JP2004041490A - Diagnostic imaging support system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a diagnostic imaging support system capable of giving more minute information for diagnosis to a user. <P>SOLUTION: A first classifying means 15 makes image analysis of an image region of an abnormal shadow candidate detected by a means for detecting the abnormal shadow candidate, and classifies it into the central part and the peripheral part. The center of gravity of the detected abnormal shadow candidate is computed. Points located at a prescribed ratio of a distance from the center of gravity to an end part of the region of the abnormal shadow candidate are found, and a boundary between the central part and the peripheral part is determined by connecting the points. The image region of the abnormal shadow candidate is classified into the central part which is an inner region between the center of gravity and the boundary and into the peripheral part which is a region from the boundary to the end part of the region of the abnormal shadow candidate. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an image diagnosis support device that detects an abnormal shadow candidate from image information obtained by imaging a subject.
[0002]
[Prior art]
In the medical field, a medical image obtained by imaging a patient as a subject using a medical imaging device such as CT (Computed Tomography) or MRI (Magnetic Resonance Imaging) is converted into digital data, and when a doctor makes a diagnosis, , A medical image has been displayed on an image display device such as a CRT, and interpretation has been performed. Particularly in recent years, an image diagnosis support called a computer assisted device (Computer-Aided Diagnostics; hereinafter, referred to as CAD) for detecting abnormal shadow candidates such as lung cancer and breast cancer for the purpose of reducing the burden on the interpreting doctor and reducing oversight of abnormal shadows. Equipment has been developed.
[0003]
[Problems to be solved by the invention]
In the conventional CAD, an abnormal shadow candidate detected when displaying an image is pointed out by a marker such as an arrow, and the position of the abnormal shadow candidate is notified to the user. However, in order to distinguish between benign and malignant abnormal shadow candidates, the state of the margin of the abnormal shadow may be important, and information on the margin of the abnormal shadow can be obtained simply by pointing out the abnormal shadow candidate with a marker. And it was not possible to efficiently determine how much the condition worsened compared to the previous condition.
[0004]
SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and provides an image diagnosis support apparatus that can provide a user with more detailed information on diagnosis.
[0005]
[Means for Solving the Problems]
The invention according to claim 1 is
In an image diagnosis support apparatus including a display unit that displays image information of an image of a subject,
Abnormal shadow candidate detection means for detecting an abnormal shadow candidate from the image information input by the image information input means,
First classification means for classifying the image area of the abnormal shadow candidate detected by the abnormal shadow candidate detection means into a plurality of image areas;
Display control means for identifying and displaying each image area classified by the first classification means when the image information is displayed by the display means;
It is characterized by having.
[0006]
According to a second aspect of the present invention, in the image diagnosis support apparatus according to the first aspect,
The display control means is characterized in that each image area classified by the first classification means is identified and displayed with a different color.
[0007]
According to a third aspect of the present invention, in the image diagnosis support apparatus according to the first or second aspect,
The first classification unit is characterized in that the image region of the abnormal shadow candidate is classified into a central portion and an image region of a peripheral portion thereof.
[0008]
According to the first, second, and third aspects of the present invention, an abnormal shadow candidate is detected from the image information, and the image area of the abnormal shadow candidate is classified into a central part and a peripheral part image area. At the time of display, each classified image area is identified and displayed with a different color, so it not only points out abnormal shadows, but also an edge index that is an important index for distinguishing benign and malignant abnormal shadows. Information can be provided to the user, and the efficiency of diagnosis can be improved and the accuracy of diagnosis can be improved.
[0009]
The invention according to claim 4 is the image diagnosis support device according to any one of claims 1 to 3,
The image processing apparatus further includes a storage unit that stores image information in which the abnormal shadow candidate is detected by the abnormal shadow candidate detecting unit.
[0010]
According to a fifth aspect of the present invention, in the image diagnosis support apparatus according to the fourth aspect,
The storage means may further store information on the abnormal shadow candidate.
[0011]
According to the fourth and fifth aspects of the present invention, since the image information of the detected abnormal shadow candidate and the information of the abnormal shadow candidate are stored, the past image information of the subject or the information of the past abnormal shadow candidate is stored. It is possible to observe the change over time by reading, and it is possible to further improve the diagnostic accuracy.
