CN114419083A - ResUnet medical image segmentation system based on edge operator improvement - Google Patents
ResUnet medical image segmentation system based on edge operator improvement Download PDFInfo
- Publication number
- CN114419083A CN114419083A CN202111542921.7A CN202111542921A CN114419083A CN 114419083 A CN114419083 A CN 114419083A CN 202111542921 A CN202111542921 A CN 202111542921A CN 114419083 A CN114419083 A CN 114419083A
- Authority
- CN
- China
- Prior art keywords
- loss function
- resunet
- medical image
- data
- obtaining
- 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.)
- Pending
Links
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
- G06T7/13—Edge detection
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic 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
- G06T2207/10081—Computed x-ray tomography [CT]
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30061—Lung
- G06T2207/30064—Lung nodule
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Probability & Statistics with Applications (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Epidemiology (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The invention provides an improved ResUnet medical image segmentation system based on an edge operator, which comprises: a data acquisition module configured to: acquiring medical image data to be processed; a lung segment segmentation module configured to: obtaining a lung segment segmentation result according to the obtained medical image data and a preset ResUnet network model; obtaining a first loss function according to the prediction result and the labeled standard data, obtaining a second loss function according to the prediction result and the labeled data processed by the edge operator, and taking the weighted sum of the first loss function and the second loss function as the loss function of the ResUnet network model; according to the method, optimization adjustment and redesign of the loss function of model training learning are carried out in the training process, intelligent feature learning can be carried out more quickly, and the organ structure segmentation accuracy is greatly improved.
Description
Technical Field
The invention relates to the technical field of medical image processing, in particular to an improved ResUnet medical image segmentation system based on an edge operator.
Background
The statements in this section merely provide background information related to the present disclosure and may not constitute prior art.
For deeper nodules of the lung, resection of the lung segment is a reliable surgical method, and complete resection of the nodules can be guaranteed. Affected by the anatomical structure of the segment gate, the operation procedures of different lung segments are different, and the segment gate of different lung segments has three major structures: the treatment of arteries, veins and trachea varies smoothly. Especially, when the anatomical structure is changed, the difficulty of lung segment operation is obviously increased. Therefore, in the diagnosis of the pulmonary nodules, more accurate information can be provided for doctors to diagnose and treat the pulmonary nodules by providing the information of the lung segments where the pulmonary nodules are located, so that more accurate diagnosis information can be obtained by better assisting doctors to live, and a more accurate treatment scheme is convenient to follow.
With the development of technologies such as deep learning, the application of the deep learning technology to medical auxiliary diagnosis becomes a necessary trend, and through the auxiliary diagnosis such as a deep learning algorithm, the working efficiency of doctors can be improved well, and more accurate diagnosis results can be provided for patients. Through deep learning, the lung CT image information is better utilized, and the parts such as lung segments and the like are better segmented.
However, the inventor finds that, in many existing deep learning-based schemes, model training is performed by using a large amount of data labeled data, and the segmentation effect of the structure edge is poor because medically related anatomical structure knowledge and edge feature information of image data are not well utilized.
Disclosure of Invention
In order to solve the defects of the prior art, the invention provides an improved ResUnet medical image segmentation system based on an edge operator, which is used for carrying out optimization adjustment and redesign of a loss function of model training learning in the training process, can carry out intelligent feature learning more quickly and greatly improves the accuracy of organ structure segmentation.
In order to achieve the purpose, the invention adopts the following technical scheme:
an improved ResUnet medical image segmentation system based on edge operators, comprising:
a data acquisition module configured to: acquiring medical image data to be processed;
a lung segment segmentation module configured to: obtaining a lung segment segmentation result according to the obtained medical image data and a preset ResUnet network model;
and obtaining a first loss function according to the prediction result and the labeled standard data, obtaining a second loss function according to the prediction result and the labeled data processed by the edge operator, and taking the weighted sum of the first loss function and the second loss function as the loss function of the ResUnet network model.
A second aspect of the present invention provides a computer-readable storage medium having stored thereon a program which, when executed by a processor, performs the steps of:
acquiring medical image data to be processed;
obtaining a lung segment segmentation result according to the obtained medical image data and a preset ResUnet network model;
and obtaining a first loss function according to the prediction result and the labeled standard data, obtaining a second loss function according to the prediction result and the labeled data processed by the edge operator, and taking the weighted sum of the first loss function and the second loss function as the loss function of the ResUnet network model.
A third aspect of the present invention provides an electronic device, including a memory, a processor, and a program stored in the memory and executable on the processor, wherein the processor executes the program to implement the following steps:
acquiring medical image data to be processed;
obtaining a lung segment segmentation result according to the obtained medical image data and a preset ResUnet network model;
and obtaining a first loss function according to the prediction result and the labeled standard data, obtaining a second loss function according to the prediction result and the labeled data processed by the edge operator, and taking the weighted sum of the first loss function and the second loss function as the loss function of the ResUnet network model.
