CN113793345A - Medical image segmentation method and device based on improved attention module - Google Patents
Medical image segmentation method and device based on improved attention module Download PDFInfo
- Publication number
- CN113793345A CN113793345A CN202111042489.5A CN202111042489A CN113793345A CN 113793345 A CN113793345 A CN 113793345A CN 202111042489 A CN202111042489 A CN 202111042489A CN 113793345 A CN113793345 A CN 113793345A
- Authority
- CN
- China
- Prior art keywords
- image
- target
- feature matrix
- attention
- neural network
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000003709 image segmentation Methods 0.000 title claims abstract description 33
- 239000011159 matrix material Substances 0.000 claims abstract description 163
- 238000003062 neural network model Methods 0.000 claims abstract description 71
- 238000010586 diagram Methods 0.000 claims abstract description 50
- 230000011218 segmentation Effects 0.000 claims abstract description 37
- 238000005070 sampling Methods 0.000 claims abstract description 29
- 238000003672 processing method Methods 0.000 claims abstract description 22
- 238000003860 storage Methods 0.000 claims abstract description 15
- 238000012545 processing Methods 0.000 claims abstract description 13
- 238000012549 training Methods 0.000 claims description 34
- 238000007781 pre-processing Methods 0.000 claims description 16
- 238000005481 NMR spectroscopy Methods 0.000 claims description 14
- 238000010606 normalization Methods 0.000 claims description 7
- 238000002591 computed tomography Methods 0.000 claims description 6
- 230000005540 biological transmission Effects 0.000 claims description 5
- 238000005520 cutting process Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 3
- 210000004556 brain Anatomy 0.000 description 12
- 230000006870 function Effects 0.000 description 8
- 230000007246 mechanism Effects 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 238000003745 diagnosis Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000012795 verification Methods 0.000 description 4
- 208000018737 Parkinson disease Diseases 0.000 description 3
- 230000004913 activation Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 3
- 230000035945 sensitivity Effects 0.000 description 3
- 239000011800 void material Substances 0.000 description 3
- 241001573498 Compacta Species 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000000586 desensitisation Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000003902 lesion Effects 0.000 description 2
- 238000002595 magnetic resonance imaging Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 210000003523 substantia nigra Anatomy 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000013519 translation Methods 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 241001633942 Dais Species 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000003190 augmentative effect Effects 0.000 description 1
- 238000004195 computer-aided diagnosis Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000002059 diagnostic imaging Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 208000030533 eye disease Diseases 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000002911 mydriatic effect Effects 0.000 description 1
- 238000013421 nuclear magnetic resonance imaging Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000002203 pretreatment Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10088—Magnetic resonance imaging [MRI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20132—Image cropping
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30016—Brain
-
- Y—GENERAL 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Epidemiology (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an image processing method, an image processing device, computer equipment and a storage medium. The method comprises the following steps: acquiring target tensor data of a target image; inputting the target tensor data into a target neural network model, wherein the target neural network model comprises an encoder, a decoder and an attention module; the encoder is used for performing down-sampling according to the target tensor data to obtain a first characteristic matrix; the attention module is used for carrying out cavity convolution according to the first feature matrix to obtain a second feature matrix; determining an original attention diagram according to the second feature matrix; determining a hole attention map according to the original attention map; determining a third feature matrix according to the first feature matrix and the hole attention diagram; a decoder for outputting the third feature matrix; the decoder is used for performing up-sampling according to the third characteristic matrix to obtain an image segmentation result; and performing segmentation feedback of the target image according to the output of the target neural network model. The accuracy of image segmentation can be improved.
Description
Technical Field
Embodiments of the present invention relate to machine learning technologies, and in particular, to an image processing method and apparatus, a computer device, and a storage medium.
Background
Medical image segmentation is an important task in computer-aided diagnosis, but the segmentation work is always challenging due to irregular shape of a target object, fuzzy boundary and the like. Therefore, accurate and reliable segmentation methods are needed for identifying these complex target objects of different scales.
With the development of the deep Convolutional Neural Networks (CNNs), U-Net type Neural Networks and a technical solution for fusing U-Net with attention mechanism are developed. However, the current U-Net combined with attention mechanism cannot accurately segment targets with different shapes and scales in medical images, and the accuracy of image segmentation is low.
Disclosure of Invention
The invention provides an image processing method, an image processing device, computer equipment and a storage medium, which are used for improving the accuracy of image segmentation.
In a first aspect, an embodiment of the present invention provides an image processing method, including:
acquiring target tensor data of a target image;
inputting the target tensor data into a target neural network model, wherein the target neural network model comprises an encoder, a decoder and an attention module;
the encoder is used for performing down-sampling according to the target tensor data to obtain a first characteristic matrix; the attention module is used for carrying out cavity convolution according to the first characteristic matrix to obtain a second characteristic matrix; determining an original attention diagram according to the second feature matrix; determining a hole attention map according to the original attention map; determining a third feature matrix according to the first feature matrix and the hole attention diagram; a decoder for outputting the third feature matrix; the decoder is used for performing up-sampling according to the third characteristic matrix to obtain an image segmentation result;
and performing segmentation feedback of the target image according to the output of the target neural network model.
In a second aspect, an embodiment of the present invention further provides an image processing apparatus, including:
the target image tensor acquisition module is used for acquiring target tensor data of a target image;
the segmentation module is used for inputting the target tensor data into a target neural network model, and the target neural network model comprises an encoder, a decoder and an attention module;
the encoder is used for performing down-sampling according to the target tensor data to obtain a first characteristic matrix; the attention module is used for carrying out cavity convolution according to the first characteristic matrix to obtain a second characteristic matrix; determining an original attention diagram according to the second feature matrix; determining a hole attention map according to the original attention map; determining a third feature matrix according to the first feature matrix and the hole attention diagram; a decoder for outputting the third feature matrix; the decoder is used for performing up-sampling according to the third characteristic matrix to obtain an image segmentation result;
and the output module is used for performing segmentation feedback on the target image according to the output of the target neural network model.
In a third aspect, an embodiment of the present invention further provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the image processing method according to the embodiment of the present invention when executing the computer program.
In a third aspect, embodiments of the present invention further provide a storage medium containing computer-executable instructions, which are used for executing the image processing method according to the embodiments of the present invention when executed by a computer processor.
The image processing method provided by the embodiment of the invention can acquire the target tensor data of the target image; inputting the target tensor data into a target neural network model, and performing down-sampling on an encoder of the target neural network model according to the target tensor data to obtain a first characteristic matrix; the attention module of the target neural network model is used for carrying out cavity convolution according to the first characteristic matrix to obtain a second characteristic matrix; determining an original attention diagram according to the second feature matrix; determining a hole attention map according to the original attention map; determining a third feature matrix according to the first feature matrix and the hole attention diagram; a decoder for outputting the third feature matrix; the decoder of the target neural network model is used for performing up-sampling according to the third characteristic matrix to obtain an image segmentation result; and performing segmentation feedback of the target image according to the output of the target neural network model. Compared with the problem that the accuracy of the U-Net image segmentation combined with the attention mechanism is low at present, according to the image processing method provided by the embodiment of the invention, the attention module of the target neural network model can process the first feature matrix of the target image based on the hole convolution, the hole convolution can improve the scope of the receptive field, further the region of the target image, which is interested by the user, can be determined more accurately, and the segmentation is carried out according to the region, so that the accuracy of the image segmentation is improved.
Drawings
FIG. 1 is a flowchart of an image processing method according to a first embodiment of the present invention;
FIG. 2 is a schematic structural diagram of an image processing apparatus according to a third embodiment of the present invention;
fig. 3 is a schematic structural diagram of a computer device in the fourth embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It is to be further noted that, for the convenience of description, only a part of the structure relating to the present invention is shown in the drawings, not the whole structure.
Example one
Fig. 1 is a flowchart of an image processing method according to an embodiment of the present invention, which is applicable to a case of segmenting a region of interest in a user in an image, in particular, a region of interest in a user such as a lesion in a medical image, and the method may be executed by a computer device providing an image segmentation function for the user, where the computer device may be a personal computer, a laptop computer, a tablet computer, or a server. The method specifically comprises the following steps:
and step 110, acquiring target tensor data of the target image.
The target image may be an image to be segmented, and the target tensor data is a target image expressed in a vector form. For example, the shape of the target tensor data may be 1 × 128 × 128, which represents 128 pixels in the horizontal direction and 128 pixels in the vertical direction in a certain color channel of the target image.
On the basis of the above embodiment, acquiring target tensor data of a target image includes:
preprocessing a target image according to a preset preprocessing mode to obtain a preprocessed image, wherein the preprocessing mode comprises an image cutting mode and an image normalization mode; target tensor data is determined from the preprocessed image.
The image cropping mode includes cropping images of different sizes into images of the same size. The image normalization mode comprises the step of normalizing the images with different formats into a unified file format. The preprocessed image with uniform specification can be obtained through a preprocessing mode. Because the number of pixels contained in the horizontal direction and the vertical direction of the preprocessed image is the same, and the number of color channels is the same, the data form of the target tensor obtained according to the preprocessed image is also the same. The preprocessed target tensor data are suitable for the target neural network model, and the segmentation accuracy of the target neural network model can be improved.
On the basis of the above embodiment, the target image is a medical image, and the medical image includes: images were taken using a computed tomography image, a nuclear magnetic resonance image, a high definition camera.
The high definition camera can be medical field high definition camera equipment. For example, a high definition camera may be a DRIVE eye disease dataset image acquisition tool. Specifically, images were acquired using a canon CR5 non-mydriatic 3CCD camera at 45 degree field of view. Each image was captured with 8 bits on a color plane of 768 x 584 pixels. The field of view of each image is circular and about 540 pixels in diameter.
On the basis of the above embodiment, acquiring target tensor data of a target image includes:
desensitizing the target image to obtain a desensitized image;
target tensor data for the target image is determined from the desensitized image.
In the embodiment, because the nuclear magnetic resonance scanning image related to the suspected patient of the clinical parkinson disease is desensitized data which is subjected to desensitization processing and has a format of nii.gz, information such as the name of the patient cannot be read, important privacy is not involved, and the information security of a user is ensured to the maximum extent.
And 120, inputting the target tensor data into the target neural network model.
Wherein the target neural network model includes an encoder, a decoder, and an attention module.
The encoder is used for performing down-sampling according to the target tensor data to obtain a first characteristic matrix; the attention module is used for carrying out cavity convolution according to the first characteristic matrix to obtain a second characteristic matrix; determining an original attention diagram according to the second feature matrix; determining a hole attention map according to the original attention map; determining a third feature matrix according to the first feature matrix and the hole attention diagram; a decoder for outputting the third feature matrix; and the decoder is used for performing up-sampling according to the third characteristic matrix to obtain an image segmentation result.
The encoder comprises a plurality of volume blocks. The convolutional layer of each convolutional block is followed by an activation function (ReLU) and a Batch Normalization (BN) layer, and then downsampled using a max pooling operation, which may be a downsampling step size of 2. After each down-sampling step, the number of feature channels will double. The first feature matrix is obtained by multiple donation blocks.
In one implementation, a data call model may be constructed from data features of clinical brain nmr images used for training. Specifically, the method comprises the following steps: account numbers and passwords with partial user rights are obtained from a PACS system of a hospital, and computer tomography images of suspected patients in the PACS system, namely Nii.gz, and corresponding label data are obtained through a communication protocol and stored.
During storage, the brain mri images may have different slice numbers according to the scan layer thickness and are stored for the patient's head. And performing down-sampling on the brain nuclear magnetic resonance scanning image through an encoder to obtain a first feature matrix.
After the first feature matrix is obtained, the first feature matrix is input to an attention module (also called AHA module). And optimizing the first feature matrix in the attention module, and outputting a third feature matrix.
The decoder and the encoder comprise a plurality of symmetrical rolling blocks, and the original image resolution is restored by up-sampling the third characteristic matrix through the decoder.
In one implementation, the attention module is configured to perform a hole convolution according to the first feature matrix to obtain a second feature matrix, and includes: and respectively performing hole convolution on the first characteristic matrix according to the hole convolutions to obtain a plurality of second characteristic matrices, wherein each hole convolution corresponds to one second characteristic matrix. Correspondingly, the original attention diagram is determined according to the second feature matrix, and the method comprises the following steps: performing mutual matrix multiplication according to the plurality of second feature matrices to obtain a matrix multiplication result; the original attention map is determined from the matrix multiplication result. Correspondingly, the method for determining the hole attention diagram according to the original attention diagram comprises the following steps: and carrying out transitive closure operation on the original attention diagram to obtain a hole attention diagram. Correspondingly, the third feature matrix is determined according to the first feature matrix and the hole attention map, and the third feature matrix comprises the following steps: and multiplying the first feature matrix by the hole attention map to obtain a third feature matrix.
In one implementation, the generation of the hole high order attention matrix is as follows: wherein, at the bottom of the coder, a characteristic diagram (also called a first characteristic matrix) is obtained from the previous stageWherein H × W × C are eachThe height, width, and number of channels of the feature map are expressed, and the feature map is sent to 4 shared weight convolution layers (r is 1, r is 2, r is 4, and r is 8, respectively), and a multi-scale feature map (also called a second feature matrix) X is generatedr=1,2,4,8. Carrying out scale transformation on the multi-scale characteristic diagram (also called a second characteristic matrix) by 1 multiplied by 1 x 1 convolution, and carrying out dot multiplication by the following formula I to obtain a preliminary attention matrix (also called an original attention diagram)
According to the theory of 'transitive closure' in the graph theory, the original attention diagram can be obtainedObtaining a high-order attention moment arrayThe calculation process is as follows:
first, an original attention matrix (also called a void attention map) after thresholding is obtained according to the following formula two:
where δ is a hyper-parameter threshold, and may be set to 0.5 in the embodiment of the present application.
Then, according to the transmission closure theory, the transmission closure calculation is carried out on the hole attention diagram, and according to the following formula III, a high-order attention moment array is obtained
Where M represents the high rank attention of the mth order. And weighting the first characteristic diagram through the high-order attention matrix to obtain a third characteristic matrix, thereby filtering the noise influence and strengthening the useful information.
The above-mentioned attention module can be embedded into any "U" -shaped encoder-decoder neural network model for use in medical image segmentation. In order to enlarge the limited reception fields of the partial convolution of the convolution layer in the down-sampling stage and the deconvolution layer in the up-sampling stage, more global information can be effectively captured in a high dimension by adopting multi-scale hole convolution. In addition, the high-order attention mechanism constructs a high-order attention feature graph through the transfer closure of the graph model so as to obtain stronger relevant context features.
Optionally, the segmentation algorithm used in the embodiment of the present invention may be a threshold segmentation algorithm, so that the influence of peripheral noise can be removed, the quality of the normalized image can be improved, and the robustness of the "U" shaped neural network model based on the improved attention module can be improved.
And step 130, performing segmentation feedback of the target image according to the output of the target neural network model.
On the basis of the above embodiment, the segmentation feedback of the target image according to the output of the target neural network model includes: and outputting the target image, and determining a highlight area of the target image according to the image segmentation result.
The user can import the target image through the computer equipment, and the computer equipment divides the target image according to the method and then displays the target image and the image division result. The segmented user regions of interest may be displayed using highlighting.
The image processing method provided by the embodiment of the invention can acquire the target tensor data of the target image; inputting the target tensor data into a target neural network model, and performing down-sampling on an encoder of the target neural network model according to the target tensor data to obtain a first characteristic matrix; the attention module of the target neural network model is used for carrying out cavity convolution according to the first characteristic matrix to obtain a second characteristic matrix; determining an original attention diagram according to the second feature matrix; determining a hole attention map according to the original attention map; determining a third feature matrix according to the first feature matrix and the hole attention diagram; a decoder for outputting the third feature matrix; the decoder of the target neural network model is used for performing up-sampling according to the third characteristic matrix to obtain an image segmentation result; and performing segmentation feedback of the target image according to the output of the target neural network model. Compared with the problem that the accuracy of the U-Net image segmentation combined with the attention mechanism is low at present, according to the image processing method provided by the embodiment of the invention, the attention module of the target neural network model can process the first feature matrix of the target image based on the hole convolution, the hole convolution can improve the scope of the receptive field, further the region of the target image, which is interested by the user, can be determined more accurately, and the segmentation is carried out according to the region, so that the accuracy of the image segmentation is improved.
On the basis of the above embodiment, before inputting the tensor data into the target neural network model, the method further includes:
acquiring training tensor data of a training image of a training set; and training the target neural network model according to the training tensor data so as to optimize the network parameters of the target neural network model.
The training images can be obtained by interfacing with a hospital image Archiving and Communication system (PACS), and can be expanded according to the images obtained by interfacing. The expansion mode comprises random turning, random rotation, random translation, random clipping and Gaussian noise addition. The target neural network model is trained by using the training tensor data obtained by expansion, so that the sensitivity of the target neural network model to unstable factors such as rotation, translation, Gaussian noise and the like can be improved, and the robustness of the U-shaped neural network model based on the improved attention module is further improved.
Example two
An embodiment of the present invention provides an image processing method, which is used to further describe the above embodiment, and includes:
and step E1, storing medical image information related to the suspected Parkinson patient. The medical imaging information includes brain magnetic resonance imaging.
In clinical practice, the head magnetic resonance imaging is standard imaging format data which is processed by desensitization and has a format of Nii.
And E2, preprocessing the target image according to a preset preprocessing mode to obtain a preprocessed image, and determining target tensor data according to the preprocessed image.
The pretreatment method comprises the following steps:
and step T1, obtaining the model of the detection machine corresponding to the brain nuclear magnetic resonance imaging, adjusting the nuclear magnetic resonance image to obtain a standard image to be detected, and normalizing to obtain a normalized image to be detected.
In this example, all magnetic resonance examinations were performed on a clinical radiology 3.0-T magnetic resonance scanner (diagnostics TM MR750 from GE healthcare) equipped with an 8-channel head matrix coil.
