KR101495489B1 - Medical image processor and medical image quality improving method for the same - Google Patents

Abstract

The present invention provides an apparatus for processing medical images which can minimize the amount of radiation to which a patient is exposed and can provide reliable high definition medical images, and a method for improving the image quality thereof. The apparatus for processing medical images according to an embodiment of the present invention comprises: a representative pattern determining unit for determining representative patterns which are criteria for image quality improvement using pixel values of a high definition medical image for reference; a representative pattern matching unit for performing pattern matching between the representative patterns to pixel values of a low definition medical image for image quality improvement; and an image quality improving unit for improving the image quality of the low definition medical image using the image quality improvement method for each representative pattern according to the pattern matching.

Description

[0001] MEDICAL IMAGE PROCESSOR AND MEDICAL IMAGE QUALITY IMPROVING METHOD FOR THE SAME [0002]

Embodiments of the present invention relate to a medical image processing apparatus and a method for improving the image quality of the medical image processing apparatus.

As electronic technology has evolved, a variety of medical imaging devices have been introduced for noninvasive diagnostic methods to improve diagnostic accuracy. For example, there are computerized tomography (CT) and magnetic resonance imaging (MRI) devices, simple radiography (X-ray), and tomosynthesis. These devices have enabled many diagnostic tests that were not possible in the past and have been a great help in making more accurate diagnoses.

In recent years, the role of diagnostic radiology has increased dramatically due to the remarkable development of diagnostic medical imaging devices. Thus, about 50-70% of patients seeking a hospital receive a noninvasive diagnosis, and the result is an appropriate treatment policy. In addition, medical imaging devices are used in the operating room, which has become possible with shorter imaging times. In the operating room, a moving medical imaging device is utilized, which is slower in imaging speed and lower in image quality than a diagnostic medical imaging device, but is increasingly utilized in the operating room due to the fact that the patient can take a picture without moving. As medical imaging devices are gradually evolving into miniaturization and portable forms, it causes degradation of resolution and image quality. Therefore, how to increase the reliability of medical images in a short time is becoming an important issue.

However, if the medical image to be displayed is difficult to read or is not suitable for diagnosis, it is difficult to greatly improve the image quality or take a long time to improve the image quality by the image processing apparatus according to the prior art. Therefore, it is urgently required to develop a medical diagnostic apparatus and method that remarkably improve the image quality of the displayed image information in a short time.

Related Prior Art Korean Unexamined Patent Application Publication No. 2003-0033561 (entitled: Video Signal Processing Apparatus and Method, Published on May 01, 2003) is available.

An embodiment of the present invention provides a medical image processing apparatus and a method for improving the quality of a medical image which can reliably provide a high-definition medical image while minimizing a radiation dose of a patient.

The problems to be solved by the present invention are not limited to the above-mentioned problem (s), and another problem (s) not mentioned can be clearly understood by those skilled in the art from the following description.

The medical image processing apparatus according to an embodiment of the present invention includes a representative pattern determining unit that determines a representative pattern that is a reference for image quality improvement using pixel values of a reference high definition medical image; A representative pattern matching unit for performing pattern matching between the representative pattern and pixel values of the low-quality medical image for improving the image quality; And an image quality enhancement unit for improving the image quality of the low-quality medical image using the image quality improvement method for each representative pattern according to the pattern matching.

Wherein the representative pattern determination unit constitutes a representative pattern table including a set of the representative patterns and a size of the representative patterns for one reference high definition medical image, and the representative pattern matching unit refers to the representative pattern table, Pattern matching with the low-quality medical image can be performed.

Wherein the representative pattern determination unit determines whether a flat region or an edge exists in a position for extracting the representative pattern from the reference high definition medical image and calculates a size of the representative pattern based on the determination result .

The representative pattern determining unit may calculate the size of the representative pattern in the reference high definition medical image using the degree of correlation.

Wherein the representative pattern table further includes at least one of an average and a standard deviation within a representative pattern necessary for the image quality improvement method for each representative pattern, and the image quality improvement unit calculates at least one of an average and a standard deviation, The quality of the low-quality medical image can be improved by determining the improvement method and using the image quality improvement method for each representative pattern.

Wherein the representative pattern determination unit determines whether the current representative pattern is similar to an existing representative pattern existing in the existing representative pattern table, and resets the current representative pattern to the representative pattern if the determination result is similar, The representative pattern table may be updated by adding the current representative pattern to the representative pattern table as a new representative pattern.

The representative pattern determination unit may measure a cosine similarity between the vector of the current representative pattern and the vector of the existing representative pattern and determine whether the current representative pattern and the existing representative pattern are similar based on the measured cosine similarity have.

The representative pattern determination unit may set an average of upper N (N is a natural number) patterns having a high degree of cosine similarity with the current representative pattern as the representative pattern.

The representative pattern determination unit may determine the representative pattern using the pixel value on the spatial domain of the reference high definition medical image or the representative pattern using the pixel value on the frequency domain of the reference high definition medical image.

Wherein when the representative pattern is determined using the pixel values on the frequency domain, the representative pattern matching unit calculates the frequency-converted transform coefficients of the representative patterns composed of the transform coefficients determined in the reference high-definition medical image and the pixel values of the low- And the image quality improving unit may improve the quality of the conversion coefficient of the low-quality medical image according to the representative pattern composed of the conversion coefficients according to the pattern matching, and invert the converted coefficients.

The representative pattern determining unit may select a reference high-definition medical image to be used for improving the image quality of the low-quality medical image among a plurality of reference high-quality medical images, based on at least one of the characteristics of the low-

The representative pattern determination unit may select any one of the medical image input by the user or the plurality of reference high definition medical images selected by the user as a reference high definition medical image to be used for improving the image quality of the low quality medical image .

The representative pattern matching unit determines a similarity degree between the representative pattern and the pattern extracted from the low-quality medical image, generates a pattern map including unique identification information of the representative pattern having the highest degree of similarity for each pattern extracted from the low- The image quality improving unit may calculate an image quality improvement coefficient for each representative pattern corresponding to the unique identification information using the pattern map and improve the image quality of the low quality medical image using the calculated image quality improvement coefficient.

