CN115330808A - Segmentation-guided automatic measurement method for key parameters of spine of magnetic resonance image - Google Patents

Segmentation-guided automatic measurement method for key parameters of spine of magnetic resonance image Download PDF

Info

Publication number
CN115330808A
CN115330808A CN202210841597.7A CN202210841597A CN115330808A CN 115330808 A CN115330808 A CN 115330808A CN 202210841597 A CN202210841597 A CN 202210841597A CN 115330808 A CN115330808 A CN 115330808A
Authority
CN
China
Prior art keywords
segmentation
regression
features
guided
attention
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202210841597.7A
Other languages
Chinese (zh)
Other versions
CN115330808B (en
Inventor
庞树茂
庞春兰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangzhou Medical University
Original Assignee
Guangzhou Medical University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Guangzhou Medical University filed Critical Guangzhou Medical University
Priority to CN202210841597.7A priority Critical patent/CN115330808B/en
Publication of CN115330808A publication Critical patent/CN115330808A/en
Application granted granted Critical
Publication of CN115330808B publication Critical patent/CN115330808B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/10Segmentation; Edge detection
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10Complex mathematical operations
    • G06F17/18Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/10Image acquisition modality
    • G06T2207/10072Tomographic images
    • G06T2207/10088Magnetic resonance imaging [MRI]
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/20Special algorithmic details
    • G06T2207/20084Artificial neural networks [ANN]
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/30Subject of image; Context of image processing
    • G06T2207/30004Biomedical image processing
    • G06T2207/30008Bone
    • G06T2207/30012Spine; Backbone
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02ATECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/30Assessment of water resources

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Algebra (AREA)
  • Operations Research (AREA)
  • Probability & Statistics with Applications (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Magnetic Resonance Imaging Apparatus (AREA)

Abstract

The invention discloses a segmentation-guided automatic measurement method for spine key parameters of a magnetic resonance image, which comprises the following steps: a segmentation branch is formed by a segmentation encoder and a segmentation decoder; the segmentation branch is used for segmenting input image data to obtain a segmentation result and multi-scale segmentation characteristics; a regression branch is formed by a regression encoder and a full connection layer; wherein the regression encoder extracts attention-aware regression features from the multi-scale segmentation features through a segmentation-guided attention module; and the full connection layer is used for outputting the key parameters obtained by estimation. The invention improves the efficiency and the precision and can be widely applied to the technical field of image processing.

