JP2001204721A - Method and device for displaying mamma image - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve an observing and radiogram-interpreting performance in a computer aided image-diagnosing device. SOLUTION: Based on a whole image data S showing the radiograph of a subject, a whole image display means 30 displays the whole radiograph. On the other hand, an abnormal shadow candidate detecting means 40 detects the candidate of the abnormal shadow of the radiograph from the whole image data S. When a deciding means 50 decides the detection of the abnormal shadow candidate, a local area extracting means 60 extracts local image data S2 expressing the image of the local area including this abnormal shadow candidate. Then, based on this base S2, a local image display means 90 displays the image P2 of the local area separately from the whole image P to concentrate the observing consciousness of an image radiogram-interpreting person.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for displaying a radiographic image of a breast used for medical diagnosis.

[0002]

2. Description of the Related Art A radiation image of a subject recorded on a stimulable phosphor sheet or a film is read to obtain image data.
After performing appropriate image processing on the image data, an image is reproduced on a display device or the like in various fields. In particular, in recent years, digital radiography technology called computed radiography by combination with a computer has been developed, and various digital radiography has been clinically applied.

[0003] This digital radiography has a fundamentally different feature from conventional analog radiography in that it can quantitatively analyze image data. Especially for medical diagnosis of the human body, computer-aided image diagnosis or CADM (Computer Aided Diagnosis of Medical) aiming to more positively utilize the features of digital radiography.
Image) has been proposed.

[0004] The computer-assisted image diagnosis or the like assists diagnosis through image interpretation at a medical site. That is, in the related art, a radiation image reproduced by a recording medium such as a film or a display device such as a CRT is visually observed and interpreted by a specialist, and an abnormal tumor shadow or a high density microcalcification shadow representing cancer or the like is displayed. (Hereinafter collectively referred to as abnormal shadows) at an early stage. However, there is a possibility that such abnormal shadows may be overlooked or a misjudgment may occur due to a subjective judgment due to a difference in image reading ability between the readers who observe and read the radiation image.

Therefore, in computer-aided image diagnosis, abnormal shadow candidates considered as abnormal shadows are detected based on image data, markings are displayed on the detected portions, and attention is given to the reader of the radiological image. Or, by quantitatively presenting the characteristic features of the abnormal shadow candidates detected as materials useful for the interpreter's objective judgment, it is possible to prevent the above-mentioned oversight or misunderstanding by the interpreter and improve the diagnostic performance. ("Detection of tumor shadow in DR image (iris filter)" IEICE Transactions on Information and Communication D-II Vol.J75-D-II No.3 P663-670 March 1992
Moon, `` Extraction of microcalcification image by morphological filter using multiple structural elements '', D-II Vol.J75-D-II No.
7 P1170-1176 July 1992).

[0006]

By the way, in the proposal of the computer-aided image diagnosis, as described above, a marking is displayed on the abnormal shadow candidate of the entire reproduced image, or a quantitative evaluation scale for the abnormal shadow candidate is displayed. Has stopped to display.

[0007] However, displaying only such markings or rating scales is effective for alerting the reader or preventing the reader's subjective misunderstanding.
In order for a radiogram interpreter to actually diagnose the abnormal shadow site, it is necessary to provide an image having high diagnostic performance suitable for radiogram interpretation.

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and has as its object to provide a computer-aided image diagnostic apparatus capable of providing images with improved observation and interpretation performance for diagnosis. .

[0009]

According to the present invention, there is provided a computer-aided image diagnostic apparatus comprising: a whole image storage unit for storing whole image data representing a radiation image of a subject; Abnormal shadow candidate detecting means for detecting a shadow candidate; determining means for determining whether or not the abnormal shadow candidate is present based on a result of detection of the abnormal shadow candidate by the abnormal shadow candidate detecting means; When the determination unit determines that the abnormal shadow candidate is present, a local extracting unit extracts local image data of a local region including the abnormal shadow candidate from the whole image data stored in the whole image storage unit. Area extracting means, local image display means for displaying an image of the local area based on the local image data extracted by the local area extracting means, and the whole image It is characterized in that an entire image display means for displaying the entire radiation image of the object based on chromatography data.

[0010] Here, the abnormal shadow is a tumor, a tumor, a calcification, a thickened pleura or a cancer such as a pneumothorax which is not recognized in a normal shadow such as a blood vessel in a medical chest X-ray image or a mammogram. Refers to the shadows showing various symptoms. The abnormal shadow candidate detecting means does not need to detect all of these abnormal shadows.For example, it may detect only a tumor shadow or only a calcified shadow as an abnormal shadow. Of these, two or more abnormal shadows may be detected. For example, when the abnormal shadow candidate detection processing by the abnormal shadow candidate detection means is processing based on an algorithm of an iris filter (hereinafter, simply referred to as iris filter processing; described later), the abnormal shadow is a tumor shadow, If the detection process is a process based on a morphology (Morphology; also referred to as morphology) algorithm (hereinafter simply referred to as morphology process; described later), the abnormal shadow is a calcification shadow.

In addition, abnormal shadow candidates are not only abnormal abnormal shadows but also abnormalities because they show the same characteristics as tumors and calcifications from the viewpoint of image features such as tumors and calcifications. It is not clear whether the shadow is a shadow or not, and it also includes a shadow similar to an abnormal shadow that needs to be finally determined by a reader.

Further, the local region including a candidate for an abnormal shadow means a region in the vicinity of the candidate including a candidate for an abnormal shadow, and various shapes such as a rectangle, a circle, and an ellipse may be employed as a peripheral shape thereof. Can be. Therefore, by displaying not only the abnormal shadow candidate itself but also the image of the neighboring area on the local image display means, the entire image can be easily recalled from the image, and the abnormal shadow candidate in the entire image can be recalled. Can be easily positioned.

Further, a local image storing means for temporarily storing local image data is provided between the local area extracting means and the local image displaying means, and a local image is provided only when a predetermined image display request is received from the outside. It is also possible to adopt a configuration further including local image display request means for outputting local image data from the storage means and displaying the image of the local area on the local image display means. When an abnormal shadow is detected in this manner, the local image data indicating the image of the local area including the abnormal shadow is temporarily stored in the storage means, so that the radiologist can read the local image as needed. Display can be requested.

Further, among the images of the local area displayed on the local image display means, the abnormal shadow candidate image is more interpreted than the entire radiation image (hereinafter simply referred to as the entire image) displayed on the entire image display means. In order to improve the performance, the image processing apparatus may further include a local image enhancement unit that performs an image enhancement process on at least the abnormal shadow image data indicating the abnormal shadow candidate among the local image data. By performing the image enhancement processing on the local image data, particularly the abnormal shadow image data therein, by such a local image enhancing unit, the image of the local region, in particular, the image of the candidate for the abnormal shadow is displayed on the entire image display unit. The image can be more suitable for image interpretation than the displayed whole image, and the diagnostic performance can be improved.

Here, the whole image displayed on the whole image display means is displayed on the whole image display means based on the whole image data without performing any image processing on the inputted whole image data. The whole image may be used, or in the case of employing a configuration further including whole image enhancing means for performing image enhancement processing such as gradation processing and frequency processing on the entire image data, The whole image may be displayed on the whole image display means after being processed.

That is, the local image emphasizing means is irrespective of whether image emphasis processing is performed on the entire image data or not.
As compared with the whole image finally displayed on the whole image display means, the image of the local region displayed on the local image display means, particularly, the image interpretation performance of the abnormal shadow candidate image is improved.
What is necessary is just to perform image enhancement processing on local image data, especially abnormal shadow image data.

As the image enhancement processing by the local image enhancement means, for example, one of gradation processing, frequency processing, and enlargement processing, or a combination thereof can be employed.

When the gradation processing is adopted as the image processing, the image of the local area displayed on the local image display means or the image of the abnormal shadow candidate in the local area is displayed on the entire image display means. It is desirable that the contrast be set to be 1.2 times or more that of the displayed whole image.

