JP2007068715A - Medical image processing apparatus and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing apparatus and an image processing method for always detecting the center position of the right and left lungs with a small operation quantity independently of changes in postures and of the presence or absence of a lesion. <P>SOLUTION: The central line of both lungs is determined by using a boundary of a neck region and an inner boundary of both lungs since the central line of both lungs is always located within the right and left boundaries of the neck region. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a symmetric tissue separation method in CAD (Computer Aided Diagnosis) technology for medical images, and more particularly to an image processing apparatus and an image processing method for separating left and right lung fields of an X-ray chest front photograph.

  The chest X-ray image temporal subtraction technique corrects subject deformation for a first image and a second image of a common subject that are part of a human body photographed at different times, performs difference processing, and both images This is a technique for rendering a part with a change as a difference image. By such a time-difference technique, it is possible to extract image components corresponding to changes in lesions from two images, ignoring common normal parts such as bones and blood vessels.

  In the time-difference technique for chest X-ray images, when the deformation of the subject is corrected, the entire area is corrected for enlargement / reduction or translation, and then the fine deformation of the lung field is aligned by a warping technique. When performing global correction, the difference object lung field is often detected. In order to detect lung fields, it is necessary to separate the left and right lung fields.

  Moreover, in the chest X-ray front image, the front view of the spine is used as an effective means for diagnosing abnormalities such as scoliosis. However, the X-ray absorption of the spinal column varies greatly depending on the site, and furthermore, it cannot be displayed with appropriate brightness due to QA processing after imaging. For this reason, it is necessary to separate the left and right lung fields to detect the longitudinal region and perform appropriate image processing.

  In order to detect a symmetrical tissue such as the left and right lung fields, there is a method of detecting the center line by calculating the degree of symmetry with respect to the temporary center using the symmetrical tissue profile described in US6263097.

  As shown in FIG. 14, a point from the start point T to the end point MT−1 is set as the temporary center C for each row of the target image of size M × N. For each temporary center, the degree of symmetry S (i, j) is calculated from equation (1).

ただし、ｃ＝Ｔ，Λ，Ｍ−Ｔ−１
g(i,j)は位置(i,j)の画素値であり、
m(i)は第ｉ行の仮中心ｃにおけるウィンドウ幅。

However, c = T, Λ, M−T−1
g (i, j) is the pixel value at position (i, j)
m (i) is the window width at the temporary center c in the i-th row.

  Next, the calculated degree of symmetry is normalized by Equation (2) with the window width m (i) and the temporary center pixel value.

最後に、前記求めた仮中心から正規化された対象度の最小値位置を第iラインの中心とする。

Finally, the minimum value position of the degree of object normalized from the calculated temporary center is set as the center of the i-th line.

However, the method of detecting the center line of the anatomical tissue according to the degree of object cannot detect the correct center position for the part where the left and right lung fields have lost the target due to the lung field posture change or the lesion. In addition, since the target degree is calculated for each temporary center, the calculation amount is large.
US 6263097

  It is an object of the present invention to provide an image processing apparatus and an image processing method that can always detect the center position of the left and right lungs with a small amount of calculation regardless of posture change and presence / absence of a lesion.

  In order to solve the above problems, the present invention is a medical image processing apparatus for detecting a central line of left and right lung fields of a chest simple X-ray front photograph, and determines a representative profile for the chest simple X-ray front photograph. Means for setting a threshold value using the representative profile, and means for determining a region including the center of the left and right lung fields using the threshold value, and determining a center position of the left and right lung fields.

  In the present invention, the means for determining the representative profile adds a predetermined horizontal and vertical width area to an upper end of the chest simple X-ray front photograph, and uses the obtained average horizontal profile as a representative profile. And

  In the present invention, the means for determining the representative profile includes means for determining a first profile for determining a central line location range of the left and right lung fields, and a means for tracking changes in the left and right lung field center lines at each vertical position. The means for determining the second profile and the weighted average of the first profile and the second profile are calculated while the vertical position of the second profile is moving to obtain a representative profile.

  In the present invention, the means for determining the first profile adds an area of a predetermined horizontal and vertical width to an upper end of the chest simple X-ray front photograph, and uses the obtained average horizontal profile as the first profile. The medical image processing apparatus according to claim 3, wherein:

  In the present invention, the means for determining the second profile includes adding a region of a predetermined horizontal and vertical width to a predetermined vertical position on the chest simple X-ray front photograph, and obtaining the obtained average horizontal profile as the second profile. The medical image processing apparatus according to claim 3, wherein:

  The present invention is characterized by further comprising means for filtering the calculated first profile, second profile or representative profile with a low-pass filter.

  In the present invention, the means for setting a threshold value using the representative profile obtains a minimum value in a central region of the representative profile, and adds the value to an empirically determined value from the minimum value as a threshold value. And

  In the present invention, the means for setting a threshold value using the representative profile obtains a minimum value in a central region of the representative profile, calculates a maximum pixel value and a weighted average value of the image, and sets the threshold value. To do.

  In the present invention, the means for setting a threshold value using the representative profile obtains a minimum value in a central region of the representative profile, calculates a maximum pixel value and a weighted average value of the profile, and sets the threshold value. To do.

  In the present invention, the means for determining the left and right lung field center position determines a region of a profile smaller than the threshold value to the left and right of the minimum value using the obtained threshold value, and the center of the region is defined as the left and right lung field center position. It is characterized by doing.

  The present invention is characterized by further comprising means for smoothing the center position of the left and right lung fields or means for interpolating and smoothing the center position of the left and right lung fields.

  Further, the present invention is a medical image processing apparatus for detecting a center line of left and right lung fields in a chest simple X-ray front photograph, and determines a center line location range of left and right lung fields in the chest simple X-ray front photograph. Means for determining a first profile; means for determining a first threshold based on the first profile; means for determining a left-right boundary position of a left-right lung field central line location range using the first threshold; Using a second threshold, means for determining a second profile for tracking changes in the left and right lung field center line at each vertical position, means for determining a second threshold based on the second profile, Means for determining the inner boundary position of the left and right lung fields at each vertical position; and means for determining the left and right lung field center positions based on the left and right boundary positions of the center line location range and the inner boundary position of the left and right lung fields. Features.

  As shown in FIG. 15, in the present invention, using the property that the brightness level of the cervical region is low and the X-ray direct irradiation regions on both sides are relatively high, using a high luminance difference at the boundary of the cervical region, The neck boundary line can be easily detected. The center line of both lungs always uses the property that is within the left and right borders of the cervical region, and uses the border between the cervical region and the inner border of both lungs, from the border position, or from the cervical region and the lung region. Synthesize the profiles and determine the midline of both lungs. This makes it possible to always detect changes in the shape of the inside of the lung field due to changes in the posture of the subject at the time of imaging, and to stabilize both lungs with a very small amount of computation even when there is no single lung due to a lesion or surgery. Can be separated.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings.

(First embodiment)
FIG. 1 is a block diagram showing an example of a lung field detection processing system according to the first embodiment of the present invention and both lung separation units in the system.

  In FIG. 1, the lung field detection processing system includes an image input unit 101, both lung separation units 102, and an output unit 103.

  Each of the lung separation units 102 includes a representative profile determination unit 102a, a threshold value determination unit 102b, and a center position determination unit 102c.

  The image input unit 101 reads an X-ray chest front image into a memory from an image database stored in a storage device (not shown).

  In both lung separation units 102, a representative profile determination unit 102a calculates a representative profile that reflects the characteristic that the luminance level at the center of both lungs is very low in the read chest front image.

  In the present embodiment, as shown in FIG. 3 square a, the representative profile integrates the cervical region from the top of the chest front image to the 0.1 h line (h is the height of the chest front image), and the average The profile is a representative profile. FIG. 4 shows the representative profile. As shown in FIG. 4, the typical profile shows that the brightness level of the cervical region is low and the X-ray direct irradiation regions on both sides are relatively high. Moreover, a high luminance difference is shown at the boundary of the cervical region. In this embodiment, the center line of both lungs is always in the left and right boundaries of the neck region, and the center of the border of the neck region is the center line of both lungs.

  Next, the threshold value determination unit 102b obtains the minimum luminance value in the central part of the representative profile, for example, within the range from 0.3w to 0.7w (w is the width of the chest front image), and the minimum luminance value A threshold value T is obtained by adding a predetermined value determined empirically. The predetermined value determined empirically is set so that an increase in luminance level can be avoided in a profile obtained by projecting a normal anatomical structure such as a trachea in the cervical region. In a chest front image quantized with 12 bits, the predetermined value may be empirically set to 500.

  The center position determining unit 102c uses the obtained threshold value T to obtain a brightness value range smaller than the threshold value T in a central part of a typical profile, that is, a range from 0.3w to 0.7w. The center of the lung is the horizontal position of both lung center lines.

  The output unit 103 outputs the calculated center line of both lungs. The output destination may be only the display unit as shown in FIG. 2, or the subsequent processing for detecting and displaying the left and right lung fields as shown in FIG. In addition, it may be a subsequent process that uses the output center line to set a downspout area and emphasize an organ in the downspout or perform other image processing.

  Further, in the present embodiment, the representative profile determination unit 102a, as shown by the square b in FIG. 3, vertically extends from the top of the chest front image to the 0.1h line, and horizontally from 0.3w to 0.7w of the chest front image. The range up to and including the analysis region, and the lines in the analysis region may be integrated to form a representative profile. Subsequent 102b and 102c processing is performed for all profiles.

  Alternatively, the representative profile determination unit 102a may integrate each line in the analysis region to obtain an average profile, and then smooth the result with a low-pass filter, and use the smoothed profile as the representative profile.

  Further, when the threshold value determination unit 102b determines the threshold value by using a high luminance difference at the boundary of the cervical region, the predetermined value determined empirically is not added to the minimum value, It may be a weighted average of the minimum value and the maximum value of the image as shown, or may be a weighted average of the minimum value and the maximum value of the profile as shown in Equation (4).

ただし、Ｉ_minはプロファイルの最小値である。MAXは画像の最大値であり、１２ビット画像において、MAX=4095。

However, I _min is the minimum value of the profile. MAX is the maximum value of an image, and MAX = 4095 for a 12-bit image.

ただし、Ｉ_minはプロファイルの最小値であり、Ｉ_maxはプロファイルの最大値である。

However, I _min is the minimum value of the profile, and I _max is the maximum value of the profile.

  FIG. 5 shows a flowchart of this embodiment.

  S101: The image input unit 101 reads an X-ray front image to be processed into a memory.

  S102: The representative profile determination unit 102a of both lung separation units 102 sets a square region from the top to the 0.1h line as an analysis region in the image to be processed.

  S103: The representative profile determination unit 102a of both lung separation units 102 integrates the set analysis regions and calculates an average profile. The profile is a representative profile.

  S104: The threshold value determination unit 102b of both lung separation units 102 searches the representative profile for a minimum value in the central region from 0.3w to 0.7w.

  S105: The threshold value determination unit 102b of both lung separation units 102 sets a threshold value T by adding a predetermined value empirically determined to the minimum value.

  S106: The central position determination unit 102c of both lung separation units 102 searches the central region from 0.3w to 0.7w in the representative profile for a region having a value smaller than the threshold value T, and determines the central position of the region. Determine the central position of the lung.

  S107: Output the detection result.

(Second embodiment)
FIG. 6 is a block diagram showing an example of a lung field detection processing system according to the second embodiment of the present invention and both lung separation units in the system.

  As in the first embodiment, the present embodiment includes an image input unit 101, both lung separation units 102, and an output unit 103. Both lung separation units 102 are configured according to the first embodiment. And different.

  The both lung separation units 102 include a first profile determination unit 102a, a second profile determination unit 102b, an average unit 102c, a threshold determination unit 102d, a center position determination unit 102e, and a vertical position setting unit 102f. . Hereinafter, the both lung separation unit 102 will be described.

  As shown in FIG. 7, as in the first embodiment, the first profile determination unit 102a uses the cervical region from the top of the read chest front image to the 0.1h line as the first profile analysis region ( h is the height of the front image of the chest), the areas are integrated, and the average profile is determined as the first profile.

  On the other hand, the vertical position setting unit 102f sets one vertical position from the top of the image. As soon as the center position of the vertical position is determined, the vertical position is increased by one line, and the processes 102b, 102c, 102d, and 102e are performed. In this way, the center position of each horizontal line is detected repeatedly to the bottom of the image.

  As shown in FIG. 7, the second profile determination unit 102b sets a region having a predetermined vertical width as a second profile analysis region at the set vertical position, integrates each line in the analysis region, and calculates an average profile. calculate. The predetermined vertical width of the analysis area has almost the same change as the line at the set vertical position, and the noise of the line can be smoothed. As an experience, 5 lines above and below are sufficient.

  The averaging unit 102c calculates a weighted average of the first profile and the second profile as shown in Expression (5). The averaged profile is a representative profile. In addition, the weights assigned are set so that both the mid-lung range determined at the cervical boundary of the first profile and the inner boundary of the lung field at each vertical position in the second profile can be reflected. As an experience, the first profile may be 0.4 and the second profile may be 0.6.

ただし、Ｉ1_jは第1プロファイルに水平位置jにおける画素値であり、Ｉ2_ｊは第2プロファイルに水平位置jにおける画素値である。

However, I1 _j is a pixel value at the horizontal position j in the first profile, and I2 _j is a pixel value at the horizontal position j in the second profile.

  The threshold value determining unit 102d obtains the minimum luminance value in the central portion of the typical profile, for example, within the range from 0.3w to 0.7w (w is the width of the chest front image), and the minimum luminance value The threshold value T is set by adding a predetermined value. Similarly, in the chest front image quantized with 12 bits, the predetermined value may be empirically set to 500.

  The center position determination unit 102e uses the obtained threshold value T to obtain a brightness value range smaller than the threshold value T in a central part of a typical profile, that is, in a range from 0.3w to 0.7w. The center is the center of both lungs in the vertical position.

  In the present embodiment, the first profile determination unit 102a, the second profile determination unit 102b, and the average unit 102c may smooth the obtained profiles with a low-pass filter.

  Similarly, as in the first embodiment, in the threshold value determination unit 102e, when the threshold value is determined using a high luminance difference at the boundary of the cervical region, a predetermined value determined empirically is added to the minimum value. Instead, a weighted average of the minimum value and the maximum value of the image or the maximum value of the profile may be used.

  Moreover, although the detected center line of both lungs may be output as it is, it may be output after being smoothed.

  FIG. 8 shows a flowchart of the present embodiment.

  S201: The image input unit 101 reads an X-ray front image to be processed into a memory.

  S202: The first profile determination unit 102a of both lung separation units 102 sets a square area from the top to the 0.1h line as the first analysis area in the image to be processed.

  S203: The first profile determination unit 102a of both lung separation units 102 integrates the set first analysis region and calculates an average profile. This profile is the first profile.

  S204: The vertical position setting unit 102f of both lung separation units 102 initializes the position pointer k to 1.

  S205: The vertical position setting unit 102f of both lung separation units 102 determines whether or not the current vertical position follows up to the image height h. When tracing, the process proceeds to step S211. Otherwise, the process proceeds to step S206.

  S206: The second profile determination unit 102b of both lung separation units 102 sets the area of the upper and lower 5 lines of the current position pointer k as the second analysis area.

  S207: The second profile determination unit 102b of both lung separation units 102 integrates the set second analysis region and calculates an average profile. The profile is a second profile.

  S208: The average unit 102c of both lung separation units 102 weights the first profile and the second profile and averages them to obtain a representative profile.

  S209: The threshold value determination unit 102d of both lung separation units 102 searches the representative profile for a minimum value in the central region from 0.3w to 0.7w.

  S210: The threshold determination unit 102d of both lung separation units 102 sets a threshold T by adding a predetermined value empirically determined to the minimum value.

  S211: The central position determination unit 102e of both lung separation units 102 searches the central region from 0.3w to 0.7w in the representative profile for a region having a value smaller than the threshold value T, and determines the central position of the region. Determine the central position of both lungs in the vertical position.

  S212: The vertical position setting unit 102f increments the position pointer k and returns to step 205.

  S213: Output the result.

(Third embodiment)
FIG. 9 is a block diagram showing an example of a lung field detection processing system according to the third embodiment of the present invention and both lung separation units in the system.

  As in the first embodiment, this embodiment includes an image input unit 101, both lung separation units 102, and an output unit 103. Both lung separation units 102 are the same as those in the first embodiment. Different.

  Both lung separation units 102 are a first profile determination unit 102a, a second profile determination unit 102b, an average unit 102c, a threshold determination unit 102d, a center position determination unit 102e, an interpolation unit 102f, a stripe division unit 102g, a vertical And a position setting unit 102h. Hereinafter, the both lung separation unit 102 will be described.

  As in the first embodiment, the first profile determination unit 102a sets the cervical region from the top of the read chest front image to the 0.1h line as the analysis region of the first profile (h is the height of the chest front image). That is, the areas are integrated, and the average profile is determined as the first profile.

  On the other hand, as shown in FIG. 10, the stripe dividing unit 102g divides the read X-ray front image into a predetermined number of stripes. The number of stripes may be the number of samples that can sufficiently track changes in both lung center lines. As an experience, any value of 5-10.

  The vertical position setting unit 102h sets the central vertical position of the stripe for each stripe. As soon as the center position of the vertical position is determined, the stripe is incremented by 1 and the processes 102b, 102c, 102d, and 102e are performed. Thus, the above process is repeated to detect the center position of each stripe.

  Since the processes of the second profile determination unit 102b, the average unit 102c, the threshold value determination unit 102d, and the center position determination unit 102e are the same as those in the second embodiment, description thereof is omitted.

  The interpolating unit 102f interpolates the central positions of both lungs for all stripes, and interpolates the central positions of both lungs in each stripe between the first and last stripes. Extrapolation is performed at the center position of the top and bottom of the image from the outside of the vertical center position of the first and last stripes.

ただし、f(i)は各ストライプにおける中央位置の内挿補間式であり、ｓ₁とｓ_nは最初と最後のストライプ中心の垂直位置である。

However, f (i) is an interpolating between expression of the central position in each stripe, s ₁ and s _n is the vertical position of the first and last stripes center.

  In the present embodiment, the first profile determination unit 102a, the second profile determination unit 102b, and the average unit 102c may smooth the obtained profiles with a low-pass filter.

  Similarly, as in the first embodiment, in the threshold value determination unit 102e, when the threshold value is determined using a high luminance difference at the boundary of the cervical region, a predetermined value determined empirically is added to the minimum value. Instead, a weighted average of the minimum value and the maximum value of the image or the maximum value of the profile may be used.

  FIG. 11 shows a flowchart of the third embodiment.

  S301: The image input unit 101 reads an X-ray front image to be processed into a memory.

  S302: The first profile determination unit 102a of both lung separation units 102 sets a square region from the top to the 0.1h line as the first analysis region in the image to be processed.

  S303: The first profile determination unit 102a of both lung separation units 102 integrates the set first analysis region and calculates an average profile. This profile is the first profile.

  S304: The stripe division unit 102g of both lung separation units 102 divides the X-ray front image into N stripes.

  S305: The vertical position setting unit 102h of both lung separation units 102 initializes the number pointer k to 1.

  S306: The vertical position setting unit 102f of both lung separation units 102 determines whether or not the stripe to be processed is the last stripe. If it is the last stripe, the process proceeds to step S314. Otherwise, the process proceeds to step S307.

Since the processing of S307 to S313 is the same as that of the second embodiment, description thereof is omitted.
S314: The interpolation unit 102f of both lung separation units 102 interpolates the central position of both lungs in each stripe between the first and last stripes.

  S315: The interpolation unit 102f of both lung separation units 102 extrapolates the outside of the first and last stripe vertical center positions at the center position of the positions.

  S316: Output the detection result.

(Fourth embodiment)
FIG. 12 is a block diagram showing an example of a lung field detection processing system according to the fourth embodiment of the present invention and both lung separation units in the system.

  As in the first embodiment, this embodiment includes an image input unit 101, both lung separation units 102, and an output unit 103. Both lung separation units 102 are the same as those in the first embodiment. Different.

  The two lung separation units 102 include a first profile determination unit 102a, a first threshold determination unit 102c, a first boundary position determination unit 102e, a second profile determination unit 102b, a second threshold determination unit 102d, And a second boundary position determination unit 102f. Hereinafter, the both lung separation unit 102 will be described.

  The first profile determination unit 102a determines the average profile of the cervical region as in the first embodiment. As in the first embodiment, the first threshold value determination unit 102c determines the first threshold value based on the minimum value in the first profile center region (0.3w to 0.7w). The first boundary position determination unit 102e searches for a range having a value smaller than the first threshold in the first profile central region (0.3w to 0.7w), and sets both ends of the range as first left and right boundary positions.

  On the other hand, the vertical position setting unit 102h sets one vertical position from the top of the image. As soon as the central position of the vertical position is determined, the vertical position is increased by one line, and the processes 102b, 102d, 102f, and 102g are performed. In this way, the center position of each horizontal line is detected repeatedly to the bottom of the image.

  Similar to the second embodiment, the second profile determination unit 102b determines a second profile that tracks changes in the left and right lung field center lines at the set vertical position. The second threshold value determination unit determines the second threshold value based on the minimum value in the second profile center region (0.3w to 0.7w). The second boundary position determining unit 102f searches the second profile center region (0.3w to 0.7w) for a range having a value smaller than the second threshold, and sets both ends of the range as the second left and right boundary positions.

  The center position determination unit 102g compares the obtained first left boundary position and the second left boundary position, and stores the more right boundary as the left end point. On the other hand, the obtained first right boundary position and the second right boundary position are compared, and the left boundary is stored as the right end point. The center position of the left and right end points is the center of the left and right lung fields.

  In the present embodiment, the first profile determination unit 102a and the second profile determination unit 102b may smooth the obtained profiles with a low-pass filter.

  Similarly, when the first threshold value determining unit 102c and the second threshold value determining unit 102d determine the threshold value by using a high luminance difference at the boundary of the cervical region, the predetermined value determined empirically is added to the minimum value. Instead, a weighted average of the minimum value and the maximum value of the image or the maximum value of the profile may be used.

  Moreover, although the detected center line of both lungs may be output as it is, it may be output after being smoothed.

  Further, in this embodiment, as in the third embodiment, the image is divided into a large number of stripes, the second profile is determined in each stripe, the center position of both lungs is obtained, and both lungs in each stripe are determined. The lungs may be separated by interpolating the central position.

  FIG. 13 shows a flowchart of the present embodiment.

  S401: The image input unit 101 reads an X-ray front image to be processed into a memory.

  S402: The first profile determination unit 102a of both lung separation units 102 sets a square area from the top to the 0.1h line as the first analysis area in the image to be processed.

  S403: The first profile determination unit 102a of both lung separation units 102 integrates the set first analysis region and calculates an average profile. This profile is the first profile.

  S404: The first threshold value determination unit 102c of both lung separation units 102 searches the representative profile for a minimum value in the central region from 0.3w to 0.7w. A first threshold value T1 is set by adding a predetermined value empirically determined to the minimum value.

  S405: The central position determination unit 102e of both lung separation units 102 searches the central region from 0.3w to 0.7w in the first profile for a region having a value smaller than the first threshold value T1, and the left and right boundary positions of the region Is stored as the first boundary position.

  S406: The vertical position setting unit 102h of both lung separation units 102 initializes the position pointer k to 1.

  S407: The vertical position setting unit 102h of both lung separation units 102 determines whether or not the current vertical position follows the lowest end of the image. When tracing, the process proceeds to step S414. Otherwise, the processes in steps S408 to S413 are repeated.

  S408 to S410: As in the second embodiment, the second profile and the second threshold T2 are determined.

  S411: The second boundary position determination unit 102f of the both lung separation units 102 searches the central region from 0.3w to 0.7w in the second profile for a region having a value smaller than the second threshold T2, and The boundary position is stored as the second boundary position.

  S412: The central position determination unit 102g of both lung separation units 102 compares the obtained first left boundary position and the second left boundary position, and stores the more right boundary as the left end point. On the other hand, the obtained first right boundary position and the second right boundary position are compared, and the left boundary is stored as the right end point. The center position of the left and right end points is the center of the left and right lung fields.

  S413: The vertical position setting unit 102h of both lung separation units 102 increments the position pointer k and returns to step S407.

  In the present specification, all positive images have been described as examples. However, the polarities of the threshold value determination unit and the center position determination unit may be changed with respect to the negative image.

  Further, the present invention may be applied to a system constituted by a plurality of devices (for example, a host computer, an interface device, an imaging device, a web application, etc.) or an apparatus constituted by a single device.

  In order to operate various devices to realize the functions of the above-described embodiments, program codes of software for realizing the functions of the above-described embodiments are provided to an apparatus or a computer in the system connected to the various devices. What is implemented by operating the various devices according to a program supplied and stored in a computer (CPU or MPU) of the system or apparatus is also included in the scope of the present invention.

  In this case, the program code of the software itself realizes the functions of the above-described embodiments, and the program code itself and means for supplying the program code to the computer, for example, the program code are stored. The recorded medium constitutes the present invention. As a recording medium for storing the program code, for example, a flexible disk, a hard disk, an optical disk, a magneto-optical disk, a CD-ROM, a magnetic tape, a nonvolatile memory card, a ROM, or the like can be used. When the present invention is applied to the recording medium, the recording medium stores program codes corresponding to the flowcharts described above.

  Further, by executing the program code supplied by the computer, not only the functions of the above-described embodiments are realized, but also the OS (operating system) or other application software in which the program code is running on the computer, etc. It goes without saying that the program code is also included in the embodiment of the present invention even when the functions of the above-described embodiment are realized in cooperation with the embodiment.

  Further, after the supplied program code is stored in the memory provided in the function expansion board of the computer or the function expansion unit connected to the computer, the CPU provided in the function expansion board or function expansion unit based on the instruction of the program code Needless to say, the present invention includes a case where the functions of the above-described embodiment are realized by performing part or all of the actual processing.

Both lungs separation block diagram. The both lung separation part and the following lung field detection part block diagram. FIG. 6 is an analysis region of a typical profile of an X-ray chest front photograph in the first embodiment. Neck region profile and central location of both lungs. The flowchart of 1st Embodiment. The block diagram of 2nd Embodiment. Analysis area of the first and second profiles of the X-ray chest front photograph. The flowchart of 2nd Embodiment. The block diagram of 3rd Embodiment. The X-ray chest front photo divided stripes and the center position of each stripe. The flowchart of 3rd Embodiment. The block diagram of 4th Example. The flowchart of 4th Embodiment. Profile symmetry calculation. First and second profiles without one lung.

Claims (16)

  A medical image processing apparatus for detecting a central line of left and right lung fields of a chest simple X-ray front photograph, comprising: means for determining a representative profile in the chest simple X-ray front photograph; and a threshold value using the representative profile. A medical image processing apparatus comprising: means for setting; and means for determining a region including the center of the left and right lung fields using the threshold value and determining a center position of the left and right lung fields.   The means for determining the representative profile adds a region of a predetermined horizontal and vertical width to the upper end of the chest X-ray front photograph, and uses the obtained average horizontal profile as a representative profile. Item 2. The medical image processing apparatus according to Item 1.   The means for determining the representative profile includes means for determining a first profile for determining the center line location range of the left and right lung fields, and a second profile for tracking changes in the left and right lung field center lines at each vertical position. The means for determining and calculating the weighted average of the first profile and the second profile while moving the vertical position of the second profile to obtain a representative profile. The medical image processing apparatus described.   The means for determining the first profile adds a region of a predetermined horizontal and vertical width to an upper end of the chest simple X-ray front photograph, and uses the obtained average horizontal profile as the first profile. The medical image processing apparatus according to claim 3.   The means for determining the second profile includes adding a predetermined horizontal and vertical width area to the chest simple X-ray front photograph at a predetermined vertical position, and setting the obtained average horizontal profile as the second profile. The medical image processing apparatus according to claim 3, wherein the medical image processing apparatus is characterized.   6. The medical image processing apparatus according to claim 2, further comprising means for filtering the calculated first profile, second profile, or representative profile with a low-pass filter.   The means for setting a threshold value using the representative profile obtains a minimum value in a central region of the representative profile, and adds the value determined empirically from the minimum value as the threshold value. The medical image processing apparatus according to claim 1.   The means for setting a threshold value using the representative profile calculates a minimum value in a central region of the representative profile, calculates a maximum pixel value and a weighted average value of the image, and sets the threshold value. The medical image processing apparatus according to 1.   The means for setting a threshold value using the representative profile obtains a minimum value in a central region of the representative profile and calculates a maximum pixel value and a weighted average value of the profile as a threshold value. The medical image processing apparatus according to 1.   The means for determining the left and right lung field center position determines a region having a profile smaller than the threshold value to the left and right of the minimum value using the obtained threshold value, and sets the center of the region as the left and right lung field center position. The medical image processing apparatus according to claim 1.   The medical image processing apparatus according to claim 1, further comprising means for smoothing the center position of the left and right lung fields, or means for smoothing by interpolating the center position of the left and right lung fields.   A medical image processing apparatus for detecting a central line of left and right lung fields in a chest simple X-ray front photograph, wherein a first profile for determining a center line location range of left and right lung fields is determined in the chest simple X-ray front photograph Means for determining the first threshold based on the first profile, means for determining the left and right boundary positions of the left and right lung field center line using the first threshold, and left and right at each vertical position Means for determining a second profile that tracks changes in the lung center line, means for determining a second threshold based on the second profile, and left and right lungs at each vertical position using the second threshold Medical image processing, comprising: means for determining an inner boundary position of a field; and means for determining a central position of the left and right lung fields based on a left and right boundary position of the center line location range and an inner boundary position of the left and right lung fields apparatus.   A medical image processing method for detecting a central line of left and right lung fields of a chest simple X-ray front photograph, comprising: determining a representative profile in the chest simple X-ray front photograph; and using the representative profile to set a threshold value A medical image processing method comprising: a setting step; and a step of determining a region including the center of the left and right lung fields using the threshold value, and determining a center position of the left and right lung fields.   A medical image processing method for detecting a central line of left and right lung fields in a chest simple X-ray front photograph, wherein a first profile for determining a center line location range of left and right lung fields is determined in the chest simple X-ray front photograph A step of determining a first threshold value based on the first profile, a step of obtaining a right and left boundary position of the left and right lung field center line using the first threshold value, and a right and left at each vertical position Determining a second profile that tracks changes in the midfield of the lung, determining a second threshold based on the second profile, and using the second threshold, the left and right lungs at each vertical position A medical image, comprising: determining an inner boundary position of the field; and determining a central position of the left and right lung fields based on the left and right boundary positions of the centerline location range and the inner boundary positions of the left and right lung fields. Processing method. A program for realizing a medical image processing method for detecting a center line of left and right lung fields of a chest simple X-ray front photograph, comprising: a program code for determining a representative profile for the chest simple X-ray front photograph; and the representative And a program code for determining a region including the center of the left and right lung fields using the threshold value, and determining a center position of the left and right lung fields. Program to do. A program for realizing a medical image processing method for detecting a center line of left and right lung fields of a chest simple X-ray front photograph, wherein a first line for determining a center line location range of left and right lung fields in the chest simple X-ray front photograph The left and right boundary positions of the left and right lung field center line using the first threshold and the program code of the step of determining the profile of the program, the program code of the step of determining the first threshold based on the first profile, and the first threshold , A program code for determining a second profile for tracking a change in the left and right lung field center line at each vertical position, and a step for determining a second threshold based on the second profile Using the second threshold value, the program code of the step of obtaining the inner boundary position of the left and right lung fields at each vertical position, and the left and right boundaries of the center line location range Based on the inner boundary position of location and the left and right lung field, the program characterized by comprising a program code for determining a lateral lung center position.

Images