Description

Segmentation-guided automatic measurement method for key parameters of spine of magnetic resonance image
Technical Field
The invention relates to the technical field of image processing, in particular to a segmentation-guided automatic measuring method for key parameters of a spine of a magnetic resonance image.
Background
Spinal stenosis can be classified by location as central spinal stenosis, lateral crypt stenosis, and foraminal stenosis. Generally, intervertebral disc herniation, osteophyte formation, and changes in the vertebral joints such as ligamentum flavum are the main causes of spinal stenosis. These pathological changes often cause nervous symptoms such as low back and leg pain, stabbing pain, intermittent claudication and the like, and seriously affect the health and living standard of people. The location and extent of spinal stenosis can influence the choice of treatment regimen. The spinal column key parameters such as the two anterior and posterior diameters of the spinal canal, the width of the left intervertebral foramen, the width of the right intervertebral foramen and the like are measured in the transverse position magnetic resonance image, and the quantitative analysis basis can be provided for spinal stenosis diagnosis and intervertebral disc herniation grading.
The existing magnetic resonance image spine key parameter measuring methods can be divided into three types: segmentation-based methods, keypoint detection-based methods, direct measurement-based methods.
The segmentation-based method generally comprises two steps of spine structure segmentation and key point detection, which can be realized manually or automatically, and the manual measurement method has low clinical applicability due to the fact that the manual segmentation and the manual key point labeling are time-consuming and highly subjective. The spine structure is automatically segmented by a deep learning method, important boundaries of the spine structure are optimized by a curve evolution method, key points are detected by a computer through designed rules, spine key parameters are calculated according to key point coordinates, and the measurement precision can reach a sub-pixel level because the optimization process of the important boundaries is performed on a super-resolution image.
The method is commonly used for calculating the Cobb angle and is rare in measuring the key parameters of the spinal column of a transverse position magnetic resonance image.
The spine key parameter measurement is regarded as a regression problem based on a direct measurement method, and the measured spine key parameter value is directly output by a prediction model.
For the existing parameter measurement method, the segmentation-based method needs to optimize the important boundary of the spine structure by a curve evolution method, the process needs iterative operation and long calculation time (5.5 seconds), and the calculation efficiency needs to be improved.
The method based on key point detection cannot achieve sub-pixel-level measurement accuracy, the measurement accuracy is directly influenced by image spatial resolution and key point detection errors, and the measurement accuracy cannot meet clinical requirements.
The method based on direct measurement is easy to over-fit, and the measurement precision cannot meet the clinical requirement.
Disclosure of Invention
In view of this, the embodiment of the present invention provides an automatic measurement method for spine key parameters of a magnetic resonance image, which has high efficiency, high precision and segmentation guidance.
One aspect of the embodiments of the present invention provides a segmentation-guided automatic measurement method for spine key parameters of a magnetic resonance image, including:
a segmentation branch is formed by a segmentation encoder and a segmentation decoder; the segmentation branch is used for segmenting input image data to obtain a segmentation result and multi-scale segmentation characteristics;
forming a regression branch circuit by a regression coder and a full connection layer; wherein the regression encoder extracts attention-aware regression features from the multi-scale segmentation features through a segmentation-guided attention module; and the full connection layer is used for outputting the estimated key parameters.
Optionally, the segmentation-guided attention module comprises a channel attention module guided by segmentation and a spatial attention module guided by segmentation;
the method further comprises the following steps:
inputting the multi-scale segmentation features into a segmentation-guided channel attention module to obtain transformed segmentation features;
and outputting a channel attention diagram according to the segmentation characteristics.
Optionally, the method further comprises:
inputting the multi-scale segmentation features into a space attention module guided by segmentation to obtain a space attention diagram;
processing the regression features through a convolution layer and a batch normalization layer to obtain transformed regression features;
and obtaining the attention-perception regression feature according to the transformed regression feature, the channel attention diagram and the space attention diagram.
Optionally, after the multi-scale segmentation features are input into the segmentation-guided channel attention module, an expression of a process of obtaining transformed segmentation features is as follows:
Figure BDA0003751302840000021
wherein ,
Figure BDA0003751302840000022
representing the transformed segmentation features; i represents the ith scale; conv () represents a convolutional layer; BN () stands for batch normalization; relu () represents a linear rectification function.
Optionally, the expression of the process of outputting the channel attention map according to the segmentation features is as follows:
Figure BDA0003751302840000023
wherein ,
Figure BDA0003751302840000024
representing a channel attention map; represents the ith scale;
Figure BDA0003751302840000025
representing the transformed segmentation features; maxPook () represents max pooling; MLP () stands for multi-layer perceptron; avgPool () stands for average pooling; sigmoid () represents a sigmoid activation function.
Optionally, after the multi-scale segmentation feature is input into a spatial attention module guided by segmentation, an expression of a process of obtaining a spatial attention map is as follows:
Figure BDA0003751302840000031
wherein ,
Figure BDA0003751302840000032
representing a spatial attention map; i represents the ith scale;
Figure BDA0003751302840000033
representing the transformed segmentation features; maxPool () represents max pooling; conv () represents a convolutional layer; BN () stands for batch normalization; sigmoid () represents a sigmoid activation function.
Optionally, the expression of the regression feature after obtaining the transformation is:
Figure BDA0003751302840000034
the expression of the regression feature for obtaining attention perception is as follows:
Figure BDA0003751302840000035
wherein ,
Figure BDA0003751302840000036
representing processed regression features of the convolutional layer and the batch normalization layer; i represents the ith scale;
Figure BDA0003751302840000037
representing pre-processing regression features of the convolutional layer and the batch normalization layer; conv () represents a convolutional layer; BN () stands for batch normalization; relu () represents a linear rectification function;
Figure BDA0003751302840000038
