JP2014104107A - Image processing apparatus and image processing program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing apparatus and an image processing program capable of converting real number data with a wide dynamic range, such as a reconstructed image created by, e.g., a nuclear medicine diagnostic apparatus, into integer data without deteriorating the data accuracy of the region of interest.SOLUTION: An image processing apparatus 1 for converting real number data into integer data comprises: a control unit 18 for causing the real number data to be displayed at an arbitrary display level; a display condition setting unit 12 for setting, on the basis of the display level, a window limit range as a conversion range of a pixel value in the conversion processing; an integer data creation condition calculation unit 14 for calculating a rescale coefficient in the conversion processing on the basis of the maximum value and the minimum value of the window limit range; and an integer data creation unit 16 for converting the medical image real number data on the basis of the rescale coefficient to create the medical image integer data.

Description

  Embodiments described herein relate generally to an image processing apparatus and a program for processing medical image data.

  As a nuclear medicine diagnostic apparatus, for example, a SPECT (Single Photon Emission Computed Tomography) apparatus is known. The SPECT apparatus is equipped with a gamma camera, detects a gamma ray emitted from a radioisotope (RI) administered to a living body by the gamma camera, and acquires a distribution image of RI in the subject. It is configured.

  That is, the SPECT apparatus captures projection data by photographing a subject injected with RI with a gamma camera, and filters the projection data with a convolution filter having a predetermined filtering characteristic, and then performs back projection calculation processing (back Reconstructed image data representing the RI density distribution in the subject is created by performing addition processing in each case. Note that a reconstruction method based on successive approximation may be used.

  In such a nuclear medicine diagnostic apparatus, in order to maintain the accuracy of the calculation processing of the collected projection data, the calculation processing is performed with real number data. Each device can also create and store reconstructed image data in its own format.

  However, in recent years, due to the popularization of PACS (Picture Archiving and Communication System) and the unified management of patient data and examination data, it is not a format specific to each data processing device, but a standardized data format such as the DICOM standard, for example. Preservation at is desired.

  Here, according to the DICOM standard, image data is stored as integer data. That is, when data is stored according to the DICOM standard, image data cannot be stored as real number data in order to maintain data accuracy. Therefore, in order to save data according to the DICOM standard, it is necessary to convert real number data (reconstructed image data) output as a result of arithmetic processing into 2-byte signed integer data or unsigned integer data.

  As a technique related to data conversion, for example, Patent Document 1 discloses a technique for converting a value representing pixel intensity from an input dynamic range to an output dynamic range.

JP 2001-155143 A

  By the way, since the nuclear medicine diagnosis apparatus images the RI distribution, there is no fixed range in which the pixel value can be taken in the projection image and the reconstructed image. Further, depending on the type of RI to be used, high accumulation may occur in a region other than the region of interest (for example, a diagnosis target organ).

  If there is a highly integrated part in a part other than the target part, when converting real number data into integer data, the pixel value of the highly integrated part raises the maximum pixel value of the entire image data. As a result, the data accuracy of the site of interest falls. As described above, when the data accuracy of the target region is reduced when converting real number data to integer data, the diagnostic accuracy and the data analysis accuracy are also reduced.

  Specifically, there are cases where it is difficult to observe the myocardium or the like, which is the original site of interest, due to the effect of RI highly accumulated in the liver or gallbladder.

  In addition, even if it is going to solve the above problems by the setting process of a reconstruction range, a solution may be difficult. That is, for example, when the liver is a highly integrated part, and this highly integrated part is covered with the heart, which is the target part, it is difficult to avoid the liver and set the reconstruction range. It is. The same applies to the PET apparatus.

  The present invention has been made in view of the above circumstances. For example, real data having a wide dynamic range, such as a reconstructed image created by a nuclear medicine diagnostic apparatus, can be obtained without reducing the data accuracy of a region of interest. An object of the present invention is to provide an image processing apparatus capable of converting to integer data.

An image processing apparatus according to an embodiment includes:
An image processing apparatus that converts medical image real number data, which is real number data acquired by imaging a subject, into integer data to create medical image integer data,
A display condition setting unit for setting a window limited range, which is a conversion range of pixel values in the conversion process, according to a pixel value of a pixel included in a region corresponding to the region of interest of the subject;
An integer data creation condition calculation unit that calculates a rescale coefficient in the conversion process based on the maximum value and the minimum value of the window limited range;
Based on the rescaling coefficient, an integer data creation unit that creates the medical image integer data by converting the medical image real number data;
It is characterized by comprising.

A program according to an embodiment is:
A program for causing a computer to realize a function of converting medical image real number data, which is real number data acquired by imaging a subject, into medical image integer data, which is integer data,
On the computer,
A display condition setting function for setting a window limited range, which is a conversion range of pixel values in the conversion process, in accordance with a pixel value of a pixel included in a region corresponding to the target region of the subject;
An integer data creation condition calculation function for calculating a rescale coefficient in the conversion process based on the maximum value and the minimum value of the range set by the display condition setting function;
A conversion processing function for creating the medical image integer data from the medical image real number data based on the rescale factor;
It is characterized by realizing.

An image processing apparatus according to an embodiment includes:
An image processing apparatus that converts first medical image integer data, which is integer data acquired by imaging a subject, and generates second medical image integer data, which is integer data having a shorter data length. ,
A display for displaying image data;
A display level setting unit for displaying the first medical image integer data on the display unit at a desired display level;
Based on the display level set by the display level setting unit, a display condition setting unit that sets a window limited range that is a conversion range of pixel values in the conversion process;
An integer data creation condition calculation unit that calculates a rescale coefficient in the conversion process based on the maximum value and the minimum value of the window limited range;
An integer data creating unit that creates the second medical image integer data by converting the first medical image real number data based on the rescaling coefficient;
It is characterized by comprising.

FIG. 1 is a block diagram illustrating a configuration example of an image processing apparatus according to an embodiment of the present invention. FIG. 2 is a graph showing an example of a histogram of pixel values and display levels of reconstructed image data. FIG. 3 is a flowchart of “reconstructed image integer data creation processing” by the control unit of the image processing apparatus according to the embodiment of the present invention. FIG. 4 is a diagram illustrating an example of two-dimensional projection data displayed on the display unit. FIG. 5 is a diagram illustrating an example of a reconstruction processing range set with a cursor on the two-dimensional projection data by the operation of the operation unit by the user. FIG. 6 is a diagram illustrating an example of an optimal display image displayed on the display unit in the window setting step.

  Hereinafter, an image processing apparatus and a program according to an embodiment of the present invention will be described.

  FIG. 1 is a block diagram illustrating a configuration example of an image processing apparatus according to an embodiment of the present invention. In the example shown in the figure, the image processing apparatus 1 according to the present embodiment is connected to a SPECT apparatus 100 and a PACS 200 including a DICOM server and a DICOM viewer. The PACS 200 is connected to the database 300.

  The SPECT apparatus 100 collects two-dimensional projection data by detecting gamma rays emitted from an RI administered to a living body in a nuclear medicine examination (SPECT examination) with a gamma camera. In the present embodiment, an example in which a SPECT apparatus is used as a nuclear medicine diagnosis apparatus will be described, but the same can be said for a PET apparatus.

  The image processing apparatus 1 includes a preprocessing unit 101, a reconstruction range setting unit 102, a reconstruction unit 103, a display condition setting unit 12, an integer data creation condition calculation unit 14, an integer data creation unit 16, A control unit 18, a display unit 20, and an operation unit 22 are included.

  The preprocessing unit 101 performs the following correction process on the two-dimensional projection data collected by the SPECT apparatus 100.

  That is, the pre-processing unit 101 corrects the influence of attenuation caused by Compton scattering when radiation propagates in the subject, and moves the rotation center position when the detector of the SPECT apparatus is driven to rotate. The rotation center correction for correcting the deviation of the projection data caused by the above and the uniformity correction for correcting the uniformity of the projection data are executed on the two-dimensional projection data according to a known procedure.

  The reconstruction range setting unit 102 sets the reconstruction processing range of the two-dimensional projection data collected by the SPECT apparatus 100 based on the operation of the operation unit 22 by the user.

  The reconstruction unit 103 reconstructs the two-dimensional projection data preprocessed by the preprocessing unit 101 by a known method such as a three-dimensional backprojection method or a three-dimensional fast Fourier transform (FFT) method. Create three-dimensional reconstructed image data. The reconstructed image data created by the reconstructing unit 103 may be in a data format of a unique format including real number data. The reconstructed data is expressed in a unique format including the real number data.

  Hereinafter, the reconstructed image data created by the reconstructing unit 103 is referred to as “reconstructed image real number data”. Further, data obtained by converting the reconstructed image real number data into integer data (hereinafter referred to as “real integer conversion”) is referred to as “reconstructed image integer data”.

  The display condition setting unit 12 defines a pixel value conversion range (referred to as a “window limited range”) in real integer conversion when generating real image integer data by converting real image data to real integer conversion. The “window maximum value WU” and “window minimum value WL” to be set are set based on the operation of the operation unit 22 by the user.

  Hereinafter, setting processing of the window maximum value WU and the window minimum value WL by the display condition setting unit 12 will be described in detail.

  FIG. 2 is a diagram illustrating an example of a histogram of pixel values and display levels of reconstructed image data. In the graph shown in the figure, the vertical axis represents the pixel value, and the horizontal axis represents the number of pixels corresponding to each pixel value. Further, the gray scale bar shown in the figure indicates the display level of the reconstructed image data.

  Specifically, the left graph of the three graphs illustrated in FIG. 2 is a graph illustrating an example of a histogram of pixel values of the reconstructed image real number data and a display level. The middle graph of the three graphs shown in FIG. 2 is a graph showing an example of a histogram of pixel values of reconstructed image integer data created by performing real integer conversion on the reconstructed image real number data corresponding to the left graph. . The right graph of the three graphs shown in FIG. 2 is an example of a display level of the reconstructed image integer data corresponding to the middle graph.

  In the “window setting step”, the user causes the display unit 20 to display the reconstructed image integer data at various display levels, and refers to these displays to confirm the visibility of the target region (for example, the diagnostic target region) of the subject. The best “optimum display image” is determined. In other words, in the “window setting step”, the user changes the display level of the reconstructed image integer data displayed on the display unit 20 in various ways, and visually recognizes the “optimal display”. Find the “image”.

  Then, the display condition setting unit 12 sets the window maximum value WU and the window minimum value WL (window limited range) in real integer conversion based on this “optimum display image”.

  The “window setting step” will be described in detail later.

  For example, as shown in FIG. 2, the window maximum value WU and the window minimum value WL corresponding to the display level of the optimal display image (the display level set by the user so that the region of interest of the subject can be viewed most easily) are “ A predetermined pixel value range having a mode value of pixel values in a pixel group corresponding to the “target region of the subject” as a substantially central value is defined as a window limited range.

  As described above, according to the image processing apparatus according to the present embodiment, the user can change the display level in the “window setting step” in various ways, so that the target level of the subject can be viewed most easily. The heading / display condition setting unit 12 calculates a rescale coefficient (details will be described later) based on the display level.

  Therefore, the user can very easily obtain the reconstructed image integer data corresponding to the display level with the best visibility of the site of interest.

  In addition, in order to find and set the display level at which the attention site of the subject is most easily viewable, for example, the user can manually determine with the operation unit 22 while referring to the image displayed on the display unit 20, for example, the attention site A region of interest (ROI) is set, a histogram of pixel values of the pixel group in the ROI is created, and automatically set based on the histogram (the mode value of the pixel value in the pixel group in the ROI is set) The display level may be set so as to be included).

  The integer data creation condition calculation unit 14 is expressed by the following (Equation 1) based on the window limited range (the window maximum value WU and the window minimum value WL) corresponding to the display condition set by the display condition setting unit 12. The value of the rescale slope (Rescale Slope) A and the value of the rescale intercept (Rescale Intercept) B in the primary conversion equation of the real integer conversion are calculated. Hereinafter, the rescale inclination A and the rescale intercept B are collectively referred to as “rescale coefficients A and B”.

