阅读说明：本技术 医用图像处理装置以及医用图像处理方法 (Medical image processing apparatus and medical image processing method ) 是由 高桥悠 后藤大雅 于 2020-06-25 设计创作，主要内容包括：本发明提供一种医用图像处理装置以及医用图像处理方法,即使是所含的噪声多的医用图像,也能够在维持构造物的边界的同时降低噪声。医用图像处理装置降低所取得的医用图像的噪声,该医用图像处理装置具备：平滑化部,其生成所述医用图像的平滑化图像；路线生成部,其生成连续位于所述平滑化图像中的像素群即满足路线条件的路线；以及噪声降低部,其在所述医用图像中提取与所述路线对应的像素群,基于提取出的像素群来降低所述医用图像的噪声。(The invention provides a medical image processing apparatus and a medical image processing method, which can reduce noise while maintaining the boundary of a structure even if the medical image contains much noise. A medical image processing apparatus for reducing noise in an acquired medical image, the medical image processing apparatus comprising: a smoothing unit that generates a smoothed image of the medical image; a route generation unit that generates a route satisfying a route condition, which is a group of pixels continuously located in the smoothed image; and a noise reduction unit that extracts a pixel group corresponding to the route from the medical image and reduces noise in the medical image based on the extracted pixel group.)

1. A medical image processing apparatus for reducing noise in an acquired medical image,

the medical image processing apparatus includes:

a smoothing unit that generates a smoothed image of the medical image;

a route generation unit that generates a route satisfying a route condition, which is a group of pixels continuously located in the smoothed image; and

and a noise reduction unit that extracts a pixel group corresponding to the route from the medical image and reduces noise in the medical image based on the extracted pixel group.

2. The medical image processing apparatus according to claim 1,

the route generation unit sets the route condition according to the smoothing intensity of the smoothing unit.

3. The medical image processing apparatus according to claim 2,

the route condition is an upper limit value of a difference between a pixel value of a start point pixel of the route and a pixel value of a pixel of interest,

the route generation unit sets the upper limit value to be smaller as the smoothing intensity is larger.

4. The medical image processing apparatus according to claim 2,

the route condition is an upper limit value of the number of consecutive pixels of the route,

the route generation unit sets the upper limit value to be smaller as the smoothing intensity is larger.

5. The medical image processing apparatus according to claim 1,

the smoothing unit generates the smoothed image by smoothing the medical image when the type of noise reduced by the noise reduction unit is granular noise,

when the type of noise reduced by the noise reduction unit is linear noise, the smoothing unit obtains smoothed projection data by smoothing projection data used for generating the medical image, and generates the smoothed image by reconstructing the smoothed projection data.

6. The medical image processing apparatus according to claim 1,

the route generation unit sets a starting point pixel selected from the smoothed image as a target pixel, and then repeats setting of a pixel having a smallest difference in pixel value from the target pixel as a new target pixel from among a pixel group adjacent to the target pixel until a route condition is not satisfied, thereby generating a pixel group from the starting point pixel to the latest target pixel as the route.

7. The medical image processing apparatus according to claim 6,

the route generation unit selects all pixels in the smoothed image as the start point pixels, and generates the route for each of the start point pixels.

8. The medical image processing apparatus according to claim 1,

the noise reduction unit calculates a representative value of the pixel values of the extracted pixel group, and replaces the pixel value of a pixel corresponding to the starting point pixel of the route in the medical image with the representative value.

9. The medical image processing apparatus according to claim 8,

the representative value is any one of an average value and a median value of pixel values of the extracted pixel group, and a weighted sum value calculated using a weighting coefficient smaller than 1, the weighting coefficient being larger as a distance from the start pixel is shorter.

10. A medical image processing method for reducing noise in an acquired medical image,

the medical image processing method includes:

a smoothing step of generating a smoothed image of the medical image;

a route generation step of generating a route satisfying a route condition as a group of pixels continuously located in the smoothed image; and

and a noise reduction step of extracting a pixel group corresponding to the route from the medical image, and reducing noise of the medical image based on the extracted pixel group.

Technical Field

The present invention relates to a medical image processing apparatus and a medical image processing method for processing a medical image obtained by a medical imaging apparatus such as an X-ray CT (Computed Tomography) apparatus, and relates to a technique for reducing noise while maintaining a boundary of a structure included in the medical image.

