JP2006325640A - Method of displaying abnormal shadow candidate and medical image processing system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide only detected information on abnormal shadow candidates which are suspicious whether the abnormal shadows are true-positive or not and/or on abnormal shadow candidates with low visibility to prevent oversights by a physician and to make reading work more efficient. <P>SOLUTION: In a medical image processing system 100, when a medical image is generated, an image processor 2 performs processing to detect abnormal shadow candidates on the medical image. Next, the abnormal shadow candidates which are obviously true-positive or false-positive by a multivariate analysis, candidates which are difficult to determine whether they are true-positive or false-positive, and candidates located in positions which are hard to detect, or the like are determined from the image characteristic amount of the detected abnormal shadow candidates and are eliminated from the detected abnormal shadow candidates. Then, only detected information on the abnormal shadow candidates after the deletion is displayed on a viewer 5. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an abnormal shadow candidate display method and a medical image processing system for detecting an abnormal shadow candidate from a medical image and displaying the detection information.

  In the medical field, a medical image obtained by imaging a patient with an imaging device such as CT (Computed Tomography) or MRI (Magnetic Resonance Imaging) is converted into digital data and displayed on a display when the doctor makes a diagnosis. I have come to do interpretation. In particular, in recent years, computer-aided diagnosis (hereinafter referred to as CAD) for detecting abnormal shadow candidates such as cancerous parts has been developed for the purpose of reducing the burden on interpreting physicians and reducing oversight of abnormal shadows.

When an abnormal shadow candidate is detected by the CAD, a method of displaying a marker (such as an arrow or a circle) indicating the detection position of the abnormal shadow candidate on the medical image as detection information is common. As the display method, (1) a method of displaying a marker directly on a medical image (see, for example, Patent Document 1), (2) a life size (equal magnification) medical image is displayed as a main image for interpretation, A method is used in which a sub-image in which a marker is displayed on a reduced image of the main image is generated and displayed together with the main image (see, for example, Patent Document 2).
Japanese Patent Laid-Open No. 2000-276587 JP 2004-230001 A

  In this situation, when the number of abnormal shadow candidates detected by CAD is large, in the method (1), many markers are displayed on the image observed by the interpreting doctor, and the medical image itself is displayed. It may be difficult to observe and may adversely affect the interpretation work.

  On the other hand, in the above method (2), since the marker is not displayed on the main image itself as compared with the method (1), the influence on the doctor's interpretation work is small, but there are still many on the reduced sub-image. Markers are displayed, and there is a problem that it is difficult to see. Further, it is difficult to correspond the positions of the main image and the sub image.

  Furthermore, when detecting an abnormal shadow candidate, the image feature amount of the candidate area is calculated and compared with a threshold value to determine whether it is an abnormal shadow, or by performing multivariate analysis using the image feature amount Various algorithms are used, such as a method for detecting a candidate area for an abnormal shadow, and the detection result clearly includes both an abnormal shadow and a suspicious one.

  In the conventional CAD, detection information regarding all of the abnormal shadow candidates detected in this way is provided to the doctor, and the final determination as to whether it is true positive or false positive is made by the doctor. However, as described above, when the number of detections is very large, the interpretation doctor must confirm all of them, which is inefficient. In particular, for experienced doctors, it may be more efficient to interpret suspicious candidates preferentially than those that can easily be determined to be apparently abnormal shadows.

  Also, if an abnormal shadow is present near the edge of the breast image, or if the area of the shadow itself is small, it is difficult for the interpreting physician to visually recognize and it is easy to overlook.

  An object of the present invention is to provide only detection information of an abnormal shadow candidate suspected of being a true positive abnormal shadow and / or an abnormal shadow candidate with low visibility, thereby preventing oversight of a doctor and improving the efficiency of interpretation work. It is.

The invention according to claim 1 is a display method of abnormal shadow candidates.
A detection step of analyzing a medical image and detecting an abnormal shadow candidate;
Of the detected abnormal shadow candidates, an extraction step of extracting abnormal shadow candidates to be displayed;
A display step for displaying detection information of the extracted abnormal shadow candidate;
It is characterized by including.

The invention according to claim 2 is the display method of the abnormal shadow candidate according to claim 1,
In the extracting step, an image feature amount of the detected abnormal shadow candidate is calculated, a priority order of the abnormal shadow candidates to be displayed is determined based on the image feature amount, and among the detected abnormal shadow candidates, It is characterized by preferentially extracting from abnormal shadow candidates having a high priority.