In step T2, since the pixel ratio of the region to be segmented is small, in particular, in this embodiment, the original image is subjected to center clipping, the middle region is taken for further processing, and the nuclear magnetic resonance image of the brain to be measured is converted into Tensor data as preprocessing data.
The shape of the Tensor data is 1 × 128 × 128.
And E3, training the neural network model as a medical image segmentation model.
Training the U-shaped neural network model based on the improved attention module comprises the following steps:
and step S1, constructing a data calling model according to the data characteristics of the clinical brain nuclear magnetic resonance image for training. Specifically, the method comprises the following steps:
the method comprises the steps of firstly obtaining an account number and a password with partial user authority from a PACS system of a hospital, obtaining a Nii.gz computed tomography image of a suspected patient in the PACS system and corresponding label data through a communication protocol, and storing the image.
In addition, during the storage process, the brain nuclear magnetic resonance scanning images can have different slice numbers according to the different scanning layer thicknesses and are stored according to the patient for the head.
Step S2, inputting the image into a down-sampling feature extraction network to obtain a first feature matrix;
and step S3, correcting the first feature matrix by using a cavity high-rank attention mechanism to obtain a second feature matrix.
Step S3 includes the following sub-steps: step S3-1, respectively performing cavity convolution on the first characteristic matrix according to the cavity convolutions to obtain a plurality of second characteristic matrices, wherein each cavity convolution corresponds to one of the second characteristic matrices, and step S3-2, performing mutual matrix multiplication according to the second characteristic matrices to obtain a matrix multiplication result; determining an original attention diagram according to a matrix multiplication result; step S2-3, carrying out transitive closure operation on the original attention map to obtain a hole attention map; and step S2-4, multiplying the first feature matrix by the hole attention map to obtain a third feature matrix.
And step S4, outputting the third feature matrix obtained in the previous step to an up-sampling segmentation network to obtain a segmentation result.
And step S4, training the U-shaped neural network model based on the improved attention module to obtain the trained U-shaped neural network model based on the improved attention module as the cavity attention segmentation neural network model.
In step S4-1, the original brain nmr image is preprocessed and converted into Tensor data as training data. Specifically, the method comprises the following steps:
the method comprises the steps of extracting a brain nuclear magnetic resonance scanning image pixel value, and carrying out normalization and pixel value truncation on the nuclear magnetic resonance image pixel, so that the influence of peripheral noise is removed, and the model robustness is improved. The middle brain image with length and width of 128 is retained using the center cropping method. And the pixel matrix is converted into the sensor Tensor data by using the sensor API in the PyTorch framework, so that the GPU can be used for parallel calculation conveniently.
And step S4-2, expanding the normalized image by a preset data expansion method to obtain an expanded image.
The data expansion method comprises random overturning, random rotating, random translating, random cutting and Gaussian noise adding.
Step S4-3, the augmented image and the tag data are input as the processing result to the "U" shaped neural network model based on the improved attention module.
In this embodiment, the setting conditions of the relevant parameters when training the void attention segmentation neural network model are as follows: the batch size is set to be 48, the epoch is 100, the gradient descent method is random gradient descent optimization (SGD), the learning rate adopts an Adam optimizer strategy, the initial value is 0.001, the momentum is 0.9, and the attenuation rate is 3E-5.
In step E4, the user-focused region is segmented from the target image (medical image) input by the user.
In this embodiment, the activation function is a commonly used activation function (ReLU), the user attention region is a key division region that requires model learning and division, and the user attention region is a Substantia Nigra Compacta (SNpc) in this embodiment.
And E5, displaying the nuclear magnetic resonance scanning image, displaying the model segmentation result in a red highlight area, and assisting a doctor in diagnosis.
All data obtained in this example were obtained by scanning with 3.0T NMR scanner (model: Discovery TM MR750, GE Healthcare). In total 188 cases of brain mri screening samples, specifically 140 cases of parkinson disease patients and 48 healthy people. The focus area of the Parkinson's disease is the substantia nigra pars compacta, so that the classification influence of Parkinson patients and healthy people does not need to be considered in the segmentation task, and all segmentation labels are independently labeled by professional doctors with eight years of clinical experience in the imaging department.
The data are divided into two independent sets, namely training for the algorithm model development process and other verification data for algorithm verification. The training set includes 152 subjects 'head nmr scan images, and the validation set includes 36 subjects' head nmr scan images.
Through verification by the method or the device, the Daiss coefficient (DSC) of the U-shaped neural network model based on the improved attention module on verification data is 0.8769, the Accuracy (ACC) is 0.9992, the Sensitivity (SE) is 0.8935, the Specificity (SP) is 0.9995, and the area under the operating characteristic curve (ROC) of the subject (AUC) is 0.9465.
Meanwhile, experiments show that the total time consumption of the test process of a batch of 32 images is about 3 seconds, the total time consumption of all tests of one patient is about 30 seconds, and the device has high time efficiency.
In summary, compared with the manual diagnosis, the medical image segmentation apparatus 1 based on the improved attention module and the medical image segmentation method based on the improved attention module of the present invention have high accuracy and high efficiency on the medical image related segmentation task, and approach the diagnosis level of the specialist, so that the present invention is particularly suitable for the situations of shortage of primary medical resources and insufficient number of specialists. In addition, the device can also effectively assist doctors to carry out rapid diagnosis, and is beneficial to improving the diagnosis and treatment efficiency of the links of the department of imaging in hospitals.
EXAMPLE III
Fig. 2 is a schematic structural diagram of an image processing apparatus according to a third embodiment of the present invention, which is applicable to a case where a region of interest in a user in an image is segmented, especially a case where a region of interest in a user, such as a lesion in a medical image, is segmented, where the method may be executed by a computer device providing an image segmentation function for the user, and the computer device may be a personal computer, a laptop, a tablet computer, or a server. The device includes: a target image tensor acquisition module 210, a segmentation module 220, and an output module 230.
