JP2010136779A - Apparatus and method for processing medical image - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To sharpen a medical image including high count areas having high pixel values and low count areas having low pixel values. <P>SOLUTION: A medical image processing apparatus 1 for photographing a subject includes an unsharp mask processing section 13 that generates a sharpened image by carrying out unsharp mask processing on a medical image, a post-processing section 15 that generates a post-processed image by masking high count pixels having pixel values higher than a predetermined threshold in the sharpened image generated by the unsharp mask processing section 13, an output section 19 that outputs the post-processed image generated by the post-processing section 15. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a technique for sharpening a medical image obtained by photographing a subject.In particular, in a medical image in which a high count area with a high pixel value and a low count area with a low pixel value exist in one image, In particular, the present invention relates to a technique for clarifying a low count area.

For example, Patent Document 1 describes image processing for performing unsharp mask processing.
JP-A-10-75395

  By the way, in the case of medical images, particularly RI images, the RI count value counted by the RI imaging device is the pixel value. For this reason, there are locations where RI is very concentrated in the subject's body and the pixel value is extremely high, and where a small amount of RI is distributed and the pixel value is very low in a single image. There is.

  Accordingly, an object of the present invention is to sharpen an RI image in which such a high count area with a high pixel value and a low count area with a low pixel value exist.

  The medical image processing apparatus according to one embodiment of the present invention includes a ROI setting unit that sets a region of interest (ROI) for a medical image obtained by imaging a subject, and the ROI setting unit for the medical image. A normalization unit that normalizes pixel values in the set ROI to a predetermined value to generate a preprocessed image, and performs an unsharp mask process on the preprocessed image generated by the normalization unit, and Contrast-enhanced image generation means for outputting a contrast-enhanced image generated in the process of sharp mask processing, the medical image, a high-count image having a pixel value larger than a predetermined threshold, and a pixel value smaller than the threshold Image separation means for separating the image into a low count image, and a high count compression by compressing the pixel value of the high count image Compression processing means for generating an image, image synthesizing means for generating a composite image based on the contrast-enhanced image, the high-count compressed image, and the low-count image.

  In a preferred embodiment, the contrast-enhanced image generation means generates a plurality of blurred images by applying a plurality of different unsharp masks to the preprocessed image, and the preprocessed image and the plurality of blurs are generated. Generating a plurality of difference images between any one of the images and any two images of the plurality of blurred images, and generating the contrast-enhanced image based on the plurality of difference images Also good.

  In a preferred embodiment, the plurality of unsharp masks are N types of unsharp masks, each having a different size, and the first, second,..., Nth unsharp masks in ascending order of size. When it is a mask, the contrast-enhanced image generation unit applies an m-th (1 ≦ m ≦ N) unsharp mask to the preprocessed image to generate an m-th blurred image, and A first difference image that is a difference between a blurred image and the preprocessed image, and an mth difference that is a difference between the mth (2 ≦ m ≦ N) blurred image and the (m−1) th blurred image. May be generated, and the contrast-enhanced image may be generated by adding the first to Nth difference images.

  In a preferred embodiment, the contrast-enhanced image generating means uses an average value of all pixels in the m-th (1 ≦ m ≦ N) unsharp mask as a value of one pixel in each unsharp mask. The m-th blurred image may be generated by assigning.

  In a preferred embodiment, the compression processing means may perform compression at a compression rate corresponding to a ratio between the maximum value of the high count image and the threshold value.

  A medical image processing apparatus according to an embodiment of the present invention includes a medical image obtained by photographing a subject, a high count image having a pixel value larger than a predetermined threshold, and a low count image having a pixel value smaller than the threshold. An image separating means for separating the image, an unsharp mask processing means for applying an unsharp mask process to the low count image to generate a low count sharpened image, the low count sharpened image and the high count image And an image synthesis means for generating a synthesized image based on the above.

  An apparatus for processing a medical image according to an embodiment of the present invention includes an unsharp mask processing unit that performs an unsharp mask process on a medical image obtained by photographing a subject and generates a sharpened image, and the unsharp mask process. A post-processing unit that generates a post-processed image by normalizing a high-count pixel having a pixel value larger than a predetermined threshold to a predetermined value with respect to the sharpened image generated by the unit, and the post-processing unit Output means for outputting the post-processed image.

Hereinafter, a medical image processing apparatus according to an embodiment of the present invention will be described with reference to the drawings. The medical image processing apparatus according to an embodiment of the present invention performs processing for sharpening processing of RI images such as SPECT images or PET images. In particular, a low count area is sharpened in an RI image in which a high count area with a high count value and a low count area with a low count value exist in one image. In the following embodiments, an RI image is described as an example of a medical image. However, the present invention can be applied to a CT image and an MRI image in addition to an RI image. The medical image processing apparatus of each embodiment described below is Both are configured by a general-purpose computer system including, for example, a processor and a memory, and individual components or functions in the medical image processing apparatus described below are realized by executing a computer program, for example. .

  FIG. 1 is a configuration diagram of a medical image processing apparatus 1 according to the first embodiment of the present invention. The medical image processing apparatus 1 includes an original image data storage unit 11, an unsharp mask processing unit 13, a post-processing unit 15, a post-processing image data storage unit 17, and an output unit 19.

  The original image data storage unit 11 stores RI original image data. In the present embodiment, for example, an original image having a high count area with a very high count value (pixel value) and a low count area with a relatively low count value is stored in the original image data storage unit 11. For example, the maximum count value in the original image may be 100 to 1000 times, 1000 to 10,000 times, or 10,000 times or more the count value of the detection range R described later.

  FIG. 2 shows a histogram of the count values of the original image stored in the original image data storage unit 11. In the figure, the horizontal axis indicates the count value and the vertical axis indicates the number of pixels. From this, it can be seen that in the original image, the number of pixels is very small, but there are pixels with a high count value. In the present embodiment, for example, attention is focused on a low-count pixel in the original image, and in particular, attention is focused on a pixel area within a predetermined low-count value range (detection range R). In the present embodiment, this focused area is sharpened by the processing described below.

  The unsharp mask processing unit 13 performs unsharp mask processing on the original image 30 to generate a sharpened image 70.

  FIG. 3 is a diagram showing a processing outline of the unsharp mask processing unit 13. Processing of the unsharp mask processing unit 13 will be described with reference to FIG.

  First, the unsharp mask processing unit 13 applies a plurality of different unsharp masks to the original image 30 to generate a plurality of blurred images. For example, the unsharp mask processing unit 13 applies a plurality of unsharp masks having different sizes and applies each to the original image 30 to perform a smoothing process, and smoothed images 40 (41, 42) that are blurred images. ,... 4N) are generated.

  The unsharp mask may be an odd number and the same number of vertical and horizontal square masks, such as 3 × 3 pixels, 5 × 5 pixels,. The unsharp mask processing unit 13 performs smoothing by assigning the average value of the pixels in each unsharp mask to the center pixel, and generates a plurality of smoothed images 40. In the case of FIG. 3, the smoothed images 41, 42,...

  Then, the unsharp mask processing unit 13 includes the difference image 50 (51, 52,..., 5N) between the original image 30 and one of the plurality of smoothed images 40 and the plurality of smoothed images 40. A plurality of difference images 50 between any two images are generated. A sharpened image 70 is generated based on the original image 30 and the plurality of difference images 50. For example, the unsharp mask processing unit 13 generates a difference image 51 that is a difference between the original image 30 and the smoothed image 41. Further, the unsharp mask processing unit 13 generates a total of N difference images 50 such as a difference image 52 that is a difference between the smoothed image 41 and the smoothed image 42.

  The unsharp mask processing unit 13 combines the difference images 51 to 5N to generate the contrast enhanced image 60. When combining the difference images 51 to 5N, the unsharp mask processing unit 13 may add each of the difference images 51 to 5N after being multiplied by a predetermined coefficient.

  The unsharp mask processing unit 13 adds the original image 30 to the contrast enhanced image 60 to generate a sharpened image 70.