represents a spatial attention map;
Figure BDA0003751302840000039
representing a channel attention map.
In another aspect, an embodiment of the present invention further provides an apparatus for automatically measuring a spine key parameter of a segmentation-guided magnetic resonance image, including:
a first module for forming a segmentation branch by a segmentation encoder and a segmentation decoder; the segmentation branch is used for segmenting input image data to obtain a segmentation result and multi-scale segmentation characteristics;
the second module is used for forming a regression branch circuit through the regression encoder and the full connection layer; wherein the regression encoder extracts attention-aware regression features from the multi-scale segmentation features through a segmentation-guided attention module; and the full connection layer is used for outputting the estimated key parameters.
Another aspect of the embodiments of the present invention further provides an electronic device, which includes a processor and a memory;
the memory is used for storing programs;
the processor executes the program to implement the method as described above.
Yet another aspect of the embodiments of the present invention provides a computer-readable storage medium, which stores a program, which is executed by a processor to implement the method as described above.
Embodiments of the present invention also disclose a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The computer instructions may be read by a processor of a computer device from a computer-readable storage medium, and the computer instructions executed by the processor cause the computer device to perform the foregoing method.
The embodiment of the invention forms a segmentation branch by a segmentation encoder and a segmentation decoder; the segmentation branch is used for segmenting input image data to obtain a segmentation result and multi-scale segmentation characteristics; forming a regression branch circuit by a regression coder and a full connection layer; wherein the regression encoder extracts attention-aware regression features from the multi-scale segmentation features through a segmentation-guided attention module; and the full connection layer is used for outputting the key parameters obtained by estimation. The invention improves the efficiency and the precision.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a block diagram of an overall method of an embodiment of the invention;
fig. 2 is a structural diagram of a segmentation-guided attention module according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
In view of the problems in the prior art, an aspect of the embodiments of the present invention provides a segmentation-guided automatic measurement method for key parameters of a spine in a magnetic resonance image, including:
forming a division branch by a division encoder and a division decoder; the segmentation branch is used for segmenting input image data to obtain a segmentation result and multi-scale segmentation characteristics;
forming a regression branch circuit by a regression coder and a full connection layer; wherein the regression encoder extracts attention-aware regression features from the multi-scale segmentation features through a segmentation-guided attention module; and the full connection layer is used for outputting the key parameters obtained by estimation.
Optionally, the segmentation-guided attention module comprises a channel attention module guided by segmentation and a spatial attention module guided by segmentation;
the method further comprises the following steps:
inputting the multi-scale segmentation features into a segmentation-guided channel attention module to obtain transformed segmentation features;
and outputting a channel attention diagram according to the segmentation characteristics.
Optionally, the method further comprises:
inputting the multi-scale segmentation features into a space attention module guided by segmentation to obtain a space attention diagram;
processing the regression features through a convolution layer and a batch normalization layer to obtain transformed regression features;
and obtaining the attention-perception regression feature according to the transformed regression feature, the channel attention diagram and the space attention diagram.
Optionally, after the multi-scale segmentation features are input into the segmentation-guided channel attention module, an expression of a process of obtaining transformed segmentation features is as follows:
Figure BDA0003751302840000051
wherein ,
Figure BDA0003751302840000052
representing the transformed segmentation features; i represents the ith scale; conv () represents a convolutional layer; BN () stands for batch normalization; relu () represents a linear rectification function.
Optionally, the expression of the process of outputting the channel attention map according to the segmentation features is as follows:
Figure BDA0003751302840000053
wherein ,
Figure BDA0003751302840000054
represents a channel attention map; i represents the ith scale;
Figure BDA0003751302840000055
representing the transformed segmentation features; maxPool () represents max pooling; MLP () stands for multilayer perceptron; avgPool () stands for average pooling; sigmoid () stands for sigmoid laserA live function.
Optionally, after the multi-scale segmentation feature is input into a spatial attention module guided by segmentation, an expression of a process of obtaining a spatial attention map is as follows:
Figure BDA0003751302840000056
wherein ,
Figure BDA0003751302840000057
representing a spatial attention map; i represents the ith scale;
Figure BDA0003751302840000058
representing the transformed segmentation features; maxPool () represents max pooling; conv () represents a convolutional layer; BN () stands for batch normalization; sigmoid () represents a sigmoid activation function.
Optionally, the expression of the regression feature after obtaining the transformation is:
Figure BDA0003751302840000059
the expression of the regression feature for obtaining attention perception is as follows:
Figure BDA00037513028400000510
wherein ,
Figure BDA0003751302840000061
representing processed regression features of the convolutional layer and the batch normalization layer; i represents the ith scale;
Figure BDA0003751302840000062
representing pre-processing regression features of the convolutional layer and the batch normalization layer; conv () represents a convolutional layer; BN () stands for batch normalization; relu () represents a linear rectification function;
Figure BDA0003751302840000063
representing a spatial attention map;
Figure BDA0003751302840000064
representing a channel attention map.
In another aspect, an embodiment of the present invention further provides an apparatus for automatically measuring a spine key parameter of a segmentation-guided magnetic resonance image, including:
a first module for forming a partition branch by a partition encoder and a partition decoder; the segmentation branch is used for segmenting input image data to obtain a segmentation result and multi-scale segmentation characteristics;
the second module is used for forming a regression branch circuit through the regression encoder and the full connection layer; wherein the regression encoder extracts attention-aware regression features from the multi-scale segmented features through a segmentation-guided attention module; and the full connection layer is used for outputting the estimated key parameters.
