JP2010227215A - Medical image processor, method of creating schema and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a schema corresponding to a patient's individual difference of a patient. <P>SOLUTION: A medical image processor includes a control means for extracting a component of an object contained in a medical image (step S1), and drawing a contour of the extracted component corresponding to the position of the extracted component in the medical image (step S2) to create a schema. The control means draws an auxiliary graphic form on the schema based on the position of the extracted component in the medical image. The medical image is a breast image of a breast as an object. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a medical image processing apparatus, a schema creation method, and a program.

  In general, an image diagnosis report created by a doctor uses a schematic diagram of a subject called a schema. The schema can convey the interpretation results to the report reader in a more understandable manner.

  When writing a report, one standard schema is used regardless of the patient, but there are individual differences in the size and position of the subject depending on the patient, and only one schema reflects the individual difference. I couldn't. Therefore, a method for creating a schema for each subject using a computer has been proposed. For example, as a schema, there are a method of creating an image obtained by reducing a captured image as it is, and a method of using an image that has been subjected to simple image processing such as edge detection so as to be easily recognized. However, these methods are difficult to use because various structures are included in the schema.

  On the other hand, a method has been disclosed in which a large number of rib model images including rib model shapes are created from a large number of chest images that have been taken in the past and used as a template for a schema (for example, Patent Documents). 1). According to this method, a template closest to the rib model shape of a medical image to be interpreted is searched for from among schema templates created from past captured images, and used as a schema.

  In addition, a method for preparing a standard schema template in which a plurality of points are set and deforming the template according to a point operation by a doctor to create a schema is also disclosed (see, for example, Patent Document 2). .

JP 2006-181146 A JP 2007-7378 A

  However, in the case of the method of Patent Document 1, many templates must be prepared in order to cope with individual differences among patients. Without a template that represents the rib model shape of a medical image to be interpreted, a schema corresponding to the patient cannot be provided.

  Further, according to the method of Patent Document 2, a schema template can be deformed into a desired shape, so that differences for each patient can be expressed in detail in the schema. However, it is cumbersome to create a schema manually by hand when creating a report. The burden on the doctor is heavy and inefficient.

  An object of the present invention is to provide a schema corresponding to individual differences among patients.

According to the invention of claim 1,
Comprising a control means for extracting a constituent part of a subject included in a medical image;
The control means is provided with a medical image processing apparatus that draws a contour of the extracted component part in accordance with a position of the extracted component part in the medical image and creates a schema.

According to invention of Claim 2,
The medical image processing apparatus according to claim 1, wherein the control unit draws an auxiliary graphic on the schema based on a position of the extracted component in the medical image.

According to invention of Claim 3,
The medical image is a breast image whose subject is a breast,
The control means extracts breast skinline, pectoral muscle, and nipple components included in the breast image, and draws the extracted skinline, pectoral muscle contour, and nipple contour to create a schema. A medical image processing apparatus according to claim 2 is provided.

According to invention of Claim 4,
The control means obtains the center of the nipple in the extracted nipple region, and based on the position of the nipple center and the extracted skin line, a boundary line for classifying the breast into upper, middle, and lower regions The medical image processing apparatus according to claim 3, wherein the image is drawn on the schema as an auxiliary figure.

According to the invention of claim 5,
The control means obtains a nipple center in the extracted nipple area, and draws a boundary line of a region within a certain distance from the nipple center as an auxiliary figure on the schema based on the position of the nipple center. A medical image processing apparatus according to 3 or 4 is provided.

According to the invention of claim 6,
A method of creating a schema by a medical image processing apparatus,
Extracting a constituent part of a subject included in the medical image by the control means;
Drawing a contour of the extracted component part according to a position of the extracted component part in the medical image by the control means, and creating a schema;
Is provided.

According to the invention of claim 7,
Computer
Extracting a constituent part of a subject included in a medical image, drawing an outline of the extracted constituent part according to the position of the extracted constituent part in the medical image, and functioning as a control unit for creating a schema A program is provided.

  According to the present invention, since a constituent part of a subject extracted from a medical image is used as a schema, it is possible to provide a schema corresponding to details of individual differences among patients. In addition, since only the contours of the constituent parts of the subject are drawn, an image of a tissue part unnecessary for the schema can be excluded, and a schema that can be easily used as a means for transmitting an interpretation result can be provided.

It is a figure which shows the functional structure of the medical image processing apparatus in this Embodiment. It is an example of a report creation screen. It is a flowchart of a schema creation process executed by the medical image processing apparatus. It is a flowchart which shows the process at the time of extracting each component. It is a figure which shows the area | region of the breast, the skin line, the pectoral muscle area | region, and the pectoral muscle line which are included in a breast image. It is a figure which shows the search line set in order to determine the search start point of a pectoral muscle line. It is a figure which shows the search line set in order to determine a pectoral muscle line. It is a figure which shows the secondary differential value calculated | required about the pixel on a skin line. It is an example of the created schema. It is a figure which shows the boundary line of the upper part of the breast calculated | required from a breast image, a center part, a lower part, and the boundary line of the area | region within a predetermined distance from the center of a nipple. 9 shows an example in which an auxiliary graphic is drawn on the schema shown in FIG.

  Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 shows a functional configuration of a medical image processing apparatus 1 in the present embodiment.
The medical image processing apparatus 1 is incorporated in, for example, a PACS (Picture Archiving and Communication System), and is used by an interpreting doctor to display a medical image for interpretation and create a report.
As shown in FIG. 1, the medical image processing apparatus 1 includes a control unit 11, an operation unit 12, a display unit 13, a communication unit 14, and a storage unit 15.

  The control unit 11 is a control unit that performs various calculations in cooperation with a program stored in the storage unit 15, performs centralized control of the operation of each unit of the medical image processing apparatus 1, and executes various processes. The control unit 11 can be configured to include a CPU (Central Processing Unit) and a RAM (Random Access Memory).

  For example, in the schema creation process described later, the control unit 11 extracts a constituent part of the subject from the medical image. A subject refers to an imaging region of a patient. The control unit 11 draws the outline of the extracted component part according to the position of the extracted component part in the medical image, and creates a schema. Further, the control unit 11 draws an auxiliary graphic on the created schema based on the position of the extracted component in the medical image. Auxiliary figures include symbols such as lines and circles that help grasp the schema.

The operation unit 12 includes a keyboard and a mouse, generates operation signals according to these operations, and outputs them to the control unit 11.
The display unit 13 includes a display, and displays various operation screens and medical images according to display control of the control unit 11. For example, the display unit 13 displays a viewer screen and displays a medical image to be interpreted on the viewer screen. The display unit 13 displays a report creation screen, and displays the schema created by the control unit 11 on the report screen.

  The communication unit 14 includes a communication interface and communicates with an external device on the network. For example, the communication unit 14 receives a medical image to be interpreted from a server that distributes the medical image.

  The storage unit 15 stores a program executed by the control unit 11 and files and data used for executing the program. A hard disk can be used as the storage unit 15.

Next, the operation of the medical image processing apparatus 1 will be described.
When a medical image to be interpreted is distributed from the server to the medical image processing apparatus 1, a viewer screen is displayed on the display unit 13, and the distributed medical image is displayed on the viewer screen. Hereinafter, an example in which a breast image with a breast as a subject is distributed and displayed will be described.
When the distributed medical image is a breast image, the breast images of the left and right breasts are displayed side by side for comparative interpretation. The imaging directions of the left and right breast images are the same. The doctor interprets the left and right breast images displayed on the viewer screen.

  After interpretation, a report creation screen is displayed on the display unit 13. FIG. 2 shows an example of a report creation screen. As shown in FIG. 2, the report creation screen displays the schema d1 and the input area d2 of the findings, so that the doctor performs the input operation of the findings on the report creation screen or the input operation of the comment attached to the schema d1. Can go. The schema d1 displayed on the report creation screen is created when the medical image processing apparatus 1 executes the schema creation process.

With reference to FIG. 3, the schema creation process executed by the medical image processing apparatus 1 will be described.
As shown in FIG. 3, in the schema creation process, the control unit 11 extracts the skin line, pectoral muscle, and nipple component parts constituting the breast from the breast image (step S <b> 1). Although the extraction method will be described in detail with reference to FIG. 4, the extraction method is an example, and any extraction method may be adopted as long as the constituent parts can be extracted.

  As shown in FIG. 4, the control unit 11 performs filter processing on the breast image and executes edge enhancement processing (step S11). For the filtering process, a Prewitt filter, a Sobel filter, or the like can be used. The breast image after the edge enhancement processing is a value in which the pixel value of each pixel indicates the edge strength.

Next, the control unit 11 extracts a skin line using the breast image in which the edge is enhanced by the edge enhancement process (step S12). The skin line is the outline of the breast.
FIG. 5 shows an example of a breast image. In FIG. 5, the X-axis is set in the direction from the nipple to the chest wall, the Y-axis is set in the direction perpendicular to the X-axis, and the position of each pixel of the breast image is indicated by coordinates (X, Y). Further, the pixel value of the pixel at the position of the coordinate (X, Y) is indicated by V (X, Y). An image end in the X-axis direction is indicated by Xmax, and an image end in the Y-axis direction is indicated by Ymax.

  The control unit 11 scans each pixel in the Y-axis direction at each X coordinate (0 to Xmax) of the breast image shown in FIG. 5, and the coordinate S (X of the pixel at which the pixel value V (X, Y) is the maximum. , Y). As shown in FIG. 5, the extracted coordinates S (X, Y) are the boundary points between the breast area Sa, which is the subject, and the non-mammal area Sb irradiated directly with X-rays without passing through the subject. is there. By connecting the coordinates S (X, Y), the outline of the breast, that is, the skin line SL can be extracted.

Next, the control unit 11 determines a pectoral muscle line search start point in the breast image (step S13). The pectoral muscle line is a boundary line between the breast region Sa and the pectoral muscle, and is an outline of the pectoral muscle.
As shown in FIG. 6, the control unit 11 is a coordinate several pixels below the coordinate S (0, Y) (the intersection of the image end where the X coordinate is 0 and the skin line SL) at which V (0, Y) is maximum. The pixel A is set as the reference point. The control unit 11 sets search lines la0 to la30 every 1 ° within an angle range of 0 ° to −30 ° with respect to the X axis with the reference point as the center. The length of the search lines la0 to la30 is 1/5 of the width of the breast image in the X-axis direction. The control unit 11 calculates the average value of the pixel values on the search lines la0 to la30 for each search line la0 to la30.

The control unit 11 repeats the process of calculating the average value while shifting the reference point pixel by pixel in the Y-axis direction starting from the coordinate A. The control unit 11 determines the reference point of the search line that maximizes the average value calculated at each reference point as the pectoral muscle line search start point B.
Note that when the pectoral muscle line search start point B is close to the lower end of the image (image end of Y = Ymax), for example, at a position within 10 pixels from the lower end, the control unit 11 has no pectoral muscle region in the breast image. Judge. Also, when the average value calculated at each reference point is smaller than a predetermined threshold, for example, 300, the control unit 11 determines that there is no pectoral muscle region. If it is determined that there is no pectoral muscle area, an error has occurred and the process is terminated.

When the search start point B is determined, the control unit 11 searches for the pectoral muscle line (step S14).
FIG. 7 is an enlarged view of the vicinity of the pectoral muscle of the breast image shown in FIG. As shown in FIG. 7, the control unit 11 sets search lines lb0 to lb18 every 1 ° within a range of an angle ± 9 ° with respect to the X axis, starting from the search start point B of the pectoral muscle line. The length of the search lines la0 to la18 is 1/5 of the width of the breast image in the X-axis direction. The control unit 11 calculates the average value of the pixel values on the search lines lb0 to lb18 for each search line lb0 to lb18, and sets the search line lbn (n = 0,..., 18) having the maximum average value as the pectoral muscle line. Determined as PL.

  Next, the control unit 11 sets a point that has moved from the search start point B by 1/10 of the width in the X-axis direction of the breast image as a new search point on the determined pectoral muscle line by the same processing as described above. Determine the muscle line. In this way, every time the pectoral muscle line is determined, the process of setting a new search point from the previous search point and determining the pectoral line is repeated until the end of the image is reached. As shown in FIG. 7, finally, a pectoral muscle line PL obtained by connecting the pectoral muscle lines determined at each search point is obtained.