[0012]
According to a sixth aspect of the present invention, in the image diagnosis support apparatus according to the fourth aspect,
A past image in which the same abnormal image candidate as the current image information in which the abnormal image candidate is detected by the abnormal image candidate detection means is the same image information as that of the same subject. When information is stored in the storage unit, the current image information and the past image information are collated to classify an abnormal shadow candidate image region into an image region having a change and an image region having no change. A second classification means,
The display control means, when displaying the image information in which the current abnormal shadow candidate is detected by the display means, identifies and displays each image area classified by the second classification means. .
[0013]
According to a seventh aspect of the present invention, in the image diagnosis support apparatus according to the sixth aspect,
The display control means is characterized in that an image area having a change classified by the second classification means is colored and identified and displayed.
[0014]
According to the sixth and seventh aspects of the present invention, the current image is compared with the stored past image, and the image area having a change in the abnormal shadow candidate image area is colored and identified and displayed. The temporal change of the abnormal shadow can be clearly recognized, and the efficiency of diagnosis can be improved and the accuracy of diagnosis can be improved.
[0015]
The invention according to claim 8 is the image diagnosis support device according to any one of claims 1 to 7,
The image display device further includes an operation unit that performs an operation instruction to perform identification display of each image region and release of the identification display by the display control unit.
[0016]
According to the eighth aspect of the present invention, the cancellation and display of the identification display of each image area can be instructed by the operating means. Therefore, when observing an image, the cancellation of the identification display is instructed, and the identification display is performed. When it is desired to obtain more detailed information on abnormal shadows, an identification display can be instructed, and a display operation can be performed as desired by the user.
[0017]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
First, the configuration will be described.
FIG. 1 is a diagram illustrating a configuration of an image diagnosis support apparatus 10 according to the present embodiment. In FIG. 1, an image diagnosis support apparatus 10 includes an image data input unit 11, an image data storage unit 12, an abnormal shadow candidate detection unit 13, a storage unit 14, a first classification unit 15, a second classification unit 16, and a display unit. 17, operation means 17a is provided.
[0018]
The image data input unit 11 scans a film on which a medical image of a patient is recorded with a laser beam using a laser digitizer, measures the amount of transmitted light, and converts the value into an analog-to-digital signal. This is input as digital image data.
[0019]
The input of the image by the image data input means 11 is not limited to the laser digitizer, and an optical sensor such as a CCD may be used. In this case, optical scanning is performed on the film, and the reflected light is photoelectrically converted by a CCD to obtain digital image data. Further, as disclosed in Japanese Patent Application Laid-Open No. 55-12429, a photographing apparatus for converting a medical image photographed using a stimulable phosphor into a digital image and outputting the digital image instead of reading an image recorded on a film is disclosed. A connection may be made to obtain image data from this photographing device. In this case, a film is not required and cost can be reduced.
[0020]
Alternatively, a medical image obtained from a flat panel detector (FPD) that captures a radiation image with a plurality of two-dimensionally arranged detection elements and outputs the image as an electric signal may be input. For example, Japanese Patent Application Laid-Open No. Hei 6-342098 discloses a photoconductive layer that generates electric charges in accordance with the intensity of irradiated radiation, and a technique for storing the generated electric charges in a plurality of two-dimensionally arranged capacitors. It has been disclosed.
[0021]
Further, as described in Japanese Patent Application Laid-Open No. 9-90048, radiation is absorbed by a phosphor layer such as an intensifying screen to generate fluorescent light, and the intensity of the fluorescent light is adjusted by a light source such as a photodiode provided for each pixel. The medical image may be input by detecting the image with the detector. As another means for detecting the fluorescence intensity, there is a method using a CCD or a C-MOS sensor. Further, a configuration in which a radiation scintillator that emits visible light by irradiation with radiation, a lens array, and an area sensor corresponding to each lens may be combined.
[0022]
When a digital medical image is obtained by the above-described various configurations, for example, for a mammogram (radiation image of the breast), the effective pixel size of the image is 200 μm or less, depending on the imaging site and the purpose of diagnosis. It is more preferable that the thickness be 100 μm or less. In order to maximize the performance of the image diagnosis support apparatus of the present invention, it is preferable to store and display image data input with an effective pixel size of about 50 μm, for example.
[0023]
Further, the image data input means 11 inputs the image data and information on the image data, for example, patient information on the patient such as the patient name, patient ID, gender, etc. of the imaged patient, the imaging site, the imaging direction, the imaging date, and the imaging. When inputting imaging information of a device or the like and outputting image data to the image data storage unit 12, the patient information and the imaging information are output in association with the image data.
The image data input means does not necessarily need to be provided. For example, the image data input means may read image data from various storage media such as a CD-ROM or a floppy (registered trademark) in which image data is stored, or may be connected via a network. Alternatively, the data may be transferred from an external device connected.
[0024]
The image data storage unit 12 stores the image data input by the image data input unit 11 after performing data compression as necessary. Here, as a data compression method, lossless compression or irreversible compression using a known method such as JPEG, DPCM, or wavelet compression is used, but lossless compression in which diagnostic information does not deteriorate due to data compression is preferable.
[0025]
The image data storage means 12 is composed of a magnetic or optical recording medium or a semiconductor memory. However, in a small-scale diagnosis, the amount of data input from the image data input means 11 is not so large. It can be stored on a magnetic disk without compression. In this case, image data can be stored and read at a much higher speed than a magneto-optical disk. At the time of image reading, a high-speed cycle time is required, so that necessary image data is also stored in a semiconductor memory.
[0026]
The abnormal shadow candidate detecting means 13 reads out the image data from the image data storage means 12 and performs image analysis to detect candidates considered as abnormal shadows such as microcalcification clusters and tumor shadows as shown in FIG. FIG. 2A shows an example of the microcalcification cluster. The presence of clustered microcalcifications is an important finding for early detection of breast cancer, as it is likely to be early stage cancer. On the mammogram, it is seen as a whitish and round shade with a generally conical structure. In addition, the tumor shadow shown in FIG. 2B is seen as a whitish round shadow close to a Gaussian distribution on a mammogram having a certain size.
[0027]
As described above, two major findings of breast cancer include a tumor shadow and a microcalcification cluster. As a tumor shadow detection method, a known detection method described in the following article can be used.
(1) tumor shadow
.Detection method by comparing left and right breasts
(Med. Phys., Vol. 21. No. 3, pp. 445-452)
・ Detection method using iris filter
(Theory of IEICE (D-11), Vol. J75-D-11, no. 3, pp. 663-670, 1992)
・ Detection method using a Quoit filter
(Theoretical theory (D-11), Vol. J76-D-11, no. 3, pp. 279-287, 1993)
.Detection method by binarization based on a histogram of pixel values of the divided breast cancer area
(JAMIT Frontier Lecture Papers, pp.84-85, 1995)
・ Minimum direction difference filter that takes the minimum output of many directional Laplacian filters)
(Theoretical theory (D-11), Vol. J76-D-11, no. 2, pp. 241-249, 1993)
[0028]
・ A method to distinguish benign or malignant tumor shadows using fractal dimension
(Medical Imaging Technology 17 (5), pp. 577-584, 1999).
[0029]
(2) Microcalcification cluster
・ A method to localize the area suspected of calcification from the breast area and remove false positive candidates from the optical density difference of the shadow image and the standard deviation value of the boundary density difference
(TEEE Transform Biomed Eng BME-26 (4): 213-219, 1979)
・ Detection method using Laplacian filtered image
(IEICE (D-11), Vol. J71-D-11, no. 10, pp. 1994-2001, 1988).
・ Detection method using morphologically analyzed image to suppress the influence of background pattern such as mammary gland
(IEICE (D-11), Vol. J71-D-11, no. 7, pp. 1170-1176, 1992)
[0030]
Here, the edge of the abnormal shadow will be described in detail with reference to FIG.
FIG. 3 is a schematic diagram showing a tumor shadow detected as an abnormal shadow. As shown in FIG. 3 (a), abnormal shadows of a tumor are classified into a circle, an ellipse, a polygon, a lobe, and an irregular shape. As shown in FIG. 3 (a), there is a characteristic that the border of the margin is clear, but in the case of a malignant tumor shadow, the border of the margin is unclear as shown in FIGS. 3 (b) and 3 (c). As shown in FIG. 3 (d), a low-concentration radial streak extending from the center of the mass called spicula to the periphery may occur, or as shown in FIG. There is a characteristic that a streak extends. Therefore, the information on the margin obtained from the detected abnormal shadow is an important index for distinguishing the benign or malignant abnormal shadow.
[0031]
The storage unit 14 stores, in an updatable file format, an abnormal shadow candidate list 141 that stores image data in which an abnormal shadow candidate has been detected by the abnormal shadow candidate detection unit 13 and information about the abnormal shadow candidate. As shown in an example of data storage in FIG. 4, the abnormal shadow candidate list 141 stores the patient ID of the imaged patient, the imaging region, the imaging direction, the information on the abnormal shadow candidate, and the imaging date in association with each other. The information such as the patient ID, the imaging region, and the imaging direction is obtained from the patient information and the imaging information added to the image data. Information candidates may be displayed in a dialog and appropriate information may be input by the operation input unit 18.