Another aspect of the embodiments of the present invention further provides an electronic device, including a processor and a memory;
the memory is used for storing programs;
the processor executes the program to implement the method as described above.
Yet another aspect of the embodiments of the present invention provides a computer-readable storage medium, which stores a program, which is executed by a processor to implement the method as described above.
Embodiments of the present invention also disclose a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The computer instructions may be read by a processor of a computer device from a computer-readable storage medium, and the computer instructions executed by the processor cause the computer device to perform the foregoing method.
The following describes in detail the implementation process of the automatic key parameter measurement method of the present invention with reference to the drawings of the specification:
aiming at the problems in the prior art, the segmentation-guided automatic measurement method for the key parameters of the spine of the magnetic resonance image provided by the invention can realize the segmentation of the spine structure and the automatic measurement of the key parameters of the spine at the same time, the inference stage does not need iteration, the running time is short (the running time on an RTX2080TI display card is less than 0.07 s), the measurement precision can reach the sub-pixel level (the average absolute error of the measurement of 4 key parameters of the spine is 0.49mm, wherein the image space resolution is 0.6875 multiplied by 0.6875 mm), and the application requirements of a spinal stenosis diagnosis system and an intervertebral disc herniation grading system can be met.
Specifically, the framework of the method of the present invention is shown in fig. 1, which is composed of a segmentation branch at the upper end and a regression branch at the lower end in the figure, the segmentation branch is composed of a segmentation encoder and a decoder, and the segmentation result is output after the input image passes through the segmentation branch
Figure BDA0003751302840000065
And multi-scale segmentation features
Figure BDA0003751302840000066
The regression branch is composed of a regression encoder and a full connectivity layer (FC), and under the guidance of multi-scale Segmentation features, the regression encoder extracts attention-aware regression features through a Segmentation-guided attention module (SGAM)
Figure BDA0003751302840000071
Figure BDA0003751302840000072
Finally, outputting the estimated key parameters through the full connection layer
Figure BDA0003751302840000073
Split-directed attention module (SGAM): the SGAM structure is shown in fig. 2, which consists of a segmentation-guided channel attention module and a segmentation-guided spatial attention module. Segmentation feature
Figure BDA0003751302840000074
After inputting the channel attention module of the segmentation guidance, obtaining the segmentation characteristics after transformation
Figure BDA0003751302840000075
Final output channel attention map
Figure BDA0003751302840000076
Can be expressed by the following formula:
Figure BDA0003751302840000077
Figure BDA0003751302840000078
Figure BDA0003751302840000079
after the space attention module guided by segmentation, a space attention diagram is obtained
Figure BDA00037513028400000710
Figure BDA00037513028400000711
Regression feature
Figure BDA00037513028400000712
Obtaining the transformed regression features through the convolution layer Conv and Batch Normalization (BN)
Figure BDA00037513028400000713
Figure BDA00037513028400000714
And finally, the mathematical expression of the regression characteristics of the attention perception is as follows:
Figure BDA00037513028400000715
in summary, compared with the prior art, the invention has the following two main characteristics:
1. segmentation-guided spine key parameter measurement framework: the invention provides a spine key parameter measuring frame guided by segmentation for the first time, can accurately and automatically measure spine key parameters of a transverse position magnetic resonance image, and provides a core algorithm and an effective tool for a spinal stenosis diagnosis system and an intervertebral disc herniation diagnosis system.
2. Segmentation-guided attention module: the invention provides an attention module for segmentation guidance for the first time, which can extract the regression characteristics of attention perception, so that the regression network focuses on the region related to the task, overfitting of the regression network is relieved, and the accuracy of spinal key parameter measurement and the interpretability of the method are improved.
The method does not need iteration in the reasoning stage, the running time of the algorithm is short (the running time on the RTX2080TI display card is less than 0.07 second), and the measurement precision of the spinal key parameters is high (the average measurement error is 0.49 mm).
In alternative embodiments, the functions/acts noted in the block diagrams may occur out of the order noted in the operational illustrations. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality/acts involved. Furthermore, the embodiments presented and described in the flow charts of the present invention are provided by way of example in order to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed, and in which sub-operations described as part of larger operations are performed independently.
Furthermore, although the present invention is described in the context of functional modules, it should be understood that, unless otherwise stated to the contrary, one or more of the described functions and/or features may be integrated in a single physical device and/or software module, or one or more functions and/or features may be implemented in a separate physical device or software module. It will also be appreciated that a detailed discussion of the actual implementation of each module is not necessary for an understanding of the present invention. Rather, the actual implementation of the various functional modules in the apparatus disclosed herein will be understood within the ordinary skill of an engineer, given the nature, function, and internal relationship of the modules. Accordingly, those skilled in the art can, using ordinary skill, practice the invention as set forth in the claims without undue experimentation. It is also to be understood that the specific concepts disclosed are merely illustrative of and not intended to limit the scope of the invention, which is defined by the appended claims and their full scope of equivalents.
The functions may be stored in a computer-readable storage medium if they are implemented in the form of software functional units and sold or used as separate products. Based on such understanding, the technical solution of the present invention or a part thereof which substantially contributes to the prior art may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Further, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
While embodiments of the invention have been shown and described, it will be understood by those of ordinary skill in the art that: various changes, modifications, substitutions and alterations can be made to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.
While the preferred embodiments of the present invention have been illustrated and described, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

Claims (10)

1. A segmentation-guided automatic measurement method for spine key parameters of a magnetic resonance image is characterized by comprising the following steps:
a segmentation branch is formed by a segmentation encoder and a segmentation decoder; the segmentation branch is used for segmenting input image data to obtain a segmentation result and multi-scale segmentation characteristics;
forming a regression branch circuit by a regression coder and a full connection layer; wherein the regression encoder extracts attention-aware regression features from the multi-scale segmented features through a segmentation-guided attention module; and the full connection layer is used for outputting the estimated key parameters.
2. The segmentation-guided automatic measurement method for spine key parameters of magnetic resonance images according to claim 1, wherein the segmentation-guided attention module comprises a segmentation-guided channel attention module and a segmentation-guided spatial attention module;
the method further comprises the following steps:
inputting the multi-scale segmentation features into a segmentation-guided channel attention module to obtain transformed segmentation features;
and outputting a channel attention diagram according to the segmentation characteristics.
3. A segmentation-guided automatic measurement method of spinal key parameters of magnetic resonance images as claimed in claim 2, further comprising:
inputting the multi-scale segmentation features into a space attention module guided by segmentation to obtain a space attention diagram;
processing the regression features by a convolution layer and a batch normalization layer to obtain transformed regression features;
and obtaining the attention-perception regression feature according to the transformed regression feature, the channel attention diagram and the space attention diagram.
4. The segmentation-guided automatic measurement method for key parameters of a spine of a magnetic resonance image according to claim 2, wherein after the multi-scale segmentation features are input into the segmentation-guided channel attention module, the expression of the process of obtaining the transformed segmentation features is as follows:
Figure FDA0003751302830000011
wherein ,
Figure FDA0003751302830000012
representing the transformed segmentation features; i represents the ith scale; comv () stands for convolutional layer; BN () stands for batch normalization; relu () represents a linear rectification function.
5. The segmentation-guided automatic measurement method for spine key parameters of magnetic resonance images according to claim 4, wherein the expression of the process of outputting channel attention maps according to the segmentation features is as follows:
Figure FDA0003751302830000013
wherein ,
Figure FDA0003751302830000014
representing a channel attention map; represents the ith scale;
Figure FDA0003751302830000015
representing the transformed segmentation features; maxPool () represents max pooling; MLP () stands for multilayer perceptron; avgPool () represents average pooling; sigmoid () represents a sigmoid activation function.
6. The segmentation-guided automatic measurement method for spinal key parameters of magnetic resonance images according to claim 3, wherein after the multi-scale segmentation features are input into a segmentation-guided spatial attention module, an expression of a process of obtaining a spatial attention map is as follows:
Figure FDA0003751302830000021
wherein ,
Figure FDA0003751302830000022
representing a spatial attention map; i represents the ith scale;
Figure FDA0003751302830000023
representing the transformed segmentation features; maxPool () represents max pooling; conv () represents a convolutional layer; BN () stands for batch normalization; sigmoid () represents a sigmoid activation function.
7. The segmentation-guided automatic measurement method for spine key parameters of magnetic resonance images according to claim 3,
the expression of the regression feature after transformation is obtained as follows:
Figure FDA0003751302830000024
the expression of the regression feature for obtaining attention perception is as follows:
Figure FDA0003751302830000025
wherein ,
Figure FDA0003751302830000026
representing processed regression features of the convolutional layer and the batch normalization layer; i represents the ith scale;
Figure FDA0003751302830000027
representing pre-processing regression features of the convolutional layers and batch normalization layers; conv () represents a convolutional layer; BN () stands for batch normalization; relu () represents a linear rectification function;
Figure FDA0003751302830000028
representing a spatial attention map;
Figure FDA0003751302830000029
representing a channel attention map.
8. A segmentation-guided automatic measuring device for key parameters of a spine of a magnetic resonance image is characterized by comprising:
a first module for forming a segmentation branch by a segmentation encoder and a segmentation decoder; the segmentation branch is used for segmenting input image data to obtain a segmentation result and multi-scale segmentation characteristics;
the second module is used for forming a regression branch circuit through the regression encoder and the full connection layer; wherein the regression encoder extracts attention-aware regression features from the multi-scale segmented features through a segmentation-guided attention module; and the full connection layer is used for outputting the estimated key parameters.
9. An electronic device comprising a processor and a memory;
the memory is used for storing programs;
the processor executing the program realizes the method of any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that the storage medium stores a program, which is executed by a processor to implement the method according to any one of claims 1 to 7.
CN202210841597.7A 2022-07-18 2022-07-18 Segmentation-guided magnetic resonance image spine key parameter automatic measurement method Active CN115330808B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210841597.7A CN115330808B (en) 2022-07-18 2022-07-18 Segmentation-guided magnetic resonance image spine key parameter automatic measurement method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210841597.7A CN115330808B (en) 2022-07-18 2022-07-18 Segmentation-guided magnetic resonance image spine key parameter automatic measurement method