Wherein the image quality improvement unit determines an average value of the representative pattern as a brightness correction coefficient and determines a ratio between a standard deviation of the representative pattern and a standard deviation of a pattern extracted from the low quality medical image as a noise correction coefficient, The image quality of the low-quality medical image can be improved by using the brightness correction coefficient and the noise correction coefficient as coefficients.

The medical image processing apparatus according to an embodiment of the present invention may include at least one of a contrast stretching and a low-pass filtering for the low-quality medical image based on a difference in radiation exposure between the reference high-quality medical image and the low- And a preprocessing unit for performing preprocessing by using the preprocessing unit.

A medical image processing apparatus according to another embodiment of the present invention includes a representative pattern database including a plurality of representative pattern tables; A representative pattern matching unit for selecting any one of the plurality of representative pattern tables and performing pattern matching between representative patterns of the selected representative pattern table and pixel values of a low-quality medical image for improving image quality; And an image quality enhancement unit for improving the image quality of the low-quality medical image using the image quality improvement method for each representative pattern according to the pattern matching.

Wherein the representative pattern database further includes a reference high definition medical image corresponding to each of the plurality of representative pattern tables, and the representative pattern matching unit uses the reference high definition medical image selected in accordance with the selection input operation among the reference high definition medical images So that any one of the plurality of representative pattern tables can be selected.

The method for improving the quality of a medical image processing apparatus according to an exemplary embodiment of the present invention includes the steps of: determining a representative pattern as a reference for image quality improvement using pixel values of a reference high definition medical image in the medical image processing apparatus; Performing pattern matching between the representative pattern and pixel values of the low-quality medical image for improving the image quality in the medical image processing apparatus; And improving the image quality of the low-quality medical image using the image quality improvement method for each representative pattern according to the pattern matching in the medical image processing apparatus.

In the medical image processing apparatus according to an embodiment of the present invention, in the medical image processing apparatus, a representative pattern table including a set of the representative patterns for the reference high- ≪ / RTI >

And performing the pattern matching may include performing pattern matching with the low-quality medical image for each representative pattern by referring to the representative pattern table.

In the medical image processing apparatus according to an embodiment of the present invention, in the medical image processing apparatus, the image quality of the low-quality medical image may be determined based on at least one of the characteristics of the low- And selecting a reference high definition medical image to be used for improving the image quality of the medical image.

Wherein the step of performing pattern matching comprises: determining a degree of similarity between the representative pattern and a pattern extracted from the low-quality medical image, and determining a pattern map including unique identification information of a representative pattern having a highest degree of similarity for each pattern extracted from the low- Wherein the improving the image quality comprises: calculating a picture quality improvement coefficient for each representative pattern corresponding to the unique identification information using the pattern map; And improving the image quality of the low-quality medical image using the calculated image quality enhancement coefficient.

Wherein the improving the image quality comprises: determining an average value of the representative pattern as a brightness correction coefficient; Determining a ratio between a standard deviation of the representative pattern and a standard deviation of a pattern extracted from the low-quality medical image as a noise correction coefficient; And improving the image quality of the low-quality medical image using the brightness correction coefficient and the noise correction coefficient as the image quality improvement coefficient.

According to another aspect of the present invention, there is provided a method for improving image quality of a medical image processing apparatus, comprising: recording a plurality of representative pattern tables in a representative pattern database in the medical image processing apparatus; Selecting one of the plurality of representative pattern tables in the medical image processing apparatus; Performing pattern matching between a representative pattern of the selected representative pattern table and pixel values of a low-quality medical image for improving image quality in the medical image processing apparatus; And improving the image quality of the low-quality medical image using the image quality improvement method for each representative pattern in the medical image processing apparatus according to the pattern matching.

The method for improving the quality of a medical image processing apparatus according to another embodiment of the present invention may further include recording a reference high definition medical image corresponding to each of the plurality of representative pattern tables in a representative pattern database, The selecting one of the tables may include selecting any one of the plurality of representative pattern tables using the reference high definition medical image selected in accordance with the selection input operation among the reference high definition medical images.

The details of other embodiments are included in the detailed description and the accompanying drawings.

According to an embodiment of the present invention, by improving the image quality of a low-quality medical image by using a high-definition medical image for reference, it is possible to provide a reliable high-quality medical image in a short time while minimizing the amount of patient exposure.

According to an embodiment of the present invention, the image quality improving performance can be improved by patterning the low-quality medical image in units of pixel blocks and then performing the image quality improving process in a different manner for each pattern.

1 is a block diagram illustrating a medical image processing apparatus according to an embodiment of the present invention.
2 is a block diagram illustrating a medical image processing apparatus according to another embodiment of the present invention.
3 is a block diagram illustrating a medical image processing apparatus according to another embodiment of the present invention.
4 is a flowchart illustrating an image quality improvement method of a medical image processing apparatus according to an exemplary embodiment of the present invention.
5 is a flowchart illustrating a process of determining representative patterns according to an embodiment of the present invention.
Fig. 6 is a representative pattern table constituting representative patterns with values in the spatial domain.
FIG. 7 is a representative pattern table in which representative patterns are formed by values in the frequency domain using a fast Fourier transform.
8 is a flowchart illustrating a process of matching representative patterns according to an embodiment of the present invention.
9 is a representative pattern table according to an embodiment of the present invention.
FIG. 10 is a flowchart illustrating a method of improving image quality of a medical image processing apparatus according to another embodiment of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS The advantages and / or features of the present invention, and how to accomplish them, will become apparent with reference to the embodiments described in detail below with reference to the accompanying drawings. It should be understood, however, that the invention is not limited to the disclosed embodiments, but is capable of many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, To fully disclose the scope of the invention to those skilled in the art, and the invention is only defined by the scope of the claims. Like reference numerals refer to like elements throughout the specification.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram illustrating a medical image processing apparatus according to an embodiment of the present invention.

1, a medical image processing apparatus 100 according to an exemplary embodiment of the present invention includes a representative pattern determining unit 110, a representative pattern matching unit 120, an image quality improving unit 130, and a controller 140 ).

The representative pattern determination unit 110 determines a representative pattern serving as a reference for image quality improvement using the pixel values of the reference high definition medical image. To this end, the representative pattern determination unit 110 can preferentially select a reference high-definition medical image for determining the representative pattern from a plurality of reference high-definition medical images.

That is, the representative pattern determining unit 110 determines, based on the characteristics of the low-quality medical image for improving the image quality, the photographing environment, and the like, the reference pattern determining unit 110 determines, among the plurality of reference high- Can be selected.

Alternatively, the representative pattern determination unit 110 may select any one of the medical image input by the user or the plurality of reference high definition medical images selected by the user as a reference high image quality to be used for improving the image quality of the low quality medical image Medical images can be selected.

The representative pattern determination unit 110 may configure a representative pattern table including a set of representative patterns determined for one selected high-definition medical image for reference and a size for each representative pattern. Here, the representative pattern table may further include an average and standard deviation within a representative pattern necessary for the image quality improvement method for each representative pattern, in addition to the representative pattern set and the representative pattern size. The average and standard deviation may be used as basic data for determining the image quality improvement method for each representative pattern in the image quality improving unit 130. [

The representative pattern determination unit 110 determines whether a flat region or an edge exists in a position for extracting the representative pattern from the reference high definition medical image, The size of each representative pattern can be calculated.

That is, when the representative pattern determining unit 110 determines that the representative pattern is to be extracted in the reference high-definition medical image, the representative pattern determining unit 110 sets the size of the representative pattern to be large and calculates its size And the size of the representative pattern can be set small and the size thereof can be calculated in the case of an area where an edge exists (edge area).

In this manner, setting the size of the representative pattern to a large size in the case of the flat area is intended to process the flat area as a background area because the flat area includes a part that is generally important. On the other hand, setting the size of the representative pattern to a small size in the case of the edge region includes a portion (an inspection object such as a liver) which is generally important in the edge region, so that the edge region can be precisely processed into the main region to be.

The representative pattern determination unit 110 may use the degree of correlation to calculate the size of the representative pattern in the reference high definition medical image. For example, the representative pattern determining unit 110 may calculate the size of the representative pattern in the reference high definition medical image using the Pearson correlation coefficient.

Regarding the configuration (update) of the representative pattern table, the representative pattern determination unit 110 can determine whether the current representative pattern is similar to the existing representative pattern existing in the existing representative pattern table.

At this time, the representative pattern determination unit 110 measures the cosine similarity between the vector of the current representative pattern and the vector of the existing representative pattern, and calculates the similarity between the current representative pattern and the existing representative pattern based on the measured cosine similarity. It is possible to judge whether or not it is possible.

If the determination result is similar, the representative pattern determination unit 110 may configure the representative pattern table by resetting the current representative pattern to the representative pattern. Alternatively, the representative pattern determination unit 110 may set an average of upper N (natural number) patterns having a high degree of cosine similarity with the current representative pattern as the representative pattern.

On the other hand, if the current representative pattern and the existing representative pattern are not similar as a result of the determination, the representative pattern determination unit 110 adds the current representative pattern to the representative pattern table as a new representative pattern, Can be updated.

The representative pattern determination unit 110 may determine the representative pattern using the pixel value on the spatial domain of the reference high definition medical image. Alternatively, the representative pattern determination unit 110 may determine the representative pattern using the pixel value on the frequency domain of the reference high definition medical image, The representative pattern may be determined. Here, the pixel values in the frequency domain may be transformed by a Fourier transform, a discrete cosine transform, a discrete sinusoidal transform, a fast Fourier transform, a Hadamard transform, (Hereinafter referred to as " conversion coefficient ") that is converted into a coefficient on the frequency by the conversion method.

The representative pattern matching unit 120 performs pattern matching between the representative pattern and pixel values of the low-quality medical image. When the representative pattern table 120 is configured by the representative pattern determination unit 120, the representative pattern matching unit 120 may perform pattern matching with the low-quality medical image for each representative pattern with reference to the representative pattern table have.

At this time, the representative pattern matching unit 120 may perform pattern matching between the spatial domain representative pattern of the reference high-definition medical image and the spatial domain representative pattern of the low-quality medical image. Alternatively, the representative pattern matching unit 120 may perform pattern matching between a frequency domain representative pattern of the reference high definition medical image and a frequency domain representative pattern of the low quality medical image.

The representative pattern matching unit 120 can determine the similarity between the representative pattern and the representative pattern extracted from the low-quality medical image through the pattern matching process as described above. Based on the result of the similarity determination, Can be generated. Here, the pattern map is generated to perform image quality improvement for each representative pattern, and includes unique identification information of a representative pattern having a highest degree of similarity (representative pattern of the reference medical image) for each representative pattern extracted from the low-quality medical image can do.

The image quality improving unit 130 improves the image quality of the low-quality medical image using the representative pattern-based image quality improving method according to the pattern matching of the representative pattern matching unit 120. [ That is, the image quality improving unit 130 may improve the image quality of the low-quality medical image by varying the image quality improving method for each representative pattern according to the representative pattern on the representative pattern table. Alternatively, the image quality improving unit 130 may improve the image quality of the low-quality medical image by varying the image quality improving method for each representative pattern according to the size of the representative pattern on the representative pattern table.

That is, when there is no representative pattern table, the image quality improvement unit 130 determines a picture quality improvement method according to the representative pattern size determined by the representative pattern determination unit 110, and uses the determined picture quality improvement method Thereby improving the image quality of the low-quality medical image.

For example, if the size of the representative pattern is small as in the flat area, the image quality improving unit 130 may improve the image quality of the low-quality medical image by determining a low-pass filter as the image quality improving method. If the size of the representative pattern is large as in the edge area, the image quality improving unit 130 determines the image quality improvement coefficient including the brightness correction coefficient and the noise correction coefficient as the image quality improvement method, Can be improved.

If the edge is unclear in the edge region, the image quality improving unit 130 may apply a high-pass filter to the representative pattern to sharpen the edge. If the edge is sharp, The quality of the low-quality medical image can be improved.

On the other hand, if there is the representative pattern table, the image quality improvement unit 130 may determine the image quality improvement method for each representative pattern based on the average and standard deviation in the representative pattern included in the representative pattern table. The image quality improving unit 130 may improve the image quality of the low-quality medical image using the image quality improving method for each representative pattern.

At this time, the image quality improvement unit 130 calculates a quality enhancement coefficient for each representative pattern corresponding to the unique identification information using the pattern map, and uses the calculated quality enhancement coefficient to determine the quality of the low- Can be improved.

Here, the image quality improvement coefficient may include a brightness correction coefficient and a noise correction coefficient, the brightness correction coefficient may be determined as an average value of the representative pattern, and the noise correction coefficient may be a standard deviation of the representative pattern, And the standard deviation of the pattern extracted from the medical image.

In other words, the image quality improving unit 130 determines the average value of the representative pattern as the brightness correction coefficient, and adjusts the ratio between the standard deviation of the representative pattern and the standard deviation of the pattern extracted from the low- Can be determined by a coefficient. The image quality improving unit 130 may improve the image quality of the low-quality medical image using the brightness correction coefficient and the noise correction coefficient as the image quality improvement coefficient.

In the meantime, when the frequency domain representative pattern is used, the image quality improvement unit 130 may improve the quality of the conversion coefficient (the value of the frequency domain representative pattern) of the low-quality medical image for each representative pattern composed of the transform coefficients, By performing inverse conversion to the pixel value, it is possible to improve the image quality for the low-quality medical image in the frequency domain.

The control unit 140 may include the representative pattern determining unit 110, the representative pattern matching unit 120, the image quality improving unit 130, and the like in the medical image processing apparatus 100 according to an exemplary embodiment of the present invention. Can be controlled in general.

2 is a block diagram illustrating a medical image processing apparatus according to another embodiment of the present invention.

2, a medical image processing apparatus 200 according to another exemplary embodiment of the present invention includes a preprocessing unit 210, a representative pattern determination unit 220, a representative pattern matching unit 230, an image quality improvement unit 240, and a control unit 250. [0028]

The representative pattern determining unit 220, the representative pattern matching unit 230, and the image quality improving unit 240 may be arranged in the same manner as in the first exemplary embodiment of the present invention. And performs the same or similar functions as those of the medical image processing apparatus 100 according to an embodiment of the present invention. Therefore, only the pre-processing unit 210 will be described in detail in another embodiment of the present invention.

The preprocessing unit 210 may perform a contrast stretching process and a low pass filtering process on the low-quality medical image based on the difference in radiation exposure between the reference high-quality medical image and the low- Pre-processing can be performed using at least one method.

That is, when selecting the reference high-definition medical image, when the medical image having a larger amount of radiation exposure than the low-quality medical image for improving the image quality is selected, the preprocessing unit 210 determines that the difference in the radiation exposure between the two medical images is larger than the threshold If it is large, preprocessing can be performed.

3 is a block diagram illustrating a medical image processing apparatus according to another embodiment of the present invention.

3, a medical image processing apparatus 300 according to another embodiment of the present invention includes a representative pattern DB 310, a representative pattern matching unit 320, an image quality improving unit 330, and a controller 340 ).

According to another embodiment of the present invention, the remaining components except for the representative pattern DB 310 and the representative pattern matching unit 320 related thereto, that is, the image quality improving unit 330, And performs the same or similar functions as the components of the medical image processing apparatus 100.

That is, another embodiment of the present invention provides a medical image processing apparatus 300 that improves the image quality of a low-quality medical image using the representative pattern DB 310 without a function of determining a representative pattern. Therefore, in another embodiment of the present invention, only the representative pattern DB 310 and the representative pattern matching unit 320 associated therewith will be described in detail.

The representative pattern DB 310 may include a plurality of representative pattern tables and a reference high-definition medical image corresponding to each of the plurality of representative pattern tables. When the radiation exposure amount of the reference high-definition medical image is larger than that of the low-quality medical image, the representative pattern DB 310 is automatically updated in the server, and the server updates the representative pattern DB 310 The representative pattern DB 310 can be configured by performing a pattern determination function.

Alternatively, when the user updates the representative pattern DB 310 and there is a medical image to be used as a medical image for reference, the medical image is manually updated to the server, and the server performs a representative pattern determination function The representative pattern DB 310 can be configured.

In the case of the automatic update of the representative pattern DB 310, the representative pattern DB 310 may have only the representative pattern table. However, in the case of the manual update, It is preferable that the representative pattern DB 310 includes the reference medical image and the representative pattern table of the reference medical image together.

The representative pattern matching unit 320 may select any one of a plurality of representative pattern tables included in the representative pattern DB 310. At this time, the representative pattern matching unit 320 can select any one of the plurality of representative pattern tables using the reference high definition medical image selected according to the selection input operation (manual) among the reference high definition medical images, Alternatively, the representative pattern table can be selected in various ways such as selecting the most suitable representative pattern table at random.

The representative pattern matching unit 320 may perform pattern matching between representative patterns of the selected representative pattern table and pixel values of a low-quality medical image for image quality improvement. The pattern matching function of the representative pattern matching unit 320 is the same as or similar to that of the representative pattern matching unit 120 shown in FIG. 1, so that a description thereof will be omitted in the present embodiment.

4 is a flowchart illustrating an image quality improvement method of a medical image processing apparatus according to an exemplary embodiment of the present invention.

Referring to FIG. 4, in step 410, the medical image processing apparatus determines a representative pattern that is a reference for image quality improvement using pixel values of a reference high definition medical image.

Before determining the representative pattern, the medical image processing apparatus can select a reference high-definition medical image for determining the representative pattern from a plurality of reference high-definition medical images.

In addition, when the medical image processing apparatus selects the medical image for reference, when the medical image having a larger radiation dose is selected than the low-quality medical image for improving the image quality, when the difference in the amount of radiation exposure between the two medical images is larger than the threshold Can be pre-processed.

At this time, as the preprocessing method, methods such as contrast stretching and low pass filtering may be used.

Next, in step 420, the medical image processing apparatus performs pattern matching between the representative pattern and the pixel values of the low-quality medical image for improving the image quality. At this time, the medical image processing apparatus may perform the pattern matching using the representative pattern table.

Through the pattern matching process as described above, the medical image processing apparatus can discriminate the similarity between the representative pattern and the representative pattern extracted from the low-quality medical image, and generate the pattern map based on the result of the similarity determination have. Here, the pattern map may be used to perform image quality enhancement for each representative pattern.

Next, in step 430, the medical image processing apparatus improves the image quality of the low-quality medical image using the image quality improvement method for each representative pattern according to the pattern matching. At this time, the medical image processing apparatus can improve the image quality of the low-quality medical image by using the pattern map, using the image quality improvement method corresponding to the unique identification number (number) of the representative pattern.

5 is a flowchart illustrating a process of determining representative patterns according to an embodiment of the present invention.

Referring to FIG. 5, in step 510, the medical image processing apparatus selects a reference medical image, and determines an improvement direction of a medical image to be improved at present as a reference medical image.

At this time, the reference medical image is relatively high quality as compared with the medical image to be improved at present, which may mean a difference in the amount of radiation exposure and may be a medical image designated by a user (doctor or diagnosis person) .

For example, when the radiation dose of the medical image A is 60 mSv and the radiation dose of the medical image to be improved is 80 mSv, if the user wants to improve the medical image A, the reference medical image may be selected as the medical image A have.

When the reference medical image is selected, the representative patterns are determined in the reference medical image, and the determined representative patterns are stored in the representative pattern table. Here, the representative pattern table may be configured to divide the reference medical image into a plurality of representative patterns and to have a representative pattern of each pattern, and to later change the image quality improving method for each representative pattern.

In addition, the essential elements constituting the representative pattern table are representative patterns and representative pattern sizes, and other components other than the representative patterns may be added. For example, the average, standard deviation, and the like in the representative pattern required in the picture quality improvement method for each representative pattern may be used.

At this time, the representative pattern may be determined as a value in the spatial domain, or may be determined as a value in the frequency domain. The representative pattern on the spatial domain may be composed of a table as shown in FIG. 6, and the representative pattern on the frequency domain may be composed of a table as shown in FIG. 6 is a representative pattern table constituting a representative pattern with a value in the spatial domain, and FIG. 7 is a representative pattern table constituting a representative pattern with values in the frequency domain using a fast Fourier transform.

Since the representative pattern can have various sizes, a representative pattern size is included as an essential element. For example, all the representative patterns of the representative pattern table may be the same size, or they may have different sizes for each representative pattern. At this time, if the representative pattern is a value in the frequency domain as shown in FIG. 7, the size of each representative pattern is ^2N .

If each representative pattern has a different size, a step of specifying the size of each representative pattern is required. For example, the representative pattern may have a size ranging from a maximum of 10 to a minimum of 5, and a representative pattern table may be formed for each size.

Alternatively, when the pattern is extracted from the reference medical image, it is possible to automatically calculate the size of the pattern after determining whether the position to extract the current pattern is a flat region or an area where an edge exists. At this time, various edge detectors such as Harris Corner Detection may be used as a method for determining whether an edge exists.

The Harris corner detection is divided into an edge area, a flat area, and a corner area by analyzing the distribution of values of an arbitrary area by a method using a Sobel operator.

Alternatively, the size of the current pattern in the reference medical image may be calculated using correlation. For example, when the Pearson correlation coefficient is used with reference to the following equations (1), (2) and (3), a similar pattern is determined only when the Auto Correlation Coefficient is 0.9 or more using Pearson correlation coefficient .

The following equation (1) is an equation of Pearson correlation coefficient, the following equation (2) is a covariance matrix for calculating a Pearson correlation coefficient, and the following equation (3) is a mathematical formula for calculating an autocorrelation coefficient. Therefore, by substituting Equation (3) into Equations (1) and (2), an autocorrelation coefficient using the Pearson correlation coefficient can be calculated.

[Equation 1]

&Quot; (2) "

&Quot; (3) "

In step 520, the medical image processing apparatus determines whether the pattern extracted from the reference medical image exists in the representative pattern table. That is, the medical image processing apparatus determines whether the pattern extracted from the reference medical image is similar to a pattern existing in the existing representative pattern table.

If it is similar to the pattern existing in the existing representative pattern table (in the "YES" direction of 520), the medical image processing apparatus resets the representative pattern in step 540. On the other hand, in the case of a pattern that is not similar to the representative patterns of the existing representative pattern table ("No" direction of 520), the medical image processing apparatus adds a new representative pattern to the representative pattern table Update the pattern table.

Here, a criterion is required in order to determine whether or not it is similar to the representative patterns existing in the existing representative pattern table. The criterion may be different according to the person providing the corresponding method. In an embodiment of the present invention, .

The cosine similarity is mainly used to measure the similarity of two vectors. The cosine value between two vectors is used and the value range is between -1 and 1. The closer to -1, If it is 0, both vectors are independent. The closer to 1, the more vector is the same vector.

&Quot; (4) "

In Equation (4), q and d denote vectors. In the embodiment of the present invention, q is a conventional representative pattern, and d is a pattern extracted from a reference medical image. The degree of similarity with the representative patterns existing in the existing representative pattern table can be determined according to the cosine similarity calculated using Equation 1. Since the calculated similarity has a value between -1 and 1 as described above , The user or the provider can set a threshold value of the degree of similarity. For example, the calculated similarity may be determined to be similar when the similarity is 0.7 or more, or may be determined to be similar when the similarity is 0.9 or more.

In addition, the representative pattern setting in step 540 is to set a more representative pattern because similar patterns are seen in the same pattern. Therefore, there are various methods for the representative pattern setting method. For example, a pattern added to the first representative pattern table may be set as a representative pattern. In this case, the representative pattern setting is set only when a new pattern is added to the representative pattern table.

Alternatively, in step 540, the medical image processing apparatus may set an average of patterns having a high degree of cosine similarity with a current representative pattern as a representative pattern. In this case, the medical image processing apparatus can calculate the average every time the representative pattern is set by adding the sum and the number of patterns similar to the representative patterns to the representative pattern table, The average between patterns may be calculated at that time and set as a representative pattern.

8 is a flowchart illustrating a process of matching representative patterns according to an embodiment of the present invention.

Referring to FIG. 8, when a medical image to be improved is input (step 810), a representative pattern (reference medical image) corresponding to the representative pattern number 'N = 0' of the representative pattern table, (Step 820), and generates (updates) a pattern map of the medical image to be improved (steps 820 to 850).

In the similarity degree determination step 840, if the determination result is that the degree of similarity is 0.9 or less (the "no" direction in 840), the representative pattern number is increased by 1 (step 860) The process of determining the degree of similarity with the pattern extracted from the medical image to be processed can be repeated up to the maximum value of N. [

This pattern map is generated in order to execute the image quality improving step for each representative pattern. In addition, the information included in the pattern map is the number (N) of the most representative pattern for each medical image pattern to be improved.

Although various methods can be used as a method of determining the degree of similarity with the representative pattern N, in the embodiment of the present invention, the cosine similarity of Equation (1) is used. The similarity determination method used in the representative pattern determination step and the similarity determination method used in the representative pattern matching step may be the same or different.

Alternatively, when a cosine similarity is used, a cosine similarity threshold value is required. The cosine similarity threshold value of the representative pattern determining step and the cosine similarity threshold value of the representative pattern matching step may have different values. For example, when the degree of similarity is determined using the cosine similarity in the representative pattern determination step, the representative pattern table is updated by identifying the representative pattern with the most similar representative pattern (in this case, the pattern of the most similar representative pattern and the reference medical image The cosine similarity degree between the representative pattern of the representative pattern table and the pattern of the medical image to be improved is 0.9 (Step 850), it is possible to perform pattern map update (step 850).

In step 810, when inputting a medical image to be improved, a method of inputting the original of the medical image and a preprocessed medical image may be input. When preprocessing is required, there is a possibility that the representative pattern matching may not be performed properly because the image quality difference between the reference medical image and the medical image to be improved is severe.

This can be determined by an arbitrary method. When selecting a reference medical image, when a medical image having a larger radiation dose than the medical image to be improved is selected, a preprocessing is performed when a difference in radiation dose between two medical images is large. Alternatively, if the user directly designates the reference medical image, the user can directly designate whether to perform the preprocessing. Alternatively, the result of performing representative pattern matching using the preprocessed medical image and the result of performing representative pattern matching using the original medical image can be provided to the user.

Here, as the pre-processing method, any method such as contrast-enhanced stretching or low-pass filtering may be used. In the embodiment of the present invention, a contrast-based stretching method using a minimum value and a maximum value will be described.

&Quot; (5) "

In Equation (5), Max _HighDose and Min _HighDose may be replaced by arbitrary values. For example, when 16-bit data is used, Max _HighDose can be fixed to 32767 and Min _HighDose to 2048. [

In addition, the pre-treatment furnace provides various methods other than one method, and a plurality of methods may be used. For example, low-pass filtering may be performed after performing a contrast-based stretching method using Equation (5) on a medical image to be improved. At this time, if the representative pattern is composed of values in the frequency domain, a representative pattern is calculated by using values after the preprocessing by using the medical image values and converting them into values in the frequency domain.

Hereinafter, the image quality improving step for each representative pattern will be described. In the image quality improving step, the image quality is improved using the image quality improving method for each representative pattern specified in the representative pattern table.

&Quot; (6) "

According to Equation (6), the improved medical image enh (x, y) is added by the brightness correction coefficient m _target , thereby increasing the pixel value of the medical image to be improved, Corrected. The brightness correction coefficient _mtarget may be calculated as an average of the representative pattern, and may be designated as an arbitrary value. The noise correction coefficient C may also be calculated using the following equation (7), or an arbitrary value may be specified.

&Quot; (7) "

The mathematical std-dev _low of formula 7 _{(x, y)} is the standard deviation in the pattern determined in a medical image to be currently _{improved, std-dev high (x,} y) is the standard deviation in the representative pattern.

9 is a representative pattern table according to an embodiment of the present invention. The representative pattern table of FIG. 9 has N, a representative pattern, a size, and an image quality improvement method as constituent elements. The representative pattern size is fixed to 6 and has different image quality improvement methods for each representative pattern N. FIG.

Alternatively, the brightness correction factor of Equation 5 m _target may be an average position by the representative pattern of equation (3) for calculating the noise correction factors C std-dev _{low (x, y)} and the std-dev _{high (x, y)} may be the standard deviation of the representative pattern by position.

In addition, when medical image quality is improved through the image quality improvement step, artifacts may occur depending on the size of the pattern. These artifacts may give a sense of rejection depending on the user, and thus may provide post processing. The post-processing may be performed using any method such as contrast-enhanced stretching or low-pass filter using Equation (2).

For example, a low-pass filter may be performed at the boundary of the pattern, or a low-pass filter may be performed if the difference between the left and right values of the pattern exceeds a certain threshold. Alternatively, a low pass filter can be performed if the difference between the left and right values of the pattern boundary exceeds the threshold and is not an actual edge.

In the case of the pattern extracted in the frequency domain, the enhancement of the image quality is performed using the transform coefficient, and then the inverse transform is performed to calculate the improved low-quality medical image pixel value.

In the representative pattern matching step, the representative pattern may have a 2D pattern or a 1D pattern. If the pattern has a 1D pattern, the representative pattern matching may be performed in the horizontal or vertical direction, in the unidirectional direction, or in both the horizontal and vertical directions.

For example, when the horizontal representative pattern matching is performed first, the vertical representative pattern matching and the image quality improvement step are performed using the images up to the horizontal representative pattern matching and the image quality improvement step. Alternatively, even when the vertical representative pattern matching is performed first, the horizontal representative pattern matching and the image quality improving step are performed using the images that have been subjected to the vertical representative pattern matching and the image quality improving step.

When bi-directional representative pattern matching using the 1D pattern is performed, it is also possible to perform post-processing after performing horizontal or vertical representative pattern matching. Here, the post-processing may use a contrast-enhanced stretching or a low-pass filter using Equation (5).

FIG. 10 is a flowchart illustrating a method of improving image quality of a medical image processing apparatus according to another embodiment of the present invention.

Referring to FIG. 10, in step 1010, the medical image processing apparatus records a plurality of representative pattern tables in the representative pattern DB.

Next, in step 1020, the medical image processing apparatus selects any one of the plurality of representative pattern tables.

Next, in step 1030, the medical image processing apparatus performs pattern matching between the representative pattern of the selected representative pattern table and the pixel values of the low-quality medical image for image quality improvement.

Next, in step 1040, the medical image processing apparatus improves the image quality of the low-quality medical image using the image quality improvement method for each representative pattern.

Hereinafter, a method for improving the image quality of the medical image processing apparatus according to another embodiment of the present invention will be described in more detail.

In an embodiment of the present invention, a reference medical image for the representative pattern determination step is required. If the representative pattern determination process using the reference medical image is omitted, the time for improving the medical image quality can be reduced.

For this, in another embodiment of the present invention, the image quality of the medical image is improved by using the representative pattern DB (database) without the representative pattern determination step. The representative pattern DB is generated in advance, and the representative pattern DB constituting the representative pattern DB can be a representative pattern table and a reference medical image.

The representative pattern DB is configured such that when a radiation dose is larger than a predetermined amount, the representative pattern is automatically updated to the server, and a representative pattern DB is formed by performing a representative pattern determination step whenever it is updated, or the user updates the representative pattern DB, If there is a medical image to be used as a representative pattern DB, the representative pattern DB can be configured by manually updating and performing a representative pattern determination step at the time of updating.

In the former method, although the representative pattern DB may have only the representative pattern table, the latter method is a case where the user selects the medical image for use as the reference medical image, so the reference medical image and the representative pattern table of the reference medical image It is desirable to have. In addition, the former method can also constitute a representative pattern table for each reference medical image together with the reference medical image, and can provide the reference medical image as a sample to the user.

In this case, the user can select a desired representative pattern table using the reference medical image sample among the plurality of representative pattern tables existing in the representative pattern DB. Therefore, when improving the image quality of the medical image, the time required for the representative pattern determination step can be shortened.

Embodiments of the present invention include computer readable media including program instructions for performing various computer implemented operations. The computer-readable medium may include program instructions, local data files, local data structures, etc., alone or in combination. The media may be those specially designed and constructed for the present invention or may be those known to those skilled in the computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape, optical recording media such as CD-ROMs and DVDs, magneto-optical media such as floppy disks, and ROMs, And hardware devices specifically configured to store and execute the same program instructions. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like.

While the present invention has been described in connection with what is presently considered to be practical exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be determined by the scope of the appended claims and equivalents thereof.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, but, on the contrary, Modification is possible. Accordingly, the spirit of the present invention should be understood only in accordance with the following claims, and all equivalents or equivalent variations thereof are included in the scope of the present invention.

110, 220: Representative pattern determination unit
120, 230, and 320: a representative pattern matching unit
130, 240, and 330:
140, 250, 340:
210:
310: Representative pattern DB

Claims (24)

A representative pattern determination unit for determining a representative pattern as a reference for image quality improvement by using pixel values of a reference high definition medical image;
A representative pattern matching unit for performing pattern matching between the representative pattern and pixel values of the low-quality medical image for improving the image quality; And
An image quality enhancement unit for improving the image quality of the low-quality medical image using the image quality improvement method for each representative pattern according to the pattern matching,
Lt; / RTI >
The representative pattern determination unit
A representative pattern table including a set of the representative patterns and sizes of the representative patterns for one reference high-definition medical image,
The representative pattern matching unit
And performs pattern matching with the low-quality medical image for each representative pattern with reference to the representative pattern table.
delete The method according to claim 1,
The representative pattern determination unit
Characterized in that it is judged whether or not a flat region or an edge exists in a position for extracting the representative pattern from the reference high definition medical image and the size of the representative pattern is calculated on the basis of the discrimination result Image processing apparatus.
The method of claim 3,
The representative pattern determination unit
And calculates the size of the representative pattern in the reference high definition medical image using the degree of correlation.
The method according to claim 1,
The representative pattern table
Further comprising at least one of an average and a standard deviation within a representative pattern necessary for the image quality improvement method for each representative pattern
The image quality improving unit
Wherein the image quality improvement method for each representative pattern is determined based on at least one of the average and standard deviation, and the image quality of the low-quality medical image is improved using the image quality improvement method for each representative pattern.
The method according to claim 1,
The representative pattern determination unit
Determining whether the current representative pattern is similar to an existing representative pattern existing in the existing representative pattern table; and if the determination result is similar to the current representative pattern, re-setting the current representative pattern to the representative pattern to construct the representative pattern table, Wherein the representative pattern table is updated by adding the current representative pattern as a new representative pattern to the representative pattern table if the similarity pattern table does not resemble the representative pattern table.
The method according to claim 6,
The representative pattern determination unit
Wherein the cosine similarity between the vector of the current representative pattern and the vector of the existing representative pattern is measured and whether or not the current representative pattern is similar to the existing representative pattern is determined based on the measured cosine similarity. Processing device.
8. The method of claim 7,
The representative pattern determination unit
Wherein an average of the upper N (N is a natural number) patterns having a higher degree of cosine similarity with the current representative pattern is set as the representative pattern among the existing representative patterns.
delete A representative pattern determination unit for determining a representative pattern as a reference for image quality improvement by using pixel values of a reference high definition medical image;
A representative pattern matching unit for performing pattern matching between the representative pattern and pixel values of the low-quality medical image for improving the image quality; And
An image quality enhancement unit for improving the image quality of the low-quality medical image using the image quality improvement method for each representative pattern according to the pattern matching,
Lt; / RTI >
The representative pattern determination unit
Determining the representative pattern using the pixel value on the spatial domain of the reference high definition medical image or determining the representative pattern using the pixel value on the frequency domain of the reference high definition medical image,
When the representative pattern is determined using the pixel value on the frequency domain,
The representative pattern matching unit
Performing pattern matching between representative patterns composed of the transform coefficients determined in the reference high-definition medical image and frequency-converted transform coefficients of pixel values of the low-quality medical image,
The image quality improving unit
Wherein the conversion coefficient of the low-quality medical image is improved in accordance with the pattern matching and the inverse conversion is performed on the conversion coefficient of the low-quality medical image for each representative pattern composed of the conversion coefficients.
A representative pattern determination unit for determining a representative pattern as a reference for image quality improvement by using pixel values of a reference high definition medical image;
A representative pattern matching unit for performing pattern matching between the representative pattern and pixel values of the low-quality medical image for improving the image quality; And
An image quality enhancement unit for improving the image quality of the low-quality medical image using the image quality improvement method for each representative pattern according to the pattern matching,
Lt; / RTI >
The representative pattern determination unit
Quality medical image to be used for improving the image quality of the low-quality medical image among the plurality of reference high-quality medical images, based on at least one of the characteristics of the low-quality medical image and the photographing environment, .
The method according to claim 1,
The representative pattern determination unit
Quality medical image to be used for improving the image quality of the low-quality medical image, the medical image selected by the user from the medical image input by the user or the plurality of reference high- Processing device.
A representative pattern determination unit for determining a representative pattern as a reference for image quality improvement by using pixel values of a reference high definition medical image;
A representative pattern matching unit for performing pattern matching between the representative pattern and pixel values of the low-quality medical image for improving the image quality; And
An image quality enhancement unit for improving the image quality of the low-quality medical image using the image quality improvement method for each representative pattern according to the pattern matching,
Lt; / RTI >
The representative pattern matching unit
Generating a pattern map including unique identification information of a representative pattern having a highest degree of similarity for each pattern extracted from the low-quality medical image, determining a similarity between the representative pattern and the pattern extracted from the low-
The image quality improving unit
The image quality improvement coefficient is calculated for each representative pattern corresponding to the unique identification information using the pattern map, and the image quality of the low-quality medical image is improved using the calculated image quality improvement coefficient.
14. The method of claim 13,
The image quality improving unit
Determining a ratio between a standard deviation of the representative pattern and a standard deviation of a pattern extracted from the low-quality medical image as a noise correction coefficient, and determining the brightness correction coefficient as the image quality enhancement coefficient, Wherein the quality of the low-quality medical image is improved by using the coefficient and the noise correction coefficient.
A representative pattern determination unit for determining a representative pattern as a reference for image quality improvement by using pixel values of a reference high definition medical image;
A representative pattern matching unit for performing pattern matching between the representative pattern and pixel values of the low-quality medical image for improving the image quality;
An image quality enhancement unit for improving the image quality of the low-quality medical image using the image quality improvement method for each representative pattern according to the pattern matching; And
A preprocessing unit that preprocesses the low-quality medical image using at least one of a contrast stretching and a low-pass filtering based on a difference in radiation exposure between the reference high-quality medical image and the low-
The medical image processing apparatus comprising:
A representative pattern database including a plurality of representative pattern tables;
A representative pattern matching unit for selecting any one of the plurality of representative pattern tables and performing pattern matching between representative patterns of the selected representative pattern table and pixel values of a low-quality medical image for improving image quality; And
An image quality enhancement unit for improving the image quality of the low-quality medical image using the image quality improvement method for each representative pattern according to the pattern matching,
Lt; / RTI >
The representative pattern database
Further comprising a reference high-definition medical image corresponding to each of the plurality of representative pattern tables,
The representative pattern matching unit
And selects one of the plurality of representative pattern tables using the reference high definition medical image selected according to the selection input operation among the reference high definition medical images.
delete A method for improving image quality of a medical image processing apparatus,
Selecting a reference high-definition medical image to be used for improving the image quality of the low-quality medical image among a plurality of reference high-definition medical images, based on at least one of the characteristics of the low-quality medical image and the photographing environment in the medical image processing apparatus;
Determining, in the medical image processing apparatus, a representative pattern serving as a reference for image quality improvement using pixel values of a reference high definition medical image;
Performing pattern matching between the representative pattern and pixel values of the low-quality medical image for improving the image quality in the medical image processing apparatus; And
In the medical image processing apparatus, improving the image quality of the low-quality medical image using the image quality improving method for each representative pattern according to the pattern matching
Wherein the image quality of the medical image processing apparatus is improved.
19. The method of claim 18,
The medical image processing apparatus comprising a representative pattern table including a set of the representative patterns for the reference high definition medical image and a size of the representative patterns,
Further comprising:
The step of performing the pattern matching
Performing pattern matching with the low-quality medical image for each representative pattern by referring to the representative pattern table
Wherein the image quality of the medical image processing apparatus is improved.
delete A method for improving image quality of a medical image processing apparatus,
Determining, in the medical image processing apparatus, a representative pattern serving as a reference for image quality improvement using pixel values of a reference high definition medical image;
Performing pattern matching between the representative pattern and pixel values of the low-quality medical image for improving the image quality in the medical image processing apparatus; And
In the medical image processing apparatus, improving the image quality of the low-quality medical image using the image quality improving method for each representative pattern according to the pattern matching
Lt; / RTI >
The step of performing the pattern matching
Determining a degree of similarity between the representative pattern and the pattern extracted from the low-quality medical image, and generating a pattern map including unique identification information of the representative pattern having the highest degree of similarity for each pattern extracted from the low-
Lt; / RTI >
The step of improving the image quality
Calculating an image quality improvement coefficient for each representative pattern corresponding to the unique identification information using the pattern map; And
And improving the image quality of the low-quality medical image using the calculated image quality enhancement coefficient
Wherein the image quality of the medical image processing apparatus is improved.
22. The method of claim 21,
The step of improving the image quality
Determining an average value of the representative pattern as a brightness correction coefficient;
Determining a ratio between a standard deviation of the representative pattern and a standard deviation of a pattern extracted from the low-quality medical image as a noise correction coefficient; And
And improving the image quality of the low-quality medical image using the brightness correction coefficient and the noise correction coefficient as the image quality improvement coefficient
Wherein the image quality of the medical image processing apparatus is improved.
A method for improving image quality of a medical image processing apparatus,
In the medical image processing apparatus, a plurality of representative pattern tables are recorded in a representative pattern database;
Recording a reference high-definition medical image corresponding to each of the plurality of representative pattern tables in a representative pattern database;
Selecting one of the plurality of representative pattern tables in the medical image processing apparatus;
Performing pattern matching between a representative pattern of the selected representative pattern table and pixel values of a low-quality medical image for improving image quality in the medical image processing apparatus; And
In the medical image processing apparatus, improving the image quality of the low-quality medical image using the image quality improving method for each representative pattern according to the pattern matching
Lt; / RTI >
Wherein the step of selecting any one of the plurality of representative pattern tables
Selecting one of the plurality of representative pattern tables using the reference high definition medical image selected in accordance with the selection input operation among the reference high definition medical images
Wherein the image quality of the medical image processing apparatus is improved. delete

Images