[0032]
The imaging part stores names of each part such as a breast, a chest, an abdomen, and the like. When the imaging part is a breast, for example, MLO (oblique direction), CC (vertical direction), ML (side direction) ) Is stored, and when the imaging part is the chest or abdomen, a symbol indicating the imaging direction such as P → A (from back to front) or LAT (side direction) is stored. You.
[0033]
Explaining the details of the abnormal shadow candidate in detail, the information on the position, size, shape, edge, and density value for specifying the abnormal shadow for each candidate, and the first classification result and the second classification result are stored. . The position information indicates the position of the center of gravity of the abnormal shadow candidate as a coordinate value (for example, (x, y) = (100, 1200) or the like) as shown in FIG. 3, but indicates, for example, an image region of the abnormal shadow candidate. It may be a coordinate value. Also, it may be indicated by a distance to a characteristic normal tissue.For example, when the imaging region is the chest, the position is determined by the distance between the center of gravity of the abnormal shadow candidate and the center of gravity of the lung field region that is a characteristic normal tissue. It may be indicated. The characteristic normal tissue is a biological tissue having a small change in position, such as an organ or bone such as the heart, lungs, and vertebrae, and is preferably a tissue that can serve as an index of a change over time in the position of an abnormal shadow candidate.
[0034]
The size information of the abnormal shadow candidate image area (for example, 225 mm) as shown in FIG. ² Etc.), but may be indicated by the average distance from the center of gravity of the abnormal shadow candidate to the edge or the longest distance (for example, 15 mm). In addition, the information on the shape of the abnormal shadow candidate is shown so that the shape (for example, a circle, a lobed shape, etc.) including the edge of the abnormal shadow candidate can be grasped. Exhibiting gender or its properties (eg, clear, unclear, spicular, etc.).
[0035]
As shown in FIG. 3, the information on the density value of the abnormal shadow candidate is represented by an average density value (for example, 300, 400, etc.) in the image area of the abnormal shadow candidate. It may be indicated by the contrast of the edge (difference in average density value of each part).
[0036]
As the first classification result, the classification result (for example, the classification result 11) when the abnormal shadow candidate is classified into the central part and the peripheral part by the first classification means 15 is stored. As the classification result, the classification result (for example, the classification result 21 or the like) when the abnormal shadow candidate is classified into a portion having a temporal change and a portion not having the change over time by the second classification means 16 is stored. It should be noted that the abnormal shadow candidate in which the classification result is not stored has no change with time and has not been classified by the second classification unit 16.
[0037]
The first classification unit 15 analyzes the image area of the abnormal shadow candidate detected by the abnormal shadow candidate detection unit, and classifies the image region into a central portion and its peripheral portion. The classification method will be described in detail with reference to FIG. As shown in FIG. 5A, first, the center of gravity of the detected abnormal shadow candidate is calculated, and a point at which the distance from the center of gravity to the end of the region of the abnormal shadow candidate is 50% is obtained. By tying, a boundary line between the center and the edge is formed. Then, the area from the center of gravity to the boundary is classified as the center, and the area from the boundary to the edge of the abnormal shadow candidate is classified as the edge. In the present embodiment, a description will be given assuming that a predetermined ratio from the center of gravity to the end of the region of the abnormal shadow candidate for determining the boundary line is 50%, but this ratio can be appropriately set.
[0038]
Further, the classification is not limited to the method described above. For example, the average distance from the center of gravity to the end of the abnormal shadow candidate area is calculated, a circle is formed around the center of gravity using the calculated average distance as the radius, and the inside area of the reduced circle obtained by reducing the circle at a rate of 50% is calculated. A method may be used in which a region from the center, outside the reduced circle to the end of the abnormal shadow candidate region is classified as a peripheral portion. Note that, here, the description will be made assuming that the ratio when forming the reduced circle is 50%, but this ratio can be set as appropriate.
[0039]
In the above description, the circle is formed. However, the present invention is not limited to this. For example, as shown in FIG. 5B, first, the distance from the center of gravity of the abnormal shadow candidate area to the end is the longest. Is set to the major axis and the shortest distance is set to the minor axis, and an ellipse is formed with the major axis set on the line where the major axis is set. Then, a reduced ellipse obtained by reducing the ellipse at a ratio of 40% is formed, and a region from the outside of the reduced ellipse to the end of the abnormal shadow candidate region is classified as a peripheral portion. It may be a method. Note that, here, the description will be given assuming that a predetermined ratio when forming the reduced ellipse is 40%, but this ratio can be set as appropriate.
[0040]
When the first classification unit 15 classifies the abnormal shadow candidate region into the center portion and the edge portion by the above-described classification method, the first classifying portion 15 outputs color information indicating the display color of each region of the center portion and the edge portion to the image data S. To be added. The color information is information indicating a display color when the abnormal shadow candidate is displayed on the display unit 17. That is, the first classifying unit 15 has a function as a display control unit described in the claims of the present invention. For example, color information indicating that the center of the abnormal shadow candidate area shown in FIG. 5A is displayed in red and the periphery is displayed in brown is added to the image data S. Although the display color is not particularly limited, it is preferable that the original image is normally displayed in monochrome, and is a color that attracts more attention than the original image. Further, the first classification unit outputs the result of the classification, specifically, coordinate values indicating the classified regions or information of the boundary line to be classified into each region to the storage unit 14 as the first classification result.
[0041]
The second classification unit 16 compares the current abnormal shadow candidate detected this time by the abnormal shadow candidate detection unit 13 with the abnormal shadow candidate detected in the past, and determines whether the abnormal shadow candidate region has a change with a region having a change. Color information indicating the display color of the changed area is added to the image data S. For example, color information indicating that an area having a change is displayed in red is added to the image data S. That is, the second classification unit 16 has a function as a display control unit described in the claims of the present invention. At this time, the color information specified by the first classifying unit and the second classifying unit is different from each other so that the classification by the first classifying unit and the class by the second classifying unit can be distinguished. Is preferred. Further, the second classification unit outputs the result of the classification, specifically, the coordinate values indicating the classified regions or the information of the boundary lines to be classified into the regions to the storage unit 14 as the second classification result.
[0042]
The first classifying unit 15 or the second classifying unit 16 performs various kinds of image processing on the image data while executing the above-described classification on the input image data S. Various types of image processing include frequency emphasis processing for adjusting image sharpness, gradation processing for adjusting contrast, and dynamic range for placing images with a wide dynamic range in a density range that is easy to view without reducing the contrast of the details of the subject. Compression processing and the like are included.
[0043]
As the display unit 17, a display unit such as a CRT (Cathode Ray Tube), a liquid crystal display, or a plasma display can be used. Among them, a high-definition CRT or a liquid crystal display dedicated to medical images is most preferable. Further, a high-definition display having a display pixel number of about 1000 × 1000 or more is preferable, and a high-definition display having a display pixel number of about 2000 × 2000 or more is preferable.
[0044]
In addition, for each of the displayed images, by performing the display position of the image, inverting the image, and rotating the image, the images can be compared and examined from various directions, and easily and easily using the medical image. Quick and accurate image diagnosis can be performed. Further, the display means 17 displays the input image data in only black and white gradations. If color information is added to the image data, the display unit 17 refers to the color information and displays the image whose display color is designated. Displays an area in its specified color.
[0045]
The display means 17 is integrally provided with an operation means 17a. The operation means 17a is constituted by a pressure-sensitive touch panel or the like in which transparent electrodes are arranged in a lattice. The operation unit 17a has an identification display button for instructing identification display of the abnormal shadow candidate area colored by the first classification unit or the second classification unit and identified and displayed by the display unit 17, and canceling the identification display. When the identification display button is turned on, the abnormal shadow candidate area is colored and identified and displayed. When the identification display button is turned off, the identification display based on the coloring of the abnormal shadow candidate area is canceled, and the original image of only black and white gradation is released. Is displayed.
[0046]
The operation unit 17a has a switching button for switching between the identification display of the area classified by the first classification unit and the identification display classified by the second classification unit. With this switching button, the classification result by the first classification unit and the classification result by the second classification unit can be displayed in order, displayed simultaneously, or both identification displays can be canceled. At the same time, when both the classification results are identified and displayed, the color is changed in each area, and an explanation indicating which color area is which area is displayed together with the image.
In the present embodiment, the operation unit 17a is described as a touch panel integrated with the image display unit 17, but is not limited to this, and may be a pointing device such as a mouse or a keyboard. .
[0047]
Next, the operation of the present embodiment will be described.
FIG. 6 is a flowchart illustrating the margin display processing by the image diagnosis support apparatus 10 according to the present embodiment.