A target image tensor acquisition module 210, configured to acquire target tensor data of a target image;
a segmentation module 220 for inputting the target tensor data to a target neural network model, the target neural network model including an encoder, a decoder and an attention module;
the encoder is used for performing down-sampling according to the target tensor data to obtain a first characteristic matrix; the attention module is used for carrying out cavity convolution according to the first characteristic matrix to obtain a second characteristic matrix; determining an original attention diagram according to the second feature matrix; determining a hole attention map according to the original attention map; determining a third feature matrix according to the first feature matrix and the hole attention diagram; a decoder for outputting the third feature matrix; the decoder is used for performing up-sampling according to the third characteristic matrix to obtain an image segmentation result;
and an output module 230, configured to perform segmentation feedback on the target image according to the output of the target neural network model.
On the basis of the above embodiment, the segmentation module 220 is configured to:
respectively performing hole convolution on the first feature matrix according to the hole convolutions to obtain a plurality of second feature matrices, wherein each hole convolution corresponds to one second feature matrix;
performing mutual matrix multiplication according to the plurality of second feature matrices to obtain a matrix multiplication result; determining an original attention diagram according to a matrix multiplication result;
carrying out transmission closure operation on the original attention diagram to obtain a hole attention diagram;
and multiplying the first feature matrix by the hole attention map to obtain a third feature matrix.
On the basis of the foregoing embodiment, the target image tensor acquisition module 210 is configured to:
preprocessing a target image according to a preset preprocessing mode to obtain a preprocessed image, wherein the preprocessing mode comprises an image cutting mode and an image normalization mode;
target tensor data is determined from the preprocessed image.
In the above embodiment, the target image is a medical image, and the medical image includes: images were taken using a computed tomography image, a nuclear magnetic resonance image, a high definition camera.
On the basis of the above embodiment, the output module 230 is configured to:
and outputting the target image, and determining a highlight area of the target image according to the image segmentation result.
On the basis of the foregoing embodiment, the acquired target image tensor acquisition module 210 is configured to:
desensitizing the target image to obtain a desensitized image;
target tensor data for the target image is determined from the desensitized image.
On the basis of the above embodiment, the training device further comprises a training module. The training module is used for:
acquiring training tensor data of a training image of a training set;
and training the target neural network model according to the training tensor data so as to optimize the network parameters of the target neural network model.
In the image processing apparatus provided in the embodiment of the present invention, the target image tensor acquisition module 210 can acquire target tensor data of a target image; the segmentation module 220 inputs the target tensor data into the target neural network model, and an encoder of the target neural network model performs down-sampling according to the target tensor data to obtain a first feature matrix; the attention module of the target neural network model is used for carrying out cavity convolution according to the first characteristic matrix to obtain a second characteristic matrix; determining an original attention diagram according to the second feature matrix; determining a hole attention map according to the original attention map; determining a third feature matrix according to the first feature matrix and the hole attention diagram; a decoder for outputting the third feature matrix; the decoder of the target neural network model is used for performing up-sampling according to the third characteristic matrix to obtain an image segmentation result; the output module 230 performs segmentation feedback of the target image according to the output of the target neural network model. Compared with the problem of low accuracy of the U-Net image segmentation combined with the attention mechanism at present, according to the image processing method provided by the embodiment of the invention, the attention module of the target neural network model can process the first feature matrix of the target image based on the hole convolution, the hole convolution can improve the scope of the receptive field, further, the region of interest of a user in the target image can be more accurately determined, and the segmentation is carried out according to the region, so that the accuracy of the image segmentation is improved.
The image processing device provided by the embodiment of the invention can execute the image processing method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.
Example four
Fig. 3 is a schematic structural diagram of a computer apparatus according to a fourth embodiment of the present invention, as shown in fig. 3, the computer apparatus includes a processor 30, a memory 31, an input device 32, and an output device 33; the number of processors 30 in the computer device may be one or more, and one processor 30 is taken as an example in fig. 3; the processor 30, the memory 31, the input device 32 and the output device 33 in the computer apparatus may be connected by a bus or other means, and the connection by the bus is exemplified in fig. 3.
The memory 31 is used as a computer-readable storage medium for storing software programs, computer-executable programs, and modules, such as program instructions/modules corresponding to the image processing method in the embodiment of the present invention (for example, the target image tensor acquisition module 210, the segmentation module 220, and the output module 230 in the image processing apparatus). The processor 30 executes various functional applications of the computer device and data processing by executing software programs, instructions and modules stored in the memory 31, that is, implements the image processing method described above.
The memory 31 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal, and the like. Further, the memory 31 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, the memory 31 may further include memory remotely located from the processor 30, which may be connected to a computer device over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input device 32 may be used to receive input numeric or character information and to generate key signal inputs relating to user settings and function controls of the computer apparatus. The output device 33 may include a display device such as a display screen.
EXAMPLE five
Embodiments of the present invention also provide a storage medium containing computer-executable instructions, which when executed by a computer processor, perform an image processing method, the method comprising:
acquiring target tensor data of a target image;
inputting the target tensor data into a target neural network model, wherein the target neural network model comprises an encoder, a decoder and an attention module;
the encoder is used for performing down-sampling according to the target tensor data to obtain a first characteristic matrix; the attention module is used for carrying out cavity convolution according to the first characteristic matrix to obtain a second characteristic matrix; determining an original attention diagram according to the second feature matrix; determining a hole attention map according to the original attention map; determining a third feature matrix according to the first feature matrix and the hole attention diagram; a decoder for outputting the third feature matrix; the decoder is used for performing up-sampling according to the third characteristic matrix to obtain an image segmentation result;
and performing segmentation feedback of the target image according to the output of the target neural network model.
On the basis of the foregoing embodiment, the attention module is configured to perform void convolution according to the first feature matrix to obtain a second feature matrix, and includes:
respectively performing hole convolution on the first feature matrix according to the hole convolutions to obtain a plurality of second feature matrices, wherein each hole convolution corresponds to one second feature matrix;
correspondingly, the original attention diagram is determined according to the second feature matrix, and the method comprises the following steps:
performing mutual matrix multiplication according to the plurality of second feature matrices to obtain a matrix multiplication result; determining an original attention diagram according to a matrix multiplication result;
correspondingly, the method for determining the hole attention diagram according to the original attention diagram comprises the following steps:
carrying out transmission closure operation on the original attention diagram to obtain a hole attention diagram;
correspondingly, the third feature matrix is determined according to the first feature matrix and the hole attention map, and the third feature matrix comprises the following steps:
and multiplying the first feature matrix by the hole attention map to obtain a third feature matrix.