The invention according to claim 3 is the display method of the abnormal shadow candidate according to claim 2,
In the extracting step, the contrast between the abnormal shadow candidate and the surrounding area is calculated as the image feature amount, and the priority of the abnormal shadow candidate having a small contrast is determined high.

The invention according to claim 4 is the display method of the abnormal shadow candidate according to claim 2,
In the extracting step, an area of an abnormal shadow candidate region is calculated as the image feature amount, and a priority order of abnormal shadow candidates having a small area is determined high.

The invention according to claim 5 is the display method of the abnormal shadow candidate according to claim 2,
In the extraction step, the distance from the image edge of the medical image to the detection position of the abnormal shadow candidate is calculated as the image feature amount, and the priority of the abnormal shadow candidate with a short distance is determined to be high. .

Invention of Claim 6 is the display method of the abnormal shadow candidate as described in any one of Claims 2-5,
Recognizing a subject area from the medical image, and classifying the subject area into a plurality of areas,
In the extracting step, a priority order is determined according to a classification region in which the abnormal shadow candidate is detected among the recognized subject regions.

The invention according to claim 7 is the display method of the abnormal shadow candidate according to claim 2,
In the extracting step, a feature amount of a shape of an abnormal shadow candidate is calculated as the image feature amount, and a priority order is determined based on the feature amount of the shape.

The invention according to claim 8 is the method of displaying an abnormal shadow candidate according to claim 2,
In the extraction step, the density of a candidate region for an abnormal shadow candidate is calculated as the image feature amount, and the priority of an abnormal shadow candidate having a feature amount with a low density is determined.

The invention according to claim 9 is the display method of the abnormal shadow candidate according to claim 2,
In the extraction step, at least the area of the candidate area of the abnormal shadow candidate, the density, the contrast with the surrounding area, the shape of the abnormal shadow candidate, and the distance from the edge of the medical image to the detection position of the abnormal shadow candidate as the image feature amount One or more of them are calculated, and the priority order is determined based on the calculated feature amount.

According to a tenth aspect of the present invention, in the medical image processing system,
An abnormal shadow candidate detecting means for analyzing a medical image and detecting an abnormal shadow candidate;
Control means for extracting abnormal shadow candidates to be displayed among abnormal shadow candidates detected by the abnormal shadow candidate detection means;
Display information for displaying detection information of the extracted abnormal shadow candidates;
It is characterized by providing.

  According to the first, second, and tenth aspects of the present invention, the display amount of detection information (such as marker display at the detection position) on the display screen can be reduced by reducing the number of abnormal shadow candidates to be displayed. Detection information can be provided in a display state that is easy for a doctor to observe. In addition, by providing only detection information for abnormal shadow candidates that are suspected to be true positives and that need to be examined by a doctor, abnormal shadow candidates that have low visibility, or abnormal shadow candidates that fall under any of the above, the result is a doctor. This makes it possible to improve the efficiency of diagnostic work and improve the diagnostic accuracy.

  According to the third aspect of the present invention, since the contrast is small, it is possible to preferentially provide the doctor with the detection information of the abnormal shadow candidate that requires the doctor's judgment as to whether or not it is true positive. The efficiency of the interpretation work can be improved. In addition, when the contrast is small, the visibility is lowered. Therefore, it is possible to prevent the doctor from overlooking by giving priority to such abnormal shadow candidates.

  According to the invention described in claim 4, since the area is small, it is possible to preferentially provide the doctor with detection information of an abnormal shadow candidate that requires the doctor's judgment as to whether or not it is true positive. The efficiency of the interpretation work can be improved. Further, when the area is small, the visibility is lowered. Therefore, by giving priority to such abnormal shadow candidates, it is possible to prevent the doctor from overlooking.

  According to the fifth aspect of the present invention, it is possible to preferentially provide the doctor with the detection information of the abnormal shadow candidate that exists at a position where the doctor can easily overlook, thereby preventing the doctor from overlooking.

  According to the sixth aspect of the present invention, detection information of abnormal shadow candidates existing in an area with low visibility can be preferentially provided to a doctor, and oversight of the doctor can be prevented.

  According to the seventh aspect of the present invention, it is possible to preferentially provide a doctor with detection information of an abnormal shadow candidate that is difficult to determine whether or not it is true positive and needs a final determination by the doctor. Can improve the efficiency of interpretation work. Moreover, since visibility may fall depending on a shape, a doctor's oversight prevention can be prevented by displaying such an abnormal shadow candidate preferentially.

  According to the invention described in claim 8, it is possible to preferentially provide the doctor with detection information of an abnormal shadow candidate having a low density and low visibility. Thereby, it is possible to prevent oversight of the doctor.

  According to the ninth aspect of the present invention, it is possible to preferentially detect detection information of an abnormal shadow candidate that is suspected to be true positive and needs a close examination by a doctor, an abnormal shadow candidate that is difficult for a doctor to visually recognize, or both. Can be provided. In addition, since the priority order is determined based on each feature amount, the priority order can be determined comprehensively in consideration of various determination factors.

  In the present embodiment, an example using a breast image obtained by photographing a breast as a medical image will be described.

First, the configuration will be described.
FIG. 1 shows a system configuration of a medical image processing system 100 in the present embodiment.
The medical image processing system 100 is a system that takes a medical image, performs image processing on the medical image, detects an abnormal shadow candidate, and provides the doctor with the detection information together with the medical image.

  As illustrated in FIG. 1, the medical image processing system 100 includes an image generation device 1, an image processing device 2, a printer 3, an image server 4, and a viewer 5. These devices 1 to 5 are connected to each other through a communication network N constructed in a medical institution such as a LAN (Local Area Network) so that data can be transmitted and received between them. The DICOM (Digital Imaging and Communication in Medicine) standard is applied to the communication network N.

Hereinafter, each component apparatus 1-5 is demonstrated.
The image generation apparatus 1 captures a human body and generates digital data of the captured image (medical image). For example, CR (Computed Radiography), FPD (Flat Panel Detector), CT, MRI, cassette-specific reading Modality such as an apparatus or a film digitizer can be applied. In the present embodiment, it is assumed that breast image data is generated by applying a breast-specific CR that performs X-ray imaging of the left and right breasts as the image generation apparatus 1.

  Note that the image generation device 1 is a device that complies with the DICOM standard described above, and includes various information attached to the generated medical image (for example, patient information regarding a patient whose medical image is captured, imaging information regarding imaging or examination, Inspection information, etc.) can be input from the outside and can be automatically generated. The image generation apparatus 1 adds the supplementary information as header information to the generated medical image and transmits it to the image processing apparatus 2 via the communication network N. In addition, when not conforming to the DICOM standard, incidental information can be input to the image generating apparatus 1 using a DICOM conversion apparatus (not shown).

  The image processing apparatus 2 performs various types of image processing on the medical image supplied from the image generation apparatus 1 and performs an image analysis of the medical image to detect an abnormal shadow candidate.

  The printer 3 outputs a medical image to a recording medium such as a film based on the medical image data received from the image processing apparatus 2 or the image server 4.

  The image server 4 includes an image DB 4a. In the image DB 4a, the medical image (original image) generated by the image generation device 1 and the image processed medical image (processed image) received from the image processing device 2 are stored. And manage its input and output.

  The viewer 5 is a display means used by a doctor for diagnosis, and includes a liquid crystal display (LCD) or the like. The viewer 5 acquires a specified medical image from the image server 4 and displays it in accordance with a doctor's operation instruction. Alternatively, the detection result of the abnormal shadow candidate detected by the image processing apparatus 2 is received and displayed.

Next, the image processing apparatus 2 according to the present invention will be described in detail.
FIG. 2 shows an internal configuration of the image processing apparatus 2.
The image processing apparatus 2 includes a control unit 21, an operation unit 22, a display unit 23, a communication unit 24, a storage unit 25, an image processing unit 26, and an abnormal shadow candidate detection unit 27.

  The control unit 21 includes a CPU (Central Processing Unit), a RAM (Random Access Memory), and the like. The CPU reads various control programs from the storage unit 25 and develops them in the RAM. Then, the execution of processing is comprehensively controlled according to the program, and the operation of each unit is centrally controlled. For example, display target determination processing (described later) is executed in accordance with the display target determination processing program according to the present invention.

  The operation unit 22 includes a keyboard, a mouse, and the like, generates an operation signal corresponding to a key operation and a mouse operation, and outputs the operation signal to the control unit 21.

  The display unit 23 includes an LCD or the like, and performs various displays such as an operation screen at the time of image processing and a medical image after processing in accordance with instructions from the control unit 21.

  The communication unit 24 includes a communication interface such as a router or a modem, and communicates with an external device on the communication network N according to an instruction from the control unit 21. For example, a medical image to be processed is received from the image generation apparatus 1 or a processed medical image is transmitted to the image server 4 or the printer 3.

  The storage unit 25 stores various control programs, parameters necessary for execution of the programs, or processing result data.

  The image processing unit 26 performs gradation conversion processing, sharpness adjustment processing, and the like on the medical image according to the image processing program. In the case of a breast image, the image processing unit 26 performs alignment processing for combining the left and right breast images side by side. Execute.

  The abnormal shadow candidate detection unit 27 performs image analysis of the medical image to extract a subject area, classify the subject area into a plurality of areas, calculate an image feature amount from the medical image, and based on the image feature amount. Detect abnormal shadow candidates.

Here, a procedure for extracting each region from the medical image will be described with reference to FIG.
FIG. 3 is a diagram showing a breast image S taken in an oblique direction (hereinafter referred to as MLO). A subject area Sa is extracted from the breast image S, and a breast area Sa1 and a pectoral muscle area Sa2 are extracted from the subject area Sa. And the breast region Sa1 is classified into three regions Da, Db, and Dc.

(1) First, the abnormal shadow candidate detection unit 27 obtains a variance histogram of pixel values in the breast image S, and uses a discriminant analysis method (intraclass variance and interclass variance in two classes when the variance histograms are classified into two classes). The threshold value is determined using a method for determining the threshold value so that the discriminant ratio (dispersion ratio) is maximized), and the breast image S is binarized using the determined threshold value.

  At this time, in the breast image S, an unexposed region (a region that does not correspond to the subject portion and is a region directly irradiated with X-rays) exhibits a high density and thus becomes “1” by binarization. The subject area is expected to be “0”. Therefore, by this binarization, the breast image S can be classified into the subject area Sa and the other missing areas Sb. When the photographing direction is the front direction (hereinafter referred to as CC), the pectoral muscle is not reflected in the subject portion, and the subject region Sa becomes the breast region Sa1. Also, the boundary between the two regions (the subject region and the missing region) by binarization is recognized as the skin line SL.

(2) On the other hand, when the shooting direction is MLO, the pectoralis muscles lie in the subject portion, so that the pectoral muscle region Sa2 is recognized next from the subject region Sa. For the pectoral muscle region Sa2, for example, a density gradient in the subject region Sa is calculated, and the subject region Sa is classified into a pectoral muscle region Sa2 and a breast region Sa1 based on the density gradient. As described in Japanese Patent Application Laid-Open No. 2001-238868, a local region is set, a threshold is set based on the pixel value in the local region, and the subject region Sa is binarized, whereby the pectoral muscles The region Sa2 and the breast region Sa1 may be recognized.

(3) Next, the breast region Sa1 is classified into three regions Da, Db, and Dc.
The breast region is a mixture of mammary gland and fat, and the concentration varies depending on the density. Therefore, when a doctor makes a diagnosis, the breast region is classified based on the concentration. Classification is high-concentration region with a high fat content and mammary gland content of less than 10%, medium-concentration region with mammary gland content of 10% or more and less than 50%, and mammary gland content of 50% or more In many cases, it is classified into three regions, which are considered to be low concentration regions. Abnormal shadows appear white and at low density on the image, so it is difficult to visually detect abnormal shadows when there is a lesion in the low-density area corresponding to the whitest density of the pectoralis major muscle or in a slightly whitish medium-density area. Become.

Therefore, the breast region Sa1 is classified into the high concentration region Da, the intermediate concentration region Db, and the low concentration region Dc with reference to the density of the pectoral muscle region Sa2. Specifically, a dispersion histogram of the pectoral muscle region Sa2 is created, a region having a relatively uniform density is extracted from the shape, and the average density is used as a threshold value and classified into each region Da to Dc. For example, the remaining area from which the area having a higher density than the threshold is deleted is denoted by Dc, and the area having a density of 30% or higher of the threshold is Db, and the area having a high density by 60% or higher is Da. By doing, it can classify | categorize into each area | region Da-Dc.

Next, a method for detecting abnormal shadow candidates will be described.
The abnormal shadow candidate detection unit 27 executes detection processing according to an algorithm detection program corresponding to the type of abnormal shadow to be detected. In the case of a breast image, a mass that is a cancerous part of breast cancer and a shadow candidate of a microcalcification cluster are detected.

  As an abnormal shadow candidate detection algorithm, a known algorithm can be applied. For example, as an algorithm for a tumor shadow candidate in a breast image, a method using an iris filter disclosed in Japanese Patent Laid-Open No. 10-91758, a method using a Laplacian filter (Journal of the Institute of Electronics, Information and Communication Engineers (D-II) , Vol. J76-D-II, no. 2, pp. 241-249, 1993), etc. are applicable. In addition, as a detection algorithm for micro-calcification cluster shadow candidates, for example, Morphology filter (The Institute of Electronics, Information and Communication Engineers (D-II), Vol.J71-D-II, no.7, pp.1170-1176, 1992), Laplacian filter (The Institute of Electrical, Information and Communication Engineers Journal (D-II), Vol. J71-D-II, no. 10, pp. 1994-2001, 1998) Applicable.

Hereinafter, as an example of a method for detecting an abnormal shadow candidate, a method for detecting a minute calcified cluster shadow candidate in a breast image will be described.
The microcalcification cluster shadow appears as a shadow in which low-density microcalcification portions having a substantially conical density change are gathered (clustered) on the breast image. A square area is sequentially set for a medical image based on such density characteristics, and each area (this area is referred to as an attention area) has a unique vector pattern as a filter for detecting a minute calcification portion. The primary detection of abnormal shadow candidates is performed by applying a double ring filter. Note that the size of the region of interest may be set according to the lesion type targeted for detection.

  The triple ring filter is composed of three ring filters in which the intensity component and the direction component of the density gradient when the density change shows an ideal conical shape are determined in advance. First, in the vicinity of a target pixel, the intensity value of the density gradient and the representative value of the direction component are obtained from the pixel values on the respective regions of the ring filters. Based on the difference between the representative value and the intensity component and direction component of the density gradient determined in advance for each ring filter, an image region having a density change close to a conical shape is first detected.

  When a candidate area is specified by the primary detection, various image feature quantities such as contrast, standard deviation, density average value, curvature, fractal dimension, circularity, and area are calculated in the candidate area, and based on these feature quantities. It is determined whether or not a true positive abnormal shadow (refers to a shadow that is truly a lesion part. On the other hand, a shadow of a normal tissue that is mistaken for a lesion is referred to as a false positive shadow). Multivariate analysis can be applied as a determination method. For example, a multivariate analysis is constructed using image features calculated from shadows that are known to be true positive in advance as learning data, and various image feature values calculated from shadow candidates to be determined are input to this multivariate analysis. Thus, an index value indicating the possibility of true positive is obtained. It is determined whether the index value is true positive or false positive by comparing it with a threshold value prepared in advance to determine that it is true positive. Then, a region where the candidate regions determined to be true positive gather within a predetermined area is output as a candidate region for the microcalcification cluster shadow.

Next, the operation of the medical image processing system 100 will be described.
First, the flow from generation to storage of a medical image will be described.
First, imaging is performed in the image generation device 1 to generate a medical image (here, an example of a breast image will be described). And as related information of the generated breast image, information about the patient such as the patient's nomination, age, sex, etc., imaging information such as tube voltage at the time of imaging, breast compression rate, examination information such as examination date and time, Detailed information related to breast image generation such as reading conditions is attached to the breast image.

The breast image with the accompanying information is output from the image generation apparatus 1 to the image processing apparatus 2.
In the image processing apparatus 2, while necessary image processing is performed on the breast image, a display target determination process described below is started for the breast image, and detection of abnormal shadow candidates and display targets from the detected abnormal shadow candidates are displayed. Candidates are extracted.

Hereinafter, the display target determination process executed by the image processing apparatus 2 will be described with reference to FIG.
In the display target determination process shown in FIG. 4, first, the abnormal shadow candidate detection unit 27 extracts a subject area, a pectoral muscle area, a breast area, and the like included in the medical image (step S <b> 1).

  Next, the abnormal shadow candidate detection unit 27 performs an abnormal shadow candidate detection process on the extracted breast region (step S2). This abnormal shadow candidate detection process will be described with reference to the flowchart shown in FIG. First, attention areas, which are processing units for performing detection processing, are sequentially set for the breast area, and primary detection is performed within the attention area (step S21). Since the primary detection method is as described above, a description thereof is omitted here.

  Next, an image feature amount is calculated for the region of the abnormal shadow candidate that is primarily detected. At least the area of the abnormal shadow candidate area, the standard deviation, the density average value, the curvature, the fractal dimension, the circularity, the contrast with the surrounding area, the complexity of the edge, and the abnormal shadow candidate are first detected. A classification area (which of the areas Da to Dc shown in FIG. 3 corresponds to) is calculated (step S22).

The contrast is the density difference between the abnormal shadow candidate area and its surrounding area, the area is the number of pixels in the abnormal shadow candidate area, and the distance from the image edge is the distance from the image edge on the chest wall side to the detection position of the abnormal shadow candidate It is obtained from (number of pixels).
The circularity is calculated by the following formula (1). As the circularity C is closer to 1, the shape of the abnormal shadow candidate is closer to a circle and is a lesion (true positive). As the value becomes smaller than 1, the figure becomes more complex and true Indicates that careful scrutiny is required.

Further, the complexity of the edge is expressed by the expansion coefficient a _k , b calculated by the following expression (2) when the periodic function indicating the contour of the abnormal shadow candidate is Fourier expanded as shown in FIG. _k . As the values of a _k and b _k are larger, the shape of the edge portion of the abnormal shadow candidate region is more distorted.

Next, these image feature quantities are input to a multivariate analysis constructed in advance to output an index value indicating the possibility of an abnormal shadow, and an index value indicating the possibility of a true positive abnormal shadow is obtained (step) S23). Next, when it is determined based on the obtained index value and threshold value whether or not it is an abnormal shadow candidate (step S24), the process proceeds to step S3 in FIG.

  In step S3 shown in FIG. 4, the control unit 21 determines the priority order of the abnormal shadow candidates to be displayed among these abnormal shadow candidates based on the image feature amount calculated for the abnormal shadow candidates (step S3). S3). Here, the abnormal shadow candidate to be displayed is a candidate excluding a candidate that is clearly determined to be true positive or false positive, and is suspected of being true positive or false positive or difficult to observe. It is a candidate that is easy to miss.

  For example, when the contrast is small and the area is small, it is difficult to determine whether it is false positive or true positive, and it is often necessary to make a final determination by an interpreting doctor. Moreover, it is difficult to say that the visibility of the area is good, and the shadow is easily overlooked by a doctor. In addition, in the abnormal shadow that should be finally determined as true positive, the border of the edge becomes unclear (unclear), or the edge shows a fine streak-like shape, or the spicule There is a case where a so-called distorted shape is exhibited, and the complexity of the edge increases, but conversely, when the complexity of the edge is small, it is difficult to determine whether the shadow is abnormal. In the case of a tumor, the greater the degree of circularity, the higher the possibility of being a tumor. However, if the degree of circularity is small, it is difficult to determine whether the tumor is true (positive). Further, even in the same breast region Sa1, the low density region Dc has a generally low density, so that the abnormal shadow appearing on the image at a low density is lower than that in the high density region Da. It is harder to discriminate if it exists in Dc. Similarly, regarding the density of the abnormal shadow candidate region itself, it is difficult to visually recognize if the density is low, and it is easy for the doctor to overlook. Furthermore, the abnormal shadow existing in the breast region near the edge of the image is easily overlooked by the doctor from that position.

  Therefore, considering priority, decision factors for contrast, area, density average value, classification area where abnormal shadow candidates are located, distance from image edge, edge complexity, etc., comprehensively based on these Determine priority.

  Specifically, learning data such as a shadow feature value that should have a higher priority and a shadow feature value that should have a lower priority are prepared in advance, and 0 to 1 (close to 0) as an index value indicating the priority. A multivariate analysis such as a neural network that outputs a lower priority and a higher priority closer to 1 is constructed. For example, the lower the contrast, the smaller the area, the lower the average density value of the candidate area, the lower the circularity, the shorter the distance from the image edge, or the lower the complexity of the edge, the higher the index value. Prepare learning data to adjust the multivariate analysis. Further, in the breast region Sa1, it is difficult to determine abnormal shadow candidates in the order of the regions Dc → Db → Da. Therefore, the multivariate analysis is constructed so that the index values have higher priority in this order.

  Then, the feature value of the abnormal shadow candidate whose priority is to be determined is input to the multivariate analysis, the index value output by the multivariate analysis is compared with the threshold value for this index value, and the abnormal value exceeding the threshold value is compared. A shadow candidate, that is, a candidate above a predetermined priority order is determined as an abnormal shadow candidate to be displayed.

  When the priority order is determined, out of the abnormal shadow candidates detected in the detection process of step S2, the abnormal shadow candidate having the higher priority order is extracted preferentially (step S4) and output as the final detection result. .

  The detection result is transmitted from the image processing device 2 to the image server 4 and stored, while the medical image (the image subjected to image processing for interpretation) and the detection result are transferred from the image processing device 2 to the viewer 5. Is transmitted and displayed in the viewer 5.

FIG. 7 shows a display example.
FIG. 7A shows an example in which only the detection results of abnormal shadow candidates whose priority order is equal to or higher than a predetermined order are displayed. FIG. 7B shows a table in which detection results of all detected abnormal shadow candidates are displayed. It is an example. FIGS. 7A and 7B show that displaying only the abnormal shadow candidates with the priority higher than the predetermined order reduces the number of display objects on the display screen and makes it easier to observe than displaying all detected ones. I understand.

FIGS. 7A and 7B show an example in which the detection result of the abnormal shadow candidate is directly displayed on the medical image by marker information (arrow in the figure) indicating the detection position of the abnormal shadow candidate. However, as shown in FIG. 7C, a sub-image (images indicated by reference signs g1 and g2 in the drawing) is created to reduce the main image (images indicated by reference signs G1 and G in the drawing). It may be displayed at a position that does not overlap the subject area. In this case, the display area of the detection result is smaller than that in the display form of FIG. 7A, but according to the present invention, the number of abnormal shadow candidates to be displayed can be reduced, which is particularly effective. In FIG. 7C, the breast images of the left and right breasts are aligned so that the image ends on the chest wall side are adjacent to each other.

  As described above, according to the present embodiment, from the abnormal shadow candidate detected by the detection process, the abnormal shadow candidate that is clearly true positive or false positive is deleted, and it is difficult to determine whether it is true positive or false positive, Since the detection information is displayed only on candidates that should be left to the doctor's final judgment or candidates that have low visibility and are easily overlooked by the doctor, the number of abnormal shadow candidates to be pointed out can be reduced. Thereby, it can be set as an easy-to-see display, and the efficiency of doctor's interpretation work can be improved.

  In addition, the doctor detects the abnormal shadow that is clearly true positive by the doctor himself / herself, and the detection of the abnormal shadow candidate provided by the medical image processing system 100 is used for the detection of the abnormal shadow that is low in visibility and difficult to detect. You can refer to the results. When all of the detected abnormal shadow candidates are pointed out and displayed, the doctor will confirm all the candidates including those that are clearly known to be true positive, which is cumbersome. Therefore, as described above, only the abnormal shadow candidate that is difficult to find is provided with the information of the detection result, so that the doctor can prevent oversight and the doctor's interpretation work becomes efficient.

In addition, embodiment mentioned above is a suitable example to which this invention is applied, and is not limited to this.
For example, as a method for determining the priority order, the multivariate analysis method is used and the priority order is determined comprehensively. However, the present invention is not limited to this, and the order of decreasing contrast and the degree of circularity are small for each abnormal shadow candidate. The priority order may be determined for each determination element, such as order, and the number of priority orders for each determination element may be added, and the priority order may be extracted from candidates having a small addition value. At this time, the priority order can be optimized by weighting and adding to the determination elements to be emphasized.

  In the above description, the abnormal shadow candidate detecting means and the control means are realized in the image processing apparatus 2 and the display means is realized in the viewer 5. However, the present invention is not limited to this, and any constituent device of the medical image processing system 100 is used. It is good also as implement | achieving each means.

  In particular, the image generation apparatus 10 takes a picture by the phase contrast method disclosed in Japanese Patent Laid-Open Nos. 2001-238871, 2001-317101, 2002-85389, 2001-299733, and the like. When executed, it is known that the edge of the lesion in the obtained image is significantly improved in visibility compared to the conventional breast image due to the phase contrast effect, and there is no clear display of the annotation. The probability of overlooking the (true positive candidate region) is preferable because it greatly decreases.

In the description of the above embodiment, the “region with low density and low visibility” indicates a region where the amount of light reaching the eyes of the interpreting doctor is small when the interpreting doctor interprets. Therefore, in the case of an image interpretation method using film and Schaukasten, it is a film area where the amount of transmitted light is reduced, which corresponds to a black portion on the film, and conversely, a blank portion on the film is “high density and high visibility. Part. On the other hand, in the case of the interpretation method by the viewer, the low density (low drive level) area is the “low density and low visibility” area.
In general, an area with a small amount of X-rays transmitted through the subject at the time of image interpretation becomes an unclear part or is reproduced and displayed with high brightness. Therefore, it is sufficient to extract a region having a large X-ray dose that has passed through the subject.

It is a figure which shows the system configuration | structure of the medical image processing system in this embodiment. It is a figure which shows the internal structure of the image processing apparatus of FIG. It is a figure which shows each area | region classified in a breast image. It is a flowchart explaining the display target judgment process performed by the image processing apparatus. It is a flowchart explaining the detection process of the abnormal shadow candidate performed with an image processing apparatus. It is a figure explaining the calculation method of the feature-value of edge complexity. It is a figure which shows the example of a display of the detection result of an abnormal shadow candidate.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Medical image processing system 1 Image generation apparatus 2 Image processing apparatus 21 Control part 26 Image processing part 27 Abnormal shadow candidate detection part 3 Printer 4 Image server 4a Image DB
5 Viewer

Claims (10)

A detection step of analyzing a medical image and detecting an abnormal shadow candidate;
Of the detected abnormal shadow candidates, an extraction step of extracting abnormal shadow candidates to be displayed;
A display step for displaying detection information of the extracted abnormal shadow candidate;
A method for displaying an abnormal shadow candidate characterized by comprising:   In the extracting step, an image feature amount of the detected abnormal shadow candidate is calculated, a priority order of the abnormal shadow candidates to be displayed is determined based on the image feature amount, and among the detected abnormal shadow candidates, 2. The method of displaying an abnormal shadow candidate according to claim 1, wherein the abnormal shadow candidate is preferentially extracted from abnormal shadow candidates having a high priority.   3. The abnormality according to claim 2, wherein in the extraction step, a contrast between an abnormal shadow candidate and a peripheral region thereof is calculated as the image feature amount, and a priority of an abnormal shadow candidate having a small contrast is determined high. How to display shadow candidates.   3. The abnormal shadow candidate according to claim 2, wherein in the extraction step, an area of an abnormal shadow candidate region is calculated as the image feature amount, and the priority of the abnormal shadow candidate having a small area is determined high. Display method.   In the extraction step, the distance from the image edge of the medical image to the detection position of the abnormal shadow candidate is calculated as the image feature amount, and the priority of the abnormal shadow candidate with a short distance is determined to be high. The abnormal shadow candidate display method according to claim 2. Recognizing a subject area from the medical image, and classifying the subject area into a plurality of areas,
The said extraction process WHEREIN: A priority is determined according to the classification | category area | region where the said abnormal shadow candidate was detected among the said to-be-photographed object area | regions. How to display abnormal shadow candidates.   3. The abnormal shadow candidate according to claim 2, wherein in the extraction step, a feature amount of a shape of the abnormal shadow candidate is calculated as the image feature amount, and a priority order is determined based on the feature amount of the shape. Display method.   3. The extraction step according to claim 2, wherein in the extraction step, a density of a candidate region of an abnormal shadow candidate is calculated as the image feature amount, and a priority order of an abnormal shadow candidate having a feature amount having a low density is determined. Method for displaying abnormal shadow candidates.   In the extraction step, at least the area of the candidate area of the abnormal shadow candidate, the density, the contrast with the surrounding area, the shape of the abnormal shadow candidate, and the distance from the edge of the medical image to the detection position of the abnormal shadow candidate as the image feature amount 3. The method for displaying an abnormal shadow candidate according to claim 2, wherein one or more are calculated, and the priority order is determined based on the calculated feature amount. 4. An abnormal shadow candidate detecting means for analyzing a medical image and detecting an abnormal shadow candidate;
Control means for extracting abnormal shadow candidates to be displayed among abnormal shadow candidates detected by the abnormal shadow candidate detection means;
Display information for displaying detection information of the extracted abnormal shadow candidates;
A medical image processing system comprising:

Images