CN107680078B

CN107680078B - Image processing method and device

Info

Publication number: CN107680078B
Application number: CN201710781177.3A
Authority: CN
Inventors: 张志伟
Original assignee: Neusoft Medical Systems Co Ltd
Current assignee: Neusoft Medical Systems Co Ltd
Priority date: 2017-09-01
Filing date: 2017-09-01
Publication date: 2020-10-13
Anticipated expiration: 2037-09-01
Also published as: CN107680078A

Abstract

The present disclosure provides an image processing method and apparatus, wherein the method includes: acquiring image data of a mixed energy image of energy spectrum Computed Tomography (CT); carrying out tissue segmentation in the mixed energy image according to the image data to obtain different tissues; for each tissue, adopting a base substance corresponding to the tissue to decompose to obtain single-energy tissue images with different energy levels corresponding to the tissue; and for each energy level, combining the single-energy tissue images of different tissues corresponding to the energy level to obtain a single-energy image corresponding to the energy level. The method and the device have the advantages that the obtained single-energy image is displayed more clearly in the aspect of substance distinguishing, and the quality of the single-energy image is improved.

Description

Image processing method and device

Technical Field

The present disclosure relates to medical imaging technologies, and in particular, to an image processing method and apparatus.

Background

In conventional CT (Computed Tomography), X-rays generated by a bulb have a continuous energy distribution, and in Multi-energy spectrum CT (Multi-energy/spectral CT) imaging, different absorption of substances at different X-ray energies is utilized to provide more image information than in conventional CT. For example, spectral CT imaging can decompose the mixed energy of conventional X-rays into multiple single energy images at different keV levels between 40-140keV, which can show CT value curves of different tissues and organs and lesions at different X-ray energy levels, thereby providing more abundant information for lesion origin and benign-malignant identification, etc.

In order to acquire the single-energy image, after the energy spectrum CT scans the subject to obtain scan data, the scan data may be analyzed by the energy image processing software of the post-processing workstation. However, the quality of the single-energy image obtained by energy decomposition by the current energy image processing software is still to be improved.

Disclosure of Invention

In view of the above, the present disclosure provides an image processing method and apparatus to improve the discrimination of different materials by a single-energy image.

Specifically, the present disclosure is realized by the following technical solutions:

in a first aspect, an image processing method is provided, the method comprising:

acquiring image data of a mixed energy image of energy spectrum Computed Tomography (CT);

carrying out tissue segmentation in the mixed energy image according to the image data to obtain different tissues;

for each tissue, adopting a base substance corresponding to the tissue to decompose to obtain single-energy tissue images with different energy levels corresponding to the tissue;

and for each energy level, combining the single-energy tissue images of different tissues corresponding to the energy level to obtain a single-energy image corresponding to the energy level.

In a second aspect, there is provided an image processing apparatus, the apparatus comprising: a memory, a processor, and computer instructions stored on the memory and executable on the processor, the processor when executing the instructions implementing the steps of:

In a third aspect, there is provided a computer-readable storage medium having instructions stored thereon, which when executed by one or more processors, cause the one or more processors to perform a method of image processing, the method comprising:

According to the image processing method and device provided by the disclosure, different tissues in the image have different material composition compositions when the image is decomposed, and different base materials are adopted for decomposing different tissues, so that the obtained single-energy image is displayed more clearly in the aspect of material distinguishing, and the quality of the single-energy image is improved.

Drawings

FIG. 1 is a flow chart illustrating a method of image processing according to an exemplary embodiment of the present disclosure;

FIG. 2 is a flow chart illustrating a process for displaying compositional differences according to an exemplary embodiment of the present disclosure;

fig. 3 is a flowchart illustrating an obtaining process of pixel point differential data according to an exemplary embodiment of the disclosure;

FIG. 4 is a flow chart illustrating a method of generating a difference curve in accordance with an exemplary embodiment of the present disclosure;

FIG. 5 is a schematic diagram illustrating an energy difference curve according to an exemplary embodiment of the present disclosure;

FIG. 6 is a flow chart illustrating a substance identification criteria definition method according to an exemplary embodiment of the present disclosure;

fig. 7 is a schematic diagram of a recognition coordinate system according to an exemplary embodiment of the disclosure.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.