In addition to the above embodiments, the acquiring of the target tensor data of the target image includes:
preprocessing a target image according to a preset preprocessing mode to obtain a preprocessed image, wherein the preprocessing mode comprises an image cutting mode and an image normalization mode;
target tensor data is determined from the preprocessed image.
In the above embodiment, the target image is a medical image, and the medical image includes: images were taken using a computed tomography image, a nuclear magnetic resonance image, a high definition camera.
In addition to the above embodiments, the method for performing segmentation feedback of a target image according to an output of a target neural network model includes:
and outputting the target image, and determining a highlight area of the target image according to the image segmentation result.
In addition to the above embodiments, the acquiring of the target tensor data of the target image includes:
desensitizing the target image to obtain a desensitized image;
target tensor data for the target image is determined from the desensitized image.
In the above embodiment, before the tensor data is input to the target neural network model, the method further includes:
acquiring training tensor data of a training image of a training set;
and training the target neural network model according to the training tensor data so as to optimize the network parameters of the target neural network model.
Of course, the storage medium containing the computer-executable instructions provided by the embodiments of the present invention is not limited to the method operations described above, and may also perform related operations in the image processing method provided by any embodiment of the present invention.
From the above description of the embodiments, it is obvious for a person skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly by hardware, but the former is a better embodiment in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute the methods according to the embodiments of the present invention.
It should be noted that, in the embodiment of the above search apparatus, the included units and modules are only divided according to functional logic, but are not limited to the above division as long as the corresponding functions can be implemented; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions without departing from the scope of the invention. Therefore, although the present invention has been described in more detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. An image processing method, comprising:
acquiring target tensor data of a target image;
inputting the target tensor data to a target neural network model, the target neural network model comprising an encoder, a decoder, and an attention module;
the encoder is used for performing down-sampling according to the target tensor data to obtain a first feature matrix; the attention module is used for carrying out cavity convolution according to the first feature matrix to obtain a second feature matrix; determining an original attention diagram according to the second feature matrix; determining a hole attention map from the original attention map; determining a third feature matrix according to the first feature matrix and the hole attention map; a decoder for decoding the third feature matrix output values; the decoder is used for performing up-sampling according to the third characteristic matrix to obtain an image segmentation result;
and performing segmentation feedback of the target image according to the output of the target neural network model.
2. The method of claim 1, wherein the attention module is configured to perform a hole convolution according to the first feature matrix to obtain a second feature matrix, and the method comprises:
performing hole convolution on the first feature matrix according to a plurality of hole convolutions respectively to obtain a plurality of second feature matrices, wherein each hole convolution corresponds to one second feature matrix;
correspondingly, the determining an original attention diagram according to the second feature matrix includes:
performing mutual matrix multiplication according to the plurality of second feature matrices to obtain a matrix multiplication result; determining an original attention diagram according to the matrix multiplication result;
correspondingly, the determining a hole attention map according to the original attention map includes:
carrying out a transmission closure operation on the original attention diagram to obtain a hole attention diagram;
correspondingly, the determining a third feature matrix according to the first feature matrix and the hole attention map includes:
and multiplying the first feature matrix and the hole attention diagram to obtain a third feature matrix.
3. The method of claim 1, wherein obtaining target tensor data for a target image comprises:
preprocessing the target image according to a preset preprocessing mode to obtain a preprocessed image, wherein the preprocessing mode comprises an image cutting mode and an image normalization mode;
and determining target tensor data according to the preprocessed image.
4. The method of claim 1, wherein the target image is a medical image comprising: images were taken using a computed tomography image, a nuclear magnetic resonance image, a high definition camera.
5. The method of claim 1, wherein the feedback of the segmentation of the target image from the output of the target neural network model comprises:
and outputting the target image, and determining a highlight area of the target image according to the image segmentation result.
6. The method of claim 1, wherein obtaining target tensor data for a target image comprises:
desensitizing the target image to obtain a desensitized image;
determining target tensor data for the target image from the desensitized image.
7. The method of claim 1, further comprising, prior to inputting the tensor data to a target neural network model:
acquiring training tensor data of a training image of a training set;
and training the target neural network model according to the training tensor data so as to optimize network parameters of the target neural network model.
8. An image processing apparatus characterized by comprising:
the target image tensor acquisition module is used for acquiring target tensor data of a target image;
a segmentation module for inputting the target tensor data to a target neural network model, the target neural network model including an encoder, a decoder, and an attention module;
the encoder is used for performing down-sampling according to the target tensor data to obtain a first feature matrix; the attention module is used for carrying out cavity convolution according to the first feature matrix to obtain a second feature matrix; determining an original attention diagram according to the second feature matrix; determining a hole attention map from the original attention map; determining a third feature matrix according to the first feature matrix and the hole attention map; a decoder for decoding the third feature matrix output values; the decoder is used for performing up-sampling according to the third characteristic matrix to obtain an image segmentation result;
and the output module is used for performing segmentation feedback on the target image according to the output of the target neural network model.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the image processing method according to any of claims 1-7 when executing the program.
10. A storage medium containing computer-executable instructions for performing the image processing method of any one of claims 1-7 when executed by a computer processor.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111042489.5A CN113793345B (en) | 2021-09-07 | 2021-09-07 | Medical image segmentation method and device based on improved attention module |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111042489.5A CN113793345B (en) | 2021-09-07 | 2021-09-07 | Medical image segmentation method and device based on improved attention module |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113793345A true CN113793345A (en) | 2021-12-14 |
CN113793345B CN113793345B (en) | 2023-10-31 |
Family
ID=78879628
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111042489.5A Active CN113793345B (en) | 2021-09-07 | 2021-09-07 | Medical image segmentation method and device based on improved attention module |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113793345B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111681273A (en) * | 2020-06-10 | 2020-09-18 | 创新奇智(青岛)科技有限公司 | Image segmentation method and device, electronic equipment and readable storage medium |
CN114387436A (en) * | 2021-12-28 | 2022-04-22 | 北京安德医智科技有限公司 | Wall coronary artery detection method and device, electronic device and storage medium |
CN114565628A (en) * | 2022-03-23 | 2022-05-31 | 中南大学 | Image segmentation method and system based on boundary perception attention |
CN115731243A (en) * | 2022-11-29 | 2023-03-03 | 北京长木谷医疗科技有限公司 | Spine image segmentation method and device based on artificial intelligence and attention mechanism |
Citations (22)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080162036A1 (en) * | 1997-10-22 | 2008-07-03 | Intelligent Technologies International, Inc. | Vehicular Map Database Management Techniques |
WO2018036293A1 (en) * | 2016-08-26 | 2018-03-01 | 杭州海康威视数字技术股份有限公司 | Image segmentation method, apparatus, and fully convolutional network system |
CN109191472A (en) * | 2018-08-28 | 2019-01-11 | 杭州电子科技大学 | Based on the thymocyte image partition method for improving U-Net network |
CN109685813A (en) * | 2018-12-27 | 2019-04-26 | 江西理工大学 | A kind of U-shaped Segmentation Method of Retinal Blood Vessels of adaptive scale information |
CN110059586A (en) * | 2019-03-29 | 2019-07-26 | 电子科技大学 | A kind of Iris Location segmenting system based on empty residual error attention structure |
WO2019153908A1 (en) * | 2018-02-11 | 2019-08-15 | 北京达佳互联信息技术有限公司 | Image recognition method and system based on attention model |
US20190355102A1 (en) * | 2018-05-15 | 2019-11-21 | Adobe Inc. | Digital Image Completion by Learning Generation and Patch Matching Jointly |
US20190362199A1 (en) * | 2018-05-25 | 2019-11-28 | Adobe Inc. | Joint blur map estimation and blur desirability classification from an image |
US20200184278A1 (en) * | 2014-03-18 | 2020-06-11 | Z Advanced Computing, Inc. | System and Method for Extremely Efficient Image and Pattern Recognition and Artificial Intelligence Platform |
CN111612807A (en) * | 2020-05-15 | 2020-09-01 | 北京工业大学 | Small target image segmentation method based on scale and edge information |
WO2020177651A1 (en) * | 2019-03-01 | 2020-09-10 | 华为技术有限公司 | Image segmentation method and image processing device |
WO2020215985A1 (en) * | 2019-04-22 | 2020-10-29 | 腾讯科技(深圳)有限公司 | Medical image segmentation method and device, electronic device and storage medium |
CN111898709A (en) * | 2020-09-30 | 2020-11-06 | 中国人民解放军国防科技大学 | Image classification method and device |
CN112017191A (en) * | 2020-08-12 | 2020-12-01 | 西北大学 | Method for establishing and segmenting liver pathology image segmentation model based on attention mechanism |
CN112017198A (en) * | 2020-10-16 | 2020-12-01 | 湖南师范大学 | Right ventricle segmentation method and device based on self-attention mechanism multi-scale features |
WO2020244108A1 (en) * | 2019-06-05 | 2020-12-10 | Boe Technology Group Co., Ltd. | Methods and apparatuses for semantically segmenting input image, and computer-program product |
CN112101363A (en) * | 2020-09-02 | 2020-12-18 | 河海大学 | Full convolution semantic segmentation system and method based on cavity residual error and attention mechanism |
WO2021031066A1 (en) * | 2019-08-19 | 2021-02-25 | 中国科学院深圳先进技术研究院 | Cartilage image segmentation method and apparatus, readable storage medium, and terminal device |
CN112541918A (en) * | 2020-12-23 | 2021-03-23 | 山东师范大学 | Three-dimensional medical image segmentation method based on self-attention mechanism neural network |
US20210089807A1 (en) * | 2019-09-25 | 2021-03-25 | Samsung Electronics Co., Ltd. | System and method for boundary aware semantic segmentation |
CN112750142A (en) * | 2021-01-26 | 2021-05-04 | 上海工程技术大学 | Ultrasonic image segmentation system and method based on side window attention mechanism |
GB202104506D0 (en) * | 2021-03-30 | 2021-05-12 | Ucl Business Plc | Medical Image Analysis Using Neural Networks |
-
2021
- 2021-09-07 CN CN202111042489.5A patent/CN113793345B/en active Active
Patent Citations (22)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080162036A1 (en) * | 1997-10-22 | 2008-07-03 | Intelligent Technologies International, Inc. | Vehicular Map Database Management Techniques |
US20200184278A1 (en) * | 2014-03-18 | 2020-06-11 | Z Advanced Computing, Inc. | System and Method for Extremely Efficient Image and Pattern Recognition and Artificial Intelligence Platform |
WO2018036293A1 (en) * | 2016-08-26 | 2018-03-01 | 杭州海康威视数字技术股份有限公司 | Image segmentation method, apparatus, and fully convolutional network system |
WO2019153908A1 (en) * | 2018-02-11 | 2019-08-15 | 北京达佳互联信息技术有限公司 | Image recognition method and system based on attention model |
US20190355102A1 (en) * | 2018-05-15 | 2019-11-21 | Adobe Inc. | Digital Image Completion by Learning Generation and Patch Matching Jointly |
US20190362199A1 (en) * | 2018-05-25 | 2019-11-28 | Adobe Inc. | Joint blur map estimation and blur desirability classification from an image |
CN109191472A (en) * | 2018-08-28 | 2019-01-11 | 杭州电子科技大学 | Based on the thymocyte image partition method for improving U-Net network |
CN109685813A (en) * | 2018-12-27 | 2019-04-26 | 江西理工大学 | A kind of U-shaped Segmentation Method of Retinal Blood Vessels of adaptive scale information |
WO2020177651A1 (en) * | 2019-03-01 | 2020-09-10 | 华为技术有限公司 | Image segmentation method and image processing device |
CN110059586A (en) * | 2019-03-29 | 2019-07-26 | 电子科技大学 | A kind of Iris Location segmenting system based on empty residual error attention structure |
WO2020215985A1 (en) * | 2019-04-22 | 2020-10-29 | 腾讯科技(深圳)有限公司 | Medical image segmentation method and device, electronic device and storage medium |
WO2020244108A1 (en) * | 2019-06-05 | 2020-12-10 | Boe Technology Group Co., Ltd. | Methods and apparatuses for semantically segmenting input image, and computer-program product |
WO2021031066A1 (en) * | 2019-08-19 | 2021-02-25 | 中国科学院深圳先进技术研究院 | Cartilage image segmentation method and apparatus, readable storage medium, and terminal device |
US20210089807A1 (en) * | 2019-09-25 | 2021-03-25 | Samsung Electronics Co., Ltd. | System and method for boundary aware semantic segmentation |
CN111612807A (en) * | 2020-05-15 | 2020-09-01 | 北京工业大学 | Small target image segmentation method based on scale and edge information |
CN112017191A (en) * | 2020-08-12 | 2020-12-01 | 西北大学 | Method for establishing and segmenting liver pathology image segmentation model based on attention mechanism |
CN112101363A (en) * | 2020-09-02 | 2020-12-18 | 河海大学 | Full convolution semantic segmentation system and method based on cavity residual error and attention mechanism |
CN111898709A (en) * | 2020-09-30 | 2020-11-06 | 中国人民解放军国防科技大学 | Image classification method and device |
CN112017198A (en) * | 2020-10-16 | 2020-12-01 | 湖南师范大学 | Right ventricle segmentation method and device based on self-attention mechanism multi-scale features |
CN112541918A (en) * | 2020-12-23 | 2021-03-23 | 山东师范大学 | Three-dimensional medical image segmentation method based on self-attention mechanism neural network |
CN112750142A (en) * | 2021-01-26 | 2021-05-04 | 上海工程技术大学 | Ultrasonic image segmentation system and method based on side window attention mechanism |
GB202104506D0 (en) * | 2021-03-30 | 2021-05-12 | Ucl Business Plc | Medical Image Analysis Using Neural Networks |
Non-Patent Citations (3)
Title |
---|
P. GUO, X. SU, H. ZHANG, M. WANG AND F. BAO: "A Multi-Scaled Receptive Field Learning Approach for Medical Image Segmentation", 《ICASSP 2020 - 2020 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP)》 * |
X. MAO ET AL.: "Deep Learning with Skip Connection Attention for Choroid Layer Segmentation in OCT Images", 《2020 42ND ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE & BIOLOGY SOCIETY (EMBC)》 * |
赵引: "基于多尺度密集网络与两级残差注意力的乳腺肿瘤分割方法研究", 《CNKI优秀硕士学位论文全文库》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111681273A (en) * | 2020-06-10 | 2020-09-18 | 创新奇智(青岛)科技有限公司 | Image segmentation method and device, electronic equipment and readable storage medium |
CN111681273B (en) * | 2020-06-10 | 2023-02-03 | 创新奇智(青岛)科技有限公司 | Image segmentation method and device, electronic equipment and readable storage medium |
CN114387436A (en) * | 2021-12-28 | 2022-04-22 | 北京安德医智科技有限公司 | Wall coronary artery detection method and device, electronic device and storage medium |
CN114565628A (en) * | 2022-03-23 | 2022-05-31 | 中南大学 | Image segmentation method and system based on boundary perception attention |
CN114565628B (en) * | 2022-03-23 | 2022-09-13 | 中南大学 | Image segmentation method and system based on boundary perception attention |
CN115731243A (en) * | 2022-11-29 | 2023-03-03 | 北京长木谷医疗科技有限公司 | Spine image segmentation method and device based on artificial intelligence and attention mechanism |
CN115731243B (en) * | 2022-11-29 | 2024-02-09 | 北京长木谷医疗科技股份有限公司 | Spine image segmentation method and device based on artificial intelligence and attention mechanism |
Also Published As
Publication number | Publication date |
---|---|
CN113793345B (en) | 2023-10-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113793345B (en) | Medical image segmentation method and device based on improved attention module | |
KR101857624B1 (en) | Medical diagnosis method applied clinical information and apparatus using the same | |
CN110853111B (en) | Medical image processing system, model training method and training device | |
RU2667879C1 (en) | Processing and analysis of data on computer-assisted tomography images | |
CN110692107B (en) | Machine learning of raw medical imaging data for clinical decision support | |
EP4089640A1 (en) | Image reconstruction method, apparatus, device, system, and computer readable storage medium | |
Yao et al. | Pneumonia Detection Using an Improved Algorithm Based on Faster R‐CNN | |
Menaouer et al. | Diabetic retinopathy classification using hybrid deep learning approach | |
Heydarheydari et al. | Auto-segmentation of head and neck tumors in positron emission tomography images using non-local means and morphological frameworks | |
Leite et al. | Etiology-based classification of brain white matter hyperintensity on magnetic resonance imaging | |
CN116862877A (en) | Scanning image analysis system and method based on convolutional neural network | |
CN112508884A (en) | Comprehensive detection device and method for cancerous region | |
CN113744271A (en) | Neural network-based automatic optic nerve segmentation and compression degree measurement and calculation method | |
CN112633404A (en) | DenseNet-based CT image classification method and device for COVID-19 patient | |
CN113240655A (en) | Method, storage medium and device for automatically detecting type of fundus image | |
CN114066798B (en) | Brain tumor nuclear magnetic resonance image data synthesis method based on deep learning | |
CN113421274B (en) | Intelligent stomach cancer staging method based on CT (computed tomography) images | |
CN112766333B (en) | Medical image processing model training method, medical image processing method and device | |
WO2022226949A1 (en) | Artificial neural network-based identification method and system for tissue lesion identification | |
Nandakumar et al. | Detection of Diabetic Retinopathy from Retinal Images Using DenseNet Models. | |
CN114359194A (en) | Multi-mode stroke infarct area image processing method based on improved U-Net network | |
WO2021021430A1 (en) | System and method for region detection in tissue sections using image registration | |
Moustari et al. | Two-stage deep learning classification for diabetic retinopathy using gradient weighted class activation mapping | |
Kumar | Intracranial Hemorrhage Detection Using Deep Learning and Transfer Learning | |
US20240296553A1 (en) | Automatic Image Quantification from Physician-Generated Reports |
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 |