CN110533573A - A kind of MRI image method for parallel processing and processing system based on JAVA language - Google Patents
A kind of MRI image method for parallel processing and processing system based on JAVA language Download PDFInfo
- Publication number
- CN110533573A CN110533573A CN201910829802.6A CN201910829802A CN110533573A CN 110533573 A CN110533573 A CN 110533573A CN 201910829802 A CN201910829802 A CN 201910829802A CN 110533573 A CN110533573 A CN 110533573A
- Authority
- CN
- China
- Prior art keywords
- image
- mri image
- data
- processing
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T1/00—General purpose image data processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- 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]
Abstract
The present invention provides a kind of MRI image method for parallel processing and processing system based on JAVA language, which is characterized in that the treating method comprises following steps: start；Upload patient's MRI image DICOM data compression packet；The compressed package is decompressed, MRI image brain structured file data are obtained；Extract MRI image brain structured file Data Analysis Services；Processing result is collected and surveyed, standard form data is corrected stage by stage according to a large amount of statistical data, provides statistic analysis result；Terminate.The invention has the beneficial effects that: the processing method and processing system, based on image processing function library of increasing income, rewrite the MRI image processing method of more machine parallel processings, and pass through a large amount of MRI datas of machine learning, it forms Chinese's MRI image and handles mould shape, it finally realizes and improves processing speed, efficiency, the purpose of precision, and realize the high-precision MRI image processing for Chinese.
Description
Technical field
The invention belongs to MRI image processing technology fields, and in particular to a kind of MRI image based on JAVA language is located parallel
Reason method and processing system.
Background technique
The existing method to MRI image processing is mostly based on MATLAB platform under single machine, and image processing template is more
Number is that American-European and Southeast Asia ethnic group is object, and image-capable and computation model are only used for research field.
But the processing method of the prior art, there is a problem of that treatment effeciency, MRI image processing accuracy are relatively low, and do not have
There is the high-precision MRI image processing method for Chinese.
Summary of the invention
In order to overcome the problems, such as the presence of the prior art, the present invention provides a kind of MRI images based on JAVA language simultaneously
Row processing method and processing system rewrite the MRI of more machine parallel processings based on image processing function library of increasing income
Image processing method, and by a large amount of MRI datas of machine learning, it forms Chinese's MRI image and handles mould shape, finally realize and mention
High processing rate, efficiency, the purpose of precision, and realize the high-precision MRI image processing for Chinese.
A kind of MRI image method for parallel processing based on JAVA language provided by the invention, comprising the following steps:
(1) start；
(2) patient's MRI image DICOM data compression packet is uploaded；
(3) aforementioned compression packet is decompressed, MRI image brain structured file data is obtained, specifically includes following 6 steps:
1) non-uniform intensity normalizes
It is uniformly processed in order to facilitate computer is carried out, needs to eliminate machine characteristic, image intensity is normalized to (0,255)
In range, the gray scale normalization method of image content-based feature is used；
By the label range image of priori knowledge acquisition tensor representation:
X∈Rp×q×r×m, α < X < β, wherein α, β are the range boundary of machine characteristic；
Directly be set to 0 and 255 after calculating, α respectively corresponded to according to content characteristic, the part range in β be mapped to (0,
255), in X1, X2... Xn is normalized respectively；
Then the contrast of foregoing tags range image can be expressed as
Wherein, XI, j, k, lIndicate the intensity of first of feature of the position three-dimensional space (i, j, k), XI, j, k, l=1 represents red,
XI, j, k, l=2 represent green, XI, j, k, l=3 represent blue, XI, j, k, l=4 represent transparency；
It is the mean intensity of foregoing tags range image, is expressed as
According to magnetic field imaging signal height, using m=1, above formula becomes
2) precision for carrying out three directions using three interpolation algorithm of convolution is unified
F (i+u, j+v)=[A] × [B] × [C]
[A]=[S (u+1) S (u+0) S (u-1) S (u-2)]
Tri- directional interpolation variables of X, Y, Z: x=abs (1- Φ x), y=abs (1- Φ y), z=abs (1- Φ z)；
Wherein, Φ x, Φ y, Φ z respectively indicate the precision in three directions；
3) coordinate system spatial alternation is carried out
Pass through Image Position (Patient)-(0020,0032) and Image in dicom standard
Two fields of Orientation (Patient)-(0020,0037) determine the space orientation of patient Patient；Wherein Image
Position (Patient) indicates (x, y, the z) coordinate of the upper left corner of image in space coordinates, and unit is millimeter；Image
The cosine value of Orientation (Patient) expression image coordinate and anatomical coordinates system respective coordinates angle；In order to obtain
With the corresponding displacement X of a voxel location point (x, y, z), Δ Y, Δ Z in hypothesis dicom image, (x is seti,yi,zi) it is the
I mark point coordinate, N are mark point number, and m takes 2 rank battens, then has
Δ y (x, y, z), Δ z (x, y, z) are calculated using the above method；
4) smooth using mean-shift algorithm
N sample point x in given d dimension space Rdi, i=1,2,3, n, the vector citation form at x are as follows:
Wherein SkWhat is indicated is data point of the point of data set to the distance of x less than radius of a ball h, that is,
Sh(x)={ y:(y-xi)T(y-xi) < h2}
Shift vectors are obtained by calculation, update the position of ball center of circle x are as follows:
X:=x+Mh
Each iteration finds the average bit that circle the inside is put so that the center of circle is mobile toward the maximum direction of data set density always
It sets as new center location, by the way that Gauss weight is added, obtains shift vectors calculation formula are as follows:
Calculate the central coordinate of circle updated every time are as follows:
It is done smoothly, is had using Gaussian kernel
5) Region growing segmentation algorithm segmentation grey matter and white matter
R is enabled to indicate entire 3d image array, then segmentation can be regarded as region R being divided into n sub-regions R1,
R2,......RnProcess, and it is necessary to meet following condition:
a:U(Ri)=R；
b:RiIt is a connected region, i=1,2,3 ... n；
c:Ri∩Rj=empty set, for any i, j；There is i ≠ j；
d:P(Ri)=Ture, to i=1,2 ... n；
e:R(Pi U Rj)=False, i ≠ j；
Above-mentioned segmentation must be the segmentation carried out as unit of voxel, otherwise it cannot be guaranteed that being divided the continuity of object,
Cause to divide imperfect；In order to improve cutting procedure efficiency, treatment process can be gone to cuda, take out knot after the completion of processing
Fruit；
6) segmentation result is registrated on Standard Segmentation brain
Shaft-like vector data f (x, y) is taken, the shaft-like vector data f (x ', y ') of standard, transformation model are registrated to are as follows:
Wherein α is zooming parameter；θ is rotation angle parameter, and (Δ x, Δ y) are translation parameters；
Image wavelet transform function are as follows:
Set H={ hnIt is low-pass filter, G={ gnIt is high-pass filter, then it is calculated and is decomposed by Mallat algorithm
Number, has
Sampled picture is registrated to standard picture by three kinds of flexible, translation, rotation transformation, wherein stretching calculates:
Translation transformation calculates: setting p=k-2m, q=l-2n, has
Rotation transformation calculates: setting p=-k cos θ+l sin θ, q=-k sin θ+l cos θ has
cM+1(2m cos θ+2n sin θ, -2m sin θ+2n cos θ)
Become through wavelet low-pass filtering:
After registration transformation, detailed brain structured file data can be gone out according to canonical statistics；
(4) MRI image data combination brain structured file Data Analysis Services are extracted；
(5) processing result is collected and surveyed, standard form data is corrected stage by stage according to a large amount of statistical data, provides statistical
Analyse result；
(6) terminate.
Preferably, abovementioned steps 5) Region growing segmentation algorithm segmentation grey matter and the treatment process of white matter can go to
Cuda further takes out result after the completion of processing.
A kind of MRI image parallel processing system (PPS) based on JAVA language, simultaneously using the aforementioned MRI image based on JAVA language
Row processing method establishes multiple subprograms in conjunction with server CPU, memory by JDK, while handling multiple image datas.
The invention has the beneficial effects that:
1. a kind of MRI image method for parallel processing based on JAVA language of the invention may be implemented accurate with open source technology
Analysis processing MRI image；
2. a kind of MRI image parallel processing system (PPS) based on JAVA language of the invention uses networking technology with more machines
Device parallel processing MRI image, significantly improves processing speed；
3. a kind of MRI image method for parallel processing and processing system based on JAVA language of the invention, with image of increasing income
Based on handling function library, the MRI image processing method of more machine parallel processings is rewritten, and a large amount of by machine learning
MRI data forms Chinese's MRI image and handles mould shape, finally realizes and improves processing speed, efficiency, the purpose of precision, and
The high-precision MRI image realized for Chinese is handled.
Detailed description of the invention
Fig. 1 is a kind of general flow chart of MRI image method for parallel processing based on JAVA language of the invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts all other
Embodiment shall fall within the protection scope of the present invention.
Embodiment 1
Referring to attached drawing 1, a kind of MRI image method for parallel processing based on JAVA language, comprising the following steps:
(1) start；
(2) patient's MRI image DICOM data compression packet is uploaded；
(3) compressed package is decompressed, MRI image brain structured file data is obtained, specifically includes following 6 steps:
1) non-uniform intensity normalizes
It is uniformly processed in order to facilitate computer is carried out, needs to eliminate machine characteristic, image intensity is normalized to (0,255)
In range, the gray scale normalization method of image content-based feature is used；
By the label range image of priori knowledge acquisition tensor representation:
X∈Rp×q×r×m, a < X < β, wherein α, β are the range boundary of machine characteristic；
Directly be set to 0 and 255 after calculating, α respectively corresponded to according to content characteristic, the part range in β be mapped to (0,
255), in X1, X2... Xn is normalized respectively；
Then the contrast of label range image can indicate are as follows:
Wherein, XI, j, k, lIndicate the intensity of first of feature of the position three-dimensional space (i, j, k), XI, j, k, l=1 represents red,
XI, j, k, l=2 represent green, XI, j, k, l=3 represent blue, XI, j, k, l=4 represent transparency；
It is the mean intensity of label range image, indicates are as follows:
According to magnetic field imaging signal height, using m=1, above formula becomes:
2) precision for carrying out three directions using three interpolation algorithm of convolution is unified
F (i+u, j+v)=[A] × [B] × [C]
[A]=[S (u+1) S (u+0) S (u-1) S (u-2)]
Tri- directional interpolation variables of X, Y, Z: x=abs (1- Φ x), y=abs (1- Φ y), z=abs (1- Φ z)；
Wherein, Φ x, Φ y, Φ z respectively indicate the precision in three directions；
3) coordinate system spatial alternation is carried out
Pass through Image Position (Patient)-(0020,0032) and Image in dicom standard
Two fields of Orientation (Patient)-(0020,0037) determine the space orientation of patient Patient；Wherein Image
Position (Patient) indicates (x, y, the z) coordinate of the upper left corner of image in space coordinates, and unit is millimeter；Image
The cosine value of Orientation (Patient) expression image coordinate and anatomical coordinates system respective coordinates angle；In order to obtain
With the corresponding displacement X of a voxel location point (x, y, z), Δ Y, Δ Z in hypothesis dicom image, (x is seti,yi,zi) it is the
I mark point coordinate, N are mark point number, and m takes 2 rank battens, then has:
θ y (x, y, z), Δ z (x, y, z) are calculated using the above method；
4) smooth using mean-shift algorithm
N sample point x in given d dimension space Rdi, i=1,2,3, n, the vector citation form at x are as follows:
Wherein SkWhat is indicated is data point of the point of data set to the distance of x less than radius of a ball h, that is,
Sh(x)={ y:(y-xi)T(y-xi) < h2}
Shift vectors are obtained by calculation, update the position of ball center of circle x are as follows:
X:=x+Mh
Each iteration finds the average bit that circle the inside is put so that the center of circle is mobile toward the maximum direction of data set density always
It sets as new center location, by the way that Gauss weight is added, obtains shift vectors calculation formula are as follows:
Calculate the central coordinate of circle updated every time are as follows:
It is done smoothly, is had using Gaussian kernel:
5) Region growing segmentation algorithm segmentation grey matter and white matter
R is enabled to indicate entire 3d image array, then segmentation can be regarded as region R being divided into n sub-regions R1,
R2,......RnProcess, and it is necessary to meet following condition:
a:U(Ri)=R；
b:RiIt is a connected region, i=1,2,3 ... n；
c:Ri∩Rj=empty set, for any i, j；There is i ≠ j；
d:P(Ri)=Ture, to i=1,2 ... n；
e:R(Pi U Rj)=False, i ≠ j；
Above-mentioned segmentation must be the segmentation carried out as unit of voxel, otherwise it cannot be guaranteed that being divided the continuity of object,
Cause to divide imperfect；In order to improve cutting procedure efficiency, treatment process can be gone to cuda, take out knot after the completion of processing
Fruit；
6) segmentation result is registrated on Standard Segmentation brain
Shaft-like vector data f (x, y) is taken, the shaft-like vector data f (x ', y ') of standard, transformation model are registrated to are as follows:
Wherein α is zooming parameter；θ is rotation angle parameter, and (Δ x, Δ y) are translation parameters；
Image wavelet transform function are as follows:
Set H={ hnIt is low-pass filter, G={ gnIt is high-pass filter, then it is calculated and is decomposed by Mallat algorithm
Number, has:
Sampled picture is registrated to standard picture by three kinds of flexible, translation, rotation transformation, wherein stretching calculates:
Translation transformation calculates:
If p=k-2m, q=l-2n have:
Rotation transformation calculates:
If p=k cos θ+l sin θ, q=-k sin θ+l cos θ, have:
cM+1(2m cos θ+2n sin θ, -2m sin θ+2n cos θ)
Become through wavelet low-pass filtering:
After registration transformation, detailed MRI image brain structured file data can be gone out according to canonical statistics；
(4) MRI image brain structured file Data Analysis Services are extracted；
(5) processing result is collected and surveyed, standard form data is corrected stage by stage according to a large amount of statistical data, provides statistical
Analyse result；
(6) terminate.
Present invention disclosed above preferred embodiment is only intended to help to illustrate the present invention.There is no detailed for preferred embodiment
All details are described, also do not limit the specific embodiment that the invention is only.Obviously, according to the content of this specification, can make
Many modifications and variations.These embodiments are chosen and specifically described to this specification, is original in order to better explain the present invention
Reason and practical application, so that skilled artisan be enable to better understand and utilize the present invention.The present invention is only authorized
The limitation of sharp claim and its full scope and equivalent.
Claims (3)
1. a kind of MRI image method for parallel processing based on JAVA language, which is characterized in that the treating method comprises following step
It is rapid:
(1) start；
(2) patient's MRI image DICOM data compression packet is uploaded；
(3) compressed package is decompressed, MRI image brain structured file data is obtained, specifically includes following 6 steps:
1) non-uniform intensity normalizes
It is uniformly processed in order to facilitate computer is carried out, needs to eliminate machine characteristic, image intensity is normalized to (0,255) range
It is interior, use the gray scale normalization method of image content-based feature；
By the label range image of priori knowledge acquisition tensor representation:
X∈Rp×q×r×m, α < X < β, wherein α, β are the range boundary of machine characteristic；
Directly be set to 0 and 255 after calculating, α respectively corresponded to according to content characteristic, the part range in β be mapped to (0,
255), in X1, X2... Xn is normalized respectively；
Then the contrast of the label range image can indicate are as follows:
Wherein, XI, j, k, 1Indicate the intensity of first of feature of the position three-dimensional space (i, j, k), XI, j, k, 1=1 represents red,
XI, j, k, l=2 represent green, XI, j, k, 1=3 represent blue, XI, j, k, 1=4 represent transparency；
It is the mean intensity of the label range image, indicates are as follows:
According to magnetic field imaging signal height, using m=1, above formula becomes:
2) precision for carrying out three directions using three interpolation algorithm of convolution is unified
F (i+u, j+v)=[A] × [B] × [C]
[A]=[S (u+1) S (u+0) S (u-1) S (u-2)]
Tri- directional interpolation variables of X, Y, Z: x=abs (1- Φ x), y=abs (1- Φ y), z=abs (1- Φ z)；
Wherein, Φ x, Φ y, Φ z respectively indicate the precision in three directions；
3) coordinate system spatial alternation is carried out
Pass through Image Position (Patient)-(0020,0032) and Image Orientation in dicom standard
(Patient)-(0020,0037) two fields determine the space orientation of patient Patient；Wherein Image Position
(Patient) (x, y, the z) coordinate of the upper left corner of image in space coordinates is indicated, unit is millimeter；Image
The cosine value of Orientation (Patient) expression image coordinate and anatomical coordinates system respective coordinates angle；In order to obtain
With the corresponding displacement X of a voxel location point (x, y, z), Δ Y, Δ Z in hypothesis dicom image, (x is seti,yi,zi) it is the
I mark point coordinate, N are mark point number, and m takes 2 rank battens, then has:
Δ y (x, y, z), Δ z (x, y, z) are calculated using the above method；
4) smooth using mean-shift algorithm
N sample point x in given d dimension space Rdi, i=1,2,3, n, the vector citation form at x are as follows:
Wherein SkWhat is indicated is data point of the point of data set to the distance of x less than radius of a ball h, that is,
Sh(x)={ y:(y-xi)T(y-xi) < h2}
Shift vectors are obtained by calculation, update the position of ball center of circle x are as follows:
X:=x+Mh
Each iteration finds the mean place that circle the inside is put and makees so that the center of circle is mobile toward the maximum direction of data set density always
Shift vectors calculation formula is obtained by the way that Gauss weight is added for new center location are as follows:
Calculate the central coordinate of circle updated every time are as follows:
It is done smoothly, is had using Gaussian kernel:
5) Region growing segmentation algorithm segmentation grey matter and white matter
R is enabled to indicate entire 3d image array, then segmentation can be regarded as region R being divided into n sub-regions R1,R2,......Rn
Process, and it is necessary to meet following condition:
a:U(Ri)=R；
b:RiIt is a connected region, i=1,2,3 ... n；
c:Ri∩Rj=empty set, for any i, j；There is i ≠ j；
d:P(Ri)=Ture, to i=1,2 ... n；
e:R(PiURj)=False, i ≠ j；
Above-mentioned segmentation must be the segmentation carried out as unit of voxel, otherwise it cannot be guaranteed that being divided the continuity of object, cause
Divide imperfect；In order to improve cutting procedure efficiency, treatment process can be gone to cuda, take out result after the completion of processing；
6) segmentation result is registrated on Standard Segmentation brain
Shaft-like vector data f (x, y) is taken, the shaft-like vector data f (x ', y ') of standard, transformation model are registrated to are as follows:
Wherein α is zooming parameter；θ is rotation angle parameter, and (Δ x, Δ y) are translation parameters；
Image wavelet transform function are as follows:
Set H={ hnIt is low-pass filter, G={ gnIt is high-pass filter, then it is calculated by Mallat algorithm and decomposes number, had:
Sampled picture is registrated to standard picture by three kinds of flexible, translation, rotation transformation, wherein stretching calculates:
Translation transformation calculates:
If p=k-2m, q=l-2n have:
Rotation transformation calculates:
If p=k cos θ+l sin θ, q=-k sin θ+l cos θ, have:
cM+1(2m cos θ+2n sin θ, -2m sin θ+2n cos θ)
Become through wavelet low-pass filtering:
After registration transformation, detailed MRI image brain structured file data can be gone out according to canonical statistics；
(4) MRI image brain structured file Data Analysis Services are extracted；
(5) processing result is collected and surveyed, corrects standard form data stage by stage according to a large amount of statistical data, provides statistical analysis knot
Fruit；
(6) terminate.
2. a kind of MRI image parallel processing system (PPS) based on JAVA language, which is characterized in that using described based on JAVA language
MRI image method for parallel processing establishes multiple subprograms in conjunction with server CPU, memory by JDK, while handling multiple images
Data.
3. a kind of MRI image method for parallel processing based on JAVA language according to claim 1, which is characterized in that institute
The treatment process of the Region growing segmentation algorithm segmentation grey matter and white matter of stating step 5) can go to cuda, take again after the completion of processing
Result out.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910829802.6A CN110533573B (en) | 2019-09-04 | 2019-09-04 | MRI image parallel processing method and processing system based on JAVA language |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910829802.6A CN110533573B (en) | 2019-09-04 | 2019-09-04 | MRI image parallel processing method and processing system based on JAVA language |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110533573A true CN110533573A (en) | 2019-12-03 |
| CN110533573B CN110533573B (en) | 2023-07-14 |
Family
ID=68666699
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910829802.6A Active CN110533573B (en) | 2019-09-04 | 2019-09-04 | MRI image parallel processing method and processing system based on JAVA language |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110533573B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111784664A (en) * | 2020-06-30 | 2020-10-16 | 广州柏视医疗科技有限公司 | Method for generating distribution map of tumor lymph nodes |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140016844A1 (en) * | 2012-07-10 | 2014-01-16 | Zakrytoe Akcionernoe Obshchestvo "Impul's" | Method for acquisition of subtraction angiograms |
| CN103942780A (en) * | 2014-03-27 | 2014-07-23 | 北京工业大学 | Fuzzy-connectedness-algorithm-based segmentation method of thalamus and substructures of thalamus |
| CN104376569A (en) * | 2014-11-28 | 2015-02-25 | 成都影泰科技有限公司 | Method for processing DICOM (digital imaging and communications in medicine) medical images on basis of versions |
| CN107507212A (en) * | 2017-08-18 | 2017-12-22 | 中国科学院深圳先进技术研究院 | Digital brain method for visualizing, device, computing device and storage medium |
| CN108122263A (en) * | 2017-04-28 | 2018-06-05 | 上海联影医疗科技有限公司 | Image re-construction system and method |
-
2019
- 2019-09-04 CN CN201910829802.6A patent/CN110533573B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140016844A1 (en) * | 2012-07-10 | 2014-01-16 | Zakrytoe Akcionernoe Obshchestvo "Impul's" | Method for acquisition of subtraction angiograms |
| CN103942780A (en) * | 2014-03-27 | 2014-07-23 | 北京工业大学 | Fuzzy-connectedness-algorithm-based segmentation method of thalamus and substructures of thalamus |
| CN104376569A (en) * | 2014-11-28 | 2015-02-25 | 成都影泰科技有限公司 | Method for processing DICOM (digital imaging and communications in medicine) medical images on basis of versions |
| CN108122263A (en) * | 2017-04-28 | 2018-06-05 | 上海联影医疗科技有限公司 | Image re-construction system and method |
| CN107507212A (en) * | 2017-08-18 | 2017-12-22 | 中国科学院深圳先进技术研究院 | Digital brain method for visualizing, device, computing device and storage medium |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111784664A (en) * | 2020-06-30 | 2020-10-16 | 广州柏视医疗科技有限公司 | Method for generating distribution map of tumor lymph nodes |
| CN111784664B (en) * | 2020-06-30 | 2021-07-20 | 广州柏视医疗科技有限公司 | Method for generating distribution map of tumor lymph nodes |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110533573B (en) | 2023-07-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107220995B (en) | Improved method of ICP (inductively coupled plasma) rapid point cloud registration algorithm based on ORB (object-oriented bounding Box) image features | |
| US9280825B2 (en) | Image processing system with registration mechanism and method of operation thereof | |
| CN107194937B (en) | Traditional Chinese medicine tongue picture image segmentation method in open environment | |
| Khagi et al. | Pixel-label-based segmentation of cross-sectional brain MRI using simplified SegNet architecture-based CNN | |
| US8861891B2 (en) | Hierarchical atlas-based segmentation | |
| CN113763314A (en) | System and method for image segmentation and classification using depth-reduced convolutional neural networks | |
| CN112183541B (en) | Contour extraction method and device, electronic equipment and storage medium | |
| Liu | Stbi-yolo: A real-time object detection method for lung nodule recognition | |
| CN112085714A (en) | Pulmonary nodule detection method, model training method, device, equipment and medium | |
| CN112750531A (en) | Automatic inspection system, method, equipment and medium for traditional Chinese medicine | |
| CN114927215B (en) | Method and system for directly predicting tumor respiratory motion based on body surface point cloud data | |
| CN110321908A (en) | Image-recognizing method, terminal device and computer readable storage medium | |
| Pawar et al. | Two-stage hybrid approach of deep learning networks for interstitial lung disease classification | |
| Xue et al. | Lung 4D CT image registration based on high-order markov random field | |
| CN113706514B (en) | Focus positioning method, device, equipment and storage medium based on template image | |
| CN110533573A (en) | A kind of MRI image method for parallel processing and processing system based on JAVA language | |
| CN115761226A (en) | Oral cavity image segmentation identification method and device, electronic equipment and storage medium | |
| CN109712124B (en) | Ultrasonic image mark removing method and device | |
| Bardosi et al. | Metacarpal bones localization in x-ray imagery using particle filter segmentation | |
| Chai et al. | Automatic segmentation of adrenal tumor in CT images based on sparse representation | |
| Wei et al. | An algorithm for segmentation of lung ROI by mean-shift clustering combined with multi-scale HESSIAN matrix dot filtering | |
| CN108388830B (en) | Animal body shape detection method and device based on convolutional neural network | |
| Liu et al. | Soft Tissue/Bone Decomposition of Conventional Chest Radiographs Using Nonparametric Image Priors | |
| CN115619835B (en) | Heterogeneous three-dimensional observation registration method, medium and equipment based on depth phase correlation | |
| Liu et al. | Using CNN With Handcrafted Features for Prostate Cancer Classification |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |