JP2009273601A - Apparatus, method, and program for distinguishing image - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To suppress a cost of distinguishing whether a front chest image is an AP (Antero-Posterior) image or a PA (Postero-Anterior) image, and also to reduce burden on a user. <P>SOLUTION: An apparatus for distinguishing an image extracts, from a front chest image 20, a central line 22 located at the center of a human body in the transverse direction. Regions each of which has a width of one fifth of the horizontal image size W of the front chest image 20, is located on both sides of the extracted central line 22, and is also located at the central part of the front chest image 20 when its vertical image size H is divided into three are set as analyzed regions 23 and 24. A difference between an average pixel value of the left analyzed region 23 and an average pixel value of the right analyzed region 24 is calculated as a feature S. The calculated feature S is then compared with a threshold value TH, and whether the front chest image 20 is an AP (Antero-Posterior) image or a PA (Postero-Anterior) image is determined on the basis of the comparison result. The apparatus for distinguishing an image carries out self-learning by using front chest images 20 within a facility as teacher images and modifies the threshold value TH. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an image discrimination device, an image discrimination method, and a program.

  In medical diagnosis, a radiation image obtained by photographing a site to be diagnosed with radiation such as X-rays is widely used. Conventionally, a technique using an intensifying screen and a film has been used, but in recent years, a monitor diagnosis using a digital X-ray imaging apparatus that does not use a film has been widely spread.

  Chest frontal imaging with simple X-rays is usually performed by a PA (Postero-Anterior) imaging method in which radiation is emitted from the back side. However, since it is difficult to perform imaging with PA imaging for patients who cannot perform imaging in a standing position, imaging is performed with AP (Antero-Posterior) imaging that irradiates radiation from the chest side. Therefore, an image captured by the PA imaging method (hereinafter referred to as a PA image) and an image captured by the AP imaging method (hereinafter referred to as an AP image) are reversed on the human body. .

  When a PA image and an AP image are mixed, it may cause a diagnosis error. Therefore, when a doctor diagnoses a chest front image with a film or a monitor, the heart of the PA image is on the right side of the image. The left and right sides of the image are reversed so that they are positioned. That is, the diagnosis is performed by aligning the images in the direction in which the doctor faces the patient. Specifically, in the case of a PA image, the film is turned over and attached to the Schacus Ten, or the image is displayed on a monitor after being subjected to a horizontal reversal process by image processing.

  Further, in an X-ray imaging system in which an imaging condition such as an imaging direction indicating whether the image is an AP image or a PA image is input in advance and the imaging condition information is embedded in the image and displayed, according to the imaging direction. A technique for determining whether to reverse an image has been proposed (see Patent Document 1).

Also, when performing radiography, a patient is photographed with a CCD (Charge Coupled Device) camera, etc., and an image photographed with a CCD camera or the like is compared with a template image associated with a patient's imaging region, imaging direction, etc. A radiation imaging system has been proposed that estimates the imaging direction and discriminates whether the chest front image that has been automatically radiographed is an AP image or a PA image (see Patent Document 2).
JP 2000-107159 A JP 2004-283258 A

  However, when the imaging conditions have to be input in advance as in the X-ray imaging system described in Patent Document 1, the man-hours are increased and complicated, and the AP may be caused by an input error in the imaging conditions. There is a risk of erroneous determination of whether the image is a PA image. In addition, the radiation imaging system described in Patent Document 2 has a problem in that it is necessary to include a CCD camera and the like, which complicates the system and is very expensive.

  The present invention has been made in view of the above-described problems in the prior art, and reduces the burden on the user while suppressing the cost for determining whether the chest front image is an AP image or a PA image. Is an issue.

  In order to solve the above problem, the invention according to claim 1 is an AP image obtained by irradiating radiation from the chest side of the chest front image based on the chest front image obtained by radiography. Or an image discriminating apparatus comprising control means for discriminating whether the image is a PA image taken by irradiating radiation from the back side, wherein the control means is a center in the left-right direction of the human body from the chest front image. A center line located at the center of the extracted center line is extracted, and an analysis target area having a predetermined width is set on each of both sides of the extracted center line, and one of the two target analysis target areas is set. The difference between the average pixel value and the average pixel value in the other analysis target area is calculated, the calculated difference between the average pixel values is compared with a predetermined threshold value, and the chest front image is calculated based on the comparison result. Is A To determine whether the PA image is an image.

  According to a second aspect of the present invention, the image discriminating apparatus according to the first aspect further includes a display unit, and the control unit determines that the chest front image is an AP image when the chest front image is determined to be an AP image. Is displayed on the display means as it is, and when it is determined that the chest front image is a PA image, the chest front image is horizontally reversed and displayed on the display means.

  According to a third aspect of the present invention, in the image discriminating apparatus according to the second aspect, correct / incorrect information indicating whether or not the discrimination result as to whether the chest front image is an AP image or a PA image is correct is input. The control means corrects the determination result based on correct / incorrect information input from the operation means.

  According to a fourth aspect of the present invention, in the image discriminating apparatus according to any one of the first to third aspects, the control means includes a difference between average pixel values calculated from the chest front image and the chest front image. The threshold value is changed based on the determination result of whether the image is an AP image or a PA image.

  According to a fifth aspect of the present invention, based on a chest front image obtained by radiography, the chest front image is an AP image taken by irradiating radiation from the chest side by the control means, or the back side A method of determining whether the image is a PA image captured by irradiating radiation from the center, wherein the control means extracts a center line located at the center in the left-right direction of the human body from the chest front image A step of setting an analysis target area having a predetermined width on each of both sides of the extracted center line by the control means; and one of the two analysis target areas set by the control means Calculating the difference between the average pixel value in the analysis target area and the average pixel value in the other analysis target area, and the difference between the calculated average pixel values by the control means Comparing a predetermined threshold value, by the control unit, comprising the steps of the front chest image it is determined whether PA image is an AP image based on the comparison result.

  In the invention according to claim 6, the computer is an AP image obtained by irradiating radiation from the chest side based on a chest front image obtained by radiography, or from the back side. A program for causing a computer to function as a means for determining whether or not a PA image is taken by irradiating radiation, and extracts a center line located at the center of the human body in the left-right direction from the chest front image Means for setting an analysis target area having a predetermined width on each of both sides of the extracted center line, and an average pixel value in one analysis target area and the other of the two set analysis target areas Means for calculating a difference from the average pixel value in the analysis target area, means for comparing the difference between the calculated average pixel value and a predetermined threshold, and based on the comparison result There the front chest image is programmed to function as a unit, for determining whether the PA image is an AP image.

  According to the first, fifth, and sixth aspects of the present invention, it is possible to determine whether the chest front image is an AP image or a PA image based on the chest front image. The burden can be reduced.

  According to the second aspect of the present invention, when it is determined that the chest front image is a PA image, the chest front image is reversed left and right, so that the chest front image can be displayed in the same orientation.

  According to the third aspect of the present invention, when the determination result is incorrect, the determination result can be corrected according to the user's determination.

  According to the fourth aspect of the present invention, the threshold value can be changed to a threshold value suitable for a chest front image for which it is determined whether the image is an AP image or a PA image.

Hereinafter, an embodiment of an image diagnosis support apparatus to which an image determination apparatus according to the present invention is applied will be described.
FIG. 1 shows a functional configuration of the image diagnosis support apparatus 10. As shown in FIG. 1, the diagnostic imaging support apparatus 10 includes a control unit 11, an operation unit 12, a display unit 13, a communication unit 14, a ROM (Read Only Memory) 15, and a storage unit 16, and each unit is a bus. 17 is connected.

  The control unit 11 includes a CPU (Central Processing Unit), a RAM (Random Access Memory), and the like, and comprehensively controls the processing operation of each unit of the diagnostic imaging support apparatus 10. Specifically, the CPU reads various processing programs stored in the ROM 15 in response to an operation signal input from the operation unit 12 or an instruction signal received by the communication unit 14, and a work piece formed in the RAM. Expand to the area and perform various processes in cooperation with the program.

  The operation unit 12 includes a keyboard having cursor keys, numeric input keys, various function keys, and the like, and a pointing device such as a mouse, and controls operation signals input by keyboard operations or mouse operations. 11 is output. For example, the operation unit 12 is used when inputting correct / incorrect information indicating whether or not the determination result of whether the chest front image 20 (see FIG. 2A) is an AP image or a PA image is correct.

  The display unit 13 is configured by an LCD (Liquid Crystal Display) and displays various screens based on display data input from the control unit 11.

  The communication unit 14 is an interface that transmits / receives data to / from an external device.

  The ROM 15 is composed of a nonvolatile semiconductor memory or the like, and stores a control program, parameters and files necessary for executing the program, and the like.

  The storage unit 16 is configured by a hard disk or the like and stores various data. Specifically, the storage unit 16 stores image data of a chest front image 20 (see FIG. 2A) obtained by imaging a human chest with radiation. The chest front image 20 may be received via the communication unit 14 from the modality where the imaging was performed, or read out via the communication unit 14 from the image server that stores the chest front image 20 after imaging. It may be a thing. The chest front image 20 includes an AP image captured by irradiating radiation from the chest side and a PA image captured by irradiating radiation from the back side. In addition, the storage unit 16 stores a threshold value TH used for determining whether the chest front image 20 is an AP image or a PA image.

  An example of the chest front image 20 is shown in FIG. In the present embodiment, a feature amount S (see formula (1) described later) that is characterized by whether the heart 21 is on the left or right side of the chest front image 20 is calculated from the chest front image 20, and the feature amount S is calculated. Based on this, it is determined whether the chest front image 20 is an AP image or a PA image. That is, when the heart 21 is on the left side of the chest front image 20, it is determined as a PA image, and when the heart 21 is on the right side of the chest front image 20, it is determined as an AP image.

  The control unit 11 extracts a center line 22 (see FIG. 3) located at the center of the human body in the left-right direction from the chest front image 20. The center line 22 does not mean the center of the image of the chest front image 20, but corresponds to the center position of the photographed human body in the left-right direction, that is, the center of the spine.

FIG. 2B is a diagram showing a horizontal profile of the chest front image 20 shown in FIG. The horizontal profile is a position in the X-axis direction (see FIG. 2A) of the chest front image 20 on the horizontal axis, and a pixel value in the Y-axis direction (see FIG. 2A) of the chest front image 20 on the vertical axis. (Signal value) is averaged. The control unit 11 obtains a position X _{0 at} which the average pixel value in the Y-axis direction is minimum in the horizontal profile shown in FIG. 2B, and extracts this position as the center line 22 shown in FIG.

  As shown in FIG. 4, the control unit 11 is a region having a width of 1/5 of the horizontal image size W of the chest front image 20 on each side of the extracted center line 22 in the chest front image 20. And the area | region of the center part which divided the vertical image size H of the chest front image 20 into 3 equal parts is set as the analysis object areas 23 and 24.

The control unit 11 calculates an average pixel value in each of the left analysis target area 23 and the right analysis target area 24 among the two set analysis target areas 23 and 24, and the left analysis target area 24 The difference between the average pixel value in 23 and the average pixel value in the analysis target area 24 on the right side is calculated as the feature amount S.
Feature S = (Average pixel value of left analysis target area 23) − (Average pixel value of right analysis target area 24) (1)

  Generally, of the analysis target regions 23 and 24, the region including the heart 21 has a smaller average pixel value than the region not including the heart 21. Therefore, for example, when the value of the feature value S is positive, the image is highly likely to be an AP image, and when the value of the feature value S is negative, the image is likely to be a PA image.

  The control unit 11 compares the calculated feature amount S with the threshold value TH, and determines whether the chest front image 20 is an AP image or a PA image based on the comparison result. Specifically, the control unit 11 determines that the chest front image 20 is an AP image when the feature amount S is greater than or equal to the threshold value TH, and if the feature amount S is less than the threshold value TH, the chest front image. It is determined that 20 is a PA image.

With reference to FIG. 5, a method of setting the initial value of threshold TH (threshold TH ₀ ) will be described.
First, the feature amount S is calculated from a plurality of AP images and PA images randomly selected as sample images. Here, the ratio of the number of AP images and PA images in the sample image is 1: 1.

For each of the AP image and the PA image, it is assumed that the distribution of the feature amount S is a one-dimensional normal distribution (mixed normal distribution model: Gaussian Mixture Model) and can be expressed by the following equation (2). Here, x is the feature quantity S, μ is the average of x, and σ ² is the variance.

The values of average μ and variance σ ² are obtained by the following formulas (3) and (4).

By substituting the feature amount S of the AP image of the sample image into x _i of Equations (3) and (4), the average μ and variance σ ² of the AP image of the sample image are calculated. A normal distribution of the AP image is estimated. Similarly, the feature value S of the PA image of the sample image is substituted for x _i in the equations (3) and (4), and the average μ and variance σ ² of the PA image of the sample image are calculated. Thus, the normal distribution of the PA image of the sample image is estimated.

As described above, the normal distribution of the AP image and PA image of the sample image is estimated, and the feature amount distribution of the sample image is created. Then, the intersection point of the normal distribution of the estimated AP image and PA image is set as a threshold value TH ₀ . Alternatively, when the normal distribution of the AP image and the PA image does not intersect, the intermediate point between the two distributions is set as the threshold value TH ₀ . The threshold value TH ₀ obtained in this way is stored in the storage unit 16 in advance.

  When determining that the chest front image 20 is an AP image, the control unit 11 displays the chest front image 20 on the display unit 13 as it is, and when determining that the chest front image 20 is a PA image, the front of the chest is displayed. The image 20 is reversed left and right and displayed on the display unit 13.

  The control unit 11 receives input of correct / incorrect information indicating whether or not the determination result of whether the chest front image 20 is an AP image or a PA image is correct from the operation unit 12. The control unit 11 corrects the determination result based on the correct / incorrect information input from the operation unit 12.

  The control unit 11 labels “AP” or “PA” with respect to each of the chest front images 20 that have been determined as an AP image or a PA image, depending on the determination result. In addition, when the determination result is corrected based on correct / incorrect information, the control unit 11 performs labeling by reflecting the correction result. In the labeling, information indicating “AP” or “PA” may be attached to the chest front image 20 as additional information, or information indicating “AP” or “PA” with each of the chest front image 20. The data may be stored in a table that associates.

  The control unit 11 acquires, as a teacher image, the chest front image 20 in which it is determined whether the image is an AP image or a PA image, and the feature amount S calculated from the teacher image and the teacher image are AP images. Or the PA image, the threshold value TH is changed. The teacher image is a learning image used for changing the threshold value TH.

Here, with reference to FIG. 6, a method for setting the threshold value TH based on learning by a teacher image in a facility where the image diagnosis support apparatus 10 is installed will be described.
The control unit 11 assigns the feature amount S of the AP image in the facility to x _i in the above formulas (3) and (4), calculates the average μ and variance σ ² of the AP image in the facility, The normal distribution of the AP image in the facility is estimated by (2). Similarly, the control unit 11 calculates the average μ and variance σ ² of the PA image in the facility by substituting the feature amount S of the PA image in the facility for x _i in the above formulas (3) and (4). Then, the normal distribution of the PA image in the facility is estimated by the above equation (2).

As described above, the control unit 11 estimates the normal distribution of the AP image and the PA image using the chest front image 20 in the facility as a teacher image, and creates a feature amount distribution in the facility. And the control part 11 makes the threshold value TH the intersection of the estimated normal distribution of AP image and PA image. Alternatively, when the normal distribution of the AP image and the PA image does not intersect, the intermediate point between the two distributions is set as the threshold value TH.
The control unit 11 determines whether the chest front image 20 is an AP image or a PA image by applying a new threshold value TH after the change.

Next, the operation in the diagnostic imaging support apparatus 10 will be described.
7 and 8 are flowcharts showing the image discrimination process. This process is realized by software processing in cooperation with the CPU of the control unit 11 and the program stored in the ROM 15.

The initial threshold value TH ₀ is set in advance in the storage unit 16 (step A1). The threshold TH ₀ is calculated by calculating the feature amount S from the AP image and PA image of the sample image according to the above equation (1), and obtaining the feature amount distribution of the sample image according to the above equations (2) to (4). Are set in advance.

The chest front image 20 to be diagnosed is selected, and the control unit 11 generates a horizontal profile of the chest front image 20 as shown in FIG. Then, the control unit 11 extracts the position X _{0 at} which the average pixel value in the Y-axis direction is minimum in the horizontal profile as the center line 22 (see FIG. 3) (step A2).

  Next, as shown in FIG. 4, the control unit 11, in the chest front image 20, has an area having a width of 1/5 of the horizontal image size W of the chest front image 20 on each side of the extracted center line 22. And the area | region of the center part which divided the vertical image size H of the chest front image 20 into 3 equally is set as analysis object area | regions 23 and 24 (step A3).

  Next, the control unit 11 calculates an average pixel value in the left analysis target region 23 out of the two analysis target regions 23 and 24 set in the chest front image 20, and calculates the average pixel in the right analysis target region 24. A value is calculated. Then, the difference between the average pixel value in the left analysis target region 23 and the average pixel value in the right analysis target region 24 is calculated as the feature amount S by the control unit 11 according to the above equation (1) (step A4). .

Next, the control unit 11 compares the calculated feature quantity S with the threshold value TH, and determines whether or not the feature quantity S is equal to or greater than the threshold value TH (step A5). The threshold value TH ₀ is used before learning using the chest front image 20 in the facility as a teacher image.

  As a result of the comparison in step A5, when the feature amount S is equal to or greater than the threshold TH (step A5; YES), the control unit 11 determines that the chest front image 20 is an AP image (step A6), and the chest front The image 20 is displayed as it is on the display unit 13 (step A7).

  On the other hand, when the feature amount S is less than the threshold value TH (step A5; NO), the control unit 11 determines that the chest front image 20 is a PA image (step A8), and the chest front image 20 is horizontally reversed. And displayed on the display unit 13 (step A9).

  After step A7 or step A9, the control unit 11 receives input of correct / incorrect information indicating whether or not the determination result of whether the chest front image 20 is an AP image or a PA image is correct from the operation unit 12 ( Step A10).

  When there is no correction by the user regarding the determination result of whether the chest front image 20 is an AP image or a PA image (step A11; NO), the control unit 11 causes the chest front image 20 to be determined according to the determination result. Are labeled “AP” or “PA” (step A12).

  On the other hand, if the determination result of whether the chest front image 20 is an AP image or a PA image is corrected by the user (Step A11; YES), the chest front image 20 is currently displayed by the control unit 11. The image is reversed left and right from the displayed state and displayed on the display unit 13 (step A13). Then, the determination result is corrected by the control unit 11, and “AP” or “PA” is labeled on the chest front image 20 in accordance with the corrected determination result (step A14).

  After step A12 or step A14, as shown in FIG. 8, the controller 11 adds labeled AP images and PA images as teacher images (step A15). Based on the feature amount S calculated from the teacher image by the control unit 11 and the determination result of whether the teacher image is an AP image or a PA image, the normal distribution of the AP image in the facility and the PA image And the threshold TH is changed (automatic learning) (step A16). Thereafter, when determining whether the chest front image 20 is an AP image or a PA image, the changed threshold TH is applied (step A17). In the present embodiment, learning is performed for each image. The control unit 11 has the feature amount S of the sample image, and performs learning by adding a teacher image to the data. Instead of learning each time one image is added, learning may be performed using the teacher image after the teacher image is stored to some extent.

  Here, when there is a chest front image 20 to be diagnosed next (step A18; YES), the process returns to step A2 in FIG. 7 and the processes of steps A2 to A17 are repeated. In step A5, it is determined whether the feature amount S is equal to or greater than the threshold value TH using the changed threshold value TH.

  In step A18, when there is no next chest front image 20 to be diagnosed (step A18; NO), the image discrimination process is terminated.

  As described above, according to the diagnostic imaging support apparatus 10, it is possible to automatically determine whether the chest front image 20 is an AP image or a PA image based on the chest front image 20. The burden can be reduced. In addition, since it is not necessary to newly add a special device for determining whether the image is an AP image or a PA image, the cost can be reduced.

  Further, since the center line 22 located at the center of the human body in the left-right direction is extracted from the chest front image 20, and the analysis target areas 23 and 24 are set based on the center line 22, the center position and the chest front image 20 on the human body are set. Even when the center position is shifted, it is possible to accurately determine whether the chest front image 20 is an AP image or a PA image.

  Further, when it is determined that the chest front image 20 is a PA image, the chest front image 20 is reversed left and right, so that the chest front image 20 can be displayed in the same orientation.

  If the determination result of whether the chest front image 20 is an AP image or a PA image is incorrect, the determination result can be corrected according to the user's determination.

  Further, since the feature amount distribution is obtained using the chest front image 20 in which the image diagnosis support apparatus 10 determines whether the image is an AP image or a PA image as a teacher image, the chest front image 20 actually handled in the facility is obtained. It is possible to change to a threshold value suitable for. Thereby, the discrimination accuracy in each facility can be improved.

  The description in the above embodiment is an example of the image discrimination apparatus according to the present invention, and the present invention is not limited to this. The detailed configuration and detailed operation of each part constituting the apparatus can be appropriately changed.

  For example, in the above embodiment, learning is performed on the assumption that the feature amount S follows a normal distribution for each of the AP image and the PA image. However, learning is performed using a neural network, a support vector machine, AdaBoost, or the like. Thus, customization may be performed so that the optimum AP image / PA image discrimination processing is performed in each facility.

  In the above embodiment, when the analysis target areas 23 and 24 are set in the chest front image 20, an area having a width of 1/5 of the horizontal image size W is taken in the horizontal direction of the chest front image 20. Although an area having a width of 1/3 of the vertical image size H is taken in the vertical direction of the image 20, the average pixel value of the analysis target areas 23 and 24 varies depending on the position where the heart 21 exists. If so, the sizes of the analysis target areas 23 and 24 can be arbitrarily changed.

It is a block diagram which shows the functional structure of an image diagnosis assistance apparatus. (A) is a figure showing an example of a chest front image. (B) is a figure which shows the horizontal profile of the chest front image shown to Fig.2 (a). It is a figure for demonstrating the centerline of a chest front image. It is a figure for demonstrating the setting method of an analysis object area | region. It is a figure which shows the feature-value distribution of a sample image. It is a figure which shows the feature-value distribution of the chest front image in a plant | facility. It is a flowchart which shows an image discrimination | determination process. It is a flowchart which shows an image discrimination | determination process.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Image diagnosis assistance apparatus 11 Control part 12 Operation part 13 Display part 14 Communication part 15 ROM
16 Storage unit 17 Bus 20 Chest front image 21 Heart 22 Center lines 23 and 24 Analysis target region

Claims (6)

Based on the chest front image obtained by radiography, the chest front image is an AP image taken by irradiating radiation from the chest side, or a PA image taken by irradiating radiation from the back side. An image discriminating apparatus having a control means for discriminating whether or not
The control means includes
A center line located at the center of the human body in the left-right direction is extracted from the chest front image,
Set an analysis target area with a predetermined width on each of both sides of the extracted center line,
Calculate the difference between the average pixel value in one analysis target area and the average pixel value in the other analysis target area of the two set analysis target areas,
Compare the calculated average pixel value difference with a predetermined threshold,
An image discrimination device that discriminates whether the chest front image is an AP image or a PA image based on the comparison result. A display means,
The control means displays the chest front image as it is on the display means when it is determined that the chest front image is an AP image, and the chest when it determines that the chest front image is a PA image. The front image is reversed left and right and displayed on the display means.
The image discrimination device according to claim 1. An operation unit for inputting correct / incorrect information indicating whether or not the determination result of whether the chest front image is an AP image or a PA image is correct;
The control means corrects the determination result based on correct / incorrect information input from the operation means.
The image discrimination device according to claim 2. The control means changes the threshold based on a difference between average pixel values calculated from the chest front image and a determination result of whether the chest front image is an AP image or a PA image.
The image discrimination device according to any one of claims 1 to 3. Based on the chest front image obtained by radiography, the chest front image is an AP image taken by irradiating radiation from the chest side or taken by irradiating radiation from the back side by the control means. An image determination method for determining whether the image is a PA image,
A step of extracting a center line located at the center of the human body in the left-right direction from the chest front image by the control means;
A step of setting an analysis target area having a predetermined width on each of both sides of the extracted center line by the control means;
Calculating a difference between an average pixel value in one analysis target region and an average pixel value in the other analysis target region of the set two analysis target regions by the control unit;
Comparing the difference between the calculated average pixel values with a predetermined threshold by the control means;
Determining by the control means whether the chest front image is an AP image or a PA image based on the comparison result;
Discriminating method including: Based on the chest front image obtained by radiography, the computer is an AP image taken by irradiating radiation from the chest side, or PA taken by irradiating radiation from the back side A program for functioning as a means for determining whether an image is an image,
The computer,
Means for extracting a center line located at the center of the human body in the horizontal direction from the chest front image;
Means for setting an analysis target region having a predetermined width on each of both sides of the extracted center line;
Means for calculating a difference between an average pixel value in one analysis target area and an average pixel value in the other analysis target area of the set two analysis target areas;
Means for comparing the difference between the calculated average pixel values with a predetermined threshold;
Means for determining whether the chest front image is an AP image or a PA image based on the comparison result;
Program to function as.

Images