JP2002330949A - Abnormal shadow candidate output system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To output an image suitable for detailed observation by giving image emphasizing processing also to a shadow important for diagnosis when the shadow important for diagnosis exists outside of a detected abnormal shadow candidate. SOLUTION: A local image preparation means 20 inputs positional information D1 showing the existing area of the abnormal shadow candidate detected based on entire image data P1 from an abnormal shadow candidate detection means 10, extracts an area obtained by expanding the existing area according to a prescribed expansion condition as a local area based on the inputted positional information D1 and performs image emphasizing processing to the image data of the extracted local area to prepare local image data P2. The local image data P2 prepared by the means 20 is inputted together with the positional information D1 to display a local image P2' and an entire image P1'.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an abnormal shadow candidate output system, and more particularly to a system for performing a predetermined image enhancement process on an image of a local region including an abnormal shadow candidate detected based on radiation image data of a subject. The present invention relates to an abnormal shadow candidate output system that outputs an abnormal shadow candidate output together with a radiation image of a subject.

[0002]

2. Description of the Related Art Conventionally, in the medical field, it has been necessary to interpret a radiographic image of a subject, find a lesion, and observe the state of the lesion to diagnose the presence or absence of a disease and the progress of the disease. Generally done. However, interpretation of radiographic images is influenced by the experience of the interpreter and the level of image interpretation ability, and is not always objective.

[0003] For example, in mammography (diagnosis radiographic image of a breast as a subject) photographed for the purpose of examining breast cancer, the image shows a tumor shadow or microcalcification shadow which is one of the features of a cancerous portion. It is necessary to detect the abnormal shadow, but it is not always possible for the reader to specify the range of the abnormal shadow accurately. For this reason, it has been demanded to accurately detect abnormal shadows such as tumor shadows and microcalcification shadows without depending on the skill of the reader.

In response to this demand, abnormal shadow candidate detection is performed by automatically detecting, using a computer, abnormal shadow candidates in an image represented by the image data based on image data of a subject acquired as a diagnostic image. A processing system (computer-assisted image diagnostic apparatus) has been proposed (Japanese Patent Laid-Open No. 8-29)
4479, JP-A-8-287230, etc.). This abnormal shadow candidate detection processing system automatically detects abnormal shadow candidates using a computer based on the density distribution characteristics and morphological characteristics of abnormal shadows, and is mainly used to detect tumor shadows. An abnormal shadow candidate is detected by using a suitable iris filter process or a morphological filter process suitable mainly for detecting microcalcification shadows.

[0005] The iris filter process compares a iris filter output value representing the maximum value of the degree of concentration of the density gradient of the image signal with a predetermined threshold value, thereby obtaining a tumor, which is one of the characteristic forms of breast cancer in the image. On the other hand, morphological filter processing is a technique that is effective for detecting a shadow candidate area. By comparing the output value with a predetermined threshold value, this is an effective method for detecting a candidate region of microcalcification shadow, which is one of the characteristic forms of breast cancer, in an image.

In this abnormal shadow candidate detection processing system, a local region including the detected abnormal shadow candidate is set as a region of interest (hereinafter referred to as ROI).
Specialized image processing such as enhancement is performed on the ROI image in accordance with the attribute of the abnormal shadow candidate (whether a tumor shadow or a microcalcification shadow) included in the OI, and the entire image of the original image is processed. Also, predetermined image processing is performed to improve image interpretation performance, and the entire image that has been subjected to image processing and the ROI image that has been subjected to the specialized image processing have been defined as one image.
The image is output to an image display means such as an RT or a printing means such as a laser printer.

[0007]

By the way, what is detected by the above-mentioned iris filter processing is the core (center portion) of the tumor, but the form of the tumor shadow in the radiographic image is not limited to that of the core alone. In some cases, the core has a linear structure called a spicula. The spicula extends radially outward from the periphery of the core and is distributed. In particular, it is said that the tumor is highly likely to be malignant when the tumor is accompanied by the spicula. For this reason, in diagnosing a tumor, it is important to observe not only the core but also the spicula distributed around the core in detail, and in the above-described abnormal shadow candidate detection processing system, the ROI image including the abnormal shadow candidate is obtained. At the time of display, it is desirable to perform specialized image processing such as emphasis processing on the spicula together with the above-mentioned cores and display them.

On the other hand, regarding microcalcification shadows, there is no important element for diagnosis outside the detected microcalcification shadow candidates, but considering the ease of interpretation, only microcalcification shadows are considered. Instead, it is desirable to display the image with a margin by spreading it slightly outside.

The present invention has been made in view of the above circumstances, and when there is a shadow important for diagnosis outside a detected abnormal shadow candidate, specialization such as enhancement processing is performed on the shadow important for the diagnosis. It is possible to output an image suitable for detailed observation by applying the image processing that has been performed, and it is suitable for observation interpretation even when there is no shadow important for diagnosis outside the abnormal shadow candidate. It is an object of the present invention to provide an abnormal shadow candidate output system capable of outputting an image.

[0010]

An abnormal shadow candidate output system according to the present invention extracts a local region including an abnormal shadow candidate detected from a radiation image based on radiation image data representing a radiation image of a subject and extracts the local region. A local image creating unit that creates local image data by performing predetermined image processing on image data of the local region, and a local image and a radiation image data based on the local image data and the radiation image data created by the local image creating unit. In the abnormal shadow candidate output system including an image output unit that outputs a radiation image, the local image creating unit extracts, as a local region, an area in which a minimum local area including the abnormal shadow candidate is enlarged according to a predetermined expansion condition. It is characterized by being.

Here, the abnormal shadow candidate means, for example, a candidate region of a tumor shadow detected by the iris filter process, a candidate region of a microcalcification shadow detected by the morphology filter process, and the like. Further, a candidate area may be further detected (selected) after the degree of malignancy or the like of each detected candidate area is evaluated based on a predetermined evaluation criterion. That is, any type may be used as long as it is detected as a candidate for an abnormal shadow based on radiation image data representing a radiation image of the subject. Here, it is particularly preferable to target a tumor shadow candidate.

The image output means means display means such as a CRT or printing means such as a printer. In outputting the local image and the radiation image, the local image and the radiation image may be laid out as one image and output simultaneously, or may be output separately.

The predetermined image processing may be any kind of image processing, and preferable image processing includes, for example, image enhancement processing such as gradation processing, frequency processing, and enlargement processing.

The minimum local region including the abnormal shadow candidate is, for example, a region of the detected abnormal shadow candidate itself or a polygon such as a rectangle circumscribing the detected abnormal shadow candidate.
It means a region surrounded by a circle, an ellipse, or the like.

The predetermined extension condition is an extension multiple for increasing the minimum local region by a predetermined factor or an extension width for increasing the minimum local region outward by a predetermined length, in a range of 1.2 to 3.0 times. It is desirable to use an expansion multiple of 0.5 to 3.0 cm, and more preferably to use an expansion multiple of 1.5 to 2.0 times. Here, the magnification is based on the similarity ratio, and the length (cm) is based on an actually measured value (not the length in the image) in the subject. The range of the extension condition is determined as a value suitable for observing the spicula distributed around the tumor shadow candidate (core) detected by the predetermined detection processing.

The "region in which the minimum local region including the abnormal shadow candidate is enlarged according to a predetermined extension condition"
This means an area in which the existence range is extended by a predetermined value within the range of the extension condition. For example, when the predetermined extension condition is set to 2.0 times, the similarity ratio having the same center as the center of the existence range The area is 2.0 times (area ratio is 4.0 times). The predetermined expansion condition is not limited to the same value in all directions, but may use different values in each direction. For example, when the existence range is rectangular, the horizontal expansion condition is increased by a factor of 1.5 (or 1.0c).
m) or the like, and the vertical expansion condition may be 2.0 times (or 2.0 cm).

In the second abnormal shadow candidate output system according to the present invention, a local region including an abnormal shadow candidate detected based on radiation image data representing a radiation image of a subject is extracted from the radiation image. A local image creating unit that creates a local image data by performing a predetermined image enhancement process on the image data of the local region; and a local image and a radiation image data based on the local image data and the radiation image data created by the local image creating unit. An abnormal shadow candidate output system including an image output unit that outputs a radiation image, further comprising a determining unit that determines a type of the abnormal shadow candidate, wherein the local image creating unit is configured to determine the type of the abnormal shadow candidate determined by the determining unit. , The minimum station including the abnormal shadow candidate in accordance with a predetermined extension condition predetermined for each type of abnormal shadow candidate It is characterized in that extracts a region to an extended region as a local region.

Here, the abnormal shadow candidate means, for example, a candidate region of a tumor shadow detected by the iris filter process, a candidate region of a microcalcification shadow detected by the morphology filter process, and the like. Further, a candidate area may be further detected (selected) after the degree of malignancy or the like of each detected candidate area is determined based on a predetermined determination criterion. That is, any type may be used as long as it is detected as a candidate for an abnormal shadow based on radiation image data representing a radiation image of the subject.
The type of abnormal shadow candidate means a type classified by characteristics such as a shape of a shadow, such as a tumor shadow candidate and a microcalcification shadow candidate.

The determination means for determining the type of the abnormal shadow candidate may be, for example, a type for automatically determining the type of the abnormal shadow candidate in the radiation image from the characteristics such as the shape of the abnormal shadow candidate, or the type of each abnormal shadow candidate. The determination may be made by inputting from an external device or the like. That is, any type can be used as long as the type of the abnormal shadow candidate can be recognized.

The "predetermined expansion condition predetermined for each type of abnormal shadow candidate" is defined as an expansion multiple or minimum expansion for increasing the minimum local region by a predetermined factor for each type of abnormal shadow candidate. An extension width that increases the local region outward by a predetermined length, and when the type of abnormal shadow candidate is a tumor shadow candidate, an extension multiple of 1.2 to 3.0 times or a range of 0.5 to 3.0 cm Is preferable, and when the type of abnormal shadow candidate is a microcalcification shadow candidate, the expansion multiple in the range of 1.0 to 1.2 times or 0.
It is desirable that the expansion width be in the range of 0 to 0.5 cm.
Further, when the type of the abnormal shadow candidate is a tumor shadow candidate, it is more preferable that the abnormal multiple is an extended multiple in the range of 1.5 to 2.0 times. Note that the description of the reference of the magnification and the length, and “the area obtained by expanding the minimum local area including the abnormal shadow candidate according to the predetermined expansion condition” is the same as that described above.

The same applies to the image output means, the predetermined image processing, and the minimum local area including the abnormal shadow candidate.

The above-mentioned abnormal shadow candidate output system according to the present invention may further comprise an extension condition setting means for setting a predetermined extension condition, and the local image creating means may be adapted to the predetermined extension condition set by the extension condition setting means. The local region may be extracted based on the local region.

The extended condition setting means may be, for example, a means for setting an extended condition for each type of abnormal shadow, a means for individually setting an extended condition for each detected abnormal shadow candidate, and a method for setting all extended abnormal conditions. The same extension condition may be set for the shadow candidate. Further, an extension condition may be set for each image reading doctor or diagnosis purpose in association with information for specifying the image reading doctor or information for specifying the diagnosis purpose.

[0024]

According to the abnormal shadow candidate output system of the present invention configured as described above, a region obtained by expanding the minimum local region including the abnormal shadow candidate is extracted as the local region including the detected abnormal shadow candidate. Since the image data of the extracted local area is subjected to predetermined image processing and output, when a shadow important for diagnosis exists outside the minimum local area including the detected abnormal shadow candidate, The above-mentioned predetermined image processing is applied to shadows important for the diagnosis, and an image suitable for detailed observation can be output.Also, a minimum local area including a detected abnormal shadow candidate It is possible to output an image suitable for observation and interpretation even when there is no shadow important for diagnosis outside the image.

The expansion condition for expanding the minimum local region including the abnormal shadow candidate is set to an expansion multiple in the range of 1.2 to 3.0 times or an expansion width in the range of 0.5 to 3.0 cm. Then, since spiculas that are important when diagnosing a tumor are also extracted as a local region together with the core, image enhancement processing is also performed on the spicula, and an image suitable for observing the spicula in detail is output be able to.

Further, according to the second abnormal shadow candidate output system of the present invention configured as described above, the type of the abnormal shadow candidate is determined, and each type is determined based on the determined type of the abnormal shadow candidate. Since an area obtained by expanding the minimum local area including the abnormal shadow candidate under different expansion conditions is used as a local area, an image expanded under appropriate expansion conditions can be output based on each type of characteristic.

When an extended condition setting means for setting a predetermined extended condition is further provided, the user can freely set the extended condition. It becomes possible to output.

[0028]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of an abnormal shadow candidate output system according to the present invention will be described below with reference to the drawings. FIG.
FIG. 1 is a diagram showing a specific embodiment of an abnormal shadow candidate output system according to the present invention.

The abnormal shadow candidate output system according to the present embodiment includes an abnormal shadow candidate detecting means 10 for detecting an abnormal shadow candidate in an image based on the whole image data (radiation image data) P1 of the subject input from an image reading device or the like. , The position information D1 indicating the existence range of the detected abnormal shadow candidate (the minimum local area including the abnormal shadow candidate) is input, and based on the input position information D1, the existence range is expanded according to a predetermined expansion condition. A local image creating means for extracting a region as a local region and performing a predetermined image enhancement process on the image data of the extracted local region to create local image data P2
And an image output means 30 such as a CRT for inputting the local image data P2 created by the local image creation means 20 together with the whole image data P1 and displaying the local image P2 'and the whole image P1'. Is done.

Next, the operation of the abnormal shadow candidate output system according to this embodiment configured as described above will be described.

When the abnormal shadow candidate detecting means 10 detects a tumor shadow candidate based on the whole image data P1 representing a radiographic image of the breast, the position information D1 of the existence range of the detected tumor shadow candidate is locally detected. It is input to the image creation means 20. Position information D of the existence range of this tumor shadow candidate
1 is detected by a detection process such as an iris filter process, and the coordinates ((x1, y1), (x1, y) of four vertices of a circumscribed rectangle surrounding a tumor shadow candidate existing in the image.
2), (x2, y1), (x2, y2)) (see FIG. 2). These four vertices ((x1, y1), (x1,
A region surrounded by (y2), (x2, y1), (x2, y2)) is referred to as a tumor shadow candidate existence range.

The local image creating means 20 extracts, as a local area, a region in which the existence range is expanded by 2.0 times in the vertical and horizontal directions based on the input position information D1 of the existence range. That is, as shown in FIG. 3, the existence range is expanded by 50% in each of the vertical and horizontal directions (four directions), and the expanded four vertices ((X1, Y1), (X1, Y
2) Extract a region surrounded by (X2, Y1), (X2, Y2)) as a local region. Here, (X1, Y1) = (x1- (x2-x1) / 2, y1-
(Y2-y1) / 2) (X1, Y2) = (x1- (x2-x1) / 2, y2 +
(Y2-y1) / 2) (X2, Y1) = (x2 + (x2-x1) / 2, y1-
(Y2-y1) / 2) (X2, Y2) = (x2 + (x2-x1) / 2, y2 +
(Y2-y1) / 2).

Next, the local image creating means 20 performs image enhancement processing such as gradation processing, frequency processing and enlargement processing on the image data of the area extracted as the local area, and performs local image data P2 Create

Specifically, these image enhancement processes are set so as to perform the following signal processes (1) to (3). That is, (1) the gradation processing is set so that the contrast of the local image P2 'displayed on the image output means 30 is 1.2 times or more that of the whole image P1'. (2) The frequency processing is set such that the local image P2 'displayed on the image output means 30 has an emphasis 1.1 times or more higher than that of the whole image P1'. (3) The enlargement process is set so that the size of the local image P2 'displayed on the image output means 30 is 1.5 times or more the size of the entire image P1'.

In the following description, an example in which enlargement processing is employed as the image enhancement processing will be described.

The image output means 30 receives the local image data P2 from the local image creating means 20, and the whole image data P1 of the subject from an image reading device or the like, and based on the local image data P2 and the whole image data P1. Then, the local image P2 'and the whole image P1' are displayed. Here, as shown in FIG. 4, a rectangular frame indicating the extracted local region is displayed at the position where the tumor shadow candidate exists in the whole image P1 ′, and the core image P2 is displayed in the local image P2 ′. ′
a and the specular image P2'b are both enlarged and displayed.

According to the abnormal shadow candidate output system according to the present embodiment, together with the core of the detected tumor shadow candidate,
Spicula, which is an important finding in diagnosis, is also displayed in an enlarged manner, which leads to improvement in diagnostic performance.

Next, a second specific embodiment of the present invention will be described. FIG. 5 is a diagram illustrating a configuration of the abnormal shadow candidate output system according to the present embodiment. The description of the same elements as those in the first embodiment will be omitted unless particularly necessary.

The abnormal shadow candidate output system according to the present embodiment includes an abnormal shadow candidate detecting means 10 for detecting an abnormal shadow candidate in an image based on the whole image data (radiation image data) P1 of the subject input from an image reading device or the like. From the type information D2 representing the type of the detected abnormal shadow candidate, and the position information D1 indicating the existence range of the detected abnormal shadow candidate from the abnormal shadow candidate detection means 10, The type information D2 representing the type of the detected abnormal shadow candidate is input from the determination means 40, and an area in which the existence range based on the position information D1 is extended is extracted as a local area according to a predetermined extension condition based on the type information D2. ,
Local image creating means 20 for creating local image data P2 by performing predetermined image enhancement processing on the image data of the extracted local area, and local image data P2 created by local image creating means 20 as whole image data. An image output means 30 such as a CRT for displaying the local image P2 'and the whole image P1', which is input together with P1, is provided.

Next, the operation of the abnormal shadow candidate output system of the present embodiment configured as described above will be described.

When the abnormal shadow candidate detecting means 10 detects an abnormal shadow candidate by predetermined detection processing such as iris filter processing or morphology filter processing, type information D2 representing the type of the detected abnormal shadow candidate is determined.
Entered in 40. In addition, since the type of the detected abnormal shadow candidate can be classified according to the type of the detection processing,
For example, when the detection process is an iris filter process, information indicating a tumor shadow candidate is input to the determination unit 40 as type information D2 when the detection process is a morphology filter process.

The local image creating means 20 inputs the position information D1 of the existence range of the abnormal shadow candidate from the abnormal shadow candidate detecting means 10 and the type information D2 from the determining means 40. Here, when the type of the abnormal shadow candidate based on the type information D2 is a tumor shadow candidate, the local image creating unit 20
Extracts a region in which the existence range is expanded by 2.0 times in the vertical direction and the horizontal direction, respectively, as a local region based on the input position information D1 of the existence range, as in the first embodiment. On the other hand, when the type of the abnormal shadow candidate is the microcalcification shadow candidate, the local image creating unit 20 determines that the area (ie, the existence (The same region as the range) is extracted as a local region.

The other operations are the same as in the first embodiment.

The manner in which the type of the abnormal shadow candidate is determined by the determining means 40 is not limited to the mode in which the type information D2 is input from the abnormal shadow candidate detecting means 10 as described above. For example, an input device connected to the outside may be used. Type information D manually from
Various modes can be adopted, such as a mode of inputting No. 2 and a mode in which the local image creating means 20 automatically detects the type based on the image data of the detected abnormal shadow candidate.

According to the second embodiment, when a tumor shadow candidate in which a shadow important for diagnosis is present in the periphery is detected as an abnormal shadow candidate, a spicula existing around the core is detected. Image enhancement processing is performed, and when a microcalcification shadow in which no significant shadow exists is detected, the image enhancement processing is performed only for the existence range of the microcalcification shadow candidate. It is possible to change the range of emphasis processing according to the type of shadow candidate,
Diagnostic performance and processing efficiency are improved.

Further, in each of the above embodiments, an extension condition setting means for setting a predetermined extension condition is further provided, and the local image creating means 20 is adapted to set the predetermined extension condition (for example, 1 A region in which the existence range of the abnormal shadow candidate is extended may be extracted as a local region based on the .5 expansion.

The extension condition setting means may be set each time the local image P2 'is displayed on the image output means 30, or may be set for each type of abnormal shadow candidate, for each doctor in charge of diagnosis,
It may be one that can be set in association with information representing various elements, such as for each inspection purpose. That is, for example, the extension condition is set to 1.
It is set to 0 times, and the extension condition is 1.5 times for the doctor B (the person in charge of the secondary examination). Alternatively, the extension condition is changed to 1.8 times for the tumor shadow candidate and 1.times. For the microcalcification shadow. For example, it may be changed twice so that it can be set as appropriate depending on the purpose of the examination, the preference of the doctor, or the characteristics of the abnormal shadow candidate in the image (eg, difference in size). In addition,
The above-mentioned primary examination means a periodic examination and the like, and the secondary examination means a detailed examination and the like. For example, in the primary examination, it is only necessary to confirm whether an abnormal shadow candidate exists in an image. Since it is only necessary to confirm the existence range of the detected abnormal shadow candidate, the secondary examination performs a detailed diagnosis on the existing abnormal shadow candidate. It is desirable to extract as a local region.

Further, the predetermined extension condition may be automatically determined according to the size of the range of the abnormal shadow candidate. In this case, the value of the extension condition that is automatically set can be set by the extension condition setting means. For example, the size of the existence range is determined by the measured values Acm to Bcm.
Various settings can be made, such as 1.5 times in the case of, and 2.0 times in the case of Ccm to Dcm.

The method of displaying the local image P2 'is as follows.
The local image P2 'created as described above is replaced with the entire image P
In addition to the form of displaying the whole image P1 'together with the rectangular frame indicating the detected abnormal shadow candidate, the local image is displayed by clicking the position of the rectangular frame with a position indicating means such as a mouse. P2 'may be displayed. This mode is particularly effective when two or more detected abnormal shadow candidates exist. in this case,
The extension condition can be changed for each abnormal shadow candidate.

The image output means 30 is not limited to one constituted by a single CRT or the like, but may be of various forms such as an apparatus which individually displays the whole image P1 'and the local image P2' on two CRTs or the like. Is possible.

In each of the above embodiments, the rectangular frame indicating the extracted local area is displayed at the position of the abnormal shadow candidate. However, the frame indicating the position of the abnormal shadow is changed to this form. The present invention is not limited to this, and may be a rectangular frame (circumscribed rectangular frame) indicating only the existence range of the abnormal shadow candidate, or a frame having a shape other than a rectangle. Further, the present invention is not limited to the frame, and may be, for example, an arrow indicating the center of the detected abnormal shadow candidate. In addition, frames and arrows indicating these positions
The user may be able to set it freely.

The abnormal shadow candidate detected by the abnormal shadow candidate detecting means 10 may be a candidate area automatically detected by iris filter processing or morphology filter processing, but is not limited to this. Such a candidate area may be further detected as having a high possibility of being malignant based on characteristics such as the shape of the candidate area.

The present invention is not limited to breast CAD, but can be applied to chest CAD and the like.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of an abnormal shadow candidate output system according to an embodiment of the present invention;

FIG. 2 is a diagram showing a range of existence of a tumor shadow candidate;

FIG. 3 is a diagram showing a local region.

FIG. 4 shows a whole image P1 ′ and a local image P as image output means.
Diagram showing the appearance of 2 '

FIG. 5 is a configuration diagram of an abnormal shadow candidate output system according to another embodiment of the present invention;

[Explanation of symbols]

 10 Abnormal shadow candidate detection means 20 Local image creation means 30 Image output means 40 Judgment means

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 4C093 AA26 CA50 DA06 FD05 FF07 FF08 FF09 FF13 FF17 FF32 FG05 FG13 5B057 AA08 BA03 CD05 CE03 CE06 CE09 CE11 DA08 DA16 DC04 DC08 5L096 AA06 BA06 BA13 CA07 EA45 JA11 MA

Claims (6)

[Claims] 1. A local region including an abnormal shadow candidate detected based on radiation image data representing a radiation image of a subject is extracted from the radiation image, and a predetermined region is extracted from the extracted image data of the local region. Local image creating means for creating local image data by performing image processing; image output for outputting a local image and the radiation image based on the local image data and the radiation image data created by the local image creating means Means for outputting an abnormal shadow candidate, wherein the local image creating means extracts, as the local area, an area obtained by increasing a minimum local area including the abnormal shadow candidate according to a predetermined expansion condition. An abnormal shadow candidate output system. 2. The method according to claim 1, wherein the predetermined expansion condition is an expansion multiple for increasing the minimum local area by a predetermined factor, and
2. An extended multiple in a range of 0 times.
The described abnormal shadow candidate output system. 3. The method according to claim 1, wherein the predetermined expansion condition is an expansion width for increasing the minimum local region outward by a predetermined length.
2. The abnormal shadow candidate output system according to claim 1, wherein the extension width is in a range of 3.0 cm. 4. A local region including an abnormal shadow candidate detected based on radiation image data representing a radiation image of a subject is extracted from the radiation image, and a predetermined region is extracted from the extracted image data of the local region. A local image creating means for creating local image data by performing image enhancement processing; and an image for outputting the local image and the radiation image based on the local image data and the radiation image data created by the local image creating means An abnormal shadow candidate output system comprising: output means; and an identification means for determining a type of the abnormal shadow candidate, wherein the local image creating means determines a type of the abnormal shadow candidate based on the type of the abnormal shadow candidate determined by the determination means. According to a predetermined expansion condition predetermined for each type of the abnormal shadow candidate, Abnormal shadow candidate output system, characterized in that extracts a region that extends frequency as the local region. 5. The predetermined expansion condition when the type of the abnormal shadow candidate is a tumor shadow candidate is an expansion multiple for increasing the minimum local region by a predetermined factor, and is in a range of 1.2 to 3.0 times. When the type of the abnormal shadow candidate is a microcalcification shadow candidate, it is an expansion width that enlarges the minimum local region outward by a predetermined length, or an expansion width in a range of 0.5 to 3.0 cm. The predetermined expansion condition is an expansion multiple for increasing the minimum local region by a predetermined factor, and is 1.0 to 1.0.
An expansion multiple of 1.2 times or an expansion width for increasing the minimum local region outward by a predetermined length, and 0.0 to 0.5c
5. The abnormal shadow candidate output system according to claim 4, wherein the extended width is in the range of m. 6. The image processing apparatus according to claim 1, further comprising: an extension condition setting unit configured to set the predetermined extension condition, wherein the local image creating unit sets the local area based on the predetermined extension condition set by the extension condition setting unit. The abnormal shadow candidate output system according to any one of claims 1 to 5, wherein the system is to extract.