JP2007151645A - Medical diagnostic imaging support system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve the efficiency of image reading work of an image reading doctor by providing detection information of abnormal shadow candidates in a display easily viewable for the image reading doctor. <P>SOLUTION: A viewer 5 of this medical diagnostic imaging support system 100 has a first display mode superimposedly displaying an annotation indicating the positions of all the abnormal shadow candidates detected based on a first threshold on a medical image in an image processor 2, a second display mode superimposedly displaying an annotation indicating the positions of abnormal shadow candidates detected based on a second threshold on the medical image, a third display mode superimposedly displaying an annotation indicating the positions of abnormal shadow candidates detected by the first threshold but non-detected by the second threshold on the medical image, and a non-display mode displaying no annotation on the medical image. The display is controlled based on a display priority order stored for every image reading doctor in a display priority order set file. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a medical image diagnosis support system that captures a medical image, performs image processing on the medical image, detects an abnormal shadow candidate, and provides the detected information together with the medical image to an interpretation doctor.

  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 when an interpreting doctor makes a diagnosis, the image is displayed on a display. It has come to display and interpret. 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 CAD, an image feature amount of the candidate region 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 such as a method for detecting a candidate region for an abnormal shadow are used. Abnormal shadow candidates detected by CAD are not only true abnormal shadows (true positives), but are also unclear whether they are abnormal shadows and are similar to abnormal shadows that ultimately need to be judged by an interpreting physician The detected shadow is detected. When an abnormal shadow candidate is detected by the CAD, an annotation (annotation) indicating a detection position of the abnormal shadow candidate (marker such as a character, an arrow, or a circle) is generally displayed on the medical image as detection information. Therefore, the interpreting doctor interprets the image while referring to the indication of the annotation.

  When the number of abnormal shadow candidates detected by CAD is large, many annotations are displayed on the image observed by the interpretation doctor, making it difficult to observe the medical image itself, which adversely affects the interpretation work. Sometimes.

  Therefore, for example, in Patent Document 1, an evaluation value indicating the degree of abnormality of a detected abnormal shadow candidate is calculated, and the evaluation value is compared with a reference value to determine whether or not the abnormal shadow candidate is abnormal. When the number of abnormal shadow candidates determined to be abnormal by the first determination unit and the first determination unit exceeds a predetermined number M, it is determined that the evaluation value from the higher rank to the predetermined number N is abnormal An abnormal shadow detection apparatus including a second determination unit is described.

Further, Patent Document 2 discloses an image diagnosis support apparatus capable of setting a display switching order for each interpreting doctor, such as display of captured images in the vertical direction of left and right breasts, display of captured images in an oblique direction, and display obtained by dividing these images. In addition, there is described an image diagnosis support apparatus that determines the rank based on the certainty of abnormal shadow candidates and the type of abnormality, and switches and displays individual abnormal shadow candidates according to the rank on the display screen.
JP 2004-195080 A JP 2000-287955 A

  By the way, depending on the experience of the interpreting doctor, etc., an interpreting doctor who can interpret images more effectively by pointing out the position of the clearly positive abnormal shadow candidate with annotation, and clearly positive positive abnormal shadow candidate by annotation Even if there is no indication, it is possible to judge visually, and there are some interpretation doctors who are difficult to see if many annotations are pointed out with annotations, and on the contrary, the efficiency of interpretation is reduced. For this reason, displaying all the abnormal shadow candidates that are uniformly detected as annotations has a problem in that the interpretation efficiency of some interpretation doctors decreases.

  In the above-mentioned Patent Document 1, when the number of abnormal shadow candidates is a predetermined number M or more, the low evaluation value is limited so as not to be detected. It does not describe a technique for displaying the detection result in an easy-to-read manner according to the interpreting doctor. Patent Document 2 describes a technique for switching the display of the left and right breasts for each interpretation doctor, but does not describe a technique for solving the above-described problem.

  An object of the present invention is to provide detection information of abnormal shadow candidates in a display that is easy to see for an interpreting doctor, and to improve the efficiency of the interpreting work of the interpreting doctor.

In order to solve the above-mentioned problem, the invention described in claim 1
In a medical image diagnosis support system comprising: an abnormal shadow candidate detecting means for analyzing a medical image and detecting an abnormal shadow candidate; and a display means for displaying detection information of the abnormal shadow candidate detected by the abnormal shadow candidate detecting means. ,
The abnormal shadow candidate detecting means detects an abnormal shadow candidate from the medical image using a first threshold value and a second threshold value,
The display means includes at least a first display mode for displaying detection information of an abnormal shadow candidate detected by the first threshold in the abnormal shadow candidate detection means, and an abnormal shadow candidate detected by the second threshold. A second display mode for displaying the detection information, and a third display mode for displaying the detection information of the abnormal shadow candidate detected by the first threshold and not detected by the second threshold,
Setting means for setting display priorities of at least the first to third display modes;
Display control means for controlling display of the abnormal shadow candidate detection information on the display means based on the setting from the setting means;
It is characterized by having.

The invention according to claim 2 is the invention according to claim 1,
The abnormal shadow candidate detection means includes index value calculation means for calculating an index value indicating a degree of difficulty in determining true positive and false positive for each of the abnormal shadow candidates detected by the first threshold. Detection is further performed based on the second threshold value and the index value calculated by the index value calculation means from among the abnormal shadow candidates detected by the threshold value of 1.

The invention according to claim 3 is the invention according to claim 1 or 2,
The setting means is capable of setting the display priority for each interpretation doctor,
Storage means for storing the display priority order set for each interpretation doctor by the setting means is provided.

The invention according to claim 4 is the invention according to any one of claims 1 to 3,
The third display mode further includes a mode for collectively displaying detection information of abnormal shadow candidates to be displayed and a mode for sequentially displaying the detection information individually.

The invention according to claim 5 is the invention according to any one of claims 1 to 4,
The detection information is displayed by an annotation indicating a detection position on the medical image of the abnormal shadow candidate detected from the medical image,
The display control unit differs between a detection position of the abnormal shadow candidate detected by the second threshold value and a detection position of the abnormal shadow candidate detected by the first threshold value and not detected by the second threshold value. It is characterized by displaying with annotation.

  According to the first aspect of the present invention, the first display mode for displaying the detection information of the abnormal shadow candidate detected by the first threshold and the detection information of the abnormal shadow candidate detected by the second threshold are displayed. In the second display mode, the detection of abnormal shadow candidates detected by the first threshold and not detected by the second threshold is set based on the display priority setting of the third display mode for displaying the detection information of the abnormal shadow candidates. Since the detection information is displayed, it is possible to provide detection information of abnormal shadow candidates in a display that is easy to see for the interpreting physician, and the efficiency of the interpretation work can be improved.

  According to the second aspect of the present invention, display of detection information of all abnormal shadow candidates detected using the first threshold value and true positive among abnormal shadow candidates detected using the first threshold value. Interpretation doctor from the display of only the abnormal shadow candidates that can be easily determined to be present, and the display of only the abnormal shadow candidates that are difficult to determine true positive or false positive among the abnormal shadow candidates detected using the first threshold Therefore, it is possible to provide detection information of abnormal shadow candidates in a display that is easy to see for the user, and it is possible to improve the efficiency of interpretation work.

  According to the third aspect of the present invention, since display priority can be set for each interpreting doctor, when a plurality of interpreting doctors use the same display means, an abnormal shadow is displayed with an easy-to-read display for each interpreting doctor. Candidate detection information can be provided.

  According to the fourth aspect of the present invention, the third display mode for displaying the detection information of the abnormal shadow candidate for which it is difficult to determine whether it is true positive or false positive is the mode in which the batch display mode and the individual display mode are sequentially displayed. Therefore, it is possible for the image interpretation doctor to provide detection information of an abnormal shadow candidate that is difficult to determine whether it is a true positive or a false positive with an easy-to-read display.

  According to the invention described in claim 5, the interpretation doctor glances at the detection information of the abnormal shadow candidate detected by the second threshold value and the detection information of the abnormal shadow candidate that is not displayed by the threshold value of the second threshold value. It becomes possible to display in an identifiable manner.

  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 diagnosis support system 100 in the present embodiment.
The medical image diagnosis support system 100 is a system that captures a medical image, performs image processing on the medical image, detects an abnormal shadow candidate, and provides detection information along with the medical image to an interpretation doctor.

  As illustrated in FIG. 1, the medical image diagnosis support 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)
), Modalities such as CT and MRI are applicable. 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 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 is an abnormal shadow candidate detection unit that performs various 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 an interpretation doctor for diagnosis, and includes a liquid crystal display (LCD) or the like. The viewer 5 acquires and displays the specified medical image from the image server 4 and receives and displays the detection result of the abnormal shadow candidate detected by the image processing device 2 according to the operation instruction of the interpretation doctor.

Next, the configuration of the image processing apparatus 2 and the viewer 5 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, and each unit is connected by a bus 28. ing.

  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, the abnormal shadow candidate detection and classification process (described later) is executed according to the abnormal shadow candidate detection and classification process 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 LAN card 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 programs, parameters necessary for execution of the programs, or processing result data.

  The image processing unit 26 includes a CPU, a RAM, and the like. The image processing unit 26 reads an image processing program stored in the storage unit 25 and performs gradation conversion processing, sharpness on an input medical image in cooperation with the program. In addition to performing the degree adjustment process and the like, if the medical image is a breast image, a registration process for combining the left and right breast images side by side is performed.

  The abnormal shadow candidate detection unit 27 includes a CPU, a RAM, and the like, reads out a subject region extraction and classification program, an abnormal shadow candidate detection program, and the like from the storage unit 25, and cooperates with these programs to detect a subject described later. Performs region extraction and classification processing, abnormal shadow candidate detection processing, etc., performs image analysis of the input medical image, extracts the subject region, classifies the subject region into a plurality of regions, and extracts image features from the medical image The amount is calculated, and an abnormal shadow candidate is detected based on the image feature amount.

Here, the subject region extraction and classification process 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). In the subject region extraction and classification process, a subject region Sa is extracted from the breast image S, and the subject region Sa is extracted. After extracting the breast region Sa1 and the pectoral muscle region Sa2 from the breast region Sa1, 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, the blank region Sb (a region that does not correspond to the subject portion and is directly irradiated with X-rays) has a high density, and thus becomes “1” by binarization. The subject area Sa 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. Further, the boundary between the two regions (the subject region Sa and the blank region Sb) obtained by binarization is recognized as the skin line SL.

(2) On the other hand, when the shooting direction is MLO, the pectoralis muscle is reflected 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 area Sa1 is classified into a high density area Da, an intermediate density area Db, and a low density area Dc with reference to the density of the pectoral muscle area 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 of the threshold is Da. By doing so, it can classify | categorize into each area | region Da-Dc.

Next, the abnormal shadow candidate detection process will be described.
The abnormal shadow candidate detection unit 27 executes detection processing according to an algorithm abnormal shadow candidate 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 the breast image based on such density characteristics, and each area (this 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. Extract abnormal shadow candidates. Multivariate analysis can be applied as an extraction method. For example, a multivariate analysis is constructed using, as learning data, an image feature amount calculated from a shadow that is known to be true positive in advance (hereinafter, this multivariate analysis is referred to as a first multivariate analysis). Various image feature quantities calculated from the shadow candidates first detected in the multivariate analysis are input to obtain an index value indicating the possibility of an abnormal shadow. This index value is compared with a threshold value prepared in advance for determining whether or not it is an abnormal shadow candidate (hereinafter, this threshold value is referred to as a first threshold value), and a candidate area that exceeds the first threshold value is determined. A region that is clustered within a predetermined area is detected as an abnormal shadow candidate for the microcalcification cluster shadow.

FIG. 4 shows the internal configuration of the viewer 5.
The viewer 5 includes a control unit 51, an operation unit 52, a display unit 53, a communication unit 54, and a storage unit 55, and each unit is connected by a bus 56.

  The control unit 51 includes a CPU, a RAM, and the like. The CPU reads various control programs from the storage unit 55 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, medical image display control processing (described later) is executed in accordance with the medical image display control program according to the present invention.

  The operation unit 52 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 51.

  The display unit 53 includes an LCD or the like, and displays various screens in accordance with instructions from the control unit 51.

  The communication unit 54 includes a communication interface such as a LAN card and communicates with an external device on the communication network N according to an instruction from the control unit 51. For example, image data of a medical image from the image server 4 is received.

  The storage unit 55 stores various programs, parameters necessary for executing the programs, or processing result data. In the present embodiment, the storage unit 55 serves as a storage unit and stores a display priority order setting file 551 that stores information on display priority order set for each interpretation doctor in association with an interpretation doctor ID for identifying the interpretation doctor. (See FIG. 10).

Next, the operation of the medical image diagnosis support 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). As related information of the generated breast image, information on the patient such as the patient's name, age, and sex, imaging information such as tube voltage at the time of imaging, breast compression rate, examination information such as the 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, necessary image processing is performed on the breast image, while the abnormal shadow candidate detection and classification process described below is started for the breast image, and detection of abnormal shadow candidates and detection of detected abnormal shadow candidates are performed. Classification is performed.

Hereinafter, the abnormal shadow candidate detection / classification process executed by the image processing apparatus 2 will be described with reference to FIG.
In the abnormal shadow candidate detection / classification process shown in FIG. 5, the abnormal shadow candidate detection unit 27 first performs subject region extraction and classification processing to extract a subject region included in the breast image, and the subject region is within the subject region. They are classified into areas such as pectoral muscle area and breast area (see FIG. 3) (step S1).

  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 each of the first detected abnormal shadow candidate regions. At least the area, standard deviation, density average value, curvature, fractal dimension, circularity, contrast with the surrounding area, marginal complexity, and abnormal shadow candidates are first detected for the image feature amount. The classified area (which of the areas Da to Dc shown in FIG. 3 corresponds) is calculated (step S22).

The contrast is the density difference between the abnormal shadow candidate area and the 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. Obtained from distance (number of pixels).
Further, the circularity is calculated by the following equation [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 equation (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.

  These image features can then be input to a first multivariate analysis constructed in advance to output an index value indicating the likelihood of being an abnormal shadow, and each of the first detected candidate regions may be an abnormal shadow An index value indicating the sex is calculated (step S23). Next, based on the comparison between the calculated index value and the first threshold value, it is determined whether or not each candidate area is an abnormal shadow candidate, thereby detecting an abnormal shadow candidate (step S24). Position information indicating the detection position of the shadow candidate region is stored in the RAM as abnormal shadow candidate detection information (step S25), and the process proceeds to step S3 in FIG.

  In step S3 shown in FIG. 5, the control unit 21 performs the second multivariate analysis on each abnormal shadow candidate extracted in the abnormal shadow candidate detection process in step S2 based on the image feature amount calculated in step S22. The index value indicating the degree of difficulty (degree of difficulty) in determining whether each abnormal shadow candidate is true positive / false positive is calculated (step S3: index value calculating means).

  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 interpretation doctor. Moreover, it is difficult to say that the visibility of the area is good, and the shadow is easily overlooked by the interpretation 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 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 or not it is an abnormal shadow. 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, the low density region Dc has a low density as a whole, so that the abnormal shadow appearing on the image at a low density is lower than that in the high density region Da. Is more difficult to distinguish. Similarly, regarding the density of the abnormal shadow candidate area itself, it is difficult to visually recognize if the density is low, and it is easy for the interpretation doctor to miss. Furthermore, the abnormal shadow existing in the breast region near the edge of the image is easily overlooked by the interpretation doctor from that position.

  Therefore, considering the contrast, area, average density value, classification area where the abnormal shadow candidate is located, the distance from the edge of the image, the complexity of the edge, etc. An index value indicating the degree of difficulty in determining false positive is calculated.

  Specifically, it is difficult to determine true positive / false positive by preparing learning data in advance, such as the feature quantity of shadow that should be increased in difficulty and the characteristic quantity of shadow that should be decreased. A second multivariate analysis such as a neural network that outputs 0 to 1 as an index value indicating degree (indicating that the degree of difficulty is lower as it is closer to 0 and that the degree of difficulty is higher as it is closer to 1) is constructed. For example, as the contrast is small, the area is small, the density average value of the candidate area is small, the circularity is small, the distance from the image edge is short, or the margin complexity is small, the true / false positive determination Learning data is prepared so that the degree of difficulty becomes an index value, and the second multivariate analysis is adjusted. Further, in the breast region, it is difficult to determine abnormal shadow candidates in the order of the regions Dc, Db, and Da. Therefore, the second multi-value is set so that the degree of difficulty in determining true positive / false positive becomes higher in this order. Build a random analysis.

  Then, the feature value of each abnormal shadow candidate is input to the second multivariate analysis, and the index value is calculated by the second multivariate analysis.

  When an index value indicating the degree of difficulty in true / false positive determination is calculated for each abnormal shadow candidate, the calculated index value is compared with a second threshold value prepared in advance, and is below the second threshold value. The abnormal shadow candidate is detected as an easy-to-determine abnormal shadow candidate (step S4). Of the abnormal shadow candidate detection information detected based on the first threshold value stored in the RAM in step S25, the second abnormal shadow candidate is detected in step S4. A flag indicating that the abnormal shadow candidate is detected based on the second threshold value is added to the detection information (position information) of the abnormal shadow candidate further detected based on the threshold value (step S5). And the detection information of an abnormal shadow candidate is matched and transmitted to the image server 4 (step S6), and this process ends.

  When the image server 4 receives the image data of the medical image and the detection information of the abnormal shadow candidate in the medical image from the image processing device 2, the image server 4 assigns an image ID for identifying the received image data. Image data and detection information are associated with each other by the assigned image ID and stored in the image DB 4a. Further, the image server 4 transmits the requested image data and corresponding detection information to the viewer 5 in response to an image transmission request from the viewer 5.

  When displaying the medical image on the display unit 53, the viewer 5 adds annotations indicating the detection positions of all the abnormal shadow candidates detected based on the first threshold value in the abnormal shadow candidate detection and classification process described above on the medical image. A first display mode for superimposing display, a second display mode for displaying an annotation indicating a detection position of an abnormal shadow candidate detected based on the second threshold on the medical image, and a medical image A third display mode in which an annotation indicating a detection position of an abnormal shadow candidate detected by the first threshold but not detected by the second threshold is superimposed, and no annotation is displayed on the medical image. Has a display mode.

  That is, in the second display mode, only the abnormal shadow candidates that are easy to determine true / false positive among the abnormal shadow candidates detected by the abnormal shadow candidate detection unit 27 are displayed as annotations. In the third display mode, only abnormal shadow candidates for which it is difficult to determine true positive / false positive among abnormal shadow candidates detected by the abnormal shadow candidate detecting means 27 are displayed as annotations.

  FIG. 8A shows a display example when a breast image is displayed in the first display mode, and FIG. 8B shows a display example when a breast image is displayed in the second display mode. FIG. 8C shows a display example when the breast image is displayed in the third mode, and FIG. 8D shows a display example when the breast image is displayed in the non-display mode.

  For example, for a veteran interpreting doctor who has a lot of interpretation experience and a high level of interpretation diagnosis technology, if all the abnormal shadow candidate annotations are displayed on the medical image, it can be clearly judged as true positive visually. Since annotations are displayed up to the candidates, the number of indications is increased, candidates that are low in visibility and easy to overlook, and candidates that are difficult to determine true / false positives and that should be left to the final decision of the interpretation doctor are concentrated. However, the interpretation efficiency deteriorates when interpreting. In such a case, first, only abnormal shadow candidates for which it is difficult to determine true positive / false positive in the third display mode are indicated by annotation, and only those candidates that can be clearly determined to be true positive are visually only visible. Then, for the candidates with high difficulty in determining true / false positives pointed out by the annotation, the interpretation is performed according to the indication of the annotation, and then the display of all abnormal shadow candidates is displayed by switching to the first display mode. Thus, if the diagnosis result is confirmed, an efficient interpretation diagnosis can be performed.

  On the other hand, when an interpreting doctor with little experience of interpretation interprets images, it is more efficient to display all abnormal shadow candidates in the first display mode and perform interpretation according to the indication of the annotation. It can be performed. Thereafter, if the display is performed in the second display mode or the third display mode, it is possible to confirm the appearance of each of the true positive candidates that can be clearly determined visually and the candidates that are difficult to determine.

  In this way, when displaying a medical image, the display mode in which the interpretation can be performed efficiently depends on the interpretation doctor. Therefore, the viewer 5 can set display priorities of the first to third display modes and the non-display mode for each interpretation doctor.

  In order to set the display priority order, the display priority order setting screen 531 is displayed on the display unit 53 by the operation of the operation unit 52, and the display priority for each interpretation doctor is operated by operating the operation unit 52 from the display priority order setting screen 531. Ranking can be set. Thereby, a setting means is realized.

  FIG. 9 shows an example of the display priority order setting screen 531. As shown in FIG. 9, the display priority setting screen 531 includes an input field 531a for inputting an interpreting doctor ID (doctor ID) for uniquely identifying an interpreting doctor, and priority (display priority) 1 An input field 531b for inputting a mode of 3, a input field 531c for inputting a mode of priority 2, an input field 531d for inputting a mode of priority 3, and a mode of priority 4 are input. Input field 531e. When the interpretation unit ID is input to the input field 531a by the operation unit 52, the mode is input to the input fields 531b to 531e, and the setting button 531f is pressed, the information input by the control unit 51 is displayed with priority as shown in FIG. It is stored in the rank setting file 551.

When the medical image is displayed on the display unit 53, display control is performed based on the setting of the display priority order setting file 551.
FIG. 11 is a flowchart showing a medical image display control process executed by the control unit 51 of the viewer 5. This process is executed when a medical image display instruction is input from the operation unit 52, and is based on cooperation between the CPU of the control unit 51 and the medical image display control processing program stored in the storage unit 55. It is executed by software processing, and display control means is realized by executing this processing. Hereinafter, the medical image display control process will be described with reference to FIG.

  First, an interpretation doctor ID input screen is displayed on the display unit 53. When an interpretation doctor ID is input from this interpretation doctor ID input screen (not shown) (step S31), the image server 4 is connected via the communication unit 54. A list of medical images is acquired from, and a list screen of the acquired medical images is displayed on the display unit 53 (step S32). When a medical image to be displayed is selected from the medical image list screen (step S33), a transmission request for image data of the selected medical image is made to the image server 4 via the communication unit 54 and selected. The image data and detection information of the medical image obtained are acquired from the image server 4 (step S34).

  Next, the display priority setting file 551 is referred to, the display priority corresponding to the interpretation doctor ID input in step S31 is referred to (step S35), and the display unit 53 displays the medical image in the priority order 1 mode. Display is performed (step S36).

  When displaying in the first display mode, the detection information (detection position) for all abnormal shadow candidate detection information detected in the medical image is displayed on the medical image acquired from the image server 4. The annotation shown is superimposed and displayed. When displaying in the second display mode, an annotation indicating detection information associated with a flag among detection information detected in the medical image is superimposed and displayed on the medical image. The third display mode includes a batch display mode for collectively displaying the abnormal shadow candidates to be displayed and a sequential display mode for sequentially displaying the abnormal shadow candidates to be displayed individually according to the cursor key operation of the operation unit 52. The two modes are switched and displayed according to a “switch” button (not shown) displayed in the third display mode. In the batch display mode, an annotation indicating detection information that is not associated with a flag among detection information is superimposed and displayed on the medical image. In the sequential display mode, detection information not associated with a flag is displayed on the medical image, an annotation indicating one of the detection information is displayed, and the cursor key (right arrow key) of the operation unit 52 is pressed. Each time, the detection information to be displayed is changed and displayed in annotation. As a result, in the third display mode, it is possible to provide detection information of abnormal shadow candidates that are more easily viewable to the interpretation doctor and are difficult to determine true positive or false positive. When displaying in the non-display mode, the medical image is displayed on the display screen.

  The display in the priority level 1 mode is performed by operating the operation unit 52, and a “next mode” button (not shown; the priority level 2 display mode is set) displayed on the display screen displayed in the priority level 1 mode. (Displayed only when it is set) or until an “end” button (not shown) is pressed (step S37; NO, step S38; NO).

  When the “end” button is pressed by the operation of the operation unit 52 from the display screen displayed in the priority level 1 mode (step S38; YES), this process ends. When the “next mode” button is pressed by operating the operation unit 52 from the display screen displayed in the priority order 1 mode (step S37; YES), the medical image of the medical image in the priority order 2 mode is displayed on the display unit 53. Display is performed (step S39). This display is displayed only when a “next mode” button (not shown; priority 3 mode is set) displayed on the display screen of the priority 2 mode by the operation of the operation unit 52. ) Or until an “end” button (not shown) is pressed (step S40; NO, step S41; NO).

  When the “end” button is pressed by the operation of the operation unit 52 from the display screen displayed in the priority order 2 mode (step S41; YES), this process ends. When the “next mode” button is pressed by operating the operation unit 52 from the display screen displayed in the priority order 2 mode (step S40; YES), the medical image of the medical image in the priority order 3 mode is displayed on the display unit 53. Display is performed (step S42). This display is displayed only when the “next mode” button (not shown; priority 4 mode is set) displayed on the display screen of the priority 3 mode by the operation of the operation unit 52. ) Or until an “end” button (not shown) is pressed (step S43; NO, step S44; NO).

  When the “end” button is pressed by the operation of the operation unit 52 from the display screen displayed in the priority order 3 mode (step S44; YES), this process ends. When the “next mode” button is pressed by operating the operation unit 52 from the display screen displayed in the priority order 3 mode (step S43; YES), the medical image of the medical image in the priority order 4 mode is displayed on the display unit 53. Display is performed (step S45). This display is continued until the “end” button is pressed by the operation of the operation unit 52 (step S46; NO). When the “end” button is pressed by the operation of the operation unit 52 (step S46; YES), this process ends.

  When displaying in the first to third display modes, as shown in FIGS. 8A to 8C, the abnormal shadow candidates detected only by the first threshold and the second threshold are detected. It is preferable to display annotations having different shapes and / or colors from the abnormal shadow candidates. Since the abnormal shadow candidate detected only with the first threshold and the abnormal shadow candidate detected with the second threshold are displayed with different annotations, the interpretation doctor displays them together so that they can be identified. The doctor can easily recognize which of the displayed abnormal shadow candidates is an abnormal shadow candidate that is difficult to determine true-positive / false-positive and requires careful observation. Can be performed.

  As described above, according to the medical image diagnosis support system 100, the viewer 5 is detected on the medical image based on the first threshold value on the medical image when displaying the medical image on the display unit 53. A first display mode in which annotations indicating the positions of all abnormal shadow candidates are superimposed and displayed, and annotations indicating the positions of abnormal shadow candidates detected based on the second threshold are superimposed and displayed on the medical image. A second display mode, and a third display mode in which an annotation indicating the position of an abnormal shadow candidate detected on the medical image at the first threshold but not detected on the second threshold is superimposed and displayed, The display priority of these modes, which has a non-display mode in which no annotation is displayed on the medical image and is stored for each interpretation doctor in the display priority setting file 551 Based on, it performs display control.

  Therefore, detection information of abnormal shadow candidates on a medical image can be provided in an easy-to-view display according to the interpretation experience of the interpretation doctor, and the interpretation work of the interpretation doctor can be made more efficient.

In addition, embodiment mentioned above is a suitable example to which this invention is applied, and is not limited to this.
For example, in the above description, the abnormal shadow candidate detection unit is realized in the image processing apparatus 2 and the display unit and the display control unit are realized in the viewer 5. Each means may be realized by a component device.

  In the above embodiment, image data and detection information are transmitted from the image processing apparatus 2 to the image server 4, and the viewer 5 sends a medical image data transmission request to the image server 4 to display a medical image. However, the image data and the detection information may be transmitted directly from the image processing apparatus 2 to the viewer 5.

  In the above embodiment, the first to third display modes for displaying the annotation on the medical image and the non-display mode for not displaying the annotation on the medical image are provided. The display priority between the four modes of the mode and the non-display mode can be set, and the medical image is displayed based on the set display priority, but the display priority between the first to third display modes is set. The order can be set, and when the display of a medical image is instructed from the operation unit, first, the selected medical image is displayed without annotation, and set when the annotation display is instructed from the operation unit. You may make it display in the 1st-3rd display mode according to a display priority.

  In addition, the detailed configuration and detailed operation of each apparatus constituting the medical image diagnosis support system 100 can be changed as appropriate without departing from the spirit of the present invention.

It is a figure which shows the system configuration | structure of the medical image diagnosis assistance system 100 in this embodiment. It is a block diagram which shows the internal structure of the image processing apparatus 2 of FIG. It is a figure which shows each area | region classified in the breast image S. FIG. It is a block diagram which shows the internal structure of the viewer 5 of FIG. 6 is a flowchart for explaining abnormal shadow candidate detection / classification processing executed by the image processing apparatus 2; 6 is a flowchart for explaining abnormal shadow candidate detection processing executed by the image processing apparatus 2; It is a figure explaining the calculation method of the feature-value of edge complexity. (A) is a diagram showing a display example of detection information of abnormal shadow candidates in the first display mode, (b) is a diagram showing a display example of detection information of abnormal shadow candidates in the second display mode, (c) (A) is a figure which shows the example of a display of the detection information of the abnormal shadow candidate in 3rd display mode, (d) is a figure which shows the example of a display of the detection information of the abnormal shadow candidate in non-display mode. FIG. 11 is a diagram showing an example of a display priority order setting screen 531. 6 is a diagram illustrating an example of data storage in a display priority setting file 551. FIG. 10 is a flowchart for explaining medical image display control processing executed in a viewer 5.

Explanation of symbols

100 Medical Image Diagnosis Support System 1 Image Generation Device 2 Image Processing Device 3 Printer 4 Image Server 4a Image DB
5 Viewer 21, 51 Control unit 22, 52 Operation unit 23, 53 Display unit 24, 54 Communication unit 25, 55 Storage unit 551 Display priority setting file 26 Image processing unit 27 Abnormal shadow candidate detection unit 28, 56 Bus N Communication network

Claims (5)

In a medical image diagnosis support system comprising: an abnormal shadow candidate detecting means for analyzing a medical image and detecting an abnormal shadow candidate; and a display means for displaying detection information of the abnormal shadow candidate detected by the abnormal shadow candidate detecting means. ,
The abnormal shadow candidate detecting means detects an abnormal shadow candidate from the medical image using a first threshold value and a second threshold value,
The display means includes at least a first display mode for displaying detection information of an abnormal shadow candidate detected by the first threshold in the abnormal shadow candidate detection means, and an abnormal shadow candidate detected by the second threshold. A second display mode for displaying the detection information, and a third display mode for displaying the detection information of the abnormal shadow candidate detected by the first threshold and not detected by the second threshold,
Setting means for setting display priorities of at least the first to third display modes;
Display control means for controlling display of the abnormal shadow candidate detection information on the display means based on the setting from the setting means;
A medical image diagnosis support system comprising:   The abnormal shadow candidate detection means includes index value calculation means for calculating an index value indicating a degree of difficulty in determining true positive and false positive for each of the abnormal shadow candidates detected by the first threshold. 2. The medical image according to claim 1, wherein detection is further performed based on the second threshold value and the index value calculated by the index value calculation unit from among abnormal shadow candidates detected based on the threshold value of 1. Diagnosis support system. The setting means is capable of setting the display priority for each interpretation doctor,
The medical image diagnosis support system according to claim 1, further comprising a storage unit that stores the display priority order set for each interpretation doctor by the setting unit.   The said 3rd display mode further has the mode which displays separately the detection information of the abnormal shadow candidate of a display object collectively, and the mode which displays separately separately, The any one of Claims 1-3 characterized by the above-mentioned. The medical image diagnosis support system described in 1. The detection information is displayed by an annotation indicating a detection position on the medical image of the abnormal shadow candidate detected from the medical image,
The display control unit differs between a detection position of the abnormal shadow candidate detected by the second threshold value and a detection position of the abnormal shadow candidate detected by the first threshold value and not detected by the second threshold value. The medical image diagnosis support system according to any one of claims 1 to 4, wherein the medical image diagnosis support system displays an annotation.

Images