JP2005028037A - Medical image processing device and medical image processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a medical image processing device and a medical image processing method capable of automatically detecting the existing direction of the chest wall in a mammography image. <P>SOLUTION: This medical image processing device is equipped with an interface 46 for inputting an image data and image supplemental information obtained by photographing for diagnostic purpose using a prescribed communication protocol, and an image rotation information producing means 41a for producing image rotation information expressing the direction of a medical image expressed by the inputted image data based on the kind of the communication protocol used in the interface or the inputted image supplemental information. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a medical image processing apparatus and a medical image processing method for processing image data obtained by imaging for diagnostic purposes such as radiography to display or output an image.

  Conventionally, imaging methods using radiation (X-rays, α-rays, β-rays, γ-rays, electron beams, ultraviolet rays, etc.) have been used in various fields, and especially in the medical field, it is the most important for diagnosis. It has become one of the means. Since the first X-ray photography was realized, the X-ray photography method has been improved many times, and at present, a method combining a fluorescent screen and an X-ray film has become mainstream. On the other hand, in recent years, various digitized apparatuses such as X-ray CT, ultrasound, and MRI have been put into practical use, and the construction of diagnostic information processing systems and the like in hospitals is being promoted. Much research has been done to digitize X-ray images, but a radiographic method using a stimulable phosphor has been established and is attracting attention as a replacement for conventional X-ray photography.

  A stimulable phosphor (storable phosphor) is a part of its radiation energy that is stored when irradiated with radiation. After that, when it is irradiated with excitation light such as visible light, it emits light according to the stored energy. It has been known for a long time. A radiographic method using this is to capture and record a radiographic image of a subject such as a human body on a sheet coated with a photostimulable phosphor using an image taking device, and use the image reading device to capture the photostimulable fluorescence. When the body sheet is scanned with excitation light such as laser light, stimulated emission light is generated. Image data is obtained by photoelectrically reading this light. Furthermore, after appropriately processing this image data using an image processing apparatus, the image data can be output to a display such as a CRT, or printed on a film by a laser printer or the like to display a radiation image as a visible image.

  Such a radiographic method has performance comparable to that of a conventional X-ray photography method in terms of imaging sensitivity and image quality. For example, compared with the conventional X-ray photography method, the exposure area is extremely wide, and the response of the photostimulated light to the exposure amount is almost proportional over the entire exposure area. For this reason, no matter what radiation dose the subject is photographed, it is possible to capture the emission region where the image exists and normalize it into a digital signal without excess or deficiency. Further, by combining the signal obtained in this way with an appropriate image processing method, it is possible to provide a steady and high-quality image even under various shooting conditions. Furthermore, since the digitized image information is obtained directly, not only can a large amount of data be stored for a long period of time without causing image deterioration, but also development into a medical diagnosis information system is possible. Become.

  The radiographic method as described above is also applied to mammography (mammography) in breast cancer screening and the like. In general, to improve the efficiency of medical diagnosis, a pair of left and right mammograms (breast) images obtained by mammography are arranged symmetrically with the ends of the mammo regions back to back and are displayed on a display screen or printed on film. The However, there is a problem that it is necessary to manually specify the presence direction of the chest wall for each mammo image in order to align the mammo region ends.

  Further, a high-luminance white area may occur in the mammo image. FIG. 23 shows a radiographic image of mammography. As shown in FIG. 23, a white area 101 is generated together with a radiographic image of the breast area 100.

  The cause of the occurrence of such a blank area in the mammography radiographic image is as follows. An image reading apparatus that reads a radiation image by conveying a sheet coated with a photostimulable phosphor may cause sheet conveyance deviation. In order to prevent the image data of the entire desired image area (for example, the breast area 100) from being obtained due to the conveyance deviation, the read start timing is delayed and adjustment is performed so that the image data of the entire desired image area is obtained. Done. However, by delaying the reading start timing, a situation in which a portion other than the sheet is read occurs, and the portion becomes a transparent blank area 101 when visualized on the film. In particular, in the mammo image, in order to obtain all the image data up to the end of the mammo region, a portion other than the sheet in the direction in which the chest wall exists is a blank region. When light is applied from behind the film, the white areas 101 become high in luminance, and there is a problem in that the visibility of the image near the chest wall is lowered and eye fatigue is caused.

  Therefore, conventionally, a blackening process is performed to reduce the luminance of the pixels in the white areas and make the areas black. However, in order to perform the blackening process on the white area, there is a problem that it is necessary to manually specify the direction in which the chest wall exists in the mammo image.

As a related technique, the following Patent Document 1 describes a breast image display device that displays a pair of left and right breast images in symmetrical positions and improves diagnostic efficiency.
According to this breast image display device, the center position of the breast region ends on the pair of left and right breast images is detected, the sides of the breast region ends of the left and right breast images face each other, and each center position is determined. Since they are displayed together, the breast regions of the left and right breast images can be symmetrically arranged and displayed on the display means, and the comparative diagnosis efficiency of the left and right breasts can be improved. However, in Patent Document 1, the designation of the direction in which the chest wall exists is not described.

Patent Document 2 below describes a computer-aided image diagnostic apparatus capable of providing an image with improved observation interpretation performance for diagnosis.
According to this computer-aided image diagnostic apparatus, it is easy to understand the position of the abnormal shadow in the entire image by simultaneously displaying the entire image and the local image including the abnormal shadow on the same display surface of a single display means. Thus, the diagnostic performance can be improved. However, in Patent Document 2, the whole image of one breast is placed on the right half of the display surface of the whole image display means, and the whole image of the other breast is placed on the left half of the display surface of the same whole image display means. However, the specification of the direction in which the chest wall exists is not described.
Japanese Patent Laid-Open No. 9-215684 (pages 1, 3 and 3) JP-A-8-294479 (pages 4, 16 and 14)

  In view of the above, an object of the present invention is to provide a medical image processing apparatus and a medical image processing method capable of automatically detecting the direction in which a chest wall exists in a mammo image.

  In order to solve the above problems, a medical image processing apparatus according to a first aspect of the present invention includes an interface for inputting image data and image supplementary information obtained by imaging for diagnostic purposes using a predetermined communication protocol; Image rotation information generating means for generating image rotation information indicating the orientation of the medical image represented by the input image data based on the type of communication protocol used in the interface or the input image supplementary information. .

  A medical image processing apparatus according to a second aspect of the present invention includes an interface for inputting image data and image supplementary information obtained by imaging for diagnostic purposes, and a plurality of mammo images represented by the image data input by the interface. By comparing the average value calculation means for calculating the average value of the luminance in the divided areas and the average value of the luminance in the plurality of divided areas calculated by the average calculation means, the chest wall in the mammo image represented by the image data Chest wall direction determining means for determining the direction.

  A medical image processing apparatus according to a third aspect of the present invention includes an interface for inputting image data obtained by imaging for diagnostic purposes and image supplementary information, and luminance of a mammo image represented by the image data input by the interface. Alternatively, a center-of-gravity calculating unit that calculates the position of the center of gravity regarding the differential value of luminance, and a chest wall direction determining unit that determines the chest wall direction in the mammo image represented by the image data based on the position of the center of gravity calculated by the center-of-gravity calculating unit. It comprises.

  The medical image processing method according to the first aspect of the present invention includes a step (a) of inputting image data and image supplementary information obtained by imaging for diagnostic purposes using a predetermined communication protocol; and (b) generating image rotation information representing the orientation of the medical image represented by the input image data based on the type of communication protocol used in a) or the input image supplementary information. .

  The medical image processing method according to the second aspect of the present invention includes a step (a) of inputting image data obtained by imaging for diagnostic purposes and image supplementary information, and a step represented by the image data input in step (a). The average value of the luminance in the plurality of divided regions of the mammo image to be performed is compared with the average value of the luminance in the plurality of divided regions calculated in step (b) by the image data. (C) determining the chest wall direction in the mammo image represented.

  The medical image processing method according to the third aspect of the present invention includes a step (a) of inputting image data obtained by imaging for diagnostic purposes and image supplementary information, and a step represented by the image data input in step (a). Calculating the position of the center of gravity regarding the luminance of the mammo image or the differential value of the luminance, and the chest wall in the mammo image represented by the image data based on the position of the center of gravity calculated in step (b) And (c) determining a direction.

  According to the present invention, by using the type of the communication protocol used when inputting the image data and the image supplementary information or the input image supplementary information and the luminance of the mammo image represented by the image data, It is possible to automatically detect the presence direction of the chest wall in the image. As a result, it becomes easy to perform blackening processing on the white areas of the mammo images and display a plurality of mammo images on one screen with the mammo area edges back to back.

Hereinafter, the best mode for carrying out the present invention will be described in detail with reference to the drawings. The same constituent elements are denoted by the same reference numerals, and the description thereof is omitted.
FIG. 1 is a block diagram showing a configuration of a medical image photographing system including a medical image processing apparatus according to the first embodiment of the present invention. As shown in FIG. 1, the medical image photographing system 10 is a medical image photographing apparatus 10 that records a radiation image on a recording sheet (stimulable phosphor sheet) 1 by irradiating a subject with radiation. A medical image reading device 20 that photoelectrically reads information such as a radiation image recorded on the recording sheet 1 to generate image data, and performs various processes by inputting image data from the medical image reading device 20 The medical image processing apparatus 30 displays or outputs an image. Furthermore, an external device 80 such as a database server, an ID card reader, or a terminal device is connected to the network N1 as necessary.

  The medical image photographing apparatus 10 includes a photographing position raising / lowering mechanism 11 that raises and lowers the photographing position of the subject by moving the position of the recording sheet 1 up and down, a photographing stand 12 that determines the position of the subject's legs, A radiation generator 13 for irradiating the examiner, an imaging controller 14 for controlling the radiation generator 13 and the like according to given imaging conditions, and an input unit 15 used for inputting various commands and imaging conditions Is included. The shooting control unit 14 is connected to the network N1, and can set shooting conditions via the network N1.

  A recording sheet 1 used for radiography is coated with a stimulable phosphor material, and information on an object is recorded by irradiation with radiation. The subject is radiographed under predetermined imaging conditions, and the radiographic image is recorded on the recording sheet 1. After photographing, the recording sheet 1 is set at a predetermined position of the medical image reading device 20.

  In the medical image reading apparatus 20, the light beam emitted from the laser light source 21 scans the surface of the recording sheet 1 through the optical scanning unit 22. By this scanning, a light beam is irradiated onto the recording sheet 1, and an amount of stimulated emission light corresponding to the radiation image information accumulated and recorded is generated from the portion irradiated with the light beam. The stimulated emission light is photoelectrically detected by a photomultiplier (photomultiplier tube) 23, output as an analog signal, amplified by an amplifier 24, and digitized by an A / D converter 25. The image data generated in this way is input to the medical image processing apparatus 30 via the network N1 together with the image supplementary information.