Similarly, when frequency processing is employed as image processing, an image of a local area displayed on the local image display means or an image of an abnormal shadow candidate in the local area is displayed on the entire image display means. It is preferable that the emphasis degree is set to be 1.1 times or more than the whole image to be processed, and further, at least an image of a candidate for an abnormal shadow among images of a local region displayed on the local image display means is set. It is desirable that the emphasis is set to be 1.1 times or more that of the whole image displayed on the whole image display means.

When the enlargement process is adopted as the image processing, the image of the local area displayed on the local image display means or the image of the candidate for the abnormal shadow among the images of the local area is displayed on the whole image display means. It is desirable that the size is set to be 1.5 times or more the size of the abnormal shadow candidate image in the displayed whole image.

This enlargement processing is not limited to the enlargement processing at a constant enlargement rate as described above, but the size of an abnormal shadow candidate detected by the abnormal shadow candidate detecting means, for example, the periphery of a tumor shadow or a calcified shadow. The processing may be set so that the enlargement ratio is changed according to the length or the total or average value of the number of pixels inside the calcified shadow by the morphological processing described later. In other words, when displaying the abnormal shadow on the local image display means, if the size of the detected abnormal shadow is small, the enlargement processing is performed at a larger magnification ratio, and if the size of the detected abnormal shadow is large, the comparison is performed. Enlargement processing may be performed at a very small enlargement ratio so that the apparent size of the abnormal shadow on the display surface of the display means is always substantially constant irrespective of the actual size of the detected abnormal shadow.

More specifically, the local image enhancing means calculates the size of the abnormal shadow candidate based on the position data of the abnormal shadow candidate obtained by the abnormal shadow candidate detecting means; An enlargement ratio setting table in which an enlargement ratio according to the size of the abnormal shadow candidate calculated by the abnormal shadow size calculation unit is set in advance so that the display size of the abnormal shadow candidate image on the display unit is substantially constant, Enlargement processing means for enlarging local image data or abnormal shadow image data according to the enlargement ratio set by the enlargement ratio setting table may be adopted.

As described above, regardless of the size of the detected abnormal shadow candidate, by changing the enlargement ratio so that the displayed image of the abnormal shadow candidate always has a substantially constant size, the detected abnormal shadow candidate is changed. Even if the shadow candidate is very small,
The abnormal shadow candidate can always be interpreted as an image having a substantially constant size, and diagnostic performance can be improved.

In addition, as described above, it is possible to further comprise a whole image enhancing means for performing a predetermined image enhancing process on the entire image data. By performing image enhancement processing such as gradation processing and frequency processing on the entire image data by such an overall image enhancement unit, the readability of the entire image displayed on the overall image display unit can be improved to some extent. .

Further, it is also possible to adopt a configuration in which the whole image display means also serves as the local image display means, and the local area image is displayed on a part of the display surface of the whole image display means. That is, while displaying an image (whole image) representing the entire radiographic image on the whole image display means, a local area image that should be displayed on a separate local image display means is
The whole image displayed on the whole image display means is superimposed on a part of the whole image (the display part of the whole image display means on which the local area image is superimposed includes a part of the whole image and the image of the local area. Is not displayed overlapping, only the image of the local area is displayed without displaying part of the entire image that is superimposed, and the rest of the entire image is displayed in the non-overlapping display part To be done). Specifically, a window area having an outline such as a rectangle, a circle, and an ellipse for displaying an image of the local area is provided on a part of the display surface of the entire image display means. The image of the area may be displayed, and the entire image in which only the area corresponding to the window area has been deleted may be displayed outside the window area.

As the display mode of the whole image and the image of the local area, various modes can be adopted depending on the subject to be imaged. That is, for example, when an X-ray image of the breast of one patient is displayed, an image of only one breast is displayed, and an abnormality of one breast is adopted by adopting a configuration including two sets of image display means. The image of the local area including the shadow is displayed on one image display means, and at the same time, the image of the local area of the part of the other breast corresponding to the local area of one breast is displayed on the other image display means. Thus, the radiographer can compare and observe the same part of the left and right breasts.

That is, the whole images of the left and right breasts are displayed as a pair on one whole image display means, or the whole images of the pair of breasts are separately displayed on the left and right on two whole image display means. You may.

When a candidate for an abnormal shadow is detected in one breast, an image of a local region including the candidate for an abnormal shadow in the one breast and the position of the local region in the one breast correspond to the position of the local region. The image of the local region of the other breast is displayed in pairs on one local image display means, or the image of the local region of the pair of breasts is separately
It may be displayed on one local image display means. In this case, it is desirable that the same local image enhancement processing be performed on the image of the local region of the pair of left and right breasts.

Further, the whole image display means also serves as the local image display means, and the images of the local areas of the left and right breasts respectively correspond to the whole images of the left and right breasts displayed in pairs on one or two whole image display means. You may make it display it.

The image of one breast and the image of the other breast photographed separately from the breast may be simultaneously displayed on the display surface of the same display means. Specifically, the image of one breast is displayed on the right half of the whole image display means, the image of the other breast is displayed on the left half of the same whole image display means, and these are displayed simultaneously facing each other or back to back. If an abnormal shadow candidate is detected from the entire image data of the breast, the image of the local area including the abnormal shadow candidate is emphasized, and a part of the entire breast is displayed on a part of the right half display surface. The image of the local region about the part of the other breast corresponding to the part of the abnormal shadow candidate in this one breast and displayed together with the image,
The whole image of the other breast may be displayed on a part of the left half display surface in a superimposed manner.

When two pieces of image data obtained separately in this way are displayed at the same time on different image display means or different display positions of the same image display means, corresponding parts are displayed. Means for storing the whole image data of one breast and the whole image data of the other breast, respectively;
A means for detecting two pieces of image data (whole image data and local image data) in correspondence with the positional relationship of the images, and a display control means for controlling these to display them on a display means. I just need.

The local area image displayed on the whole image display means is displayed in a display area corresponding to the local area image and different from the local area in the radiation image displayed on the whole image display means. By doing so, the local region image is not superimposed on the local region in the displayed whole image and displayed, and the local region image displayed simultaneously while recognizing the position of the local region in the whole image is observed. And diagnostic performance is improved.

Further, the display area of the local area image displayed on the whole image display means is determined so as to be included in a display area different from the subject displayed on the whole image display means, so that the local area image is determined. It is possible not only to prevent the region image from overlapping the local region in the entire image, but also to overlap the entire image of the subject in the entire image.

It is conceivable that the display area of the local area image determined as described above changes in size according to the size of the subject displayed on the entire image display means. In other words, if the area where no subject is displayed is very small, the display area must be very small.

In such a case, the size of the local area image itself may be reduced and displayed in accordance with the size of the display area. The effect of extracting and displaying only the image is reduced.

Thus, when the size of the local area image in the display area of the whole image display means is smaller than the size of the local area image, the local area image which can be displayed in the display area of the local area image is displayed. Is displayed, and the local area image can be scrolled in the display area, so that only a part of the local area image can be displayed in the display area. ), It is desirable to scroll the local area image within the window so that a necessary portion of the local area image can be observed.

The number of local area images displayed simultaneously with the whole image on the whole image display means is not limited to one. If a plurality of abnormal shadows are detected, a plurality of local shadows corresponding to the number are detected. Can be displayed.

Further, as the processing for detecting an abnormal shadow candidate by the above-described abnormal shadow candidate detecting means, specifically, an iris filter processing, a morphology processing, or a processing combining these can be used.

Here, the iris filter calculates the gradient of an image signal represented by the density value of the radiation image as a gradient vector and outputs the degree of concentration of the gradient vector. This is a method of detecting a candidate for a tumor shadow based on the degree of concentration of the gradient vector.