Compared with the prior art, the invention has the beneficial effects that:
according to the ResUnet medical image segmentation system based on the improvement of the edge operator, in the training process, a first loss function is obtained according to the prediction result and the labeled standard data, a second loss function is obtained according to the prediction result and the labeled data processed by the Canny operator, the weighted sum of the first loss function and the second loss function is used as the loss function of the ResUnet network model, intelligent feature learning can be carried out more quickly, and the organ structure segmentation accuracy is greatly improved.
The edge detection loss function adopted in the invention can effectively improve the speed of model training, reduce the training time, reduce the requirement of a large amount of data of model data, reduce the amount of labeled data and further improve the segmentation accuracy.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
Fig. 1 is a schematic workflow diagram of an improved respnet medical image segmentation system based on an edge operator according to embodiment 1 of the present invention.
Fig. 2 is a schematic diagram of a network structure provided in embodiment 1 of the present invention.
Fig. 3 is a schematic flowchart of an algorithm provided in embodiment 1 of the present invention.
Detailed Description
The invention is further described with reference to the following figures and examples.
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
The embodiments and features of the embodiments of the present invention may be combined with each other without conflict.
Example 1:
as shown in fig. 1, fig. 2 and fig. 3, embodiment 1 of the present invention provides an improved respnet medical image segmentation system based on edge operators, including:
a data acquisition module configured to: acquiring medical image data to be processed;
a lung segment segmentation module configured to: obtaining a lung segment segmentation result according to the obtained medical image data and a preset ResUnet network model;
and obtaining a first loss function according to the prediction result and the labeled standard data, obtaining a second loss function according to the prediction result and the labeled data processed by the Canny operator, and taking the weighted sum of the first loss function and the second loss function as the loss function of the ResUnet network model.
Specifically, the work flow is as follows:
(1) constructing a data set of lung segment segmentation: the structure of each lung segment of the lung is labeled by using image CT data of the lung and experienced doctors.
(2) And (3) checking the labeled data: and (3) submitting the marked data to a principal physician of another three hospitals for examination, and when two experts have no objection to the marked data, taking the marked data as standard marked data, discussing the objected data by the experts, and confirming the marking modification opinions again until the data has no objection, namely, taking the marked data as standard marked data.
(3) Data preprocessing: and (3) carrying out window value calculation on the marked data according to HU values of [ -1200, 600], carrying out normalization processing, and processing output into a format of [128, 128, 128] as a model input standard. And secondly, performing edge detection calculation on the labeled data by using a Canny operator to obtain labeled data of relatively obvious organ segmentation edge characteristics such as oblique fissure, bronchi and the like in CT. And the data is divided into a training set and a test set, wherein the test set does not participate in the training.
(4) Designing a Loss function, wherein Loss is (1-beta). CELoss + beta. CYLoss;
CELOSs is a cross-entropy loss function of
Calculating a cross entropy loss function by the prediction result and the labeled standard data;
the CYLoss also adopts a cross entropy loss function, and the loss function is calculated through the prediction result and the labeled data processed by the Canny operator;
the beta factor is a variable factor, and is adjusted in the training process to guide the training effect of the model on the edge learning and the overall characteristics, for example, the training process is set to be 0.9, 0.7, 0.5, 0.3 and 0.1, and the weight is attenuated along with the training.
(5) Network training: inputting the preprocessed data (1, 128, 128, 128) into a ResUnet network for training, and simultaneously adopting a random sampling mode in the training process to sample the whole data, thereby increasing the diversity of the data and enhancing the generalization capability of the model.
(6) Prediction of lung segment segmentation model: after the iterative training is finished, the model is used for carrying out segmentation prediction on the lung segment on the data of the test set, and the segmentation prediction result is returned to a doctor as a lung nodule auxiliary result to assist the diagnosis of the doctor.
The following is presented as a specific example:
s1: selecting image data without defects of lung CT data, carrying out data labeling work by experts, labeling different lung segments with different colors and labels, storing the lung segments in a ni.gz format, and totally dividing the lung segments into 19 parts, wherein 0 is set as a background label.
S2: and after the preliminary annotation is finished, the annotated data is audited by another group of experts, the audited data is used as a final result if the conclusion is consistent, and the annotation data is modified or the original annotation data is kept if the opinion is left.
S3: and (4) the data after auditing is carried out according to the following steps of 7: and 3, dividing the training set into a training set and a testing set, wherein the training set is used by a KFold method. Before training begins, data are preprocessed, a HU value of lung CT data is windowed, lung data are obtained, resampling is carried out, the data spacing is guaranteed to be 1mm in thickness in three dimensions, and data are normalized and mapped to 255 gray scale data. And then converted into unit8 format. And adopting a random window for three dimensions of the data, and taking out the data of the [128, 128, 128] dimension as the input of the network.