As a premise of the explanation, a user (for example, a radiologist) who operates the image diagnosis support apparatus 10 selects a margin display mode for discriminating and displaying the center and the margin of the abnormal shadow by the operation unit 17a, and performs the margin display processing. Has been executed.
[0048]
In FIG. 6, first, when image data S obtained by imaging a patient as a subject is input to the image diagnosis support apparatus 10 by the image information input unit 11, the image information input unit 11 converts the input image data into patient information and imaging information. Are stored in the image data storage unit 12 in association with each other. The abnormal shadow candidate detection unit 13 reads out the image data S stored in the image data storage unit 12, performs image analysis, and detects an abnormal shadow candidate (step S1). At this time, the abnormal shadow candidate detecting unit 13 outputs information on the detected abnormal shadow candidate to the storage unit 14. The storage unit 14 stores the output information of the abnormal shadow candidate in the abnormal shadow candidate list 141 in association with the patient ID, the imaging region, the imaging direction, and the imaging date.
[0049]
Next, the image data S in which the abnormal shadow candidate has been detected is output to the first classifying unit 15, and the first classifying unit 15 analyzes the image area of the abnormal shadow candidate in the input image data S and performs center analysis. (Step S2), the color information specifying the display color of each classified area is added to the image data S, and the classified result is stored as a first classification result in the storage unit 14. Output to
[0050]
Next, the first classifying means outputs the image data to which the color information is added to the display means 17, and the display means 17 displays the input image data S based on the color information, and FIG. Alternatively, as shown in FIG. 5 (b), the image data S in which the regions of the center and the edge of the abnormal shadow candidate are identified and displayed is displayed (step S3), and the edge display processing ends.
[0051]
Next, with reference to FIG. 7, a description will be given of a temporal change display process performed by the image diagnosis support apparatus 10 of the present embodiment.
As a premise of the explanation, it is assumed that a user (for example, a radiologist) operating the image diagnosis support apparatus 10 has selected a temporal change display mode for displaying temporal change information in detail by the operating means 17a, and that the temporal change display processing has been executed. I do.
[0052]
In FIG. 7, first, when image data S obtained by imaging a patient as a subject is input to the image diagnosis support apparatus 10 by the image information input unit 11, the image information input unit 11 converts the patient information and the imaging information of the input image. The image data is stored in the image data storage unit 12 in association with the image. Next, the abnormal shadow candidate detection unit 13 reads out the image data S stored in the image data storage unit 12, performs image analysis, and detects an abnormal shadow candidate (step S101). At this time, the abnormal shadow candidate detecting unit 13 outputs information on the detected abnormal shadow candidate to the storage unit 14. The storage unit 14 stores the abnormal shadow candidate information in the abnormal shadow candidate list 141 in association with the patient ID, the imaging region, the imaging direction, and the imaging date.
[0053]
Next, the image data S in which the abnormal shadow candidate has been detected is output to the second classifying unit 16, and when the image data S is input, the second classifying unit 16 receives the abnormal shadow stored in the storage unit 14. With reference to the candidate list 141, it is determined whether or not an image in which an abnormal shadow candidate has been detected in the past in the same imaging region of the same patient is stored (step S102).
[0054]
If the image of the abnormal shadow candidate detected in the past in the same part of the same patient is not stored, the second classifying means 16 outputs the image data S in which the abnormal shadow candidate is detected to the display means 17, and the display means In step S108, the input image data S is displayed and the abnormal shadow candidate area in the image data S is indicated by a marker such as an arrow (step S108).
[0055]
On the other hand, when the image of the abnormal shadow candidate detected in the past in the same part of the same patient is stored, the second classifying means 16 stores the current abnormal shadow candidate area detected this time in the image data S and the storing means 14. Is compared with the past abnormal shadow candidate area in the past image data stored in (step S103), and it is determined whether or not the same abnormal shadow is photographed (step S104). At the time of matching, the position information of the abnormal shadow candidate in the abnormal shadow candidate list 141 is referred to, and the center of gravity of each abnormal shadow candidate is matched to perform the matching. If the same abnormal shadow has not been photographed, the process proceeds to step S108 to output the image data S in which the abnormal shadow candidate has been detected this time to the display unit 17, and the display unit 17 displays the input image data S At the same time, the abnormal shadow candidate area in the image data S is indicated by a marker such as an arrow (step S108).
[0056]
On the other hand, if the same abnormal shadow is photographed, the second classifying means 16 compares the current abnormal shadow candidate detected this time with the abnormal shadow candidate detected in the past, and compares their sizes (steps). S105), it is determined whether or not there is a change (step S106). Specifically, the size information of the abnormal shadow candidate list 141 is read out, and the area of the abnormal shadow candidate stored in the past is compared with the area of the currently detected abnormal shadow candidate. If the value is exceeded, it is determined that the size has changed.
[0057]
In the present embodiment, the presence or absence of a change in the abnormal shadow candidate is determined based on its size. However, for example, the shape, the edge, or the density value of the abnormal shadow candidate may be compared for determination. For example, match the center of gravity of the saved abnormal shadow candidate with the position of the center of gravity of the detected abnormal shadow candidate, match the image area of the abnormal shadow candidate, and if the shape or margin has changed, there is a change over time. The determination may be made. When comparing the density values, for example, the density value of the past abnormal shadow candidate is compared with the density value of the detected abnormal shadow candidate, and when the difference between the density values exceeds a predetermined value, the change with time is observed. It is determined that there is. At this time, the radiation dose varies between the past and present images depending on the photographing conditions at the time of photographing. Specifically, in the past and present abnormal shadow candidate images, the difference between the density values of the radiation-free portion (the portion where the radiation has reached without passing through the subject) is calculated, and the difference between the calculated density values is calculated. Is added to the density value of the image with the lower density value to make a correction and match. Alternatively, at the time of imaging, a substance whose radiation transmission amount is known in advance may be photographed together with the subject, and the same correction may be performed based on the density value of the substance to perform matching.
[0058]
If there is no change in the size in step S106, the process proceeds to step S108, where the second classifying means outputs the image data S in which the abnormal shadow candidate is detected this time to the display means 17, and the display means 17 The displayed image data S is displayed, and an abnormal shadow candidate area in the image data S is indicated by a marker (step S108).
[0059]
On the other hand, if there is a change in the size, the second classifying unit 16 refers to the abnormal shadow candidate list 141, and refers to the centroid position of the abnormal shadow candidate saved in the past and the centroid of the currently detected abnormal shadow candidate. The image areas of the abnormal shadow candidates are collated by matching the positions, and classified into overlapping areas (areas with no temporal change) and non-overlapping areas (areas with temporal change) (step S107). The color information specifying the display color is added to the image data S, and the classified result is output to the storage unit 14 as a second classified result. Then, the image data S to which the color information has been added is output to the display means 17, and the display means 17 displays the input image data S based on the color information, and as shown in FIG. An area having a temporal change in size between the current time and the present time is identified and displayed (step S108), and the temporal change display processing ends.
[0060]
Specifically, the identification display indicating the temporal change of the abnormal shadow candidate described above will be described with reference to FIG. 8. For example, the past image including the abnormal shadow candidate area stored in the storage unit 14 is shown in FIG. If the current image including the abnormal shadow candidate area detected this time is shown in FIG. 8B, the past image and the current image are collated, and the area where the abnormal shadow candidate area has changed is displayed in red, for example. Is identified and displayed.
[0061]
As described above, when the edge display mode for identifying and displaying the detailed information of the edge is selected, the detected abnormal shadow candidate area is divided into the center and the edge by the first classifying means. Since the areas of the respective parts are classified and displayed in different colors, it is possible to provide the information on the margin which is an important index for discriminating the benign and malignant of abnormal shadows, to the user, a radiologist. Thus, the efficiency of diagnosis can be improved.
[0062]
When the temporal change display mode for identifying and displaying the temporal change is selected, the current abnormal shadow candidate area detected this time is compared with the past abnormal shadow candidate area detected last time to reduce the size. If there is a change in the size, the second classifying means classifies the area into a changed area and a non-changed area, and colors and displays the changed area, so that there is a change with time. This can be notified to the user, a radiologist, and the efficiency of diagnosis can be improved.
[0063]
Further, since the operation means 17a has an identification display button for canceling or displaying the identification display by color, after the user recognizes that there is information on the periphery or change with time by the identification display, the identification display button is used. By canceling the identification display, the original image can be read, and a display operation can be performed as desired by the user. Therefore, operability can be improved.
[0064]
【The invention's effect】
According to the first, second and third aspects of the present invention, an abnormal shadow candidate is detected from the image information, and the image area of the abnormal shadow candidate is classified into a center area and an edge area, and the image information is displayed. When performing the classification, each image area is classified and displayed with a different color, so that not only the abnormal shadows are pointed out, but also information on the marginal area, which is an important index for distinguishing benign and malignant abnormal shadows. Can be provided to the user, and the efficiency of diagnosis can be improved and the accuracy of diagnosis can be improved.
[0065]
According to the fourth and fifth aspects of the present invention, since the image information of the detected abnormal shadow candidate and the information of the abnormal shadow candidate are stored, the past image information of the subject or the information of the past abnormal shadow candidate is read. As a result, changes over time can be observed, and diagnostic accuracy can be further improved.
[0066]
According to the sixth and seventh aspects of the present invention, the current image is compared with the stored past image, and the image area having a change in the abnormal shadow candidate image area is colored and identified and displayed. It is possible to clearly recognize the temporal change of the shadow, and to improve the efficiency of the diagnosis and the diagnostic accuracy.
[0067]
According to the eighth aspect of the present invention, the cancellation and display of the identification display of each image region can be instructed by the operating means. Therefore, when observing an image, the cancellation of the identification display is instructed, and the identification display is performed. When it is desired to obtain more detailed information on the abnormal shadow, an identification display can be instructed, and a display operation can be performed as desired by the user.
[Brief description of the drawings]
FIG. 1 is a diagram showing a functional configuration of an image diagnosis support apparatus 10 according to an embodiment to which the present invention is applied.
2A is a diagram showing an image of a tumor shadow, and FIG. 2B is a diagram showing an image of a microcalcification cluster.
3A is a schematic diagram showing classification of benign tumor shadows, and FIGS. 3B, 3C, 3D, and 3E are schematic diagrams showing classification of malignant tumor shadows. FIG.
FIG. 4 is a diagram showing an example of data storage of an abnormal shadow candidate list 141 stored in a storage unit 14 of FIG. 1;
5A is a diagram illustrating an example of a method of classifying an abnormal shadow candidate area by a first classifying unit, and FIG. 5B is an example of another classifying method of an abnormal shadow candidate area by a first classifying unit; FIG.
FIG. 6 is a flowchart illustrating a margin display process performed by the image diagnosis support apparatus 10.
FIG. 7 is a flowchart illustrating a temporal change display process performed by the image diagnosis support apparatus 10.
8A is a diagram of a past image in which an abnormal shadow candidate region detected in the past is identified and displayed, and FIG. 8B is a diagram of a current image in which an abnormal shadow candidate region detected in this time is identified and displayed. (C) is a diagram showing an image in which a past image and a current image are collated and a portion having a temporal change is identified and displayed.
[Explanation of symbols]
10 Image diagnosis support device
11 Image data input means
12. Image data storage means
13 Abnormal shadow candidate detection means
14 Storage means
15 First Classification Means
16 Second Classification Means
17 Display means
17a Operation means

Claims (8)

In an image diagnosis support apparatus including a display unit that displays image information of an image of a subject,
Abnormal shadow candidate detecting means for detecting an abnormal shadow candidate from the image information,
First classification means for classifying the image area of the abnormal shadow candidate detected by the abnormal shadow candidate detection means into a plurality of image areas;
Display control means for identifying and displaying each image area classified by the first classification means when the image information is displayed by the display means;
An image diagnosis support device comprising: 2. The image diagnosis support apparatus according to claim 1, wherein the display control unit applies a different color to each of the image areas classified by the first classification unit and causes the image areas to be identified and displayed. The apparatus according to claim 1, wherein the first classifying unit classifies the image area of the abnormal shadow candidate into an image area of a central part and an image area of a peripheral part thereof. The image diagnosis support apparatus according to claim 1, further comprising a storage unit configured to store image information in which an abnormal shadow candidate is detected by the abnormal shadow candidate detection unit. The apparatus according to claim 4, wherein the storage unit further stores information on the abnormal shadow candidate. A past image in which the same abnormal image candidate as the current image information in which the abnormal image candidate is detected by the abnormal image candidate detection means is the same image information as that of the same subject. When information is stored in the storage unit, the current image information and the past image information are collated to classify an abnormal shadow candidate image region into an image region having a change and an image region having no change. A second classification means,
The display control means, when displaying the image information in which the current abnormal shadow candidate is detected by the display means, identifies and displays each image area classified by the second classification means. The image diagnosis support device according to claim 4. 7. The image diagnosis support apparatus according to claim 6, wherein the display control unit adds a color to an image area having a change classified by the second classification unit and causes the image area to be identified and displayed. The image diagnosis support apparatus according to any one of claims 1 to 7, further comprising an operation unit that performs an operation instruction to perform identification display of each image region and release of the identification display by the display control unit.

Images