JP2003000575A - Image diagnosis supporting system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image diagnosis supporting system, with which accuracy in diagnosis support is improved by deciding whether the same disease is benignant or malignant from a plurality of diagnostic methods. SOLUTION: This system is provided with a CAD 4 for MMG image for detecting the presence/absence of fine calcifying and tumor on the basis of MMG image data 2 and generating a stochastic value for the detected fine calcified part or tumor to be malignant on the basis of the feature data of the detected calcified part and tumor, a CAD 5 for US image for detecting the presence/absence of a tumor on the basis of US image data 3 and generating a stochastic value for the detected tumor to be malignant on the basis of the feature data of the detected tumor, and a total CAD 6 for calculating a total stochastic value yout by weighing stochastic values y1 and y2 with a weighing coefficient m1 for MMG and a weighing coefficient m2 for US corresponding to the combination of the contents of a signal showing the presence/absence of fine calcifying and tumor from the CAD 4 for MMG image and a signal showing the presence/absence of the tumor from the CAD 5 for US image.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image diagnosis support system for assisting a doctor's diagnosis by automatically processing a computer to process an X-ray image or an ultrasonic image to determine whether the examination site is benign or malignant. is there.

[0002]

2. Description of the Related Art For example, the morbidity rate of breast cancer tends to increase year by year. Conventionally, diagnosis of breast cancer has been performed mainly by palpation, but this method is difficult to detect early because it is detected when a lump is formed. Therefore, in recent years, computer-aided diagnosis (Co
mputer-Aided Diagnosis (hereinafter referred to as CAD) system is being researched and developed.

In the field of mammography, mammography (Mmogram) by mammogram is used.
CAD of MG image or ultrasound (US: Ultra Son)
ic) Research and development such as CAD of images is being conducted. In the CAD of the mammography (MMG) image and the CAD of the ultrasonic (US) image, a computer automatically detects a lesion and determines whether a tumor is benign or malignant based on each image.

[0004]

By the way, the CAD of a mammography (MMG) image and the CAD of an ultrasonic (US) image have the following advantages and disadvantages.

That is, CAD of a mammography (MMG) image is suitable for a breast having a large amount of fat (strong), has an advantage that a detection rate of microcalcification is high, and is not suitable for a breast having a large number of mammary glands (weak). ), The detection rate of the tumor is low.

On the other hand, CA of ultrasonic (US) images
D has the advantage that it is suitable for breasts with many mammary glands (strong), has a high detection rate for tumors, is unsuitable for breasts with a lot of fat (weak), and has a low detection rate for microcalcifications. Have.

As described above, the CAD of the mammography (MMG) image and the CAD of the ultrasonic (US) image have opposite advantages and disadvantages with respect to the same lesion. Therefore, in the conventional CAD in the area of breast diagnosis, a highly reliable judgment result may be obtained in some cases, but a reliable judgment result cannot always be obtained depending on the lesion or the characteristics of the breast of the subject. Is the current situation, and CA with high accuracy of judgment (diagnosis) by CAD of a single modality
Development of D is desired.

The present invention has been made in view of the above circumstances, and an object thereof is to perform image diagnosis with high accuracy in diagnosis support with which diagnosis of benignity or malignancy can be made for a same lesion from a plurality of diagnostic methods. To provide a support system.

[0009]

In order to achieve the above object, a first aspect of the present invention is to process a plurality of image data of a region to be inspected of a subject imaged by different image pickup devices, respectively. An image diagnosis support system for generating judgment data of whether a part is good or bad, wherein presence / absence of a lesion is detected based on image data of a corresponding imaging device, characteristic data of the detected lesion is generated, and the generated characteristic Based on the data, a plurality of captured image diagnosis support units that respectively generate a probability value of a detected lesion being malignant, and a plurality of malignant probability values by the plurality of captured image diagnosis support units The total probability value is calculated by weighting with a weighting coefficient according to the combination of the contents of the signal indicating the presence or absence of a lesion by the image diagnosis support unit, and is output as the final malignancy probability value. And a diagnostic support unit.

Further, according to the first aspect, the weighting coefficient is set in advance in a look-up table according to a combination of presence / absence of a lesion, and the comprehensive diagnosis support unit determines the content of a signal indicating the presence / absence of a lesion. The weighting coefficient corresponding to the combination is read from the look-up table to perform the weighting calculation.

A second aspect of the present invention is an image for processing a plurality of pieces of image data of a region to be examined of a subject imaged by different image pickup devices to generate benign / malignant determination data of the region to be examined. A diagnostic support system, which detects the presence or absence of a lesion based on image data obtained by a corresponding imaging device,
Generates characteristic data of the detected lesion, and inputs the characteristic data of the subject and a plurality of diagnostic support units for captured images that generate probability values of malignancy of the detected lesion based on the generated characteristic data Input unit and a plurality of probability values that are malignant by the plurality of captured image diagnosis support units are respectively weighted by weighting coefficients according to the combination of the contents of the feature data by the input unit, and the total probability value is calculated. And a comprehensive diagnosis support unit that outputs a final probability value of malignancy.

Further, according to a second aspect, the weighting coefficient is set in advance in a look-up table according to the content of the characteristic data, and the comprehensive diagnosis support unit according to the content of the input characteristic data. The weighting coefficient is read from the look-up table to perform weighting calculation.

Further, a third aspect of the present invention is an image for processing a plurality of image data of a site to be examined of a subject imaged by different image pickup devices to generate benign / malignant judgment data of the site to be examined. A diagnostic support system, which detects the presence or absence of a lesion based on image data obtained by a corresponding imaging device,
Generates characteristic data of the detected lesion, and inputs the characteristic data of the subject and a plurality of diagnostic support units for captured images that generate probability values of malignancy of the detected lesion based on the generated characteristic data Input unit and a plurality of probability values that are malignant by the plurality of captured image diagnosis support units, the content of the signal indicating the presence or absence of a lesion by each captured image diagnosis support unit, and the content of the feature data by the input unit And a total diagnosis support unit that calculates a total probability value by weighting each with a weighting coefficient corresponding to the combination and outputs the final probability value as a probability value of malignancy.

According to a third aspect, the weighting coefficient is preset in a look-up table in accordance with the presence / absence of a lesion and the contents of the characteristic data, and the comprehensive diagnosis support unit determines the presence / absence of a lesion. The weighting coefficient corresponding to the combination of the content of the signal shown and the content of the characteristic data is read from the look-up table to perform the weighting calculation.

An image diagnosis support system according to a fourth aspect of the present invention detects the presence or absence of microcalcifications and tumors based on mammography image data, and detects the feature data of the detected microcalcifications and the detection. Generating the characteristic data of the generated tumor, the detected microcalcification portion based on the generated characteristic data, the diagnosis support unit for mammography image that generates the first probability value that the tumor is malignant, and the ultrasonic image The presence / absence of a tumor is detected based on the data, and based on the characteristic data of the detected tumor, an ultrasonic image diagnostic support unit that generates a second probability value that the detected tumor is malignant, and a mammography image The first probability value of being malignant and the second probability value of the image of being malignant by the ultrasonic image are respectively used as a signal indicating presence / absence of microcalcification by the diagnosis support unit for mammography images. , A signal indicating the presence / absence of a tumor, and a weighting factor for a mammography image and a second weighting factor for an ultrasound image according to the combination of the contents of the signal indicating the presence / absence of a tumor by the ultrasound image diagnostic support unit. Then, it has a comprehensive diagnosis support unit that calculates a comprehensive probability value and outputs it as a final probability value of malignancy.

According to a fourth aspect of the present invention, the first weighting coefficient for mammography image and the second weighting coefficient for ultrasonic image are the presence / absence of microcalcification by a diagnosis support unit for mammography image and tumor mass. A lookup table is set in advance according to a combination of presence / absence and presence / absence of a tumor by the ultrasound image diagnosis support unit, and the comprehensive diagnosis support unit indicates presence / absence of microcalcification by the mammography image diagnosis support unit. A mammography image first weighting coefficient and an ultrasound image second weighting coefficient according to a combination of the signal, the signal indicating the presence or absence of a tumor, and the content of the signal indicating the presence or absence of a tumor by the ultrasound image diagnostic support unit. The weighting calculation is performed by reading from the look-up table.

An image diagnosis support system according to a fifth aspect of the present invention detects the presence or absence of microcalcification and a tumor based on mammography image data, and detects the feature data of the detected microcalcification part and the detection. Generating the characteristic data of the generated tumor, the detected microcalcification portion based on the generated characteristic data, the diagnosis support unit for mammography image that generates the first probability value that the tumor is malignant, and the ultrasonic image Based on the data, the presence / absence of a tumor is detected, and based on the characteristic data of the detected tumor, an ultrasonic imaging diagnostic support unit that generates a second probability value that the detected tumor is malignant, and The input unit for inputting the characteristic data, the first probability value of being malignant by the mammography image and the second probability value of the image of being malignant by the ultrasonic image are Comprehensive diagnostic support unit that calculates a total probability value by weighting with a first weighting coefficient for mammography image and a second weighting coefficient for ultrasonic image according to the combination of Have and.

According to a fifth aspect of the present invention, the first weighting coefficient for mammography image and the second weighting coefficient for ultrasonic image are set in advance in a look-up table according to the contents of the characteristic data. And
The comprehensive diagnosis support unit reads the first weighting coefficient for mammography image and the second weighting coefficient for ultrasonic image according to the content of the input feature data from the lookup table and performs weighting calculation.

An image diagnosis support system according to a sixth aspect of the present invention detects the presence or absence of microcalcification and a tumor based on mammography image data, and detects the feature data of the detected microcalcification part and the detection. Generating the characteristic data of the generated tumor, the detected microcalcification portion based on the generated characteristic data, the diagnosis support unit for mammography image that generates the first probability value that the tumor is malignant, and the ultrasonic image Based on the data, the presence / absence of a tumor is detected, and based on the characteristic data of the detected tumor, an ultrasonic imaging diagnostic support unit that generates a second probability value that the detected tumor is malignant, and An input unit for inputting characteristic data, a first probability value of a malignant mammographic image and a second probability value of an malignant image of an ultrasonic image are respectively used as a mammographic image diagnostic image. Depending on the combination of the signal indicating the presence or absence of microcalcification by the support unit, the signal indicating the presence or absence of a tumor, the content of the signal indicating the presence or absence of a tumor by the ultrasound image diagnosis support unit, and the content of the feature data by the input unit It has a comprehensive diagnosis support unit that calculates a total probability value by weighting it with a first weighting coefficient for mammography image and a second weighting coefficient for ultrasound image, and outputs it as a final probability value of malignancy.

According to a sixth aspect of the present invention, the first weighting coefficient for mammography image and the second weighting coefficient for ultrasonic image are the presence / absence of microcalcification by the diagnosis support unit for mammography image and the tumor mass. Presence / absence, presence / absence of a tumor by the ultrasound image diagnosis support unit, and a lookup table set in advance according to a combination of the contents of the characteristic data, and the comprehensive diagnosis support unit is a mammography image diagnosis support unit. The signal indicating the presence or absence of microcalcification, the signal indicating the presence or absence of a tumor, the content of the signal indicating the presence or absence of a tumor by the ultrasound image diagnostic support unit, and the first mammography image for the combination of the contents of the characteristic data described above. The weighting coefficient and the second weighting coefficient for ultrasonic images are read from the look-up table and weighted Do.

According to the present invention, for example, different regions of the subject are imaged by different image pickup devices, and the plurality of image data of the imaged regions of the subject are respectively associated with the corresponding diagnostic assistance unit for picked-up images. Is supplied to. In each of the captured image diagnosis support units, the presence or absence of a lesion is detected based on the image data of the corresponding image capturing apparatus, and the feature data of the detected lesion is generated. Then, in each of the captured image diagnosis support units, a probability value of malignancy of the detected lesion is generated based on the generated feature data and is supplied to the comprehensive diagnosis support unit. Further, for example, the feature data of the subject is input from the input unit to the comprehensive diagnosis support unit. In the comprehensive diagnosis support unit, the plurality of probability values that are malignant by the plurality of captured image diagnosis support units indicate the content of the signal indicating the presence or absence of a lesion by each captured image diagnosis support unit and the content of the feature data by the input unit. The total probability value is calculated by weighting each with a weighting coefficient according to the combination. The comprehensive diagnosis support unit outputs the calculated comprehensive probability value as a final probability value of malignancy.

Further, according to the present invention, mammography image data obtained by, for example, a mammography device is supplied to a mammography image diagnosis support unit. The mammography image diagnosis support unit detects the presence or absence of microcalcifications and tumors based on the mammography image data, and generates the feature data of the detected microcalcifications and the feature data of the detected tumors. Then, the mammography image diagnosis support unit generates a first probability value that the detected microcalcification part and tumor are malignant based on the generated feature data, and outputs the first probability value to the comprehensive diagnosis support unit. Further, for example, ultrasonic image data obtained by the ultrasonic imaging apparatus is supplied to the ultrasonic image diagnostic support unit. The ultrasound image diagnosis support unit detects the presence or absence of a tumor based on the ultrasound image data, and generates a second probability value that the detected tumor is malignant, based on the feature data of the detected tumor. Is output to the comprehensive diagnosis support unit. Further, for example, the feature data of the subject is input from the input unit to the comprehensive diagnosis support unit. In the comprehensive diagnosis support unit, the first probability value that is malignant by the mammography image and the second probability value of the image that is malignant by the ultrasonic image are signals indicating the presence or absence of microcalcification by the diagnosis support unit for mammography images, respectively. , A signal indicating the presence / absence of a tumor, the content of the signal indicating the presence / absence of a tumor by the ultrasound image diagnosis support unit, and the first weighting coefficient for the mammography image and the ultrasound image according to the combination of the content of the feature data by the input unit The total probability value is calculated by weighting with the second weighting coefficient.
The comprehensive diagnosis support unit outputs the calculated comprehensive probability value as a final probability value of malignancy.

[0023]

BEST MODE FOR CARRYING OUT THE INVENTION An image diagnosis support system according to an embodiment of the present invention will be described below with reference to the drawings.

First Embodiment FIG. 1 is a block diagram showing a first embodiment of a computer image diagnosis support system according to the present invention. Also, FIG.
FIG. 3 is a diagram for explaining the functions of the CAD for MMG image, the CAD for US image, and the comprehensive CAD in FIG. 1.

A computer image diagnosis support (CAD) system 1 according to this embodiment, as shown in FIG. 1, is a mammography (hereinafter, MMG) device 2, an ultrasonic wave (hereinafter, U).
S) Imaging device 3, MM as MMG image diagnosis support unit
CAD4 for G image, US as diagnostic support unit for US image
CAD5 for image and comprehensive C as a comprehensive diagnosis support unit
It has AD6.

The MMG apparatus 2 converts the X-ray photograph data of the region to be examined of the subject (patient) 10, for example, the breast, into digital M data.
It is converted into the MG image data S2 and output to the MMG image CAD 4 and the comprehensive CAD 6.

The US imaging device 3 transmits an ultrasonic wave to a region to be examined of a subject (patient) 10, for example, a breast, generates an image based on a harmonic echo due to a nonlinear effect in the subject, and generates the image. The resulting image is output to the US image CAD 5 and the comprehensive CAD 6 as digital US image data S3.

The MMG image CAD 4 receives the MMG image data S2 from the MMG device 2, detects the presence or absence of microcalcification as shown in FIG. 2, and outputs the result to the general CAD 6 as a signal S41. In parallel with this, the presence or absence of a tumor is detected, and the result is output as a signal S42 to the comprehensive CAD 6. Furthermore, CAD for MMG images
4 is based on the detected shape and agglomerate (cluster) characteristics and position information of the microcalcifications, and the detected abdominal shape and contrast value characteristics and position information,
A probability value that the detected microcalcification part or a tumor is malignant is generated, and the generated probability value y1 is output to the comprehensive CAD6.

FIG. 3 is a block diagram showing a specific example of the structure of the MMG image CAD 4 shown in FIG.

As shown in FIG. 3, the present MMG image CAD 4 includes a microcalcification detecting section 41, a tumor detecting section 42, a first characteristic detecting section 43, a second characteristic detecting section 44, and a probability value calculating section. Has 45.

The microcalcification detecting unit 41 receives the MMG image data S2 from the MMG device 2, detects the presence or absence of microcalcification, and outputs the result as a signal S41 to the comprehensive CAD6.
Output to. Then, when it is determined that there is microcalcification, the microcalcification detecting unit 41 extracts, for example, a candidate area and generates position information S411 in which the microcalcification exists and luminance data 412 in a predetermined area. The generated position information S411 and the brightness data 412 in the predetermined area are output to the first feature detection unit 43, and the position information S411 is output to the probability value calculation unit 45.

The tumor mass detection unit 42 uses the MMG device 2
Upon receipt of the MG image data S2, the presence / absence of a tumor is detected, and the result is output to the comprehensive CAD 6 as a signal S42.
When it is determined that there is a tumor, the tumor detection unit 42 extracts, for example, a candidate region, the positional information S421 in which the tumor exists, and the brightness data 42 in the predetermined region.
2 is generated, the generated position information S421 and the brightness data 422 in the predetermined area are output to the second feature detection unit 44, and the position information S421 is calculated as the probability value calculation unit 45.
Output to.

The first feature detecting section 43 is provided with the position information S411 and the brightness data 412 by the microcalcification detecting section 41.
In response to this, the features such as the shape and agglomeration (cluster) of the microcalcification portion are detected and the first feature data S431 is output to the probability value calculation unit 45.

The second feature detector 44 is a tumor detector 42.
Upon receiving the position information S421 and the brightness data 422 according to, the features such as the shape of the tumor and the contrast value are detected, and the second feature data S441 is output to the probability value calculation unit 45. As the feature parameter of the tumor, specifically, the aspect ratio, the minimum width, the circularity, the average contrast inside the candidate, the average contrast at the candidate center, the standard deviation of the unsharp mask image at the candidate center, and the center of gravity are used. The intensity ratio by direction can be mentioned (for example, Jour.
nal of computer AidedDiag
nosis of Medical Images V
ol. 3 No. 2 Apr 1999). The aspect ratio is the ratio of the posterior aspect to the lateral length when the candidate is a tumor shadow. In addition, the average contrast inside the candidate becomes large around the inside of the candidate when the candidate is a tumor shadow. The average contrast of the candidate center is
When the candidate is a tumor shadow, the pixel value changes little at the center of the candidate and the contrast becomes small. In the intensity ratio by direction using the center of gravity, when the candidate is a tumor shadow, a strong position appears in a shape that draws a circle around the center of gravity.

The probability value calculation unit 45 includes a microcalcification detection unit 4
1 position information S411 by the first feature detection unit 43, the first feature data S431 by the first feature detection unit 43, the position information S421 by the tumor detection unit 42, and the second feature data S441 by the second feature detection unit 44, and detected. A probability value that the malignant microcalcification portion or tumor is malignant is generated, and the generated probability value y1 is output to the comprehensive CAD6. Probability value calculation unit 4
5 is, for example, position information S411, first characteristic data S
431, position information S421, and second characteristic data S
Substituting 441 into each predetermined discriminant set correspondingly, each disease is scored according to the result, and the score is added for each disease, and the disease with a high score is used as the discriminant result of the discriminant. , The probability value y1 is obtained.

The CAD for US image 5 receives the US image data S3 from the US imaging device 3, detects the presence or absence of a tumor, for example, as shown in FIG. 2, and outputs the result as a signal S51.
Is output to the comprehensive CAD 6. Further, the CAD 5 for US image generates a probability value that the detected tumor is malignant based on the characteristic data such as the shape and position of the detected tumor, and outputs the generated probability value y2 to the comprehensive CAD 6. .

FIG. 4 is a block diagram showing a specific example of the construction of the US image CAD 5 shown in FIG.

As shown in FIG. 4, the CAD 5 for US image has a tumor detection section 51, a feature detection section 52, and a probability value calculation section 53.

The tumor detection unit 51 receives the US image data S3 from the US imaging device 3, detects the presence or absence of a tumor, and outputs the result as a signal S51 to the comprehensive CAD 6.
When the tumor detection unit 51 determines that there is a tumor, the tumor detection unit 51 extracts, for example, a candidate region, the positional information S511 in which the tumor exists, and the brightness data 51 in the predetermined region.
2 is generated, and the generated position information S511 and the brightness data 512 in the predetermined area are output to the feature detection unit 52.

The feature detection unit 52 receives the position information S511 and the brightness data 512 from the tumor detection unit 51, detects the features such as the shape and position of the tumor, and detects the feature data S521.
Is output to the probability value calculation unit 53. The characteristic parameters of a tumor can be roughly classified into five types: shape, morphology, position, echo, and texture. For example, contour extraction processing is performed in order to obtain the characteristic parameters of the shape or morphology of the tumor.
For the features relating to the echo, for example, the density average value and variance at the same depth, inside, outside, posterior, posterior lateral shadow, and the same depth are obtained, and the difference or ratio thereof is used as one characteristic parameter. In addition, the texture has different internal compositions depending on the type of tumor. Texture feature parameters include uniformity, contrast, entropy,
Concentration correlation is used. To quantify the texture,
For example, the Fourier transform method, the simultaneous co-occurrence matrix method, the fractal method, etc. are used. The feature detection unit 52 determines that the position parameter, which is the feature parameter of the tumor, is the upper or lower end of the tumor,
The skin surface is detected based on the contour data of the tumor.
Specifically, as the position parameter, for example, the distance from the skin to the upper end of the mammary gland, the distance from the upper end of the mammary gland to the upper end of the tumor, and the distance from the lower end of the tumor to the pleural pleura are used.

The probability value calculation unit 53 receives the feature data S521 from the feature detection unit 52, generates a probability value that the detected tumor is malignant, and outputs the generated probability value y2 to the general CAD 6. The probability value calculation unit 53 substitutes, for example, the characteristic data S521 into a predetermined discriminant set correspondingly, scores each disease according to the result, and adds the score for each disease to obtain a score. The probability value y2 is obtained by using a disease with a high degree of discrimination as the discrimination result of the discriminant.

The comprehensive CAD 6 gives the probability value y1 of being malignant by the MMG image supplied from the CAD 4 for MMG image and the probability value y 2 of being malicious by the US image supplied from CAD 5 for US image, respectively, to the CAD for MMG image.
A signal S41 supplied from 4 indicating the presence or absence of microcalcification,
Signal S42 indicating presence / absence of tumor mass, and CAD for US image
5 are weighted by the MMG weighting coefficient m1 and the US weighting coefficient m2 which are set in advance in accordance with the combination of the contents of the signal S51 indicating the presence or absence of a tumor, and the total probability value shown by the following equation (1) Yout is calculated and output as the probability value of the final malignancy of this device.

[0043]

[Equation 1]         yout = m1 · y1 + m2 · y2 (1)

The weighting coefficient is stored in the memory as a look-up table 61 as shown in FIG. 5, for example, and the signal S41 supplied from the CAD 4 for MMG image indicates the presence or absence of microcalcification and the signal indicating the presence or absence of a tumor. S4
2, and is selectively read according to the combination of the contents of the signal S51 indicating the presence or absence of a tumor supplied from the CAD 5 for US image, and is substituted into the above equation (1).

As shown in FIG. 5, in the first embodiment, when the MMG image has microcalcification, the MMG image has a tumor, and the US image has a tumor, the weighting coefficient m1 for MMG is used. Is set to 0.5, and the US weighting coefficient m2 is also set to 0.5. When the MMG image has microcalcification, the MMG image has a tumor, and the US image has no tumor, the MMG weighting coefficient m1 is 0.9.
And the US weighting coefficient m2 is set to 0.1. If the MMG image has microcalcification, the MMG image does not have a mass, and the US image has a mass, then MMG
The weighting coefficient m1 for US is set to 0.4, and the weighting coefficient m2 for US is set to 0.6. When the MMG image has microcalcification, the MMG image has no tumor, and the US image has no tumor, the weighting coefficient m1 for MMG is 0.
9, and the US weighting coefficient m2 is set to 0.1. If the MMG image has no microcalcification, the MMG image has a mass, and the US image has a mass, then MM
The G weighting coefficient m1 is set to 0.3, and the US weighting coefficient m2 is set to 0.7. When the MMG image has no microcalcification, the MMG image has a tumor, and the US image has no tumor, the MMG weighting coefficient m1 is set to 0.4 and the US weighting coefficient m2 is set to 0.6. Has been done. There is no microcalcification in the MMG image,
If the G image has no tumor and the US image has a tumor,
The MMG weighting coefficient m1 is set to 0.0, and the US weighting coefficient m2 is set to 1.0. There is no microcalcification in the MMG image, no mass in the MMG image,
If there is no tumor in the image, MMG weighting coefficient m1
Is set to 0.5, and the US weighting coefficient m2 is also 0.5.
Is set to.

Next, the operation of the above configuration will be described with reference to the flowcharts of FIGS. 6 and 7. Here, first, an example will be described in which an MMG image is obtained from the subject and then a US image is obtained to obtain the total probability value yout. However, the present invention is not limited to this, and the US image is not limited to this. Needless to say, it is also possible to obtain the MMG image and then to obtain the total probability value yout.

First, the MMG apparatus 2 obtains X-ray photograph data of the breast, which is the region to be examined of the subject (patient).
This data is converted into digital MMG image data S2 (ST1), the MMG image CAD 4 and the total CA are obtained.
It is output to D6.

In the MMG image CAD 4, the presence or absence of microcalcification is detected by the microcalcification detecting unit 41 based on the MMG image data S2 (ST2), and the result is output to the comprehensive CAD 6 as a signal S41. And
If the microcalcification detecting unit 41 determines that there is microcalcification, the position information S411 indicating that the microcalcification exists.
And the brightness data 412 in the predetermined area are generated (ST3), the generated position information S411 and the brightness data 412 in the predetermined area are output to the first feature detection unit 43, and the position information S411 is output to the probability value calculation unit 45. It

In the first feature detecting section 43, the position information S411 and the brightness data 41 obtained by the microcalcification detecting section 41 are used.
Based on 2, the shape and agglomeration (cluster) of the microcalcification part
Is detected (ST4), and the first characteristic data S4 is detected.
31 is output to the probability value calculation unit 45.

In the MMG image CAD4,
The presence or absence of a tumor is detected by the tumor detection unit 42 based on the MMG image data S2 (ST5), and the result is signal S4.
2 is output to the comprehensive CAD 6. When the tumor detection unit 42 determines that there is a tumor, position information S421 in which the tumor exists and brightness data 422 in a predetermined area are generated (ST6), and the generated position information S421 and brightness data in the predetermined area are generated. 422 is output to the second feature detection unit 44, and the position information S421 is output to the probability value calculation unit 45.

In the second feature detecting section 44, the tumor detecting section 4
Based on the position information S421 and the brightness data 422 based on 2, the features such as the shape of the tumor and the contrast value are detected (ST7), and the second feature data S441 is output to the probability value calculation unit 45.

In the probability value calculation unit 45, the positional information S411 by the microcalcification detection unit 41, the first characteristic data S431 by the first characteristic detection unit 43, the positional information S421 by the tumor detection unit 42, and the second characteristic data. Based on the second feature data S441 by the detection unit 44, a probability value of malignancy of a microcalcification part or a tumor detected based on the second feature data S441 is generated (ST
8) The generated probability value y1 is output to the comprehensive CAD6.

Next, the US imaging device 3 transmits ultrasonic waves to the breast, which is the region to be examined of the subject (patient), and a US image is generated based on the harmonic echo due to the nonlinear effect in the subject. The generated image is output as digital US image data S3 to the US image CAD 5 and the integrated CAD 6 (ST9).

In the US image CAD 5, the presence / absence of a tumor is detected by the tumor detecting unit 51 based on the US image data S3 by the US imaging device 3, and the result is output to the comprehensive CAD 6 as a signal S51 (ST10). . When the tumor detection unit 51 determines that there is a tumor, position information S511 in which the tumor exists and luminance data 512 in a predetermined area are generated (ST11), and the generated position information S511 and luminance data in the predetermined area are generated. 512 is output to the feature detection unit 52.

The feature detection unit 52 detects features such as the shape and position of the tumor based on the position information S511 and the brightness data 512 from the tumor detection unit 51, and outputs the feature data S521 to the probability value calculation unit 53. .

The probability value calculation unit 53 generates a probability value that the tumor detected based on the feature data S521 by the feature detection unit 52 is malignant (ST12), and the generated probability value y2 is output to the general CAD 6. It

In the comprehensive CAD6, each MMG
Signal S41 indicating presence / absence of microcalcification supplied from CAD 4 for image, signal S42 indicating presence / absence of tumor, and US
Signal S indicating presence / absence of a tumor supplied from CAD 5 for image
According to the combination of the contents of 51, the MMG weighting coefficient m1 and the US weighting coefficient m2 which are preset in the lookup table 61 are selected (S
T13). In the comprehensive CAD6, the C for MMG image is used.
The probability value y1 of being malignant by the MMG image supplied from AD4 and the probability value y2 of being malignant by the US image supplied from CAD5 for the US image are respectively selected as the weighting coefficient m1 for MMG and the weighting coefficient m for US.
Weighted by 2, the total probability value yout is calculated and output as the final probability value that the device is malignant (ST14).

As described above, according to the first embodiment, the presence / absence of microcalcification is detected based on the MMG image data S2, the result is output as the signal S41, and the presence / absence of a tumor is detected. The detection result is output as a signal S42, and based on the detected shape and agglomerate (cluster) characteristics and position information of the microcalcifications, and the detected shape and contrast value characteristics and position information of the tumor. ,
Based on the MMG image CAD4 that generates a probability value that the detected microcalcifications and tumors are malignant, the presence or absence of tumors is detected based on the US image data S3, and the result is output as a signal S51, and detected. A CAD5 for US image that generates a probability value that a detected tumor is malignant based on feature data such as a shape and a position of the tumor, a probability value y1 that is malignant by the MMG image, and a probability value that is malignant by the US image. y2 is a signal S41 indicating the presence or absence of microcalcification by the CAD4 for MMG images, and a signal S indicating the presence or absence of a tumor.
42 and the weighting coefficient m1 for MMG and the weighting coefficient m2 for US preset according to the combination of the contents of the signal S51 indicating the presence or absence of a tumor by the CAD 5 for US image, and the total probability value yout
Is provided and a comprehensive CAD6 that outputs the probability value of the final malignancy of this device is provided. Therefore, it is possible to determine benign or malignant from a plurality of diagnostic methods with respect to the same lesion, and a single modality Compared with CAD, there is an advantage that an image diagnosis support system with high accuracy of diagnosis support can be realized.

Second Embodiment FIG. 8 is a block diagram showing the second embodiment of the computer image diagnosis support system according to the present invention.

The second embodiment is different from the above-mentioned first embodiment in that the presence or absence of microcalcification by the MMG image CAD 4 is set as a condition for setting and selecting the weighting factors m1 and m2. Presence / absence of tumor mass and C for US image
Rather than depending on the combination of the presence / absence of a tumor due to AD5, for example, the age and constitution of the subject (patient) input in the results of the interview, such as the amount of fat and the amount of mammary glands that are characteristic of the breast It is configured to set and select according to the combination of (patient) individual characteristic data.

Therefore, FIG. 8 showing the second embodiment of the present invention.
Then, the personal characteristic data input unit 7 is provided to input the personal characteristic data S7 into the comprehensive CAD 6A. Also, MMG
A signal indicating the presence or absence of microcalcification by the CAD4 for images, a signal indicating the presence or absence of a tumor, and a signal indicating the presence or absence of a tumor by the CAD5 for US images are not input to the general CAD6A.

FIG. 9 is a diagram showing an example of a look-up table in which weighting coefficients according to the second embodiment are set.

In this example, as shown in FIG. 9, the age is 5
For people over the age of 0, regardless of breast characteristics, MMG
The weighting coefficient m1 for the probability value y1 related to the image is 0.
8 and the weighting coefficient m2 for the probability value y2 related to the US image is set to 0.2. When the age is less than 50 years old and the breast has a large amount of fat, the weighting coefficient m1 for the probability value y1 for the MMG image is set to 0.9, and the probability value y2 for the US image is set.
Is set to 0.1.
When the age is less than 50 years and there are many mammary glands as a feature of the breast, the weighting coefficient m1 for the probability value y1 related to the MMG image is set to 0.2, and the weighting for the probability value y2 related to the US image is performed. The coefficient m2 is set to 0.8.

The other structure is the same as that of the above-described first embodiment, and the same effect as that of the above-described first embodiment can be obtained.

Third Embodiment FIG. 10 is a block diagram showing a third embodiment of the computer image diagnosis support system according to the present invention.

The second embodiment differs from the first embodiment described above in that the presence or absence of microcalcification by the MMG image CAD 4 is set as a condition for setting and selecting the weighting factors m1 and m2. Presence / absence of tumor mass and C for US image
Not only the combination of the contents of the presence or absence of a tumor by AD5, but also the subject (patient) such as the age and constitution of the subject (patient) input in the result of the interview, for example, the amount of fat that is a characteristic of the breast, the amount of mammary gland ) Is set and selected according to the combination including the individual characteristic data of).

Therefore, FIG. 1 showing the third embodiment.
In 0, unlike the second embodiment, the personal characteristic data input unit 7 is provided to input the personal characteristic data S7 into the comprehensive CAD 6B, and at the same time, a signal indicating the presence or absence of microcalcification by the MMG image CAD 4 and a tumor mass. A signal indicating the presence / absence and a signal indicating the presence / absence of a tumor by the CAD for US image 5 are also input to the general CAD 6B.

FIG. 11 is a diagram showing an example of a look-up table in which weighting coefficients are set according to the third embodiment.

In this example, as shown in FIG. 11, when the age is 50 years or older, the characteristics of the breast and the presence or absence of microcalcification by CAD4 for MMG images, the presence or absence of a tumor, and the CAD5 for US images are used. Irrespective of the presence / absence of a tumor due to the MMG image, the weighting coefficient m1 for the probability value y1 for the MMG image is set to 0.8, and the probability value y for the US image is
The weighting coefficient m2 for 2 is set to 0.2.

When the age is less than 50 years old and the breast has a large amount of fat, the following settings are made. When the MMG image has microcalcification, the MMG image has a tumor, and the US image has a tumor, the MMG weighting coefficient m1 is set to 0.6 and the US weighting coefficient m2 is set to 0.4. Has been done. When the MMG image has microcalcification, the MMG image has a tumor, and the US image has no tumor, the MMG weighting coefficient m1 is set to 0.9 and the US weighting coefficient m2 is set to 0.1. Has been done. When the MMG image has microcalcification, the MMG image has no tumor, and the US image has a tumor, the MMG weighting coefficient m1 is set to 0.5 and the US weighting coefficient m2 is also set to 0.5. Has been done. When the MMG image has microcalcification, the MMG image has no tumor, and the US image has no tumor, the MMG weighting coefficient m1 is set to 1.0 and the US weighting coefficient m2 is set to 0.0. Has been done. When the MMG image has no microcalcification, the MMG image has a tumor, and the US image has a tumor, the MMG weighting coefficient m1 is set to 0.4 and the US weighting coefficient m2 is set to 0.6. Has been done. When the MMG image has no microcalcification, the MMG image has a tumor, and the US image has no tumor, the weighting coefficient m1 for MMG is 0.5.
And the US weighting coefficient m2 is also set to 0.5. If there is no microcalcification in the MMG image, no mass in the MMG image, and a mass in the US image, then MMG
The weighting coefficient m1 for US is set to 0.3, and the weighting coefficient m2 for US is set to 0.7. When the MMG image has no microcalcification, the MMG image has no tumor mass, and the US image has no tumor mass, the weighting coefficient m1 for MMG is 0.7.
And the US weighting coefficient m2 is set to 0.3.

If the age is less than 50 and the breast has many mammary glands, the following settings are made. When the MMG image has microcalcification, the MMG image has a tumor, and the US image has a tumor, the MMG weighting coefficient m1 is set to 0.4 and the US weighting coefficient m2 is set to 0.6. Has been done. When the MMG image has microcalcification, the MMG image has a tumor, and the US image has no tumor, the MMG weighting coefficient m1 is set to 0.7 and the US weighting coefficient m2 is set to 0.3. Has been done. When the MMG image has microcalcification, the MMG image has no tumor, and the US image has a tumor, the MMG weighting coefficient m1 is set to 0.3 and the US weighting coefficient m2 is also set to 0.7. Has been done. When the MMG image has microcalcification, the MMG image has no tumor, and the US image has no tumor, the MMG weighting coefficient m1 is set to 0.7 and the US weighting coefficient m2 is set to 0.3. Has been done. When the MMG image has no microcalcification, the MMG image has a tumor, and the US image has a tumor, the MMG weighting coefficient m1 is set to 0.1 and the US weighting coefficient m2 is set to 0.9. Has been done. When the MMG image has no microcalcification, the MMG image has a tumor, and the US image does not have a tumor, the weighting coefficient m1 for MMG is 0.2.
And the US weighting coefficient m2 is set to 0.8. If there is no microcalcification in the MMG image, no mass in the MMG image, and a mass in the US image, then MMG
The weighting coefficient m1 for US is set to 0.0, and the weighting coefficient m2 for US is set to 1.0. When the MMG image has no microcalcification, the MMG image has no tumor mass, and the US image has no tumor mass, the weighting coefficient m1 for MMG is 0.3.
And the US weighting coefficient m2 is set to 0.7.

The other structure is the same as that of the first embodiment.

According to the third embodiment, the above-mentioned first embodiment
Not only can the same effects as those of the embodiment described above be obtained, but there is an advantage that the quality can be determined more accurately.

[0074]

As described above, according to the present invention,
There is an advantage that it is possible to determine benignity or malignancy with respect to the same lesion from a plurality of diagnostic methods, and to realize an image diagnosis support system with high accuracy in diagnosis support.

[Brief description of drawings]

FIG. 1 is a configuration diagram showing an embodiment of a computer image diagnosis support system according to the present invention.

FIG. 2 is a CAD for MMG image and a CA for US image of FIG.
It is a figure for demonstrating the function of D and general CAD.

FIG. 3 is a block diagram showing a specific configuration example of the MMG image CAD shown in FIG. 1.

FIG. 4 is a block diagram showing a specific configuration example of the CAD for US image of FIG. 1.

FIG. 5 is a diagram showing an example of a lookup table in which weighting coefficients are set according to the first embodiment.

FIG. 6 is a flowchart for explaining the operation of the first embodiment.

FIG. 7 is a flowchart for explaining the operation of the first embodiment.

FIG. 8 is a configuration diagram showing a second embodiment of a computer image diagnosis support system according to the present invention.

FIG. 9 is a diagram showing an example of a lookup table in which weighting coefficients according to the second embodiment are set.

FIG. 10 is a configuration diagram showing a third embodiment of a computer image diagnosis support system according to the present invention.

FIG. 11 is a diagram showing an example of a lookup table in which weighting coefficients are set according to the third embodiment.

[Explanation of symbols]

1, 1A, 1B ... Computer image diagnosis support (CAD)
System, 2 ... Mammography (MMG) device, 3 ... Ultrasound (US) imaging device, 4 ... Mammography (MMG image CAD, 41 ... Microcalcification detection unit, 42 ... Tumor detection unit, 43 ... First feature detection Part, 44 ... Second feature detecting part, 45 ... Probability value calculating part, 5 ... Ultrasonic (US) image C
AD, 51 ... Tumor detection section, 52 ... Feature detection section, 53 ... Probability value detection section, 6, 6A, 6B ... Comprehensive CAD.

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Takao Chiba             127, 4-7 Asahigaoka, Hino City, Tokyo             GE Yokogawa Medical System Co., Ltd.             Within F-term (reference) 4C093 AA01 AA26 CA31 DA06 FD12                       FF17 FF19 FF20                 4C301 DD24 EE11 EE14 JC20 KK31                       LL20                 5B057 AA08 BA03 BA05 CA08 CB08                       CC02 CH01 CH07 DA02 DA07                       DC05                 5L096 BA06 BA13 DA02 EA39 FA09                       FA69 GA53

Claims (12)

[Claims] 1. An image diagnosis support system, which processes a plurality of image data of a test region of a subject imaged by different imaging devices to generate benign / malignant determination data of the test region, Based on the image data by the imaging device, the presence or absence of a lesion is detected, characteristic data of the detected lesion is generated,
Based on the generated feature data, a plurality of captured image diagnosis support units that respectively generate probability values of malignancy of detected lesions, and a plurality of probability values of malignancy by the plurality of captured image diagnosis support units are set. , Comprehensive diagnosis that calculates the overall probability value by weighting with a weighting coefficient according to the combination of the contents of the signals indicating the presence or absence of lesions by each captured image diagnosis support unit, and outputs it as the final malignancy probability value An image diagnosis support system having a support unit. 2. The weighting coefficient is set in a look-up table in advance according to a combination of presence / absence of a lesion, and the comprehensive diagnosis support unit determines a weighting coefficient according to a combination of contents of signals indicating the presence / absence of a lesion. The image diagnosis support system according to claim 1, wherein is read out from the lookup table to perform weighting calculation. 3. An image diagnosis support system for processing a plurality of image data of a test region of a subject imaged by different imaging devices to generate benign / malignant determination data of the test region, which is provided. Based on the image data by the imaging device, the presence or absence of a lesion is detected, characteristic data of the detected lesion is generated,
Based on the generated feature data, a plurality of captured image diagnostic support units that respectively generate probability values of malignancy of the detected lesion, an input unit that inputs the feature data of the subject, and the plurality of captured image data A plurality of probability values that are malignant by the diagnosis support unit are respectively weighted by weighting coefficients according to the combination of the contents of the feature data by the input unit, and an overall probability value is calculated, and as a final probability value that is malignant. An image diagnosis support system having a comprehensive diagnosis support unit for outputting. 4. The weighting coefficient is set in advance in a look-up table according to the content of the characteristic data, and the comprehensive diagnosis support unit uses the weighting coefficient according to the content of the input characteristic data as the look-up table. The image diagnosis support system according to claim 3, wherein the weighting calculation is performed by reading out from the up table. 5. An image diagnosis support system for processing a plurality of image data of a test region of a subject imaged by different imaging devices to generate benign / malignant determination data of the test region, which is provided. Based on the image data by the imaging device, the presence or absence of a lesion is detected, characteristic data of the detected lesion is generated,
Based on the generated feature data, a plurality of captured image diagnostic support units that respectively generate probability values of malignancy of the detected lesion, an input unit that inputs the feature data of the subject, and the plurality of captured image data A plurality of probability values that are malignant by the diagnosis support unit are weighted by the weighting coefficient according to the combination of the content of the signal indicating the presence or absence of the lesion by each captured image diagnosis support unit and the content of the feature data by the input unit. And an overall diagnosis support unit that calculates a total probability value and outputs it as a final malignancy probability value. 6. The weighting coefficient is preset in a look-up table in accordance with the presence or absence of a lesion and the content of the characteristic data, and the comprehensive diagnosis support unit includes the content of a signal indicating the presence or absence of a lesion,
The image diagnosis support system according to claim 5, wherein a weighting coefficient corresponding to a combination of the contents of the characteristic data is read from the look-up table and weighted. 7. Detecting the presence or absence of microcalcifications and tumors based on mammography image data, generating feature data of the detected microcalcifications and feature data of the detected tumors, and adding the feature data to the generated feature data. Based on the detected microcalcifications, the mammography image diagnosis support unit that generates the first probability value that the tumor is malignant, and the presence or absence of the tumor based on the ultrasonic image data, and the detected tumor Based on the characteristic data of, the ultrasonic image diagnostic support unit that generates a second probability value that is malignant of the detected tumor, and the first probability value that is malignant by the mammography image and the malignancy by the ultrasonic image. The second probability value of an image
Each of the mammography image diagnosis signals corresponding to the combination of a signal indicating the presence or absence of microcalcification by the mammography image diagnosis support unit, a signal indicating the presence or absence of a tumor, and a signal indicating the presence or absence of a tumor by the ultrasound image diagnosis support unit. An image diagnosis support system comprising: a weighting coefficient of 1 and a second weighting coefficient for ultrasonic images to calculate a total probability value, and output the final probability value as malignancy. 8. The mammography image first weighting coefficient and the ultrasound image second weighting coefficient are the presence or absence of microcalcification by a mammography image diagnosis support unit,
The look-up table is set in advance according to the combination of the presence / absence of a tumor and the presence / absence of a tumor by the ultrasound image diagnosis support unit. The comprehensive diagnosis support unit determines whether microcalcification occurs by the mammography image diagnosis support unit. , A signal indicating the presence or absence of a tumor, and a first weighting coefficient for a mammography image and a second weighting for an ultrasound image according to the combination of the contents of the signal indicating the presence or absence of a tumor by the ultrasound image diagnostic support unit. The image diagnosis support system according to claim 7, wherein the coefficient is read from the look-up table and weighted. 9. The presence or absence of microcalcification and a tumor is detected based on the mammography image data, the feature data of the detected microcalcification portion and the feature data of the detected tumor are generated, and the generated feature data is added to the generated feature data. Based on the detected microcalcifications, the mammography image diagnosis support unit that generates the first probability value that the tumor is malignant, and the presence or absence of the tumor based on the ultrasonic image data, and the detected tumor Based on the characteristic data of the above, the ultrasonic image diagnostic support unit that generates the second probability value that the detected tumor is malignant, the input unit that inputs the characteristic data of the subject, and the malignancy by the mammography image The first probability value and the second probability value of the image that is malignant by the ultrasonic image are
The overall probability value is calculated by weighting with the first weighting coefficient for mammography image and the second weighting coefficient for ultrasonic image according to the combination of the contents of the characteristic data by the input unit, and the final probability of malignancy is calculated. An image diagnosis support system having a comprehensive diagnosis support unit for outputting as a value. 10. The first weighting coefficient for mammography image and the second weighting coefficient for ultrasonic image are preset in a lookup table according to the contents of the characteristic data, and the comprehensive diagnosis support unit 10. The image diagnosis support system according to claim 9, wherein the first weighting coefficient for mammography image and the second weighting coefficient for ultrasonic image according to the content of the input feature data are read from the look-up table to perform weighting calculation. . 11. Detecting the presence or absence of microcalcifications and tumors based on mammography image data, generating feature data of the detected microcalcifications and feature data of the detected tumors, and adding the generated feature data to the generated feature data. Based on the detected microcalcifications, the mammography imaging diagnosis support unit that generates the first probability value that the tumor is malignant, and the presence or absence of the tumor based on the ultrasound image data, and the detected tumor Based on the feature data of the above, the ultrasonic image diagnostic support unit that generates the second probability value that the detected tumor is malignant, the input unit that inputs the feature data of the subject, and the malignant mammographic image The first probability value and the second probability value of the image that is malignant by the ultrasonic image are
Signals indicating the presence or absence of microcalcification by the diagnostic support unit for mammography images, signals indicating the presence or absence of a tumor, the contents of the signal indicating the presence or absence of a tumor by the diagnostic support unit for ultrasound images, respectively.
And the weighting is performed by the first weighting coefficient for mammography image and the second weighting coefficient for ultrasonic image according to the combination of the contents of the feature data by the input unit, and the total probability value is calculated to determine the final malignancy. An image diagnosis support system having a comprehensive diagnosis support unit that outputs a probability value. 12. The first weighting coefficient for mammography image and the second weighting coefficient for ultrasonic image are the presence / absence of microcalcification, the presence / absence of a tumor, and the ultrasonic imaging diagnostic assistance by the mammography image diagnostic assistance unit. Whether or not there is a tumor by the part, and is preset in the lookup table according to the combination of the contents of the characteristic data, the comprehensive diagnosis support unit, a signal indicating the presence or absence of microcalcification by the diagnosis support unit for mammography images, A signal indicating the presence / absence of a tumor, the contents of the signal indicating the presence / absence of a tumor by the ultrasound image diagnosis support unit, and the first weighting coefficient for mammography images and the second for ultrasound images according to the combination of the contents of the characteristic data. 2. The weighting coefficient is read from the look-up table to perform the weighting calculation.
1. The image diagnosis support system described in 1.