CN110211089A - Extracting method, device and the storage medium of clear cell carcinoma of kidney transfer judging characteristic - Google Patents
Extracting method, device and the storage medium of clear cell carcinoma of kidney transfer judging characteristic Download PDFInfo
- Publication number
- CN110211089A CN110211089A CN201910327345.0A CN201910327345A CN110211089A CN 110211089 A CN110211089 A CN 110211089A CN 201910327345 A CN201910327345 A CN 201910327345A CN 110211089 A CN110211089 A CN 110211089A
- Authority
- CN
- China
- Prior art keywords
- image
- cell carcinoma
- clear cell
- kidney
- frequency
- 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
- 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
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20048—Transform domain processing
-
- 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
- G06T2207/20092—Interactive image processing based on input by user
- G06T2207/20104—Interactive definition of region of interest [ROI]
-
- 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
- G06T2207/30096—Tumor; Lesion
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses extracting method, device and the storage medium of a kind of clear cell carcinoma of kidney transfer judging characteristic, tumor region has been delineated in the first CT image of the clear cell carcinoma of kidney patient obtained in the present invention;The frequency-domain transform method intermediate frequency domain conversion coefficient used has non-Gaussian feature, and each frequency domain sub-band coefficient is symmetrical all near zero, so carrying out frequency-domain transform to the first CT image based on preset frequency-domain transform method obtains corresponding 2nd CT image, it can be regarded as carrying out the first CT image of clear cell carcinoma of kidney patient the description under a variety of resolution ratio, so as to provide local physical feature of the first CT image under different spectral for user, so that the tumor region in the 2nd CT image becomes apparent from, to be easier and more accurately shift judging characteristic to the preset clear cell carcinoma of kidney of the 2nd CT image zooming-out, data are provided for clear cell carcinoma of kidney transfer judgement to support, be conducive to be promoted the accuracy of clear cell carcinoma of kidney transfer judgement.
Description
Technical field
This application involves technical field of medical image processing more particularly to a kind of clear cell carcinoma of kidney to shift judging characteristic
Extracting method, device and storage medium.
Background technique
Clear-cell carcinoma is one of most common malignant tumour of urinary system, and annual increase newly in the whole world is diagnosed as more than 350,000 people
Renal cell carcinoma patients, and the death of 140,000 people or more is caused every year.According to clear-cell carcinoma pathological, clear-cell carcinoma is mainly wrapped
Include: clear cell carcinoma of kidney, papillary cell carcinoma, chromophobe cell tumor and unfiled clear-cell carcinoma etc., wherein clear cell carcinoma of kidney is
Main subtypes account for 70% or so of RCC total amount.The cause of disease of clear cell carcinoma of kidney is unknown, mainly takes radical-ability at present
Kidney surgery or the scheme of Part Nephrectomy operation are treated, and the median survival interval of patients with terminal is about 18-24 months.To the greatest extent
Opposite life cycle after pipe clear cell carcinoma of kidney patient treatment in the past 30 years is constantly promoted, but the transfer wind of clear cell carcinoma of kidney
Danger cures to patient and brings great challenge.
Clinical research discovery, there are far-end transfers by the clear cell carcinoma of kidney patient more than 17%.Common metastasis site packet
Lung, bone and brain tissue are included, furthermore clear cell carcinoma of kidney may also occur at positions such as adrenal gland, opposite side kidney and livers
Far-end transfer.The poor prognosis of metastatic clear cell carcinoma of kidney patient, median survival interval only have about 12 months, 5 years survival rates
Also below 10%, the patient's triennial for not carrying out any treatment deposits rate less than 5%.The judgement that clear cell carcinoma of kidney shifts risk can
Foundation is provided to formulate targeted treatment schemes in advance for doctor, can effectively improve the pre- of metastatic clear cell carcinoma of kidney patient
Afterwards.
Summary of the invention
Extracting method, device and the storage that the embodiment of the present application provides a kind of clear cell carcinoma of kidney transfer judging characteristic are situated between
Matter, the judgement that can be shifted risk for clear cell carcinoma of kidney patient provide data support, are promoted and turned to clear cell carcinoma of kidney patient
Move the accuracy of judgement degree of risk.
The embodiment of the present application first aspect provides a kind of extracting method of clear cell carcinoma of kidney transfer judging characteristic, comprising:
CT image based on clear cell carcinoma of kidney patient, obtain the first CT image, wherein the first CT image by
Tumor region is delineated in the CT image to obtain;
Frequency-domain transform is carried out to the first CT image based on preset frequency-domain transform method and obtains corresponding 2nd CT figure
Picture, wherein in the preset frequency-domain transform method, frequency-domain transform coefficient has non-Gaussian feature, and each frequency domain sub-band coefficient
It is symmetrical near zero;
Based on the tumor region, judging characteristic is shifted to the preset clear cell carcinoma of kidney of the 2nd CT image zooming-out.
The embodiment of the present application second aspect provides a kind of extraction element of clear cell carcinoma of kidney transfer judging characteristic, comprising:
Image collection module obtains the first CT image, wherein institute for the CT image based on clear cell carcinoma of kidney patient
The first CT image is stated to obtain by delineating tumor region in the CT image;
Frequency-domain transform module is obtained for carrying out frequency-domain transform to the first CT image based on preset frequency-domain transform method
To corresponding 2nd CT image, wherein in the preset frequency-domain transform method, frequency-domain transform coefficient has non-Gaussian feature,
And each frequency domain sub-band coefficient is symmetrical near zero;
Characteristic extracting module, it is transparent to the preset kidney of the 2nd CT image zooming-out thin for being based on the tumor region
Born of the same parents' metastasis of cancer judging characteristic.
The embodiment of the present application third aspect provides a kind of extraction element of clear cell carcinoma of kidney transfer judging characteristic, comprising:
Memory, processor and it is stored in the computer program that can be run on the memory and on the processor, the processing
When device executes the computer program, the step in the method for the embodiment of the present application first aspect offer is realized.
The embodiment of the present application fourth aspect provides a kind of storage medium, is stored thereon with computer program, which is characterized in that
When the computer program is executed by processor, the step in the method for the embodiment of the present application first aspect offer is realized.
The embodiment of the invention provides a kind of extracting method of clear cell carcinoma of kidney transfer judging characteristic, device and storages to be situated between
Matter has delineated tumor region in the first CT image for the clear cell carcinoma of kidney patient that the present invention obtains;The frequency-domain transform side used
Method intermediate frequency domain conversion coefficient has non-Gaussian feature, and each frequency domain sub-band coefficient is symmetrical all near zero, so based on pre-
If frequency-domain transform method to the first CT image carry out frequency-domain transform obtain corresponding 2nd CT image, can be regarded as saturating to kidney
The first CT image of clear cell carcinoma patient carries out the description under a variety of resolution ratio, so as to provide the first CT image for user
Local physical feature under different spectral, so that the tumor region in the 2nd CT image becomes apparent from, thus more easily and more
Judging characteristic accurately is shifted to the preset clear cell carcinoma of kidney of the 2nd CT image zooming-out, is mentioned for clear cell carcinoma of kidney transfer judgement
It is supported for data, is conducive to the accuracy for promoting clear cell carcinoma of kidney transfer judgement.
Detailed description of the invention
Fig. 1 is the stream for the extracting method that a kind of clear cell carcinoma of kidney that the application first embodiment provides shifts judging characteristic
Journey schematic diagram;
Fig. 2 is that the principle for the extracting method that the clear cell carcinoma of kidney that the application first embodiment provides shifts judging characteristic is shown
It is intended to;
Fig. 3 is the CT image before the image registration that the application first embodiment provides;
Fig. 4 is the CT image that the CT image in Fig. 3 obtains after image registration;
Fig. 5 is the knot for the extraction element that a kind of clear cell carcinoma of kidney that the application second embodiment provides shifts judging characteristic
Structure schematic diagram;
Fig. 6 is the extraction element that another clear cell carcinoma of kidney that the application second embodiment provides shifts judging characteristic
Structural schematic diagram.
Specific embodiment
To enable present invention purpose, feature, advantage more obvious and understandable, below in conjunction with the application
Attached drawing in embodiment, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described reality
Applying example is only some embodiments of the present application, and not all embodiments.Based on the embodiment in the application, those skilled in the art
Member's every other embodiment obtained without making creative work, shall fall in the protection scope of this application.
First embodiment:
Precisely judged to realize to shift risk to clear cell carcinoma of kidney patient, the present invention provides one kind to be directed to kidney
The Features extracting method of the CT image of clear cell carcinoma patient can use priori knowledge library model of mind, to patient
CT image carry out frequency domain conversion, pass through feature extraction operator and realize that the clear cell carcinoma of kidney in CT image shifts judging characteristic
It extracts, provides data for the judgement judgement that clear cell carcinoma of kidney shifts risk and support.
Referring to Fig. 1, the embodiment of the present invention proposes a kind of extracting method of clear cell carcinoma of kidney transfer judging characteristic, the extraction
Method includes:
Step 101, the CT image based on clear cell carcinoma of kidney patient obtain the first CT image, wherein the first CT image is logical
It crosses and delineates tumor region in CT image and obtain;
Step 102 obtains corresponding second to the first CT image progress frequency-domain transform based on preset frequency-domain transform method
CT image, wherein in preset frequency-domain transform method, frequency-domain transform coefficient has non-Gaussian feature, and each frequency domain sub-band coefficient
It is symmetrical near zero;
Step 103 is based on tumor region, shifts judging characteristic to the preset clear cell carcinoma of kidney of the 2nd CT image zooming-out.
In the present embodiment, the CT image in step 101 can be the CT image of arbitrary sequence, including but not limited to unenhanced,
The CT image of the timing such as arterial phase, venous phase and parenchymal phase.
In one embodiment, may there was only a kind of CT image in step 101, such as plain CT image.But in another reality
It applies in example, in order to which the judgement shifted to clear cell carcinoma of kidney is more acurrate, it is possible that will use the CT image of several different timing,
Such scene can guarantee different timing CT images by some preprocessing means to CT image before step 101
Consistency.Optionally, if in the CT image of clear cell carcinoma of kidney patient including the CT image of different timing, transparent based on kidney
The CT image of carcinoma patients, before obtaining the first CT image, further includes: the CT of several timing of clear cell carcinoma of kidney patient is schemed
As carrying out image registration.Image registration processing in the present embodiment, may be implemented the matching on the unified scale to CT image, really
The CT image for protecting the different timing of same patient is consistent in the number of plies and resolution ratio.
Optionally, it includes: transparent to kidney for carrying out image registration to the CT image of several timing of clear cell carcinoma of kidney patient
The CT image of several timing of carcinoma patients carries out image registration by benchmark image of wherein top-quality CT image.CT
The quality of image can be measured with image definition, noise, tumour clarity and tumor's profiles etc..In one example,
Arterial phase CT image be can choose as benchmark image.
In the present embodiment, to principle such as Fig. 2 of the feature extraction of the CT image of several timing of clear cell carcinoma of kidney patient
Shown, image preprocessing section point includes the tumor region contouring steps in above-mentioned CT process of image registration and step 101;CT
Image characteristics extraction part includes that step is extracted in the calculating of frequency-domain transform step (step 102) and feature based on priori knowledge
Suddenly (step 103).
In view of the advantage and disadvantage of CT image detection cost, time and each timing CT image, an example of the present embodiment
In, the CT image that can choose unenhanced, arterial phase and venous phase these three timing implements the clear cell carcinoma of kidney of the present embodiment
Shift the extracting method of judging characteristic.Optionally, image is carried out to the CT image of several timing of clear cell carcinoma of kidney patient to match
Standard includes: the CT image to three the unenhanced of clear cell carcinoma of kidney patient, arterial phase and venous phase timing, with arterial phase CT image
As benchmark image, image registration is carried out.Wherein, in the mistake for carrying out image registration using arterial phase CT image as benchmark image
Cheng Zhong, the present embodiment convert unenhanced and venous phase CT image to benchmark image (arterial phase CT image), this is transformed
Remained in journey it is each to collimation, i.e., change of scale coefficient uniformity in each direction, so that it is guaranteed that same patient
Different timing CT images be consistent in the number of plies and resolution ratio.Referring to the CT figure that Fig. 3 and Fig. 4, Fig. 3 are before image registration
Picture;Fig. 4 is the CT image after image registration.
In the present embodiment, when carrying out image registration based on benchmark image, algorithm in the prior art can be used, such as flat
Absolute difference algorithm (MAD), absolute error and algorithm (SAD), error sum of squares algorithm (SSD) etc., the present embodiment does not have this
It is restricted.
Optionally, the tumor region in the present embodiment can be delineated by professional image department doctor, further, hooked
It, can also be swollen to distinguish by the way of setting different figure layers for other regions other than tumor region and tumor region after picture
Tumor region and other regions.
Optionally, based on the CT image of clear cell carcinoma of kidney patient, obtaining the first CT image includes:
Show the CT image of clear cell carcinoma of kidney patient;
The figure layer of tumor region is arranged different from CT image as tumor region in the region that acquisition user delineates in CT image
In other regions figure layer, obtain the first CT image.
In the above scheme, the CT image for showing clear cell carcinoma of kidney patient includes: to show kidney by ITK-SNAP software
The CT image of clear cell carcinoma patient.Professional image department doctor can carry out tumor area on the CT image that ITK-SNAP software is shown
Domain is delineated, and can be indicated the tumor region in CT image with figure layer Label1 by ITK-SNAP software, with CT
Other image pixel points in image are distinguished, and have delineated the CT image of tumor region as the first CT image in subsequent step
It is used in rapid.
There is also the methods to tumor region automatic identification in the prior art, so, in another example, each CT figure
As upper tumor region can delineate completion, optionally, base by the extraction element that clear cell carcinoma of kidney shifts judging characteristic automatically
In the CT image of clear cell carcinoma of kidney patient, the CT figure that the first CT image includes: automatic identification clear cell carcinoma of kidney patient is obtained
Tumor region as in is simultaneously delineated, and figure layer of the figure layer different from other regions in CT image of tumor region is arranged, obtains the
One CT image.
It is understood that in above-mentioned two scheme, the timing of CT image has multiple, then needs the CT in multiple timing
Tumor's profiles are sketched out in image respectively.
The embodiment of the present invention is based on priori knowledge library, proposes on the basis of studying knowledge experience obtained by a large amount of natural images
A kind of frequency-domain transform method, when carrying out frequency-domain transform proposed by the invention, transformation coefficient has non-Gaussian feature, and
Each frequency domain sub-band coefficient is symmetrical all near zero.When the CT image to clear cell carcinoma of kidney patient carries out this kind of frequency-domain transform
When, it is equivalent to and CT image is redescribed under a variety of resolution ratio, so as to provide CT image under different spectral
Local physical feature, provide qualitative picture data for the feature extraction of final step.
Optionally, frequency-domain transform is carried out to the first CT image based on preset frequency-domain transform method and obtains corresponding 2nd CT
Image includes:
According to formula Fu,v(z)=I (z) * Gu,v(z) convolutional calculation is carried out to the first CT image, obtains corresponding 2nd CT
Image, wherein Fu,v(z) indicate that the 2nd CT image, I (z) indicate that the first CT image, z are some pixel in the first CT image
Coordinate value in space;* convolution algorithm, G are indicatedu,v(z) it indicates to pixel z frequency-domain transform any in the first CT image
Kernel function,ku,vIndicate the center frequency of filter
Rate;The direction of u expression filter;V indicates the number of plies (scale parameter) that frequency-domain transform decomposes;I is complex operator, and the π of σ=2 is indicated
The ratio between Gauss window width and wavelength value are 2 π.
It is understood that in the present embodiment, before step 101, further includes: be configured to clear cell carcinoma of kidney patient
Specifically to pixel z any in CT image, the core letter under its current state is arranged in the kernel function of the frequency-domain transform of CT image
Number isThe definition of parameter in kernel function is referring to upper
The description in content is stated, details are not described herein.
Optionally, in embodiment,Wherein, i ∈ x,
J ∈ y indicates that the pixel coordinate sequence of the CT plane of delineation, x, y are coordinate value of the pixel z in the CT plane of delineation.
In the present embodiment, operator can be extracted using the image feature pre-set to extract the office in the 2nd CT image
Portion's physical feature shifts judging characteristic as clear cell carcinoma of kidney, and optionally, image feature extracts operator and kidney hyaline cell
The number amount and type of metastasis of cancer judging characteristic can judge in unrestricted choice, extraction clear cell carcinoma of kidney transfer according to the actual needs
The timing CT image of feature can also unrestricted choice, the present embodiment do not limit this.
Optionally, it in a specific example, can be extracted from the CT image of the unenhanced and arterial phase in the 2nd CT image
Feature.The clear cell carcinoma of kidney transfer judging characteristic extracted from the 2nd CT image in the present embodiment can be used as kidney it is transparent thin
The judgement of born of the same parents' cancer far-end transfer provides the image sign object of reliable basis.
Optionally, in one example, the quantity that the image feature used extracts operator is 4 four operators under
The calculation formula of column is based on tumor region, includes: to the 2nd CT image zooming-out preset clear cell carcinoma of kidney transfer judging characteristic
The image feature arteryfirstorderMinimum of the arterial phase CT image in the 2nd CT image is extracted, is extracted
Calculation formula is as follows:
ArteryfirstorderMinmum=min (X)
In formula, X indicates the tumor region of arterial phase CT image;
With, the image feature arterygldmSDLGLE of the arterial phase CT image in the 2nd CT image of extraction, extraction calculating
Formula is as follows:
In formula, (i, j) ∈ (x, y), x, y respectively indicate the number and dependence of the discrete intensity values of the image X of tumor region
The quantity in region, P (i, j) then indicate to rely on matrix, wherein the number of discrete intensity values is the pixel in image X-plane
Number, the quantity of domain of dependence are the number for the pixel that the outer edge one of tumor region in described image X encloses;
With, the image feature arteryglcmAutocorrelation of the arterial phase CT image in the 2nd CT image is extracted,
It is as follows to extract calculation formula:
In formula, i ∈ g, j ∈ g, g indicate the quantity of the discrete intensity grade of image X, and p (i, j) is standardization Consistent Matrix,
Specifically,Discrete intensity grade is the tonal gradation in image X;
With, the image feature plainfirstorderMaximum of the plain CT image in the 2nd CT image of extraction, extraction
Calculation formula is as follows:
PlainfirstorderMaximum=max (Y)
In formula, Y indicates the tumor region of plain CT image.
It is understood that it is special image can also to be calculated using above-mentioned formula for the 2nd CT image of other timing
Levy arteryfirstorderMinimum, image feature arterygldmSDLGLE and image feature
ArteryglcmAutocorrelation etc., the present embodiment are not limited in this respect.
The embodiment of the invention provides a kind of extracting method of clear cell carcinoma of kidney transfer judging characteristic, obtained in the present invention
Clear cell carcinoma of kidney patient the first CT image in delineated tumor region;The frequency-domain transform method frequency domain transformation series used
Number has non-Gaussian feature, and each frequency domain sub-band coefficient is symmetrical all near zero, so being based on preset frequency-domain transform side
Method carries out frequency-domain transform to the first CT image and obtains corresponding 2nd CT image, can be regarded as to clear cell carcinoma of kidney patient's
First CT image carries out the description under a variety of resolution ratio, so as to provide the first CT image under different spectral for user
Local physical feature, so that the tumor region in the 2nd CT image becomes apparent from, to be easier and more accurately to the 2nd CT
The preset clear cell carcinoma of kidney of image zooming-out shifts judging characteristic, provides data for clear cell carcinoma of kidney transfer judgement and supports have
Conducive to the accuracy for promoting clear cell carcinoma of kidney transfer judgement.
Second embodiment:
Referring to Fig. 5, the present embodiment provides a kind of extraction element of clear cell carcinoma of kidney transfer judging characteristic, feature exists
In, comprising:
Image collection module 501 obtains the first CT image for the CT image based on clear cell carcinoma of kidney patient, wherein
First CT image is obtained by delineating tumor region in CT image;
Frequency-domain transform module 502 is obtained for carrying out frequency-domain transform to the first CT image based on preset frequency-domain transform method
To corresponding 2nd CT image, wherein in preset frequency-domain transform method, frequency-domain transform coefficient has non-Gaussian feature, and each
Frequency domain sub-band coefficient is symmetrical near zero;
Characteristic extracting module 503 turns the 2nd preset clear cell carcinoma of kidney of CT image zooming-out for being based on tumor region
Move judging characteristic.
Optionally, the extraction element of clear cell carcinoma of kidney transfer judging characteristic further includes image registration module 504, if for
It include the CT image of several timing in the CT image of clear cell carcinoma of kidney patient, then it is transparent to be based on kidney in image collection module 501
The CT image of carcinoma patients before obtaining the first CT image, carries out the CT image of several timing of clear cell carcinoma of kidney patient
Image registration.
Further, image registration module 504 is schemed specifically for the CT of several timing to clear cell carcinoma of kidney patient
Picture carries out image registration by benchmark image of wherein top-quality CT image.
Further, image registration module 504, specifically for the unenhanced of clear cell carcinoma of kidney patient, arterial phase and quiet
The CT image of three timing of arteries and veins phase carries out image registration using arterial phase CT image as benchmark image.
Optionally, frequency-domain transform module 502, for according to formula Fu,v(z)=I (z) * Gu,v(z) to the first CT image into
Row convolutional calculation obtains corresponding 2nd CT image, wherein Fu,v(z) indicate that the 2nd CT image, I (z) indicate the first CT image,
Z is the coordinate value of some pixel in space in the first CT image, and * indicates convolution algorithm, Gu,v(z) it indicates to the first CT
The kernel function of any pixel z frequency-domain transform in image,
ku,vIndicate the centre frequency of filter, u indicates the direction of filter, and v indicates the number of plies that frequency-domain transform decomposes, and i is complex operation
Symbol, the π of σ=2 indicate that the ratio between Gauss window width and wavelength value are 2 π.
Optionally, characteristic extracting module 503, specifically for extracting the image spy of the arterial phase CT image in the 2nd CT image
ArteryfirstorderMinimum is levied, it is as follows to extract calculation formula:
ArteryfirstorderMinmum=min (X)
In formula, X indicates the tumor region of arterial phase CT image;
With, the image feature arterygldmSDLGLE of the arterial phase CT image in the 2nd CT image of extraction, extraction calculating
Formula is as follows:
In formula, (i, j) ∈ (x, y), x, y respectively indicate the number of the discrete intensity values of image X and the quantity of domain of dependence,
P (i, j) then indicates to rely on matrix, wherein the number of discrete intensity values is the number of the pixel in image X-plane, relies on area
The quantity in domain is the number for the pixel that the outer edge one of tumor region in described image X encloses;
With, the image feature arteryglcmAutocorrelation of the arterial phase CT image in the 2nd CT image is extracted,
It is as follows to extract calculation formula:
In formula, i ∈ g, j ∈ g, g indicate the quantity of the discrete intensity grade of image X, and p (i, j) is standardization Consistent Matrix,
Specifically,Discrete intensity grade is the tonal gradation in image X;
With, the image feature plainfirstorderMaximum of the plain CT image in the 2nd CT image of extraction, extraction
Calculation formula is as follows:
PlainfirstorderMaximum=max (Y)
In formula, Y indicates the tumor region of plain CT image.
Optionally, image collection module 501, for showing the CT image of clear cell carcinoma of kidney patient;
The figure layer of tumor region is arranged different from CT image as tumor region in the region that acquisition user delineates in CT image
In other regions figure layer, obtain the first CT image.
Further, the present embodiment also provides a kind of extraction element of clear cell carcinoma of kidney transfer judging characteristic, referring to figure
6, which includes: memory 601, processor and 602 is stored on memory 601 and can run on processor 602
Computer program, processor 602 execute computer program when, realize first embodiment method in step.
Further, the embodiment of the present application also provides a kind of storage medium, the storage medium can be set to it is above-mentioned
In the extraction element of clear cell carcinoma of kidney transfer judging characteristic in each embodiment, which be can be shown in earlier figures 6
Memory in embodiment.It is stored with computer program on the storage medium, such as first is realized when which is executed by processor
Step in the method for embodiment description.Further, which can also be USB flash disk, mobile hard disk, read-only memory
The various media that can store program code such as (ROM, Read-Only Memory), RAM, magnetic or disk.
The embodiment of the invention provides a kind of extraction elements of clear cell carcinoma of kidney transfer judging characteristic, which can
Obtain the first CT image for delineating the clear cell carcinoma of kidney patient of tumor region;The frequency-domain transform method frequency domain transformation used
Coefficient has non-Gaussian feature, and each frequency domain sub-band coefficient is symmetrical all near zero, so being based on preset frequency-domain transform
Method carries out frequency-domain transform to the first CT image and obtains corresponding 2nd CT image, can regard as to clear cell carcinoma of kidney patient
The first CT image carry out the description under a variety of resolution ratio, so as to provide the first CT image under different spectral for user
Local physical feature so that the tumor region in the 2nd CT image becomes apparent from, to be easier and more accurately to second
The preset clear cell carcinoma of kidney of CT image zooming-out shifts judging characteristic, provides data for clear cell carcinoma of kidney transfer judgement and supports,
Be conducive to be promoted the accuracy of clear cell carcinoma of kidney transfer judgement.In several embodiments provided herein, it should be appreciated that
It arrives, disclosed device and method may be implemented in other ways.For example, Installation practice described above is only
It is schematically, for example, the division of module, only a kind of logical function partition, can there is other division in actual implementation
Mode, such as multiple module or components can be combined or can be integrated into another system, or some features can be ignored, or
It does not execute.Another point, shown or discussed mutual coupling, direct-coupling or communication connection can be by some
The indirect coupling or communication connection of interface, device or module can be electrical property, mechanical or other forms.
Module may or may not be physically separated as illustrated by the separation member, show as module
Component may or may not be physical module, it can and it is in one place, or may be distributed over multiple networks
In module.Some or all of the modules therein can be selected to achieve the purpose of the solution of this embodiment according to the actual needs.
It, can also be in addition, can integrate in a processing module in each functional module in each embodiment of the application
It is that modules physically exist alone, can also be integrated in two or more modules in a module.Above-mentioned integrated mould
Block both can take the form of hardware realization, can also be realized in the form of software function module.
If integrated module is realized and when sold or used as an independent product in the form of software function module, can
To be stored in a computer readable storage medium.Based on this understanding, the technical solution of the application 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 readable storage medium storing program for executing, including some instructions are used so that a computer
Equipment (can be personal computer, server or the network equipment etc.) execute each embodiment method of the application whole or
Part steps.And readable storage medium storing program for executing above-mentioned includes: that USB flash disk, mobile hard disk, ROM, RAM, magnetic or disk etc. are various can be with
Store the medium of program code.
It should be noted that for the various method embodiments described above, describing for simplicity, therefore, it is stated as a series of
Combination of actions, but those skilled in the art should understand that, the application is not limited by the described action sequence because
According to the application, certain steps can use other sequences or carry out simultaneously.Secondly, those skilled in the art should also know
It knows, the embodiments described in the specification are all preferred embodiments, and related actions and modules might not all be this Shen
It please be necessary.
In the above-described embodiments, it all emphasizes particularly on different fields to the description of each embodiment, there is no the portion being described in detail in some embodiment
Point, it may refer to the associated description of other embodiments.
The above are extracting method, device and the storages to clear cell carcinoma of kidney transfer judging characteristic provided herein to be situated between
The description of matter, for those skilled in the art, according to the thought of the embodiment of the present application, in specific embodiment and application range
Upper there will be changes, and to sum up, the contents of this specification should not be construed as limiting the present application.
Claims (10)
1. a kind of extracting method of clear cell carcinoma of kidney transfer judging characteristic characterized by comprising
CT image based on clear cell carcinoma of kidney patient obtains the first CT image, wherein the first CT image passes through described
Tumor region is delineated in CT image to obtain;
Frequency-domain transform is carried out to the first CT image based on preset frequency-domain transform method and obtains corresponding 2nd CT image,
In, in the preset frequency-domain transform method, frequency-domain transform coefficient has non-Gaussian feature, and each frequency domain sub-band coefficient is zero
It is symmetrical nearby;
Based on the tumor region, judging characteristic is shifted to the preset clear cell carcinoma of kidney of the 2nd CT image zooming-out.
2. the extracting method of clear cell carcinoma of kidney transfer judging characteristic according to claim 1, which is characterized in that if described
It include the CT image of several timing in the CT image of clear cell carcinoma of kidney patient, then described based on clear cell carcinoma of kidney patient's
CT image, before obtaining the first CT image, further includes:
Image registration is carried out to the CT image of several timing of the clear cell carcinoma of kidney patient.
3. the extracting method of clear cell carcinoma of kidney transfer judging characteristic according to claim 2, which is characterized in that the kidney
The CT image of clear cell carcinoma patient includes the CT image of three timing of unenhanced, arterial phase and venous phase, described saturating to the kidney
The CT image of several timing of clear cell carcinoma patient carries out image registration
To the CT image of three the unenhanced of clear cell carcinoma of kidney patient, arterial phase and venous phase timing, with arterial phase CT figure
As being used as benchmark image, image registration is carried out.
4. the extracting method of clear cell carcinoma of kidney transfer judging characteristic according to claim 3, which is characterized in that the base
In the tumor region, include: to the 2nd CT image zooming-out preset clear cell carcinoma of kidney transfer judging characteristic
The image feature arteryfirstorderMinimum of the arterial phase CT image in the 2nd CT image is extracted, is extracted
Calculation formula is as follows:
ArteryfirstorderMinmum=min (X)
In formula, X indicates the tumor region of the arterial phase CT image;
With, the image feature arterygldmSDLGLE of the arterial phase CT image in extraction the 2nd CT image, extraction calculating
Formula is as follows:
In formula, (i, j) ∈ (x, y), x, y respectively indicate the number and dependence of the discrete intensity values of the image X of the tumor region
The quantity in region, P (i, j) then indicate to rely on matrix, wherein the number of discrete intensity values is the pixel in described image X-plane
The number of point, the quantity of domain of dependence are the number for the pixel that the outer edge one of tumor region in described image X encloses;
With, the image feature arteryglcmAutocorrelation of the arterial phase CT image in the 2nd CT image is extracted,
It is as follows to extract calculation formula:
In formula, i ∈ g, j ∈ g, g indicate the quantity of the discrete intensity grade of described image X, and p (i, j) is standardization Consistent Matrix,
Specifically,The discrete intensity grade is the tonal gradation in described image X;
With, the image feature plainfirstorderMaximum of the plain CT image in extraction the 2nd CT image, extraction
Calculation formula is as follows:
PlainfirstorderMaximum=max (Y)
In formula, Y indicates the tumor region of the plain CT image.
5. the extracting method of clear cell carcinoma of kidney transfer judging characteristic according to claim 1-4, feature exist
In described to obtain corresponding 2nd CT image to the first CT image progress frequency-domain transform based on preset frequency-domain transform method
Include:
According to formula Fu,v(z)=I (z) * Gu,v(z) convolutional calculation is carried out to the first CT image, obtains corresponding 2nd CT
Image, wherein Fu,v(z) indicate that the 2nd CT image, I (z) indicate that the first CT image, z are some pixel in the first CT image
Coordinate value in space, * indicate convolution algorithm, Gu,v(z) it indicates to pixel z frequency-domain transform any in the first CT image
Kernel function,ku,vIndicate the center frequency of filter
Rate, u indicate the direction of filter, and v indicates the number of plies that frequency-domain transform decomposes, and i is complex operator, and the π of σ=2 indicates Gauss window width
It is 2 π with the ratio between wavelength value.
6. the extracting method of clear cell carcinoma of kidney transfer judging characteristic according to claim 1-4, feature exist
In the CT image based on clear cell carcinoma of kidney patient, obtaining the first CT image includes:
Show the CT image of the clear cell carcinoma of kidney patient;
The figure layer of the tumor region is arranged different from institute as tumor region in the region that acquisition user delineates in the CT image
The figure layer for stating other regions in CT image obtains the first CT image.
7. a kind of extraction element of clear cell carcinoma of kidney transfer judging characteristic characterized by comprising
Image collection module obtains the first CT image for the CT image based on clear cell carcinoma of kidney patient, wherein described the
One CT image is obtained by delineating tumor region in the CT image;
Frequency-domain transform module obtains pair for carrying out frequency-domain transform to the first CT image based on preset frequency-domain transform method
The 2nd CT image answered, wherein in the preset frequency-domain transform method, frequency-domain transform coefficient has non-Gaussian feature, and each
Frequency domain sub-band coefficient is symmetrical near zero;
Characteristic extracting module, for being based on the tumor region, to the preset clear cell carcinoma of kidney of the 2nd CT image zooming-out
Shift judging characteristic.
8. the extraction element of clear cell carcinoma of kidney transfer judging characteristic according to claim 7, which is characterized in that
The frequency-domain transform module, for according to formula Fu,v(z)=I (z) * Gu,v(z) convolution is carried out to the first CT image
It calculates, obtains corresponding 2nd CT image, wherein Fu,v(z) indicate that the 2nd CT image, I (z) indicate the first CT image, z the
The coordinate value of some pixel in space in one CT image, * indicate convolution algorithm, Gu,v(z) it indicates in the first CT image
The kernel function of any pixel z frequency-domain transform,
ku,vIndicate the centre frequency of filter, u indicates the direction of filter, and v indicates the number of plies that frequency-domain transform decomposes, and i is complex operation
Symbol, the π of σ=2 indicate that the ratio between Gauss window width and wavelength value are 2 π.
9. a kind of extraction element of clear cell carcinoma of kidney transfer judging characteristic, comprising: memory, processor and be stored in described deposit
On reservoir and the computer program that can run on the processor, which is characterized in that the processor executes the computer
When program, the step in any one of claim 1-6 the method is realized.
10. a kind of storage medium, is stored thereon with computer program, which is characterized in that the computer program is held by processor
When row, the step in any one the method in claim 1-6 is realized.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910327345.0A CN110211089A (en) | 2019-04-23 | 2019-04-23 | Extracting method, device and the storage medium of clear cell carcinoma of kidney transfer judging characteristic |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910327345.0A CN110211089A (en) | 2019-04-23 | 2019-04-23 | Extracting method, device and the storage medium of clear cell carcinoma of kidney transfer judging characteristic |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110211089A true CN110211089A (en) | 2019-09-06 |
Family
ID=67786153
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910327345.0A Pending CN110211089A (en) | 2019-04-23 | 2019-04-23 | Extracting method, device and the storage medium of clear cell carcinoma of kidney transfer judging characteristic |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110211089A (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106469449A (en) * | 2016-08-31 | 2017-03-01 | 上海联影医疗科技有限公司 | The display packing of focus and device in a kind of medical image |
CN106530236A (en) * | 2015-09-11 | 2017-03-22 | 上海联影医疗科技有限公司 | Medical image processing method and system |
CN106709967A (en) * | 2016-12-23 | 2017-05-24 | 天津恒宇医疗科技有限公司 | Endoscopic imaging algorithm and control system |
CN106815481A (en) * | 2017-01-19 | 2017-06-09 | 中国科学院深圳先进技术研究院 | A kind of life cycle Forecasting Methodology and device based on image group |
CN107292312A (en) * | 2017-06-19 | 2017-10-24 | 中国科学院苏州生物医学工程技术研究所 | Tumour recognition methods |
CN109074639A (en) * | 2015-10-19 | 2018-12-21 | 上海联影医疗科技有限公司 | Figure registration system and method in medical image system |
-
2019
- 2019-04-23 CN CN201910327345.0A patent/CN110211089A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106530236A (en) * | 2015-09-11 | 2017-03-22 | 上海联影医疗科技有限公司 | Medical image processing method and system |
CN109074639A (en) * | 2015-10-19 | 2018-12-21 | 上海联影医疗科技有限公司 | Figure registration system and method in medical image system |
CN106469449A (en) * | 2016-08-31 | 2017-03-01 | 上海联影医疗科技有限公司 | The display packing of focus and device in a kind of medical image |
CN106709967A (en) * | 2016-12-23 | 2017-05-24 | 天津恒宇医疗科技有限公司 | Endoscopic imaging algorithm and control system |
CN106815481A (en) * | 2017-01-19 | 2017-06-09 | 中国科学院深圳先进技术研究院 | A kind of life cycle Forecasting Methodology and device based on image group |
CN107292312A (en) * | 2017-06-19 | 2017-10-24 | 中国科学院苏州生物医学工程技术研究所 | Tumour recognition methods |
Non-Patent Citations (5)
Title |
---|
ZHI-CHENG LI ET AL.: "Differentiation of clear cell and non-clear cell renal cell carcinomas by all-relevant radiomics features from multiphase CT: a VHL mutation perspective", 《COMPUTER APPLICATIONS》 * |
中国航空学会信息融合分会著: "《第六届中国信息融合大会论文集 上》", 30 November 2014 * |
余胜威等著: "《MATLAB图像滤波去噪分析及其应用》", 30 June 2015, 北京航空航天出版社 * |
郑勋烨著: "《小波理论及其在图像处理中的应用》", 30 April 2018, 西安交通大学出版社 * |
韩志艳著: "《面向语音与面部表情信号的多模式情感识别技术研究》", 31 January 2017, 东北大学出版社 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10398344B2 (en) | Apparatus, methods and articles for four dimensional (4D) flow magnetic resonance imaging | |
Dalmış et al. | Using deep learning to segment breast and fibroglandular tissue in MRI volumes | |
Danilov et al. | Methods of graph network reconstruction in personalized medicine | |
EP3128919B1 (en) | Lung segmentation and bone suppression techniques for radiographic images | |
CN105096310B (en) | Divide the method and system of liver in magnetic resonance image using multi-channel feature | |
WO2021109987A1 (en) | Electronic apparatus and non-transient computer-readable storage medium | |
Kurugol et al. | Automated quantitative 3D analysis of aorta size, morphology, and mural calcification distributions | |
Lu et al. | Nonlocal means-based denoising for medical images | |
Goceri | Automatic labeling of portal and hepatic veins from MR images prior to liver transplantation | |
Saadatmand-Tarzjan | Self-affine snake for medical image segmentation | |
CN109447963A (en) | A kind of method and device of brain phantom identification | |
Guo et al. | Automatic segmentation of the pectoral muscle based on boundary identification and shape prediction | |
El‐Rewaidy et al. | Nonrigid active shape model–based registration framework for motion correction of cardiac T1 mapping | |
CN109767448A (en) | Parted pattern training method and device | |
Kim et al. | Automated segmentation of liver and liver cysts from bounded abdominal MR images in patients with autosomal dominant polycystic kidney disease | |
Jiang et al. | Nested block self‐attention multiple resolution residual network for multiorgan segmentation from CT | |
CN110211089A (en) | Extracting method, device and the storage medium of clear cell carcinoma of kidney transfer judging characteristic | |
CN116844734A (en) | Method and device for generating dose prediction model, electronic equipment and storage medium | |
Ng et al. | Spatially dependent filtering for removing phase distortions at the cortical surface | |
Lafitte et al. | Accelerating multi-modal image registration using a supervoxel-based variational framework | |
Li et al. | Medical image fusion based on local Laplacian decomposition and iterative joint filter | |
Jin et al. | Nonlocal means denoising of self-gated and k-space sorted 4-dimensional magnetic resonance imaging using block-matching and 3-dimensional filtering: implications for pancreatic tumor registration and segmentation | |
Yi et al. | Global multi-level attention network for the segmentation of clinical target volume in the planning CT for cervical cancer | |
Dahdouh et al. | Infants and young children modeling method for numerical dosimetry studies: application to plane wave exposure | |
Nikravanshalmani et al. | Three-dimensional semi-automatic segmentation of intracranial aneurysms in CTA |
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 |
Application publication date: 20190906 |
|
RJ01 | Rejection of invention patent application after publication |