  Reference is again made to FIG. The post-processing unit 15 is masking means for masking high count pixels having a count value larger than a predetermined threshold for the sharpened image 70 as post-processing of unsharp mask processing. For example, the post-processing unit 15 assigns and normalizes the threshold T to pixels having a count value equal to or greater than the threshold T among the pixels of the sharpened image 70, and removes the high count region. Then, the post-processing unit 15 generates a post-processed image 80 in which the high count area is normalized as an output image. The post-processing unit 15 stores the data of the post-process image 80 in the post-process image data storage unit 17. The threshold T may be a value slightly larger than the detection range R, for example, about 1.1 to 1.5 times the maximum value of the detection range R. For example, if the detection range R is a range of 20 to 90 counts, the threshold T may be about 100 counts.

  The output unit 19 refers to the post-process image data storage unit 17 and outputs a post-process image 80. The output unit 19 may be, for example, a printer or a display device.

  As a result, it is possible to sharpen a particularly low count area of an RI image in which a high count area with a high count value and a low count area with a low count value exist in one image.

  Next, a second embodiment of the present invention will be described.

  FIG. 4 is a configuration diagram of the medical image processing apparatus 2 according to the second embodiment. In the second embodiment, differences from the first embodiment will be mainly described, and components or functions common to the first embodiment may be denoted by the same reference numerals and description thereof may be omitted. .

  The medical image processing apparatus 2 according to the present embodiment includes an original image data storage unit 11, an image separation processing unit 21, an unsharp mask processing unit 13, an image composition unit 23, a composite image data storage unit 25, and an output. Part 19.

  The image separation processing unit 21 separates the original image 30 into a high count image 110 including a high count area and a low count image 120 including a low count area.

  With reference to FIG. 5, the image separation processing by the image separation processing unit 21 will be described. As illustrated in FIG. 5A, the image separation processing unit 21 obtains a difference between the original image 30 and the mask image 130 and generates a high count image 110. That is, the high count image 110 is an image having only pixels whose pixel value in the original image 30 has a value larger than the threshold value of the mask image. In the high count image 110, the pixel values of other pixels, that is, pixels whose pixel value in the original image 30 is equal to or less than the threshold value of the mask image 130 are set to predetermined values (for example, a threshold value or zero).

  On the other hand, the low count image 120 is a difference image obtained by subtracting the high count image 110 from the original image 30, as shown in FIG. 5B. That is, the low count image 120 is an image having only pixels whose pixel value in the original image 30 is equal to or less than the threshold value of the mask image 130. In the low count image 120, pixel values of other pixels, that is, pixels having a pixel value in the original image 30 that is larger than the threshold value of the mask image 130 are set to predetermined values (for example, a threshold value or zero).

  Note that the threshold value used in the post-processing unit 15 of the first embodiment and the threshold value of the mask image 130 may be the same or different.

  Reference is again made to FIG. The unsharp mask processing unit 13 performs the above-described unsharp mask processing on the low count image 120 to generate a low count sharpened image 140.

  The image composition unit 23 synthesizes the low count image 120 and the low count sharpened image 140 to generate a composite image 150. For example, the image composition unit 23 adds the pixel value of the corresponding low count image 120 and the pixel value of the low count sharpened image 140 to generate the composite image 150. When the image composition unit 23 generates the composite image 150, the pixel value of the high count image 110 may be multiplied by a predetermined coefficient and then added to the pixel value of the low count sharpened image 140. The image composition unit 23 stores the generated composite image 150 data in the composite image data storage unit 25.

  The output unit 19 refers to the composite image data storage unit 25 and outputs a composite image 150.

  Next, a third embodiment of the present invention will be described.

  FIG. 6 is a configuration diagram of the medical image processing apparatus 3 according to the third embodiment. In the third embodiment, differences from the first embodiment or the second embodiment will be mainly described, and the same constituent elements or functions as those in the first embodiment or the second embodiment are the same. A description may be omitted with reference numerals.

  The medical image processing apparatus 3 according to the present embodiment includes an original image data storage unit 11, a preprocessing unit 210, an unsharp mask processing unit 13, an image composition unit 250, an image separation processing unit 21, and a compression processing unit. 240, an image composition unit 250, a composite image data storage unit 260, and an output unit 19.

  The preprocessing unit 210 performs preprocessing on the original image 30 to generate a preprocessed image 310. The preprocessing unit 210 includes an ROI setting unit 212 and a normalization unit 214 for preprocessing.

  The ROI setting unit 212 sets an ROI (Region of Interest) for the original image 30. The ROI setting unit 212 sets, for example, an ROI for a predetermined area designated by the user or an automatically extracted processing area. For example, when the original image 30 is an RI image of a region including a region where a subject is injected with RI, the count value of the injected region is very high, but this is irrelevant to the disease and is a nearby lesion site. May be buried. In such a case, the ROI setting unit 212 sets the ROI for such a high count region that is clearly unrelated to the disease.

  The normalizing unit 214 normalizes (masks) the pixel value in the ROI region to a predetermined value (for example, 0). An image obtained by normalizing the pixel values in the ROI region by the normalizing unit 214 is the preprocessed image 310.

  The unsharp mask processing unit 13 performs the same processing as the processing described in FIG. That is, in the present embodiment, the unsharp mask processing unit 13 performs processing as shown in FIG. 3 using the preprocessed image 310. However, in this embodiment, the unsharp mask processing unit 13 outputs a contrast-enhanced image 320 instead of a sharpened image based on the preprocessed image 310. That is, in this embodiment, the unsharp mask processing unit 13 functions as a contrast-enhanced image generation unit.

  The image separation processing unit 21 separates the original image 30 into a high count image 110 and a low count image 120.

  The compression processing unit 240 generates a high count compressed image 340 by compressing the high count image 110. For example, the compression processing unit 240 divides the total count value of the high count image 110 by a predetermined value to compress the count value. For example, if the maximum value of the high count image 110 is K and the image separation threshold in the image separation processing unit 21 is k, the compression processing unit 240 sets the total count value of the high count image 110 to (k / K). The high count compressed image 340 is generated by compressing to about twice.

  The image composition unit 250 generates a composite image based on the contrast enhanced image 320, the high count compressed image 340, and the low count image 120. That is, the image composition unit 250 adds the corresponding pixel values of the contrast enhanced image 320, the high count compressed image 340, and the low count image 120, respectively, to obtain a composite image. When the image synthesizing unit 250 synthesizes the images, the image may be synthesized after multiplying any one or more of the contrast-enhanced image 320, the high count compressed image 340, and the low count image 120 by a predetermined coefficient. . The image data of the composite image generated by the image composition unit 250 is stored in the composite image data storage unit 260.

  The output unit 19 outputs an output image based on the composite image data stored in the composite image data storage unit 260.

  The above-described embodiments of the present invention are examples for explaining the present invention, and are not intended to limit the scope of the present invention only to those embodiments. Those skilled in the art can implement the present invention in various other modes without departing from the gist of the present invention.

  In the above-described embodiment, the case where the present invention is applied to a two-dimensional image (including a two-dimensional image cut out from a three-dimensional image) has been described as an example. However, the present invention is not cut out into a two-dimensional image, It can also be applied directly to a three-dimensional image.

1 is a configuration diagram of a medical image processing apparatus 1 according to a first embodiment of the present invention. The histogram of the count value of an original image is shown. It is a figure which shows the process outline | summary of the unsharp mask process part. It is a block diagram of the medical image processing apparatus 2 which concerns on the 2nd Embodiment of this invention. An image separation process performed by the image separation processing unit 21 will be described. It is a block diagram of the medical image processing apparatus 3 which concerns on the 3rd Embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1, 2 Image processing apparatus 11 Original image data memory | storage part 13 Unsharp mask process part 15 Post process part 17 Post process image data memory part 19 Output part 21 Image separation process part 23 Image composition part 25 Composite image data memory part 30 Original image 60 Contrast-enhanced image 70 Sharpened image 80 Post-processed image 110 High-count image 120 Low-count image 130 Mask image 140 Low-count processed image 150 Composite image 210 Pre-processing unit 212 ROI setting unit 214 Normalization unit 240 Compression processing unit 250 Image composition Unit 260 Composite Image Data Storage Unit

Claims (9)

ROI setting means for setting a ROI (Region of Interest) for a medical image obtained by imaging a subject;
Normalization means for normalizing pixel values in the ROI set by the ROI setting means to a predetermined value for the medical image, and generating a preprocessed image;
A contrast-enhanced image generating unit that performs an unsharp mask process on the preprocessed image generated by the normalizing unit and outputs a contrast-enhanced image generated in the course of the unsharp mask process;
Image separating means for separating the medical image into a high count image having a pixel value larger than a predetermined threshold and a low count image having a pixel value smaller than the threshold;
Compression processing means for compressing pixel values of the high count image to generate a high count compressed image;
A medical image processing apparatus comprising: an image composition unit configured to generate a composite image based on the contrast-enhanced image, the high-count compressed image, and the low-count image. The contrast-enhanced image generating means includes
Applying a plurality of different unsharp masks to the preprocessed image to generate a plurality of blurred images,
Generating a plurality of difference images between the preprocessed image and any one of the plurality of blurred images, and a difference image between any two images of the plurality of blurred images;
The medical image processing apparatus according to claim 1, wherein the contrast-enhanced image is generated based on the plurality of difference images. The plurality of unsharp masks are N types of unsharp masks having different sizes from each other, and are the first, second,..., Nth unsharp masks in order from the smallest size,
The contrast-enhanced image generating means includes
Applying an m-th (1 ≦ m ≦ N) unsharp mask to the preprocessed image to generate an m-th blurred image;
A first difference image that is a difference between the first blurred image and the preprocessed image, and a difference between the mth (2 ≦ m ≦ N) blurred image and the (m−1) th blurred image. A certain m-th difference image is generated,
The medical image processing apparatus according to claim 2, wherein the contrast-enhanced image is generated by adding the first to Nth difference images. The contrast-enhanced image generating means includes
Assigning an average value of all pixels in the m-th (1 ≦ m ≦ N) unsharp mask as a value of one pixel in each unsharp mask to generate the m-th blurred image. The medical image processing apparatus according to claim 3, wherein: The compression processing means includes
5. The medical image processing apparatus according to claim 1, wherein compression is performed at a compression rate corresponding to a ratio between a maximum value of the high count image and the threshold value. 6. An image separating means for separating a medical image obtained by photographing a subject into a high count image having a pixel value larger than a predetermined threshold and a low count image having a pixel value smaller than the threshold;
An unsharp mask processing means for applying an unsharp mask process to the low count image to generate a low count sharpened image;
An apparatus for processing a medical image, comprising: an image compositing unit that generates a composite image based on the low count sharpened image and the high count image. Unsharp mask processing means for generating a sharpened image by performing unsharp mask processing on a medical image obtained by photographing a subject,
A post-processing unit that normalizes a high count pixel having a pixel value larger than a predetermined threshold with respect to the sharpened image generated by the unsharp mask processing unit to generate a post-processed image;
A medical image processing apparatus comprising: output means for outputting a post-processed image generated by the post-processing means. ROI setting means for setting an ROI for a medical image obtained by photographing a subject;
Normalizing pixel values in the ROI set by the ROI setting means to a predetermined value for the medical image, and generating a preprocessed image;
A step of outputting a contrast-enhanced image generated in the process when unsharp mask processing is applied to the generated preprocessed image;
Separating the medical image into a high count image having a pixel value greater than a predetermined threshold and a low count image having a pixel value less than the threshold;
Compressing pixel values of the high count image to generate a high count compressed image;
A method for processing a medical image, comprising: generating a composite image based on the contrast-enhanced image, the high-count compressed image, and the low-count image. A computer program for performing image processing of a medical image obtained by photographing a subject,
ROI setting means for setting an ROI for a medical image;
Normalizing pixel values in the ROI set by the ROI setting means to a predetermined value for the medical image, and generating a preprocessed image;
A step of outputting a contrast-enhanced image generated in the process when unsharp mask processing is applied to the generated preprocessed image;
Separating the medical image into a high count image having a pixel value greater than a predetermined threshold and a low count image having a pixel value less than the threshold;
Compressing pixel values of the high count image to generate a high count compressed image;
A computer program for causing a computer to execute a step of generating a composite image based on the contrast-enhanced image, the high-count compressed image, and the low-count image.

Images