  In addition, when the maximum average value calculated | required in each search point is smaller than a predetermined threshold value, for example, 300, the control part 11 judges that it is an unclear pectoral muscle line. When there are two or more search points where the maximum average value of the search lines lb0 to lb18 is larger than the threshold value, the control unit 11 draws a quadratic approximate curve from these search points, and determines this approximate curve as a pectoral muscle line. When the search point where the maximum average value of the search lines lb0 to lb18 is greater than or equal to the threshold is 1 or less, the control unit 11 shifts the search start point B by one pixel in the X-axis direction and executes the above processing.

  When the pectoral muscle line PL is determined, as shown in FIG. 5, the control unit 11 determines a region surrounded by the determined pectoral muscle line PL and the image end (the image end on the chest wall side) as a pectoral muscle region. Extracted as M1 (step S15).

Next, the control unit 11 extracts a nipple region (step S16).
The control unit 11 calculates a second derivative value with respect to the pixel values of the pixels on the skin line SL. As shown in FIG. 8, the secondary differential value is a substantially constant value except for the nipple, but rapidly changes at the boundary point of the nipple region. The control unit 11 extracts the area surrounded by the straight line connecting the inflection points Q1 and Q2 of the secondary differential value and the skin line SL as the nipple area Sc.

  The control unit 11 extracts the above-described components from the left and right breast images. Thereafter, as shown in FIG. 3, the control unit 11 draws a skin line (breast contour), a pectoral muscle contour (thoracic muscle line), and a nipple contour to create a schema (step S2). Drawing is performed so as to correspond to the position of each component in the original breast image. FIG. 9 shows an example of the created schema. As shown in FIG. 9, the contours of the skin line SL, the pectoral muscle line PL (the contour of the pectoral muscle region M1), and the contour of the nipple region Sc, which are respectively extracted from the left and right breast images, are drawn.

Next, the control unit 11 obtains the center of the nipple in the nipple area Sc, and based on the position of the nipple center and the skin line SL, the boundary line for classifying the breast area Sa into the upper, middle, and lower areas. Then, it is drawn on the created schema (step S3). The boundary line is an auxiliary figure drawn so that each region can be easily identified, and there are three boundary lines: an upper end, a center, and a lower end.
As shown in FIG. 10, first, the control unit 11 obtains, as the nipple center Q3, the pixel having the shortest distance between the inflection points Q1 and Q2 among the pixels on the skin line SL in the nipple region Sc. . The control unit 11 draws a line parallel to the Y axis passing through the nipple center Q3 as a center boundary line LM.

Next, the control unit 11 detects a portion where the shape of the skin line SL of the breast image is a concave shape. For example, the curvature of an arc (an arc in which the chest wall side is recessed) with respect to the skin line SL is calculated, and a portion where the curvature is maximum or a portion where the curvature is larger than a threshold (for example, 1/1000 (mm)) is defined as a concave shape. What is necessary is just to detect. The curvature is represented by the reciprocal of the radius of the arc, and the larger the curvature (the smaller the radius of the arc), the larger the curve of the skin line SL. The control unit 11 draws a line parallel to the Y axis passing through the pixel corresponding to the concave bottom portion as the lower boundary line LL. When the concave portion is not detected, the image end where X = Xmax may be set as the lower end of the breast.
Next, the control unit 11 draws the upper boundary line LU above the central boundary line LM at a position moved by the distance between the central boundary line LM and the lower boundary line LL.

  Next, based on the position of the nipple center Q3, the control unit 11 draws a boundary line of a region within a predetermined distance, for example, 2 cm from the nipple center Q3 on the created schema (step 4). The control unit 11 draws an arc having a radius of 2 cm from the nipple center Q3 as a boundary line CL in the breast region Sa.

  Finally, the control unit 11 draws on the created schema a symbol indicating whether the image is the left or right breast image, a symbol indicating the upper, middle, and lower regions of the breast, and a symbol indicating the pectoral muscle region. And complete the schema.

FIG. 11 shows an example of a schema on which auxiliary graphics are drawn. In the schema shown in FIG. 11, each boundary line and symbol are drawn on the schema shown in FIG. In FIG. 11, the symbol R indicates the right breast, and the symbol L indicates the left breast. The symbol X represents the pectoral muscle region, the symbol U represents the upper part of the breast region, the symbol M represents the center of the breast, and the symbol L represents the lower part of the breast. The symbol S indicates an area within 2 cm from the center of the nipple.
The created schema is displayed on the report creation screen as described above and used for the report.

  As described above, according to the present embodiment, the control unit 11 extracts each component part of the breast skinline, pectoral muscles, and nipples included in the breast image, and positions each extracted component part in the breast image. In response, the outline of the extracted component is drawn to create a schema. Since the component extracted from the breast image is drawn, it is possible to create a schema corresponding in detail to individual differences among patients. The doctor can save the trouble of creating the schema and can improve the efficiency of the interpretation work. Further, since only the components of the breast necessary for the schema are drawn, it is possible to provide a schema that can be easily used as a means for transmitting the interpretation result.

  Further, the control unit 11 draws an auxiliary graphic on the schema based on the extracted positions of the constituent parts in the breast image. For example, the center of the nipple is obtained in the nipple area, and based on the position of the nipple center and the extracted skin line, a boundary line for classifying the breast into the upper, middle, and lower areas is drawn as an auxiliary figure. Further, the center of the nipple is obtained in the nipple area, and based on the position of the nipple center, the boundary line of the area within a certain distance from the nipple center is drawn on the schema as an auxiliary figure. Accordingly, it is possible to provide a schema that can easily grasp each region of the breast and that can be easily used as a means for transmitting an interpretation result.

In addition, the above-mentioned embodiment is a suitable example of this invention, and is not limited to this.
For example, in the above-described embodiment, a schema is created for a breast image whose photographing direction is the oblique direction (MLO). Therefore, the constituent parts of the breast included in the breast image in the oblique direction are extracted and drawn. Since the component part of the breast may change depending on the imaging direction, the component part to be extracted may be changed depending on which component part is included in the breast image. The same applies to the auxiliary figure, and the classification of the subject area differs depending on the shooting direction, and therefore a necessary auxiliary figure may be drawn according to the classification.

  Further, the present invention can be applied not only to the breast image schema but also to creating a schema for a medical image of another subject such as a chest image. A constituent part of a subject necessary for creating a schema may be extracted from a medical image and drawn, and an auxiliary figure necessary for the schema may be calculated from a positional relationship with each constituent part and drawn.

  Further, the present invention can be applied not only to reports but also to schemas used in electronic medical records.

As a computer-readable medium for the program according to the present invention, a non-volatile memory such as a ROM and a flash memory, and a portable recording medium such as a CD-ROM can be applied.
Further, a carrier wave (carrier wave) is also applied to the present invention as a medium for providing program data according to the present invention via a communication line.

DESCRIPTION OF SYMBOLS 1 Medical image processing apparatus 11 Control part 12 Operation part 13 Display part 14 Communication part 15 Memory | storage part SL Skin line PL Pectoral muscle line Sa Breast area Sc Nipple area M1 Pectoral muscle area Q3 Nipple center

Claims (7)

Comprising a control means for extracting a constituent part of a subject included in a medical image;
The said control means is a medical image processing apparatus which draws the outline of the said extracted component part according to the position in the said medical image of the said extracted component part, and produces a schema.   The medical image processing apparatus according to claim 1, wherein the control unit draws an auxiliary graphic on the schema based on a position of the extracted component in the medical image. The medical image is a breast image whose subject is a breast,
The control means extracts breast skinline, pectoral muscle, and nipple components included in the breast image, and draws the extracted skinline, pectoral muscle contour, and nipple contour to create a schema. The medical image processing apparatus according to claim 2.   The control means obtains the center of the nipple in the extracted nipple region, and based on the position of the nipple center and the extracted skin line, a boundary line for classifying the breast into upper, middle, and lower regions The medical image processing apparatus according to claim 3, wherein the image is drawn on the schema as an auxiliary figure.   The control means obtains a nipple center in the extracted nipple area, and draws a boundary line of a region within a certain distance from the nipple center as an auxiliary figure on the schema based on the position of the nipple center. The medical image processing apparatus according to 3 or 4. A method of creating a schema by a medical image processing apparatus,
Extracting a constituent part of a subject included in the medical image by the control means;
Drawing a contour of the extracted component part according to a position of the extracted component part in the medical image by the control means, and creating a schema;
How to create a schema that includes Computer
Extracting a constituent part of a subject included in a medical image, drawing an outline of the extracted constituent part according to the position of the extracted constituent part in the medical image, and functioning as a control unit for creating a schema Program for.

Images