Publications (2)

Publication Number Publication Date
CN115330808A true CN115330808A (en) 2022-11-11
CN115330808B CN115330808B (en) 2023-06-20

Family

ID=83917376

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210841597.7A Active CN115330808B (en) 2022-07-18 2022-07-18 Segmentation-guided magnetic resonance image spine key parameter automatic measurement method

Country Status (1)

Country Link
CN (1) CN115330808B (en)

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170193655A1 (en) * 2015-12-30 2017-07-06 Case Western Reserve University Radiomic features on diagnostic magnetic resonance enterography
CN111598030A (en) * 2020-05-21 2020-08-28 山东大学 Method and system for detecting and segmenting vehicle in aerial image
CN111612756A (en) * 2020-05-18 2020-09-01 中山大学 Coronary artery specificity calcification detection method and device
CN111671454A (en) * 2020-06-08 2020-09-18 平安科技(深圳)有限公司 Spine bending angle measuring method and device, computer equipment and storage medium
CN112132816A (en) * 2020-09-27 2020-12-25 北京理工大学 Target detection method based on multitask and region-of-interest segmentation guidance
CN112700444A (en) * 2021-02-19 2021-04-23 中国铁道科学研究院集团有限公司铁道建筑研究所 Bridge bolt detection method based on self-attention and central point regression model
WO2021147257A1 (en) * 2020-01-20 2021-07-29 上海商汤智能科技有限公司 Network training method and apparatus, image processing method and apparatus, and electronic device and storage medium
CN113205538A (en) * 2021-05-17 2021-08-03 广州大学 Blood vessel image segmentation method and device based on CRDNet
CN114581662A (en) * 2022-02-17 2022-06-03 华南理工大学 Method, system, device and storage medium for segmenting brain tumor image
CN114639102A (en) * 2022-05-11 2022-06-17 珠海横琴圣澳云智科技有限公司 Cell segmentation method and device based on key point and size regression

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170193655A1 (en) * 2015-12-30 2017-07-06 Case Western Reserve University Radiomic features on diagnostic magnetic resonance enterography
WO2021147257A1 (en) * 2020-01-20 2021-07-29 上海商汤智能科技有限公司 Network training method and apparatus, image processing method and apparatus, and electronic device and storage medium
CN111612756A (en) * 2020-05-18 2020-09-01 中山大学 Coronary artery specificity calcification detection method and device
CN111598030A (en) * 2020-05-21 2020-08-28 山东大学 Method and system for detecting and segmenting vehicle in aerial image
CN111671454A (en) * 2020-06-08 2020-09-18 平安科技(深圳)有限公司 Spine bending angle measuring method and device, computer equipment and storage medium
CN112132816A (en) * 2020-09-27 2020-12-25 北京理工大学 Target detection method based on multitask and region-of-interest segmentation guidance
CN112700444A (en) * 2021-02-19 2021-04-23 中国铁道科学研究院集团有限公司铁道建筑研究所 Bridge bolt detection method based on self-attention and central point regression model
CN113205538A (en) * 2021-05-17 2021-08-03 广州大学 Blood vessel image segmentation method and device based on CRDNet
CN114581662A (en) * 2022-02-17 2022-06-03 华南理工大学 Method, system, device and storage medium for segmenting brain tumor image
CN114639102A (en) * 2022-05-11 2022-06-17 珠海横琴圣澳云智科技有限公司 Cell segmentation method and device based on key point and size regression

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
ALA S. AL-KAFRI 等: "Boundary Delineation of MRI Images for Lumbar Spinal Stenosis Detection Through Semantic Segmentation Using Deep Neural Networks" *
CHUNLAN PANG 等: "Automated measurement of spine indices on axial MR images for lumbar spinal stenosis diagnosis using segmentation-guided regression network" *
FRISKA NATALIA 等: "Automated measurement of anteroposterior diameter and foraminal widths in MRI images for lumbar spinal stenosis diagnosis" *
郭璠 等: "YOLOv3-A:基于注意力机制的交通标志检测网络" *

Also Published As

Publication number Publication date
CN115330808B (en) 2023-06-20

Similar Documents

Publication Publication Date Title
Sobhaninia et al. Fetal ultrasound image segmentation for measuring biometric parameters using multi-task deep learning
Ortiz-Ramón et al. Identification of the presence of ischaemic stroke lesions by means of texture analysis on brain magnetic resonance images
US10366488B2 (en) Image processing used to estimate abnormalities
Zotin et al. Techniques for medical images processing using shearlet transform and color coding
Hsiao et al. A deep learning-based precision and automatic kidney segmentation system using efficient feature pyramid networks in computed tomography images
CN109767448B (en) Segmentation model training method and device
CN113298831B (en) Image segmentation method and device, electronic equipment and storage medium
US11257210B2 (en) Method and system of performing medical treatment outcome assessment or medical condition diagnostic
CN114219755B (en) Intelligent pulmonary tuberculosis detection method and system based on image and clinical data
CN112862830A (en) Multi-modal image segmentation method, system, terminal and readable storage medium
CN104732520A (en) Cardio-thoracic ratio measuring algorithm and system for chest digital image
CN116958217B (en) MRI and CT multi-mode 3D automatic registration method and device
CN118172614B (en) Ordered ankylosing spondylitis rating method based on supervised contrast learning
Shrestha et al. Early stage detection of scoliosis using machine learning algorithms
CN115809998A (en) Based on E 2 Glioma MRI data segmentation method based on C-Transformer network
CN115330817A (en) Pneumothorax lesion segmentation method based on Swin-Unet and morphological processing and related equipment
CN117495882B (en) Liver tumor CT image segmentation method based on AGCH-Net and multi-scale fusion
CN114972266A (en) Lymphoma ultrasonic image semantic segmentation method based on self-attention mechanism and stable learning
CN114359671A (en) Multi-target learning-based ultrasonic image thyroid nodule classification method and system
CN117853413A (en) Bile duct calculus identification system based on deep neural network
CN117274603A (en) Liver focus image description method based on semantic segmentation network
CN116664594A (en) Three-dimensional medical image two-stage segmentation method and device based on sharing CNN
CN115330808A (en) Segmentation-guided automatic measurement method for key parameters of spine of magnetic resonance image
CN116491968A (en) Acute ischemic stroke collateral circulation assessment method based on single-phase CTA
CN116740386A (en) Image processing method, apparatus, device and computer readable storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant