitting. In response to the user's interactive operation on the first sampling ratio of the current round, the first sampling ratio of the current round is adjusted, and the intermediate candidate surface fitting of the current round is performed based on the adjusted first sampling ratio. In this way, the user's experience can be taken into consideration during multiple rounds of surface fitting, which can not only make the head midsagittal plane fitting more accurate, but also speed up the convergence of the entire recognition process in some cases.

[0040] In step 2012, based on the subset of centroid points / symmetric points, a surface fitting algorithm is used to fit the intermediate candidate surface of this round.

[0041] In some embodiments, in each round, multiple random sampling can be performed in various tissues at the same first sampling ratio to obtain multiple subsets of centroid points / symmetry points, and surface fitting can be performed more than once to obtain multiple intermediate candidate surfaces in the same round.

[0042] In step 2013, based on the matching degree threshold range, the matching degree between other centroid points / symmetric points outside the subset of centroid points / symmetric points and the intermediate candidate curved surface is verified, and the fitting quality of the intermediate candidate curved surface is evaluated.

[0043] In some embodiments, the matching threshold can be set as needed while considering the iterative convergence speed, and this application does not impose specific restrictions. The specific calculation method of the fitting quality can, for example, substitute each other centroid point / symmetry point outside the subset of centroid points / symmetry points into the analytical expression of the intermediate candidate surface of this round, and use the points whose calculated values ​​do not exceed the matching threshold as the inner points of the intermediate candidate surface, and use the points that exceed the matching threshold range as the outer points of the intermediate candidate surface, and then evaluate the fitting quality of the intermediate candidate surface based on the number of inner points or outer points, or the proportion of inner points or outer points in the total number of points. For example, the larger the number or proportion of inner points, the higher the fitting quality, which will not be listed here one by one.

[0044] Next, in step 202, from the intermediate candidate curved surfaces obtained by each round of fitting in step 201, the intermediate candidate curved surface with the highest fitting quality is selected as the final midsagittal plane of the head.

[0045] In some embodiments, the first sampling ratio of this round is dynamically adjusted according to the verification result of the matching degree between other centroid points / symmetric points other than the subset of centroid points / symmetric points in the previous round and the intermediate candidate surface, which can be specifically performed according to the following steps: First, when performing each round of verification, the centroid points / symmetric points other than the subset of centroid points / symmetric points and the centroid points / symmetric points whose distances from the intermediate candidate curved surface of the round exceed the matching degree threshold range are taken as the outer points of the intermediate candidate curved surface of the round; Then, based on the core tissue / left-right symmetrical tissue to which each external point belongs in the previous round of verification, the first sampling ratio of this round is adjusted, and the core tissue / left-right symmetrical tissue with a larger proportion of external points in the previous round of verification has a greater degree of reduction in the sampling ratio value in this round.

[0046] According to the above method, multiple rounds of surface fitting can be reasonably used to identify tissues with lower quality due to potential reasons such as segmentation quality or the presence of lesions. In addition, considering that the lesions in these tissues may be local, the centroid points / symmetry points therein are not necessarily completely unavailable. Therefore, in an embodiment of the present application, the sampling ratio of the tissue is reduced instead of abandoning all the centroid points / symmetry points of the tissue. In this way, even when the centroid point / symmetry point resources of the tissue are scarce, accurate fitting of the midsagittal plane of the head can still be achieved.

[0047] According to an embodiment of the present application, a device for identifying the midsagittal plane of the head is also provided. Figure 3 FIG. 2 is a schematic diagram showing the partial components of a device for identifying the midsagittal plane of the head according to an embodiment of the present application. Figure 3As shown, the identification device 300 may at least include an interface 301 and a processor 302, wherein the interface 301 may be configured to receive a 3D medical image of the head, for example, wherein the 3D medical image of the head may be, for example, a plain scan CT image of the head, an enhanced scan CT image of the head, or a magnetic resonance imaging (MRI) image of the head, etc., and the present application does not impose any limitation on this.

[0048] In some embodiments, the processor 302 may be configured to execute the steps in the head midsagittal plane recognition method as described in various embodiments of the present application, and finally output the recognized head midsagittal plane. The processor 302 may be a processing device including one or more general-purpose processing devices, such as a microprocessor, a central processing unit (CPU), a graphics processing unit (GPU), etc. More specifically, the processor may be a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, a processor running other instruction sets, or a processor running a combination of instruction sets. The processor may also be one or more special-purpose processing devices, such as an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), a digital signal processor (DSP), a system on a chip (SoC), etc.

[0049] In some embodiments, the recognition device 300 may further include a display terminal 303, which may be configured to display the 3D medical image of the head acquired by the processor 302. In other embodiments, the display terminal 303 may be further configured to display the 3D medical image of the head, the segmentation of each core tissue and bilaterally symmetrical tissue, and the midsagittal plane of the head obtained by the processor 302, and the presentation method of each image in the display terminal 303 is not specifically limited in this application.

[0050] In some embodiments, the display terminal 303 may include a liquid crystal display (LCD), a light emitting diode display (LED), a plasma display, or any other type of display. In addition, a graphical user interface (GUI) may be presented on the display for user input and image / data display, etc., which will not be elaborated here.

[0051] According to an embodiment of the present application, a system for identifying the midsagittal plane of the head is also provided. Figure 4 A schematic diagram showing the composition of a system for identifying the midsagittal plane of the head according to an embodiment of the present application is shown.

[0052] like Figure 4As shown, the recognition system 400 includes a head 3D medical image imaging device 401 and a head midsagittal plane recognition device 402 as described in various embodiments of the present application. The head 3D medical image imaging device 401 may, for example, include a CT scanner with CT plain scan and enhanced CT functions, as well as a nuclear magnetic resonance imaging device, etc., which are not listed here one by one. The head midsagittal plane recognition device 402 has been described in detail in various embodiments of the present application, and will not be repeated here.

[0053] In some embodiments, the recognition system 400 may also include a non-transitory computer-readable storage medium (not shown), which may be used to store computer-executable instructions, and when the computer-executable instructions are executed by the processor, various operations of the head midsagittal plane recognition method described in various embodiments of the present application are implemented. In addition, the storage medium may also be used to store the head 3D medical image received from the head 3D medical image imaging device 401, and / or the point cloud set of centroid points / symmetric points obtained by segmenting and identifying each core tissue / left-right symmetrical tissue, etc., which are not listed here one by one.

[0054] In some embodiments, the above-mentioned non-transitory computer-readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a phase-change random access memory (PRAM), a static random access memory (SRAM), a dynamic random access memory (DRAM), an electrically erasable programmable read-only memory (EEPROM), other types of random access memory (RAM), a flash disk or other forms of flash memory, a cache, a register, a static memory, a compact disk read-only memory (CD-ROM), a digital versatile disk (DVD) or other optical storage, a cassette or other magnetic storage device, or any other possible non-transitory medium used to store information or instructions that can be accessed by a computer device.

[0055] According to the method, device and system for identifying the mid-sagittal plane of the head of the embodiment of the present application, the centroid point of the core tissue and the symmetric point of the bilaterally symmetrical tissue are fully utilized. The prior information that the centroid point of the core tissue and the symmetric point of the bilaterally symmetrical tissue should theoretically be located on the mid-sagittal plane, and the centroid point / symmetric point information of multiple core tissues / laterally symmetrical tissues is utilized, which not only makes the surface fitting have richer and more diverse spatial information, but also effectively avoids excessive dependence on the quality of the head 3D image and the quality of the single tissue segmentation. Moreover, in the iterative process of multiple rounds of surface fitting, some centroid points / symmetric points with poor quality that may be caused by congenital malformations or pathological changes can be screened out, thereby further reducing the influence of various unfavorable factors and ensuring that the optimal mid-sagittal plane of the head is always obtained by fitting the best set of centroid points / symmetric points. In addition, in the process of fitting the mid-sagittal plane of the head, the doctor user can also promote the efficient and accurate recognition of the mid-sagittal plane of the head according to his experience through interaction. Therefore, the present application can provide efficient, high-precision and robust mid-sagittal plane recognition of the head with good generalization ability for different recognition objects, can provide strong support for the diagnosis of related diseases and pathology, and has broad application prospects.

[0056] In addition, although exemplary embodiments have been described herein, the scope includes any and all embodiments based on the present application with equivalent elements, modifications, omissions, combinations (e.g., various embodiments intersecting schemes), adaptations or changes. The elements in the claims will be interpreted broadly based on the language adopted in the claims, and are not limited to the examples described in this specification or during the execution of the application, and the examples will be interpreted as non-exclusive. Therefore, this specification and examples are intended to be considered as examples only, and the true scope and spirit are indicated by the claims and the full scope of their equivalents.

[0057] The above description is intended to be illustrative rather than restrictive. For example, the above examples (or one or more schemes thereof) may be used in combination with each other. For example, a person of ordinary skill in the art may use other embodiments when reading the above description. In addition, in the above-mentioned specific embodiments, various features may be grouped together to simplify the present application. This should not be interpreted as an intention that a disclosed feature that is not required to be protected is necessary for any claim. On the contrary, the subject matter of the present application may be less than all the features of a particular disclosed embodiment. Thus, the claims are incorporated herein into the specific embodiments as examples or embodiments, wherein each claim is independently a separate embodiment, and it is considered that these embodiments may be combined with each other in various combinations or arrangements. The scope of the present application should be determined with reference to the claims and the full scope of equivalent forms granted by these claims.

[0058] The above embodiments are only exemplary embodiments of the present application and are not intended to limit the present application. The protection scope of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and protection scope of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the protection scope of the present application.

Claims

1. A method for identifying the midsagittal plane of the head, characterized in that: The identification method comprises, by a processor: Obtain 3D medical images of the head; Normalizing and resizing the acquired 3D head medical image; Selecting at least two core tissues and / or bilaterally symmetrical tissues from the preprocessed head 3D medical image as segmentation objects, wherein the core tissues have points on the midsagittal plane of the head as centroid points, and the bilaterally symmetrical tissues have points on the midsagittal plane of the head as symmetry points; performing segmentation processing on the selected core tissues / laterally symmetrical tissues based on the first learning network to generate 3D segmentation masks for the respective core tissues / laterally symmetrical tissues; Determine the centroid point of each core tissue in each image layer based on the 3D segmentation mask of each core tissue, determine the symmetric point of each bilaterally symmetric tissue in each image layer based on the 3D segmentation mask of each bilaterally symmetric tissue, and use the set of the centroid point of each core tissue / the symmetric point of the bilaterally symmetric tissue as the point cloud set of the corresponding tissue; Based on the point cloud set of each tissue, multiple rounds of surface fitting are performed using a surface fitting algorithm, so as to screen out the midsagittal plane of the head obtained based on the fitting of some centroid points / symmetric points.

2. The identification method according to claim 1, characterized in that: The point cloud set based on each tissue is subjected to multiple rounds of surface fitting using a surface fitting algorithm, thereby screening out the midsagittal plane of the head obtained based on partial centroid point / symmetric point fitting, further comprising: In each round of surface fitting: randomly sample the point cloud set of each tissue according to the first sampling ratio of this round to obtain a subset of centroid points / symmetric points; based on the subset of centroid points / symmetric points, use the surface fitting algorithm to fit to obtain the intermediate candidate surface of this round; based on the matching degree threshold range, verify the matching degree of other centroid points / symmetric points outside the subset of centroid points / symmetric points with the intermediate candidate surface of this round, and evaluate the fitting quality of the intermediate candidate surface of this round; The intermediate candidate surface with the highest fitting quality is selected as the final midsagittal plane of the head.

3. The identification method according to claim 2, characterized in that: The first sampling ratio of the first round of surface fitting is determined as follows: A first ratio of the number of centroid points / symmetric points in the point cloud set of each tissue to the total number of centroid points / symmetric points of all tissues is used as the sampling ratio value of the corresponding tissue in the first sampling ratio.

4. The identification method according to claim 2, characterized in that: The first sampling ratio of the first round of surface fitting is determined as follows: Calculate a first ratio of the number of centroid points / symmetric points in the point cloud set of each tissue to the total number of centroid points / symmetric points of all tissues, assign a corresponding proportional weight to each tissue based on the segmentation quality of each tissue, and determine the sampling proportion value of the corresponding tissue in the first sampling proportion based on the first ratio and the proportional weight.

5. The identification method according to claim 3 or 4, characterized in that: The first sampling ratio of this round is dynamically adjusted according to the verification result of the matching degree between other centroid points / symmetric points other than the subset of centroid points / symmetric points in the previous round and the intermediate candidate surface, including: When performing each round of verification, the centroid points / symmetric points other than the subset of centroid points / symmetric points and the centroid points / symmetric points whose distances from the intermediate candidate curved surface of the round exceed the matching degree threshold range are taken as the outer points of the intermediate candidate curved surface of the round; Based on the core tissue / left-right symmetrical tissue to which each external point belongs in the previous round of verification, the first sampling ratio of this round is adjusted, and the core tissue / left-right symmetrical tissue with a larger proportion of external points in the previous round of verification has a greater degree of reduction in the sampling ratio value in this round.

6. The identification method according to claim 3 or 4, characterized in that: The identification method further comprises: After each round of fitting, the intermediate candidate surface is obtained and presented to the user; In response to a user's interactive operation on the first sampling ratio of the current round, adjusting the first sampling ratio of the current round; Based on the adjusted first sampling ratio, the intermediate candidate surface fitting of this round is performed.

7. The identification method according to any one of claims 1 to 4, characterized in that: The centroid point of the selected core tissue / the symmetry point of the left and right symmetrical tissues have different depths in the front-to-back direction of the midsagittal plane of the head.

8. The identification method according to any one of claims 1 to 5, characterized in that: The surface fitting algorithm includes the least squares method.

9. A device for identifying the midsagittal plane of the head, characterized in that: include: An interface configured to receive a 3D medical image of the head; and A processor configured to execute the method for identifying the midsagittal plane of the head as described in any one of claims 1-8.

10. A system for identifying the midsagittal plane of the head, characterized in that: include: A head 3D medical imaging device and a head midsagittal plane recognition device as claimed in claim 9.

Images