CN105286912A - System and method for quantization and visualization of echo texture features - Google Patents
System and method for quantization and visualization of echo texture features Download PDFInfo
- Publication number
- CN105286912A CN105286912A CN201410365753.2A CN201410365753A CN105286912A CN 105286912 A CN105286912 A CN 105286912A CN 201410365753 A CN201410365753 A CN 201410365753A CN 105286912 A CN105286912 A CN 105286912A
- Authority
- CN
- China
- Prior art keywords
- value
- pixel
- local
- parameter
- local variations
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000013139 quantization Methods 0.000 title claims abstract description 22
- 238000012800 visualization Methods 0.000 title claims abstract description 20
- 230000000007 visual effect Effects 0.000 claims description 42
- 238000012545 processing Methods 0.000 claims description 29
- 230000015572 biosynthetic process Effects 0.000 claims description 5
- 238000011002 quantification Methods 0.000 description 20
- 241000700560 Molluscum contagiosum virus Species 0.000 description 9
- 206010028980 Neoplasm Diseases 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 6
- 238000003745 diagnosis Methods 0.000 description 5
- 208000024770 Thyroid neoplasm Diseases 0.000 description 4
- 208000013076 thyroid tumor Diseases 0.000 description 4
- 238000000205 computational method Methods 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- KRQUFUKTQHISJB-YYADALCUSA-N 2-[(E)-N-[2-(4-chlorophenoxy)propoxy]-C-propylcarbonimidoyl]-3-hydroxy-5-(thian-3-yl)cyclohex-2-en-1-one Chemical compound CCC\C(=N/OCC(C)OC1=CC=C(Cl)C=C1)C1=C(O)CC(CC1=O)C1CCCSC1 KRQUFUKTQHISJB-YYADALCUSA-N 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000003759 clinical diagnosis Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 125000001475 halogen functional group Chemical group 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000001613 neoplastic effect Effects 0.000 description 1
- 230000000505 pernicious effect Effects 0.000 description 1
- 238000001507 sample dispersion Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 238000002604 ultrasonography Methods 0.000 description 1
Landscapes
- Ultra Sonic Daignosis Equipment (AREA)
Abstract
The invention discloses a system and method for quantization and visualization of echo texture features. The method includes the steps: acquiring an ultrasonic image; calculating numerical values of all pixels in a selected region of the ultrasonic image to obtain a regional standard deviation; excluding the pixels, with the numerical values smaller than a ratio value of the regional standard deviation, in the selected region; aiming at the remaining pixels, respectively counting a group of pixels centering on an nth pixel to calculate an nth local average value, an nth local variation value and an nth local variation coefficient of numerical values of the corresponding group of pixels, wherein n refers to a number from 1 to the number of the remaining pixels; acquiring echo texture indexes according to the local average values, the local variation values, the local variation coefficients or combinations of the values and the coefficients. Visualization is realized by displaying texture distribution conditions through presentation of continuous color gradations and adjusting contrast ratios presented by different visualizations according to visualization contrast intensity parameters.
Description
Technical field
The present invention relates to a kind of quantification and the quantification of method for visualizing and system, particularly a kind of echo textural characteristics and method for visualizing and system.
Background technology
In recent years, no matter all have significant progress due to medical ultrasonic image technology in the resolution of image and on image digitization, so medical ultrasonic image technology is used in the judgement of various types of neoplastic state step by step, such as thyroid tumor.And due to the non-intruding imaging characteristic of medical ultrasonic image technology, doctor also judges the disposal options of tumor character and evaluate subsequent gradually by the assistance of medical ultrasonic image technology.
The character of tumor can be assessed by the echo texture of ultrasonograph, is usually all directly to be carried out assessing and judge with naked eyes by doctor.Especially the echo texture of tumor is only the change slightly of the GTG Grad that some pixel has in gray scale ultrasound ripple image, for the naked eyes of doctor, these features are also not easy to pick out, with feeling, doctor only can judge whether these features exist, and then cause the judgement of tumor character only based on the subjective feeling of doctor, and accurately cannot can judge based on the fact.
Subjective impression and experience can affect the result of diagnosis, even same doctor, also likely have different diagnostic results because of spirit or the tired degree difference of eyes.That is, judge that the program of tumor character there is no a kind of objectively mechanism and exists by ultrasonograph at present.
Therefore, industry needs a kind of quantification of effectively and reliably echo texture and method for visualizing and system.
Summary of the invention
The object of this invention is to provide a kind of quantification of echo textural characteristics and method for visualizing and system, obtain an echo texture index thus, to reach the object of objective quantification, if Image Creation is presented on clinical image more in addition, for doctor diagnosed assessment, can auxiliary diagnosis be reached, reduce doctor's work load.
Quantification and the method for visualizing of a kind of echo textural characteristics of the present invention comprise: obtain a ultrasonograph, ultrasonograph comprises several pixel, and each pixel has a pixel numerical value; Receive the chosen area in ultrasonograph, and to calculate in chosen area all pixel numerical value to obtain the standard deviation in this region; Get rid of pixel numerical value in chosen area and be less than the pixel of a reference value, wherein reference value is the multiplier value of the standard deviation in this region; For residual pixel point, add up the local statistical parameter of one group of pixel centered by one n-th pixel respectively, this n-th local statistical parameter comprises the n-th local mean values, the n-th local variations numerical value, the n-th local variations coefficient or its combination, and n is the number of 1 to residual pixel point; And obtain at least one echo texture index according to described local mean values, described local variations numerical value, described local variations coefficient or its combination.
The quantization system of a kind of echo textural characteristics of the present invention comprises:
One ultrasonograph capture device, for obtaining a ultrasonograph, this ultrasonograph comprises several pixel, and each this pixel has a pixel numerical value;
One image processing module, connects this ultrasonograph capture device, and for receiving and processing this ultrasonograph, described image processing module comprises an analytic unit and an arithmetic element, and this analytic unit and this arithmetic element are interconnected;
One display device, connects image processing module, for showing this ultrasonograph; And
One input equipment, connects image processing module, for choosing the chosen area in this ultrasonograph;
Wherein, analytic unit receives this chosen area, arithmetic element to calculate in this chosen area this pixel numerical value all to obtain a regional standard difference, this analytic unit gets rid of this pixel that pixel numerical value in this chosen area is less than a reference value, wherein said reference value is the multiplier value of this regional standard difference, this arithmetic element is for residual pixel point, add up one group of pixel centered by one n-th pixel respectively, calculate one n-th local statistical parameter of one group of corresponding pixel numerical value, n is the number of 1 to residual pixel point, wherein n is natural number, this arithmetic element obtains at least one echo texture index according to those local statistical parameters.
The visualization system of a kind of echo textural characteristics of the present invention comprises:
One ultrasonograph capture device, for obtaining a ultrasonograph, described ultrasonograph comprises several pixel, and each pixel has a pixel numerical value;
One image processing module, connects this ultrasonograph capture device, and for receiving and processing described ultrasonograph, this image processing module comprises an analytic unit and an arithmetic element, and analytic unit and arithmetic element are interconnected;
One display device, connects described image processing module, in order to show this ultrasonograph; And
One input equipment, connects described image processing module, in order to choose the chosen area in this ultrasonograph;
Wherein this analytic unit receives this chosen area, this arithmetic element to calculate in this chosen area this pixel numerical value all to obtain a regional standard difference, this analytic unit gets rid of this pixel that pixel numerical value in this chosen area is less than a reference value, wherein this reference value is the multiplier value of this regional standard difference, this arithmetic element is for residual pixel point, add up one group of pixel centered by one n-th pixel respectively, calculate one n-th local statistical parameter of one group of corresponding pixel numerical value, n is the number of 1 to residual pixel point, wherein n is natural number, described local statistical parameter comprises several local mean values, this arithmetic element obtains at least one echo texture pointer according to described local statistical parameter, this arithmetic element calculates residual pixel point value, to obtain a region averages, and the absolute difference calculating described local mean values and this region averages is respectively to obtain the visual value of an echo texture strength of each this pixel, this analytic unit in conjunction with the visual value of this echo texture strength and this echo texture index in order to draw an echo texture image color range figure.
By enforcement of the present invention, the echo texture pointer quantizing echo textural characteristics can be obtained.Echo texture index is applied in clinical diagnosis, particularly thyroid tumor ultrasonic diagnosis, because having good specificity, confirms that the accuracy rate that can improve diagnosis is to 74.8%.
Below in conjunction with accompanying drawing, also by embodiment, the present invention is described in detail, so that those skilled in the art understand above-mentioned relevant object of the present invention and advantage.
Accompanying drawing explanation
Fig. 1 is the quantization method flow chart of a kind of echo textural characteristics of the embodiment of the present invention;
Fig. 2 is a kind of chosen area of the embodiment of the present invention and the schematic diagram of one group of pixel;
Fig. 3 is that the continuous color range of one of the embodiment of the present invention presents figure;
Fig. 4 A is a kind of original ultrasonic image of the embodiment of the present invention;
Fig. 4 B is the even schematic diagram of a kind of Image Creation echo grain distribution of the embodiment of the present invention;
Fig. 5 A is a kind of original ultrasonic image of the embodiment of the present invention;
Fig. 5 B is the mixed and disorderly schematic diagram of a kind of Image Creation echo grain distribution of the embodiment of the present invention;
Fig. 6 A is a kind of original ultrasonic image of the embodiment of the present invention;
Fig. 6 B is a kind of schematic diagram contrasting intensive parameter S1=1.125 and S2=3 of the embodiment of the present invention;
Fig. 6 C is a kind of schematic diagram contrasting intensive parameter S1=1.5 and S2=4 of the embodiment of the present invention;
Fig. 6 D is a kind of schematic diagram contrasting intensive parameter S1=1.875 and S2=5 of the embodiment of the present invention;
Fig. 7 A is a kind of original ultrasonic image of the embodiment of the present invention;
Fig. 7 B is a kind of Image Creation echo grain distribution schematic diagram of the embodiment of the present invention;
Fig. 7 C is known a kind of Image Creation echo grain distribution schematic diagram;
Fig. 8 is quantification and the visualization system of a kind of echo textural characteristics of the embodiment of the present invention.
Description of reference numerals: the quantification of A-echo textural characteristics and visualization system; 10-chosen area; 20-ultrasonograph capture device; 30-image processing module; 31-analytic unit; 32-arithmetic element; 40-input equipment; 50-display device; S110 ~ S150-step.
Detailed description of the invention
The invention provides a kind of quantification and method for visualizing of echo textural characteristics, and a kind of quantification of echo textural characteristics and visualization system, the quantification of echo textural characteristics disclosed in the embodiment of the present invention and method for visualizing are except the quantification of the echo textural characteristics that can be applicable to disclosed by the present invention and visualization system, directly can also be applied in ultrasonic instrument, or be applied in and can be connected in the computer system of ultrasonic instrument.The execution step of the embodiment of the present invention can be write as software program, software program can be stored in any microprocessing unit can identification, deciphering recording medium, or include article and the device of above-mentioned programmed recording medium.Be not defined as any form, above-mentioned article can be hard disk, floppy disk, CD, ZIP, magneto-optical device (MO), IC chip, random access memory (RAM), or any be familiar with this operator the spendable article including above-mentioned programmed recording medium.
Computer system can comprise display device, processor, internal memory, input equipment and storage device.Wherein, input equipment can in order to input the data such as image, word, instruction to computer system.Storage device system is such as hard disk, CD-ROM drive or the remote data base by Internet connection, in order to stocking system program, application program and user data etc., can also store the software program that the embodiment of the present invention is write as.Internal memory system is in order to the program of temporal data or execution.Processor is in order to computing and process data etc.Display device is then in order to the data of display translation.When computer system performs quantification and the method for visualizing of the echo textural characteristics of the embodiment of the present invention, corresponding program is just loaded internal memory, with the appraisal procedure coordinating processor to perform the embodiment of the present invention.Finally, then display device will be the results are shown in or be stored in storage device.
In order to for simplicity, be applied as example below with ultrasonic instrument, method and system of the present invention is described.
As shown in Figure 1, the embodiment of the present invention is a kind of quantization method of echo textural characteristics, and it can obtain echo texture pointer, to represent the complexity of the echo texture in range areas.The quantization method of echo textural characteristics comprises: obtain a ultrasonograph, and this ultrasonograph comprises several pixel, and each pixel has a pixel numerical value, and pixel numerical value can be gray-scale intensity values (step S110); Receive the chosen area in this ultrasonograph, and to calculate in chosen area all pixel numerical value to obtain this regional standard difference (step S120); Get rid of in chosen area, pixel numerical value is less than the pixel of a reference value, and wherein reference value is the multiplier value (step S130) of this regional standard difference; For remaining pixel, add up with one n-th local statistical parameter respectively, this n-th local statistical parameter comprises the n-th local mean values, the n-th local variations numerical value, the n-th local variations coefficient or its combination, and n is 1 to the number (step S140) of remaining pixel; And obtain at least one echo texture index (step S150) according to the n-th local statistical parameter.
As shown in Figure 2, when ultrasonic unit has captured ultrasonograph, microprocessing unit in ultrasonic unit or computer system can directly obtain above-mentioned ultrasonograph, ultrasonograph comprises a plurality of pixel, and each pixel has a pixel numerical value (step S110).In ultrasonograph, enclose choosing when user or define a chosen area 10; Step S120, receives the chosen area 10 (region that in Fig. 2, white closed curve encloses) in ultrasonograph.Chosen area 10 can be region or the interested region (RegionOfInterest, ROI) of user at focus place usually, and such as thyroid tumor region, the present invention is not as limit.
Then, all pixel numerical value is calculated in chosen area 10 to obtain a regional standard difference.Each pixel can be defined as one as the lattice of in chosen area in figure 10, and the pixel numerical value of lattice is GR (i, j).Therefore regional standard difference (σ) is exactly the standard deviation of pixel numerical value GR (i, j) of pixel in chosen area 10, σ=
std(GR (i, j)).In one embodiment, all pixel numerical value in chosen area 10 can be used to carry out the calculating of regional standard difference.
In a preferred embodiment, low echo band (halo) or the variation value in order to remove the choosing of different users's profile circle is enclosed in order to get rid of one outside tumor, the profile of chosen area 10 more can inwardly be reduced a set ring-type scope by the embodiment of the present invention, such as 5 ~ 10 pixel wide.After this revises, then add up the pixel numerical value of remaining pixel, comparatively can be had the regional standard difference quantizing meaning, the present invention is not as limit.
In order to the numerical value quantizing meaning comparatively can be had, need first get rid of the pixel in chosen area 10 with meaningless interval value.Meaningless interval value can be the pixel numerical value in region (Anechoicregion) of having no reply, its can because of the difference of tissue difference to some extent.Therefore, step S130 system gets rid of in chosen area 10, pixel numerical value is less than the pixel of a reference value, reference value is the multiplier value (r) of this regional standard difference (σ), be namely r × σ, wherein multiplier value (r) can allow user according to different tissues to adjust setting.
After above-mentioned eliminating step, in chosen area 10 in remaining pixel, analyze the statistical parameter in each pixel subrange, such as step S140 system adds up one group of pixel centered by one n-th pixel respectively.Add up one group of pixel in the subrange of the n-th pixel, one n-th local statistical parameter can be obtained after calculating one group of corresponding pixel numerical value, this n-th local statistical parameter comprises the n-th local mean values, the n-th local variations numerical value, the n-th local variations coefficient or its combination, and n is 1 to the number of remaining pixel.Subrange defines with a shielding (mask), and the center of shielding is current pixel, and remaining pixel is vicinity points, and namely the scope of one group of pixel that shielding covers is subrange, and it can be the square formation of k × k.As shown in Figure 2, current pixel is P (i, j), and the pixel numerical value of pixel is GR (i, j), k=5, and k=2q+1, n are 1 to the number of remaining pixel.K is used to the parameter defining square formation (matrix) width.Square formation
component can be expressed as:
Step S150, according to above-mentioned this statistical parameter calculating each pixel, in order to obtain at least one echo texture index, this statistical parameter comprises the local mean values (localMean) of each pixel, the local variations numerical value (localVariance) of each pixel, the local variations coefficient (localCoefficientofVariance, localCV) of each pixel or its combination.Be positioned at
the sample mean of interior each point is defined as local mean values (localmean), and it can be expressed as
and can calculate by following formula:
Be positioned at
the sample changed of interior each point is defined as local variations numerical value (localvariance), and it can be expressed as
and can calculate by following formula:
Be positioned at
the sample dispersion degree of interior each point is defined as local variations coefficient (localcoefficientofvariance), and it can be expressed as
and can calculate by following formula:
But, when
when comprising the point enclosing and select beyond scope 10,
above-mentioned parameter can not be calculated.
In a preferred embodiment, quantization method of the present invention can obtain an overall echo texture index, in order to assess the complexity of the echo grain distribution of tumor entirety.Such as, echo texture index can by calculating those local mean values and those local variations numerical value to obtain a local variations mean values (MeanoflocalVariance, and the variation value (VarianceoflocalVariance of a local variations MOV), VOV), can using the variation value of this local variations mean values or this local variations as echo texture index, also or using the value of the square root of the variation value of local variations divided by local variations mean values as echo texture index, namely echo texture index is Sqrt (VOV)/MOV.Echo texture pointer value is larger, and to represent the distribution of echo texture in chosen area 10 more mixed and disorderly more inconsistent.
In a preferred embodiment, echo texture index can for calculating a local variations mean coefficient (MeanoflocalCV that those local variations coefficients obtain, MOCV) obtain, namely echo texture index is MOCV, and echo texture index value is larger, and to represent the distribution of echo texture in chosen area 10 more mixed and disorderly more inconsistent.
In order to simplified characterization, we known n is the number of remaining pixel, and therefore local variations mean values (MOV) can be calculated as follows:
The variation value (VOV) of local variations can be calculated as follows:
Local variations mean coefficient (MOCV) can be calculated as follows:
The echo texture index of each pixel subrange can be obtained, in order to assess the complexity of the echo grain distribution of tumor by local at quantization method of the present invention.The echo texture pointer Image Creation of each pixel subrange can be judged easily the echo texture of tumor by image, promote diagnosis accuracy and alleviate the work load of doctor.
Method for visualizing of the present invention is except above-mentioned quantization step, comprise further and calculate remaining pixel numerical value, to obtain a region averages (μ ') and regional standard difference parameter (σ '), region averages (μ ') is after deducting the pixel that has no reply in chosen area 10, the average gray-scale intensity values of remaining pixel.And the visual value of an echo texture strength of each pixel can calculate the absolute difference of those local mean values and region averages respectively and obtain, and can in conjunction with the echo texture index of above-mentioned gained in order to draw an echo texture image color range figure, to show the echo texture level of chosen area 10, it is even that such as blueness belongs to echo texture, and it is mixed and disorderly that redness belongs to echo texture.The visual value of echo texture strength of above-mentioned pixel (i, j) can be expressed as T
i,j=Abs (localMean
i,j-μ ').
In a preferred embodiment, method for visualizing of the present invention can comprise that to compare dividing value (U) and a floor value (L) on the visual value of each echo texture strength and one visual in present worth (V with obtain each pixel one further
i,j), when the visual value of echo texture strength of arbitrary pixel is greater than upper dividing value, it is visual in present worth (V
i,j) be 0; When the visual value of echo texture strength of arbitrary pixel is less than floor value, it is visual is color stage value in present worth; When the echo texture index of arbitrary pixel is less than upper dividing value and is greater than floor value, it is visual is that upper dividing value deducts the visual value of echo texture strength and deducts this floor value divided by upper dividing value in present worth, then is multiplied by color range value.
If color range value is for 230, the visual of each pixel (i, j) is present worth (V
i,j) can be expressed as follows:
Method for visualizing of the present invention can comprise further and calculates those local mean values and those local variations numerical value with the variation value (VOV) of the local variations mean values (MOV) obtained and a local variations, upper dividing value can be the variation value (VOV) of local variations mean values (MOV) divided by local variations, then is multiplied by this regional standard difference parameter (σ ') divided by a upper bound parameter (S
1); And floor value can be the variation value (VOV) of local variations mean values (MOV) divided by local variations, then be multiplied by this regional standard difference parameter (σ ') divided by a lower bound parameter (S
2), wherein S
1and S
2be a pair specific strength parameter, between 0.5 ~ 10.0, and S
2be greater than S
1, user can from Row sum-equal matrix S
1and S
2.Upper dividing value (U) and floor value (L) can be expressed as follows:
U＝(MOV/VOV)×(σ')/(S
1)
L＝(MOV/VOV)×(σ')/(S
2)
In another preferred embodiment, method for visualizing of the present invention can also comprise those local variations coefficients of calculating further with the local variations mean coefficient (MOCV) obtained, upper dividing value can be the inverse (1/MOCV) of this local variations mean coefficient, then is multiplied by this regional standard difference parameter (σ ') divided by a upper bound parameter (S
1); And floor value can the inverse (1/MOCV) of this local variations mean coefficient, then be multiplied by this regional standard difference parameter (σ ') divided by a lower bound parameter (S
2), wherein S
1and S
2be a pair specific strength parameter, between 0.5 ~ 10.0, and S
2be greater than S
1, user can from Row sum-equal matrix S
1and S
2.Upper dividing value (U) and floor value (L) can be expressed as follows:
As shown in Figure 3, finally, visual in present worth (V by each pixel
i,j), through the mode of HSV colour model conversion rgb color model, the color corresponding to each pixel is calculated, will be presented by red, orange, yellow, green, blue color range mode Image Creation continuously with one.If visual acquired by all pixels combined in present worth, a texture image color range figure can be drawn, to show the distribution of this ultrasonograph heterogeneous tissue.
As shown in Fig. 4 A and Fig. 4 B, Fig. 5 A and Fig. 5 B, obviously can find out, the COLOR COMPOSITION THROUGH DISTRIBUTION that the image of method for visualizing of the present invention presents is comparatively continuous, and really the comparatively mixed and disorderly region of original echo distribution more clearly can be highlighted.Fig. 4 B is the more uniform case of ultrasonograph echo grain distribution, and Image Creation presents and more tends to blue.Fig. 5 B is the comparatively mixed and disorderly case of ultrasonograph echo grain distribution, and Image Creation presents more trend red area and the more, represents that echo grain distribution is more mixed and disorderly.
As shown in Fig. 6 A to Fig. 6 D, the method for visualizing of the embodiment of the present invention can also utilize the contrast intensive parameter S of above-mentioned echo grain distribution Image Creation further
1and S
2the display comparison of value adjustment image, can find S
1and S
2be worth larger, then contrast more obvious.
The quantification of the embodiment of the present invention and method for visualizing first can get rid of meaningless region in area-of-interest, the echo texture index obtained is made to have more quantification meaning, utilize the contrast that echo grain distribution Image Creation contrast intensive parameter presents to adjust different Image Creation simultaneously, add the mode through HSV colour model conversion rgb color model, original image energy is made therefore to provide the image feature of high contrast, as shown in figs. 7 a and 7b, apply quantification of the present invention and method for visualizing, original ultrasonograph becomes can make healthcare givers more heterogeneous aura be done to the high-contrast image diagnosed.Compared to the known method as Fig. 7 C the echo texture comparison image made, this case is applied on the good pernicious discriminant analysis of clinical thyroid tumor case, and its Average Accuracy reaches 74.8%, has the lifting of suitable amplitude than known 60%.Therefore this case obtains the technology effectively improved Image Creation and present, and more highlights the situation of echo grain distribution.
The invention provides a kind of quantification and visualization system A of echo textural characteristics, in order to implement quantification and the method for visualizing of aforementioned echo textural characteristics, quantification and the visualization system A of echo textural characteristics can divide into quantization system and visualization system, both have identical assembly, only be the method difference to some extent implemented, quantization system can perform the quantization method of aforementioned echo textural characteristics, and visualization system can perform the method for visualizing of aforementioned echo textural characteristics.As shown in Figure 8, in an embodiment, system of the present invention comprises: ultrasonograph capture device 20, image processing module 30, display device 50 and an input equipment 40.
Ultrasonograph acquisition unit 20 is in order to obtain a ultrasonograph, ultrasonograph comprises a plurality of pixel, each pixel has a pixel numerical value, pixel numerical value can be gray-scale intensity values, and the ultrasonograph acquisition unit 20 of indication of the present invention contains all ultrasonic instruments obtaining ultrasonograph.
Image processing module 30 is connect ultrasonograph capture device 20, in order to receive and process ultrasonograph, image processing module 30 comprises analytic unit 31 and an arithmetic element 32, analytic unit 31 and arithmetic element 32 are interconnected, the image processing module 30 of indication of the present invention is contained and is allly carried out image processing, and comprise software or the hardware of analysis and calculation function, such as be installed in ultrasonograph acquisition unit 20 in the mode of software or embedded system, or directly using main frame or image processor as image processing module 30, the present invention is not as limit.
Display device 50 is connect image processing module 30, in order to show ultrasonograph.Input equipment 40 is image processing module 30, in order to choose the chosen area in ultrasonograph, input equipment 40 contain all can in order to input the instrument of image, word or instruction, such as mouse, keyboard, handwriting pad, also or can in conjunction with display device 50 reach input object instrument, the mike of such as Touch Screen or acoustic control, the present invention is not as limit.
In a preferred embodiment, in the quantization system of echo textural characteristics, user watches ultrasonograph on the display device 50, and chosen the chosen area in ultrasonograph by input equipment 40 after, analytic unit 31 receives chosen area, arithmetic element 32 to calculate in chosen area all pixel numerical value to obtain a regional standard difference, analytic unit 31 is got rid of in chosen area, pixel numerical value is less than the pixel of a reference value, wherein reference value is the multiplier value of regional standard difference, then, arithmetic element 32 is for remaining pixel, add up one group of pixel centered by one n-th pixel respectively, calculate one n-th local statistical parameter of one group of corresponding pixel numerical value, n is 1 to the number of this pixel remaining, wherein n is natural number, finally, arithmetic element 32 obtains at least one echo texture index according to those local statistical parameters.Those local statistical parameters can comprise a plurality of local variations coefficient, a plurality of local mean values, a plurality of local variations numerical value or its combination, statistical parameter and echo texture minute hand system are obtained by arithmetic element 32 calculating, its computational methods, as described in the quantization method of aforementioned echo textural characteristics, are not added to repeat at this.
In a preferred embodiment, in the visualization system of echo textural characteristics, user watches ultrasonograph on the display device 50, and choose the chosen area in ultrasonograph through input equipment 40 after, analytic unit 31 receives chosen area, arithmetic element 32 to calculate in chosen area all pixel numerical value to obtain a regional standard difference, analytic unit 31 is got rid of in chosen area, pixel numerical value is less than the pixel of a reference value, wherein reference value is the multiplier value of regional standard difference, then, arithmetic element 32 is for remaining pixel, add up one group of pixel centered by one n-th pixel respectively, calculate one n-th local statistical parameter of one group of corresponding pixel numerical value, n is 1 to the number of this pixel remaining, wherein n is natural number, and local statistical parameter at least comprises a plurality of local mean values, arithmetic element 32 obtains at least one echo texture pointer according to those local statistical parameters, then, arithmetic element 32 calculates remaining pixel numerical value, to obtain a region averages, and the absolute difference calculating those local mean values and region averages is respectively to obtain the visual value of an echo texture strength of each pixel, finally, analytic unit 31 in conjunction with the visual value of echo texture strength and echo texture index in order to draw an echo texture image color range figure.Those local statistical parameters can comprise a plurality of local variations coefficient and a plurality of local variations numerical value further, statistical parameter, echo texture index and the visual value of echo texture strength and correlation values system are obtained by arithmetic element 32 calculating, its computational methods, as described in the method for visualizing of aforementioned echo textural characteristics, are not added to repeat at this.
In a further preferred embodiment, it is visual in present worth to obtain one that analytic unit 31 can compare dividing value and a floor value on the visual value of each echo texture strength and one further, wherein comparative approach and the visual acquisition mode in present worth are as described in the method for visualizing of aforementioned echo textural characteristics, do not add to repeat at this.And upper dividing value, floor value and correlation values system are obtained by arithmetic element 32 calculating, its computational methods similarly, as described in the method for visualizing of aforementioned echo textural characteristics, are not added to repeat at this.
Although above to invention has been detailed description, the present invention is not limited thereto, those skilled in the art of the present technique can carry out various amendment according to principle of the present invention.Therefore, all amendments done according to the principle of the invention, all should be understood to fall into protection scope of the present invention.
Claims (18)
1. a quantization method for echo textural characteristics, comprises:
Obtain a ultrasonograph, this ultrasonograph comprises several pixel, and each this pixel has a pixel numerical value;
Receive the chosen area in this ultrasonograph, and to calculate in this chosen area this pixel numerical value all to obtain a regional standard difference;
Get rid of pixel numerical value in this chosen area and be less than the pixel of a reference value, wherein said reference value is the multiplier value of this regional standard difference;
For remaining pixel, add up one group of pixel centered by one n-th pixel respectively, calculate one n-th local statistical parameter of one group of corresponding pixel numerical value, n is 1 to the number of this pixel remaining, and wherein n is natural number; And
At least one echo texture index is obtained according to described local statistical parameter.
2. quantization method as claimed in claim 1, wherein said local statistical parameter comprises several local variations coefficient, and described echo texture index obtains to obtain a local variations mean coefficient by calculating those local variations coefficients.
3. quantization method as claimed in claim 1, wherein said local statistical parameter comprises several local mean values and several local variations numerical value, the square root of the variation value of described local variations by calculating described local mean values and described local variations numerical value to obtain the variation value of a local variations mean values and a local variations, then obtains divided by this local variations mean values by described echo texture index.
4. quantization method as claimed in claim 1, wherein the scope of this group pixel is k × k square formation, and wherein k is natural number.
5. a method for visualizing for echo textural characteristics, comprises:
Obtain a ultrasonograph, described ultrasonograph comprises several pixel, and each this pixel has a pixel numerical value;
Receive the chosen area in described ultrasonograph, and to calculate in this chosen area this pixel numerical value all to obtain a regional standard difference;
Get rid of pixel numerical value in this chosen area and be less than the pixel of a reference value, wherein said reference value is the multiplier value of this regional standard difference;
For remaining pixel, add up one group of pixel centered by one n-th pixel respectively, calculate one n-th local statistical parameter of one group of corresponding pixel numerical value, n is the number of 1 to residual pixel point, wherein n is natural number, and described local statistical parameter comprises several local mean values;
At least one echo texture index is obtained according to described local statistical parameter;
Calculate residual pixel point value, to obtain a region averages, and the absolute difference calculating described local mean values and this region averages is respectively to obtain the visual value of an echo texture strength of each this pixel; And
In conjunction with the visual value of echo texture strength and echo texture index, draw an echo texture image color range figure.
6. method for visualizing as claimed in claim 5, comprises further:
Relatively dividing value and a floor value in the visual value of each echo texture strength and, visual in present worth to obtain one;
When the visual value of echo texture strength of arbitrary pixel is greater than described upper dividing value, visual is 0 in present worth;
When the visual value of echo texture strength of arbitrary pixel is less than described floor value, visual is color stage value in present worth; And
When the visual value of echo texture strength of arbitrary pixel is less than described upper dividing value and is greater than described floor value, visually deduct described floor value in present worth for dividing value on this deducts the visual value of echo texture strength divided by described upper dividing value, then be multiplied by described color range value.
7. method for visualizing as claimed in claim 6, wherein said local statistical parameter comprises several local variations numerical value further, and described method comprises further: calculate residual pixel point value, to obtain a regional standard difference parameter; And calculate described local mean values and described local variations numerical value to obtain the variation value of a local variations mean values and a local variations, described upper dividing value is the variation value of local variations mean values divided by local variations, be multiplied by regional standard difference parameter again divided by a upper bound parameter, and described floor value is the variation value of this local variations mean values divided by this local variations, then be multiplied by described regional standard difference parameter divided by a lower bound parameter.
8. method for visualizing as claimed in claim 6, wherein said local statistical parameter comprises several local variations coefficient further, and described method comprises further: calculate residual pixel point value, to obtain a regional standard difference parameter; And calculate described local variations coefficient to obtain a local variations mean coefficient, described upper dividing value is the inverse of this local variations mean coefficient, be multiplied by described regional standard difference parameter again divided by a upper bound parameter, and this floor value is the inverse of described local variations mean coefficient, then be multiplied by described regional standard difference parameter divided by a lower bound parameter.
9. quantization method as claimed in claim 7 or 8, wherein said upper bound parameter and lower bound parameter are a pair specific strength parameter, and described upper bound parameter and this lower bound parameter are between 0.5 ~ 10.0, and this lower bound parameter is greater than this upper bound parameter.
10. method for visualizing as claimed in claim 5, wherein the scope of this group pixel is k × k square formation, and wherein k is natural number.
The quantization system of 11. 1 kinds of echo textural characteristics, comprises:
One ultrasonograph capture device, for obtaining a ultrasonograph, this ultrasonograph comprises several pixel, and each this pixel has a pixel numerical value;
One image processing module, connects this ultrasonograph capture device, and for receiving and processing this ultrasonograph, described image processing module comprises an analytic unit and an arithmetic element, and this analytic unit and this arithmetic element are interconnected;
One display device, connects image processing module, for showing this ultrasonograph; And
One input equipment, connects image processing module, for choosing the chosen area in this ultrasonograph;
Wherein, analytic unit receives this chosen area, arithmetic element to calculate in this chosen area this pixel numerical value all to obtain a regional standard difference, this analytic unit gets rid of this pixel that pixel numerical value in this chosen area is less than a reference value, wherein said reference value is the multiplier value of this regional standard difference, this arithmetic element is for residual pixel point, add up one group of pixel centered by one n-th pixel respectively, calculate one n-th local statistical parameter of one group of corresponding pixel numerical value, n is the number of 1 to residual pixel point, wherein n is natural number, this arithmetic element obtains at least one echo texture index according to those local statistical parameters.
12. quantization systems as claimed in claim 11, wherein said local statistical parameter comprises several local variations coefficient, and described arithmetic element calculates described local variations coefficient and obtains this echo texture index to obtain a local variations mean coefficient.
13. quantization systems as claimed in claim 11, wherein said local statistical parameter comprises several local mean values and several local variations numerical value, described arithmetic element calculates described local mean values and those local variations numerical value to obtain the variation value of a local variations mean values and a local variations, then the square root of the variation value of described local variations is obtained this echo texture index divided by described local variations mean values.
The visualization system of 14. 1 kinds of echo textural characteristics, comprises:
One ultrasonograph capture device, for obtaining a ultrasonograph, described ultrasonograph comprises several pixel, and each pixel has a pixel numerical value;
One image processing module, connects this ultrasonograph capture device, and for receiving and processing described ultrasonograph, this image processing module comprises an analytic unit and an arithmetic element, and analytic unit and arithmetic element are interconnected;
One display device, connects described image processing module, in order to show this ultrasonograph; And
One input equipment, connects described image processing module, in order to choose the chosen area in this ultrasonograph;
Wherein this analytic unit receives this chosen area, this arithmetic element to calculate in this chosen area this pixel numerical value all to obtain a regional standard difference, this analytic unit gets rid of this pixel that pixel numerical value in this chosen area is less than a reference value, wherein this reference value is the multiplier value of this regional standard difference, this arithmetic element is for residual pixel point, add up one group of pixel centered by one n-th pixel respectively, calculate one n-th local statistical parameter of one group of corresponding pixel numerical value, n is the number of 1 to residual pixel point, wherein n is natural number, described local statistical parameter comprises several local mean values, this arithmetic element obtains at least one echo texture pointer according to described local statistical parameter, this arithmetic element calculates residual pixel point value, to obtain a region averages, and the absolute difference calculating described local mean values and this region averages is respectively to obtain the visual value of an echo texture strength of each this pixel, this analytic unit in conjunction with the visual value of this echo texture strength and this echo texture index in order to draw an echo texture image color range figure.
15. visualization systems as claimed in claim 15, it is visual in present worth to obtain one that wherein said analytic unit compares dividing value and a floor value in each visual value of this echo texture strength and one further, when the visual value of echo texture strength of arbitrary pixel is greater than upper dividing value, this is visual is 0 in present worth; When the visual value of this echo texture strength of arbitrary pixel is less than floor value, this is visual is color stage value in present worth; And when the visual value of echo texture strength of arbitrary pixel is less than upper dividing value and is greater than floor value, this is visual is that upper dividing value deducts the visual value of this echo texture strength and deducts floor value divided by upper dividing value in present worth, then is multiplied by this color range value.
16. as the visualization system of claim 15, wherein said local statistical parameter comprises several local variations numerical value further, described arithmetic element calculates residual pixel point value, to obtain a regional standard difference parameter, and calculate described local mean values and described local variations numerical value to obtain the variation value of a local variations mean values and a local variations, described arithmetic element is by the variation value of described local variations mean values divided by described local variations, be multiplied by described regional standard difference parameter again divided by a upper bound parameter to obtain dividing value on this, and by this local variations mean values divided by this local variations online variation value, be multiplied by described regional standard difference parameter again divided by a lower bound parameter to obtain this floor value.
17. visualization systems as claimed in claim 15, wherein said local statistical parameter comprises several local variations coefficient further, described arithmetic element calculates residual pixel point value, to obtain a regional standard difference parameter, and calculate described local variations coefficient to obtain a local variations mean coefficient, this arithmetic element is by the inverse of described local variations mean coefficient, be multiplied by this regional standard difference parameter again divided by a upper bound parameter to obtain dividing value on this, and by the inverse of this local variations mean coefficient, be multiplied by this regional standard difference parameter again divided by a lower bound parameter to obtain this floor value.
18. visualization systems as described in claim 16 or 17, wherein said upper bound parameter and described lower bound parameter are a pair specific strength parameter, and described upper bound parameter and described lower bound parameter are between 0.5 ~ 10.0, and lower bound parameter is greater than upper bound parameter.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410365753.2A CN105286912B (en) | 2014-07-29 | 2014-07-29 | The quantization of echo textural characteristics and method for visualizing and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410365753.2A CN105286912B (en) | 2014-07-29 | 2014-07-29 | The quantization of echo textural characteristics and method for visualizing and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105286912A true CN105286912A (en) | 2016-02-03 |
| CN105286912B CN105286912B (en) | 2019-03-01 |
Family
ID=55185235
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410365753.2A Expired - Fee Related CN105286912B (en) | 2014-07-29 | 2014-07-29 | The quantization of echo textural characteristics and method for visualizing and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105286912B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107169929A (en) * | 2016-03-07 | 2017-09-15 | 安克生医股份有限公司 | A kind of bearing calibration of have no reply region and high echo bright spot quantization characteristic suitable for different ultrasonic wave type images |
| CN107516306A (en) * | 2016-06-15 | 2017-12-26 | 安克生医股份有限公司 | Apnea disease appraisal procedure and its system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5656763A (en) * | 1994-08-16 | 1997-08-12 | Flextech Systems Inc. | Method for evaluating imaging capabilities of an ultrasound system |
| CN1235323A (en) * | 1998-04-14 | 1999-11-17 | 通用电气公司 | Method and apparatus for analyzing image structures |
-
2014
- 2014-07-29 CN CN201410365753.2A patent/CN105286912B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5656763A (en) * | 1994-08-16 | 1997-08-12 | Flextech Systems Inc. | Method for evaluating imaging capabilities of an ultrasound system |
| CN1235323A (en) * | 1998-04-14 | 1999-11-17 | 通用电气公司 | Method and apparatus for analyzing image structures |
Non-Patent Citations (1)
| Title |
|---|
| YAIR ZIMMER ET AL: "AN AUTOMATIC APPROACH FOR MORPHOLOGICAL ANALYSIS AND MALIGNANCY EVALUATION OF OVARIAN MASSES USING B-SCANS", 《ULTRASOUND IN MEDICINE AND BIOLOGY》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107169929A (en) * | 2016-03-07 | 2017-09-15 | 安克生医股份有限公司 | A kind of bearing calibration of have no reply region and high echo bright spot quantization characteristic suitable for different ultrasonic wave type images |
| CN107516306A (en) * | 2016-06-15 | 2017-12-26 | 安克生医股份有限公司 | Apnea disease appraisal procedure and its system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105286912B (en) | 2019-03-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Verdun et al. | Image quality in CT: From physical measurements to model observers | |
| EP2102789B1 (en) | System and method for feature score mapping and visualization of medical images | |
| CN102881010B (en) | Method for evaluating perception sharpness of fused image based on human visual characteristics | |
| CN102743181B (en) | Image processing apparatus and image processing method | |
| US20090082637A1 (en) | Multi-modality fusion classifier with integrated non-imaging factors | |
| CN102740114B (en) | A kind of nothing ginseng appraisal procedure of Subjective video quality | |
| KR101711228B1 (en) | Method for quantifying the development of diseases involving changes in the volume of bodies, in particular tumours | |
| JP4383352B2 (en) | Histological evaluation of nuclear polymorphism | |
| Chawla et al. | Effect of dose reduction on the detection of mammographic lesions: a mathematical observer model analysis | |
| Sindhwani et al. | Semi‐automatic outlining of levator hiatus | |
| Trivedi et al. | A no-reference image quality index for contrast and sharpness measurement | |
| CN105286912A (en) | System and method for quantization and visualization of echo texture features | |
| US9773307B2 (en) | Quantification and imaging methods and system of the echo texture feature | |
| Grob et al. | Measuring echogenicity and area of the puborectalis muscle: method and reliability | |
| Albiol et al. | Automatic intensity windowing of mammographic images based on a perceptual metric | |
| EP2980757B1 (en) | Quantification and imaging methods of the echo-texture feature | |
| Brennan et al. | BREAST: a novel method to improve the diagnostic efficacy of mammography | |
| Abbey et al. | Optimal shifted estimates of human-observer templates in two-alternative forced-choice experiments | |
| US11561273B2 (en) | Medical image processing apparatus and method | |
| CN111784637B (en) | Prognostic characteristic visualization method, system, equipment and storage medium | |
| KR100825960B1 (en) | Method of assigning a landmark on a cephalometric radiograph | |
| Platiša et al. | Volumetric detection tasks with varying complexity: human observer performance | |
| CN100431494C (en) | Method and apparatus for analyzing biological tissue images | |
| TWI474284B (en) | Quantification method of the feature of a tumor and an imaging method of the same | |
| CN107440687A (en) | A kind of pain grade appraisal procedure and use its pain grade apparatus for evaluating |
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 |
Granted publication date: 20190301 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |