CN110163870A - A kind of abdomen body image liver segmentation method and device based on deep learning - Google Patents
A kind of abdomen body image liver segmentation method and device based on deep learning Download PDFInfo
- Publication number
- CN110163870A CN110163870A CN201910334931.8A CN201910334931A CN110163870A CN 110163870 A CN110163870 A CN 110163870A CN 201910334931 A CN201910334931 A CN 201910334931A CN 110163870 A CN110163870 A CN 110163870A
- Authority
- CN
- China
- Prior art keywords
- abdomen body
- body image
- threshold
- liver
- segmentation
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformation in the plane of the image
- G06T3/40—Scaling the whole image or part thereof
- G06T3/4038—Scaling the whole image or part thereof for image mosaicing, i.e. plane images composed of plane sub-images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration by the use of more than one image, e.g. averaging, subtraction
-
- 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
- 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
- G06T2207/10—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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30056—Liver; Hepatic
Abstract
The present invention relates to preoperative planning fields, and in particular to a kind of abdomen body image liver segmentation method and device based on deep learning.This method and device carry out sliding window sampling to pretreated abdomen body image and obtain splicing volume data, each splicing volume data is sent into trained liver segmentation convolutional neural networks and obtains corresponding segmentation result, and multiple segmentation results are spliced into complete abdomen body image data, go out liver area convenient for subsequent singulation, that the present invention is based on the abdomen body image liver segmentation method and device splitting speeds of deep learning is fast, segmentation precision is high.
Description
Technical field
The present invention relates to preoperative planning fields, in particular to a kind of abdomen body image liver based on deep learning
Dividing method and device.
Background technique
Important organ is rebuild from medical volume data such as CT or MR, be for clinical medicine it is highly important,
Accurately segmentation not only contributes to that then area-of-interest is quantitatively evaluated, and also helps Precise Diagnosis, prognosis prediction, hand
It is instructed in art plan and art.As the maximum organ of abdomen, the accurate segmentation of liver is liver neoplasm excision and Minimally Invasive Surgery
Prerequisite.
Nowadays, the goldstandard of liver segmentation is obtained by experienced doctor's manual segmentation, but for upper
The manual mark of the medical 3 D volume data of hundred slices is very boring and wastes time, and has the subjectivity of doctor
Experience, therefore full automatic partitioning algorithm is very necessary.
There is several points challenge below for the full-automatic dividing algorithm of 3 D medical data:
1. the difference between the boundary of liver and its surrounding tissue is very fuzzy, contrast is insufficient；
2. the organs differences between different patients are larger；
Previous liver algorithm specifically include that active contour model, statistical shape model, level set, multichannel chromatogram, figure cut with
And the machine learning algorithm based on manual feature, the above algorithm are bad for above-mentioned two o'clock challenge performance.
Convolutional neural networks yielded unusually brilliant results in computer vision field in recent years, however although depth convolutional neural networks
Surprising effect is achieved on two-dimensional natural image, for three-dimensional medical image volume data, convolutional neural networks
Still it is faced with some challenges below:
1. 3 d medical images have more complicated anatomical environment than the image under 2 D natural scene, therefore usually require
The variant of Three dimensional convolution neural network with more parameters captures more characteristic features；
2. there is need the problem of optimizing aspect for the such convolutional neural networks of training: over-fitting, gradient disappearance with
Explosion and slow convergence rate.
3. the scarcity of medical image data amount makes one deep neural network of training become more difficult.
Summary of the invention
The abdomen body image liver segmentation method and device based on deep learning that the embodiment of the invention provides a kind of, so that
The low technical problem of existing liver segmentation System Partition precision is solved less.
An embodiment according to the present invention provides a kind of abdomen body image liver segmentation method based on deep learning,
The following steps are included:
Multiple abdomen body images of input are pre-processed；
Sliding window sampling is carried out to pretreated abdomen body image and obtains splicing volume data；
Each splicing volume data is sent into trained liver segmentation convolutional neural networks and obtains corresponding segmentation result；
The segmentation result of the splicing volume data of all acquisitions is spliced into a complete abdomen body image data.
