JP2005168962A - Diagnostic support apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a diagnostic support apparatus that can extract information for supporting a diagnosis and can use it to help the diagnosis. <P>SOLUTION: In the diagnostic support apparatus relating to this prevention, a frequency coefficient resolving means resolves an image into a plurality of frequency bands and calculates the coefficient. A feature analyzing means calculates prescribed featured values from a prescribed region of a component of the coefficient calculated by the frequency coefficient resolving means. In addition, in the diagnostic support apparatus relating to this prevention, an indicating means displays the image, a frequency coefficient resolving means resolves the image into a plurality of frequency bands, and calculates the coefficient. The feature analyzing means calculates prescribed featured values from a component of the coefficient calculated by the frequency coefficient resolving means in a prescribed region determined based on a region indicated by the indication means. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a diagnosis support apparatus, and more particularly to a diagnosis support apparatus that quantifies image properties using a plurality of frequency components.

  As disclosed in Patent Document 1, it is possible to convert a radiographic image into a digital image signal due to recent advances in digital technology, perform image processing on the digital image signal, and display or print out on a CRT or the like. Has been done.

  By the way, the purpose of the radiographic image is to diagnose a subject. Therefore, it is necessary to create an image suitable for diagnosis, and to add information necessary for diagnosis.

  For such purposes, conventionally, image processing has been subjected to frequency processing such as sharpening processing and image processing such as dynamic range compression.

JP-A-6-342099

  However, the conventional image processing suitable method only discloses the idea of improving the image quality of the entire image, and there is a problem that special processing is performed only on the specific area and information useful for diagnosis cannot be obtained. Therefore, information necessary for diagnosis can be obtained only from the display of the general image.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a diagnosis support apparatus that can extract information for supporting diagnosis and use the information for diagnosis.

  In order to achieve the above object, the diagnosis support apparatus according to claim 1, wherein the frequency coefficient decomposing means decomposes the image into a plurality of frequency bands, calculates the coefficients, and the characteristic analyzing means calculates the coefficients calculated by the frequency coefficient decomposing means. A predetermined feature amount is calculated from a predetermined region of the component.

  The diagnosis support apparatus according to claim 2, wherein the instruction means displays an image, the frequency coefficient decomposition means decomposes the image into a plurality of frequency bands, calculates the coefficients, and the feature analysis means indicates an area designated by the instruction means. The predetermined feature amount is calculated from the coefficient component calculated by the frequency coefficient decomposing means in the predetermined region determined by

  The diagnosis support apparatus according to claim 3, wherein the instruction means displays an image, the frequency coefficient decomposition means decomposes the image into a plurality of frequency bands, calculates the coefficients, and the feature analysis means indicates an area designated by the instruction means. A predetermined feature amount is calculated from the coefficient component calculated by the frequency coefficient decomposing means in a predetermined region determined by the following, and the display means displays the feature amount calculated by the feature analyzing means.

  According to a fourth aspect of the present invention, the diagnosis support apparatus uses a wavelet transform method as a method of differentiating into frequencies.

  According to a fifth aspect of the present invention, there is provided a diagnostic support apparatus for displaying the coefficient component for each frequency band in a graph by the display means.

  According to a sixth aspect of the present invention, the target for calculating the feature amount is a tumor.

  According to the first aspect of the present invention, information that cannot be obtained only by normal image display can be obtained by decomposing and characterizing the image for each frequency band. By using this information, there is an effect that the characteristics of the image can be objectively grasped.

  According to the second aspect of the present invention, since a specific area of the image can be indicated, there is an effect that information on the attention area can be accurately extracted.

  According to the third aspect of the invention, there is an effect that the characteristic of the image can be easily grasped by displaying the feature amount for each frequency band of the predetermined region as an image.

  According to the fourth aspect of the present invention, by using the wavelet for frequency decomposition, it is possible to extract local frequency characteristics and to extract local characteristics of an image.

  According to the fifth aspect of the invention, there is an effect that the characteristics of the image can be easily grasped by graphing the characteristics for each frequency band.

  According to the sixth aspect of the invention, when the object is a tumor, the characteristics of the tumor can be objectively grasped, which is useful for diagnosis.

  Embodiments of the present invention will be described below with reference to the accompanying drawings.

  FIG. 1 shows an X-ray imaging apparatus 100 according to an embodiment of the present invention. That is, the X-ray imaging apparatus 100 is an X-ray imaging apparatus having a diagnosis support function, and includes a preprocessing circuit 106, a CPU 108, a main memory 109, an operation panel 110, an image display 111, and a diagnosis support circuit 112. Data is exchanged with each other via the CPU bus 107.

  In the diagnosis support circuit 112, reference numeral 113 is an instruction circuit circuit for indicating a predetermined area in the image, and 114 is a frequency coefficient decomposition circuit for decomposing the image into a plurality of frequency bands and calculating respective frequency coefficients. Is a feature analysis circuit that calculates the feature quantity from the frequency coefficient calculated by the frequency coefficient decomposition circuit 114 in a predetermined area determined by the area specified by the instruction circuit 113. The display device 111 displays an image and the feature amount calculated by the feature analysis circuit 115.

  The X-ray imaging apparatus 100 includes a data acquisition circuit 105 connected to the preprocessing circuit 106, a two-dimensional X-ray sensor 104 and an X-ray generation circuit 101 connected to the data acquisition circuit 105. These circuits are also connected to the CPU bus 107.

  FIG. 2 is a flowchart showing a process flow of the diagnosis support circuit 112 according to the embodiment of the present invention. FIG. 3 is a diagram showing an original image obtained by photographing, where 301 indicates a malignant tumor and 302 indicates a benign tumor. There are two types of tumors, benign and malignant, such as cancer, and the characteristics are said to appear in the tumor periphery. For example, in malignant cases, there is an example in which blood vessels and the like are drawn into the tumor and the periphery of the tumor is jagged. In benign, there is an example in which there is no drawing of blood vessels and the whole tumor is rounded. Thus, the distribution of pixel values around the tumor is an important factor for tumor determination.

  FIG. 4 is a diagram in which the tumors 301 and 302 in FIG. 3 are divided and displayed.

  4A shows a malignant tumor 301, and FIG. 4B shows a predetermined region 401 of the tumor 301. (C) shows the tumor 302, and (d) shows a predetermined region 401 of the tumor 302.

  FIG. 5 is a diagram illustrating a frequency decomposition method using wavelet transform, where (a) is a configuration of the decomposition method, and (b) is a diagram illustrating coefficients for each decomposed frequency. 6A and 6B are diagrams showing frequency components in the predetermined regions 401 and 402. FIG. 6A corresponds to the tumor 301, and FIG.

  In the X-ray imaging apparatus 100 as described above, first, the main memory 109 stores various data necessary for processing by the CPU 108 and also includes a work memory for work of the CPU 108.

  The CPU 108 uses the main memory 109 to perform operation control of the entire apparatus according to the operation from the operation panel 110. Thereby, the X-ray imaging apparatus 100 operates as follows.

  First, the X-ray generation circuit 101 emits an X-ray beam 102 to the inspection object 103. The X-ray beam 102 emitted from the X-ray generation circuit 101 passes through the object 103 while being attenuated, reaches the two-dimensional X-ray sensor 104, and is output as an X-ray image by the two-dimensional X-ray sensor 104. . Here, the X-ray image output from the two-dimensional X-ray sensor 104 is, for example, a human body image.

  The data acquisition circuit 105 converts the X-ray image output from the two-dimensional X-ray sensor 104 into an electrical signal and supplies it to the preprocessing circuit 106. The preprocessing circuit 106 performs preprocessing such as offset correction processing and gain correction processing on the signal (X-ray image signal) from the data acquisition circuit 105. The X-ray image signal preprocessed by the preprocessing circuit 106 is transferred as an original image to the main memory 109 and the diagnosis support circuit 112 via the CPU bus 107 under the control of the CPU 108.

  Here, the processing flow of the diagnosis support circuit 112 will be described with reference to the flowchart of FIG.

  First, the display 111 displays an original image obtained by shooting on the display screen (s201). Then, the frequency coefficient decomposition circuit 114 performs frequency decomposition processing using discrete wavelet transform. This operation is as follows.

  The frequency coefficient decomposition circuit 114 performs a two-dimensional discrete wavelet transform process on the input image, calculates a transform coefficient, and outputs it. The processing configuration of the frequency coefficient decomposition circuit 114 in the present embodiment is shown in FIG. In the figure, an input image signal is separated into even address and odd address signals by a combination of a delay element and a downsampler, and subjected to filter processing by two filters p and u. In the figure, s and d represent low-pass coefficients and high-pass coefficients when one-level decomposition is performed on a one-dimensional image signal, respectively, and are calculated by the following equations.

d (n) = x (2 * n + 1) −floor (x (2 * n) + x (2 * n + 2)) / 2)
... (Formula 1)
s (n) = x (2 * n + floor ((d (n-1) + d (n)) / 4) (Expression 2)
However, x (n) is an image signal to be converted.

  With the above processing, one-dimensional discrete discrete wavelet transform processing is performed on the image signal. The two-dimensional discrete wavelet transform is a one-dimensional transform that is sequentially performed in the horizontal and vertical directions of the image, and details thereof are publicly known, and thus description thereof is omitted here.

  FIG. 5B is a configuration example of a two-level transform coefficient group obtained by two-dimensional transform processing, and an image signal is decomposed into coefficient sequences HH1, HL1, LH1,..., LL in different frequency bands (s202 ). In the present embodiment, frequency resolution processing up to five levels is performed, and coefficient sequences HH1, HH2, HH3, HH4, and HH5 in different frequency bands are used. The discrete wavelet transform is suitable for high-speed decomposition into local domain frequencies. Therefore, the use of wavelet transform has the effect of extracting the frequency characteristics of the local region of the image.

  Next, the instruction unit 113 instructs a specific area from the image displayed on the display 111. For example, the tumor 301 in FIG. 3 is instructed, and the cut out image is FIG. 4A (s203). Since the point of interest on the image can be instructed by the instruction means 113, the region of interest can be extracted appropriately.

  Then, the feature analysis circuit 115 extracts frequency coefficients HH1, HH2, HH3, HH4, and HH5 corresponding to regions within a certain region from the point indicated by the instruction means 113 (s204). Here, the predetermined region is a region that is concentric from the point instructed by the instruction means 113, and 401 and 402 in FIG. 4 correspond to this, for example. Coefficients are extracted from the coefficient spaces HH1, HH2,... Corresponding to this area. This concentric area corresponds to the peripheral area of the tumor.

  Then, the feature analysis circuit 115 calculates an average value of frequency coefficients in the region corresponding to the region 401 (s205). Thereby, a component for each frequency region representing the region 401 is calculated. Note that the value representing the frequency component is not limited to an average value but may be a statistic such as an intermediate value.

  Then, the feature amount for each frequency band calculated by the feature analysis circuit 115 is displayed (s206). For example, FIG. 6A is an example, and the magnitude of the feature amount is graphed for each frequency band. Here, the coefficient (which also indicates the frequency band) used by HH1 or the like is shown, and the length of the graph indicates the size of the feature amount.

  By displaying in this way, it is possible to refer to the frequency characteristics of the peripheral portion of the image rather than silently viewing only the image, which is useful for tumor diagnosis.

  Furthermore, when it is desired to see the components of different tumor frequency bands, the processing from s203 is repeated. For example, FIG. 6B is a diagram in which the tumor 302 is indicated, the same processing is performed, and the coefficient components are displayed.

  As described above, according to the present embodiment, by decomposing and characterizing an image for each frequency band, information that cannot be obtained only by normal image display can be obtained. By using this information, there is an effect that the characteristics of the image can be objectively grasped. When the subject is a tumor, the characteristics of the tumor can be objectively grasped, which is useful for diagnosis.

  Furthermore, since a specific area of the image can be indicated, there is an effect that information on the attention area can be accurately extracted. In addition, displaying the feature amount for each frequency band of the predetermined region as an image has an effect of easily grasping the characteristics of the image. Then, by graphing the characteristics for each frequency band, there is an effect that the characteristics of the image can be easily grasped. By using a wavelet for frequency decomposition, it is possible to extract local frequency characteristics and to extract local characteristics of an image.

  The present invention is applied particularly to a diagnosis support apparatus that quantifies the properties of an image using a plurality of frequency components, and information that cannot be obtained only by normal image display can be obtained. It is possible to produce an effect that the characteristics of the can be objectively grasped.

It is a block diagram which shows the structure of the diagnosis assistance apparatus which concerns on this invention. It is a flowchart which shows the process sequence of the diagnosis assistance apparatus which concerns on this invention. It is a figure which shows the image in which the tumor is reflected. It is a figure which shows the area | region which calculates the extracted tumor and the feature-value. It is explanatory drawing of a discrete wavelet transform. It is the figure which graphed the feature-value for every frequency band.

Explanation of symbols

114 Frequency coefficient decomposition circuit 115 Feature analysis circuit 113 Instruction circuit 111 Display

Claims (6)

  Frequency coefficient decomposing means for decomposing an image into a plurality of frequency bands and calculating coefficients thereof, and feature analyzing means for calculating a predetermined feature amount from a predetermined area of the coefficient component calculated by the frequency coefficient decomposing means A diagnosis support apparatus characterized by the above.   An instruction means for displaying an image and indicating a specific area in the displayed image, a frequency coefficient decomposition means for decomposing the image into a plurality of frequency bands and calculating coefficients thereof, and an area indicated by the instruction means A diagnostic support apparatus comprising a feature analysis unit that calculates a predetermined feature amount from a coefficient component calculated by the frequency coefficient decomposition unit in a predetermined region.   An instruction means for displaying an image and indicating a specific area in the displayed image, a frequency coefficient decomposition means for decomposing the image image into a plurality of frequency bands and calculating coefficients thereof, and an area indicated by the instruction means A feature analysis unit that calculates a predetermined feature amount from a coefficient component calculated by the frequency coefficient decomposition unit in a predetermined region that is determined, and a display unit that displays the feature amount calculated by the feature analysis unit Diagnosis support device.   The diagnosis support apparatus according to any one of claims 1 to 3, wherein a wavelet transform method is used as a method of differentiating into frequencies by the frequency coefficient decomposing means.   The diagnosis support apparatus according to any one of claims 1 to 3, wherein the display unit displays the coefficient component for each frequency band in a graph.   The diagnosis support apparatus according to claim 1, wherein a target for calculating a feature amount is a tumor.

Images