JP2010025758A - Image processing apparatus and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To identify a specific tissue from another tissue without applying special staining. <P>SOLUTION: For example, an elastic fiber consists of an aggregate of fine fibers and is formed into a strip like or bundle-like shape. A cause is obscure, but, by this morphorogical feature, a phenomenon such that intensity of light is conversely increased under a certain condition without being attenuated in the elastic fiber or the region adjacent to the elastic fiber is experimentally confirmed. Then, an illumination light image wherein only illumination light is photographed in a pervious observation system and an H&E stained specimen image are acquired on the basis of the optical feature possessed by the elastic fiber and a region where the intensity of light is increased as compared with the illumination light image among the respective image positions of the acquired H&E stained specimen image is extracted or displayed to identify the specific tissue from another tissue without applying special staining. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an image processing apparatus and an image processing method for a pathological specimen image of a living body stained with a pigment, for example.

  In biological tissue specimens, especially pathological specimens, block specimens obtained by organ excision and specimens obtained by needle biopsy are sliced into several μm, and are then widely observed with a microscope to obtain various findings. It has been broken. In particular, transmission observation using an optical microscope is one of the most popular observation methods because the equipment is relatively inexpensive and easy to handle and has been used historically. . In this case, the sliced biological specimen hardly absorbs and scatters light and is almost colorless and transparent. Therefore, it is general that the specimen is stained with a dye prior to observation.

  Various dyeing methods have been proposed, and the total number thereof reaches 100 or more. Especially, regarding pathological specimens, hematoxylin-eosin staining using two of blue-purple hematoxylin and red eosin as pigments is proposed. (Hereinafter referred to as “H & E staining”) is used as standard.

  The main diagnostic item is whether or not the H & E-stained pathological specimen is cancer, and in the case of cancer, what level of malignancy is. Here, it is necessary to confirm the grade and depth of cancer and the presence or absence of invasion into blood vessels and lymphatic vessels. In particular, the presence or absence of invasion into blood vessels or lymph vessels is an index for diagnosing the presence or absence of metastasis to other organs, and greatly affects the selection of a treatment method.

  However, blood vessels and lymphatic vessels can be identified morphologically in a normal tissue state, but when the cancer progresses, the tissue shape collapses, making it difficult to visually recognize it. For this reason, a technique for visually enhancing the color by applying special staining to a tissue (elastic fiber) constituting a blood vessel wall that is difficult to visually recognize by H & E staining is clinically performed.

JP 2006-153742 A

  However, in the case of a technique for performing special staining, there is a problem that costs and work processes at a clinical site are increased.

  For this reason, an attempt to extract a specific tissue by image processing without actually performing special staining has been proposed (see, for example, Patent Document 1). That is, the feature amount of each tissue is determined based on color information (hue, saturation, brightness) and average luminance of tissues (cell nucleus, cytoplasm, stroma, pores) in the H & E stained specimen, and classification used for cell classification I try to create a table.

  That is, in Patent Document 1, color information and an average luminance value are used as feature amounts. However, in the case of extracting an H & E stained specimen, such as an elastic fiber, in which the difference in color and brightness is difficult to understand compared to the surrounding tissue, it is difficult to extract, and eventually, special staining must be performed. There are many cases.

  The present invention has been made in view of the above, and an object thereof is to provide an image processing apparatus and an image processing method capable of distinguishing a specific tissue from other tissues without performing special staining.

  In order to solve the above-described problems and achieve the object, an image processing apparatus according to the present invention includes an image acquisition unit that acquires an observation target image obtained by photographing an observation target in a transmission observation system using illumination light, and Extraction means for extracting, as a region to be extracted, a region in which the light intensity is increased at a position in the acquired observation target image as compared with the case where the illumination light is photographed.

  The image processing apparatus according to the present invention includes an image acquisition unit that acquires an observation target image obtained by photographing an observation target with a transmission observation system using illumination light, and the illumination at a position in the acquired observation target image. Display means for displaying, as an extraction target area, an area in which the light intensity is higher than when light is photographed.

  The image processing method according to the present invention includes an image acquisition step of acquiring an observation target image obtained by photographing an observation target with a transmission observation system using illumination light, and the illumination at a position in the acquired observation target image. An extraction step of extracting, as an extraction target region, a region in which the light intensity is increased as compared with a case where light is photographed.

  The image processing method according to the present invention includes an image acquisition step of acquiring an observation target image obtained by photographing an observation target with a transmission observation system using illumination light, and the illumination at a position in the acquired observation target image. A display step of displaying, as an extraction target area, an area where the light intensity is increased as compared with a case where light is photographed.

  According to the image processing apparatus and the image processing method of the present invention, there is an effect that the observation target can be distinguished from other tissues by image processing by using the optical characteristics of the observation target. .

  Hereinafter, an image processing apparatus and an image processing method which are the best mode for carrying out the present invention will be described with reference to the drawings. The present invention is not limited to the embodiment, and various modifications can be made without departing from the spirit of the present invention.

  First, referring to FIGS. 1-1 and 1-2, an outline of the present embodiment will be described in comparison with the conventional method. FIG. 1A is an explanatory diagram of an image example of a conventional special staining method. The H & E-stained image shown in FIG. 1-1 shows an example of an image including elastic fibers that have undergone a shape change with the progression of cancer. Since it is difficult to visually recognize the elastic fiber in the H & E dyed image, conventionally, as shown in the special dyed image in FIG. 1-1, the elastic fiber can be visually recognized even if the structure is unclear. It is what is visualized. A jagged black part around the whitish area in the center is visualized as an elastic fiber.

  On the other hand, in the present embodiment, the morphological characteristics of the elastic fiber are used. In general, when light passes through a translucent material, it attenuates based on the eigenvalue of the material and its thickness (Lambert-Beer law). That is, the light transmitted through the stained specimen is attenuated by the dye contained in the stained specimen. Therefore, the phenomenon in which the light intensity increases in the stained specimen image is considered to be noise due to photographing conditions or the like. However, in an elastic fiber or a region adjacent to the elastic fiber, a specific phenomenon (optical characteristic) in which the light intensity increases conversely without being attenuated under certain conditions was experimentally confirmed. The cause of such a unique phenomenon is not necessarily clear, but the morphological characteristics of the observation object are considered as factors. For example, a specific tissue (for example, elastic fiber) that is an observation object is composed of a collection of fine fibers, and is assumed to be in a band shape or a bundle shape. Such a morphological feature is considered to cause the unique phenomenon described above.

  Therefore, based on the morphological characteristics or optical characteristics of the observation target (a specific tissue such as an elastic fiber in this embodiment), as shown in FIG. Acquire a photographed image (image of a slide glass on which an H & E-stained stained specimen is placed) and an H & E-stained image, and compare the illumination light image and the H & E-stained image to obtain the light intensity ratio at each image position. In general, a region where the light intensity, which is considered to be noise, is increased is extracted or displayed as a light intensity increased region image. The image example on the right side in FIG. 1-2 is an image example showing the light intensity increasing region extracted by taking the difference, and the whitish part is a narrow region adjacent to the elastic fiber or the elastic fiber having the increased light intensity. Is shown. In this way, by utilizing the morphological characteristics and optical characteristics of the elastic fibers, the elastic fibers can be clearly distinguished from other tissues without special staining.

  FIG. 2 is a schematic block diagram illustrating a configuration example of an image processing apparatus that performs the image processing method of the present embodiment. The image processing apparatus according to the present embodiment is an image processing apparatus mounted on an optical microscope whose observation target is a pathologically stained specimen that has been subjected to H & E staining. The image processing unit 100, the operation unit 109, the memory 110, the control unit 111, and the like. Is provided.

  The image processing unit 100 includes an image capturing unit 101, a target region extracting unit 102, a light intensity ratio calculating unit 103, a focal position determining unit 104, a light intensity increasing region creating unit 105, a light intensity ratio threshold processing unit 106, and a superimposed image creating unit. 107 and a display unit 108. Here, an outline function of each unit will be described. The image capturing unit 101 captures an image of a stained specimen that has been subjected to H & E staining and an image of a glass-only portion (hereinafter simply referred to as glass) in which no specimen tissue exists (a glass image obtained by photographing only illumination light). In order to acquire a stained specimen image and a glass image. The target area extraction unit 102 is for extracting a processing target area from the glass image and the stained specimen image. The light intensity ratio calculation unit 103 is for calculating the light intensity ratio at each image position corresponding to the glass image and the stained specimen image. The focal position determination unit 104 is for determining an optimal focal position from a plurality of changed focal positions. The light intensity increasing region image creating unit 105 is for selecting a region where the light intensity is increasing at the determined optimum focal position and performing an image processing. The light intensity ratio threshold processing unit 106 performs threshold processing on the calculated light intensity and extracts an optimum light intensity increasing region. The superimposed image creating unit 107 creates a superimposed image by associating the position determined as the light intensity increasing region with the position of the stained specimen image. The display unit 106 is for displaying an image and various other information, and includes a flat panel display such as an LCD or an ELD or a CRT display.

  The memory 109 temporarily or permanently stores various data such as glass images, stained specimen images, and processed data. The control unit 111 is for overall control of the whole. The operation unit 109 is for performing an instruction process related to various processes.

  Hereinafter, functions and the like of each unit will be described in detail. First, the image capturing unit 101 captures a multiband image by a frame sequential method while switching by switching 16 band pass filters with a filter wheel. Such an imaging method is disclosed and widely known, for example, in Japanese Patent Laid-Open No. 7-120324. As a result, a multiband image having 16 band pixel values for each point of the sample is obtained. The pixel value of each band corresponds to the light intensity in the wavelength band λ.

  Here, although the dye is originally distributed three-dimensionally within the stained specimen to be observed, it cannot be regarded as a three-dimensional image as it is in a normal transmission observation system, and illumination that has passed through the stained specimen. It is observed as a two-dimensional image in which light is projected onto the image sensor of the camera. Therefore, the “each point of the specimen” here represents a point on the stained specimen corresponding to each pixel of the projected image sensor, and the point x on the stained specimen corresponds to the position x on the multiband image. Suppose you are.

  The image capturing unit 101 only needs to be able to acquire light intensity information at each image position of the glass image and the stained specimen image, and may be an image capturing system that can obtain a grayscale image or an RGB color image. Hereinafter, in the present embodiment, it is assumed that the stained specimen image indicates a 16-band multiband image.

  Next, a method for photographing a stained specimen and glass in the image capturing unit 101 will be described with reference to FIGS. In the present embodiment, the stained specimen is photographed at a plurality of focal positions. First, as shown in FIG. 3A, the focal point for the stained specimen of the optical system 121 is searched for the glass on which the stained specimen is placed, and the stained specimen is imaged at this position to obtain a stained specimen image. To do. At this time, a position where the tissue is focused on a tissue other than the elastic fiber in the stained specimen, for example, a cell nucleus or a red blood cell, is set as a focal point, and a glass-only portion is photographed at the same focal position to obtain a glass image. Glass images taken in advance under the same illumination conditions may be used. Further, a pixel value that can be taken by the glass image is set in advance from a glass image taken in advance under the same illumination condition, and the glass image having the pixel value may be held as a model and used in the subsequent processing. . Further, even when the illumination conditions are different, the glass image and the stained specimen image may be normalized and used.

  Next, as shown in FIG. 3-2, the stained specimen is imaged at a focal position close to the stained specimen by a distance α from the focal point, and a stained specimen image is acquired. Further, as shown in FIG. 3C, the stained specimen is imaged at a focal position close to the stained specimen by a distance β from the focal point, and a stained specimen image is acquired. Then, as shown in FIG. 3-4, the stained sample is imaged at a focal position away from the stained sample by a distance γ from the focal point, and a stained sample image is acquired. In this way, the focus position is changed a predetermined number N times (not including the focal point), and imaging is repeated. The captured image is stored in the memory 110. The predetermined number N is 1 or more, that is, the focus is shifted one or more times from the focal point. The predetermined number N may be set in advance, or may be arbitrarily changed between one and N times.

  The target area extraction unit 102 performs a process of removing an area that is not a processing target from the glass image and the stained specimen image, that is, a background area and a boundary portion with the background area. Here, the boundary portion with the background region corresponds to the boundary between the glass and the tissue, and there is a possibility that incremental interference in which the light intensity increases due to the influence of the incident angle of light and the thickness of the tissue may occur. Therefore, in the present embodiment, in order to distinguish from the elastic fiber or the like, the boundary portion with the background region is excluded from the target region even if the light intensity is increased. This process may be performed after extracting a light intensity increasing region described later.

Hereinafter, a background area and a background area extraction method will be described. First, a stained specimen image is converted into a gray scale image, and a background region is extracted by performing threshold processing on the pixel value at each image position. Since the background region corresponds to glass and can be distinguished as a high-intensity region compared to the surrounding tissue region, the background region can be roughly extracted by performing threshold processing on the pixel values of the grayscale image. it can. Here, the threshold Th _g can be automatically determined based on a histogram of pixel values. For example, when the pixel average of a grayscale image is Gmean and the standard deviation is Gstd, the threshold Th _g is
Th _g = Gmean + Gstd
Is calculated as Then, an image position having a pixel value larger than the threshold Th _g is extracted as a background area. Further, the background area may be specified based on the pixel value of the glass image. For example, a glass image may be converted into a grayscale image, and a pixel having a difference within a predetermined range with respect to a pixel value at each image position may be regarded as a background region.

In order to distinguish the background area thus extracted from other areas, an index is assigned. This index is referred to as a target area label. The target region label S ₀ meaning non-processing target is at the image position of the stained specimen image corresponding to the background region of the gray scale image, and the target region label S ₁ is at the image position of the stained specimen image corresponding to the other region. Is granted. Furthermore, it can be displayed on the display unit 108 as a binary image based on the target area labels S ₀ and S ₁ . At this time, the threshold value may be directly set by the operation unit 109 while observing the binary image displayed on the display unit 108.

Subsequently, an image position serving as a boundary with the extracted background area is obtained. This can be done by extracting edges from the binary image described above, and using a generally known edge extraction filter (for example, Sobel filter) or the like. As in the background area, the target area label S ₀ is assigned to the image position of the stained specimen image corresponding to the boundary position with the background area in the binary image extracted in this manner, and is distinguished from the processing target. Information on the target area label S ₀ or S ₁ assigned to each image position is stored in the memory 110. In the following process, the processing target image position with the target area label S _1.

First, the light intensity ratio calculation unit 103, a predetermined number N of times taken by the imaging unit 101 from the image of the stained sample and glass image at the focus position of the (focus point not including) the image position with the target area label S ₁ The light intensity ratio is calculated. Here, the light intensity of the stained specimen image and the glass image at the focal position j (j = 1, 2, 3,..., N) at the image position P (x, y) is represented by I _{i, j} (x, y), When W _{i, j} (x, y), the light intensity ratio R _{i, j} (x, y) is
R _{i, j} (x, y) = I _{i, j} (x, y) / W _{i, j} (x, y)
Is calculated as Note that the subscript i (i = 1, 2, 3,..., 16) indicates a band number.

  Here, FIG. 4 and FIG. 5 show a light intensity graph and a light intensity ratio graph at an image position P (x, y) of a certain stained specimen image including elastic fibers. FIG. 4 is a graph showing the light intensity (brightness) of the glass image and the stained specimen image for each band, and FIG. 5 is a graph showing the result shown in FIG. 4 converted into a ratio. In the present embodiment, a region where the light intensity ratio is 1.0 or more, that is, a region where the light intensity is increased as compared with the glass image in the stained specimen image is extracted. The region where the light intensity increases corresponds to an elastic fiber or a narrow region adjacent to the elastic fiber.

  6 and 7 extract and expand the light intensity graph and the light intensity ratio graph in 13 to 16 bands in which the light intensity ratio is 1.0 or more in FIGS. 4 and 5. These 13 to 16 bands correspond to image examples taken with illumination light in the wavelength band of 630 nm to 780 nm, and it has been empirically confirmed that the light intensity ratio is often 1.0 or more. . Therefore, a region where the light intensity increases may be extracted by calculating the light intensity ratio by narrowing down to a plurality of specific wavelength bands in which the light intensity increase phenomenon is likely to occur.

Here, the target area label of the image position whose light intensity ratio is smaller than 1.0 is changed from S ₁ to S ₀ and excluded from the processing target area. In this case, the image position with the target area label S ₁ is, the incremental region of the light intensity. The light intensity ratio for each band at each image position and the target area label information are stored in the memory 110.

Next, the focal position determination unit 104 determines an optimum focal position from the light intensity ratio of each image position for each focal position. First, the average E _{i, j} of the light intensity ratio at the focal position j (j = 1, 2, 3,..., N) is obtained for the image position having the target region label S ₁ in the stained specimen image. When the number of image positions having the target area label S ₁ is Count _j and the total light intensity ratio of the image positions P (x, y) having the target area label S ₁ is ΣR _{i, j} (x, y), The average intensity ratio E _{i, j} is
E _{i, j} = ΣR _{i, j} (x, y) / Count _j
Is calculated by Note that the subscript i (i = 1, 2, 3,..., 16) indicates a band number. Therefore, the focal position k that maximizes the average light intensity ratio E _{i, j} is determined as the optimum focal position. In this way, by acquiring target images at a plurality of focal positions and finally determining the optimum focal position, a more optimal light intensity increasing region can be extracted from an image in which the light intensity increase is most noticeable. Is possible.

  FIG. 8 is an explanatory diagram illustrating an example of the in-focus point, the focus position N times the predetermined number of times, and the optimum focus position k determined by the focus position determination unit 104. Information on the determined optimum focus position k is stored in the memory 110.

Subsequently, the light intensity increasing region creating unit 105 creates an image of the light intensity increasing region based on the target region labels S ₀ and S ₁ at the optimum focal position k determined by the focal position determining unit 104. FIG. 9 shows an example of a light intensity increasing region image U ₀ created by binarizing the light intensity increasing region and the other region (light intensity increasing region: white, other region: black). . The information of the light intensity increase area image U ₀ is stored in the memory 110, and is sent to the display unit 108 for display as needed.

Further, the light intensity threshold processing unit 106, extracts to light intensity of the image position with the target area labels S _1, by performing threshold processing on the basis of the light intensity of the glass image, a more optimal light intensity increasing region To do.

When the standard deviation of the light intensity of the entire image with respect to the light intensity W _{i, k} (x, y) of the glass image at the optimum focus position k and the image position P (x, y) is Wstd _i , the intensity threshold Th _b is
Th _b = Wstd _i / 2
To be determined.

The difference Δ = I _{i, k} between the light intensity I _{i, k} (x, y) of the stained specimen image and the light intensity W _{i, k} (x, y) of the glass image at the image position P (x, y). Only when (x, y) −W _{i, k} (x, y) is larger than the intensity threshold Th _b , the image position corresponding to the light intensity increasing region is set.

On the other hand, the target area label of the image position below the intensity threshold Th _b is changed from S ₁ to S ₀ . As will be described later, the intensity threshold Th _b may be set directly from the operation unit 109 while confirming the light intensity increase region image U ₀ created in advance on the display unit 108. Based on the updated target region labels S ₀ and S ₁ , the light intensity increasing region creating unit 105 creates a light intensity increasing region image U ₁ . Figure 10 is an explanatory view showing an image example of the light intensity increase area image U _1. Information of the light intensity increase area image U _1, while being stored in the memory 110, is subjected to the display are sent to the display unit 108 as needed.

When the target area label at each image position of the stained specimen image is determined in this way, the position information of the elastic fiber can be provided. That is, when the light intensity increasing region corresponds to an elastic fiber, the image position is directly applied. Further, if the narrow region adjacent to the elastic fibers applicable, can be considered a region surrounded by the target area labeled S ₁ and elastic fibers. In addition to the presentation by imaging as described later, for example, it is also possible to newly calculate the feature amount of the area.

Further, the superimposed image creating unit 107 creates a superimposed image from the stained specimen image at the focal point stored in the memory 110 and the light intensity increasing region image U _1, and sends information about the created superimposed image to the display unit 108. . That superimposes and displays the light intensity increase area image U ₁ on the image of the stained sample. FIG. 11 is an explanatory diagram illustrating a display example of a superimposed image. According to this, similarly to the case of the special dyeing method, it is possible to present an image in which the position of the elastic fiber is visible.

As described above, the display unit 108 displays the stained specimen image at each focal position (predetermined number of times N + 1 including the focused point) stored in the memory 110, the light intensity increased region images U ₀ and U _1, and the superimposition. Display an image. Furthermore, a light intensity graph and a light intensity ratio graph at each image position can be displayed. Here, when a plurality of images and graphs are stored in the memory 110, the user may be able to select an image or graph to be displayed through the operation unit 109.

The operation unit 109 provides the user with a function of designating the intensity threshold Th _b in the light intensity threshold processing unit 106. That is, the user can specify an arbitrary threshold value based on the light intensity graph using a mouse or the like by observing the light intensity increasing region image U ₀ displayed on the display unit 108. In addition, a function of directly specifying the boundary position of the background area is also provided to the target area extraction unit 102. Furthermore, all the displayable images can be displayed or a display image can be selected by operating the operation unit 109.

It is explanatory drawing which shows the example of an image of the conventional special dyeing | staining system. It is explanatory drawing which shows the outline | summary of the example of an image of the extraction process of the elastic fiber of this invention. It is a schematic block diagram which shows the structural example of the image processing apparatus of embodiment of this invention. It is explanatory drawing which shows the method to image | photograph the dyeing | staining sample and glass in a some focus position. It is explanatory drawing which shows the method to image | photograph the dyeing | staining sample and glass in a some focus position. It is explanatory drawing which shows the method to image | photograph the dyeing | staining sample and glass in a some focus position. It is explanatory drawing which shows the method to image | photograph the dyeing | staining sample and glass in a some focus position. It is a graph which shows each light intensity (brightness) of the glass image and stained specimen image for every band. It is a graph which converts and shows the result shown in FIG. 4 in the ratio. It is a graph which expands and shows each light intensity (brightness) of the glass image of 13-16 bands, and a dyed specimen image. 7 is a graph showing the results shown in FIG. It is explanatory drawing which shows the example of a focal point, the focus position of N times of predetermined times, and the determined optimal focus position k. And binary image is an explanatory view showing an image example of the light intensity increase area image U ₀ that was created. It is an explanatory view showing an image example of the light intensity increase area image U _1. It is explanatory drawing which shows the example of a display of a superimposed image.

Explanation of symbols

DESCRIPTION OF SYMBOLS 101 Image pick-up part 102 Target area extraction part 103 Light intensity ratio calculation part 104 Focus position determination part 105 Light intensity increase area | region image creation part 106 Light intensity ratio threshold value processing part 107 Superimposed image creation part 108 Display part

Claims (16)

Image acquisition means for acquiring an observation target image obtained by photographing an observation target in a transmission observation system using illumination light;
An extraction means for extracting a region where the light intensity is increased as compared with the case where the illumination light is captured at a position in the acquired observation target image;
An image processing apparatus comprising:   The extraction unit extracts, as the extraction target region, a region that indicates a pixel value corresponding to a value larger than a pixel value corresponding to a pixel value when the illumination light is captured at a position in the observation target image. The image processing apparatus according to claim 1. The image acquisition means acquires an illumination light image obtained by photographing the illumination light,
The extraction means compares the observation target image with the illumination light image, and extracts a region showing a value larger than a pixel value of the illumination light image at a position in the observation target image, thereby extracting the extraction The image processing apparatus according to claim 2, wherein a target area is extracted. The image acquisition means acquires an illumination light image obtained by photographing the illumination light,
The image processing apparatus according to claim 1, wherein the extraction unit extracts the extraction target region based on a light intensity ratio at each image position of the illumination light image and the observation target image.   The image processing apparatus according to claim 4, wherein the extraction unit extracts a region where the light intensity ratio is larger than a predetermined threshold as the extraction target region.   The said extraction means extracts the area | region which excluded the predetermined high-intensity area | region in the said observation object image, and the area | region of the boundary part of the said high-intensity area | region as said extraction object area | region. The image processing apparatus according to claim 5.   The image processing apparatus according to claim 1, wherein the extraction unit extracts a region other than the observation target in the observation target image as the extraction target region.   The image processing apparatus according to claim 1, wherein the observation target image is an image corresponding to an image photographed with illumination light having a wavelength of 630 nm to 780 nm.   The image processing apparatus according to claim 1, wherein the observation object is a pathological specimen stained with a pigment.   The image processing apparatus according to claim 1, wherein the extraction target region is an elastic fiber or a region adjacent to the elastic fiber.   The image processing apparatus according to claim 1, wherein the image acquisition unit acquires the observation target image by photographing the observation target at a plurality of focal positions. Image acquisition means for acquiring an observation target image obtained by photographing an observation target in a transmission observation system using illumination light;
Display means for displaying, as a region to be extracted, a region where the light intensity is increased at a position in the acquired observation target image as compared with the case where the illumination light is captured;
An image processing apparatus comprising:   The image processing apparatus according to claim 12, wherein the display unit superimposes and displays the extraction target region on the pathological specimen image. An image acquisition step of acquiring an observation target image obtained by photographing an observation target in a transmission observation system using illumination light;
An extraction step of extracting a region where the light intensity is increased as compared with the case where the illumination light is photographed at a position in the acquired observation target image;
An image processing method comprising:   The image processing method according to claim 14, wherein the image acquisition step acquires the observation target image by photographing the observation target at a plurality of focal positions. An image acquisition step of acquiring an observation target image obtained by photographing an observation target in a transmission observation system using illumination light;
A display step of displaying an area where the light intensity is increased as compared with a case where the illumination light is captured at a position in the acquired observation target image, as an extraction target area;
An image processing method comprising: