CN109308477A - A kind of medical image automatic division method, equipment and storage medium based on rough sort - Google Patents
A kind of medical image automatic division method, equipment and storage medium based on rough sort Download PDFInfo
- Publication number
- CN109308477A CN109308477A CN201811110633.2A CN201811110633A CN109308477A CN 109308477 A CN109308477 A CN 109308477A CN 201811110633 A CN201811110633 A CN 201811110633A CN 109308477 A CN109308477 A CN 109308477A
- Authority
- CN
- China
- Prior art keywords
- medical image
- organ
- neural network
- rough sort
- interest
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/285—Selection of pattern recognition techniques, e.g. of classifiers in a multi-classifier system
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/03—Recognition of patterns in medical or anatomical images
Abstract
The invention belongs to medical images and field of computer technology, are related to automatic division method, equipment and the storage medium of a kind of medical image.This method comprises the following steps: if the organ in medical image is divided into Ganlei；Training judges the class categories belonging to it for carrying out the deep learning neural network of rough sort to body medical image after so that medical image is input to rough sort deep learning neural network；Medical image to be sorted is input to trained rough sort deep learning neural network and carries out rough sort, output includes the medical image layer of organ of interest；Medical image layer comprising organ of interest is input in the matched fine segmentation neural network for being used to be split specific organ of interest, completes the Accurate Segmentation to the organ of interest on medical image.The present invention enables the automatic cutting procedure of radiotherapy structure to save the time that segmentation is predicted；Method provided by the invention does not depend on specific neural network, has very strong universality.
Description
Technical field
The invention belongs to medical images and field of computer technology, are related to a kind of automatic division method of medical image, set
Standby and storage medium.
Background technique
During hospital carries out radiotherapy to patient, the segmentation of target target area is often related to, at present doctor master
If segmentation is time-consuming and laborious by hand by the way of dividing by hand, the working efficiency of doctor is influenced, more influence patient's controls in time
It treats.
It has developed in the prior art and has completed automatic division method and organ that human body multiple location jeopardizes organ and target area
Neural network model training.However the organ of human body has very much, how to judge that whom inputted medical image includes
Body organ is simultaneously transported to progress Accurate Segmentation (delineating) in matching neural network, then needs the medicine shadow to input
As the organ for being included is classified.It is not possible to the sorter network of a positioning is individually trained for each organ, it is main
Reason is wanted to be that many target class account for that overall specific gravity is too low, when encountering some shorter and smaller organs, such case can become
It is all the more serious；And time-consuming will increase when individually training positioning sorter network has another disadvantage that prediction.
Summary of the invention
It is an object of the invention to overcome the shortcomings of existing technologies and provide a kind of medical image based on rough sort from
Dynamic dividing method, equipment and storage medium.
To achieve the above object, the invention adopts the following technical scheme: from top to bottom (or from top to bottom) by human organ
According to the apparent organ of boundary successively thick divide into several classes, trained rough segmentation neural network is passed through according to above-mentioned classification method
Classification belonging to target organ on medicine layer is determined first；Then according to rough sort as a result, in the medicine layer containing target organ
The accurate segmentation of the U-shaped segmentation neural fusion organ of upper use 3D corresponding with this layer.
A kind of medical image automatic division method based on rough sort including walks as follows suitable for executing in calculating equipment
It is rapid:
(1) if, the organ in medical image is successively divided into Ganlei according to the high and low position locating for it；
(2), based on organ classes' mode training in step (1) for carrying out the depth of rough sort to body medical image
Learning neural network judges the class categories belonging to it after so that medical image is input to rough sort deep learning neural network；
(3), medical image to be sorted is input to trained rough sort deep learning neural network and carries out rough sort,
Output includes the medical image layer of organ of interest；
(4), the medical image layer comprising organ of interest is input to matched be used for specific device interested
In the fine segmentation neural network that official is split, the Accurate Segmentation to the organ of interest on medical image is completed.
The present invention is it is further preferred that the medical image is a variety of doctors such as CT images, MR image or ultrasonograph
Learn image.
In step (1), it is by the obvious of human body that the rough segmentation neural network, which is that 2D classifies convolutional neural networks more,
The training of feature organ.
In step (1), body medical image is divided into ten classes according to the position of human organ, respectively opens and begins from first
To the crown, from the crown to eyes on top layer, from eyes top layer under top layer to eyes, under eyes top layer under top layer to cerebellum,
Under cerebellum top layer to lower jaw the last layer, from lower jaw the last layer to lung top layer, lung top layer to stomach top layer, stomach top layer to kidney
Bottom, kidney bottom to bladder top, bladder top to foot.
In step (4), the fine segmentation neural network is to utilize this with the U-shaped neural network of 3D of expansion convolution
Resolution capability of the neural network on human body Z axis carries out Accurate Segmentation to the organ of interest on medical image, and wherein Z axis is behaved
The up and down direction of body organ, the training data of the network are the data block comprising several continuous medical images.
The present invention also provides a kind of calculating equipment, comprising:
One or more processors；
Memory；And
One or more programs, wherein one or more of programs are stored in the memory and are configured as by one
A or multiple processors execute, and one or more programs include executing the side of segmentation automatically of the medical image based on rough sort
The instruction of method.
The present invention also provides a kind of computer readable storage medium for storing one or more programs, described one or more
A program includes instruction, and described instruction is suitable for being loaded by memory and executing the above-mentioned medical image based on rough sort dividing automatically
Method.
The present invention has following beneficial outcomes:
The present invention first trains a coarse positioning network according to the apparent human organ of boundary, to determine target organ substantially
Classify (position), fine segmentation is then carried out according to positioning again.Therefore, the method provided through the invention: (1) make radiotherapy structure
Automatic cutting procedure in accuracy between organ generic it is secure (because distinguished with the apparent organ of boundary between classification
) and be suitable for systemic organs, it can be used as the omnibus algorithm that systemic organs divide first step coarse positioning, and save segmentation prediction
Time；(2) after rough sort, the data of the U-shaped neural network of 3D are inputted because need not consider the medicine shadow of the outer organ of generic
Picture, specific gravity whole shared by target organ is just greatly improved in this, no matter is all very favorable to training or prediction；
(3) method provided by the invention does not depend on specific neural network, has very strong universality.
Detailed description of the invention
Fig. 1 is that human body is classified (10 class) schematic diagram totally from top to bottom in an example of the present invention embodiment.
Fig. 2 is the medical image automatic division method based on rough sort in an example of the present invention embodiment to oesophagus top
The segmentation result schematic diagram of starting end.
Fig. 3 is the medical image automatic division method based on rough sort in an example of the present invention embodiment to oesophagus lower part
The segmentation result schematic diagram of ending segment.
Fig. 4 is the medical image automatic division method flow chart based on rough sort in a preferred embodiment of the invention.
Specific embodiment
The present invention is further illustrated below in conjunction with drawings and examples.
A kind of medical image automatic division method (as shown in Figure 4) based on rough sort, suitable for being executed in calculating equipment,
Include the following steps:
Step 210, if the organ in medical image is successively divided into Ganlei according to the high and low position locating for it；
Preferably, medical image can be selected from plurality of medical images such as CT images, MR image or ultrasonographs.
Preferably, the apparent organ of human body boundary is as human body device in rough segmentation neural network using in CT images for the present embodiment
Human body CT images are divided into ten classes by the foundation of official's classification from top to bottom or from top to bottom；Wherein in one example, the present embodiment
Human body CT images are sequentially divided into ten classes from top to bottom, as shown in table 1 and Fig. 1.
Table 1
Step 220, with training ten classification for human body CT images to be carried out with rough sort according to the classification method in step 210
Deep learning neural network judges the classification class belonging to it after so that medical image is input to rough sort deep learning neural network
Not；In the present embodiment preferably, rough sort Web vector graphic is depth convolutional neural networks；
Step 230, it by CT images to be sorted, is input to trained rough sort deep learning neural network and is divided
Class；Output includes the medical image layer of organ of interest；
Step 240, by the medical image layer comprising organ of interest be input to it is matched be used for it is emerging to specific sense
In the fine segmentation neural network that interesting organ is split, the Accurate Segmentation to the organ of interest on medical image is completed.?
In one example embodiment, fine segmentation neural network is to utilize the nerve net with the U-shaped neural network of 3D of expansion convolution
Resolution capability of the network on human body Z axis carries out Accurate Segmentation to the organ of interest on medical image, and wherein Z axis is human organ
Up and down direction, the training data of the network is the data block comprising several continuous medical images.With more classification convolution of 2D
Neural network is compared, and the U-shaped neural network of 3D considers the information on Z axis.Such as the Z-direction of oesophagus be above connect it is pharyngeal, it is lower with
The cardia of stomach is connected.Training 2D classify more convolutional neural networks when input network be each medical image, and each
It is mutually broken up between medical image, so just without Z axis information；The medicine of network is inputted when the U-shaped neural network of training 3D
Image be it is blocking, each piece has several medical images, such as 32, mutually broken up between block and block, so the U- of 3D
Net network just saves Z axis information in each piece, so the result of training is exactly to have certain resolution capability to Z axis.
In a preferred embodiment, illustrate this by taking the automatic segmentation of the oesophagus (Esophagus) in CT images as an example
The medical image automatic division method based on rough sort provided is provided, is included the following steps:
If the organ in medical image is successively divided into Ganlei according to the high and low position locating for it；
Based on the training of above-mentioned organ classes' mode for carrying out the deep learning nerve net of rough sort to body medical image
Network judges the class categories belonging to it, wherein rough sort after so that medical image is input to rough sort deep learning neural network
Neural network is ten Classification Neurals as shown in figure 1 and table 1；
The CT images of patient are input in rough segmentation neural network, it, can be in order to from big by rough segmentation neural network
The CT image of organ of interest is obtained in the CT training data of amount, for example oesophagus is located at [6,7] class of rough sort network, then
Training data fetch bit for fine segmentation neural network is in the CT images of [6,7] class.Such as existing 73 sets of CT (are shared
5731 CT images), and the CT images containing esophagus have 3906, then including a large amount of unrelated CT images in above-mentioned 73 sets of CT
(1825) can be removed by the screening of rough segmentation neural network, to improve the specific gravity that target data accounts for overall data；
Remove to shear every medical image object machine in [6,7] class in table 1 with the rectangle frame of a fixed window size again
Medical image layer comprising organ of interest is input to matched be used for specific organ of interest by the CT images of official
In the fine segmentation neural network being split, the Accurate Segmentation to the organ of interest on medical image is completed.Fig. 2 and Fig. 3
The respectively segmentation effect schematic diagram of the segmentation effect of oesophagus top starting end and oesophagus lower part ending segment.If only depended on thick
Sorter network as a result, then delineating for oesophagus can be since [6] class lower jaw the last layer, this is possible to hook several layers of cause more
False sun；End is the end stomach top layer of [7], is thick with various alvus organs here, and HU value is close, is also easy to delineate not
Standard causes false sun.And method provided by the invention is used, after completing rough sort, the U-shaped nerve net of 3D is used in [6,7]
Network can guarantee that the data of input are largely target datas first, and signal-to-noise ratio significantly gets a promotion；Secondly for object machine
The beginning of official delineated terminates, and the U-shaped neural network of 3D can also learn corresponding information, this is very for delineating for junction
It is necessary.
The present invention also provides a kind of calculating equipment, comprising:
One or more processors；
Memory；And
One or more programs, the storage of wherein one or more programs in the memory and be configured as by one or
Multiple processors execute, and said one or multiple programs include the localization method divided automatically for human organ medical image
Instruction, this method comprises the following steps:
If the organ in medical image is successively divided into Ganlei according to the high and low position locating for it；
Based on the training of above-mentioned organ classes' mode for carrying out the deep learning nerve net of rough sort to body medical image
Network judges the class categories belonging to it after so that medical image is input to rough sort deep learning neural network；
Medical image to be sorted is input to trained rough sort deep learning neural network and carries out rough sort, output
Medical image layer comprising organ of interest；
By the medical image layer comprising organ of interest be input to it is matched be used for specific organ of interest into
In the fine segmentation neural network of row segmentation, the Accurate Segmentation to the organ of interest on medical image is completed.
The present invention also provides a kind of computer readable storage medium for storing one or more programs, above-mentioned one or more
A program includes instruction, which is suitable for load by memory and being executed the above-mentioned human organ medical image of being used for and divides automatically
Localization method, this method comprises the following steps:
If the organ in medical image is successively divided into Ganlei according to the high and low position locating for it；
Based on the training of above-mentioned organ classes' mode for carrying out the deep learning nerve net of rough sort to body medical image
Network judges the class categories belonging to it after so that medical image is input to rough sort deep learning neural network；
Medical image to be sorted is input to trained rough sort deep learning neural network and carries out rough sort, output
Medical image layer comprising organ of interest；
By the medical image layer comprising organ of interest be input to it is matched be used for specific organ of interest into
In the fine segmentation neural network of row segmentation, the Accurate Segmentation to the organ of interest on medical image is completed.
It should be appreciated that various technologies described herein are realized together in combination with hardware or software or their combination.From
And some aspects or part of the process and apparatus of the present invention or the process and apparatus of the present invention can take the tangible matchmaker of insertion
It is situated between, such as the program code in floppy disk, CD-ROM, hard disk drive or other any machine readable storage mediums (refers to
Enable) form, wherein when program is loaded into the machine of such as computer etc, and when being executed by the machine, which becomes real
Trample equipment of the invention.
By way of example and not limitation, computer-readable medium includes computer storage media and communication media.It calculates
Machine storage medium stores the information such as computer readable instructions, data structure, program module or other data.Communication media one
As with the modulated message signals such as carrier wave or other transmission mechanisms embody computer readable instructions, data structure, program
Module or other data, and including any information transmitting medium.Above any combination is also included within computer-readable
Within the scope of medium.
It will be understood to those skilled in the art that can adaptively be changed to the module in the equipment in embodiment
Become and they are arranged in one or more devices different from this embodiment.It can be the module or unit in embodiment
Or component is combined into a module or unit or component, and furthermore they can be divided into multiple submodule or subelements or son
Component.Other than such feature and/or at least some of process or unit exclude each other, any group can be used
It closes to all features disclosed in this specification (including adjoint claim, abstract and attached drawing) and so disclosed any
All process or units of method or apparatus are combined.Unless expressly stated otherwise, this specification is (including adjoint right
It is required that, abstract and attached drawing) disclosed in each feature can carry out generation with an alternative feature that provides the same, equivalent, or similar purpose
It replaces.
As used in this, unless specifically stated, come using ordinal number " first ", " second ", " third " etc.
Description plain objects, which are merely representative of, is related to the different instances of similar object, and is not intended to imply that the object being described in this way must
Must have the time it is upper, spatially, sequence aspect or given sequence in any other manner.
In addition, it will be appreciated by those of skill in the art that although some embodiments described herein include other embodiments
In included certain features rather than other feature, but the combination of the feature of different embodiments mean it is of the invention
Within the scope of and form different embodiments.For example, in the following claims, embodiment claimed is appointed
Meaning one of can in any combination mode come using.
This hair can be understood and applied the above description of the embodiments is intended to facilitate those skilled in the art
It is bright.Person skilled in the art obviously easily can make various modifications to these embodiments, and described herein
General Principle is applied in other embodiments without having to go through creative labor.Therefore, the present invention is not limited to implementations here
Example, those skilled in the art's announcement according to the present invention, improvement and modification made without departing from the scope of the present invention all should be
Within protection scope of the present invention.
Claims (7)
1. a kind of medical image automatic division method based on rough sort, suitable for being executed in calculating equipment, it is characterised in that: packet
Include following steps:
(1) if, the organ in medical image is successively divided into Ganlei according to the high and low position locating for it；
(2), based on organ classes' mode training in step (1) for carrying out the deep learning of rough sort to body medical image
Neural network judges the class categories belonging to it after so that medical image is input to rough sort deep learning neural network；
(3), medical image to be sorted is input to trained rough sort deep learning neural network and carries out rough sort, output
Medical image layer comprising organ of interest；
(4), by the medical image layer comprising organ of interest be input to it is matched be used for specific organ of interest into
In the fine segmentation neural network of row segmentation, the Accurate Segmentation to the organ of interest on medical image is completed.
2. the medical image automatic division method according to claim 1 based on rough sort, it is characterised in that: the doctor
Image is CT images, MR image or ultrasonograph.
3. the medical image automatic division method according to claim 1 based on rough sort, it is characterised in that: step (1)
In, the rough segmentation neural network is that 2D classifies convolutional neural networks more.
4. the medical image automatic division method according to claim 1 based on rough sort, it is characterised in that: step (1)
In, body medical image is divided into ten classes: be respectively from first open begin to the crown, from the crown to eyes on top layer, from
Top layer under top layer to eyes on eyes, under eyes top layer under top layer to cerebellum, under cerebellum top layer to lower jaw the last layer, from
Lower jaw the last layer is to lung top layer, lung top layer to stomach top layer, stomach top layer to kidney bottom, kidney bottom to bladder top, bladder top to foot.
5. the medical image automatic division method according to claim 1 based on rough sort, it is characterised in that: step (4)
In, the fine segmentation neural network is the three-dimensional U-shape neural network with expansion convolution, using the neural network in human body
Resolution capability on Z axis carries out Accurate Segmentation to the organ of interest on medical image, and wherein Z axis is the upper and lower of human organ
To the training data of the neural network is the data block comprising several continuous medical images.
6. a kind of calculating equipment, comprising:
One or more processors；
Memory；And
One or more programs, wherein the storage of one or more of programs in the memory and be configured as by one or
Multiple processors execute, one or more programs include perform claim require it is any described based on rough sort in 1-5
Medical image automatic division method instruction.
7. a kind of computer readable storage medium for storing one or more programs, one or more programs include referring to
It enables, described instruction is suitable for being loaded by memory and being executed any medicine based on rough sort in the claims 1-5
Image automatic division method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811110633.2A CN109308477A (en) | 2018-09-21 | 2018-09-21 | A kind of medical image automatic division method, equipment and storage medium based on rough sort |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811110633.2A CN109308477A (en) | 2018-09-21 | 2018-09-21 | A kind of medical image automatic division method, equipment and storage medium based on rough sort |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109308477A true CN109308477A (en) | 2019-02-05 |
Family
ID=65224137
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811110633.2A Pending CN109308477A (en) | 2018-09-21 | 2018-09-21 | A kind of medical image automatic division method, equipment and storage medium based on rough sort |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109308477A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110047577A (en) * | 2019-04-23 | 2019-07-23 | 杭州依图医疗技术有限公司 | Image distributor and image distribution method |
| CN110197716A (en) * | 2019-05-20 | 2019-09-03 | 广东技术师范大学 | Processing method, device and the computer readable storage medium of medical image |
| CN111127444A (en) * | 2019-12-26 | 2020-05-08 | 广州柏视医疗科技有限公司 | Method for automatically identifying radiotherapy organs at risk in CT image based on depth semantic network |
| CN111462086A (en) * | 2020-03-31 | 2020-07-28 | 北京推想科技有限公司 | Image segmentation method and device and training method and device of neural network model |
| CN111724893A (en) * | 2019-03-20 | 2020-09-29 | 宏碁股份有限公司 | Medical image recognition device and medical image recognition method |
| CN115761221A (en) * | 2022-09-21 | 2023-03-07 | 阿里巴巴（中国）有限公司 | Image processing method, device, equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2157905A1 (en) * | 2007-05-15 | 2010-03-03 | Val-Chum, S.E.C. | A method for tracking 3d anatomical and pathological changes in tubular-shaped anatomical structures |
| JP2011092677A (en) * | 2009-09-30 | 2011-05-12 | Fujifilm Corp | Medical image diagnostic apparatus and method using liver angiographic image, and program |
| CN105447872A (en) * | 2015-12-03 | 2016-03-30 | 中山大学 | Method for automatically identifying liver tumor type in ultrasonic image |
| CN105719278A (en) * | 2016-01-13 | 2016-06-29 | 西北大学 | Organ auxiliary positioning segmentation method based on statistical deformation model |
| CN107705305A (en) * | 2017-10-20 | 2018-02-16 | 上海联影医疗科技有限公司 | A kind of method and device of Medical Image Processing |
| CN108447062A (en) * | 2018-02-01 | 2018-08-24 | 浙江大学 | A kind of dividing method of the unconventional cell of pathological section based on multiple dimensioned mixing parted pattern |
-
2018
- 2018-09-21 CN CN201811110633.2A patent/CN109308477A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2157905A1 (en) * | 2007-05-15 | 2010-03-03 | Val-Chum, S.E.C. | A method for tracking 3d anatomical and pathological changes in tubular-shaped anatomical structures |
| JP2011092677A (en) * | 2009-09-30 | 2011-05-12 | Fujifilm Corp | Medical image diagnostic apparatus and method using liver angiographic image, and program |
| CN105447872A (en) * | 2015-12-03 | 2016-03-30 | 中山大学 | Method for automatically identifying liver tumor type in ultrasonic image |
| CN105719278A (en) * | 2016-01-13 | 2016-06-29 | 西北大学 | Organ auxiliary positioning segmentation method based on statistical deformation model |
| CN107705305A (en) * | 2017-10-20 | 2018-02-16 | 上海联影医疗科技有限公司 | A kind of method and device of Medical Image Processing |
| CN108447062A (en) * | 2018-02-01 | 2018-08-24 | 浙江大学 | A kind of dividing method of the unconventional cell of pathological section based on multiple dimensioned mixing parted pattern |
Non-Patent Citations (2)
| Title |
|---|
| REZA MORADI RAD 等: "Multi-Resolutional Ensemble of Stacked Dilated U-Net for Inner Cell Mass Segmentation in Human Embryonic Images", 《2018 25TH IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP)》 * |
| 金春兰 等: "基于多重分形的医学图像分割方法", 《中国组织工程研究》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111724893A (en) * | 2019-03-20 | 2020-09-29 | 宏碁股份有限公司 | Medical image recognition device and medical image recognition method |
| CN111724893B (en) * | 2019-03-20 | 2024-04-09 | 宏碁股份有限公司 | Medical image identification device and medical image identification method |
| CN110047577A (en) * | 2019-04-23 | 2019-07-23 | 杭州依图医疗技术有限公司 | Image distributor and image distribution method |
| CN110197716A (en) * | 2019-05-20 | 2019-09-03 | 广东技术师范大学 | Processing method, device and the computer readable storage medium of medical image |
| CN111127444A (en) * | 2019-12-26 | 2020-05-08 | 广州柏视医疗科技有限公司 | Method for automatically identifying radiotherapy organs at risk in CT image based on depth semantic network |
| CN111127444B (en) * | 2019-12-26 | 2021-06-04 | 广州柏视医疗科技有限公司 | Method for automatically identifying radiotherapy organs at risk in CT image based on depth semantic network |
| CN111462086A (en) * | 2020-03-31 | 2020-07-28 | 北京推想科技有限公司 | Image segmentation method and device and training method and device of neural network model |
| CN111462086B (en) * | 2020-03-31 | 2024-04-26 | 推想医疗科技股份有限公司 | Image segmentation method and device, and training method and device of neural network model |
| CN115761221A (en) * | 2022-09-21 | 2023-03-07 | 阿里巴巴（中国）有限公司 | Image processing method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109308477A (en) | A kind of medical image automatic division method, equipment and storage medium based on rough sort | |
| CN107909581B (en) | Lobe of the lung section dividing method, device, system, storage medium and the equipment of CT images | |
| CN109035255B (en) | Method for segmenting aorta with interlayer in CT image based on convolutional neural network | |
| CN109410188B (en) | System and method for segmenting medical images | |
| CN106204587A (en) | Multiple organ dividing method based on degree of depth convolutional neural networks and region-competitive model | |
| CN103180878B (en) | System for the segmentation of a medical image | |
| CN109872333A (en) | Medical image dividing method, device, computer equipment and storage medium | |
| CN108268870A (en) | Multi-scale feature fusion ultrasonoscopy semantic segmentation method based on confrontation study | |
| CN109447976A (en) | A kind of medical image cutting method and system based on artificial intelligence | |
| CN109285200A (en) | A kind of conversion method of the Multimodal medical image based on artificial intelligence | |
| CN107563434B (en) | Brain MRI image classification method and device based on three-dimensional convolutional neural network | |
| CN103562960B (en) | For generating the assigned unit between the image-region of image and element class | |
| CN110310287A (en) | It is neural network based to jeopardize the automatic delineation method of organ, equipment and storage medium | |
| CN110047082A (en) | Pancreatic Neuroendocrine Tumors automatic division method and system based on deep learning | |
| CN109727235A (en) | A kind of automatic delineation algorithms of organ based on deep learning | |
| CN105389811A (en) | Multi-modality medical image processing method based on multilevel threshold segmentation | |
| CN108986115A (en) | Medical image cutting method, device and intelligent terminal | |
| CN109461161A (en) | A method of human organ in medical image is split based on neural network | |
| Aygün et al. | Multi modal convolutional neural networks for brain tumor segmentation | |
| CN109215040A (en) | A kind of tumor of breast dividing method based on multiple dimensioned weighting study | |
| Snaauw et al. | End-to-end diagnosis and segmentation learning from cardiac magnetic resonance imaging | |
| CN111127487B (en) | Real-time multi-tissue medical image segmentation method | |
| Zimmer et al. | A multi-task approach using positional information for ultrasound placenta segmentation | |
| CN110415252A (en) | A kind of eye circumference organ segmentation method, equipment and storage medium based on CNN | |
| CN113192069A (en) | Semantic segmentation method and device for tree structure in three-dimensional tomography image |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190205 |