S4: the network consists of an encoder module, a down _ conv module, a decoder module, an up _ conv module and a map module.
The encoder module mainly comprises 3D convolutions and 3 PRule activating functions, the down _ conv module mainly comprises one 3D convolution and one PRule activating function, the decoder module mainly comprises 3D convolutions and 3 PRule activating functions, and the up _ conv module mainly comprises one 3D transposed convolution and one PRule activating function. The map module consists of a 3D convolution and a UPSample upsampling layer. Finally, the final result is obtained by a softmax function.
S5: the Adam algorithm is selected in the training process to optimize the model, and the initial learning rate is 0.0001. And when the number of training epoch rounds reaches 50, multiplying the initial learning rate by an attenuation coefficient to obtain a new learning rate. The model weights are saved every 50 epochs.
S6: and setting the total number of training rounds as 2000 rounds, recording the variation of the loss value, checking the loss variation in the training, verifying the effect of the model, and taking the model as the use model when the loss of the model tends to be stable.
S7: and verifying and calculating evaluation parameters such as accuracy and dice value on the test set data by using the stored model, outputting a prediction result, performing 3D reconstruction on the generated mask segmentation result besides the evaluation standard, and comparing whether the comparison result is consistent or not and whether the result meets the use standard or not.
Example 2:
embodiment 2 of the present invention provides a computer-readable storage medium on which a program is stored, the program implementing, when executed by a processor, the steps of:
acquiring medical image data to be processed;
obtaining a lung segment segmentation result according to the obtained medical image data and a preset ResUnet network model;
and obtaining a first loss function according to the prediction result and the labeled standard data, obtaining a second loss function according to the prediction result and the labeled data processed by the Canny operator, and taking the weighted sum of the first loss function and the second loss function as the loss function of the ResUnet network model.
The detailed steps are the same as the working method provided in embodiment 1, and are not described again here.
Example 3:
embodiment 3 of the present invention provides an electronic device, including a memory, a processor, and a program stored in the memory and capable of running on the processor, where the processor executes the program to implement the following steps:
acquiring medical image data to be processed;
obtaining a lung segment segmentation result according to the obtained medical image data and a preset ResUnet network model;
and obtaining a first loss function according to the prediction result and the labeled standard data, obtaining a second loss function according to the prediction result and the labeled data processed by the Canny operator, and taking the weighted sum of the first loss function and the second loss function as the loss function of the ResUnet network model.
The detailed steps are the same as the working method provided in embodiment 1, and are not described again here.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. An improved ResUnet medical image segmentation system based on edge operators is characterized in that:
the method comprises the following steps:
a data acquisition module configured to: acquiring medical image data to be processed;
a lung segment segmentation module configured to: obtaining a lung segment segmentation result according to the obtained medical image data and a preset ResUnet network model;
and obtaining a first loss function according to the prediction result and the labeled standard data, obtaining a second loss function according to the prediction result and the labeled data processed by the edge operator, and taking the weighted sum of the first loss function and the second loss function as the loss function of the ResUnet network model.
2. The improved ResUnet medical image segmentation system based on edge operators as claimed in claim 1, wherein:
the first loss function and the second loss function are both cross entropy loss functions.
3. The improved ResUnet medical image segmentation system based on edge operators as claimed in claim 1, wherein:
the loss function of the ResUnet network model is the product of a first loss function and a first coefficient, and then the sum of the product of a second loss function and a second coefficient, wherein the sum of the first coefficient and the second coefficient is 1.
4. The improved ResUnet medical image segmentation system based on edge operators as claimed in claim 1, wherein:
the first coefficient is a variable factor.
5. The improved ResUnet medical image segmentation system based on edge operators as claimed in claim 1, wherein:
when the ResUnet network model is trained, the method comprises the following steps:
carrying out window value calculation on the annotation data of the image according to the HU value, and carrying out normalization processing to obtain the annotation data in a preset format; and (3) carrying out edge detection calculation on the labeled data in the preset format by using a Canny operator to obtain labeled data of organ segmentation edge features with the oblique fissure and the obvious bronchi in the CT.
6. The improved ResUnet medical image segmentation system based on edge operators as claimed in claim 1, wherein:
the ResUnet network model comprises: the device comprises an encoder module, a down _ conv module, a decoder module, an up _ conv module and a map module, wherein the encoder module comprises 3D convolutions and 3 PRule activating functions.
7. The improved ResUnet medical image segmentation system based on edge operators as claimed in claim 6, wherein:
the down _ conv module includes a 3D convolution and a PRule activation function, and the decoder module includes 3D convolutions and 3 PRule activation functions.
8. The improved ResUnet medical image segmentation system based on edge operators as claimed in claim 6, wherein:
the up _ conv module includes a 3D transposed convolution and a PRule activation function, and the map module includes a 3D convolution and a UPSample upsampling layer.
9. A computer-readable storage medium having a program stored thereon, the program, when executed by a processor, implementing the steps of:
acquiring medical image data to be processed;
obtaining a lung segment segmentation result according to the obtained medical image data and a preset ResUnet network model;
and obtaining a first loss function according to the prediction result and the labeled standard data, obtaining a second loss function according to the prediction result and the labeled data processed by the edge operator, and taking the weighted sum of the first loss function and the second loss function as the loss function of the ResUnet network model.
10. An electronic device comprising a memory, a processor, and a program stored on the memory and executable on the processor, wherein the processor implements the following steps when executing the program:
acquiring medical image data to be processed;
obtaining a lung segment segmentation result according to the obtained medical image data and a preset ResUnet network model;
and obtaining a first loss function according to the prediction result and the labeled standard data, obtaining a second loss function according to the prediction result and the labeled data processed by the edge operator, and taking the weighted sum of the first loss function and the second loss function as the loss function of the ResUnet network model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111542921.7A CN114419083A (en) | 2021-12-16 | 2021-12-16 | ResUnet medical image segmentation system based on edge operator improvement |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111542921.7A CN114419083A (en) | 2021-12-16 | 2021-12-16 | ResUnet medical image segmentation system based on edge operator improvement |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114419083A true CN114419083A (en) | 2022-04-29 |
Family
ID=81267079
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111542921.7A Pending CN114419083A (en) | 2021-12-16 | 2021-12-16 | ResUnet medical image segmentation system based on edge operator improvement |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114419083A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115170795A (en) * | 2022-05-13 | 2022-10-11 | 深圳大学 | Image small target segmentation method, device, terminal and storage medium |
| CN115272377A (en) * | 2022-09-27 | 2022-11-01 | 松立控股集团股份有限公司 | Vehicle segmentation method fusing image edge information |
-
2021
- 2021-12-16 CN CN202111542921.7A patent/CN114419083A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115170795A (en) * | 2022-05-13 | 2022-10-11 | 深圳大学 | Image small target segmentation method, device, terminal and storage medium |
| CN115272377A (en) * | 2022-09-27 | 2022-11-01 | 松立控股集团股份有限公司 | Vehicle segmentation method fusing image edge information |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108921851B (en) | Medical CT image segmentation method based on 3D countermeasure network | |
| Selvan et al. | Lung segmentation from chest X-rays using variational data imputation | |
| CN114419083A (en) | ResUnet medical image segmentation system based on edge operator improvement | |
| CN109087306A (en) | Arteries iconic model training method, dividing method, device and electronic equipment | |
| CN110517238B (en) | AI three-dimensional reconstruction and human-computer interaction visualization network system for CT medical image | |
| CN109801272A (en) | Liver neoplasm divides localization method, system and storage medium automatically | |
| CN109215035B (en) | Brain MRI hippocampus three-dimensional segmentation method based on deep learning | |
| CN112102276B (en) | Low-field-intensity MR stomach segmentation method based on transfer learning image enhancement | |
| CN110110723B (en) | Method and device for automatically extracting target area in image | |
| CN111986206A (en) | Lung lobe segmentation method and device based on UNet network and computer-readable storage medium | |
| KR102270934B1 (en) | Apparatus and method for medical image reading assistant providing representative image based on medical use artificial neural network | |
| CN110910404B (en) | Anti-noise data breast ultrasonic nodule segmentation method | |
| CN111260705A (en) | Prostate MR image multi-task registration method based on deep convolutional neural network | |
| CN112885453A (en) | Method and system for identifying pathological changes in subsequent medical images | |
| CN111932512B (en) | Intracranial hemorrhage detection method applied to CT image based on CNN and NLSTM neural network | |
| Tan et al. | An approach for pulmonary vascular extraction from chest CT images | |
| Walvekar et al. | Efficient medical image segmentation of covid-19 chest ct images based on deep learning techniques | |
| CN115311193A (en) | Abnormal brain image segmentation method and system based on double attention mechanism | |
| CN115330817A (en) | Pneumothorax lesion segmentation method based on Swin-Unet and morphological processing and related equipment | |
| CN112686849B (en) | Lung lobe segmentation system based on three-dimensional convolutional neural network | |
| Zhang et al. | C2‐GAN: Content‐consistent generative adversarial networks for unsupervised domain adaptation in medical image segmentation | |
| Ghali et al. | Vision transformers for lung segmentation on cxr images | |
| CN113421228A (en) | Thyroid nodule identification model training method and system based on parameter migration | |
| CN112967254A (en) | Lung disease identification and detection method based on chest CT image | |
| CN116309615A (en) | Multi-mode MRI brain tumor image segmentation method |
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 |