  The medical image processing apparatus 30 includes an input unit 31 used for inputting various commands and settings for display formats, a display unit 32 for displaying an image, a printer 33 for printing an image on a film, and the like. Part 40. The processing unit 40 processes the input image data and configures image data for simultaneously displaying a plurality of images on one screen.

  FIG. 2 is a block diagram showing a detailed configuration of the medical image processing apparatus shown in FIG. The processing unit 40 includes a central processing unit (hereinafter referred to as a CPU) 41, a memory 42 for temporarily storing input image data and image supplementary information, a hard disk 43 as a recording medium, a hard disk control unit 44, and an interface 45. And a network interface 46. These are connected to each other via a bus line BL.

  The CPU 41 is connected to an input unit 31 such as a keyboard and a mouse, a display unit 32 such as a CRT display, and a printer 33 via an interface 45, and the medical image photographing apparatus 10 via a network interface 46 and a network N1. And a medical image reading device 20. The hard disk 43 stores software (program) for causing the CPU 41 to perform an operation. In addition to the built-in hard disk 43, an external hard disk, flexible disk, MO, MT, RAM, CD-ROM, DVD-ROM, or the like can be used as the recording medium.

  Next, functional blocks 41a to 41f configured by the CPU 41 and software (program) will be described. FIG. 3 shows an example of a mammo image represented by image data input from the medical image reading apparatus to the medical image processing apparatus. In this mammo image, there are a mammo area 2, an area outside the mammo 3, and a white area 4 that is generated due to the delay in the timing of reading.

  The image rotation information generation unit 41a generates image rotation information representing the chest wall direction or the chest wall direction and the front and back of the captured image based on the communication protocol between the medical image reading device 20 and the medical image processing device 30. To do. The communication protocol is set when each device is connected to the network.

  For example, when the communication protocol between the medical image reading device 20 and the medical image processing device 30 is FRUP (Fuji reader unit protocol), the medical image reading device 20 outputs the communication protocol. Since the chest wall direction of the image represented by the image data is the downward direction, the image rotation information generation unit 41a generates image rotation information indicating that the chest wall direction is “down” and the image is in the “front” direction.

  On the other hand, when the protocol between the medical image reading apparatus 20 and the medical image processing apparatus 30 is a DICOM (digital imaging and communications in medicine) protocol, the image rotation information generation unit 41a displays image data. Image rotation information representing the chest wall direction and the front and back is generated based on an IIP (Internet imaging protocol) internal rotation reversal flag attached to.

  The DICOM standard is a standard developed to enable transmission / reception of medical image data and the like on a network regardless of whether it is in a hospital or a medical image equipment manufacturer. In a medical image reading apparatus conforming to the DICOM standard, image data is generated based on the arrangement direction and the scanning direction of a recording sheet, and an image display pattern represented by the image data is indicated by an IIP internal rotation inversion flag.

  Generally, the recording sheet is stored in a cassette, and when a mammo image is taken, a mammo image taking cassette with one long side frame removed is used as shown in FIG. As shown in FIG. 5, the mammo image is shot by placing the chest wall against the long side of the mammo image shooting cassette.

  Further, in a medical image reading apparatus conforming to the DICOM standard, image data is generated by scanning a recording sheet in a cassette inserted in a direction determined for each apparatus, and therefore an image represented by the image data And the IIP internal rotation reversal flag uniquely correspond to each other.

  FIG. 6 is a diagram showing images corresponding to the IIP internal rotation inversion flag and the IIP internal rotation inversion flag in the DICOM standard. FIG. 6A shows an IIP internal rotation inversion flag attached to image data taken using a recording sheet having a short side of 18 cm and a long side of 24 cm, and an image corresponding to the IIP internal rotation inversion flag. Show. The thick line in the image represents the chest wall direction of the mammo image, and “F” is used instead of the mammo image to represent the front and back of the image. Here, an image in which “F” is drawn in the front direction is set to face up.

  As shown in FIG. 6A, in the image corresponding to the IIP internal rotation inversion flag (001), the chest wall direction is represented as “left” and the image is represented as “front”. Further, in the image corresponding to the IIP internal rotation inversion flag (101), the chest wall direction is represented as “right” and the image is represented as “front”. Similarly, an IIP internal rotation inversion flag is set corresponding to images represented by a total of eight patterns combining front and back and upper, lower, left, and right chest wall directions.

  FIG. 6B shows an IIP internal rotation inversion flag attached to image data taken using a recording sheet having a short side of 24 cm and a long side of 30 cm, and an image corresponding to the IIP internal rotation inversion flag. Show. The method of drawing an image in this figure is the same as that described with reference to FIG.

  As shown in FIG. 6B, in the image corresponding to the IIP internal rotation inversion flag (100), the chest wall direction is represented as “left” and the image is represented as “front”. In the image corresponding to the IIP internal rotation inversion flag (000), the chest wall direction is represented as “right” and the image is represented as “front”. Similarly, an IIP internal rotation inversion flag is set corresponding to a total of eight patterns of images combining the upper and lower chest wall directions and the front and back. Therefore, in the present embodiment, based on the image rotation information representing the chest wall direction and the front and back, the white area is blackened, or the rotation process is performed in which the mammo area ends are back to back in order to improve the efficiency of medical diagnosis. It is done automatically.

  Referring to FIG. 2 again, the blank area identification unit 41b identifies a blank area from the captured image based on the image rotation information and the QL value of the image data. Here, the QL value is a value obtained by quantizing a luminance value representing the intensity of the stimulated emission light generated from the recording sheet. That is, in this embodiment, the value obtained by A / D-converting the electrical signal obtained for each pixel by photoelectrically detecting the stimulated emission light generated by irradiating the recording sheet with the light beam is the QL value. Equivalent to. When the QL value is small, the luminance of the displayed image increases. When the QL value is large, the luminance of the displayed image decreases.

Here, the identification of the blank area will be described. White spot area identifying unit 41b, first, the QL value difference processing along the direction _{D 1} shown in FIG. 7, and so on, the direction _{D 2, D 3, ···,} subtracting the QL value along Dn To process. The plurality of directions D _{1 to} Dn are radial directions from the center position O of the side on the chest wall direction side determined based on the image rotation information to the end on the other side in the image represented by the image data. In this embodiment, they are set at equiangular intervals.

Further, the blank area identification unit 41b, based on the difference value of the QL values in the plurality of directions D _{1 to} Dn, the boundary between the blank area 4 and the mammo area 2, and the blank area 4 and the outside mammo area 3 Boundary candidate points that are estimated to be the boundaries are obtained. That, QL value in the breast region 2 and the breast out area 3, since the clearly higher than QL value in the white spot region 4 stimulated emission does not occur, for example, an image signal in the direction D ₂ level Has changes as shown in FIG. Therefore, the difference value of the QL value changes greatly at the boundary between the whiteout region 4 and the non-mammo region 3 as shown in FIG. 8B.

  Therefore, the blank area identifying unit 41b detects the point closest to the center position O among the points (boundary candidate points) where the absolute value of the difference between the QL values exceeds the threshold value, and obtains the boundary point. Further, the blank region identification unit 41b estimates several boundary lines represented by a straight line or a predetermined curve based on the detected coordinates of the plurality of boundary points, and is more in the chest wall direction than the boundary lines. The area is recognized as a blank area 4.

  Next, the blackening process will be described. The blackening processing unit 41c shown in FIG. 2 performs blackening processing on the image data corresponding to the blank area 4 on the input image data. In other words, the blackening processing unit 41 c sets the QL value of the pixels in the whiteout region 4 to be the same as the maximum value (black level at the time of display) or the QL value of the pixels in the region outside the mammo 3. FIG. 9 is a diagram illustrating an image obtained by performing blackening processing on the image illustrated in FIG. 7. Thus, by automatically detecting the chest wall direction, it becomes easy to blacken the white area 4.

  The image processing unit 41d illustrated in FIG. 2 performs various types of image processing such as normalization processing and gradation processing on the image data that has been subjected to blackening processing. Image data that has been subjected to various types of image processing in the image processing unit 41d is supplied from the output unit 41f to the display unit 32 via the interface 45, and an image is displayed on the display unit 32.

  Next, image rotation processing will be described. The operator confirms the image displayed on the display unit 32 shown in FIG. 2, and inputs an instruction to display the mammo images back to back using the input unit 31 as shown in FIG. When an instruction to display the mammo images back to back is input, the image composition unit 41e displays the image rotation information generated by the image rotation information generation unit 41a and the white area identified by the white area identification unit 41b. Based on the above, a layout of an image to be displayed on the display unit 32 or output to the printer 33 is configured.

  When an instruction to display the mammo images back to back is input, the image construction unit 41e cuts out image data corresponding to the white areas from the image data representing the left and right mammo images, and uses the image rotation information as image rotation information. Based on this, image rotation processing is performed on each image data, and an image layout combining the chest wall side is constructed. Further, the image construction unit 41e automatically aligns the corresponding positions of the two mammo images on the display screen by moving one mammo image in the vertical direction on the display screen.

  The image construction unit 41e includes the image rotation information generated by the image rotation information generation unit 41a, and part information (included in the image supplementary information) indicating whether the image is a right mammo image or a left mammo image. Based on the above, image rotation processing is performed on the image data. In the image rotation processing, image data is processed so that the image rotates in the display screen, or the image data is rotated so that the image is rotated around an axis parallel to the side on the chest wall direction side so that the front and back sides are reversed. Is processed.

  First, the process for creating the left mammo image will be described. Here, in order to display the left mammo image on the right side of the screen, the image information is processed so that the chest wall direction is on the left side and the image is face up.

  When the image rotation information indicates “left” and “table”, the image rotation processing is not necessary. On the other hand, when the image rotation information represents “right” and “table”, the image composition unit 41e processes the image data so that the mammo image rotates 180 degrees in the display screen, and the image rotation information When “right” and “back” are represented, the image data is processed so that the mammo image rotates 180 degrees about an axis parallel to the chest wall side, and the image rotation information is “left”. When “back” is indicated, the mammo image is rotated 180 degrees in the display screen, and further, the image data is rotated 180 degrees around the axis parallel to the side on the chest wall direction side. Process.

  Alternatively, when the image rotation information represents “down” and “table”, the image configuration unit 41e processes the image data so that the mammo image rotates 90 degrees in the clockwise direction in the display screen, When the image rotation information represents “up” and “table”, the image data is processed so that the mammo image rotates 90 degrees counterclockwise in the display screen, and the image rotation information is “down” and In the case of “back”, the mammo image is rotated 90 degrees counterclockwise in the display screen, and the mammo image is rotated 180 degrees about an axis parallel to the chest wall side. When the image data is processed and the image rotation information indicates “up” and “back”, the mammo image is rotated 90 degrees clockwise in the display screen, and is further parallel to the chest wall side. Mammo image centered on a special axis Processing the image data to rotate 180 degrees.

  Next, processing for creating the right mammo image will be described. Here, in order to display the right mammo image on the left side of the screen, the image information is processed so that the chest wall direction is on the right side and the image is face up.

  When the image rotation information indicates “right” and “table”, the image rotation processing is not necessary. On the other hand, when the image rotation information represents “left” and “table”, the image composition unit 41e processes the image data so that the mammo image is rotated 180 degrees in the display screen, and the image rotation information is When “right” and “back” are represented, the mammo image is rotated 180 degrees in the display screen, and the mammo image is rotated 180 degrees around an axis parallel to the side on the chest wall direction side. When the image data is processed and the image rotation information indicates “left” and “back”, the image data is rotated so that the mammo image is rotated 180 degrees around an axis parallel to the chest wall side. Process.

  Alternatively, when the image rotation information indicates “down” and “table”, the image composition unit 41e processes the image data so that the mammo image rotates 90 degrees counterclockwise in the display screen. When the image rotation information represents “up” and “table”, the image data is processed so that the mammo image rotates 90 degrees clockwise in the display screen, and the image rotation information is “down” and In the case of “back”, the mammo image is rotated 90 degrees clockwise in the display screen, and the mammo image is rotated 180 degrees about an axis parallel to the chest wall side. When image data is processed and the image rotation information indicates “up” and “back”, the mammo image is rotated 90 degrees counterclockwise in the display screen, and is further parallel to the chest wall side. Mammo image centered on a special axis Processing the image data to rotate 180 degrees.

  In the above description, the left and right mammo images are displayed with the chest wall side aligned, but two left (or right) mammo images taken at different times are displayed with the chest wall side aligned on the front and back. You may do it. In addition, the image configuration unit 41e sets a display window pattern for displaying a plurality of images side by side in the horizontal direction or the vertical direction, or automatically sets the image to be displayed in a desired display format according to the imaging region. It is also possible.

  The operator can use the input unit 31 to set one type of display format from among a plurality of types of display formats. If it is possible to determine for each operator which display format is to be set, the display format can be automatically set according to the preference of each operator. Further, when a plurality of medical image processing apparatuses 30 are connected to the network N1, the display format set in one medical image processing apparatus 30 is recorded for each operator, and the other medical image processing apparatuses 30 If the same display format is automatically set for each operator, the operation of the operator becomes very easy.

  The image data in which the image layout is configured is output from the output unit 41f to the display unit 32 or the printer 33 via the interface 45, and the image is displayed on a display or printed on a film or the like. Note that image processing such as normalization processing and gradation processing may be performed before blackening processing. Further, when the white area is not a problem, the blackening process may be omitted.

  In the present embodiment, the image rotation information generation unit 41a, the white area identification unit 41b, the blackening processing unit 41c, the image processing unit 41d, the image configuration unit 41e, and the output unit 41f are configured by a CPU and software. Alternatively, it may be constituted by a digital circuit or an analog circuit.

  Next, the operation of the medical image photographing system including the medical image processing apparatus according to the present embodiment will be described with reference to FIGS. FIG. 11 is a flowchart showing the operation of this medical image photographing system.

  First, in step S <b> 11, the medical image capturing apparatus 10 captures a radiation image by irradiating the subject with radiation, and records the radiation image on the recording sheet 1. In step S <b> 12, the medical image reading device 20 reads the radiation image recorded on the recording sheet 1 and generates image data.

  In step S13, when the communication protocol is FRUP, the image rotation information generation unit 41a of the medical image processing apparatus 30 displays image rotation information indicating that the chest wall direction is “down” and the image is “front”. When the communication protocol is the DICOM protocol, the image rotation information corresponding to the IIP internal rotation / inversion flag is obtained based on the IIP internal rotation / inversion flag attached to the image data input from the image reading device 20. Generate.

  In step S14, the blank area identification unit 41b identifies a blank area from the image represented by the image data based on the image rotation information and the QL value. In step S15, the blackening processing unit 41c performs blackening processing in the white area on the input image data.

  In step S16, the image processing unit 41d performs various image processing such as normalization processing and gradation processing. In step S <b> 17, the operator inputs an instruction to display an image in a desired display format using the input unit 31. In step S18, the image construction unit 41e performs necessary image rotation processing on the image data representing the mammo image in accordance with the instructed display format, thereby constructing a mammo image layout. In step S <b> 19, the mammo image is displayed on the display unit 32 via the interface 45 or printed on the printer 33.

  Next, an image information processing apparatus according to the second embodiment of the present invention will be described. FIG. 12 is a block diagram showing a configuration of a medical image processing apparatus according to the second embodiment of the present invention. In this medical image processing apparatus, an image rotation information generation unit 50 generates image rotation information instead of the image rotation information generation unit 41a. Other configurations are the same as those of the medical image processing apparatus shown in FIG.

  The image rotation information generation unit 50 divides the mammo image represented by the image data output from the image reading device 20 into a plurality of regions, and calculates a QL value average calculation unit 51 that calculates an average of the QL values in each region; A chest wall direction determination unit 52 that determines the chest wall direction based on the average of the QL values in each region calculated by the QL value average calculation unit 51. Each of the QL value average calculation unit 51 and the chest wall direction determination unit 52 may be configured by a CPU and software, or may be configured by a digital circuit or an analog circuit.

  Next, the operation of the medical image photographing system including the medical image processing apparatus according to the present embodiment will be described with reference to FIGS. FIG. 13 is a flowchart showing the operation of this medical image photographing system. In the present embodiment, image rotation information is generated in steps S21 and S22 instead of step S13 shown in FIG. Here, it is assumed that the image data is not generated so that the image is face down, and the image is not reversed. However, when the image is displayed face down, the image data may be processed so that the image is face up according to “lead characters” that allow visual confirmation of the front and back of the image. Other operations are the same as those shown in FIG.

  In step S21, the QL value average calculation unit 51 divides the mammo image into a plurality of regions, and calculates the average of the QL values in the image data corresponding to each region. FIG. 14 is a diagram illustrating an example of division of a mammo image. As shown in FIG. 14, the mammo image is divided into nine regions. The QL value average calculation unit 51 calculates the average of the QL values in the areas for the divided areas 5a to 5d among the divided areas.

  Next, in step S22, the chest wall direction determination unit 52 determines the direction of the region having the smallest average as the chest wall direction based on the average of the QL values obtained by the QL value average calculation unit 51, and sets the image rotation information. Generate. The area outside the mammo has a large QL value due to the large amount of the stimulated emission light, and the white area has a small QL value because no stimulated emission light is generated. In addition, the mammo region takes the middle of those QL values. In other words, the chest wall direction region that largely includes the mammo region and the whiteout region has a smaller average QL value than the other regions. Therefore, in the example shown in FIG. 14, the average of the QL values is the smallest in the divided area 5c including a part of the mammo area and a part of the blank area, so that it can be seen that the lower side of the image is the chest wall direction.

  Since the chest wall exists in the long side direction of the image, when the long side direction of the image is known in advance, the average of the QL values in the divided region 5a is compared with the average of the QL values in the divided region 5c. Anyway. Further, the areas of the divided regions may not be equal, and the image may be divided in any way as long as the dividing method is such that the chest wall direction can be understood as shown in FIG. Further, the image rotation information may be calculated based on image data subjected to image processing such as normalization processing and gradation processing.

  Next, an image information processing apparatus according to the third embodiment of the present invention will be described. FIG. 16 is a block diagram showing a configuration of a medical image processing apparatus according to the third embodiment of the present invention. In this medical image processing apparatus, instead of the image rotation information generation unit 50, an image rotation information generation unit 60 generates image rotation information. Other configurations are the same as those of the medical image processing apparatus shown in FIG.

  The image rotation information generation unit 60 considers the differential value of the QL value based on the input image data, and the absolute value of the QL value as the density. The center-of-gravity calculation unit 62 that calculates the chest wall direction determination unit 63 that determines the chest wall direction based on the center of gravity calculated by the center-of-gravity calculation unit 62 is included. Each of the differential processing unit 61, the gravity center calculating unit 62, and the chest wall direction determining unit 63 may be configured by a CPU and software, or may be configured by a digital circuit or an analog circuit.

  Next, the operation of the medical image photographing system including the medical image processing apparatus according to the present embodiment will be described with reference to FIGS. FIG. 17 is a flowchart showing the operation of this medical image photographing system. In the present embodiment, image rotation information is generated in steps S31 to S33 instead of steps S21 and S22 shown in FIG. Other operations are the same as those shown in FIG.

  In step S <b> 31, the differential processing unit 61 obtains differential data by performing a filtering process on the image data input from the medical image reading apparatus using a linear first-order spatial differential filter (Previt filter). Here, linear first-order spatial differentiation means that a total of nine QL values of a pixel of interest and upper, lower, left, and right adjacent pixels are calculated by summing QL values respectively multiplied by a predetermined coefficient. A differential value corresponding to a change in the QL value in the pixel is obtained.

  FIG. 18 shows the coefficients of the Previt filter. 18A is a coefficient for differentiating in the horizontal direction, and FIG. 18B is a coefficient for differentiating in the vertical direction. In each direction, the portion where the change in the QL value is large, such as the mammo region, has a large absolute value of the differential value, and the portion where the change in the QL value is small, such as the other regions, is Get smaller. Therefore, the center of the mammo region can be obtained by calculating the centroid of the absolute value of the differential value.

  Depending on the characteristics of the image, either one of the coefficients shown in FIGS. 18A and 18B may be used, or both may be used in combination. For example, it is possible to calculate the center of gravity based on the differential value obtained using the coefficient shown in FIG. 18 (a), or use the coefficients shown in FIG. 18 (a) and (b). It is also possible to calculate the center of gravity based on the obtained differential value. In place of the Previt filter, a Roberts filter shown in FIG. 19 or a Sobel filter shown in FIG. 20 may be used.

但し、ｉは画像の水平方向の座標、ｊは画像の垂直方向の座標、Ｉは画像の水平方向の画素数、Ｊは画像の垂直方向の画素数、ｄ_ｉｊは座標（ｉ，ｊ）の画素における微分値の絶対値、Ｄは次式（２）に示す全画素における微分値の絶対値の総和である。

Referring to FIG. 17 again, in step S32, the centroid calculating unit 62 calculates the centroid G1 (G1 _i , G1 _j ) of the absolute value of the differential value using the following equation (1).

Where i is the horizontal coordinate of the image, j is the vertical coordinate of the image, I is the number of pixels in the horizontal direction of the image, J is the number of pixels in the vertical direction of the image, and _dij is the coordinate (i, j) The absolute value of the differential value in the pixel, D is the sum of the absolute values of the differential values in all the pixels shown in the following equation (2).

  Next, in step S33, the chest wall direction determination unit 63 calculates the distance between the center of gravity G1 and each side of the image, and compares these distances to determine the direction of the side closest to the center of gravity G1 as the chest wall direction. And the image rotation information is generated based on the direction.

  When the long side direction of the image is known in advance, the image rotation information may be generated by comparing the distance between the center of gravity G1 and the two long sides, When the length of the short side is known in advance, the image rotation information may be generated by comparing the distance between the center of gravity G1 and one long side with half the length of the short side.

  Next, an image information processing apparatus according to the fourth embodiment of the present invention will be described. FIG. 21 is a block diagram showing a configuration of a medical image processing apparatus according to the fourth embodiment of the present invention. In this medical image processing apparatus, instead of the image rotation information generation unit 60, an image rotation information generation unit 70 generates image rotation information. Other configurations are the same as those of the medical image processing apparatus shown in FIG.

  The image rotation information generation unit 70 subtracts the QL value from a predetermined value, and considers the value obtained by subtracting the QL value from the predetermined value as the density. A calculation unit 72 and a chest wall direction determination unit 73 that determines the chest wall direction based on the center of gravity calculated by the center of gravity calculation unit 72 are included. Each of the subtraction processing unit 71, the center of gravity calculation unit 72, and the chest wall direction determination unit 73 may be configured with a CPU and software, or may be configured with a digital circuit or an analog circuit.

  Next, the operation of the medical image photographing system including the medical image processing apparatus according to the present embodiment will be described with reference to FIGS. FIG. 22 is a flowchart showing the operation of this medical image photographing system. In the present embodiment, image rotation information is generated in steps S41 to S43 instead of steps S31 to S33 shown in FIG. Other operations are the same as those shown in FIG.

  In step S41, the subtraction processing unit 71 obtains subtraction data by subtracting the QL value from a predetermined value. Here, the predetermined value may be larger than the average of the QL values in all pixels, but is preferably equal to or larger than the maximum value of the QL values.

但し、ｉは画像の水平方向の座標、ｊは画像の垂直方向の座標、Ｉは画像の水平方向の画素数、Ｊは画像の垂直方向の画素数、ｓ_ｉｊは座標（ｉ，ｊ）の画素における減算値、Ｓは次式（４）に示す全画素における減算値の総和である。

In step S42, the center-of-gravity calculation unit 72 calculates the center of gravity G2 (G2 _i , G2 _j ) of the subtraction value using the following equation (3).

Where i is the horizontal coordinate of the image, j is the vertical coordinate of the image, I is the number of pixels in the horizontal direction of the image, J is the number of pixels in the vertical direction of the image, and s _ij is the coordinate (i, j) The subtraction value in the pixel, S is the sum of the subtraction values in all the pixels shown in the following equation (4).

  Next, in step S43, the chest wall direction determination unit 73 calculates the distance between the center of gravity G2 and each side of the image, and compares these distances to determine the direction of the side closest to the center of gravity G2 as the chest wall direction. And the image rotation information is generated based on the direction.

  When the long side direction of the image is known in advance, the image rotation information may be generated by comparing the distance between the center of gravity G2 and the two long sides, When the length of the short side is known in advance, the image rotation information may be generated by comparing the distance between the center of gravity G2 and one long side with half the length of the short side.

  As described above, according to the present invention, it is possible to automatically detect the presence direction of a chest wall in a mammo image, and it can be used in a medical image processing apparatus that automatically performs blackening processing and image rotation processing. .

1 is a block diagram illustrating a configuration of a medical image photographing system including a medical image processing apparatus according to a first embodiment of the present invention. It is a block diagram which shows the detailed structure of the medical image processing apparatus shown in FIG. It is a figure which shows the example of the image represented by the image data input into a medical image processing apparatus from a mammo medical image reading apparatus. It is a figure which shows the cassette for mammo image photography. It is a figure which shows arrangement | positioning with the cassette for mammo images, and a mammo. It is a figure which shows the image corresponding to the IIP internal rotation inversion flag and IIP internal rotation inversion flag in a DICOM standard. FIG. 3 is a diagram for explaining a method for recognizing a white area in the medical image processing apparatus shown in FIG. 2. It is a figure which shows the change of the QL value in one direction of the image displayed based on image data, and a difference value, when a white-out area | region exists. It is a figure which shows the image which performed the blackening process to the image shown in FIG. It is a figure which shows the display format which displays a mammo image back to back. It is a flowchart which shows operation | movement of the medical image imaging system containing the medical image processing apparatus which concerns on the 1st Embodiment of this invention. It is a block diagram which shows the structure of the medical image processing apparatus which concerns on the 2nd Embodiment of this invention. It is a flowchart which shows operation | movement of the medical image imaging system containing the medical image processing apparatus which concerns on the 2nd Embodiment of this invention. It is a figure which shows an example of the division | segmentation of a mammo image. It is a figure which shows an example of the division | segmentation of a mammo image. It is a block diagram which shows the structure of the medical image processing apparatus which concerns on the 3rd Embodiment of this invention. It is a flowchart which shows operation | movement of the medical image imaging system containing the medical image processing apparatus which concerns on the 3rd Embodiment of this invention. It is a figure which shows the coefficient of a Previt filter. It is a figure which shows the coefficient of a Roberts filter. It is a figure which shows the coefficient of a Sobel filter. It is a block diagram which shows the structure of the medical image processing apparatus which concerns on the 4th Embodiment of this invention. It is a flowchart which shows operation | movement of the medical image imaging system containing the medical image processing apparatus which concerns on the 4th Embodiment of this invention. It is a figure which shows the radiographic image of mammography.

Explanation of symbols

1 Recording sheet (stimulable phosphor sheet)
2 mammo area 3 non-mammo area 4 white areas 5a to 5d, 6a to 6d divided area 10 medical image photographing device 11 photographing position raising / lowering mechanism 12 photographing stand 13 radiation generating unit 14 photographing control unit 15 input unit 20 medical image reading device 21 Laser light source 22 Optical scanning unit 23 Photomultiplier (photomultiplier tube)
24 Amplifier 25 A / D Converter 30 Medical Image Processing Device 31 Input Unit 32 Display Unit 33 Printer 40 Processing Unit 41 Central Processing Unit (CPU)
41a, 50, 60, 70 Image rotation information generation unit 41b White area identification unit 41c Blackening processing unit 41d Image processing unit 41e Image composition unit 41f Output unit 42 Memory 43 Hard disk 44 Hard disk control unit 45 Interface 46 Network interface 51 QL value Average calculation units 52, 63, 73 Chest wall direction determination unit 61 Differential processing units 62, 72 Center of gravity calculation unit 71 Subtraction processing unit 80 External device 100 Breast region 101 Blank region

Claims (16)

An interface for inputting image data and image supplementary information obtained by imaging for diagnostic purposes using a predetermined communication protocol;
Image rotation information generating means for generating image rotation information representing the orientation of the medical image represented by the input image data, based on the type of communication protocol used in the interface or the input image supplementary information;
A medical image processing apparatus comprising:   The blackening processing unit according to claim 1, further comprising a blackening processing unit that performs blackening processing on a part of the medical image represented by the input image data based on the image rotation information generated by the image rotation information generation unit. Medical image processing apparatus.   The medical image processing according to claim 1, further comprising: an image composing unit that forms a layout for displaying a plurality of medical images on a single screen based on the image rotation information generated by the image rotation information generating unit. apparatus.   When the communication protocol is a DICOM (Medical Digital Image Communication) protocol, the image rotation information generating means is a medical device represented by input image data based on a predetermined flag included in the image supplementary information. The medical image processing apparatus according to any one of claims 1 to 3, which generates image rotation information representing an orientation in an image plane and an image front and back. An interface for inputting image data obtained by imaging for diagnostic purposes and image supplementary information;
An average value calculating means for calculating an average value of luminance in a plurality of divided areas of the mammo image represented by the image data input by the interface;
A chest wall direction determining means for comparing the average value of the luminance in the plurality of divided regions calculated by the average calculating means to determine the chest wall direction in the mammo image represented by the image data;
A medical image processing apparatus comprising: An interface for inputting image data obtained by imaging for diagnostic purposes and image supplementary information;
Centroid calculating means for calculating the position of the centroid relating to the luminance of the mammo image represented by the image data input by the interface or a differential value of luminance;
Chest wall direction determining means for determining the chest wall direction in the mammo image represented by the image data based on the position of the center of gravity calculated by the center of gravity calculating means;
A medical image processing apparatus comprising:   The medical image processing according to claim 5 or 6, further comprising blackening processing means for performing blackening processing on a part of the mammo image represented by the input image data based on the determination result of the chest wall direction determination means. apparatus.   The medical image processing apparatus according to claim 5 or 6, further comprising an image composing unit that forms a layout for displaying a plurality of mammo images on a single screen based on a determination result of the chest wall direction determining unit.   The medical image processing apparatus according to claim 7, further comprising a display unit configured to display an image that has been blackened by the blackening processing unit, or an image having a layout configured by the image composing unit. A step (a) of inputting image data and image supplementary information obtained by imaging for diagnostic purposes using a predetermined communication protocol;
(B) generating image rotation information indicating the orientation of the medical image represented by the input image data based on the type of communication protocol used in step (a) or the input image supplementary information;
A medical image processing method comprising:   The medical image according to claim 10, further comprising a step (c) of performing blackening processing on a part of the medical image represented by the input image data based on the image rotation information generated in step (b). Processing method.   The medical image processing method according to claim 10, further comprising a step (c) of configuring a layout for displaying a plurality of medical images on one screen based on the image rotation information generated in step (b). . A step (a) of inputting image data and image supplementary information obtained by imaging for diagnostic purposes;
Calculating an average value of luminance in a plurality of divided regions of the mammo image represented by the image data input in step (a);
A step (c) of determining a chest wall direction in a mammo image represented by the image data by comparing average values of luminance in the plurality of divided regions calculated in step (b);
A medical image processing method comprising: A step (a) of inputting image data and image supplementary information obtained by imaging for diagnostic purposes;
Calculating the position of the center of gravity regarding the luminance of the mammo image represented by the image data input in step (a) or the differential value of the luminance; and
Determining the chest wall direction in the mammo image represented by the image data based on the position of the center of gravity calculated in step (b);
A medical image processing method comprising:   The medical image processing method according to claim 13 or 14, further comprising a step (d) of performing blackening processing on a part of the mammo image represented by the input image data based on the determination result in step (c). .   The medical image processing method according to claim 13 or 14, further comprising a step (d) of configuring a layout for displaying a plurality of mammo images on one screen based on the determination result in step (c).

Images