Background

An X-ray CT apparatus, which is an example of a medical imaging apparatus, irradiates X-rays from the periphery of a subject to acquire projection data at a plurality of projection angles, and performs back-projection processing on the projection data to generate a tomographic image, i.e., a so-called reconstructed image, of the subject. The generated reconstructed image is used as a medical image for diagnosis of the subject. Since noise included in medical images hinders diagnosis, various methods for reducing noise have been developed.

Patent document 1 discloses that a group of pixels is obtained on a previously set route formed by pixels consecutive from the start position of an adjacent pixel of a target pixel until an end condition is not satisfied, and noise is removed from the target pixel using the obtained group of pixels. According to patent document 1, noise can be reduced regardless of the presence or absence of a structure.

However, in patent document 1, for example, in a reconstructed image generated by low-dose X-ray irradiation in order to reduce exposure, a pixel group until the end condition is not satisfied may not be accurately obtained for a medical image containing a large amount of noise. In this case, the boundary of the structure cannot be maintained, or the noise reduction is insufficient.

Patent document 1: japanese patent No. 4688269

Disclosure of Invention

Therefore, an object of the present invention is to provide a medical image processing apparatus and a medical image processing method that can reduce noise while maintaining the boundary of a structure even in a medical image containing a large amount of noise.

In order to achieve the above object, the present invention relates to a medical image processing apparatus for reducing noise in an acquired medical image, the medical image processing apparatus including: a smoothing unit that generates a smoothed image of the medical image; a route generation unit that generates a route satisfying a route condition, which is a group of pixels continuously located in the smoothed image; and a noise reduction unit that extracts a pixel group corresponding to the route from the medical image and reduces noise in the medical image based on the extracted pixel group.

The present invention also relates to a medical image processing method for reducing noise in an acquired medical image, the medical image processing method including: a smoothing step of generating a smoothed image of the medical image; a route generation step of generating a route satisfying a route condition as a group of pixels continuously located in the smoothed image; and a noise reduction step of extracting a pixel group corresponding to the route from the medical image, and reducing noise of the medical image based on the extracted pixel group.

According to the present invention, it is possible to provide a medical image processing apparatus and a medical image processing method capable of reducing noise while maintaining the boundary of a structure even in a medical image containing a large amount of noise.

Drawings

Fig. 1 is an overall configuration diagram of a medical image processing apparatus.

Fig. 2 is an overall configuration diagram of an X-ray CT apparatus as an example of a medical imaging apparatus.

Fig. 3 is a functional block diagram of embodiment 1.

Fig. 4 shows an example of the flow of the processing of example 1.

Fig. 5 shows an example of the flow of the smoothing step in fig. 4.

Fig. 6 shows another example of the flow of the smoothing step in fig. 4.

Fig. 7 shows an example of the flow of the route generation step in fig. 4.

Fig. 8 shows an example of the flow of the noise reduction step of fig. 4.

Fig. 9 shows that the result of route generation varies due to noise.

Fig. 10 shows that the normal direction of the contour line does not change due to smoothing.

Fig. 11 shows contours of a smoothed image of a noise-added image.

Fig. 12 shows an example of the effect of embodiment 1.

Detailed Description

Hereinafter, embodiments of a medical image processing apparatus and a medical image processing method according to the present invention will be described with reference to the drawings. In the following description and the drawings, the same reference numerals are given to the components having the same functional configurations, and redundant description thereof is omitted.

[ example 1 ]

Fig. 1 shows a hardware configuration of a medical image processing apparatus 1. The medical image Processing apparatus 1 is configured by connecting a CPU (Central Processing Unit) 2, a main memory 3, a storage device 4, a display memory 5, a display device 6, a controller 7, an input device 8, and a network adapter 10 via a system bus 11 so as to be capable of transmitting and receiving signals. The medical image processing apparatus 1 is connected to a medical image capturing apparatus 13 and a medical image database 14 via a network 12 so as to be capable of transmitting and receiving signals. Here, "receivable/receivable signal" indicates a state in which signals can be transmitted/received to/from each other or from one side to the other side electrically or optically, whether wired or wireless.

The CPU2 controls the operations of the components. The CPU2 loads and executes the program stored in the storage device 4 and data necessary for program execution into the main memory 3. The storage device 4 is a device that stores programs executed by the CPU2 and data necessary for the execution of the programs, and specifically, is a Hard Disk Drive (HHD) or a Solid State Drive (SSD) or the like. Various data are transmitted and received via a Network 12 such as a Local Area Network (LAN). The main memory 3 stores intermediate passes of programs and arithmetic processing executed by the CPU 2.

The display memory 5 temporarily stores display data for display in a display device 6 such as a liquid crystal display. The input device 8 is an operation device for an operator to give an operation instruction to the medical image processing apparatus 1, and specifically includes a keyboard, a mouse, a touch panel, and the like. The mouse may be a track pad or other pointing device such as a trackball. The controller 7 detects the state of the mouse, acquires the position of the mouse pointer on the display device 6, and outputs the acquired position information and the like to the CPU 2. The network adapter 10 is used to connect the medical image processing apparatus 1 to a network 12 such as a LAN, a telephone line, and the internet.

The medical imaging apparatus 13 is an apparatus that acquires a medical image such as a tomographic image of a subject. The medical imaging apparatus 13 is, for example, an X-ray CT apparatus, and will be described later with reference to fig. 2. The medical image database 14 is a database system that stores medical images acquired by the medical imaging apparatus 13.

The overall configuration of an X-ray CT apparatus 100 as an example of the medical imaging apparatus 13 will be described with reference to fig. 2. In fig. 2, the lateral direction is an X axis, the vertical direction is a Y axis, and the direction perpendicular to the paper surface is a Z axis. The X-ray CT apparatus 100 includes a scanner 200 and an operation unit 250. The scanner 200 includes an X-ray tube 211, a detector 212, a collimator 213, a drive unit 214, a central control unit 215, an X-ray control unit 216, a high voltage generation unit 217, a scanner control unit 218, a bed control unit 219, a collimator control unit 221, a preamplifier 222, an a/D converter 223, a bed 240, and the like.

The X-ray tube 211 is a device that irradiates the subject 210 placed on the bed 240 with X-rays. The X-ray tube 211 is irradiated with X-rays from the X-ray tube 211 by applying a high voltage generated by the high voltage generator 217 in accordance with a control signal transmitted from the X-ray controller 216 to the X-ray tube 211.

The collimator 213 is a device that limits the irradiation range of the X-rays irradiated from the X-ray tube 211. The X-ray irradiation range is set in accordance with a control signal transmitted from the collimator control unit 221.

The detector 212 is a device that measures the spatial distribution of the transmitted X-rays by detecting the X-rays transmitted through the object 210. The detector 212 is disposed to face the X-ray tube 211, and many detection elements are two-dimensionally arranged in a plane facing the X-ray tube 211. The signal measured by the detector 212 is amplified by the preamplifier 222 and then converted into a digital signal by the a/D converter 223. Then, various correction processes are performed on the digital signal to acquire projection data.

The drive unit 214 rotates the X-ray tube 211 and the detector 212 around the subject 210 in accordance with a control signal transmitted from the scanner control unit 218. The X-ray tube 211 and the detector 212 rotate and irradiate and detect X-rays, thereby acquiring projection data from a plurality of projection angles. The unit of data collection for each projection angle is called the view angle (view). Regarding the arrangement of the respective detection elements of the two-dimensionally arranged detector 212, the rotation direction of the detector 212 is referred to as a channel, and the direction orthogonal to the channel is referred to as a column. The projection data is identified by view, channel, column.

The bed controller 219 controls the movement of the bed 240, and keeps the bed 240 stationary or moves the bed 240 at a constant speed in the Z-axis direction while the X-ray irradiation and detection are performed. Scanning in which the bed 240 is kept stationary is called axial scanning, and scanning in which the bed 240 is moved is called helical scanning.

The central control unit 215 controls the operation of the scanner 200 described above in accordance with an instruction from the operation unit 250. Next, the operation unit 250 is explained. The operation unit 250 includes a reconstruction processing unit 251, an image processing unit 252, a storage unit 254, a display unit 256, an input unit 258, and the like.

The reconstruction processor 251 generates a reconstructed image by performing back projection processing on the projection data acquired by the scanner 200. The image processing unit 252 performs various image processing operations to convert the reconstructed image into an image suitable for diagnosis. The storage unit 254 stores projection data, a reconstructed image, and an image after image processing. The display unit 256 displays the reconstructed image and the image after the image processing. The input unit 258 is used when the operator sets conditions for acquiring projection data (tube voltage, tube current, scanning speed, and the like) and reconstruction conditions for reconstructing an image (reconstruction filter, FOV size, and the like).

Further, the operation unit 250 may be the medical image processing apparatus 1 shown in fig. 1. In this case, the image processing unit 252 corresponds to the CPU2, the storage unit 254 corresponds to the storage device 4, the display unit 256 corresponds to the display device 6, and the input unit 258 corresponds to the input device 8.

The important part of the present embodiment will be described with reference to fig. 3. These important components may be configured by dedicated hardware or may be configured by software running on the CPU 2. In the following description, a case where an important part of the present embodiment is constituted by software will be described.

The present embodiment includes a smoothing unit 301, a route generation unit 302, and a noise reduction unit 303. The storage device 4 stores a reconstructed image and projection data generated by the X-ray CT apparatus 100.

Hereinafter, each component will be described.

The smoothing unit 301 generates a smoothed image of a medical image, for example, a reconstructed image. In order to generate a smoothed image, the pixel value of a target pixel to be smoothed in a medical image is changed so as to be smoothed with respect to each pixel value in a pixel group adjacent to the target pixel. Most simply, an average value or a median value is calculated from the target pixel and the adjacent pixel group, and the pixel value of the target pixel is replaced by the calculated average value or median value. In addition, the pixel value of the target pixel and each pixel value in the pixel group around the target pixel may be subjected to weighted summation using a weighting coefficient smaller than 1, which increases as the distance from the target pixel decreases, and the pixel value of the target pixel may be replaced by the calculated weighted summation value. Further, the smoothed image may be generated by a known method other than replacing the pixel value of the target pixel by the average value, the center value, or the weighted sum value. For example, the smoothed image generated by the replacement of the average value may be weighted-summed with the original medical image using a weight coefficient smaller than 1. Alternatively, a noise reduction method based on a known successive approximation method or a noise reduction method using artificial intelligence may be used as the smoothing method.

The projection data used for generating the reconstructed image may be smoothed to calculate smoothed projection data, and the smoothed projection data may be subjected to reconstruction processing to generate a smoothed image. The projection data and the reconstructed image are read out from the storage device 4 or transmitted from the outside via the network adapter 10.

Further, whether the reconstructed image is smoothed to generate the smoothed image or the smoothed projection data calculated by smoothing the projection data is subjected to reconstruction processing and smoothed to generate the smoothed image may be selected according to the type of noise to be reduced by the noise reduction unit 303 described later. For example, when reducing granular noise due to statistical fluctuations of X-rays, it is preferable to smooth the reconstructed image, and when reducing linear noise, so-called texture artifacts, it is preferable to smooth the projection data. The type of noise to be reduced by the noise reduction unit 303 can be specified by an operator via the input device 8. The type of noise may be determined based on the contour shape of the object 210 and the positional deviation of a site having a large X-ray attenuation coefficient such as a bone. For example, if the contour shape is close to a perfect circle, it is determined that granular noise is a reduction target, and if the flatness ratio of the contour shape is high or the arrangement of bones is deviated in a specific direction, it is determined that linear noise is a reduction target.

The route generating unit 302 generates a route satisfying a route condition, which is a group of pixels continuously located in the smoothed image generated by the smoothing unit 301. More specifically, after the starting point pixel selected from the smoothed image is set as the target pixel, the pixel having the smallest difference from the pixel value of the target pixel is set as the new target pixel from the pixel group adjacent to the target pixel repeatedly until the route condition is not satisfied. Then, a pixel group from the start point pixel to the latest pixel of interest that is the pixel of interest when the route condition is not satisfied is generated as a route. In addition, since a new target pixel is set from among the pixel groups adjacent to the target pixel based on the difference in pixel value from the target pixel, the generated route is not set in advance.

The adjacent pixel group, which is a pixel group adjacent to the target pixel, may be set to 4 pixels located above, below, on the left, and on the right of the target pixel in the cross section of the medical image, or may be set to 8 pixels to which 4 pixels located in an oblique direction are further added. In addition, not only 4 pixels in the top, bottom, left, and right directions in the cross section, 6 pixels to which 2 pixels in the front and rear directions orthogonal to the cross section are added may be set as the adjacent pixel group, and 26 pixels to which 20 pixels in the oblique direction are further added may be set as the adjacent pixel group.

The route condition is an upper limit value of a difference between the pixel value of the start point pixel and the pixel value of the target pixel, an upper limit value of the number of continuous pixels in the route, that is, an upper limit value of the number of times of updating the target pixel, or the like. The route condition may be set according to the smoothing intensity of the smoothing unit 301. For example, the storage device 4 may store in advance a relationship between a smoothing intensity obtained from data obtained by imaging a phantom of a known structure and a route condition, and the route condition may be set by comparing the smoothing intensity of the smoothing unit 301 with the relationship.

In addition, in a smoothed image with a large smoothing intensity, the generated route may be expanded in the normal direction of the boundary of the structure. Therefore, the higher the smoothing intensity of the smoothing unit 301 is, the smaller the upper limit value of the difference between the pixel value of the start point pixel and the pixel value of the target pixel, which is the route condition, and the upper limit value of the number of pixels continuing the route can be set. By setting the upper limit value to be small, the spread of the route in the normal direction of the boundary of the structure can be suppressed.

The noise reduction unit 303 extracts a pixel group corresponding to the route generated in the smoothed image by the route generation unit 302 from the medical image, and reduces the noise of the medical image based on the extracted pixel group. More specifically, a pixel group having the same coordinates as the route of the smoothed image is extracted from the medical image, and the pixel value of the starting point pixel of the route in the medical image is replaced by a representative value calculated using the pixel values of the extracted pixel group. The representative value is, for example, an average value or a median value of pixel values of a pixel group extracted from the medical image, or a weighted sum value calculated using a weighting coefficient smaller than 1 whose distance from the starting point pixel is shorter. The noise reduction may be performed for the entire medical image, or may be performed for a partial region of the medical image. The medical image with reduced noise is displayed on the display device 6 and used for diagnosis of the subject 210.

An example of the flow of the processing executed in the present embodiment will be described with reference to fig. 4.

(S401)

The smoothing unit 301 generates a smoothed image of the medical image. The medical image and the smoothed image have the same number of pixels, and pixels having the same coordinates correspond to each other. Even when the noise included in the medical image is large, the noise is reduced in the smoothed image generated in this step.

Fig. 5 shows an example of the flow of the processing of this step, and the steps of fig. 5 will be described later. The medical image of fig. 5 is a reconstructed image generated by the X-ray CT apparatus 100.

(S501)

The smoothing unit 301 acquires a reconstructed image. The reconstructed image is read out from the storage device 4, or transmitted from the outside via the network adapter 10.

(S502)

The smoothing unit 301 smoothes the reconstructed image to generate a smoothed image. For example, the pixel value of the target pixel is replaced with the average value of the pixel values of the pixel group adjacent to the target pixel, thereby smoothing the reconstructed image.

Fig. 6 shows another example of the flow of the processing of S401, and the steps of fig. 6 will be described later. The medical image of fig. 6 is a reconstructed image generated from projection data acquired by the X-ray CT apparatus 100.

(S601)

The smoothing unit 301 acquires projection data used for generating a reconstructed image. The projection data is read out from the storage device 4, or the projection data is transmitted from the outside via the network adapter 10.

(S602)

The smoothing unit 301 smoothes the projection data to generate smoothed projection data. For example, the projection data is smoothed by replacing the projection value of the object data with the average value of the projection values of the data groups adjacent to the object data in the channel direction or the column direction in the projection data. Further, the projection value of the target data may be replaced with the calculated weighted sum value by weighted sum of the projection value of the target data and the respective projection values of the data group around the target data using a weighting coefficient smaller than 1, which increases as the distance from the target data decreases.

(S603)

The smoothing unit 301 generates a smoothed image by performing reconstruction processing on the smoothed projection data generated in S602.

The explanation returns to fig. 4.

(S402)

The route generation unit 302 generates a route in the smoothed image generated in S401. The route generated in the smoothed image is found more accurately than the route generated in the medical image that is not smoothed.

Fig. 7 shows an example of the flow of the processing of this step, and the steps of fig. 7 will be described later.

(S701)

The route generation unit 302 selects a start point pixel that is a pixel that becomes a start point in the smoothed image. The starting point pixel may be any pixel, and for example, a pixel at the upper left end is selected as the starting point pixel.

(S702)

The route generation unit 302 sets the start point pixel as the target pixel. The target pixel is a pixel to be processed later.

(S703)

The route generation unit 302 selects a most similar pixel that is most similar to the target pixel from an adjacent pixel group that is a pixel group adjacent to the target pixel. Specifically, a pixel having the smallest difference from the pixel value of the target pixel is selected from the adjacent pixel group.

(S704)

The route generation unit 302 sets the most similar pixel selected in S703 as a new pixel of interest. I.e. updating the pixel of interest.

(S705)

The route generation unit 302 determines whether or not the latest pixel of interest satisfies the route condition. When the route condition is not satisfied, the process proceeds to S706, and when the route condition is satisfied, the process returns to S703. In addition, when the processing returns to S703, the last attention pixel is not included in the most similar pixels selected in S703. Since the most similar pixels do not include the previous attention pixel, the generated route can be prevented from going back and forth between the same pixels.

(S706)

The route generation unit 302 generates a route from the start pixel to the latest pixel of interest. The route generated by connecting the most similar pixels selected in S703 includes the boundary of the structure.

(S707)

The route generation unit 302 determines whether or not all pixels of the smoothed image are set as the starting pixels. When all the pixels have been set as the start point pixels, the flow of the processing ends, and when there are pixels that have not become the start point pixels, the processing proceeds to S708. That is, the route is generated for all pixels. When noise reduction is performed on a partial region of a medical image, it is determined whether or not all pixels included in the region have become starting pixels.

(S708)

The route generation unit 302 updates the start point pixel by selecting an arbitrary pixel from a pixel group that has not yet become the start point pixel. After the start point pixel is updated, the process returns to S702.

The explanation returns to fig. 4.

(S403)

The noise reduction unit 303 reduces the noise of the medical image based on the route generated in the smoothed image in S402. By reducing the noise of the medical image based on the route determined more accurately, the noise can be reduced while maintaining the boundary of the structure.

Fig. 8 shows an example of the flow of the processing of this step, and the steps of fig. 8 will be described later.

(S801)

The noise reduction unit 303 selects one of the routes generated in the smoothed image, and extracts the pixel value of the pixel of the coordinate corresponding to the selected route from the medical image. If the number of pixels on the route is N, N pixel values are extracted from the medical image.

(S802)

The noise reduction unit 303 calculates a representative value of the pixel values extracted in S801. The representative value is, for example, an average value or a median value of the extracted pixel values, a weighted sum value calculated using a weight coefficient smaller than 1, the distance from the start pixel being shorter, or the like.

(S803)

The noise reduction unit 303 replaces the pixel value of the pixel of the coordinate corresponding to the start point pixel of the route in the medical image with the representative value calculated in S802. Noise of a medical image is reduced by replacing a pixel value of a starting point pixel in the medical image with a representative value. The pixel value of the start pixel in the medical image may be sequentially replaced by the representative value, or the representative value may be embedded in a storage area prepared separately from the medical image.

(S804)

The noise reduction unit 303 determines whether all the routes generated in the smoothed image have been processed. When all the routes have been processed, the processing flow ends, and when there is a route that has not been processed, the process proceeds to S805.

(S805)

The noise reduction unit 303 updates the route by selecting an arbitrary route from routes that have not been processed yet. After updating the route, the process returns to S801.

With the above-described processing flow, even in a medical image containing a large amount of noise, the noise of the medical image is reduced based on the route generated in the smoothed image, and therefore, the noise can be reduced while maintaining the boundary of the structure.

Effects of the present embodiment will be described with reference to fig. 9 to 12. First, the change in the result of route generation due to noise will be described with reference to fig. 9. Fig. 9 (a) is a simulated image in which two types of circular structures having different pixel values are arranged on a computer, and fig. 9 (b) is an image in which contour lines in which pixels having the same pixel value are connected by lines on the simulated image are superimposed on the simulated image. Fig. 9 (c) is a noise-added image in which random noise is added to the simulation image, and fig. 9 (d) is an image in which contour lines are superimposed on the noise-added image. The route of the present embodiment is generated roughly along the contour.

The contour lines in fig. 9 (d) do not coincide with the boundaries of the structure, and are scattered in the interior of the structure, for example, with respect to the contour lines in fig. 9 (b) along the boundaries of the structure. That is, while noise can be reduced while maintaining the boundary of the structure in fig. 9 (a) with less random noise, noise can be reduced in fig. 9 (c) with more random noise, but the boundary of the structure is blurred.

The following describes, with reference to fig. 10, that the normal direction of the contour line does not change due to smoothing. Fig. 10 (a) is a simulated image similar to fig. 9 (a), fig. 10 (b) is an image in which a contour line and an arrow indicating the normal direction of the contour line are superimposed on the simulated image, fig. 10 (c) is a smoothed image of the simulated image, and fig. 10 (d) is an image in which a contour line and an arrow indicating the normal direction of the contour line are superimposed on the smoothed image.

As is clear from comparison between fig. 10 (b) and 10 (d), the intervals between the contour lines are widened by smoothing the simulated image. On the other hand, it is found that the normal direction of the contour lines coincides between fig. 10 (b) and fig. 10 (d), and that even if the simulated image is smoothed, the route along the structure boundary can be generated.

Contours of the smoothed image of the noise-added image will be described with reference to fig. 11. Fig. 11 (a) is a noise-added image similar to fig. 9 (c), fig. 11 (b) is a smoothed image of the noise-added image, and fig. 11 (c) is an image in which contours are superimposed on the smoothed image of fig. 11 (b).

In fig. 11 (c), the contour lines do not match the contour lines of the boundary of the structure as in fig. 9 (d), but follow the boundary of the structure as in fig. 9 (b). That is, by using the route generated in the smoothed image of the noise-added image, the noise can be reduced while maintaining the boundary of the structure.

The effects of the present embodiment will be described with reference to fig. 12. Fig. 12 (a) shows a noise-added image similar to fig. 9 (c), fig. 12 (b) shows an image in which noise is reduced from the noise-added image by the method of patent document 1, and fig. 12 (c) shows an image in which noise is reduced from the noise-added image by the present embodiment. Fig. 12 (d) is a difference image between fig. 12 (a) and 12 (b), and fig. 12 (e) is a difference image between fig. 12 (a) and 12 (c).

When fig. 12 (b) is compared with fig. 12 (c), SD (Standard Deviation) values of pixel values, which are one of the noise indexes, are almost equal to each other, and both have almost equal noise reduction effects. On the other hand, when fig. 12 (d) is compared with fig. 12 (e), in fig. 12 (d), a portion having a large difference in pixel value from the surrounding portion exists in the boundary of the structure, whereas such a portion is not visible in fig. 12 (e). A region having a large difference from the surrounding pixel values is a result of blurring of the boundary of the structure due to noise reduction. That is, in the present embodiment, it is shown by (e) of fig. 12 that noise can be reduced while maintaining the boundary of the structure, as compared with the method of patent document 1 in which noise is reduced while blurring the boundary of the structure.

The medical image processing apparatus and the medical image processing method according to the present invention are not limited to the above-described embodiments, and the components may be modified and embodied without departing from the scope of the invention. Further, a plurality of constituent elements disclosed in the above embodiments may be appropriately combined. Further, some of the components may be deleted from all the components shown in the above embodiments.

Reference numerals:

1: medical image processing apparatus, 2: CPU, 3: main memory, 4: storage device, 5: display memory, 6: display device, 7: controller, 8: input device, 10: network adapter, 11: system bus, 12: network, 13: medical imaging apparatus, 14: medical image database, 100: x-ray CT apparatus, 200: scanner, 210: subject, 211: x-ray tube, 212: detector, 213: collimator, 214: drive unit, 215: central control unit, 216: x-ray control unit, 217: high voltage generation unit, 218: scanner control unit, 219: bed control unit, 221: collimator control unit, 222: preamplifier, 223: a/D converter, 240: bedding, 250: operation unit, 251: reconstruction processing unit, 252: image processing unit, 254: storage unit, 256: display unit, 258: input unit, 301: smoothing unit, 302: route generation unit, 303: a noise reduction unit.