CN105741241B - Tumor region image enchancing method and system based on synthesis enhancing image - Google Patents
Tumor region image enchancing method and system based on synthesis enhancing image Download PDFInfo
- Publication number
- CN105741241B CN105741241B CN201610054646.7A CN201610054646A CN105741241B CN 105741241 B CN105741241 B CN 105741241B CN 201610054646 A CN201610054646 A CN 201610054646A CN 105741241 B CN105741241 B CN 105741241B
- Authority
- CN
- China
- Prior art keywords
- image
- region
- window
- covering
- enhancing
- 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.)
- Expired - Fee Related
Links
- 206010028980 Neoplasm Diseases 0.000 title claims abstract description 177
- 230000002708 enhancing effect Effects 0.000 title claims abstract description 112
- 238000000034 method Methods 0.000 title claims abstract description 72
- 238000003786 synthesis reaction Methods 0.000 title claims abstract description 54
- 230000015572 biosynthetic process Effects 0.000 title claims abstract description 51
- 230000004927 fusion Effects 0.000 claims abstract description 23
- 238000012545 processing Methods 0.000 claims abstract description 21
- 238000011946 reduction process Methods 0.000 claims abstract description 6
- 238000005070 sampling Methods 0.000 claims description 58
- 238000009792 diffusion process Methods 0.000 claims description 56
- 230000009467 reduction Effects 0.000 claims description 14
- 238000005549 size reduction Methods 0.000 claims description 14
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 claims description 6
- 150000001875 compounds Chemical class 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 6
- 238000007500 overflow downdraw method Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 6
- 230000011218 segmentation Effects 0.000 abstract description 13
- 238000005516 engineering process Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000002591 computed tomography Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- 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
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The invention discloses a kind of tumor region image enchancing methods and system based on synthesis enhancing image, are related to field of medical image processing.Present invention firstly provides " synthesis enhancing image " this concepts, in original CT or MR images, noise reduction process and enhancing processing are carried out respectively to the ROI region of the whole tumor regions of covering, obtain the noise-reduced image and enhancing image of ROI region, fusion is weighted to noise-reduced image and enhancing image, obtains synthesis enhancing image.The present invention can enhance the ROI region in original CT or MR images, synthesize weighted factor different in enhancing image by selecting, so that the surface of tumour and smeared out boundary in original CT or MR images is become apparent from, doctor is facilitated to observe the surface and boundary of tumour;Threshold segmentation method, energy Accurate Segmentation is used to go out tumor region on synthesis enhancing image.
Description
Technical field
The present invention relates to field of medical image processing, are specifically related to a kind of tumor region figure based on synthesis enhancing image
Image intensifying method and system.
Background technology
Doctor is during practical diagnosing tumour relevant disease, by CT (Computed Tomography, electronics
Computed tomography) or the medical images such as MR (Magnetic Resonance, magnetic resonance) in tumour boundary, section face
Product and volume are measured and are analyzed, and can be helped the clinical definite state of an illness, at this moment be needed in the medical images such as CT or MR
Tumor boundaries are split.The Accurate Segmentation of tumor boundaries is extremely important to treatment plan, this part work at present relies primarily on
In hand drawing, accuracy is not high.Due to the noise and ambiguity in the medical images such as CT or MR, lead to the side of many tumours
Boundary is relatively fuzzyyer, takes traditional image partition method, such as using threshold value point on the medical images such as original CT or MR
Cut method, it is difficult to which Accurate Segmentation is carried out to the boundary of tumour.
Invention content
The purpose of the invention is to overcome the shortcomings of above-mentioned background technology, provide a kind of based on the swollen of synthesis enhancing image
Tumor area image Enhancement Method and system can enhance the areas ROI that whole tumor regions are covered in original CT or MR images
Domain synthesizes weighted factor different in enhancing image by selecting, makes the surface of tumour in original CT or MR images and obscure
Boundary becomes apparent from, and doctor is facilitated to observe the surface and boundary of tumour;Threshold segmentation side is used on synthesis enhancing image
Method, can Accurate Segmentation go out tumor region, accuracy is apparently higher than in original CT or MR images directly into row threshold division
Accuracy.
The present invention provides a kind of tumor region image enchancing method based on synthesis enhancing image, includes the following steps:
A, in the frame original CT image comprising tumour or MR images, the ellipse of the whole tumor regions of a covering is chosen
Rounded interested area ROI;Pixel other than ROI region is all set to zero, window width is carried out to the pixel value inside ROI region
The adjusting of window position is allowed to the requirement of compound mankind's observation;
B, using anisotropy parameter method, the ROI region in the image after being adjusted by window width and window level is dropped
It makes an uproar processing, obtains the noise-reduced image for covering the ROI region of whole tumor regions;
C, using multi-scale enhancement method, the ROI region in the image after being adjusted by window width and window level is enhanced
Processing, obtains the enhancing image for covering the ROI region of whole tumor regions;
D, the covering that the noise-reduced image to the ROI region of the obtained whole tumor regions of covering of step B and step C are obtained is complete
The enhancing image of the ROI region of portion's tumor region is weighted fusion, obtains the synthesis for covering the ROI region of whole tumor regions
Enhance image.
Based on the above technical solution, the adjusting of window width and window level is carried out in step A to the pixel value inside ROI region
Process be:Centered on elliptical barycenter, a rectangular area, the length r of the rectangle are selectedlengthLess than elliptical long axis,
The width r of the rectanglewidthLess than elliptical short axle;Choose the maximum that the maximum gradation value in the rectangle is window width in window technique
Value Cmax, choose the minimum value C that the minimum gradation value in the rectangle is window width in window techniquemin;Then following three are arbitrarily selected
A formula in a formula, maps the pixel value in ellipse:
Wherein, IoriFor original CT MR images, the coefficient of α image overall brightnesses in order to control, γ1It is adjustable
Coefficient, by changing γ1The mapping curve for being worth to different modes, I is the image after being adjusted by window width and window level.
Based on the above technical solution, in step A, γ1Value be less than 1 when, formula (3) is relatively narrow by a range
Input value is transformed into the output valve an of wider range.
Based on the above technical solution, the anisotropy parameter method in step B is:Use anisotropy parameter mould
Type removes noise, and the formula of Anisotropic Diffusion Model is:
Wherein, div () is divergence operator, and c (x, y, t) is diffusion coefficient, and diffusion coefficient is selected as the letter of image gradient
Number, t is the time,Gradient operator is represented,It is then the gradient of diffusion coefficient c,It is the gradient of image I, Δ represents La Pu
Laplacian operater, Δ I are the Laplace operators of image I;
To the Solving Partial Differential Equations that formula (4) indicates, the solution of the equation is expressed as Irn, IrnThe covering obtained for step B
The noise-reduced image of the ROI region of whole tumor regions.
Based on the above technical solution, in step B, the diffusion coefficient in the formula of the Anisotropic Diffusion Model
There are two types of expression ways for the function of c (x, y, t):
Wherein,Indicate the mould of the gradient of image I, K is constant, the two diffusion coefficient functions are with the gradient of image I
Mould as diffusion speed foundation, it is small in the big position diffusion coefficient of gradient, to play the purpose of Protect edge information.
Based on the above technical solution, the multi-scale enhancement method in step C is:
The enhancing processing under different scale is carried out to the ROI region of the whole tumor regions of covering:
First, by formula (7), down-sampling is carried out to the ROI region of the whole tumor regions of covering, after obtaining down-sampling
Image Ire;
Ire=REDUCE (I) (7)
Wherein, REDUCE () indicates down-sampling operation;
Then, by formula (8), to the image I after down-samplingreZoom in and out enhancing processing;
Iu=(Ire-Ire*G)γ2 (8)
Wherein, IuIt is the down-sampling enhancing image of size reduction, IreIt is the image after down-sampling, G is Gaussian kernel, Ire-
Ire* the radio-frequency component of G representative images, γ2The coefficient of tumor region and the difference of non-tumor region in order to control, IuSize under
Image I after samplingreSize it is identical, respectively less than pass through window width and window level adjust after image I;
Then, by formula (9), to the down-sampling enhancing image I for the size reduction that formula (8) obtainsuIt is up-sampled,
Obtain Iex:Iex=EXPAND (Iu) (9)
Wherein, IexFor the enhancing image of the ROI region of the obtained whole tumor regions of covering of step C, IexSize and warp
The size for crossing the image I after window width and window level is adjusted is identical;EXPAND () indicates up-sampling operation.
Based on the above technical solution, the EXPAND () in step C is obtained by extension interpolation operator, if to extend
One times of size then enhances one times of pixel, i.e., often needs to insert between row any two pixel in both the horizontal and vertical directions
Enter a value, every two in the ranks need to be inserted into a line;Interpolation operator uses bi-cubic interpolation method.
Based on the above technical solution, it is weighted and is melted using the weighted average fusion method of Pixel-level in step D
It closes, formula is as follows:
Ien=wIrn+(1-w)·Iex (10)
Wherein, IrnFor the noise-reduced image of the ROI region of the obtained whole tumor regions of covering of step B, IexIt is obtained for step C
The enhancing image of the ROI region of the whole tumor regions of covering arrived, w is weighted factor, IenFor the ROI of the whole tumor regions of covering
The synthesis in region enhances image;By selecting weighted factor w different in formula (10), to facilitate doctor to observe the surface of tumour
The boundary and.
The present invention also provides a kind of tumor region Image Intensified Systems based on synthesis enhancing image, which includes ROI
Region selection unit, window width and window level adjust unit, noise reduction unit, enhancement unit, Weighted Fusion unit, wherein:
The ROI region selection unit is used for:In the frame original CT image comprising tumour or MR images, one is chosen
The oval region of interest ROI of a whole tumor regions of covering;
The window width and window level adjusts unit and is used for:Pixel other than ROI region is all set to zero, to ROI region inside
Pixel value carry out the adjusting of window width and window level, be allowed to the requirement of compound mankind observation;
The noise reduction unit is used for:Using anisotropy parameter method, in the image after being adjusted by window width and window level
ROI region carry out noise reduction process, obtain the noise-reduced image for covering the ROI region of whole tumor regions;
The enhancement unit is used for:Using multi-scale enhancement method, in the image after being adjusted by window width and window level
ROI region carries out enhancing processing, obtains the enhancing image for covering the ROI region of whole tumor regions;
The Weighted Fusion unit is used for:Noise-reduced image and covering to the ROI region of the whole tumor regions of covering are all
The enhancing image of the ROI region of tumor region is weighted fusion, and the synthesis for obtaining covering the ROI region of whole tumor regions increases
Strong image.
Based on the above technical solution, the window width and window level adjusts unit and is carried out to the pixel value inside ROI region
The process of the adjusting of window width and window level is:Centered on elliptical barycenter, a rectangular area, the length r of the rectangle are selectedlength
Less than elliptical long axis, the width r of the rectanglewidthLess than elliptical short axle;It is window to choose the maximum gradation value in the rectangle
The maximum value C of window width in technologymax, choose the minimum value C that the minimum gradation value in the rectangle is window width in window techniquemin;Appoint
A formula in meaning selection following three formula, maps the pixel value in ellipse:
Wherein, IoriFor original CT MR images, the coefficient of α image overall brightnesses in order to control, γ1It is adjustable
Coefficient, by changing γ1The mapping curve for being worth to different modes, I is the image after being adjusted by window width and window level.
Based on the above technical solution, γ1Value be less than 1 when, formula (3) by the relatively narrow input value of a range turn
Change to the output valve an of wider range.
Based on the above technical solution, the anisotropy parameter method that the noise reduction unit uses for:Using it is each to
Anisotropic diffusion model removes noise, and the formula of Anisotropic Diffusion Model is:
Wherein, div () is divergence operator, and c (x, y, t) is diffusion coefficient, and diffusion coefficient is selected as the letter of image gradient
Number, t is the time,Gradient operator is represented,It is then the gradient of diffusion coefficient c,It is the gradient of image I, Δ represents La Pu
Laplacian operater, Δ I are the Laplace operators of image I;
To the Solving Partial Differential Equations that formula (4) indicates, the solution of the equation is expressed as Irn, IrnFor the whole tumor areas of covering
The noise-reduced image of the ROI region in domain.
Based on the above technical solution, the diffusion coefficient c (x, y, t) in the formula of the Anisotropic Diffusion Model
Function there are two types of expression way:
Wherein,Indicate the mould of the gradient of image I, K is constant, the two diffusion coefficient functions are with the gradient of image I
Mould as diffusion speed foundation, it is small in the big position diffusion coefficient of gradient, to play the purpose of Protect edge information.
Based on the above technical solution, the multi-scale enhancement method that the enhancement unit uses for:
The enhancing processing under different scale is carried out to the ROI region of the whole tumor regions of covering:
First, by formula (7), down-sampling is carried out to the ROI region of the whole tumor regions of covering, after obtaining down-sampling
Image Ire;
Ire=REDUCE (I) (7)
Wherein, REDUCE () indicates down-sampling operation;
Then, by formula (8), to the image I after down-samplingreZoom in and out enhancing processing;
Iu=(Ire-Ire*G)γ2 (8)
Wherein, IuIt is the down-sampling enhancing image of size reduction, IreIt is the image after down-sampling, G is Gaussian kernel, Ire-
Ire* the radio-frequency component of G representative images, γ2The coefficient of tumor region and the difference of non-tumor region in order to control, IuSize under
Image I after samplingreSize it is identical, respectively less than pass through window width and window level adjust after image I;
Then, by formula (9), to the down-sampling enhancing image I for the size reduction that formula (8) obtainsuIt is up-sampled,
Obtain Iex:Iex=EXPAND (Iu) (9)
Wherein, IexFor the enhancing image of the ROI region of the obtained whole tumor regions of covering of step C, IexSize and warp
The size for crossing the image I after window width and window level is adjusted is identical;EXPAND () indicates up-sampling operation.
Based on the above technical solution, the EXPAND () is obtained by extension interpolation operator, if to extend one times
Size, then enhance one times of pixel in both the horizontal and vertical directions, i.e., often need insertion one between row any two pixel
A value, every two in the ranks need to be inserted into a line;Interpolation operator uses bi-cubic interpolation method.
Based on the above technical solution, the Weighted Fusion unit using Pixel-level weighted average fusion method into
Row Weighted Fusion, formula are as follows:
Ien=wIrn+(1-w)·Iex (10)
Wherein, IrnFor the noise-reduced image of the ROI region of the whole tumor regions of covering, IexFor the whole tumor regions of covering
The enhancing image of ROI region, w are weighted factor, IenSynthesis for the ROI region of the whole tumor regions of covering enhances image;It is logical
Weighted factor w different in selection formula (10) is crossed, to facilitate doctor to observe the surface and boundary of tumour.
Compared with prior art, advantages of the present invention is as follows:
Present invention firstly provides " synthesis enhancing image " this concepts, in original CT or MR images, all to covering
The ROI region of tumor region carries out noise reduction process and enhancing processing respectively, obtains the noise-reduced image and enhancing image of ROI region,
Fusion is weighted to noise-reduced image and enhancing image, obtains synthesis enhancing image.Compared with original CT or MR images, synthesis
The boundary for enhancing tumour in image is apparent, while the information loss of tumor surface and few.The present invention can enhance original CT
Or the ROI region of whole tumor regions is covered in MR images, weighted factor different in enhancing image is synthesized by selecting,
So that the surface of tumour and smeared out boundary in original CT or MR images is become apparent from, facilitate doctor observe tumour surface and
Boundary;Synthesis enhancing image on use threshold segmentation method, can Accurate Segmentation go out tumor region, accuracy is apparently higher than
Directly into the accuracy of row threshold division in original CT or MR images.
Description of the drawings
Fig. 1 is the flow chart of the tumor region image enchancing method based on synthesis enhancing image in the embodiment of the present invention.
Fig. 2 is the image after window width and window level adjusting in the embodiment of the present invention.
Fig. 3 is the ROI region in the image in the embodiment of the present invention after window width and window level adjusting.
Fig. 4 is the noise-reduced image of ROI region in the embodiment of the present invention.
Fig. 5 is the enhancing image of ROI region in the embodiment of the present invention.
Fig. 6 is the synthesis enhancing image of ROI region in the embodiment of the present invention.
Specific implementation mode
Below in conjunction with the accompanying drawings and specific embodiment the present invention is described in further detail.
The problem of being difficult to divide for the obscurity boundary of tumour in original CT or MR images, the embodiment of the present invention provide one
Tumor region image enchancing method of the kind based on synthesis enhancing image, shown in Figure 1, this approach includes the following steps:
A, in the frame original CT image comprising tumour or MR images, the ellipse of the whole tumor regions of a covering is chosen
Round ROI (Region Of Interest, area-of-interest);Pixel other than ROI region is all set to zero, to the areas ROI
Pixel value inside domain carries out the adjusting of window width and window level, is allowed to the requirement of compound mankind's observation.
In practical applications, ROI region is first selected, then adjusts window width and window level, still, due to the pixel of original medical image
Value can not be shown between -1000 to+1000HU (Hounsfield Unit, Korea Spro Sen Feierde units) in regular display
Show, so the image in the embodiment of the present invention after vision-control window width and window level.Window width and window level adjust after image referring to
Shown in Fig. 2;ROI region in the image adjusted by window width and window level is shown in Figure 3.
B, using anisotropy parameter method (Anisotropic diffusion), after being adjusted by window width and window level
Image in ROI region carry out noise reduction process, the noise-reduced image for covering the ROI region of whole tumor regions is obtained, referring to Fig. 4
It is shown;
C, using multi-scale enhancement method, the ROI region in the image after being adjusted by window width and window level is enhanced
Processing, obtains the enhancing image for covering the ROI region of whole tumor regions, shown in Figure 5;
D, the covering that the noise-reduced image to the ROI region of the obtained whole tumor regions of covering of step B and step C are obtained is complete
The enhancing image of the ROI region of portion's tumor region is weighted fusion, obtains the synthesis for covering the ROI region of whole tumor regions
Enhance image, it is shown in Figure 6.
Compared with original CT or MR images, the boundary of tumour is apparent in synthesis enhancing image, while tumor surface
Information loss is simultaneously few.The synthesis enhances image and can be adopted on synthesis enhancing image as the input of threshold segmentation method
With threshold segmentation method, can Accurate Segmentation go out tumor region, accuracy is apparently higher than in original CT or MR images directly
Into the accuracy of row threshold division.Meanwhile can by selecting to synthesize different weighted factor in enhancing image, make original CT or
The surface of tumour and smeared out boundary become apparent from person's MR images, and doctor is facilitated to observe the surface and boundary of tumour.
It is to the process of the adjusting of the pixel value progress window width and window level inside ROI region in step A:It is with elliptical barycenter
Center selects a rectangular area, the length r of the rectanglelengthLess than elliptical long axis, the width r of the rectanglewidthLess than ellipse
Round short axle;Choose the maximum value C that the maximum gradation value in the rectangle is window width in window techniquemax, choose in the rectangle most
Small gray value is the minimum value C of window width in window techniquemin;Then a formula arbitrarily in selection following three formula, to ellipse
Pixel within the circle value is mapped:
Wherein, IoriFor original CT MR images, the coefficient of α image overall brightnesses in order to control, γ1It is adjustable
Coefficient, by changing γ1Value can obtain the mapping curves of different modes, γ1Value be less than 1 when, formula (3) can be by one
The relatively narrow input value of a range is transformed into the output valve an of wider range, and I is the image after being adjusted by window width and window level.
Anisotropy parameter method in step B is:Noise, anisotropy parameter are removed using Anisotropic Diffusion Model
The formula of model is:
Wherein, div () is divergence operator, and c (x, y, t) is diffusion coefficient, and diffusion coefficient is typically selected to be image gradient
Function, t is the time,Gradient operator is represented,It is then the gradient of diffusion coefficient c,It is the gradient of image I, Δ represents
Laplace operator, Δ I are the Laplace operators of image I.
The formula of above-mentioned Anisotropic Diffusion Model proposes by two scholars of Perona and Malik (two names),
Perona and Malik are also proposed:There are two types of expression ways for the function of diffusion coefficient c (x, y, t):
Indicate the mould of the gradient of image I, K is constant.The two diffusion coefficient functions are made with the mould of the gradient of image I
It is small in the big position diffusion coefficient of gradient to spread the foundation of speed, to play the purpose of Protect edge information.
To the Solving Partial Differential Equations that formula (4) indicates, the solution of the equation is expressed as Irn, IrnThe covering obtained for step B
The noise-reduced image of the ROI region of whole tumor regions.
Multi-scale enhancement method in step C is:
Clearly tumor boundaries, the ROI region to covering whole tumor regions carry out the increasing under different scale in order to obtain
It manages strength:
First, by formula (7), down-sampling is carried out to the ROI region of the whole tumor regions of covering, after obtaining down-sampling
Image Ire;
Ire=REDUCE (I) (7)
Wherein, REDUCE () indicates down-sampling operation;
Then, by formula (8), to the image I after down-samplingreZoom in and out enhancing processing;
Iu=(Ire-Ire*G)γ2 (8)
Wherein, IuIt is the down-sampling enhancing image of size reduction, IreIt is the image after down-sampling, G is Gaussian kernel, Gaussian kernel
Line number and columns can adjust, generally between 20 to 90.
Ire-Ire* the radio-frequency component of G representative images, tumor boundaries therein are simultaneously unchanged clear, swollen in order to further expand
The difference in tumor region and non-tumor region will calculate the γ of the radio-frequency component2Power, γ2Tumor region and non-tumor area in order to control
The coefficient of the difference in domain, due to IreIt is the image after down-sampling, size is less than the image I after being adjusted by window width and window level,
By formula (8), the down-sampling enhancing image I of size reduction is obtainedu, IuSize and down-sampling after image IreSize phase
Together, respectively less than pass through the image I after window width and window level is adjusted.
Then, by formula (9), to the down-sampling enhancing image I for the size reduction that formula (8) obtainsuIt is up-sampled,
Obtain Iex:Iex=EXPAND (Iu) (9)
Wherein, IexFor the enhancing image of the ROI region of the obtained whole tumor regions of covering of step C, IexSize and warp
The size for crossing the image I after window width and window level is adjusted is identical;EXPAND () indicates up-sampling operation.
EXPAND () is obtained by extension interpolation operator, if to extend one times of size, in horizontal and vertical two sides
The pixel of one times of enhancing upwards often needs to be inserted into a value between row any two pixel, every two in the ranks need to be inserted into a line;Interpolation
Operator uses bi-cubic interpolation method (Bicubic interpolation).
Fusion is weighted using the weighted average fusion method of Pixel-level in step D, formula is as follows:
Ien=wIrn+(1-w)·Iex (10)
Wherein, IrnFor the noise-reduced image of the ROI region of the obtained whole tumor regions of covering of step B, IexIt is obtained for step C
The enhancing image of the ROI region of the whole tumor regions of covering arrived, w is weighted factor, IenFor the ROI of the whole tumor regions of covering
The synthesis in region enhances image.It can be by selecting weighted factor w different in formula (10), to facilitate doctor to observe tumour
Surface and boundary.
The reason of synthesis enhances image is established in step D is:Noise in the noise-reduced image of step B has one relative to artwork
Fixed reduction, but the smeared out boundary of its tumour is not apparent from, and is unfavorable for dividing.Although and the enhancing image side in step C
Boundary is apparent from, but due to change of scale, the information loss of tumor surface is more, is unfavorable for doctor's inspections and examinations.In order to simultaneously
The inspections and examinations for taking into account tumor image segmentation and doctor, to the noise reduction of the ROI region of the obtained whole tumor regions of covering of step B
The enhancing image of the ROI region for the whole tumor regions of covering that image and step C are obtained is weighted fusion, obtains covering all
The synthesis of the ROI region of tumor region enhances image, and the tumor boundaries of synthesis enhancing image are apparent from, and tumor surface is believed
Breath loss is few.
Next by taking CT images as an example, the result handled by this method is demonstrated.
Since the value range of the pixel value in original CT image is between -1000 and+1000HU, in regular display
Effect is bad.Therefore, the image after being adjusted outside window by window width is only provided below.
Image after window width and window level in step A is adjusted is shown in Figure 2, in the image adjusted by window width and window level
ROI region is shown in Figure 3, it can be seen that the wherein obscurity boundary of tumour.
The noise-reduced image of ROI region in step B is shown in Figure 4, and compared with Fig. 3, the noise in Fig. 4 has to a certain degree
Reduction, but tumor boundaries are not obviously improved.
The enhancing image of ROI region in step C is shown in Figure 5, and compared with Fig. 3, the boundary of tumour becomes clear in Fig. 5
It is clear, but the information loss of tumor surface is more.
The synthesis enhancing image of ROI region in step D is shown in Figure 6, and compared with Fig. 4 and Fig. 5, Fig. 6 is Fig. 4 and figure
A kind of 5 half-way house, tumor boundaries therein are apparent from, and tumor surface information loss is few.
The embodiment of the present invention also provides a kind of tumor region Image Intensified System based on synthesis enhancing image, the system packet
ROI region selection unit, window width and window level adjusting unit, noise reduction unit, enhancement unit, Weighted Fusion unit are included, wherein:
ROI region selection unit is used for:In the frame original CT image comprising tumour or MR images, chooses one and cover
The oval ROI (Region Of Interest, area-of-interest) of the whole tumor regions of lid;
Window width and window level adjusts unit and is used for:Pixel other than ROI region is all set to zero, to the picture inside ROI region
Element value carries out the adjusting of window width and window level, is allowed to the requirement of compound mankind's observation.
In practical applications, ROI region is first selected, then adjusts window width and window level, still, due to the pixel of original medical image
Value can not be shown, so vision-control window in the embodiment of the present invention between -1000 to+1000HU in regular display
Image after wide window position.Image after window width and window level is adjusted is shown in Figure 2;In the image adjusted by window width and window level
ROI region is shown in Figure 3.
Noise reduction unit is used for:Using anisotropy parameter method (Anisotropic diffusion), to passing through window width window
The ROI region in image after the adjusting of position carries out noise reduction process, obtains the noise reduction figure for covering the ROI region of whole tumor regions
Picture, it is shown in Figure 4;
Enhancement unit is used for:Using multi-scale enhancement method, to the areas image Zhong ROI after being adjusted by window width and window level
Domain carries out enhancing processing, obtains the enhancing image for covering the ROI region of whole tumor regions, shown in Figure 5;
Weighted Fusion unit is used for:Noise-reduced image to the ROI region of the whole tumor regions of covering and the whole tumours of covering
The enhancing image of the ROI region in region is weighted fusion, obtains the synthesis enhancing figure for covering the ROI region of whole tumor regions
Picture, it is shown in Figure 6.
Window width and window level, which adjusts unit, to carry out the pixel value inside ROI region the process of adjusting of window width and window level and is:With ellipse
Centered on round barycenter, a rectangular area, the length r of the rectangle are selectedlengthLess than elliptical long axis, the width of the rectangle
rwidthLess than elliptical short axle;Choose the maximum value C that the maximum gradation value in the rectangle is window width in window techniquemax, choosing should
Minimum gradation value in rectangle is the minimum value C of window width in window techniquemin;Then one arbitrarily in selection following three formula
A formula maps the pixel value in ellipse:
Wherein, IoriFor original CT MR images, the coefficient of α image overall brightnesses in order to control, γ1It is adjustable
Coefficient, by changing γ1Value can obtain the mapping curves of different modes, γ1Value be less than 1 when, formula (3) can be by one
The relatively narrow input value of a range is transformed into the output valve an of wider range, and I is the image after being adjusted by window width and window level.
The anisotropy parameter method that noise reduction unit uses for:Noise is removed using Anisotropic Diffusion Model, respectively to different
The formula of property diffusion model is:
Wherein, div () is divergence operator, and c (x, y, t) is diffusion coefficient, and diffusion coefficient is typically selected to be image gradient
Function, t is the time,Gradient operator is represented,It is then the gradient of diffusion coefficient c,It is the gradient of image I, Δ represents
Laplace operator, Δ I are the Laplace operators of image I.
The formula of above-mentioned Anisotropic Diffusion Model proposes by two scholars of Perona and Malik (two names),
Perona and Malik are also proposed:There are two types of expression ways for the function of diffusion coefficient c (x, y, t):
Wherein,Indicate the mould of the gradient of image I, K is constant.The two diffusion coefficient functions are with the gradient of image I
Mould as diffusion speed foundation, it is small in the big position diffusion coefficient of gradient, to play the purpose of Protect edge information.
To the Solving Partial Differential Equations that formula (4) indicates, the solution of the equation is expressed as Irn, IrnFor the whole tumor areas of covering
The noise-reduced image of the ROI region in domain.
The multi-scale enhancement method that enhancement unit uses for:
Clearly tumor boundaries, the ROI region to covering whole tumor regions carry out the increasing under different scale in order to obtain
It manages strength:
First, by formula (7), down-sampling is carried out to the ROI region of the whole tumor regions of covering, after obtaining down-sampling
Image Ire;
Ire=REDUCE (I) (7)
Wherein, REDUCE () indicates down-sampling operation;
Then, by formula (8), to the image I after down-samplingreZoom in and out enhancing processing;
Iu=(Ire-Ire*G)γ2 (8)
Wherein, IuIt is the down-sampling enhancing image of size reduction, IreIt is the image after down-sampling, G is Gaussian kernel, Gaussian kernel
Line number and columns can adjust, generally between 20 to 90.
Ire-Ire* the radio-frequency component of G representative images, tumor boundaries therein are simultaneously unchanged clear, swollen in order to further expand
The difference in tumor region and non-tumor region will calculate the γ of the radio-frequency component2Power, γ2Tumor region and non-tumor area in order to control
The coefficient of the difference in domain, due to IreIt is the image after down-sampling, size is less than the image I after being adjusted by window width and window level,
By formula (8), the down-sampling enhancing image I of size reduction is obtainedu, IuSize and down-sampling after image IreSize phase
Together, respectively less than pass through the image I after window width and window level is adjusted.
Then, by formula (9), to the down-sampling enhancing image I for the size reduction that formula (8) obtainsuIt is up-sampled,
Obtain Iex:Iex=EXPAND (Iu) (9)
Wherein, IexFor the enhancing image of the ROI region of the whole tumor regions of covering, IexSize with pass through window width and window level
The size of image I after adjusting is identical;EXPAND () indicates up-sampling operation.
EXPAND () is obtained by extension interpolation operator, if to extend one times of size, in horizontal and vertical two sides
The pixel of one times of enhancing upwards often needs to be inserted into a value between row any two pixel, every two in the ranks need to be inserted into a line;Interpolation
Operator uses bi-cubic interpolation method (Bicubic interpolation).
Weighted Fusion unit is weighted fusion using the weighted average fusion method of Pixel-level, and formula is as follows:
Ien=wIrn+(1-w)·Iex (10)
Wherein, IrnFor the noise-reduced image of the ROI region of the whole tumor regions of covering, IexFor the whole tumor regions of covering
The enhancing image of ROI region, w are weighted factor, IenSynthesis for the ROI region of the whole tumor regions of covering enhances image.It can
With by selecting weighted factor w different in formula (10), to facilitate doctor to observe the surface and boundary of tumour.
From the foregoing, it will be observed that tumor region image enchancing method and system proposed by the present invention based on synthesis enhancing image, it can
For making the surface of tumour and smeared out boundary in original CT or MR images become apparent from, doctor to be facilitated to observe the table of tumour
Face and boundary.Threshold segmentation method, energy Accurate Segmentation is used to go out tumor region on synthesis enhancing image.
Those skilled in the art can be carry out various modifications to the embodiment of the present invention and modification, if these modifications and change
For type within the scope of the claims in the present invention and its equivalent technologies, then these modifications and variations are also in protection scope of the present invention
Within.
The prior art that the content not being described in detail in specification is known to the skilled person.
Claims (14)
1. a kind of tumor region image enchancing method based on synthesis enhancing image, which is characterized in that include the following steps:
A, in the frame original CT image comprising tumour or MR images, the ellipse of the whole tumor regions of a covering is chosen
Region of interest ROI;Pixel other than ROI region is all set to zero, window width and window level is carried out to the pixel value inside ROI region
Adjusting, be allowed to the requirement of compound mankind observation;
B, using anisotropy parameter method, the ROI region in the image after being adjusted by window width and window level is carried out at noise reduction
Reason, obtains the noise-reduced image for covering the ROI region of whole tumor regions;
C, using multi-scale enhancement method, enhancing processing is carried out to the ROI region in the image after being adjusted by window width and window level,
Obtain covering the enhancing image of the ROI region of whole tumor regions;
Multi-scale enhancement method in step C is:
The enhancing processing under different scale is carried out to the ROI region of the whole tumor regions of covering:
First, by formula (7), down-sampling is carried out to the ROI region of the whole tumor regions of covering, obtains the image after down-sampling
Ire;
Ire=REDUCE (I) (7)
Wherein, REDUCE () indicates down-sampling operation;
Then, by formula (8), to the image I after down-samplingreZoom in and out enhancing processing;
Wherein, IuIt is the down-sampling enhancing image of size reduction, IreIt is the image after down-sampling, G is Gaussian kernel, Ire-Ire* G generations
The radio-frequency component of table image, γ2The coefficient of tumor region and the difference of non-tumor region in order to control, IuSize and down-sampling after
Image IreSize it is identical, respectively less than pass through window width and window level adjust after image I;
Then, by formula (9), to the down-sampling enhancing image I for the size reduction that formula (8) obtainsuIt is up-sampled, is obtained
Iex:Iex=EXPAND (Iu) (9)
Wherein, IexFor the enhancing image of the ROI region of the obtained whole tumor regions of covering of step C, IexSize with pass through window
The size of image I after wide window position adjusting is identical;EXPAND () indicates up-sampling operation;
D, the covering that the noise-reduced image to the ROI region of the obtained whole tumor regions of covering of step B and step C are obtained is all swollen
The enhancing image of the ROI region in tumor region is weighted fusion, obtains the synthesis enhancing for covering the ROI region of whole tumor regions
Image.
2. the tumor region image enchancing method as described in claim 1 based on synthesis enhancing image, it is characterised in that:Step
It is to the process of the adjusting of the pixel value progress window width and window level inside ROI region in A:Centered on elliptical barycenter, one is selected
A rectangular area, the length r of the rectanglelengthLess than elliptical long axis, the width r of the rectanglewidthLess than elliptical short axle;Choosing
It is the maximum value C of window width in window technique to take the maximum gradation value in the rectanglemax, the minimum gradation value chosen in the rectangle is
The minimum value C of window width in window techniquemin;Then a formula arbitrarily in selection following three formula, to ellipse pixel within the circle
Value is mapped:
Wherein, IoriFor original CT MR images, the coefficient of α image overall brightnesses in order to control, γ1For adjustable coefficient,
By changing γ1The mapping curve for being worth to different modes, I is the image after being adjusted by window width and window level.
3. the tumor region image enchancing method as claimed in claim 2 based on synthesis enhancing image, it is characterised in that:Step
In A, γ1Value be less than 1 when, a relatively narrow input value of range is transformed into the output valve an of wider range by formula (3).
4. the tumor region image enchancing method as claimed in claim 2 based on synthesis enhancing image, it is characterised in that:Step
Anisotropy parameter method in B is:Noise, the formula of Anisotropic Diffusion Model are removed using Anisotropic Diffusion Model
For:
Wherein, div () is divergence operator, and c (x, y, t) is diffusion coefficient, and diffusion coefficient is selected as the function of image gradient, t
For the time,Gradient operator is represented,It is then the gradient of diffusion coefficient c,It is the gradient of image I, Δ represents Laplce's calculation
Son, Δ I are the Laplace operators of image I;
To the Solving Partial Differential Equations that formula (4) indicates, the solution of the equation is expressed as Irn, IrnThe covering obtained for step B is whole
The noise-reduced image of the ROI region of tumor region.
5. the tumor region image enchancing method as claimed in claim 4 based on synthesis enhancing image, it is characterised in that:Step
In B, there are two types of expression ways for the function of the diffusion coefficient c (x, y, t) in the formula of the Anisotropic Diffusion Model:
Wherein,Indicate the mould of the gradient of image I, K is constant, the two diffusion coefficient functions are made with the mould of the gradient of image I
It is small in the big position diffusion coefficient of gradient to spread the foundation of speed, to play the purpose of Protect edge information.
6. the tumor region image enchancing method as described in claim 1 based on synthesis enhancing image, it is characterised in that:Step
EXPAND () in C is obtained by extension interpolation operator, if one times of size is extended, in horizontal and vertical directions
The pixel of upper one times of enhancing often needs to be inserted into a value between row any two pixel, every two in the ranks need to be inserted into a line;Interpolation is calculated
Son uses bi-cubic interpolation method.
7. the tumor region image enchancing method as described in claim 1 based on synthesis enhancing image, it is characterised in that:Step
Fusion is weighted using the weighted average fusion method of Pixel-level in D, formula is as follows:
Ien=wIrn+(1-w)·Iex (10)
Wherein, IrnFor the noise-reduced image of the ROI region of the obtained whole tumor regions of covering of step B, IexIt is covered for what step C was obtained
The enhancing image of the ROI region of the whole tumor regions of lid, w is weighted factor, IenFor the ROI region of covering whole tumor regions
Synthesis enhancing image;By selecting weighted factor w different in formula (10), to facilitate doctor to observe the surface and side of tumour
Boundary.
8. a kind of tumor region Image Intensified System based on synthesis enhancing image, it is characterised in that:The system includes ROI region
Selection unit, window width and window level adjust unit, noise reduction unit, enhancement unit, Weighted Fusion unit, wherein:
The ROI region selection unit is used for:In the frame original CT image comprising tumour or MR images, chooses one and cover
The oval region of interest ROI of the whole tumor regions of lid;
The window width and window level adjusts unit and is used for:Pixel other than ROI region is all set to zero, to the picture inside ROI region
Element value carries out the adjusting of window width and window level, is allowed to the requirement of compound mankind's observation;
The noise reduction unit is used for:Using anisotropy parameter method, to the ROI in the image after being adjusted by window width and window level
Region carries out noise reduction process, obtains the noise-reduced image for covering the ROI region of whole tumor regions;
The enhancement unit is used for:Using multi-scale enhancement method, to the areas image Zhong ROI after being adjusted by window width and window level
Domain carries out enhancing processing, obtains the enhancing image for covering the ROI region of whole tumor regions;
The multi-scale enhancement method that the enhancement unit uses for:
The enhancing processing under different scale is carried out to the ROI region of the whole tumor regions of covering:
First, by formula (7), down-sampling is carried out to the ROI region of the whole tumor regions of covering, obtains the image after down-sampling
Ire;
Ire=REDUCE (I) (7)
Wherein, REDUCE () indicates down-sampling operation;
Then, by formula (8), to the image I after down-samplingreZoom in and out enhancing processing;
Wherein, IuIt is the down-sampling enhancing image of size reduction, IreIt is the image after down-sampling, G is Gaussian kernel, Ire-Ire* G generations
The radio-frequency component of table image, γ2The coefficient of tumor region and the difference of non-tumor region in order to control, IuSize and down-sampling after
Image IreSize it is identical, respectively less than pass through window width and window level adjust after image I;
Then, by formula (9), to the down-sampling enhancing image I for the size reduction that formula (8) obtainsuIt is up-sampled, is obtained
Iex:Iex=EXPAND (Iu) (9)
Wherein, IexFor the enhancing image of the ROI region of the obtained whole tumor regions of covering of step C, IexSize with pass through window
The size of image I after wide window position adjusting is identical;EXPAND () indicates up-sampling operation;
The Weighted Fusion unit is used for:Noise-reduced image to the ROI region of the whole tumor regions of covering and the whole tumours of covering
The enhancing image of the ROI region in region is weighted fusion, obtains the synthesis enhancing figure for covering the ROI region of whole tumor regions
Picture.
9. the tumor region Image Intensified System as claimed in claim 8 based on synthesis enhancing image, it is characterised in that:It is described
Window width and window level, which adjusts unit, to carry out the pixel value inside ROI region the process of adjusting of window width and window level and is:With elliptical barycenter
Centered on, select a rectangular area, the length r of the rectanglelengthLess than elliptical long axis, the width r of the rectanglewidthIt is less than
Elliptical short axle;Choose the maximum value C that the maximum gradation value in the rectangle is window width in window techniquemax, choose in the rectangle
Minimum gradation value is the minimum value C of window width in window techniquemin;A formula in arbitrary selection following three formula, to ellipse
Interior pixel value is mapped:
Wherein, IoriFor original CT MR images, the coefficient of α image overall brightnesses in order to control, γ1For adjustable coefficient,
By changing γ1The mapping curve for being worth to different modes, I is the image after being adjusted by window width and window level.
10. the tumor region Image Intensified System as claimed in claim 9 based on synthesis enhancing image, it is characterised in that:γ1
Value be less than 1 when, a relatively narrow input value of range is transformed into the output valve an of wider range by formula (3).
11. the tumor region Image Intensified System as claimed in claim 9 based on synthesis enhancing image, it is characterised in that:Institute
State the anisotropy parameter method that noise reduction unit uses for:Noise, anisotropy parameter are removed using Anisotropic Diffusion Model
The formula of model is:
Wherein, div () is divergence operator, and c (x, y, t) is diffusion coefficient, and diffusion coefficient is selected as the function of image gradient, t
For the time,Gradient operator is represented,It is then the gradient of diffusion coefficient c,It is the gradient of image I, Δ represents Laplce's calculation
Son, Δ I are the Laplace operators of image I;
To the Solving Partial Differential Equations that formula (4) indicates, the solution of the equation is expressed as Irn, IrnFor the whole tumor regions of covering
The noise-reduced image of ROI region.
12. the tumor region Image Intensified System as claimed in claim 11 based on synthesis enhancing image, it is characterised in that:Institute
Stating the function of the diffusion coefficient c (x, y, t) in the formula of Anisotropic Diffusion Model, there are two types of expression ways:
Wherein,Indicate the mould of the gradient of image I, K is constant, the two diffusion coefficient functions are made with the mould of the gradient of image I
It is small in the big position diffusion coefficient of gradient to spread the foundation of speed, to play the purpose of Protect edge information.
13. the tumor region Image Intensified System as claimed in claim 8 based on synthesis enhancing image, it is characterised in that:Institute
It states EXPAND () to be obtained by extension interpolation operator, if to extend one times of size, increase in both the horizontal and vertical directions
Strong one times of pixel often needs to be inserted into a value between row any two pixel, every two in the ranks need to be inserted into a line;Interpolation operator is adopted
With bi-cubic interpolation method.
14. the tumor region Image Intensified System as claimed in claim 8 based on synthesis enhancing image, it is characterised in that:Institute
It states Weighted Fusion unit and fusion is weighted using the weighted average fusion method of Pixel-level, formula is as follows:
Ien=wIrn+(1-w)·Iex (10)
Wherein, IrnFor the noise-reduced image of the ROI region of the whole tumor regions of covering, IexFor the areas ROI of the whole tumor regions of covering
The enhancing image in domain, w are weighted factor, IenSynthesis for the ROI region of the whole tumor regions of covering enhances image;Pass through selection
Different weighted factor w in formula (10), to facilitate doctor to observe the surface and boundary of tumour.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610054646.7A CN105741241B (en) | 2016-01-27 | 2016-01-27 | Tumor region image enchancing method and system based on synthesis enhancing image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610054646.7A CN105741241B (en) | 2016-01-27 | 2016-01-27 | Tumor region image enchancing method and system based on synthesis enhancing image |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105741241A CN105741241A (en) | 2016-07-06 |
CN105741241B true CN105741241B (en) | 2018-08-17 |
Family
ID=56247645
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610054646.7A Expired - Fee Related CN105741241B (en) | 2016-01-27 | 2016-01-27 | Tumor region image enchancing method and system based on synthesis enhancing image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105741241B (en) |
Families Citing this family (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106600550B (en) * | 2016-11-29 | 2020-08-11 | 深圳开立生物医疗科技股份有限公司 | Ultrasonic image processing method and system |
CN106709929A (en) * | 2016-12-30 | 2017-05-24 | 上海联影医疗科技有限公司 | Method and device for displaying interesting region of medical image |
CN106600568B (en) * | 2017-01-19 | 2019-10-11 | 东软医疗系统股份有限公司 | A kind of low-dose CT image de-noising method and device |
US11100632B2 (en) * | 2018-04-13 | 2021-08-24 | Elekta, Inc. | Image synthesis using adversarial networks such as for radiation therapy |
CN110163862B (en) * | 2018-10-22 | 2023-08-25 | 腾讯科技(深圳)有限公司 | Image semantic segmentation method and device and computer equipment |
CN109492587B (en) * | 2018-11-12 | 2021-06-22 | 浙江宇视科技有限公司 | Image processing method and device |
CN110135247B (en) * | 2019-04-03 | 2021-09-24 | 深兰科技(上海)有限公司 | Data enhancement method, device, equipment and medium in pavement segmentation |
CN110162953A (en) * | 2019-05-31 | 2019-08-23 | Oppo(重庆)智能科技有限公司 | Biometric discrimination method and Related product |
CN110706791B (en) * | 2019-09-30 | 2020-10-16 | 杭州依图医疗技术有限公司 | Medical image processing method and device |
CN111402261B (en) * | 2020-02-25 | 2021-06-08 | 四川大学青岛研究院 | Improved skull segmentation method based on Hessian matrix enhancement |
CN111489314A (en) * | 2020-04-16 | 2020-08-04 | 东软医疗系统股份有限公司 | Image enhancement method and device and terminal equipment |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1980321A (en) * | 2005-12-09 | 2007-06-13 | 逐点半导体(上海)有限公司 | Picture reinforcing treatment system and treatment method |
CN101393644A (en) * | 2008-08-15 | 2009-03-25 | 华中科技大学 | Hepatic portal vein tree modeling method and system thereof |
CN101587586A (en) * | 2008-05-20 | 2009-11-25 | 株式会社理光 | Device and method for processing images |
CN101706843A (en) * | 2009-11-16 | 2010-05-12 | 杭州电子科技大学 | Interactive film Interpretation method of mammary gland CR image |
CN101727658A (en) * | 2008-10-14 | 2010-06-09 | 深圳迈瑞生物医疗电子股份有限公司 | Image processing method and device |
CN103150714A (en) * | 2013-03-12 | 2013-06-12 | 华东师范大学 | Method and device for real-time interactive enhancement of magnetic resonance image |
-
2016
- 2016-01-27 CN CN201610054646.7A patent/CN105741241B/en not_active Expired - Fee Related
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1980321A (en) * | 2005-12-09 | 2007-06-13 | 逐点半导体(上海)有限公司 | Picture reinforcing treatment system and treatment method |
CN101587586A (en) * | 2008-05-20 | 2009-11-25 | 株式会社理光 | Device and method for processing images |
CN101393644A (en) * | 2008-08-15 | 2009-03-25 | 华中科技大学 | Hepatic portal vein tree modeling method and system thereof |
CN101727658A (en) * | 2008-10-14 | 2010-06-09 | 深圳迈瑞生物医疗电子股份有限公司 | Image processing method and device |
CN101706843A (en) * | 2009-11-16 | 2010-05-12 | 杭州电子科技大学 | Interactive film Interpretation method of mammary gland CR image |
CN103150714A (en) * | 2013-03-12 | 2013-06-12 | 华东师范大学 | Method and device for real-time interactive enhancement of magnetic resonance image |
Non-Patent Citations (2)
Title |
---|
基于CR图像的多尺度对比增强算法研究;葛日波;《公安海警高等专科学校学报》;20060815;第5卷(第3期);第16-17页 * |
改进的各向异性扩散图像去噪模型;郑满满;《中国优秀硕士学位论文全文数据库 信息科技辑》;20150115(第1期);第9-10页 * |
Also Published As
Publication number | Publication date |
---|---|
CN105741241A (en) | 2016-07-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105741241B (en) | Tumor region image enchancing method and system based on synthesis enhancing image | |
US10997701B1 (en) | System and method for digital image intensity correction | |
Liu et al. | Deep iterative reconstruction estimation (DIRE): approximate iterative reconstruction estimation for low dose CT imaging | |
US8705867B2 (en) | Devices and methods for processing images using scale space | |
EP2545527B1 (en) | Probabilistic refinement of model-based segmentation | |
CN103781416B (en) | Use the comprehensive cardiovascular analysis of body phase-contrast MRI | |
US11599982B2 (en) | Systems and methods for image processing | |
CN105488781B (en) | A kind of dividing method based on CT images liver neoplasm lesion | |
Hashimoto et al. | Automated segmentation of 2D low-dose CT images of the psoas-major muscle using deep convolutional neural networks | |
US10140715B2 (en) | Method and system for computing digital tomosynthesis images | |
CN104574284A (en) | Digital X-ray image contrast enhancement processing method | |
CN109844815A (en) | The image procossing based on feature is carried out using from the characteristic image of different iterative extractions | |
Lin et al. | Denoising of multi b-value diffusion-weighted MR images using deep image prior | |
US20160275653A1 (en) | Medical image processing apparatus and method | |
EP1945102B1 (en) | Image processing system and method for silhouette rendering and display of images during interventional procedures | |
Huang | An unsupervised convolutional neural network method for estimation of intravoxel incoherent motion parameters | |
Gloger et al. | Fully automatized renal parenchyma volumetry using a support vector machine based recognition system for subject-specific probability map generation in native MR volume data | |
WO2022207238A1 (en) | Methods and systems for biomedical image segmentation based on a combination of arterial and portal image information | |
Lin et al. | A general-threshold filtering method for improving intravoxel incoherent motion parameter estimates | |
Wang et al. | Adaptive patch-based POCS approach for super resolution reconstruction of 4D-CT lung data | |
US8335809B2 (en) | Procedure for deriving a three-dimensional digital mask starting from a series of two-dimensional masks, plus a device for doing this | |
Yang et al. | A skull segmentation method for brain MR images based on multiscale bilateral filtering scheme | |
Yin et al. | Scalable edge enhancement with automatic optimization for digital radiographic images | |
Wu et al. | Spatial and frequency-based super-resolution of ultrasound images | |
Guo et al. | LRENet: a location-related enhancement network for liver lesions in CT images |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180817 Termination date: 20210127 |