That is, in the radiation image, for example, X
It is known that, in a radiographic image (an image that outputs an image signal with a high density and a high signal level) on a line film, the density of the tumor shadow is slightly lower than that of the surrounding area,
The gradient vector at any pixel in the tumor shadow points near the center of the tumor shadow. On the other hand, in an elongated shadow such as a blood vessel shadow, the gradient vector does not concentrate on a specific point. Therefore, if the distribution of the direction of the gradient vector is locally evaluated and a region concentrated at a specific point is extracted,
It becomes a candidate for a mass shadow. This is the basic concept of iris filter processing. The specific algorithm steps are shown below.

(Step 1) Calculation of Gradient Vector For all the pixels constituting the target image, the direction θ of the gradient vector of the image data is calculated for each pixel j based on the calculation formula shown in the following equation (1). Ask.

[0042]

(Equation 1) Here, as shown in FIG. 18, f1 to f16 are pixel values (image data) corresponding to pixels on the outer periphery of the mask of 5 pixels vertically and 5 pixels horizontally around the pixel j.

(Step 2) Calculation of Concentration of Gradient Vector Next, for all the pixels constituting the target image,
For each pixel, the degree of concentration C of the gradient vector having that pixel as the pixel of interest is calculated according to the following equation (2).

[0044]

(Equation 2) Here, N is the number of pixels existing in a circle having a radius R around the pixel of interest, θj is a straight line connecting the pixel of interest and each pixel j in the circle, and the above equation (1) for each pixel j (See FIG. 19). Therefore, the degree of concentration C represented by the above equation (2) becomes a large value when the direction of the gradient vector of each pixel j is concentrated on the target pixel.

By the way, the gradient vector of each pixel j in the vicinity of the tumor shadow is determined regardless of the contrast of the tumor shadow.
Since the pixel of interest is directed substantially toward the center of the tumor shadow, the pixel of interest having a large value of the degree of concentration C can be said to be the pixel at the center of the tumor shadow. On the other hand, in the shadow of a linear pattern such as a blood vessel, the value of the degree of concentration C is small because the direction of the gradient vector is biased in a certain direction. Therefore, the value of the degree of concentration C with respect to the pixel of interest is calculated for each of all the pixels constituting the image, and whether or not the value of the degree of concentration C exceeds a predetermined threshold value is evaluated. Can be detected. In other words, this filter is less susceptible to blood vessels and mammary glands than a normal differential filter,
It has the feature that tumor shadows can be detected efficiently.

Further, in the actual processing, in order to achieve a detection power independent of the size and shape of the tumor, a device for adaptively changing the size and shape of the filter is used. FIG. 20 shows the filter. This filter differs from the filter shown in FIG.
/ M types of directions for each / M (in FIG. 20, 32 for each 11.25 degrees)
The evaluation of the degree of concentration is performed only with the pixels on the radial line (example of direction).

Here, the coordinates ([x], [y]) of the n-th pixel on the i-th line and from the pixel of interest are as follows if the coordinates of the pixel of interest are (k, l): Equation (3),
Given by (4).

[0048]

(Equation 3) Here, [x] and [y] are the maximum integers not exceeding x and y.

Further, the output value up to the pixel at which the maximum concentration is obtained for each line on the radial line is defined as the concentration in that direction, and the concentration is averaged in all directions (32 directions in the example). Then, the concentration degree C of the gradient vector group for the target pixel is set. Specifically, the concentration Ci (n) obtained from the pixel of interest to the n-th pixel on the i-th radial line is obtained by the following equation (5).

[0050]

(Equation 4) Here, Rmin and Rmax are the minimum and maximum values of the radius of the tumor shadow to be extracted.

Next, the concentration C of the gradient vector group is calculated by the following equations (6) and (7).

[0052]

(Equation 5) Here, since the expression (6) is the maximum value of the concentration Ci (n) for each radial line obtained by the expression (5), the pixel having the maximum value of the concentration Ci (n) from the image of interest. The region up to is a candidate region for a tumor shadow in the direction of the line.
By obtaining the area of all the radial lines by the equation (6), the shape of the outer peripheral edge of the area that can be a candidate for the tumor shadow can be specified.

In equation (7), a value is obtained by averaging the maximum value of the degree of concentration given by equation (6) in this area in all directions of the radial line. By comparing the obtained value with a preset threshold value T, it is determined whether or not an area around the target pixel is likely to be an abnormal shadow candidate.

The area for evaluating the degree of concentration C of the gradient vector group of the equation (7) is similar to the manner in which the iris (iris) of the human eye expands and contracts according to the brightness of the outside world. Since the size and the shape are adaptively changed according to the distribution of the vector, it is referred to as an iris filter.

The above calculation of the degree of concentration Ci (n) may be performed by using the following equation (5 ') instead of the equation (5).

[0056]

(Equation 6) (Step 3) Evaluation of shape of tumor shadow candidate In general, the shadow of a malignant tumor is 1) its edge is irregular 2) it has a shape close to a circle 3) its morphology has an uneven density distribution inside It has various features.

Therefore, normal tissue is removed from the detected abnormal shadow candidates, and a more definite “abnormal shadow candidate”
That is, in order to extract only shadow candidates having a very high probability of being a tumor, shape determination is performed in consideration of these features. The feature amount used here is a spread degree (Spread
ness), Elongation, Roughnes
s), circularity (Circularity) and internal roughness (En
tropy).

According to the above steps, the iris filter can effectively detect the tumor shadow from the radiographic image.

Next, the morphology processing will be described. The morphological processing is a method for detecting a candidate for a microcalcification image, which is a characteristic form in breast cancer, together with a tumor shadow, and a multi-scale λ and a structural element (mask) B
(1) that it is effective for extracting the calcified image itself, (2) that it is not easily affected by complicated background information, (3) that the extracted calcified image is not distorted,
There are such features. That is, this method can perform detection while maintaining geometric information such as the size, shape, and density distribution of a calcified image better than general differentiation processing. The outline is described below.

(Basic Operation of Morphology) Morphology is generally developed as a set theory in an N-dimensional space. However, for intuitive understanding, a description will be given of a two-dimensional gray image.

The point of coordinates (x, y) is represented by a density value f
It is regarded as a space having a height corresponding to (x, y). Here, the density value f (x, y) is a high-luminance high-signal level signal which becomes a larger image signal as the density is lower (the brightness is higher when displayed on a CRT).

First, for simplicity, consider a one-dimensional function f (x) corresponding to the cross section. The structural element g used in the morphological operation is a symmetric function symmetric about the origin, as shown in the following equation (8).

(Equation 7) And the value is 0 in the domain, and the domain is represented by the following equation (9).

[0063]

(Equation 8) At this time, the basic form of the morphological operation is as shown in Equation (10).
As shown in (13), the calculation becomes very simple.

[0064]

(Equation 9) That is, the dilation process is a process of searching for a maximum value within a range of ± m (a value determined according to the structural element B) around the target pixel (FIG. 21).
On the other hand, the erosion process is a process of searching for a minimum value within a width of ± m around the target pixel (see FIG. 21B). The opening (or closing) process corresponds to searching for the maximum value (or minimum value) after searching for the minimum value (or maximum value). That is, the opening process smoothes the density curve f (x) from the low brightness side, and removes the convex density variation portion (the portion having higher brightness than the surrounding portion) that fluctuates in a spatially narrower range than the mask size 2 m. This corresponds to removal (see FIG. 21C). On the other hand, the closing (closing) processing is performed from the high-luminance side to the density curve f (x)
Is equivalent to removing a concave density variation portion (a portion having a lower luminance than the surrounding portion) which fluctuates in a spatially narrower range than the mask size 2 m (see FIG. 21D).

When the structural element g is not symmetrical with respect to the origin, the dilation operation in Expression (10) is called Minkowski sum, and the erosion operation in Expression (11) is called Minkowski difference.

Here, in the case of a signal with a high density and a high signal level, the higher the density, the higher the density, the higher the density value f
Since the magnitude relationship is reversed with respect to the case where the image signal value of (x) is at the high luminance and high signal level, the dilation processing on the signal at the high density and high signal level is performed by the erosion processing at the high luminance and high signal level (see FIG. B)), the erosion processing for the signal with the high density and high signal level is performed by the dilation processing at the high luminance and high signal level (FIG. 21).
(A)), the opening process for a signal with a high density and high signal level matches the closing process for a signal with a high luminance and high signal level (FIG. 21D), , And the opening process at a high luminance and high signal level (FIG. 21C).

In this section, the case of an image signal (luminance value) at a high luminance and high signal level will be described.

(Application to Calculated Shading Detection) For the detection of calcified shading, a difference method of subtracting a smoothed image from an original image can be considered. Because it is difficult to distinguish calcified shadows and elongated non-calcified shadows (such as mammary glands, blood vessels, and mammary gland supporting tissues) using simple smoothing, Obata et al. We have proposed a morphological filter expressed by the following equation (14) based on the (opening) operation ("Extraction of microcalcification image by morphological filter using multiple structural elements" IEICE Transactions D- II Vol.J75-D-II No.7 P1170-1176 July 1992
Tsuki, "Basics of Morphology and Its Application to Mammogram Processing" MEDICAL IMAGING TECHNOLOGY Vol.12 No.1 Jan
uary 1994).

[0069]

(Equation 10) Here, Bi (i = 1, 2, 3, 4) is the four linear structural elements B shown in FIG. If the structural element B is set to be larger than the calcified shadow to be detected, the calcification is a convex signal change portion (an image portion that fluctuates in a narrow spatial range) finer than the structural element B in the opening process. The image is removed. On the other hand, the elongated non-calcified shadow has a longer length than the structural element B, and if its inclination (extending direction) matches any of the four structural elements Bi, the opening process (the equation (14)) Even if the two-term operation is performed, it remains as it is. Therefore, by subtracting the smoothed image (image from which the calcified shadow has been removed) obtained by the opening process from the original image f, an image including only small calcified image candidates is obtained. This is the idea of equation (14).

As described above, in the case of a signal with a high density and a high signal level, the calcified shadow has a lower density value than the surrounding image portion, and the calcified shadow has a concave signal change with respect to the surrounding portion. Therefore, the closing processing is applied instead of the opening processing, and the equation (14 ′) is applied instead of the equation (14).

[0071]

[Equation 11] However, even in this case, a part of the non-calcified shadow having the same size as the calcified shadow may remain. In such a case, the differential information based on the morphological operation of the following equation (15) is used. The non-calcified image included in P in equation (14) is further removed.

[0072]

(Equation 12) Here, the larger the value of Mgrad, the greater the possibility of calcification shadows, so the calcification candidate image Cs can be obtained by the following equation (16).

[0073]

(Equation 13) Note that T1 and T2 are preset threshold values that are determined experimentally.

However, for non-calcified shadows different from the size of the calcified shadow, P in equation (14) and a predetermined threshold T1
In the case where a non-calcified shadow having the same size as the calcified shadow does not remain because it can be removed only by comparing with the condition (P (i, j) .Gtoreq.T1).

Finally, the cluster Cc of the calcified shadow is detected by a combination of the multi-scale opening operation and the closing operation shown in the equation (17).

[0076]

[Equation 14] Here, λ1 and λ2 are determined by the maximum distance of the calcified shadow to be fused and the maximum radius of the isolated shadow to be removed, respectively, and λ3 = λ1 + λ2.

As described above, these morphological filters are described in the case of image data of high luminance and high signal level. However, image data of high density and high signal level (higher density pixels have larger digital data). In the case of (image data having a value), the opening operation and the closing operation have the opposite relationship.

[0078]

According to the computer-aided image diagnostic apparatus of the present invention, the whole image display means is provided with a radiographic image based on the whole image data directly inputted or the whole image data inputted through the whole image storage means. Display the whole of. On the other hand, this whole image data is also input to the abnormal shadow candidate detecting means directly or via the whole image storing means, and the abnormal shadow candidate detecting means is estimated to be an abnormal shadow such as a tumor shadow based on the whole image data. Detect abnormal shadow candidates.
When detecting an abnormal shadow candidate, the abnormal shadow candidate detecting unit specifies a pixel of the image data indicating the abnormal shadow candidate, and outputs position data indicating the specified pixel to the determining unit. On the other hand, when no abnormal shadow candidate is detected, such position data is not output.

When the position data is inputted, the judging means judges that an abnormal shadow candidate has been detected, and outputs the inputted position data to the local area extracting means. On the other hand, if it is determined that no abnormal shadow candidate has been detected, the processing ends because such position data is not output from the abnormal shadow candidate detection means.

On the other hand, the whole image data stored in the whole image storage means is also input to the local area extraction means, and the local area extraction means makes the abnormal shadow based on the input position data out of the whole image data. After specifying neighboring pixels including a pixel of image data indicating a candidate (a local region as a set of these pixels) according to a preset processing procedure, local image data representing an image of the local region is extracted. .

The extracted local image data is output to the local image display means, and the local image display means displays an image of the local area including the abnormal shadow based on the data.

As described above, since only the image of the local region including the candidate for the abnormal shadow is displayed on the local image display means separately from the entire image, the observer is displayed on the local image display means. It is possible to concentrate the observation consciousness on the image of the localized region, improve the diagnostic performance, and reduce the time required for the diagnosis.

A local image storing means for temporarily storing local image data is provided between the local area extracting means and the local image displaying means, and a local image storing means is provided only when a predetermined image display request is received from the outside. In a computer-aided image diagnostic apparatus having a configuration further including local image display requesting means for outputting local image data from the means and displaying an image of the local area on the local image display means, the local area extracting means The obtained local image data is input to the local image storage means and temporarily stored therein, and the processing is temporarily stopped here.

Here, when the observer of the image inputs a signal indicating a request to display the local image to the local image display requesting means, the signal indicating the request to display the local image is transmitted from the local image display requesting means to the local image storage. The local image storage means outputs the stored local image data to the local image display means in response to the input of the signal, and the processing is restarted.

According to the computer-aided image diagnostic apparatus having such a configuration, the local image data of the local area including the detected abnormal shadow is temporarily stored in the local image storage means. Whether or not to display an image can be arbitrarily determined by the observation interpreter so as to match the interpretation time and the like.

That is, the radiologist usually observes the whole image first, recognizes information of the whole image to some extent, and then observes the detailed structure. Therefore, the interpreter first observes the whole image displayed on the whole image display means, and inputs a display request to the local image display request means after the observation or at a desired time during the observation. The image of the local region including the candidate for the abnormal shadow of can be immediately displayed on the local image display means,
It can be operated according to the actual situation in the medical field.

Further, among the images of the local area displayed on the local image display means, at least an image of an abnormal shadow candidate is
Computer support having a configuration further including local image enhancement means for performing image enhancement processing on at least abnormal shadow image data among local image data so that image reading performance is improved over the overall image displayed on the overall image display means. According to the diagnostic imaging apparatus, the local image data extracted by the local region extracting unit is input to the local image enhancing unit, and the local image enhancing unit performs image enhancement processing such as gradation processing, frequency processing, and enlargement processing. Will be applied.

The purpose of these image enhancement processes is to improve the observation and interpretation performance of the image of the local area to be diagnosed. For example, in the case of gradation processing, the image is displayed on the local image display means. The contrast of the local image (or the image of the abnormal shadow) should be higher than that of the whole image displayed on the whole display means, and it is particularly desirable that the contrast be 1.2 times or more. In the case of frequency processing, the local image (or the image of the abnormal shadow) displayed on the local image display means may have a higher degree of emphasis than the whole image displayed on the whole display means. It is desirable to do. further,
In the case of the enlargement processing, the local image (or the image of the abnormal shadow) displayed on the local image display means is compared with the whole image displayed on the whole display means so that the fine structure can be easily observed.
It is desirable to enlarge the display to 1.5 times or more.

As the enlargement processing, according to the size of the abnormal shadow candidate detected by the abnormal shadow candidate detecting means,
According to the diagnostic apparatus adopting the configuration for performing the processing set so that the enlargement ratio changes, the displayed image of the abnormal shadow candidate is always substantially constant regardless of the size of the detected abnormal shadow candidate. By changing the enlargement ratio so that the size becomes the size, even if the detected abnormal shadow candidate is a minute size, the abnormal shadow candidate can always be interpreted as an image of a substantially constant size, thereby improving diagnostic performance. Can be achieved.

Further, in the diagnostic apparatus having the whole image enhancing means, the image reading performance of the whole image can be improved with respect to the image data representing the whole image displayed on the whole image display means. The image enhancement processing such as gradation processing or frequency processing is performed by the whole image enhancement means. Even when the whole image is enhanced by the whole image enhancing means, at least the image of the local area including the abnormal shadow is displayed on the local image display means. In order to make the observation and interpretation performance better than the image displayed on the display means, the image enhancement processing of the local image displayed on the local image display means is compared with the image enhancement processing of the entire image displayed on the entire display means. In contrast, the gradation processing is higher than the contrast, the frequency processing is higher in the degree of emphasis, and the enlargement processing is higher so that the fine structure can be easily observed. In particular, it is desirable that the contrast increase by the gradation processing be 1.2 times or more, the emphasis degree by the frequency processing is 1.1 times or more, and the enlargement ratio by the enlargement processing is 1.5 times or more.

In the diagnostic apparatus in which the whole image display means also serves as the local image display means, while displaying the whole image on the display surface of the whole image display means, the display request is issued from the local image display request means. Then, an image of a local region including an abnormal shadow is displayed in a part of the entire image.

Thus, on the display surface of one display means,
By displaying the entire image and the local image including the abnormal shadow, the positioning of the abnormal shadow in the entire image is further facilitated, and the diagnostic performance can be improved.

The local area image displayed on the whole image display means is displayed in a display area corresponding to the local area image and different from the local area in the radiation image displayed on the whole image display means. By doing so, the local region image is not superimposed on the local region in the displayed whole image and displayed, and the local region image displayed simultaneously while recognizing the position of the local region in the whole image is observed. And diagnostic performance is improved.

Further, by determining that the display area of the local area image displayed on the whole image display means falls within a display area different from the subject displayed on the whole image display means, It is possible not only to prevent the region image from overlapping the local region in the entire image, but also to overlap the entire image of the subject in the entire image. Note that by making the local area image scrollable within the window area of the whole image display means, even if the size of the local area image in the display area of the whole image display means is smaller than the size of the local area image, It is possible to display only a part of the region image and scroll the local region image within the window to observe a necessary portion of the local region image.

Further, in a configuration in which a plurality of such regions are set, even when a plurality of abnormal shadows are detected, all of them can be observed simultaneously with the entire image, which is more effective for diagnosis. It is.

[0096]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, specific embodiments of the computer-aided image diagnostic apparatus of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing a schematic configuration of a first embodiment of a computer-aided image diagnostic apparatus according to the present invention. The computer-aided image diagnostic apparatus shown in FIG. 1 includes a whole image memory 10 for storing digital image data (whole image data) S representing a whole radiation image (whole image) P of a subject, Based on the whole image display means 30 such as a CRT for displaying the whole image P of the subject based on the whole image data S stored in the whole image memory 10, based on the whole image data S stored in the whole image memory 10,
An abnormal shadow candidate detecting means 40 for detecting an abnormal shadow candidate P1 in the radiation image P; a determining means 50 for determining whether or not the abnormal shadow candidate P1 is detected by the abnormal shadow candidate detecting means 40; Is detected by the determination means 50, the candidate P1 of the abnormal shadow in the entire image data S stored in the entire image memory 10 is determined.
, A local region extracting unit 60 for extracting image data (local image data) S2 representing an image P2 of a local region including the local image data S2 extracted by the local region extracting unit 60
And a local image display means 90 such as a CRT for displaying an image P2 of the local area based on the image.

In the following, in the present embodiment, an image of a breast having a tumor inside is used as a radiation image P, and a tumor shadow indicating the tumor is used as an abnormal shadow candidate P1.
A description will be given of a case where an iris filter is applied as an abnormal shadow candidate detecting unit, but the computer-aided image diagnostic apparatus of the present invention is not limited to such an embodiment.

Here, the iris filter calculates the gradient of the image signal (density value) for detecting a tumor shadow in a radiographic image as a gradient vector, and outputs the degree of concentration of the gradient vector. The iris filter processing refers to the degree of concentration of this gradient vector (formula (1
The iris filter in the present embodiment does not refer to this algorithm itself, but means for detecting a tumor shadow candidate by this algorithm. Shall mean.

The above-mentioned local region is a region including a tumor portion which is a candidate for an abnormal shadow and including the vicinity of the tumor portion.

Hereinafter, the operation of the computer-aided image diagnostic apparatus according to the present embodiment will be described.

The whole image data S representing a radiation image P of a breast having a tumor therein is input from an external storage medium such as a magneto-optical disk or an image reading device to the whole image memory 10. The whole image data S is directly inputted to the whole image display means 30 (path A in FIG. 1), or is inputted to the whole image display means 30 as once stored in the whole image memory 10 (FIG. 1). 2B), the whole image display means 30 displays the whole radiation image P based on the whole image data S as shown in FIG.

Since the radiation image P also includes the above-mentioned candidate P1 for a tumor shadow, the whole image display means 30
Displays the whole image P including the tumor shadow candidate P1.

On the other hand, the whole image data S stored in the whole image memory 10 is also inputted to the abnormal shadow candidate detecting means (iris filter in this embodiment) 40. The iris filter 40 obtains a gradient vector based on image data at an arbitrary pixel over the entire input entire image data S according to the above-described procedure, and evaluates the degree of concentration of the direction of the gradient vector. As a result, the degree of concentration of the gradient vector given by a predetermined calculation formula is calculated, and the distribution is evaluated to detect image data S1 indicating a tumor shadow candidate P1.

As described above, the pixel (position) of the image data (hereinafter referred to as tumor image data) S1 indicating the tumor shadow candidate P1 is specified by the iris filter 40.

The determination means 50 determines that the iris filter 40 has detected the tumor image data indicating the tumor shadow candidate P1, and determines the tumor image data S1.
The position data (hereinafter referred to as tumor pixel position data) D1 for specifying the pixel position is input to the local region extracting means 60. On the other hand, when the iris filter 40 determines that the tumor image data indicating the tumor shadow candidate P1 has not been detected by the iris filter 40, the tumor image data S
The process ends without outputting the tumor pixel position data D1 specifying the pixel position of No. 1.

When it is determined that the tumor image data has been detected, the local image extracting unit 60 also receives the whole image data S stored in the entire image memory 10 and inputs the local image data.
Is a neighboring pixel (a local region as a set of these pixels) including the pixel of the tumor image data S1 based on the tumor pixel position data D1 in the input whole image data S.
Is specified in accordance with a preset processing procedure, and then the local image data S
2 is extracted.

The extracted local image data S2 is input to the local image display means 90. On the display surface of the local image display means 90, based on the input local image data S2,
An image P2 of the local area including the tumor shadow P1 is displayed (see FIG. 2B).

As described above, only the image of the candidate P1 of the tumor shadow among the whole images is separately displayed on the local image display means 90. Can be concentrated, and diagnostic performance can be improved. In addition, the time required for diagnosis can be shortened.

In the present embodiment, the whole image data S inputted to the abnormal shadow candidate detecting means 40 is inputted through the whole image memory 10 (path shown in FIG. 1B).
However, the present invention is not limited to such an embodiment, and the information may be directly input from a magneto-optical disk, an image reading device, or the like as shown in the path of FIG. The same applies to the following embodiments.

In the computer-aided image diagnostic apparatus according to the present embodiment, a breast having a tumor inside is a subject, a tumor shadow indicating the tumor is an abnormal shadow candidate, and an iris filter is an abnormal shadow candidate detecting means. However, the computer-aided image diagnostic apparatus of the present invention is not limited to this mode. For example, as a subject, a breast having a cancerous part showing a calcified image in which an extremely fine high-density region is densely distributed to some extent, a calcified image as an abnormal shadow candidate, and an abnormal shadow candidate detecting means It is also possible to adopt a mode in which the above-described morphological filters for performing a process of detecting a calcified image are applied.

Further, when displaying each of these abnormal shadow candidates on the local image display means, the feature amount as an abnormal shadow candidate which has been detected as an abnormal shadow candidate by the abnormal shadow candidate detection means, ie, a tumor For the shadow, the value of the degree of concentration of the gradient vector and the degree of the irregularity of the tumor shadow may be displayed, and for the calcified image, the calcification density may be displayed as quantification information together with the image of the abnormal shadow. .

FIG. 3 is a block diagram showing a schematic configuration of a second embodiment of the computer-aided image diagnostic apparatus according to the present invention. The illustrated computer-aided image diagnostic apparatus differs from the first embodiment shown in FIG. 1 in that the local image data S2 is temporarily stored between the local region extracting means 60 and the local image displaying means 90. The local image data S2 is output from the image memory 15 and the local image memory 15 only when a predetermined image display request is received from the outside, and the image P2 of the local region is displayed on the local image display means 90. The configuration is the same as that of the first embodiment except that the configuration further includes an image display request unit 80.

That is, the local area extracting means 60 extracts the local image data S2 based on the judgment result of the judging means 50 as in the first embodiment. Then, the extracted local image data S2 is input to the local image memory 15 and is temporarily stored, and the processing is temporarily stopped at this stage. Here, when the observer of the image inputs a signal indicating a request to display a local image to the local image display requesting means 80, a signal indicating a request to display a local image is sent from the local image display requesting means 80 to the local image memory The local image memory 15 receives this signal and outputs the stored local image data S2 to the local image display means 90, and the processing is restarted.

As described above, according to the computer-aided image diagnostic apparatus of the present embodiment, the local image data S2 of the local area including the detected abnormal shadow is temporarily stored in the local image memory. Whether or not to display the local image at 90 can be arbitrarily determined by the observing radiologist so as to adapt to the tracing time and the like. That is, the interpreter
Normally, after observing the whole image first and recognizing the information of the whole image to some extent, it is usual to observe the detailed structure. Therefore, the interpreter first observes the whole image displayed on the whole image display means, and inputs a display request to the local image display request means after the observation or at a desired time during the observation. The image of the local region including the abnormal shadow candidate can be immediately displayed on the local image display means, and can be operated in accordance with the actual situation in the medical field.

FIG. 4 is a block diagram showing a schematic configuration of a computer-aided image diagnostic apparatus according to a third embodiment of the present invention. The illustrated computer-aided image diagnostic apparatus is different from the first embodiment shown in FIG. 1 in that at least the abnormal shadow candidate image P1 of the local area image P2 displayed on the local image display means 90 is entirely The local image data S is selected so that the interpretation performance is higher than that of the whole image P displayed on the image display means 30.
2 is the same as that of the first embodiment except that at least the local image enhancing means 70 for performing the image enhancing process on the abnormal shadow image data S1 is further added.

That is, the whole image data S is input to the whole image display means 30 via the whole image memory 10 (path A shown in FIG. 4) or directly (path B shown in FIG. 4) and The image P is displayed (FIG. 5A). On the other hand, the local region extracting unit 60
The local image data S2 is extracted as in the first embodiment. Then, the extracted local image data S2 is input to the local image enhancing means 70, and the local image enhancing means 70 performs image enhancement processing such as gradation processing, frequency processing, and enlargement processing.

The image enhancement processing is specifically set to perform the following signal processing (1) to (3). That is, (1) the gradation processing is set such that the contrast of the local image P2 displayed on the local image display means 90 is 1.2 times or more higher than that of the whole image P displayed on the whole image display means 30.

(2) In the frequency processing, the local image display means 90
Is set so that the degree of enhancement is 1.1 times or more higher than that of the whole image P displayed on the whole image display means 30.

(3) The enlargement process is set so that the size of the local image P2 displayed on the local image display means 90 is 1.5 times or more the size of the whole image P displayed on the whole image display means 30.

The local image data S2 after the image enhancement processing is input to the local image display means 90, and the local image display means 90 converts the image P2 of the local area based on the input local image data S2. Display (Fig. 5
(B)). The image P2 of the local area displayed on the local image display means 90 has a contrast of 1.2 times or more and a degree of enhancement with respect to the entire image P displayed on the entire image display means 20 by setting of the local image enhancement means 70. 1.1 times or more in size
Since it is 1.5 times or more, the observation and interpretation performance is superior to the whole image P displayed on the whole image display means 20, and only the image P2 of the local region is displayed separately from the whole image P. Can concentrate the observation consciousness on the image P2 of the local area including the tumor shadow P1, improve the diagnostic performance, and shorten the time required for the diagnosis.

Note that the image enhancement processing such as the gradation processing, frequency processing, and enlargement processing by the local image enhancement means 70 is as follows.
The local image data S2 is not limited to being applied to the entire area, but may be applied only to the abnormal shadow image data S1 excluding the image data of the neighboring area in the local image data S2. Thus, the abnormal shadow image data S
In the case where the image enhancement processing is performed only on the image No. 1, the values of the contrast, the degree of enhancement, and the degree of improvement of the size are values for the abnormal shadow candidate image for the entire image as described above. In other words, the image enhancement processing by the local image enhancement means 70 is as follows: (1 ′) The gradation processing is the whole image P that the candidate image P1 of the abnormal shadow displayed on the local image display means 90 is displayed on the overall image display means 30 The contrast is set to be 1.2 times or more.

(2 ') Frequency processing is performed by local image display means.
The emphasis degree of the candidate image P1 of the abnormal shadow displayed at 90 is set to be 1.1 times or more higher than that of the entire image P displayed at the entire image display means 30.

(3 ') Enlargement processing is performed by the local image display means 90
Is set to be 1.5 times or more the size of the whole image P displayed on the whole image display means 30.

The above-described enlargement processing is performed, for example, as described above.
The processing is not limited to the enlargement processing with a constant enlargement rate such as 1.5 times, and is a processing set so that the enlargement rate changes according to the size of the abnormal shadow candidate detected by the abnormal shadow candidate detection means 40. You may. That is, when displaying the image P1 of the abnormal shadow candidate on the local image display means 90, if the size of the detected abnormal shadow candidate is small, enlargement processing is performed at a higher magnification, and the size of the detected abnormal shadow candidate is changed. If the size is large, the enlargement processing is performed at a relatively small enlargement ratio, and the apparent size of the abnormal shadow on the display surface of the display means 90 is always substantially constant irrespective of the actual size of the detected abnormal shadow. You may do so.

More specifically, the local image enhancement means 70 includes an abnormal shadow size calculating means for calculating the size of the abnormal shadow candidate based on the position data D1 of the abnormal shadow candidate obtained by the abnormal shadow candidate detecting means 40. In order that the display size of the abnormal shadow candidate image on the local image display unit 90 is always substantially constant, an enlargement ratio corresponding to the size of the abnormal shadow candidate calculated by the abnormal shadow size calculation unit is set in advance. The configuration may include a ratio setting table (see FIG. 6) and an enlargement processing unit that performs enlargement processing on the local image data S2 or the abnormal shadow image data S1 according to the enlargement ratio set by the enlargement ratio setting table.

Thus, regardless of the size of the detected abnormal shadow candidate, the magnification is changed so that the image P1 of the abnormal shadow candidate displayed on the local image display means 90 always has a substantially constant size. Thus, even if the detected abnormal shadow candidate has a very small size, the interpreter can always read the image P1 of the abnormal shadow candidate as an image having a substantially constant size, thereby improving diagnostic performance. Can be.

The size of the abnormal shadow candidate calculated by the abnormal shadow size calculating means is calculated based on the peripheral length of the abnormal shadow candidate such as a tumor shadow or a calcified shadow, or an abnormal shadow candidate detecting means. When 40 is a morphology filter, it can be calculated by the following method.

That is, the value (the density value of the calcified shadow by the morphology filter) P of the following equation (14) based on the above-described opening calculation is a preset abnormal shadow (calcified shadow) candidate. Is larger than the threshold value Th (P ≧ Th), a value of “1” is assigned to the pixel as a calcification shadow candidate, and if the pixel value is smaller than the threshold value Th (P <Th), the pixel is A binarization process is performed by assigning a value of “0” as not being a simulated shadow candidate, and an abnormal shadow candidate is determined based on the sum or average value of the number of pixels having the value of “1”, which are calcified shadow candidates. Can be calculated.

[0130]

(Equation 15) Further, the above-described local image enhancing means 70 can be applied to the computer-aided image diagnostic apparatus of the second embodiment. Specifically, the image enhancement processing is performed on the local image data S2 output from the local region extracting means 60 (see FIG. 7), or the local image data S2 output from the local image memory 15 is processed. As shown in FIG.
), The local image enhancing means 70 may be provided.

As described above, the local image data S2 after the image enhancement processing by the local image enhancement means 70 is temporarily stored in the local image memory in the case of the configuration shown in FIG. After that, the image data is input to the local image display means 90. On the other hand, in the case of the configuration shown in FIG.
Is displayed on the local image display means 90 as shown in FIG. 5B.

FIG. 9 is a block diagram showing a schematic configuration of a fourth embodiment of the computer-aided image diagnostic apparatus according to the present invention. The computer-aided image diagnostic apparatus shown in the drawing is different from the third embodiment shown in FIG. 4 in that gradation processing and frequency processing are also performed on the entire image data S indicating the entire image P displayed on the entire image display means 30. The third embodiment is the same as the third embodiment except that the whole image enhancement means 20 for performing image enhancement processing such as processing or a combination thereof is further added.

That is, the operation of displaying the image P2 of the local area on the local image display means 90 is the same as that of the third embodiment, while the whole image data S input to this device is stored in the whole image memory. The data is input to the whole image enhancement means 20 via the path 10 (path A shown in FIG. 9) or directly (path B shown in FIG. 9). The whole image enhancement means 20
The above-described gradation processing for the input whole image data S,
After inputting an image enhancement process such as a frequency process or a process combining them, the image is input to the whole image display means 30. The whole image display means 30 receives the whole image data S
Is displayed based on the whole image P.

As described above, by performing a certain degree of image enhancement processing on the whole image, the whole image P suitable for observation and interpretation can be displayed.

Even when the image enhancement processing is applied to the entire image P, the entire image enhancement means 20 aims at improving the readability of the entire image P. Image P2 (or abnormal shadow P1)
The local image P2 (or the abnormal shadow P1) subjected to the image enhancement processing by the local image enhancement means 70 is superior in the image interpretation performance of (1).

FIG. 10 shows the whole image display means 30 according to the third embodiment of the present invention shown in FIG.
In the fourth configuration, image enhancement processing such as gradation processing, frequency processing, or a combination thereof is also performed on the entire image data S indicating the entire image P displayed on the fourth image.
FIG. 10 is a block diagram showing a schematic configuration of an embodiment similar to that of the third embodiment, and is the same as the third embodiment shown in FIG. 7 or FIG. 8 except that the entire image enhancing means 20 is further added.

The operation of these embodiments is the same as that of the third embodiment shown in FIG. 7 or 8 in that the operation of displaying the image P2 of the local area on the local image display means 90 is the same as that of the third embodiment. The operation of displaying the image P on the whole image display means 30 is the same as that of the fourth embodiment shown in FIG.

FIG. 12 shows an embodiment in which the entire image display means 30 also serves as the local image display means 90 in the embodiment shown in FIG. That is, in the embodiment shown in FIG. 11, a window area W for superimposing and displaying the local area image P2 is provided on a part of the display surface of the entire image display means 30 on which the entire image P is displayed. . Therefore,
As shown in FIG. 13, the entire image display means 30 displays the entire image P
Is displayed, the local region image P2 originally displayed on the local image display means 90, which is originally different from this, is displayed so as to overlap a part of the whole image P displayed on the whole image display means 30 ( In the window region W of the whole image display means 30 on which the local region image P2 is superimposed, a part of the whole image P and the local region image P2 are not displayed in an overlapping manner. This means that only the local area image P2 is displayed without displaying a part of the entire image P, and the remaining part of the entire image P is displayed on a display portion that is not overlapped.) Such a window area W is
This is set by display control means (not shown).

In the case of the embodiment shown in FIG. 12, before the display request for the local image display request 80 is input from the interpreter, the entire image P is displayed on the entire image display means 30. Only after the display request is input, the window area W is provided in the whole image display means 30 and the local area image P2 is displayed as described above (FIG. 13).

As described above, the entire image P and the local image P2 including the abnormal shadow P1 are displayed on the same display surface of the single display means.
Are displayed at the same time, it becomes much easier to understand the positioning of the abnormal shadow P1 in the entire image P,
This can also improve the diagnostic performance.

The entire image P and the image P of the local area
Various modes can be adopted as the display mode 2 depending on the subject to be imaged. For example, when displaying an X-ray image of a breast or the like of one patient, in addition to displaying only one breast image, a configuration including two sets of whole image display means and two sets of local image display means is employed. Image P2 of the local area including the abnormal shadow of the breast
Is displayed on one of the local image display means, and at the same time, an image P2 ′ of the local region of the other breast corresponding to the local region of the one breast is converted to an image P2 of the local region of the one breast.
By performing the same emphasizing process as above and displaying the same on the other local image display means, the radiologist can compare and observe the same portions P2 and P2 'of the left and right breasts, and further diagnose Performance can be improved.

The whole image display means may be configured to display the whole image of the pair of breasts separately on the left and right sides on separate whole image display means.

Further, the above-mentioned window areas are provided in the two whole image display means, respectively, and the images of the local areas corresponding to the whole images of the left and right breasts respectively displayed on both the whole image display means are displayed. You may do so.

The image of one breast and the image of the other breast photographed separately from the breast may be simultaneously displayed on the same display surface of a single image display means. Specifically, as shown in FIG. 14, the whole image P of one breast is displayed on the right half of the display surface of the whole image display means 30, and the whole image P 'of the other breast is displayed on the same whole image display means 30. In the left half of the face, these are displayed simultaneously (for example, face-to-face or back-to-back) at the same time (FIG. (A)), and an abnormal shadow candidate P1 is detected from the whole image data S of one breast. In this case, the image P2 of the local area including the abnormal shadow candidate P1 is displayed in the window area W provided on the right half display surface after the image P2 of the local area including the abnormal shadow candidate P1 is emphasized. The image P2 'of the local region of the other breast corresponding to P1 is subjected to the same enhancement processing, and then the window region W' provided on the left half display surface is displayed.
((B) in the figure).

In the embodiment configured as described above,
Both breasts can be directly compared and observed, and an effect is obtained that a detailed difference between the two breasts, for example, a difference in the presence or absence of an abnormal shadow and a form of the abnormal shadow between the two breasts can be easily diagnosed.

Note that various modes can be adopted as examples in which the window region W is provided in the whole image display means 30 and the whole image and the image of the local region are displayed on a single display surface. A method of setting the size and position of the window area at this time will be specifically described below.

FIG. 15 shows a whole image display means 30 in which the whole images of the left and right breasts are back-to-back and simultaneously displayed on a single display surface as shown in FIG. The description of the left half is omitted.

The whole image display means 30 shown in FIG. 15 has a vertical length Y of the display surface and a distance y from the upper edge of the display surface in the whole image P of the detected abnormal shadow P1. The window area W is set to the following position by display control by display control means (not shown).

That is, when y <Y / 2, the position of the window area W is set to an area below the vertical center on the display surface, and conversely, when y> Y / 2, Is set in an area above the center in the vertical direction on the display surface.

As described above, when the position of the abnormal shadow P1 is on the upper side, by providing the window area W on the lower side, the display of the abnormal shadow P1 on the entire image P is not obstructed by the display of the window area W. While confirming the position of the abnormal shadow P1 in the image P, it is possible to observe the processed image of the abnormal shadow P1 displayed in the window region W and subjected to the image processing with excellent observation performance, and the diagnostic performance of the image is improved. improves. The abnormal shadow P1
Is located on the lower side, the window region W is provided on the upper side, and the same effect as described above can be obtained.

FIG. 16A shows a case where the window area W does not obstruct the display of not only the abnormal shadow P1 but also the image of the subject itself.

That is, the whole image display means in FIG. 16A is controlled by a display control means (not shown) so that the window area W is positioned in the entire image P at a position where there is no image of the subject itself (in the drawing, the breast) (for example, a so-called element of radiation). (An image area indicating a missing portion).

This is performed by a display control means (not shown) calculating an area where the subject itself does not exist based on the sizes a and b of the area occupied by the subject itself on the display surface. Therefore, not only the position of the window region W but also its size changes according to the region occupied by the subject itself.

By the way, the local area to be displayed in the window area W may be enlarged to a desired size by the local image enhancing means, and the entire abnormal shadow P1 may not fit in the window area W. Is also conceivable. In such a case, the window area W is set to have a size changed according to the area occupied by the subject itself, and corresponds to the side and bottom ends of the window area W in the vertical and horizontal directions. Scroll bar S
B, the image of the local area to be displayed in the window area W can be freely moved in the window area W by, for example, clicking the scroll bar SB with a screen operation mouse or the like. The whole image of the region can be sequentially observed in the window.

That is, as shown in FIG. 16 (B), the window w indicated by the broken line can be moved relative to the image of the local area to be displayed in the window area W. May be displayed in the window area W. The above operation may be performed by the above-described display control means.

FIG. 17 shows a case where a plurality of abnormal shadows P1
, P11, P21, and P31 are detected, and a plurality of local region images corresponding to these abnormal shadows are displayed separately in a plurality of window regions W1, W2, W3, and W4.

FIG. 17A shows all the window regions W1, W2, W3, W regardless of the size of the local region image.
FIG. 4B shows the same size as that of FIG.
Represents each window region W according to the size of the local region image.
In the figure, there is shown a display in which the sizes of 1, W2, W3, and W4 are changed.

In FIG. 17A, each window area W1
, W2, W3, W4 abnormal shadow P1
, P11, P21, and P31 are displayed in that size when the size is set in advance by the local image enhancing means, and when displayed in such a size, the abnormal shadows P1, P11, P21, and P31 are displayed. Is larger than the window regions W1, W2, W3, and W4, the scroll bar SB as shown in FIG. 16 (A) is displayed in each of the window regions W1, W2, W3, and W3.
What is necessary is just to set for every W4.

The window areas W1, W2, W3
, W4, the size of the abnormal shadows P1, P11, P21, P31 may be forcibly enlarged or reduced.

On the other hand, as shown in FIG. 17B, each of the window regions W1, W2, W
3 and W4, the window areas W1, W2, W3, W4
In the case where the object blocks the display of the subject,
As in the case of (A), the window regions W1, W2,
The sizes of W3 and W4 may be forcibly reduced.
In this case, the window areas W1, W2, W3, W
4, abnormal shadows P1, P11, P21,
P31 also has window areas W1, W2, W3, W4
May be reduced in accordance with the reduction ratio, or the scroll bar SB may be provided with the size of the abnormal shadows P1, P11, P21, and P31 unchanged.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a schematic configuration of a first embodiment of a computer-aided image diagnostic apparatus according to the present invention;

FIG. 2A shows an entire image P displayed on the entire image display means.
(B) A diagram showing a state where the local image P2 including the abnormal shadow P1 is displayed on the local image display means.

FIG. 3 is a block diagram showing a schematic configuration of a second embodiment of the computer-aided image diagnostic apparatus of the present invention.

FIG. 4 is a block diagram showing a schematic configuration of a computer-aided image diagnostic apparatus according to a third embodiment of the present invention;

FIG. 5 (A) shows the whole radiographic image P on the whole image display means.
(B) A diagram showing a state where the local image P2 including the abnormal shadow P1 is displayed on the local image display means.

FIG. 6 is a graph showing an enlargement ratio setting table in which an enlargement ratio according to the size of an abnormal shadow candidate calculated by the abnormal shadow size calculation unit is set in advance.

FIG. 7 shows a local image enhancement unit for the computer-aided image diagnostic apparatus of the second embodiment shown in FIG. 3 so as to perform image enhancement processing on local image data S2 output from a local region extraction unit 60; Block diagram showing a schematic configuration of an embodiment to which 70 is applied

FIG. 8 shows a local image emphasizing unit 70 for the computer-aided image diagnostic apparatus of the second embodiment shown in FIG. 3 so as to perform an image emphasizing process on local image data S2 output from the local image memory 15. Block diagram showing a schematic configuration of an embodiment to which the present invention is applied.

FIG. 9 is a block diagram showing a schematic configuration of a computer-aided image diagnostic apparatus according to a fourth embodiment of the present invention;

10 is a configuration in which image enhancement processing is performed on the entire image data S indicating the entire image P displayed on the entire image display means 30 in the third embodiment of the present invention shown in FIG. Block diagram showing a schematic configuration of the embodiment of FIG.

11 is different from the third embodiment of the present invention shown in FIG. 8 in that image enhancement processing is performed on the entire image data S indicating the entire image P displayed on the entire image display means 30. Block diagram showing a schematic configuration of the embodiment of FIG.

FIG. 12 is a block diagram showing a schematic configuration of an embodiment in which the whole image display means 30 also serves as the local image display means 90 in the embodiment shown in FIG. 11;

FIG. 13 is a diagram showing a display state of an entire image and a local image in a configuration in which the entire image display unit also serves as a local image display unit.

14A is a diagram showing a state in which two breast images are simultaneously displayed on the whole image display means. FIG. 14B is a view showing two breast images simultaneously displayed on the whole image display means and a local image display means. Showing the display state of the entire image and the local image in a configuration that also serves as

FIG. 15 is a diagram showing a state where a window area W is set on the display surface of the whole image display means.

16A is a diagram illustrating a state in which a scroll bar is provided in a window area. FIG. 16B is a diagram illustrating a moving state of a window.

17A is a diagram showing a state in which a plurality of window regions are displayed at the same size regardless of the size of the local region image. FIG. 17B is a diagram showing a state in which a plurality of window regions change according to the size of the local region image. Diagram showing the state displayed

FIG. 18 is a view showing a 5 × 5 mask for calculating a direction θ of a gradient vector in the iris filter.

FIG. 19 is a diagram showing the concept of the degree of concentration of a gradient vector.

FIG. 20 is a diagram showing an evaluation of the degree of concentration, centering on a pixel of interest.
Diagram showing pixels on radial lines in 32 directions

FIG. 21 is a view showing a basic operation of a morphological filter. FIG. 21 shows (A) dilation processing, (B) erosion processing, (C) opening processing, and (D) closing processing, respectively.

FIG. 22 is a diagram showing four linear structural elements in a morphological filter.

[Explanation of symbols]

 10 Whole image memory 15 Local image memory 20 Whole image enhancement means 30 Whole image display means 40 Abnormal shadow candidate detection means 50 Judgment means 60 Local area extraction means 70 Local image enhancement means 80 Local image display request means 90 Local image display means P subject Radiation image P1 Abnormal shadow candidate P2 Image of local area including abnormal shadow candidate S Overall image data S2 Local image data A, B, a, b Input path of overall image data S

Claims (2)

[Claims] When displaying the respective radiographic images of a pair of left and right breasts side by side, the display positions of the respective breasts are adjusted so that the images of the pair of right and left breasts can be observed with the chest wall of each breast back to back. A breast image display method characterized by displaying the image on a breast. 2. A breast image display device for displaying a pair of left and right breast images side by side, the breast images being displayed such that the chest wall of each breast can be observed back to back. A breast image display device comprising image position adjusting means for adjusting a display position of a breast.