Further, this method further include:
Largest connected region is extracted in complete abdomen body image data, is partitioned into liver area after removing unrelated results
Domain.
Further, the training step of liver segmentation convolutional neural networks includes:
Gray threshold truncation is carried out to multiple abdomen body images of input；
Spacing normalization is carried out to the abdomen body image data after gray threshold is truncated；
Multiple training samples are obtained to the abdomen body image sliding sampling after progress spacing normalization；
Construct the liver segmentation convolutional neural networks based on deep learning and using multiple training samples to liver point
Convolutional neural networks are cut to be trained.
Further, carrying out gray threshold truncation to multiple abdomen body images of input includes:
Two threshold values, respectively first threshold and second threshold are chosen, first threshold is greater than second threshold, to the more of input
A abdomen body image carries out gray threshold truncation, will be above first threshold and excludes lower than the non-liver area of second threshold
It goes.
Further, the liver segmentation convolutional neural networks based on deep learning are constructed and make multiple training samples
Liver segmentation convolutional neural networks are trained with stochastic gradient descent optimization algorithm.
Further, carrying out pretreatment to multiple abdomen body images of input includes:
Gray threshold truncation is carried out to multiple abdomen body images of input；
Spacing normalization is carried out to the abdomen body image data after gray threshold is truncated.
Further, carrying out gray threshold truncation to multiple abdomen body images of input includes:
Two threshold values, respectively first threshold and second threshold are chosen, first threshold is greater than second threshold, to the more of input
A abdomen body image carries out gray threshold truncation, will be above first threshold and excludes lower than the non-liver area of second threshold
It goes.
Further, abdomen body image includes CT or MR abdomen body image.
According to another embodiment of the present invention, a kind of abdomen body image liver segmentation dress based on deep learning is provided
It sets, comprising:
Pretreatment unit, for being pre-processed to multiple abdomen body images of input；
Window sliding unit obtains spliceosome number for carrying out sliding window sampling to pretreated abdomen body image
According to；
Training cutting unit is obtained for each splicing volume data to be sent into trained liver segmentation convolutional neural networks
Corresponding segmentation result；
Segmentation result concatenation unit, for the segmentation result of the splicing volume data of all acquisitions to be spliced into one completely
Abdomen body image data.
Further, the device further include:
Connected region extraction unit removes nothing for extracting largest connected region in complete abdomen body image data
Liver area is partitioned into after closing result.
The abdomen body image liver segmentation method and device based on deep learning in the embodiment of the present invention, after pretreatment
Abdomen body image carry out sliding window sampling and obtain splicing volume data, each splicing volume data is sent into trained liver and is divided
It cuts convolutional neural networks and obtains corresponding segmentation result, and multiple segmentation results are spliced into complete abdomen body image data,
Go out liver area convenient for subsequent singulation, the present invention is based on the abdomen body image liver segmentation method and device of deep learning segmentation speed
Degree is fast, segmentation precision is high.
Detailed description of the invention
The drawings described herein are used to provide a further understanding of the present invention, constitutes part of this application, this hair
Bright illustrative embodiments and their description are used to explain the present invention, and are not constituted improper limitations of the present invention.In the accompanying drawings:
Fig. 1 is that the present invention is based on the flow charts of the abdomen body image liver segmentation method of deep learning；
Fig. 2 is that the present invention is based on the preferred flow charts of the abdomen body image liver segmentation method of deep learning；
Fig. 3 is the training flow chart of liver segmentation convolutional neural networks in the present invention；
Fig. 4 is that the present invention is based on the module maps of the abdomen body image liver segmentation device of deep learning；
Fig. 5 is that the present invention is based on the preferred module figures of the abdomen body image liver segmentation device of deep learning.
Specific embodiment
In order to enable those skilled in the art to better understand the solution of the present invention, below in conjunction in the embodiment of the present invention
Attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is only
The embodiment of a part of the invention, instead of all the embodiments.Based on the embodiments of the present invention, ordinary skill people
The model that the present invention protects all should belong in member's every other embodiment obtained without making creative work
It encloses.
It should be noted that description and claims of this specification and term " first " in above-mentioned attached drawing, "
Two " etc. be to be used to distinguish similar objects, without being used to describe a particular order or precedence order.It should be understood that using in this way
Data be interchangeable under appropriate circumstances, so as to the embodiment of the present invention described herein can in addition to illustrating herein or
Sequence other than those of description is implemented.In addition, term " includes " and " having " and their any deformation, it is intended that cover
Cover it is non-exclusive include, for example, the process, method, system, product or equipment for containing a series of steps or units are not necessarily limited to
Step or unit those of is clearly listed, but may include be not clearly listed or for these process, methods, product
Or other step or units that equipment is intrinsic.
Embodiment 1
An embodiment according to the present invention provides a kind of abdomen body image liver segmentation method based on deep learning, ginseng
See Fig. 1, comprising the following steps:
S101: multiple abdomen body images of input are pre-processed；
S102: sliding window sampling is carried out to pretreated abdomen body image and obtains splicing volume data；
S103: each splicing volume data is sent into trained liver segmentation convolutional neural networks and obtains corresponding segmentation knot
Fruit；
S104: the segmentation result of the splicing volume data of all acquisitions is spliced into a complete abdomen body image data.
The abdomen body image liver segmentation method based on deep learning in the embodiment of the present invention, to pretreated abdomen
Body image carries out sliding window sampling and obtains splicing volume data, and each splicing volume data is sent into trained liver segmentation convolution
Neural network obtains corresponding segmentation result, and multiple segmentation results are spliced into complete abdomen body image data, after being convenient for
It is continuous to be partitioned into liver area, fast, segmentation essence that the present invention is based on the abdomen body image liver segmentation method splitting speeds of deep learning
Degree is high.
In as a preferred technical scheme, referring to fig. 2, this method further include:
S105: extracting largest connected region in complete abdomen body image data, is partitioned into liver after removing unrelated results
Dirty district domain.
In as a preferred technical scheme, referring to Fig. 3, the training step of liver segmentation convolutional neural networks includes:
S201: gray threshold truncation is carried out to multiple abdomen body images of input；
S202: spacing normalization is carried out to the abdomen body image data after gray threshold is truncated；
S203: multiple training samples are obtained to the abdomen body image sliding sampling after progress spacing normalization；
S204: liver segmentation convolutional neural networks of the building based on deep learning simultaneously use multiple training samples to liver
Dirty segmentation convolutional neural networks are trained.
In as a preferred technical scheme, to multiple abdomens of input in the training step of liver segmentation convolutional neural networks
Portion's body image carries out gray threshold truncation
Two threshold values, respectively first threshold and second threshold are chosen, first threshold is greater than second threshold, to the more of input
A abdomen body image carries out gray threshold truncation, will be above first threshold and excludes lower than the non-liver area of second threshold
It goes.
In as a preferred technical scheme, the liver segmentation convolutional neural networks based on deep learning are constructed and by multiple nets
Network training sample is trained liver segmentation convolutional neural networks using stochastic gradient descent optimization algorithm.
Specifically, it is as follows to be somebody's turn to do the liver segmentation convolutional neural networks training step based on deep learning:
Step 1: multiple CT or MR abdomen body images (three-dimensional data) to input carry out gray threshold truncation, exclude
The influence of non-liver area.
Step 2: carrying out spacing normalization to CT the or MR abdomen body image data after gray threshold is truncated, make
The physical extent for obtaining each CT or MR abdomen body image data is approximately equal.
Step 3: sliding sampling obtains multiple training samples of liver segmentation convolutional neural networks.
Step 4: constructing the liver segmentation convolutional neural networks based on deep learning and using multiple networks in third step
Training sample is trained liver segmentation convolutional neural networks.
Below with specific embodiment, liver segmentation convolutional neural networks training of the invention is described in detail:
Step 1: gray threshold truncation is carried out to multiple CT or MR abdomen body images of input, chooses two suitable thresholds
Value, a high threshold a, Low threshold will be above high threshold and be excluded away lower than the non-liver area of Low threshold, leave
The effective coverage of abdomen body image.
Step 2: being averaged for the concentration of the CT or MR abdomen body image data after gray threshold is truncated is calculated
Spacing carries out spacing normalization to abdomen body image data, the physical extent of all abdomen body image datas is carried out
It is unified, so that the physical extent of each CT or MR abdomen body image data is approximately equal.
Step 3: sliding window sampling is carried out to CT the or MR abdomen body image data after spacing normalized and is made
For the model training data of liver segmentation convolutional neural networks.
Step 4: the model training data obtained using step 3 kind, and using stochastic gradient descent optimization algorithm to liver
Dirty segmentation convolutional neural networks are trained.
In as a preferred technical scheme, in the abdomen body image liver segmentation method based on deep learning, to input
Multiple abdomen body images carry out pretreatment include:
Gray threshold truncation is carried out to multiple abdomen body images of input；
Spacing normalization is carried out to the abdomen body image data after gray threshold is truncated.
In as a preferred technical scheme, in the abdomen body image liver segmentation method based on deep learning, to input
Multiple abdomen body images carry out gray threshold truncation include:
Two threshold values, respectively first threshold and second threshold are chosen, first threshold is greater than second threshold, to the more of input
A abdomen body image carries out gray threshold truncation, will be above first threshold and excludes lower than the non-liver area of second threshold
It goes.
In as a preferred technical scheme, abdomen body image includes CT or MR abdomen body image.
Specifically, trained liver segmentation convolution can be used after the training of liver segmentation convolutional neural networks
Neural network model carries out liver segmentation to CT the or MR abdomen body image of input, and steps are as follows:
Step 1: data prediction is carried out to multiple CT or MR abdomen body images (three-dimensional data) of input, it is specific to wrap
It includes:
Gray threshold truncation is carried out to multiple CT or MR abdomen body images of input, excludes the influence of non-liver area；
Spacing normalization is carried out to CT the or MR abdomen body image data after gray threshold is truncated, so that each
The physical extent of CT or MR abdomen body image data is approximately equal.
Step 2: sliding window sampling is carried out to each CT or MR abdomen body image data by data prediction, and
Each spliceosome patch that sampling is come out is sent into liver segmentation convolutional neural networks and obtains corresponding segmentation result.
Step 3: the segmentation result of the spliceosome patch of all acquisitions is spliced into a complete abdomen body image number
According to.
Below with specific embodiment, the abdomen body image liver segmentation method of the invention based on deep learning is carried out detailed
It describes in detail bright:
Step 1: according to the gray threshold and spacing selected in liver segmentation convolutional neural networks training step,
Gray threshold truncation and spacing normalization are carried out to data to be split (multiple CT or MR abdomen body images of input).
Step 2: carrying out sliding window sampling to CT the or MR abdomen body image data after spacing normalized, will
Each spliceosome patch is sent into trained liver segmentation convolutional neural networks model and obtains corresponding segmentation result.
Step 3: the segmentation result of the spliceosome patch of all acquisitions is spliced into a complete abdomen body image number
According to.
Step 4: extracting largest connected region in complete abdomen body image data, removes unrelated results, is partitioned into liver
Dirty district domain.
Embodiment 2
According to another embodiment of the present invention, a kind of abdomen body image liver segmentation dress based on deep learning is provided
It sets, referring to fig. 4, comprising:
Pretreatment unit 100, for being pre-processed to multiple abdomen body images of input；
Window sliding unit 200 obtains spliceosome for carrying out sliding window sampling to pretreated abdomen body image
Data；
Training cutting unit 300, for each splicing volume data to be sent into trained liver segmentation convolutional neural networks
Obtain corresponding segmentation result；
Segmentation result concatenation unit 400, for by the segmentation result of the splicing volume data of all acquisitions be spliced into one it is complete
Whole abdomen body image data.
The abdomen body image liver segmentation device based on deep learning in the embodiment of the present invention, to pretreated abdomen
Body image carries out sliding window sampling and obtains splicing volume data, and each splicing volume data is sent into trained liver segmentation convolution
Neural network obtains corresponding segmentation result, and multiple segmentation results are spliced into complete abdomen body image data, after being convenient for
It is continuous to be partitioned into liver area, fast, segmentation essence that the present invention is based on the abdomen body image liver segmentation device splitting speeds of deep learning
Degree is high.
In as a preferred technical scheme, referring to Fig. 5, the device further include:
Connected region extraction unit 500 is removed for extracting largest connected region in complete abdomen body image data
Divisible liver area out, segmentation precision are high after unrelated results.Compared with existing liver segmentation system, the present invention is based on depth
The advantages of abdomen body image liver segmentation method and device of study, is:
1. the liver segmentation convolutional neural networks model training time is short；
2. splitting speed is fast；
3. segmentation precision is high.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.
In the above embodiment of the invention, it all emphasizes particularly on different fields to the description of each embodiment, does not have in some embodiment
The part of detailed description, reference can be made to the related descriptions of other embodiments.
In several embodiments provided herein, it should be understood that disclosed technology contents can pass through others
Mode is realized.Wherein, system embodiment described above is only schematical, such as the division of unit, can be one kind
Logical function partition, there may be another division manner in actual implementation, such as multiple units or components can combine or can
To be integrated into another system, or some features can be ignored or not executed.Another point, shown or discussed is mutual
Coupling, direct-coupling or communication connection can be through some interfaces, the indirect coupling or communication connection of unit or module,
It can be electrical or other forms.
Unit may or may not be physically separated as illustrated by the separation member, shown as a unit
Component may or may not be physical unit, it can and it is in one place, or may be distributed over multiple units
On.It can some or all of the units may be selected to achieve the purpose of the solution of this embodiment according to the actual needs.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of software functional units.
It, can if integrated unit is realized in the form of SFU software functional unit and when sold or used as an independent product
To be stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention substantially or
Say that all or part of the part that contributes to existing technology or the technical solution can embody in the form of software products
Out, which is stored in a storage medium, including some instructions are used so that a computer equipment
(can be personal computer, server or network equipment etc.) executes all or part of step of each embodiment method of the present invention
Suddenly.And storage medium above-mentioned includes: USB flash disk, read-only memory (ROM, Read-Only Memory), random access memory
The various media that can store program code such as (RAM, Random Access Memory), mobile hard disk, magnetic or disk.
The above is only a preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications are also answered
It is considered as protection scope of the present invention.
Claims (10)
1. a kind of abdomen body image liver segmentation method based on deep learning, which comprises the following steps:
Multiple abdomen body images of input are pre-processed；
Sliding window sampling is carried out to pretreated abdomen body image and obtains splicing volume data；
Each splicing volume data is sent into trained liver segmentation convolutional neural networks and obtains corresponding segmentation result；
The segmentation result of the splicing volume data of all acquisitions is spliced into a complete abdomen body image data.
2. the method according to claim 1, wherein the method also includes:
Largest connected region is extracted in complete abdomen body image data, is partitioned into liver area after removing unrelated results.
3. the method according to claim 1, wherein the training step packet of the liver segmentation convolutional neural networks
It includes:
Gray threshold truncation is carried out to multiple abdomen body images of input；
Spacing normalization is carried out to the abdomen body image data after gray threshold is truncated；
Multiple training samples are obtained to the abdomen body image sliding sampling after progress spacing normalization；
It constructs the liver segmentation convolutional neural networks based on deep learning and liver segmentation is rolled up using multiple training samples
Product neural network is trained.
4. according to the method described in claim 3, it is characterized in that, multiple abdomen body images of described pair of input carry out gray scale threshold
Value is truncated
Two threshold values, respectively first threshold and second threshold are chosen, first threshold is greater than second threshold, to multiple abdomens of input
Portion's body image carries out gray threshold truncation, will be above first threshold and excludes lower than the non-liver area of second threshold.
5. according to the method described in claim 3, it is characterized in that, liver segmentation convolutional Neural net of the building based on deep learning
Multiple training samples are simultaneously trained liver segmentation convolutional neural networks using stochastic gradient descent optimization algorithm by network.
6. the method according to claim 1, wherein multiple abdomen body images of described pair of input pre-process
Include:
Gray threshold truncation is carried out to multiple abdomen body images of input；
Spacing normalization is carried out to the abdomen body image data after gray threshold is truncated.
7. according to the method described in claim 6, it is characterized in that, multiple abdomen body images of described pair of input carry out gray scale threshold
Value is truncated
Two threshold values, respectively first threshold and second threshold are chosen, first threshold is greater than second threshold, to multiple abdomens of input
Portion's body image carries out gray threshold truncation, will be above first threshold and excludes lower than the non-liver area of second threshold.
8. the method according to claim 1, wherein the abdomen body image includes CT or MR abdomen body image.
9. a kind of abdomen body image liver segmentation device based on deep learning characterized by comprising
Pretreatment unit, for being pre-processed to multiple abdomen body images of input；
Window sliding unit obtains splicing volume data for carrying out sliding window sampling to pretreated abdomen body image；
Training cutting unit is corresponded to for each splicing volume data to be sent into trained liver segmentation convolutional neural networks
Segmentation result；
Segmentation result concatenation unit, for the segmentation result of the splicing volume data of all acquisitions to be spliced into a complete abdomen
Body image data.
10. device according to claim 9, which is characterized in that described device further include:
Connected region extraction unit removes unrelated knot for extracting largest connected region in complete abdomen body image data
Liver area is partitioned into after fruit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910334931.8A CN110163870A (en) | 2019-04-24 | 2019-04-24 | A kind of abdomen body image liver segmentation method and device based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910334931.8A CN110163870A (en) | 2019-04-24 | 2019-04-24 | A kind of abdomen body image liver segmentation method and device based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110163870A true CN110163870A (en) | 2019-08-23 |
Family
ID=67638693
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910334931.8A Pending CN110163870A (en) | 2019-04-24 | 2019-04-24 | A kind of abdomen body image liver segmentation method and device based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110163870A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111862056A (en) * | 2020-07-23 | 2020-10-30 | 东莞理工学院 | Retinal vessel image segmentation method based on deep learning |
| CN112419343A (en) * | 2019-11-27 | 2021-02-26 | 上海联影智能医疗科技有限公司 | System and method for image segmentation |
| CN114708255A (en) * | 2022-04-29 | 2022-07-05 | 浙江大学 | Multi-center children X-ray chest image lung segmentation method based on TransUNet model |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005245830A (en) * | 2004-03-05 | 2005-09-15 | Jgs:Kk | Tumor detecting method, tumor detecting device, and program |
| CN105574859A (en) * | 2015-12-14 | 2016-05-11 | 中国科学院深圳先进技术研究院 | Liver tumor segmentation method and device based on CT (Computed Tomography) image |
| CN108038513A (en) * | 2017-12-26 | 2018-05-15 | 北京华想联合科技有限公司 | A kind of tagsort method of liver ultrasonic |
| CN108364294A (en) * | 2018-02-05 | 2018-08-03 | 西北大学 | Abdominal CT images multiple organ dividing method based on super-pixel |
| CN109102506A (en) * | 2018-08-20 | 2018-12-28 | 东北大学 | A kind of automatic division method carrying out abdominal CT hepatic disease image based on three-stage cascade network |
| CN109447120A (en) * | 2018-09-26 | 2019-03-08 | 佛山市幻云科技有限公司 | A kind of method, apparatus and computer readable storage medium of Image Automatic Segmentation |
-
2019
- 2019-04-24 CN CN201910334931.8A patent/CN110163870A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005245830A (en) * | 2004-03-05 | 2005-09-15 | Jgs:Kk | Tumor detecting method, tumor detecting device, and program |
| CN105574859A (en) * | 2015-12-14 | 2016-05-11 | 中国科学院深圳先进技术研究院 | Liver tumor segmentation method and device based on CT (Computed Tomography) image |
| CN108038513A (en) * | 2017-12-26 | 2018-05-15 | 北京华想联合科技有限公司 | A kind of tagsort method of liver ultrasonic |
| CN108364294A (en) * | 2018-02-05 | 2018-08-03 | 西北大学 | Abdominal CT images multiple organ dividing method based on super-pixel |
| CN109102506A (en) * | 2018-08-20 | 2018-12-28 | 东北大学 | A kind of automatic division method carrying out abdominal CT hepatic disease image based on three-stage cascade network |
| CN109447120A (en) * | 2018-09-26 | 2019-03-08 | 佛山市幻云科技有限公司 | A kind of method, apparatus and computer readable storage medium of Image Automatic Segmentation |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112419343A (en) * | 2019-11-27 | 2021-02-26 | 上海联影智能医疗科技有限公司 | System and method for image segmentation |
| CN111862056A (en) * | 2020-07-23 | 2020-10-30 | 东莞理工学院 | Retinal vessel image segmentation method based on deep learning |
| CN114708255A (en) * | 2022-04-29 | 2022-07-05 | 浙江大学 | Multi-center children X-ray chest image lung segmentation method based on TransUNet model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110163870A (en) | A kind of abdomen body image liver segmentation method and device based on deep learning | |
| CN104992430B (en) | Full automatic three-dimensional liver segmentation method based on convolutional neural networks | |
| EP2027566B1 (en) | Automatic recognition of preneoplastic anomalies in anatomic structures based on an improved region-growing segmentation, and computer program therefor | |
| CN111179237B (en) | Liver and liver tumor image segmentation method and device | |
| CN106355582B (en) | A method of the liver medical image segmentation based on shape prior | |
| CN110263724A (en) | Image identification method, identification model training method, device and storage medium | |
| CN110495877A (en) | A kind of Multi resolution feature extraction method and device based on ECG | |
| CN104573309A (en) | Apparatus and method for computer-aided diagnosis | |
| CN109872325B (en) | Full-automatic liver tumor segmentation method based on two-way three-dimensional convolutional neural network | |
| CN102663824A (en) | Method and system for heart isolation in cardiac computed tomography volumes | |
| CN112017185B (en) | Focus segmentation method, device and storage medium | |
| CN109801268B (en) | CT radiography image renal artery segmentation method based on three-dimensional convolution neural network | |
| CN105913432A (en) | Aorta extracting method and aorta extracting device based on CT sequence image | |
| CN104035563B (en) | W-PCA (wavelet transform-principal component analysis) and non-supervision GHSOM (growing hierarchical self-organizing map) based electrocardiographic signal identification method | |
| CN109452959B (en) | Traceless layered extraction method and device | |
| CN110009656B (en) | Target object determination method and device, storage medium and electronic device | |
| CN109584223A (en) | Pulmonary vascular dividing method in CT image | |
| CN110428887A (en) | A kind of brain tumor medical image three-dimensional reconstruction shows exchange method and system | |
| CN111383215A (en) | Focus detection model training method based on generation of confrontation network | |
| CN109785399B (en) | Synthetic lesion image generation method, device, equipment and readable storage medium | |
| CN109124620A (en) | A kind of atrial fibrillation detection method, device and equipment | |
| CN109544528A (en) | A kind of small pulmonary artery image-recognizing method and device | |
| CN114365188A (en) | Analysis method and product based on VRDS AI inferior vena cava image | |
| CN113012249A (en) | Method, device and storage medium for generating focus on CT image | |
| CN104915989A (en) | CT image-based blood vessel three-dimensional 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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190823 |