Integer data = A * Real number data + B (Formula 1)
Here, in the calculation of the rescale coefficients A and B, based on the window maximum value WU and the window minimum value WL, the processing result (that is, the reconstructed image integer) by the primary conversion equation of the real integer conversion expressed by (Equation 1) The values of the rescale coefficients A and B are determined so that the data is within a range that can be expressed by a 2-byte signed integer or unsigned integer.

Specifically, for example, when the window maximum value WU = 90 and the window minimum value WL = 10 are set, the rescale coefficients A and B are set so that the pixel value data within this range is assigned to 0 to 1024, for example. You just have to decide. Therefore, using (Equation 1),
1024 = A * 90 + B
0 = A * 10 + B
As
A = 12.8
B = -128.0
Is calculated.

If only rescale slope A is defined,
1024 = A * 90 + 0
As
A = 11.38
Is calculated.

  By the way, for data relating to pixel values that are out of the window limited range, the rescale coefficients A and B are taken into consideration, and the maximum value when the maximum value (32767) or minimum value (−32768) of 16 bits is exceeded. The value and the minimum value may be replaced.

  The integer data creation unit 16 is based on the values of the rescale coefficients A and B calculated by the integer data creation condition calculation unit 14 and the primary conversion formula of real number integer conversion (Formula 1). The real number data of the reconstructed image created by the above is subjected to real integer conversion to create reconstructed image integer data.

  Note that the window limited range determined by the user is “expanded by a predetermined constant multiple” (with a margin set), a real integer conversion process is executed, and the reconstructed image integer data created thereby is saved. You may comprise as follows. Specifically, for example, a margin may be set on both the window maximum value WU side and the window minimum value WL side, or a margin may be set only on one side.

  In addition, the “predetermined constant” that determines the expansion width when setting the margin may be arbitrarily set by the operation of the operation unit 22 by the user, or the non-volatile provided in the image processing apparatus 1 as a default value. It may be recorded in a memory (not shown) or the like.

  Specifically, for example, when the window maximum value WU = 90 and the window minimum value WL = 10 are set, and “3 times” is set as the “predetermined constant” when setting the margin. Assigns data within a range obtained by multiplying the window limited range by 3 so that the maximum value (32767) of 16 bits is approached.

  As described above, the reconstructed image integer data created by expanding the window limited range by a predetermined constant multiple (with a margin set) is saved, so that the window limited range can be made original by, for example, an erroneous operation. Even if the setting is narrower than intended, the data can be restored by the amount that the window limited range is expanded.

  The control unit 18 is a CPU (Central Processing Unit) that comprehensively controls each unit of the image processing apparatus 1. The display unit 20 is a monitor that displays various image data and the like. The operation unit 22 is input operation means for various operations by the user.

  The flow of the “reconstructed image integer data” creation process by the control unit 18 of the image processing apparatus 1 according to the present embodiment will be described below with reference to FIG.

  FIG. 3 is a diagram showing a flowchart of the “reconstructed image integer data” creation process by the control unit 18 of the image processing apparatus 1 according to the present embodiment. FIG. 4 is a diagram illustrating an example of two-dimensional projection data displayed on the display unit 20. FIG. 5 is a diagram illustrating an example of the reconstruction processing range set by the cursor C on the two-dimensional projection data by the operation of the operation unit 22 by the user. FIG. 6 is a diagram illustrating an example of an optimal display image displayed on the display unit 20 in the window setting step.

  First, when the two-dimensional projection data collected by the SPECT apparatus 100 is input to the image processing apparatus 1, the control unit 18 performs the above-described various correction processes on the two-dimensional projection data by the preprocessing unit 101. (Step S1).

  Subsequently, the user sets the cursor C as shown in FIG. 5 by the operation unit 22 on the two-dimensional projection data (for example, see FIG. 4) of the subject displayed on the display unit 20, and performs a desired reconstruction process. Specify a range. In the example shown in FIG. 5, the range from position 47 to position 72 in the two-dimensional projection data is designated by the cursor C as the reconstruction processing range.

  Here, the control unit 18 sets the reconstruction processing range by the reconstruction range setting unit 102 based on the designation of the reconstruction processing range by the user, and the reconstruction unit 103 performs two-dimensional projection on the reconstruction processing range. The data is reconstructed to create three-dimensional reconstructed image real number data (step S2).

  And the control part 18 displays this reconstruction image real number data on the display part 20, and transfers to the window setting step (step S3) by a user.

  In this window setting step, the user operates the operation unit 22 to display the reconstructed image real number data on the display unit 20 at a desired display level, and further operates the operation unit 22 while viewing them to display the display level. Will be changed in various ways.

  The control unit 18 causes the display unit 20 to display the reconstructed image real number data at the display level set by the user in this way. Finally, the user causes the display unit 20 to display an optimal display image with a display level at which the region of interest of the subject is most easily viewed (see, for example, FIG. 6).

  Note that the user sets a region of interest (ROI) in the region of interest, and the display condition setting unit 12 creates a histogram of pixel values of the pixel group in the ROI and automatically sets the display level based on the histogram. Of course, you may comprise.

  And a user performs operation which performs real number integer conversion based on the said optimal display image with the operation part 22. FIG. In response to this operation, the control unit 18 uses the display condition setting unit 12 to set the window limited range (the window maximum value WU and the window minimum value WL) corresponding to the optimum display image displayed on the display unit 20 ( Step S3: Window setting step).

  Next, the control unit 18 calculates the rescale coefficients A and B by the integer data creation condition calculation unit 14 based on the window limited range (the window maximum value WU and the window minimum value WL) set in the window setting step ( Step S4; rescale coefficient calculation step). The details of the processing for calculating the rescale coefficients A and B are as described above.

  In addition, you may provide the button for a user to perform the calculation process of the rescaling coefficients A and B. FIG. Further, the rescale coefficients A and B may be calculated based on the display level when the display report image (secondary capture) is created.

  Subsequently, the control unit 18 converts the reconstructed image real number data into real integer conversion based on the rescale coefficients A and B calculated in the rescale coefficient calculation step and the primary conversion expression (Equation 1) of real integer conversion. Then, reconstructed image integer data is created (step S5; integer data creation step).

  Then, the control unit 18 transmits the reconstructed image integer data created in the integer data creation step to the database 300 via the PACS 200 and stores it.

  When the reconstructed image integer data created by the above process is used for actual image diagnosis, the window maximum value WU and the window minimum value WL used for calculating the rescale coefficients A and B are set as default values. The reconstructed image integer data may be displayed.

  By the way, the above-described series of processing by the image processing apparatus according to the present embodiment is a software product that is independent of the image processing apparatus by programming or by reading the program into a storage medium after programming. Sales and distribution as a single unit are facilitated, and the technology according to the present embodiment can be used on other hardware.

  As described above, according to the present embodiment, for example, real number data having a wide dynamic range, such as a reconstructed image created by a nuclear medicine diagnostic apparatus, can be converted into an integer without degrading the “data accuracy” of the region of interest. An image processing apparatus and a program that can be converted into data can be provided.

  In this embodiment, “data accuracy” is data accuracy as a concept that includes not only the accuracy of image quality parameters at the time of image quality adjustment but also “accuracy in image calculation and image processing”.

  According to the image processing apparatus and the program according to the present embodiment, the user actually confirms the optimal display image of the display level at which the region of interest of the subject is most easily viewed, and the optimal window limited range corresponding to the display level. Since real integer conversion can be performed, it is possible to easily and reliably create reconstructed image integer data with the accuracy intended by the user.

  According to the prior art, when processing image data with a wide dynamic range such as reconstructed image real number data created by a nuclear medicine diagnostic apparatus, there is a case where a highly integrated portion exists in the image data. When the target region (diagnosis target region) has a lower count value than the highly integrated region, the data accuracy of the target region is lowered due to the influence of the highly integrated region during the real integer conversion process.

  On the other hand, according to the image processing apparatus and the program according to the present embodiment, the window limited range (window maximum value) is set according to the pixel value of the pixel group included in the region corresponding to the target region (diagnosis target region) of the subject. WU, window minimum value WL) are set, and based on this, a rescale coefficient (rescale slope A, rescale intercept B) in real integer conversion processing is calculated, and based on this, medical image real number data is converted into the medical image integer data. Convert it. Thereby, medical image integer data can be created while maintaining the data accuracy of the region of interest, and can be stored in the DICOM standard.

  As a specific provisional setting / determination method of the window limited range, it may be determined by the window maximum value WU and the window minimum value WL as in the above example, or by the median value WC and the width WW. Of course, it may be determined, or other methods may be adopted.

  It is very effective to apply the image processing apparatus and program according to the present embodiment to processing of medical image data having a wide dynamic range such as an MR image and a CT image.

  The “real number data” in the above-described case is data including both single-precision real numbers and double-precision real numbers. By the way, for integer data, the current DICOM data is 16-bit data. The above-described embodiment can also be applied to a case where 32-bit double precision integer or integer data having a data length longer than that is converted into 16-bit integer data. In other words, the above-described embodiment can also be applied to the case where double precision integer data larger than a 16-bit integer or integer data having a data length longer than that is converted into 16-bit integer data.

  Although several embodiments of the present invention have been described, these embodiments are presented by way of example and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalents thereof.

    WU: Window maximum value, WL: Window minimum value, A: Rescale slope, B: Rescale intercept, 1 ... Image processing device, 12 ... Display condition setting unit, 14 ... Integer data creation condition calculation unit, 16 ... Integer data creation unit DESCRIPTION OF SYMBOLS 18 ... Control part, 20 ... Display part, 22 ... Operation part, 100 ... SPECT apparatus, 101 ... Pre-processing part, 102 ... Reconstruction range setting part, 103 ... Reconstruction part, 200 ... PACS, 300 ... Database.

Claims (5)

An image processing apparatus that converts medical image real number data, which is real number data acquired by imaging a subject, into integer data to create medical image integer data,
A display for displaying image data;
A display level setting unit for displaying the medical image real number data on the display unit at a desired display level;
Based on the display level set by the display level setting unit, a display condition setting unit that sets a window limited range that is a conversion range of pixel values in the conversion process;
An integer data creation condition calculation unit that calculates a rescale coefficient in the conversion process based on the maximum value and the minimum value of the window limited range;
Based on the rescaling coefficient, an integer data creation unit that creates the medical image integer data by converting the medical image real number data;
An image processing apparatus comprising: The display level setting unit
A setting operation unit for the user to set a desired display level;
An automatic setting unit that sets a display level in accordance with a pixel value of a pixel included in a region corresponding to the region of interest of the subject;
The image processing apparatus according to claim 1, further comprising: The display condition setting unit sets a range obtained by expanding a window limited range based on the display level set by the display level setting unit by a predetermined constant multiple as a window limited range for the conversion process. The image processing apparatus according to claim 2. A program that causes a computer to realize a function of converting medical image real number data, which is real number data acquired by imaging a subject, into integer data to create medical image integer data,
On the computer,
A display level setting function for displaying the medical image real number data at a desired display level;
Based on the display level set by the display level setting function, the window limited range, which is the pixel value conversion range in the conversion process, is set according to the pixel value of the pixel included in the region corresponding to the region of interest of the subject. Display condition setting function to set
An integer data creation condition calculation function for calculating a rescale coefficient in the conversion process based on the maximum value and the minimum value of the range set by the display condition setting function;
A conversion processing function for creating the medical image integer data from the medical image real number data based on the rescale factor;
A program characterized by realizing. An image processing apparatus that converts first medical image integer data, which is integer data acquired by imaging a subject, and generates second medical image integer data, which is integer data having a shorter data length. ,
A display for displaying image data;
A display level setting unit for displaying the first medical image integer data on the display unit at a desired display level;
Based on the display level set by the display level setting unit, a display condition setting unit that sets a window limited range that is a conversion range of pixel values in the conversion process;
An integer data creation condition calculation unit that calculates a rescale coefficient in the conversion process based on the maximum value and the minimum value of the window limited range;
An integer data creating unit that creates the second medical image integer data by converting the first medical image real number data based on the rescaling coefficient;
An image processing apparatus comprising:

Images