One of the imaging techniques of the energy spectrum CT can be double-bulb double-energy imaging, and the technique can be that two sets of bulbs and detectors are embedded in a CT frame, the two bulbs are arranged at a certain angle, the two bulbs simultaneously generate X rays during imaging, one bulb generates X rays with high kVp, and the other bulb generates X rays with low kVp. The two systems respectively and independently acquire data information, and are matched in an image space to perform dual-energy subtraction analysis. For example, the lowest voltage (80kVp) and the highest voltage (140kVp) can be used to achieve maximum energy separation to maximally distinguish different species.

In one example, the CT images obtained when scanned at 80kVp and 140kVp, respectively, are also each mixed energy images, and are obtained with X-ray scans containing multiple energies. The mixed energy image at 80kVp and the mixed energy image at 140kVp can be integrated to obtain single energy images with different energy levels between 80kVp and 140 kVp. Fig. 1 illustrates an image processing method that can process a mixed energy image of spectral CT to obtain a series of single energy images for disease diagnosis by a doctor.

In step 101, image data of a mixed energy image of a spectral CT is acquired.

In this step, the energy spectrum CT scans the object to be examined to obtain scan data, and performs image reconstruction according to the scan data to obtain a mixed energy image. The image data of the mixed energy image may be input to energy image processing software, for example, the mixed energy image may be loaded into the energy image processing software to decompose the mixed energy image to produce single energy images of different energy levels.

In step 102, tissue segmentation in the mixed energy image is performed according to the image data to obtain different tissues.

In the mixed energy image obtained by scanning the subject, different tissues (Tissue) may be included, for example, the Tissue may be a blood vessel, a bone, or the like.

In this step, the mixed energy image may be subjected to tissue segmentation, for example, a mixed energy image corresponding to 80kVp may be segmented into different tissues such as blood vessels and bones, and a mixed energy image corresponding to 140kVp may be segmented into different tissues. For example, Tissue such as Tissue1, Tissue2, Tissue3, etc.

In step 103, for each tissue, a base material corresponding to the tissue is used for decomposition, and single-energy tissue images with different energy levels corresponding to the tissue are obtained.

In this example, different base substances may be used to decompose different tissues according to their composition characteristics, and the selected base substance may be more favorable for accurately and clearly distinguishing tissues. The above-mentioned base material is a material used for separating materials from a mixed energy image, and according to the principle of energy spectrum imaging, any tissue can be expressed by generating the same attenuation effect through the combination of two base materials in the energy spectrum imaging, namely, an X-ray attenuation image subjected to high and low two sets of voltage scanning can be expressed as a density map of the two base materials.

For example, table 1 below illustrates one base material selection approach:

TABLE 1 organization and corresponding base material pairs

Tissue of	Base material pair
		Tissue 1: blood vessel	Water iodine
Tissue 2: skeleton(s)	Calcium iodine
		Others	Water iodine

According to the base materials in table 1, the mixed energy image can be decomposed to obtain single energy tissue images of different energy levels corresponding to the tissue. For example, a series of single-energy images of different energy levels corresponding to blood vessel tissue also correspond to single-energy images of different energy levels of bone, and in this example, the single-energy images of tissue may be referred to as single-energy tissue images.

In step 104, for each energy level, the single-energy tissue images of different tissues corresponding to the energy level are combined to obtain a single-energy image corresponding to the energy level.

In this step, the single-energy tissue images of the respective tissues may be recombined to obtain an overall single-energy image. For example, the single-energy tissue images of different tissues at the same energy level may be combined to obtain a single-energy image corresponding to the energy level; according to the method, a plurality of single energy images with different energy levels can be obtained.

In the image processing method of the example, when the image is decomposed, different tissues in the image have different material composition compositions, and different base materials are adopted for the different tissues to be decomposed, so that the obtained single-energy image is displayed more clearly in the aspect of material distinction, and the quality of the single-energy image is improved.

In another example, even the same tissue may have different components that may be of interest to a physician, such as tumor components, cystic components, etc. in a tissue. The image processing method of the present example can also display the above-mentioned difference components in the tissue significantly, which is more helpful to assist the doctor to find the lesion quickly and conveniently. Fig. 2 illustrates a processing flow of displaying composition differences, which may be performed after decomposing to obtain individual single-energy images:

in step 201, for each pixel point in the target organization, differential data of the pixel point is obtained.

In this example, the tissue to be observed by the doctor may be referred to as a target tissue, and the target tissue may be one of a plurality of tissues divided in the flow shown in fig. 1, or may be a part of a tissue region instead of the entire tissue.

For the target tissue, a series of images corresponding to the target tissue is included, for example, the target tissue is included in the mixed energy image, and the target tissue is also included in a series of single energy images obtained by energy decomposition. Different material compositions may have different absorption characteristics at different energy levels, and therefore the image representation of the target tissue may be different in the mixed energy image and the plurality of single energy images described above. For example, the HU (hounsfield unit, a unit of measure for measuring the density of a local tissue or organ of a human body) values of the pixels in the target tissue are relatively close to each other in the mixed energy image, but as the X-ray energy changes, the HU values of the pixels in the tissue in the single energy image at different energy levels may be different if the target tissue contains different material components at different energy levels. The method of the present example will identify differences in tissue composition based on such differences in image display.

The differentiation data in this step may be data of a pixel point in the target tissue, and the differentiation data may be used to represent a change in the mixed energy image and each single energy image of a difference between the HU value of the pixel point and the tissue average HU value. For example, the change may be that the HU value of the pixel point is not much different from the HU values of other organized pixel points in the mixed energy image, but in each single energy image with different energy levels, the HU value of the pixel point may be greatly different from the HU values of other organized pixel points, that is, the display of the pixel point in the single energy image and the mixed energy image is different.

In an example, fig. 3 illustrates an obtaining process of differentiated data of a pixel, which may include the following processing, but in actual implementation, the execution sequence of each step is not limited:

in step 2011, an average HU value of each pixel of the tissue in the mixed energy image is obtained.

For example, the average HU value of a tissue is expressed as AvgHu, and is calculated according to the following formula:

CtHu represents the Hu value of the pixel points in the mixed energy image, and N represents the number of the pixel points in the whole organization.

In step 2012, the difference value between the HU value of the pixel in the mixed energy image and the average HU value is used as the original difference value.

For example, the original disparity value may be calculated as follows, where PixHu represents the Hu value of the pixel currently being calculated, and the absolute value of the result calculated by the following formula may be obtained as the original disparity value:

PixHu-AvgHu..........(2)

in step 2013, for the single energy image of each energy level, respectively acquiring an energy level HU mean value of each pixel point of the tissue in the single energy image, and a difference value between an energy level HU value of the pixel point in the single energy image and the energy level HU mean value, as an energy level difference value.

For example, taking a single energy image of a certain energy level as an example, the average value of the energy level HU of the target tissue in the single energy image may be obtained first, and may be obtained by accumulating and averaging HU values of each pixel point in the target tissue, which is expressed by KevCtHuAvg. Secondly, the PixKevHu represents the Hu value of the current pixel point in the single energy image of the energy level, and the Hu value can be called as the Hu value of the energy level.

Calculating according to the following formula, and solving the absolute value of the calculation result to obtain the energy level difference value of the pixel point under the energy level:

KevCtHuAvg-PixKevHu..........(3)

in step 2014, an average value of the energy level differences is obtained according to the average value of the energy level differences of the pixel points at each energy level.

For example, the energy level difference values calculated by the pixel point under each energy level can be averaged:

as in the above formula, N represents the number of energy levels, and AvgKevDiff represents the average value of the energy level difference.

In step 2015, difference data of the pixel points is obtained according to the average value of the energy level differences of the pixel points and the original difference value.

For example,

wherein, V_DiffAnd the difference data corresponding to the pixel point is represented, and R is an adjusting coefficient, so that the condition that the effectiveness of a calculation result is lost due to too small difference between the average value of the energy level difference and the original difference value can be avoided, and the calculation sensitivity is improved.

In step 202, a difference value image is generated according to the difference data of each pixel point in the target tissue.

For example, the calculated V for each pixel point of the target organization_DiffNormalization processing may be performed, and a difference value image may be generated from the normalized difference data and displayed as a result of the calculation. The disparity value image can also be displayed in a fusion manner together with the target tissue.

In an example of practical application, a user may select an indication option for generating a difference value image from the energy image processing software, and trigger the energy image processing software to generate a difference value image of the selected target tissue according to the single energy image and the mixed energy image after performing energy decomposition to obtain the single energy image, so as to clearly observe different components in the target tissue through the difference value image.

The image processing method of the present example calculates the difference value image of the target tissue by combining the mixed energy image and each single energy image, so that different components in a specific tissue region are obviously displayed in a differentiated manner, which is helpful for a doctor to distinguish lesion components more quickly and accurately.

In yet another example, the image processing method of the present disclosure also provides a way to compare between different tissue regions. For example, one region may be given as a standard, the other region may be compared with the standard, the difference between the values of the images Hu of the two regions at different energy levels is observed, and the difference may also be displayed graphically by a difference curve.

For example, taking the above-mentioned tissue having different components as an example, a normal tissue region in the tissue may be used as a standard tissue, and the standard tissue may be used as a reference; differential tissue may also be determined, which may be a region of tissue that produces a difference (e.g., the region has a greater difference in Hu from the average Hu of the tissue) determined from the difference value image. Fig. 4 illustrates a generation method of a difference curve, which may include:

in step 401, average HU values of the pixel points in the standard organization at different energy levels are obtained as reference HU values.

For example, the reference HU may be AVG_KevThe value can be obtained by accumulating and averaging the HU values of the respective pixels in the standard organization.

In step 402, the difference between the average HU value of each pixel point in the differential organization at different energy levels and the reference HU value is taken as the ordinate.

For example, the HU values of the PIXELs in the differential organization at different energy levels may be accumulated and averaged to obtain an average HU value, which may be obtained by using PIXEL_KevAnd (4) showing. Also, PIXEL can be used_Kev-AVG_KevAs a coordinate of the vertical axis when PIXEL_Kev>AVG_KevWhen the difference is positive, when PIXEL_Kev<AVG_KevWhen, the difference is negative.

In step 403, the different energy levels are taken as abscissa.

In this example, the abscissa may be different energy levels, with different energy levels having different kev.

In step 404, an energy difference curve is drawn according to the horizontal axis coordinate and the vertical axis coordinate, and the energy difference curve is used for representing the change of the vertical axis coordinate at different energy levels.

For example, fig. 5 illustrates an energy difference curve that may illustrate the difference in HU values between a certain difference tissue and a selected standard tissue as a function of energy level. The variation curves of two different tissues compared to the standard tissue are illustrated in fig. 5.

The image processing method of the present example shows the difference comparison value between the difference tissue and the standard tissue in the form of an energy difference curve, so that the change of the difference comparison value with different energy levels can be clearly seen, which is helpful for more conveniently observing the comparison situation between different regions. In practical implementation, a user can select a standard tissue and can also determine a differential tissue to be observed, and after the generation of the energy difference curve is triggered, the change of the difference value between the standard tissue and the differential tissue selected by the user can be automatically displayed in the form of the energy difference curve through the method.

In yet another example, examples of the present disclosure also provide a method for customizing substance identification criteria. Fig. 6 illustrates the method, which may include:

in step 601, substance identification criteria are stored, the substance identification criteria including: the identification method comprises the steps of identifying a parameter coordinate system, obtaining a substance area range according to the identification parameter coordinate system and identifying substances corresponding to the substance area range.

This example may support user-defined criteria for substance identification, which may be defined as follows:

for example, as illustrated in FIG. 7, the user may select the X-axis criteria and the Y-axis criteria that identify the parameter coordinate system, and as an example, the X-axis criteria may be 60kev and the Y-axis criteria may be 80 kev.

In the coordinate system, a scatter diagram of the image in the coordinate system may be displayed based on the image in which the substance partition has been determined. For example, for one of the pixels in the image, the HU value of the pixel in the 60kev single-energy image, that is, the X coordinate in the coordinate system, is obtained, the HU value of the pixel in the 80kev single-energy image, that is, the Y coordinate in the coordinate system, is obtained, and the X coordinate and the Y coordinate are combined to determine the corresponding position point of the pixel in the coordinate system of fig. 7.

For example, if a certain material region in the image is already determined, the user can select the known material region, and the energy image processing software instructs the region to be converted into the scatter diagram in the set coordinate system shown in fig. 7, the corresponding scatter diagram of the material region in fig. 7 can be obtained, and each position point in the scatter diagram obtained by converting the same material is usually located in the same region, which can be referred to as a cluster region of the position points.

The cluster region may be referred to as a substance region range corresponding to a substance, which may be referred to as an identified substance. When other images are identified, if the images are subjected to scatter diagram conversion, the pixel points located in the material area range can be identified as the identified materials, and therefore material identification is achieved. Of course, if the above-mentioned coordinate standards of 60kev and 80kev are not good enough in clustering the scatter plot, for example, if the converted position points are relatively scattered, the user may switch to another coordinate system standard, which is not limited in this example as long as the scatter plot obtained according to the coordinate system standard can achieve a good scatter clustering effect of the same substance. Further, the clustering region may have various shapes such as a circle and a straight line.

After a good clustering effect is achieved and a good used coordinate system is determined, the identification parameter coordinate system can be saved in the step, and the identification parameter coordinate system, the substance area range and the corresponding identification substance can be called as substance identification standards according to the substance area range (i.e. the scattered point clustering range of the same substance) obtained by the coordinate system and the identification substance corresponding to the range (e.g. storing what substance the certain area range corresponds to).

In step 602, when a hybrid energy image to be identified is acquired, a corresponding scatter diagram is displayed in the identification parameter coordinate system for each pixel point in the hybrid energy image.

For example, the coordinate system determined in step 601 may be used to convert the pixel points in the image into a corresponding scatter plot. For example, taking the X-axis standard as 60kev and the Y-axis standard as 80kev as examples, the energy image processing software may decompose the mixed energy image into individual single energy images, and obtain the 60kev single energy image and the 80kev single energy image according to the definition of the coordinate system. For each pixel point in the image, the HU value of the pixel point in the single-energy image of 60kev is used as an x coordinate, and the HU value of the pixel point in the single-energy image of 80kev is used as a y coordinate, so that a corresponding scatter diagram is obtained.

In step 603, each pixel point clustered in the material region range in the scatter diagram is obtained.

For example, in step 601, the correspondence between the substance region range formed by clustering in the scatter diagram and the corresponding identified substance has been stored, for example, all the pixels falling within a certain region range are identified as a corresponding certain substance. Therefore, through the scatter diagram conversion in step 602, the scatters falling within the predetermined material region range, that is, the corresponding pixel points in the image, can be obtained.

In step 604, the pixels of the cluster region are identified as the substance.

As described above, each pixel point clustered in a certain substance region can be identified as a corresponding substance, for example, an image pixel point corresponding to a scatter clustered in a predetermined region represents a substance corresponding to the region.

The image processing method of the example provides a mode capable of supporting a user to define the substance identification standard, so that the substance identification means is more flexible and diversified.

The functions of the image processing method of the present disclosure, if implemented in the form of software functional units and sold or used as a separate product, may be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present disclosure or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing an image processing apparatus to execute all or part of the steps of the method according to the embodiments of the present disclosure. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

The present example provides a computer-readable storage medium having instructions stored thereon, which when executed by one or more processors, cause the one or more processors to perform a method of data processing, the method comprising: .

The present disclosure also provides an image processing apparatus, which may be, for example, a computer. The image processing device may run thereon energy image processing software, which may be computer instructions stored on a memory and executable on a processor, the processor of the image processing device may execute the instructions to implement the image processing method of the present disclosure:

In one example, the processor, when executing the instructions, is further configured to: after the single energy image corresponding to the energy level is obtained, generating a difference value image corresponding to a target tissue in different tissues;

the generating a disparity value image corresponding to a target tissue in the different tissues includes:

for each pixel point in the target tissue, acquiring differentiation data of the pixel point, wherein the differentiation data is used for representing the change of the difference between the HU value of the pixel point and the tissue average HU value in a mixed energy image and each single energy image;

and generating the difference value image according to the difference data of each pixel point in the target organization.

In one example, the processor, when executing the instructions for obtaining differentiated data of the pixel points, includes:

acquiring an average HU value of each pixel point of the tissue in the mixed energy image;

taking a difference value between the HU value of the pixel point in the mixed energy image and the average HU value as an original difference value;

respectively acquiring the energy level HU mean value of each pixel point of the tissue in the single energy image and the difference value between the energy level HU value of the pixel point in the single energy image and the energy level HU mean value as energy level difference values for the single energy image of each energy level;

obtaining an average value of energy level difference according to the average value of the energy level difference of the pixel points at each energy level;

and obtaining differentiation data of the pixel points according to the average energy level difference value and the original differentiation value of the pixel points.

In one example, the processor, when executing the instructions, is further configured to:

determining a standard organization as a benchmark and a difference organization for comparison with the standard organization;

acquiring average HU values of all pixel points in the standard organization under different energy levels, and taking the average HU values as reference HU values;

taking the difference between the average HU value of each pixel point in the differential organization under different energy levels and the reference HU value as a longitudinal axis coordinate;

and drawing an energy difference curve according to the vertical axis coordinate at different energy levels, wherein the energy difference curve is used for representing the change of the vertical axis coordinate at different energy levels, and the horizontal axis coordinate of the energy difference curve is different energy levels.

storing substance identification criteria, the substance identification criteria comprising: identifying a parameter coordinate system, obtaining a substance area range according to the identification parameter coordinate system and identifying substances corresponding to the substance area range;

when a mixed energy image to be identified is obtained, displaying a corresponding scatter diagram in the identification parameter coordinate system for each pixel point in the mixed energy image;

and identifying each pixel point clustered in the substance area range in the scatter diagram as the substance.

The above description is only exemplary of the present disclosure and should not be taken as limiting the disclosure, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present disclosure should be included in the scope of the present disclosure.

Claims

1. An image processing method, characterized in that the method comprises:

for each energy level, combining the single-energy tissue images of different tissues corresponding to the energy level to obtain a single-energy image corresponding to the energy level;

after the obtaining of the single energy image corresponding to the energy level, the method further comprises: generating a difference value image corresponding to a target tissue in the different tissues;

2. The method of claim 1, wherein the obtaining the differentiated data of the pixel points comprises:

3. The method of claim 1, further comprising:

4. The method of claim 1, further comprising:

5. An image processing apparatus, characterized in that the apparatus comprises: a memory, a processor, and computer instructions stored on the memory and executable on the processor, the processor when executing the instructions implementing the steps of:

the processor, when executing the instructions, is further configured to: after the single energy image corresponding to the energy level is obtained, generating a difference value image corresponding to a target tissue in different tissues;

6. The apparatus of claim 5, wherein the processor when executing the instructions for obtaining the differentiated data of the pixel points comprises:

7. The apparatus of claim 5, wherein the processor, when executing the instructions, is further configured to:

8. The apparatus of claim 5, wherein the processor, when executing the instructions, is further configured to:

9. A computer-readable storage medium having instructions stored thereon, which when executed by one or more processors, cause the one or more processors to perform the image processing method of any one of claims 1 to 4.