CN109978823B - Elastography image analysis and tissue viscoelasticity detection method and device - Google Patents
Elastography image analysis and tissue viscoelasticity detection method and device Download PDFInfo
- Publication number
- CN109978823B CN109978823B CN201910117697.3A CN201910117697A CN109978823B CN 109978823 B CN109978823 B CN 109978823B CN 201910117697 A CN201910117697 A CN 201910117697A CN 109978823 B CN109978823 B CN 109978823B
- Authority
- CN
- China
- Prior art keywords
- stripe
- determining
- main
- image
- main stripe
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10132—Ultrasound image
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a method and a device for elastography image analysis and tissue viscoelasticity detection, which determine main stripes by carrying out image analysis on an E hypergraph; determining an evaluation parameter for evaluating the image quality of the E-hypergraph based on the main stripes; and determining the quality of the E-hypergraph according to the evaluation parameters. The evaluation parameters obtained after the E-hypergraph is analyzed are objectively evaluated based on a unified evaluation standard, and the objectivity and the uniformity of the evaluation of the image quality of the E-hypergraph are guaranteed; the qualified E-hypergraph is adopted to judge the hardness grade of the human tissue, so that the accuracy of the judgment result is ensured.
Description
Technical Field
The invention relates to an image analysis technology, in particular to a method and a device for elastography image analysis and tissue viscoelasticity detection.
Background
The E ultrasonic imaging, namely Elasticity (Elasticity) imaging, called E ultrasonic map for short, is two-dimensional real-time shear wave Elasticity imaging, and can be applied to superficial tissues, abdominal tissues, intracavity tissues and the like to measure and image the hardness of the tissues. In E-mode ultrasound imaging, the horizontal axis represents time, the vertical axis represents scan depth, and the gray scale value of the image is the intensity of elastic ultrasound detected at a certain depth at a certain time. The basic feature of the E-map is that, depending on the propagation of several elastic shear waves in soft tissue, the E-map has 3 to 5 alternating black and white stripes of substantially similar width and spacing, the stripes generally following a distribution extending from the upper left to the lower right of the E-map. For a high-quality E-hypergram, the stripes are clear and long enough, and the width change is small, so that the propagation speed of the elastic shear wave calculated by the stripes can be more accurately used for evaluating the hardness level of human tissues.
However, in the conventional E-ultrasonography, it is usually necessary to rely on the medical staff to visually evaluate the imaging quality of the E-ultrasonogram, scan for a plurality of times continuously, select the E-ultrasonogram with better visual inspection, and then obtain the hardness value of a tissue according to the E-ultrasonogram subjectively determined by the medical staff. Because different testers exist, the evaluation hardly has an objective and unified standard, and further, the accuracy of judging the hardness grade of the human tissue is hardly ensured.
Disclosure of Invention
In order to solve the technical problems that the evaluation of the E-hypergraph in the prior art is lack of objectivity and the accurate evaluation of the hardness level of human tissues is difficult to ensure, the invention provides the method and the device for analyzing the elastic imaging image and detecting the tissue viscoelasticity, which can ensure the objectivity and the uniformity of the evaluation of the E-hypergraph and the accuracy of the judgment of the hardness level of the human tissues.
The invention provides an elastography image analysis method, which comprises the following steps:
performing image analysis on the elastic imaging image to determine a main stripe;
determining an evaluation parameter for evaluating the image quality of the elastography image based on the main stripe;
and determining the quality of the elastography image according to the evaluation parameters.
Optionally, the performing image analysis on the elastography image to determine the main stripe includes:
according to a first gray threshold, carrying out image segmentation on the elastic imaging image based on a region growing method;
and determining the stripe with the largest area as the main stripe in the elastography image after image segmentation.
Optionally, before determining an evaluation parameter for evaluating the image quality of the elastography image based on the main stripe, the method further includes:
performing principal component analysis on the main stripe to determine the direction of the main stripe;
determining a coordinate axis of the main stripe according to the direction of the main stripe, and projecting each pixel on the main stripe onto the coordinate axis; the direction of the main stripe is a first main component axis, and the direction perpendicular to the main stripe is a second main component axis.
Optionally, the evaluation parameter includes at least one of the following parameters:
the method comprises the following steps of main stripe symmetry, main stripe width standard difference, elasticity imaging image definition, parallelism between a main stripe and other stripes in an elasticity imaging image, main stripe length, size of each stripe in the elasticity imaging image, main stripe direction and position of the main stripe in the elasticity imaging image.
Optionally, the evaluation parameter is the symmetry of the main stripe; correspondingly, the determining an evaluation parameter for evaluating the image quality of the elastography image based on the main stripe comprises:
determining a statistical boundary of the main stripe according to the outline of the main stripe;
and determining the symmetry of the main stripe relative to the symmetry axis within the statistical boundary range by taking the first main component axis as the symmetry axis.
Optionally, the evaluation parameter is the standard deviation of the width of the main stripe; correspondingly, the determining an evaluation parameter for evaluating the image quality of the elastography image based on the main stripe comprises:
determining a statistical boundary of the main stripe according to the outline of the main stripe;
within the statistical boundary range, obtaining the width mean value of the main stripe and obtaining the tail end position of the main stripe;
the tail end position points to the direction of the head end position of the main stripe, and the width value of the main stripe is obtained;
and determining the position of which the width value is greater than or equal to the width mean value as a starting position, starting from the starting position to the head end of the main stripe of the statistical boundary, and calculating the width standard deviation of the main stripe.
Optionally, the evaluation parameter is the sharpness of the elasticity imaging image; correspondingly, the determining an evaluation parameter for evaluating the image quality of the elastography image based on the main stripe comprises:
determining the proportion value of pixel points meeting the preset gray scale range in all pixel points in the elastic imaging image according to the preset gray scale range, and generating a definition value;
determining the central position of the main stripe, and if the area value of the main stripe is less than or equal to a preset area threshold and the distance between the central position of the main stripe and the central position of the elastography image is greater than a preset distance threshold, determining a first penalty weight of the elastography image;
determining the central position of the main stripe in the elastic imaging image, wherein the central position is perpendicular to the projection axis of the first main component shaft;
projecting each pixel point with the gray value larger than a second gray threshold value to the projection axis; dividing the distance of each preset number of pixels at two sides into a statistical band by taking the main component shaft as a central axis;
determining the ratio of the number of the pixel points with the gray value in each statistical band larger than the second gray threshold value to the number of all the pixel points in the whole statistical band, and generating a frequency value;
forming a column statistical chart by taking the number of the statistical bands as an abscissa and the frequency value as an ordinate;
smoothing the histogram, calculating the ratio of the standard deviation of the height of each statistical band to the mean value of the height of the band after smoothing, and determining a second penalty weight according to the ratio;
and determining the definition of the elastic imaging image according to the definition value, the first penalty weight and the second penalty weight.
Optionally, the evaluation parameter is the parallelism between the main stripe and other stripes in the elastic imaging image; correspondingly, the determining an evaluation parameter for evaluating the image quality of the elastography image based on the main stripe comprises:
determining stripes with the second largest area as secondary main stripes in an elastic imaging image after image segmentation;
performing principal component analysis on the secondary main stripe to determine the direction of the secondary main stripe;
and acquiring an included angle between the direction of the main stripe and the direction of the secondary main stripe, and determining the parallelism between the main stripe and the secondary main stripe according to the included angle.
Optionally, the evaluation parameter is the length of the main stripe; correspondingly, the determining an evaluation parameter for evaluating the image quality of the elastography image based on the main stripe comprises:
taking the direction of the main stripe as the length direction of the main stripe to obtain the length value of the main stripe;
acquiring the length value of an image diagonal of the elastic imaging image;
and calculating the ratio of the length value of the main stripe to the image diagonal length value of the elastography image, and determining the evaluation parameter of the length of the main stripe.
Optionally, the determining the quality of the elastography image according to the evaluation parameter includes:
scoring one or more items in the evaluation parameters according to a preset evaluation threshold, and if the sum of the scores is lower than the preset score threshold, determining that the quality of the elastic imaging image is unqualified; and if the sum of the scores is higher than or equal to a preset score threshold value, determining that the quality of the elastography image is qualified.
The present invention also provides an elastography image analysis apparatus, comprising:
the determining module is used for carrying out image analysis on the elastic imaging image and determining the main stripe;
the evaluation module is used for determining an evaluation parameter for evaluating the image quality of the elastography image based on the main stripe;
the determining module is further configured to determine the quality of the elastography image according to the evaluation parameter.
Optionally, the determining module includes:
the segmentation submodule is used for carrying out image segmentation on the elastic imaging image based on a region growing method according to a first gray threshold;
and the first determining submodule is used for determining the stripe with the largest area as the main stripe in the elasticity imaging image after image segmentation.
Optionally, the determining module is further configured to perform principal component analysis on the main stripe, and determine a direction of the main stripe; determining a coordinate axis of the main stripe according to the direction of the main stripe, and projecting each pixel on the main stripe onto the coordinate axis; the direction of the main stripe is a first main component axis, and the direction perpendicular to the main stripe is a second main component axis.
Optionally, the evaluation parameter includes at least one of the following parameters:
the method comprises the following steps of main stripe symmetry, main stripe width standard difference, elasticity imaging image definition, parallelism between a main stripe and other stripes in an elasticity imaging image, main stripe length, size of each stripe in the elasticity imaging image, main stripe direction and position of the main stripe in the elasticity imaging image.
Optionally, the evaluation parameter is the symmetry of the main stripe; accordingly, the determining module comprises:
the second determining submodule is used for determining the statistical boundary of the main stripe according to the outline of the main stripe; and determining the symmetry of the main stripe relative to the symmetry axis within the statistical boundary range by taking the first main component axis as the symmetry axis.
Optionally, the evaluation parameter is the standard deviation of the width of the main stripe; accordingly, the determining module comprises:
the third determining submodule is used for determining the statistical boundary of the main stripe according to the outline of the main stripe; within the statistical boundary range, obtaining the width mean value of the main stripe and obtaining the tail end position of the main stripe; the tail end position points to the direction of the head end position of the main stripe, and the width value of the main stripe is obtained; and determining the position of which the width value is greater than or equal to the width mean value as a starting position, starting from the starting position to the head end of the main stripe of the statistical boundary, and calculating the width standard deviation of the main stripe.
Optionally, the evaluation parameter is the sharpness of the elasticity imaging image; accordingly, the determining module comprises:
the fourth determining submodule is used for determining the proportion value of pixel points meeting the preset gray scale range in all the pixel points in the elastic imaging image according to the preset gray scale range and generating a definition value; determining the central position of the main stripe, and if the area value of the main stripe is less than or equal to a preset area threshold and the distance between the central position of the main stripe and the central position of the elastography image is greater than a preset distance threshold, determining a first penalty weight of the elastography image; determining the central position of the main stripe in the elastic imaging image, wherein the central position is perpendicular to the projection axis of the first main component shaft; projecting each pixel point with the gray value larger than a second gray threshold value to the projection axis; dividing the distance of each preset number of pixels at two sides into a statistical band by taking the main component shaft as a central axis; determining the ratio of the number of the pixel points with the gray value in each statistical band larger than the second gray threshold value to the number of all the pixel points in the whole statistical band, and generating a frequency value; forming a column statistical chart by taking the number of the statistical bands as an abscissa and the frequency value as an ordinate; smoothing the histogram, calculating the ratio of the standard deviation of the height of each statistical band to the mean value of the height of the band after smoothing, and determining a second penalty weight according to the ratio; and determining the definition of the elastic imaging image according to the definition value, the first penalty weight and the second penalty weight.
Optionally, the evaluation parameter is the parallelism between the main stripe and other stripes in the elastic imaging image; accordingly, the determining module comprises:
the fifth determining submodule is used for determining the stripe with the second largest area as a secondary main stripe in the elastic imaging image after image segmentation; performing principal component analysis on the secondary main stripe to determine the direction of the secondary main stripe; and acquiring an included angle between the direction of the main stripe and the direction of the secondary main stripe, and determining the parallelism between the main stripe and the secondary main stripe according to the included angle.
Optionally, the evaluation parameter is the length of the main stripe; accordingly, the determining module comprises:
a sixth determining submodule, configured to obtain a length value of the main stripe with the direction of the main stripe as the length direction of the main stripe; acquiring the length value of an image diagonal of the elastic imaging image; and calculating the ratio of the length value of the main stripe to the image diagonal length value of the elastography image, and determining the evaluation parameter of the length of the main stripe.
Optionally, the determining module includes:
the evaluation submodule is used for scoring one or more items in the evaluation parameters according to a preset evaluation threshold value, and if the sum of the scores is lower than the preset evaluation threshold value, the quality of the elastic imaging image is determined to be unqualified; and if the sum of the scores is higher than or equal to a preset score threshold value, determining that the quality of the elastography image is qualified.
The invention also provides a tissue viscoelasticity detection method, which comprises the following steps:
performing tissue viscoelasticity detection, outputting one or more elastographic images;
evaluating the quality of the one or more elastographic images according to any of the methods described above; and determining tissue viscoelasticity information according to the elastography image of which the evaluation result meets the preset standard.
The present invention also provides a tissue viscoelasticity detection apparatus comprising:
a detection module for performing tissue viscoelasticity detection;
an output module for outputting one or more elastographic images;
a quality evaluation module for evaluating the quality of the one or more elastographic images according to any of the above-mentioned means;
and the tissue viscoelasticity determining module is used for determining tissue viscoelasticity information according to the elastography image of which the evaluation result meets the preset standard.
According to the method and the device for analyzing the elastography image and detecting the tissue viscoelasticity, the main stripe is determined by performing image analysis on the elastography image; determining an evaluation parameter for evaluating the image quality of the elastography image based on the main stripe; and determining the quality of the elastography image according to the evaluation parameters. The method realizes objective evaluation of the evaluation parameters obtained after the elastography image is analyzed based on a uniform evaluation standard, and ensures the objectivity and uniformity of the evaluation of the image quality of the elastography image; the qualified elastography image is adopted to judge the hardness grade of the human tissue, so that the accuracy of the judgment result is ensured.
Drawings
FIG. 1 is a flow chart illustrating a method for E-hypergraph analysis of the present invention in an exemplary embodiment;
FIG. 2 is a schematic diagram of an E-hypergraph image of the embodiment shown in FIG. 1;
FIG. 3 is a flow chart of a method of the present invention for E-hypergraph analysis in accordance with another exemplary embodiment;
FIG. 4 is a schematic diagram of a main stripe image of the embodiment shown in FIG. 3;
FIG. 5 is a schematic view of another main fringe image of the embodiment shown in FIG. 3;
FIG. 6 is a schematic diagram of a main stripe profile image of the embodiment shown in FIG. 3;
FIG. 7 is a schematic diagram of an E-hypergraph image of the embodiment shown in FIG. 3;
FIG. 8 is a schematic illustration of a histogram statistical representation of the E-hypergraph of the embodiment shown in FIG. 3;
FIG. 9a is a schematic band diagram of the embodiment of FIG. 3 before E-hypergraph smoothing;
FIG. 9b is a schematic band diagram of the embodiment of FIG. 3 after E-map smoothing;
FIG. 10 is a schematic structural diagram of an E-hypergraph analysis apparatus of the present invention according to an exemplary embodiment;
FIG. 11 is a schematic structural diagram of an E-hypergraph analysis apparatus according to another exemplary embodiment of the present invention;
FIG. 12 is a flow chart illustrating a method of tissue viscoelasticity detection in accordance with an exemplary embodiment;
fig. 13 is a schematic structural diagram of a tissue viscoelasticity detection apparatus according to an exemplary embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a flowchart illustrating an E-hypergraph analysis method according to an exemplary embodiment of the present invention, and as shown in fig. 1, the E-hypergraph analysis method according to the present invention includes:
and 101, carrying out image analysis on the E-hypergraph to determine a main stripe.
Specifically, as shown in FIG. 2, the E-map is basically characterized by 3 to 5 alternating black and white stripes having substantially similar widths and spacings, and the stripes generally follow a distribution extending from the top left corner to the bottom right corner of the E-map. In the stripes, a representative stripe can be determined as a main stripe according to the stripe characteristics, such as the stripe width, length, area, contour and the like, and the image characteristics of the whole E-super map can be obtained by analyzing the main stripe due to the similarity among the stripes.
And 102, determining an evaluation parameter for evaluating the image quality of the E-hypergraph based on the main stripes.
And 103, determining the quality of the E-hypergraph according to the evaluation parameters.
Specifically, stripe morphological feature analysis is carried out on the main stripes determined in the previous step to obtain some evaluation parameters which can be used for evaluating the image quality of the E-hypergraph, so that the evaluation parameters are evaluated according to a uniform evaluation standard, and the objective evaluation on the image quality of the E-hypergraph is realized.
According to the E hypergraph analysis method, the main stripes are determined by carrying out image analysis on the E hypergraph; determining an evaluation parameter for evaluating the image quality of the E-hypergraph based on the main stripes; and determining the quality of the E-hypergraph according to the evaluation parameters. The evaluation parameters obtained after the E-hypergraph is analyzed are objectively evaluated based on a unified evaluation standard, and the objectivity and the uniformity of the evaluation of the image quality of the E-hypergraph are guaranteed; the qualified E-hypergraph is adopted to judge the hardness grade of the human tissue, so that the accuracy of the judgment result is ensured.
Fig. 3 is a flowchart of an E-hypergraph analysis method according to another exemplary embodiment of the present invention, and as shown in fig. 3, the E-hypergraph analysis method according to the present embodiment includes:
and 301, according to the first gray threshold, carrying out image segmentation on the E-hypergraph based on a region growing method.
In particular, region growing (region growing) is a process that develops groups of pixels or regions into larger regions. Starting from the set of seed points, the region from these points grows by merging into this region neighboring pixels with similar properties like intensity, grey level, texture color, etc. as each seed point. The first gray threshold is a gray value determined from gray levels of 0 to 255, generally, white is 255, black is 0, the first gray threshold can be set to 240, for example, and pixels exceeding the first gray threshold 240 will show a whitish color, so as to extract white main stripes for study; of course, the first gray threshold may be modified, that is, a division standard with a gray level smaller than a certain gray threshold is adopted, so as to extract the black main stripe for study. The gray color between black and white in the E-mode super image after image segmentation is greatly reduced, and the E-mode super image shows clear black and white color, as shown in fig. 2.
Optionally, before step 201, a picture size adjustment may be performed on the obtained E-hypergraph, so as to adjust the picture size of the E-hypergraph to be analyzed to the same standard size, for example, a size of 100 × 100. The step of adjusting the size has little influence on the evaluation result of the quality of the E-hypergraph, but the analysis time of the E-hypergraph with uniform size can be greatly saved.
And step 302, determining the stripe with the largest area as a main stripe in the E-hypergraph after image segmentation.
Specifically, in fig. 2 after image division, the stripe having the largest area among the respective stripes is determined as the main stripe.
And step 303, performing principal component analysis on the main stripe to determine the direction of the main stripe.
Specifically, Principal Component Analysis (PCA) is performed on the main stripe, and is a statistical method that converts a set of variables that may have correlation into a set of linearly uncorrelated variables through orthogonal transformation, and the set of converted variables is called a main Component. The direction of the main stripe (shown as the X-axis in fig. 4) is obtained by performing a main component analysis on the main stripe.
And 304, determining a coordinate axis of the main stripe according to the direction of the main stripe, and projecting each pixel on the main stripe onto the coordinate axis.
Specifically, the direction of the main stripe is a first main component axis (i.e., X axis in fig. 4), and the direction perpendicular to the main stripe is a second main component axis (i.e., Y axis in fig. 4). All pixels on the main stripe are projected to the two directions (X-axis and Y-axis) to obtain the projected E-super map, as shown in fig. 5.
Step 305 determines an evaluation parameter for evaluating the image quality of the E-hypergraph based on the main streak.
Specifically, the evaluation parameter may include at least one of the following parameters:
the main stripe symmetry, the main stripe width standard difference, the definition of the E hypergraph, the parallelism between the main stripe and other stripes in the E hypergraph, the length of the main stripe, the size of each stripe in the E hypergraph, the direction of the main stripe and the position of the main stripe in the E hypergraph.
How to determine the evaluation parameters for evaluating the image quality of the E-hypergraph based on the main stripes is described in detail below according to different evaluation parameters:
step 305A, if the evaluation parameter is the symmetry of the main stripe, determining a statistical boundary of the main stripe according to the outline of the main stripe; and determining the symmetry of the main stripe relative to the symmetry axis within the statistical boundary range by taking the first main component axis as the symmetry axis.
As shown in fig. 6, the maximum value and the minimum value of the Y coordinate of the contour line corresponding to each scale on the X axis are counted based on the contour of the main stripe. Where the two points a and B are boundary points to the left and right along the first principal component axis X, respectively. As can be seen from fig. 6, the left side of the boundary line (Y axis) is an area affected by the image boundary, and the left side of the boundary line Y is a severely asymmetric area, so that in order to eliminate the boundary effect, the statistical boundary of the main stripe is determined as a CB zone, and the symmetry of the main stripe with respect to the symmetry axis is determined within the statistical boundary CB range by taking the first principal component axis X as the symmetry axis.
sym＝mean(|ymax(x)+ymin(x) I) x ∈ (C, B), where sym represents a calculated value of symmetry (symmetry), and mean represents an average of the maximum and minimum values of y for each x in the CB interval.
Step 305B, if the evaluation parameter is the standard difference of the width of the main stripe, determining the statistical boundary of the main stripe according to the outline of the main stripe; within the statistical boundary range, obtaining the width mean value of the main stripe and obtaining the tail end position of the main stripe; the width value of the main stripe is obtained from the direction from the tail end position to the head end position of the main stripe; and determining the position of which the width value is greater than or equal to the width mean value as a starting point position, starting from the starting point position to the head end of the main stripe of the statistical boundary, and calculating the width standard deviation of the main stripe.
Specifically, the measure of the standard deviation of the width of the main stripe can be calculated by the following formula:
var＝std(ymax(x)-ymin(x) X is belonged to (C, B), wherein var represents a variance value (variance), which measures the variation amplitude of the boundary width of the main stripe; std represents the Standard Deviation (also known as mean square error or Standard Deviation), which reflects the degree of dispersion of a data set. It is wideThe width standard deviation can measure the width variation range by measuring the difference of the average values, namely, even if two main stripes with the same width average value exist, the width variation of the two main stripes can be different. In some cases, the tail end of the main stripe may be very long and narrow, and in order to eliminate the influence of the tail end, a search is performed from the tail end position B to the head end C pointing to the statistical boundary, and a point with an outline width larger than the width average is searched as a starting point position, and the standard deviation of the main stripe width is counted starting from the starting point position to the head end C of the main stripe of the statistical boundary.
Step 305C, if the evaluation parameter is E-hypergraph definition; determining the proportion value of the pixel points meeting the preset gray scale range in all the pixel points in the E-hypergraph according to the preset gray scale range to generate a definition value; determining the central position of the main stripe, and if the area value of the main stripe is less than or equal to a preset area threshold and the distance between the central position of the main stripe and the central position of the E-hypergraph is greater than a preset distance threshold, determining a first penalty weight of the E-hypergraph; in the E-ultrasonic image, determining the central position of a main stripe, and the central position is vertical to the projection axis of the first main component shaft; projecting each pixel point with the gray value larger than the second gray threshold value to the projection axis; dividing the distance of each preset number of pixels at two sides into a statistical band by taking the main component axis as a central axis; determining the ratio of the number of the pixel points with the gray value in each statistical band larger than the second gray threshold value to the number of all the pixel points in the whole statistical band, and generating a frequency value; forming a bar-shaped statistical graph by taking the number of the statistical bands as an abscissa and the frequency value as an ordinate; smoothing the histogram, calculating the ratio of the standard deviation of the height of each statistical band to the mean value of the height of the band after smoothing, and determining a second punishment weight according to the ratio; and determining the definition of the E-hypergraph according to the definition value, the first penalty weight and the second penalty weight.
Specifically, for the clarity of the E-map, the following formula can be used for measurement,
quality＝clear×k0×krwhere quality represents the evaluation value of the definition of the E-scope, clear represents the definition value, k0Represents a first penalty weight, krRepresents the secondAnd punishing the weight.
Considering that the pixels with the gray scale between 0-50 and 200-255 are black and white pixels, the percentage of the part of pixels to all pixels is the clear value. The preset gray scale range may be a gray scale between 0-50 and 200-255, or other gray scale ranges determined by those skilled in the art according to statistical requirements.
First penalty weight k0The penalty weight value for the clear value is introduced when the area of the main stripe is too small and the distance between the center position of the main stripe and the center of the image is too far.
Second penalty weight krIs a coefficient between 0 and 1 and is also used to penalize poor quality E-hypergraph samples.
The second penalty weight k is explained belowrAs shown in fig. 7, the Z-axis is a central position O of the main stripe and is perpendicular to the projection axis of the first principal component axis X. Projecting all pixel points (hereinafter referred to as white points) with the gray levels larger than a second gray level threshold (for example, the gray level is 150) to the Z axis, taking the distance between every two preset pixel numbers (for example, every 2 pixels) on both sides of the X axis as a statistical band, recording the white point number in each statistical band divided by the total pixel number, which is called as a frequency value, forming a histogram (as shown in fig. 8) with the statistical band number as the horizontal coordinate and the frequency value as the vertical coordinate, and simultaneously recording the average gray level of the white points in the statistical band divided by 256. According to the histogram shown in fig. 8, the significant depressions on both sides of the main stripe are divided into a main stripe and left and right stripes, such as 4 stripes shown in fig. 7, one stripe is located on the left side of the main stripe (i.e., left stripe) and two stripes are located on the right side of the main stripe (i.e., right stripe). And smoothing the belt height in each belt area, and when a certain belt height is larger than the left adjacent belt height and the right adjacent belt height, making the belt height be the average value of the two adjacent belt heights. Fig. 9a and 9b are graphs showing comparison of effects before and after the smoothing process. And judging whether the belt zone only has one independent clear stripe, otherwise, continuously judging whether the distribution of the belt height in the belt zone has obvious interruption, wherein the interruption is obvious, each stripe in the belt zone is clear, and the direction consistency with the main stripe is good. For this purpose, the respective band height standard deviations and the average of the left and right band regions are recordedThe ratio of the values is that of,
rate ═ var/mean; then correspondingly, a second penalty weight krAs determined by the rate, the rate is,
it can be seen that when the white dot distribution on the left and right sides of the main stripe is less striped, or the stripe direction is less uniform, the rate is smaller, krThe smaller the penalty on E-map sharpness (quality value), the better the image quality is represented.
Step 305D, if the evaluation parameter is the parallelism between the main stripe and other stripes in the E-hypergraph; determining the stripe with the second largest area as a secondary main stripe in the E-hypergraph after image segmentation; performing principal component analysis on the secondary main stripes to determine the direction of the secondary main stripes; and acquiring an included angle between the direction of the main stripe and the direction of the secondary main stripe, and determining the parallelism between the main stripe and the secondary main stripe according to the included angle.
The parallelism can be a sine value of an included angle between central lines of the two largest main stripes, and the sine value is in direct proportion to the included angle when the included angle is small, so that sine is more suitable for calculation than cosine, for example, when the evaluation parameters are used as input of a neural network, the sine value of the included angle can be used as input of the parallelism between the main stripe and other stripes in the E-hypergraph.
Step 305E, if the evaluation parameter is the main stripe length; then, the length value of the main stripe (i.e. the length of AB in fig. 6) is obtained by taking the direction of the main stripe as the length direction of the main stripe; obtaining the length value of an image diagonal line of the E-hypergraph; and calculating the ratio of the length value of the main stripe to the length value of the diagonal line of the image of the E-hypergraph, and determining the evaluation parameter of the length of the main stripe.
And 307, if the sum of the scores is lower than a preset score threshold, determining that the quality of the E-hypergraph is unqualified.
And 308, if the sum of the scores is higher than or equal to a preset score threshold value, determining that the quality of the E-hypergraph is qualified.
Specifically, one or more of the evaluation parameters may be used to measure the quality of the E-hypergraph, and the quality of the E-hypergraph is determined by integrating the scores of the one or more evaluation parameters. Where a decision tree approach may be used, samples of particularly poor quality are given a score of 0 (or a 0-point rating if some "normal" sample is scored less than 0.5 by the neural network) before the E-hypergraph sample is fed into the neural network. When the following cases occur, the area of 1, the area size column of which is the area of the secondary main stripe, is too small, which is calculated as an abnormal case with extremely poor quality; 2. the included angle between the main stripe and the secondary main stripe is more than 30 degrees; 3. the main stripes are distributed in a reverse oblique manner, namely from the upper right to the lower left; 4. the center of the main stripe is too close to the image boundary; 5. the main stripe symmetry is greater than a certain threshold; 6. the standard deviation of the width of the main stripe is larger than a certain threshold value; 7. the main stripe is too short; 8. image sharpness less than a threshold, and so on.
A person skilled in the art can use the E-hypergraph analysis method described in each of the above embodiments to divide a sample image of the E-hypergraph into a training sample and a test sample, extract each of the above evaluation parameters, train the learning network with the training sample, then score the test sample, and compare the score with a manual score. The selected learning network can be a BP neural network, and the evaluation parameters are input by the network and scored as output. It should be noted that the machine learning method includes not only the neural network mentioned in this patent, but also supervised or unsupervised learning methods such as decision trees, support vector machines, and the like.
Instead of manually checking the quality of the scan, the quality of the E-hypergraph is automatically evaluated by employing machine learning. The E-hypergraph analysis method has the advantages of being rapid and unified in standard.
Fig. 10 is a schematic structural diagram of an E-hypergraph analysis apparatus according to an exemplary embodiment of the present invention, and as shown in fig. 10, the E-hypergraph analysis apparatus according to the present embodiment includes:
the determining module 1 is used for carrying out image analysis on the E-hypergraph and determining main stripes;
the evaluation module 2 is used for determining an evaluation parameter for evaluating the image quality of the E-hypergraph based on the main stripes;
the determining module 1 is further configured to determine the quality of the E-hypergraph according to the evaluation parameters.
The embodiment can be used to implement the embodiment shown in fig. 1, and the implementation principle is similar, which is not described herein again.
The E-hypergraph analysis device of the embodiment determines the main stripe by performing image analysis on the E-hypergraph; determining an evaluation parameter for evaluating the image quality of the E-hypergraph based on the main stripes; and determining the quality of the E-hypergraph according to the evaluation parameters. The evaluation parameters obtained after the E-hypergraph is analyzed are objectively evaluated based on a unified evaluation standard, and the objectivity and the uniformity of the evaluation of the image quality of the E-hypergraph are guaranteed; the qualified E-hypergraph is adopted to judge the hardness grade of the human tissue, so that the accuracy of the judgment result is ensured.
Fig. 11 is a schematic structural diagram of an E-map analysis apparatus according to another exemplary embodiment of the present invention, and as shown in fig. 11, based on the above-mentioned embodiment, in the E-map analysis apparatus according to this embodiment,
the determination module 1 includes:
the segmentation submodule 11 is used for carrying out image segmentation on the E-hypergraph based on a region growing method according to a first gray threshold;
the first determining submodule 12 is configured to determine, in the E-hypergraph after image segmentation, a stripe with the largest area as a main stripe.
Optionally, the determining module 1 is further configured to perform principal component analysis on the main stripe, and determine a direction of the main stripe; determining a coordinate axis of the main stripe according to the direction of the main stripe, and projecting each pixel on the main stripe onto the coordinate axis; the direction of the main stripe is a first main component axis, and the direction perpendicular to the main stripe is a second main component axis.
Optionally, the evaluation parameter includes at least one of the following parameters:
the main stripe symmetry, the main stripe width standard difference, the definition of the E hypergraph, the parallelism between the main stripe and other stripes in the E hypergraph, the length of the main stripe, the size of each stripe in the E hypergraph, the direction of the main stripe and the position of the main stripe in the E hypergraph.
Optionally, the evaluation parameter is the symmetry of the main stripe; accordingly, the determination module 1 includes:
the second determining submodule 13 is configured to determine a statistical boundary of the main stripe according to the contour of the main stripe; and determining the symmetry of the main stripe relative to the symmetry axis within the statistical boundary range by taking the first main component axis as the symmetry axis.
Optionally, the evaluation parameter is a main stripe width standard difference; accordingly, the determination module 1 includes:
a third determining submodule 14, configured to determine a statistical boundary of the main stripe according to the contour of the main stripe; within the statistical boundary range, obtaining the width mean value of the main stripe and obtaining the tail end position of the main stripe; the width value of the main stripe is obtained from the direction from the tail end position to the head end position of the main stripe; and determining the position of which the width value is greater than or equal to the width mean value as a starting point position, starting from the starting point position to the head end of the main stripe of the statistical boundary, and calculating the width standard deviation of the main stripe.
Optionally, the evaluation parameter is E-hypergraph definition; accordingly, the determining module includes:
the fourth determining submodule 15 is configured to determine, according to the preset gray scale range, a ratio value of pixel points satisfying the preset gray scale range to all pixel points in the E-hypergraph, and generate a sharpness value; determining the central position of the main stripe, and if the area value of the main stripe is less than or equal to a preset area threshold and the distance between the central position of the main stripe and the central position of the E-hypergraph is greater than a preset distance threshold, determining a first penalty weight of the E-hypergraph; in the E-ultrasonic image, determining the central position of a main stripe, and the central position is vertical to the projection axis of the first main component shaft; projecting each pixel point with the gray value larger than the second gray threshold value to the projection axis; dividing the distance of each preset number of pixels at two sides into a statistical band by taking the main component axis as a central axis; determining the ratio of the number of the pixel points with the gray value in each statistical band larger than the second gray threshold value to the number of all the pixel points in the whole statistical band, and generating a frequency value; forming a bar-shaped statistical graph by taking the number of the statistical bands as an abscissa and the frequency value as an ordinate; smoothing the histogram, calculating the ratio of the standard deviation of the height of each statistical band to the mean value of the height of the band after smoothing, and determining a second punishment weight according to the ratio; and determining the definition of the E-hypergraph according to the definition value, the first penalty weight and the second penalty weight.
Optionally, the evaluation parameter is the parallelism between the main stripe and other stripes in the E-hypergraph; accordingly, the determination module 1 includes:
a fifth determining submodule 16, configured to determine, in the E-hypergraph after image segmentation, a stripe with a second largest area as a secondary main stripe; performing principal component analysis on the secondary main stripes to determine the direction of the secondary main stripes; and acquiring an included angle between the direction of the main stripe and the direction of the secondary main stripe, and determining the parallelism between the main stripe and the secondary main stripe according to the included angle.
Optionally, the evaluation parameter is a main stripe length; accordingly, the determination module 1 includes:
a sixth determining submodule 17, configured to obtain a length value of the main stripe using the direction of the main stripe as the length direction of the main stripe; obtaining the length value of an image diagonal line of the E-hypergraph; and calculating the ratio of the length value of the main stripe to the length value of the diagonal line of the image of the E-hypergraph, and determining the evaluation parameter of the length of the main stripe.
Optionally, the determining module 1 includes:
the evaluation submodule 18 is used for scoring one or more items in the evaluation parameters according to a preset evaluation threshold value, and if the sum of the scores is lower than the preset score threshold value, determining that the quality of the E-hypergraph is unqualified; and if the sum of the scores is higher than or equal to a preset score threshold value, determining that the quality of the E-hypergraph is qualified.
This embodiment can be used to implement the embodiment shown in fig. 3, and the implementation principle thereof is similar and will not be described herein again.
Fig. 12 is a flowchart of a tissue viscoelasticity detection method according to an exemplary embodiment, and as shown in fig. 12, based on the E-hypergraph analysis method according to the foregoing embodiments, in a viscoelasticity detection scenario, poor quality may be removed by evaluating E-hypergraphs obtained through multiple triggers in one detection process, so that final viscoelasticity is calculated according to the remaining E-hypergraphs, and a result of determining the viscoelasticity of a tissue is more accurate. In particular, can be prepared by
the tissue viscoelasticity detection method is based on the principle that a certain correlation exists between the propagation state of shear waves in tissues and the tissue viscoelasticity, low-frequency pulses are excited by a probe to generate shear waves in the tissues, and then the motion states of the shear waves along the time and space directions are observed by exciting ultrasonic waves, so that the tissue viscoelasticity is detected. The focus of the present invention is not on the viscoelastic detection technique and will not be described in detail.
the evaluation of the quality of the E-hypergraph can be performed based on the method described in any one of the embodiments in fig. 1 to 9b, and details are not repeated here.
And step 1204, determining tissue viscoelasticity information according to the E-hypergraph with the evaluation result meeting the preset standard.
The preset standard of the E-hypergraph can be set by a person skilled in the art according to the tissue viscoelasticity determination standard, for example, the preset standard can be the preset score threshold described in the above embodiments in step 307 and step 308, and the E-hypergraph meeting the preset score threshold is a qualified E-hypergraph, so that the tissue viscoelasticity information is determined according to the qualified E-hypergraph.
Fig. 13 is a schematic structural diagram of a tissue viscoelasticity detection apparatus according to an exemplary embodiment, and as shown in fig. 13, based on the E-hypergraph analysis apparatus according to the foregoing embodiments, in a viscoelasticity detection scenario, poor quality may be removed by evaluating E-hypergraphs obtained through multiple triggers in one detection process, so that final viscoelasticity is calculated according to the remaining E-hypergraphs, and a result of determining the viscoelasticity of a tissue is more accurate. Specifically, the tissue viscoelasticity detection device includes:
a detection module 3 for performing tissue viscoelasticity detection;
the output module 4 is used for outputting one or more E hypergraphs;
a quality evaluation module 5, configured to evaluate the quality of the one or more E-hypergraphs;
the evaluation of the quality of the E-hypergraph can be performed based on the apparatus described in any one of the embodiments of fig. 10 to 11, and is not described herein again.
And the tissue viscoelasticity determining module 6 is used for determining tissue viscoelasticity information according to the E-hypergraph with the evaluation result meeting the preset standard.
This embodiment can be used to implement the embodiment shown in fig. 12, and the implementation principle is similar, and will not be described herein again.
Those of ordinary skill in the art will understand that: all or a portion of the steps of implementing the above-described method embodiments may be performed by hardware associated with program instructions. The program may be stored in a computer-readable storage medium. When executed, the program performs steps comprising the method embodiments described above; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (20)
1. An elastography image analysis method, comprising:
performing image analysis on the elastic imaging image to determine a main stripe, wherein the main stripe is the stripe with the largest area;
determining, based on the primary streak, an evaluation parameter for evaluating image quality of the elastographic image, the evaluation parameter including at least one of: main stripe symmetry, main stripe width standard difference, elasticity imaging image definition, parallelism between a main stripe and other stripes in an elasticity imaging image, main stripe length, size of each stripe in the elasticity imaging image, main stripe direction and position of the main stripe in the elasticity imaging image;
and determining the quality of the elastography image according to the evaluation parameters.
2. The method of claim 1, wherein the image analyzing the elastographic image to determine the dominant streak comprises:
according to a first gray threshold, carrying out image segmentation on the elastic imaging image based on a region growing method;
and determining the stripe with the largest area as the main stripe in the elastography image after image segmentation.
3. The method of claim 2, wherein prior to determining an evaluation parameter for evaluating image quality of the elastographic image based on the primary fringe, further comprising:
performing principal component analysis on the main stripe to determine the direction of the main stripe;
determining a coordinate axis of the main stripe according to the direction of the main stripe, and projecting each pixel on the main stripe onto the coordinate axis; the direction of the main stripe is a first main component axis, and the direction perpendicular to the main stripe is a second main component axis.
4. The method of claim 3, wherein the evaluation parameter is the primary stripe symmetry; correspondingly, the determining an evaluation parameter for evaluating the image quality of the elastography image based on the main stripe comprises:
determining a statistical boundary of the main stripe according to the outline of the main stripe;
and determining the symmetry of the main stripe relative to the symmetry axis within the statistical boundary range by taking the first main component axis as the symmetry axis.
5. The method of claim 1, wherein the evaluation parameter is the primary fringe width standard deviation; correspondingly, the determining an evaluation parameter for evaluating the image quality of the elastography image based on the main stripe comprises:
determining a statistical boundary of the main stripe according to the outline of the main stripe;
within the statistical boundary range, obtaining the width mean value of the main stripe and obtaining the tail end position of the main stripe;
the tail end position points to the direction of the head end position of the main stripe, and the width value of the main stripe is obtained;
and determining the position of which the width value is greater than or equal to the width mean value as a starting position, starting from the starting position to the head end of the main stripe of the statistical boundary, and calculating the width standard deviation of the main stripe.
6. The method of claim 4, wherein the evaluation parameter is the elastographic image sharpness; correspondingly, the determining an evaluation parameter for evaluating the image quality of the elastography image based on the main stripe comprises:
determining the proportion value of pixel points meeting the preset gray scale range in all pixel points in the elastic imaging image according to the preset gray scale range, and generating a definition value;
determining the central position of the main stripe, and if the area value of the main stripe is less than or equal to a preset area threshold and the distance between the central position of the main stripe and the central position of the elastography image is greater than a preset distance threshold, determining a first penalty weight of the elastography image;
determining the central position of the main stripe in the elastic imaging image, wherein the central position is perpendicular to the projection axis of the first main component shaft;
projecting each pixel point with the gray value larger than a second gray threshold value to the projection axis; dividing the distance of each preset number of pixels at two sides into a statistical band by taking the main component shaft as a central axis;
determining the ratio of the number of the pixel points with the gray value in each statistical band larger than the second gray threshold value to the number of all the pixel points in the whole statistical band, and generating a frequency value;
forming a column statistical chart by taking the number of the statistical bands as an abscissa and the frequency value as an ordinate;
smoothing the histogram, calculating the ratio of the standard deviation of the height of each statistical band to the mean value of the height of the band after smoothing, and determining a second penalty weight according to the ratio;
and determining the definition of the elastic imaging image according to the definition value, the first penalty weight and the second penalty weight.
7. The method of claim 1, wherein the evaluation parameter is a degree of parallelism between the primary fringe and other fringes in the elastographic image; correspondingly, the determining an evaluation parameter for evaluating the image quality of the elastography image based on the main stripe comprises:
determining stripes with the second largest area as secondary main stripes in an elastic imaging image after image segmentation;
performing principal component analysis on the secondary main stripe to determine the direction of the secondary main stripe;
and acquiring an included angle between the direction of the main stripe and the direction of the secondary main stripe, and determining the parallelism between the main stripe and the secondary main stripe according to the included angle.
8. The method of claim 1, wherein the evaluation parameter is the primary stripe length; correspondingly, the determining an evaluation parameter for evaluating the image quality of the elastography image based on the main stripe comprises:
taking the direction of the main stripe as the length direction of the main stripe to obtain the length value of the main stripe;
acquiring the length value of an image diagonal of the elastic imaging image;
and calculating the ratio of the length value of the main stripe to the image diagonal length value of the elastography image, and determining the evaluation parameter of the length of the main stripe.
9. The method according to any one of claims 1 to 8, wherein the determining the quality of the elastographic image according to the evaluation parameter comprises:
scoring one or more items in the evaluation parameters according to a preset evaluation threshold, and if the sum of the scores is lower than the preset score threshold, determining that the quality of the elastic imaging image is unqualified; and if the sum of the scores is higher than or equal to a preset score threshold value, determining that the quality of the elastography image is qualified.
10. An elastography image analysis device, comprising:
the determining module is used for carrying out image analysis on the elastic imaging image and determining a main stripe, wherein the main stripe is the stripe with the largest area;
an evaluation module, configured to determine an evaluation parameter for evaluating an image quality of the elastography image based on the main fringe, where the evaluation parameter includes at least one of the following parameters: main stripe symmetry, main stripe width standard difference, elasticity imaging image definition, parallelism between a main stripe and other stripes in an elasticity imaging image, main stripe length, size of each stripe in the elasticity imaging image, main stripe direction and position of the main stripe in the elasticity imaging image;
the determining module is further configured to determine the quality of the elastography image according to the evaluation parameter.
11. The apparatus of claim 10, wherein the determining module comprises:
the segmentation submodule is used for carrying out image segmentation on the elastic imaging image based on a region growing method according to a first gray threshold;
and the first determining submodule is used for determining the stripe with the largest area as the main stripe in the elasticity imaging image after image segmentation.
12. The apparatus of claim 11,
the determining module is further configured to perform principal component analysis on the main stripe to determine a direction of the main stripe; determining a coordinate axis of the main stripe according to the direction of the main stripe, and projecting each pixel on the main stripe onto the coordinate axis; the direction of the main stripe is a first main component axis, and the direction perpendicular to the main stripe is a second main component axis.
13. The apparatus of claim 12, wherein the evaluation parameter is the primary stripe symmetry; accordingly, the determining module comprises:
the second determining submodule is used for determining the statistical boundary of the main stripe according to the outline of the main stripe; and determining the symmetry of the main stripe relative to the symmetry axis within the statistical boundary range by taking the first main component axis as the symmetry axis.
14. The apparatus of claim 10, wherein the evaluation parameter is the primary fringe width standard deviation; accordingly, the determining module comprises:
the third determining submodule is used for determining the statistical boundary of the main stripe according to the outline of the main stripe; within the statistical boundary range, obtaining the width mean value of the main stripe and obtaining the tail end position of the main stripe; the tail end position points to the direction of the head end position of the main stripe, and the width value of the main stripe is obtained; and determining the position of which the width value is greater than or equal to the width mean value as a starting position, starting from the starting position to the head end of the main stripe of the statistical boundary, and calculating the width standard deviation of the main stripe.
15. The apparatus of claim 13, wherein the evaluation parameter is the elastographic image sharpness; accordingly, the determining module comprises:
the fourth determining submodule is used for determining the proportion value of pixel points meeting the preset gray scale range in all the pixel points in the elastic imaging image according to the preset gray scale range and generating a definition value; determining the central position of the main stripe, and if the area value of the main stripe is less than or equal to a preset area threshold and the distance between the central position of the main stripe and the central position of the elastography image is greater than a preset distance threshold, determining a first penalty weight of the elastography image; determining the central position of the main stripe in the elastic imaging image, wherein the central position is perpendicular to the projection axis of the first main component shaft; projecting each pixel point with the gray value larger than a second gray threshold value to the projection axis; dividing the distance of each preset number of pixels at two sides into a statistical band by taking the main component shaft as a central axis; determining the ratio of the number of the pixel points with the gray value in each statistical band larger than the second gray threshold value to the number of all the pixel points in the whole statistical band, and generating a frequency value; forming a column statistical chart by taking the number of the statistical bands as an abscissa and the frequency value as an ordinate; smoothing the histogram, calculating the ratio of the standard deviation of the height of each statistical band to the mean value of the height of the band after smoothing, and determining a second penalty weight according to the ratio; and determining the definition of the elastic imaging image according to the definition value, the first penalty weight and the second penalty weight.
16. The apparatus of claim 10, wherein the evaluation parameter is a degree of parallelism between the primary fringe and other fringes in the elastographic image; accordingly, the determining module comprises:
the fifth determining submodule is used for determining the stripe with the second largest area as a secondary main stripe in the elastic imaging image after image segmentation; performing principal component analysis on the secondary main stripe to determine the direction of the secondary main stripe; and acquiring an included angle between the direction of the main stripe and the direction of the secondary main stripe, and determining the parallelism between the main stripe and the secondary main stripe according to the included angle.
17. The apparatus of claim 10, wherein the evaluation parameter is the primary stripe length; accordingly, the determining module comprises:
a sixth determining submodule, configured to obtain a length value of the main stripe with the direction of the main stripe as the length direction of the main stripe; acquiring the length value of an image diagonal of the elastic imaging image; and calculating the ratio of the length value of the main stripe to the image diagonal length value of the elastography image, and determining the evaluation parameter of the length of the main stripe.
18. The apparatus of any one of claims 10-17, wherein the means for determining comprises:
the evaluation submodule is used for scoring one or more items in the evaluation parameters according to a preset evaluation threshold value, and if the sum of the scores is lower than the preset evaluation threshold value, the quality of the elastic imaging image is determined to be unqualified; and if the sum of the scores is higher than or equal to a preset score threshold value, determining that the quality of the elastography image is qualified.
19. A method for detecting tissue viscoelasticity, comprising:
performing tissue viscoelasticity detection, outputting one or more elastographic images;
evaluating the quality of the one or more elastographic images according to any of the methods of claims 1-9; and determining tissue viscoelasticity information according to the elastography image of which the evaluation result meets the preset standard.
20. A tissue viscoelasticity detecting apparatus, comprising:
a detection module for performing tissue viscoelasticity detection;
an output module for outputting one or more elastographic images;
a quality evaluation module for evaluating the quality of the one or more elastographic images according to any of the apparatus claims 10-18;
and the tissue viscoelasticity determining module is used for determining tissue viscoelasticity information according to the elastography image of which the evaluation result meets the preset standard.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910117697.3A CN109978823B (en) | 2019-02-15 | 2019-02-15 | Elastography image analysis and tissue viscoelasticity detection method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910117697.3A CN109978823B (en) | 2019-02-15 | 2019-02-15 | Elastography image analysis and tissue viscoelasticity detection method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109978823A CN109978823A (en) | 2019-07-05 |
| CN109978823B true CN109978823B (en) | 2021-08-06 |
Family
ID=67076967
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910117697.3A Active CN109978823B (en) | 2019-02-15 | 2019-02-15 | Elastography image analysis and tissue viscoelasticity detection method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109978823B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103845074A (en) * | 2012-11-28 | 2014-06-11 | 深圳迈瑞生物医疗电子股份有限公司 | Ultrasonic elastography system and method |
| CN105899140A (en) * | 2013-08-12 | 2016-08-24 | 三星电子株式会社 | Method for producing elastic image and ultrasonic diagnostic apparatus |
| CN107405133A (en) * | 2014-12-24 | 2017-11-28 | 超声成像 | Suitable for the shearing wave elastograph imaging method and device of anisotropic medium imaging |
| CN107666863A (en) * | 2015-05-14 | 2018-02-06 | 三星电子株式会社 | Show the method for elastogram image and perform the ultrasonic diagnostic equipment of this method |
| CN107833218A (en) * | 2017-11-22 | 2018-03-23 | 深圳中科乐普医疗技术有限公司 | A kind of biological tissue's shearing wave elastic image quality evaluation display methods |
| CN108733857A (en) * | 2017-04-21 | 2018-11-02 | 深圳迈瑞生物医疗电子股份有限公司 | Ultrasonic elastograph imaging device and elastogram evaluation of result method |
-
2019
- 2019-02-15 CN CN201910117697.3A patent/CN109978823B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103845074A (en) * | 2012-11-28 | 2014-06-11 | 深圳迈瑞生物医疗电子股份有限公司 | Ultrasonic elastography system and method |
| CN105899140A (en) * | 2013-08-12 | 2016-08-24 | 三星电子株式会社 | Method for producing elastic image and ultrasonic diagnostic apparatus |
| CN107405133A (en) * | 2014-12-24 | 2017-11-28 | 超声成像 | Suitable for the shearing wave elastograph imaging method and device of anisotropic medium imaging |
| CN107666863A (en) * | 2015-05-14 | 2018-02-06 | 三星电子株式会社 | Show the method for elastogram image and perform the ultrasonic diagnostic equipment of this method |
| CN108733857A (en) * | 2017-04-21 | 2018-11-02 | 深圳迈瑞生物医疗电子股份有限公司 | Ultrasonic elastograph imaging device and elastogram evaluation of result method |
| CN107833218A (en) * | 2017-11-22 | 2018-03-23 | 深圳中科乐普医疗技术有限公司 | A kind of biological tissue's shearing wave elastic image quality evaluation display methods |
Non-Patent Citations (3)
| Title |
|---|
| " Pattern classification of ShearWave Elastography images for differential diagnosis between benign and malignant solid breast masses";Tozaki M;《Acta Radiologica》;20110911;第1069-1075页 * |
| "First step for computer assisted evaluation of qualitative supersonic shear wave elastography characteristics in breast tissue";Skerl, K;《IEEE》;20161231;第481-484页 * |
| "乳腺弹性成像原理及质量控制";李俊来;《中国超声医学杂志》;20170731;第667-670页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109978823A (en) | 2019-07-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wilson et al. | A new metric for grey-scale image comparison | |
| Zhang et al. | A three-dimensional fractal analysis method for quantifying white matter structure in human brain | |
| US10405834B2 (en) | Surface modeling of a segmented echogenic structure for detection and measurement of anatomical anomalies | |
| US10572987B2 (en) | Determination of localised quality measurements from a volumetric image record | |
| Rahmatullah et al. | Quality control of fetal ultrasound images: Detection of abdomen anatomical landmarks using adaboost | |
| US9892507B2 (en) | Processing optical coherence tomography scans of a subjects skin | |
| Sioma | Assessment of wood surface defects based on 3D image analysis | |
| CN110322554B (en) | Brain dynamic function mode stability calculation method | |
| Lavoué et al. | Perceptual quality metrics for 3D meshes: towards an optimal multi-attribute computational model | |
| JP2012132754A (en) | Texture evaluation device, texture evaluation method | |
| US10229494B2 (en) | Automated analysis of intravascular OCT image volumes | |
| MacGillivray et al. | The resolution integral: visual and computational approaches to characterizing ultrasound images | |
| Khan et al. | Semiautomatic quantification of carotid plaque volume with three-dimensional ultrasound imaging | |
| CN116977331B (en) | 3D model surface detection method based on machine vision | |
| CN109978823B (en) | Elastography image analysis and tissue viscoelasticity detection method and device | |
| Wang et al. | Quantitative assessments of cup-to-disk ratios in spectral domain optical coherence tomography images for glaucoma diagnosis | |
| CN109965910B (en) | Sagittal plane projection imaging method based on three-dimensional ultrasonic spine body data | |
| EP2693397B1 (en) | Method and apparatus for noise reduction in an imaging system | |
| Jamaludin et al. | Predicting spine geometry and scoliosis from DXA scans | |
| KR20110111751A (en) | Quantitative evaluation of scratch-induced damages on polymeric and coating materials | |
| Aliahmad et al. | Adaptive Higuchi's dimension-based retinal vessel diameter measurement | |
| DE102006043348A1 (en) | System and method for grouping airways and arteries for quantitative analysis | |
| JP6810619B2 (en) | Skin condition evaluation method | |
| Attar et al. | Ebiqa: An edge based image quality assessment | |
| TW201535320A (en) | Aging analyzing method and aging analyzing device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |