JP7434923B2 - Image diagnosis support device and program - Google Patents

Description

The present invention relates to an image diagnosis support device and a program.

2. Description of the Related Art Conventionally, image diagnosis support devices are known that display medical images on a monitor for diagnosis by a doctor. Image diagnosis support devices display areas of predetermined structures automatically identified or manually specified in medical images with lines or colors, or perform measurements to provide quantitative information. .

For example, Patent Document 1 discloses that a lung field region is automatically identified from each of a plurality of frame images of a dynamic image obtained by dynamically photographing a subject's chest, and is identified from a functional information image of the subject's lungs. It is described that the image is aligned with the lung field area, integrated, and displayed.

Furthermore, for example, Patent Document 2 describes that segmentation is performed by automatically or manually selecting a lung field.

Japanese Patent Application Publication No. 2018-183943 Special table 2018-537175 publication

However, in the case of automatic discrimination, for example, the edges of a medical image are extracted and the region of a given structure is determined according to a predetermined algorithm, but when the edge of the given structure is incorrectly judged, a completely different region is selected. will be done. Furthermore, there is also a device that allows the operator to adjust the automatically determined area by manually moving a movable point plotted on the automatically determined area, as shown in FIG. 6(a). However, adjusting only one point results in a clearly distorted state as shown in FIG. 6(b). Furthermore, when attempting to make detailed adjustments, it is necessary to adjust many points, which increases the number of operations and takes time and effort.

On the other hand, when manually specifying the region of a predetermined structure, the region is often specified subjectively by the operator, and there is a possibility that a line different from the actual edge portion may be specified depending on the appearance. Furthermore, manually specifying an area accurately takes time and is inefficient.

An object of the present invention is to prevent errors caused by visual region designation and automatic region discrimination in medical images, to support highly reliable diagnosis, and to significantly reduce the effort required by the operator to designate regions. The objective is to improve the efficiency of diagnosis.

In order to solve the above problems, the image diagnosis support device of the present invention includes:
an operation unit that performs input according to the operator's operations;
a display unit that displays medical images;
Discriminating a region of a predetermined structure from the medical image displayed on the display section using a predetermined algorithm, displaying the discrimination result on the display section, and discriminating the region of the predetermined structure on the medical image displayed on the display section. When the region of the predetermined structure is specified by moving one point on the outline of the region of the predetermined structure using the operation unit, the edge where the point after the movement is located is set to the edge of the region of the predetermined structure. a control unit that regards the edge as an outline of the region of the predetermined structure and corrects the determination result of the region of the predetermined structure;
Equipped with.

Further, the program of the present invention is
The computer of the image diagnosis support device is equipped with an operation section that performs input according to the operator's operations, and a display section that displays medical images.
Discriminating a region of a predetermined structure from the medical image displayed on the display section using a predetermined algorithm, displaying the discrimination result on the display section, and discriminating the region of the predetermined structure on the medical image displayed on the display section. When the region of the predetermined structure is specified by moving a point on the outline of the region of the predetermined structure using the operation unit, the edge where the point after the movement is located is set to the edge of the region of the predetermined structure. a control unit that regards the edge as an outline of the region of the predetermined structure and corrects the determination result of the region of the predetermined structure;
function as

According to the present invention, it is possible to prevent errors caused by visual area designation or automatic area discrimination in a medical image, and to support highly reliable diagnosis. Further, it is possible to significantly reduce the amount of effort required by the operator to specify the area, and to improve diagnostic efficiency.

FIG. 1 is a block diagram showing the functional configuration of an image diagnosis support device in an embodiment of the present invention. 2 is a flowchart showing the flow of area optimization processing A executed by the control unit of FIG. 1. FIG. (a) is a diagram showing an example of the diagnostic screen in the automatic discrimination state of the region of a predetermined structure, and (b) is a diagram showing an example of the diagnostic screen after the movable point displayed in (a) has been moved. The figure (c) is a diagram showing an example of the diagnostic screen after the area of the predetermined structure has been optimized based on the moved movable point and edge information. 2 is a flowchart showing the flow of area optimization processing B executed by the control unit of FIG. 1. FIG. (a) is a diagram showing an example of the diagnostic screen when specifying the area of the predetermined structure, and (b) is a diagram showing an example of the diagnostic screen after the area of the predetermined structure has been optimized. FIG. 3 is a diagram illustrating adjustment of a region of a predetermined structure in the prior art.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. However, the scope of the invention is not limited to the illustrated example.

<First embodiment>
A first embodiment of the present invention will be described below. In the first embodiment, an example will be described in which an image diagnosis support apparatus optimizes a region of a predetermined structure automatically determined from a medical image.

[Configuration of image diagnosis support device 1]
First, the configuration of the first embodiment of the present invention will be described.
The image diagnosis support device 1 identifies the area of a predetermined structure from the medical image generated by the modality, highlights the identified area, and performs measurements regarding the identified area, thereby helping doctors perform image diagnosis. It is a supporting device. In this embodiment, a case where the medical image is a chest X-ray image will be explained as an example, but it may also be an X-ray image taken of another body part, or it may be an X-ray image taken from another region, or it may be an X-ray image taken from other modalities such as a CT image or an MRI image. It may also be a medical image taken by.

FIG. 1 shows an example of the functional configuration of the image diagnosis support device 1. As shown in FIG.
As shown in FIG. 1, the image diagnosis support apparatus 1 includes a control section 11, an operation section 12, a display section 13, a storage section 14, and a communication section 15, and each section is connected by a bus 16.

The control unit 11 includes a CPU (Central Processing Unit), a RAM (Random Access Memory), and the like. The CPU of the control unit 11 reads out various programs such as system programs and processing programs stored in the storage unit 14, expands them into the RAM, and executes various processes according to the expanded programs.

For example, when the communication unit 15 receives a medical image from a modality, the control unit 11 transmits the received medical image to patient information (patient ID, name, age, gender, etc.) included in the accompanying information of the medical image, examination information ( (examination ID, examination site, examination date and time, modality type, etc.) and stored in the image DB 141. Further, the control unit 11 executes various processes including area optimization process A, which will be described later, in cooperation with a program stored in the storage unit 14.

The operation unit 12 includes a keyboard with character input keys, numeric input keys, various function keys, etc., and a pointing device such as a mouse, and receives press signals of keys pressed on the keyboard by the operator and the mouse. The operation signal is output to the control unit 11 as an input signal.

The display unit 13 includes a monitor such as a CRT (Cathode Ray Tube) or an LCD (Liquid Crystal Display), and displays various screens according to instructions from display signals input from the control unit 11.

The storage unit 14 is composed of, for example, an HDD (Hard Disk Drive), a semiconductor nonvolatile memory, or the like. The storage unit 14 stores various programs and various data as described above.

Furthermore, the storage unit 14 is provided with an image DB (Data Base) 141. The image DB 141 stores a medical image transmitted from a modality, and patient information and examination information of the medical image in association with each other.

The communication unit 15 is configured with a LAN card or the like, and transmits and receives data to and from an external device connected to the communication network N via a switching hub.

[Operation of image diagnosis support device 1]
Next, the operation of the image diagnosis support device 1 will be explained.
In the image diagnosis support device 1, when a medical image (still image) to be diagnosed is selected by the operation unit 12 from among the medical images stored in the image DB 141, the control unit 11, for example, An image diagnostic screen 131 is displayed on the display unit 13. On the diagnostic screen 131, when a predetermined structure to be diagnosed or measured (for example, lung field, heart, bone, etc.) is selected by the operation unit 12, the control unit 11 selects the selected structure according to a predetermined algorithm. Extracts edges from medical images, automatically determines the area of a given structure based on the extracted edge information (edge information), and makes the determination result distinguishable (for example, by surrounding it with a line) indicate. In this state, when area optimization is instructed by the operation unit 12, the control unit 11 executes area optimization processing A.

FIG. 2 is a flowchart showing the area optimization process A. The area optimization process A is executed in cooperation with the control unit 11 and a program stored in the storage unit 14.

First, the control unit 11 displays a movable point that can be moved by operating the operating unit 12 on the outline of the region of the predetermined structure that has been automatically determined in the displayed medical image (step S1). The position adjustment (movement) of the movable point is accepted (step S2).

When one of the movable points is moved by the operation of the operation unit 12 (step S3; YES), the control unit 11 moves the movable point based on the position of the moved movable point and the edge information of the displayed medical image. , the area of the predetermined structure is optimized (step S4).
In step S4, the control unit 11 regards the edge on which the moved movable point is located in the medical image as an edge of a region of a predetermined structure, and, for example, the control unit 11 considers the edge where the moving point is located after the movement as an edge of a region of a predetermined structure, and (within a predetermined range) is appropriated as the outline of the region of the predetermined structure.
That is, in steps S1 to S4, when a region different from the discrimination result of the automatic discrimination is specified as a region of a predetermined structure by the operation section 12, the control section 11 compares the specified region with the edge information of the medical image. Based on this, the area of the predetermined structure is optimized.

Then, the control unit 11 moves and displays the movable point on the contour of the optimized region of the predetermined structure (step S5), and proceeds to step S6.

FIG. 3A shows an example in which a movable point is displayed on the outline of a region of a predetermined structure (left lung) automatically determined in step S1. FIG. 3(b) shows an example in which one of the movable points is moved by the operation unit 12 in step S3. FIG. 3C shows an example in which the movable point is placed on the contour of the optimized predetermined structure region after one point is moved in step S5.

As shown in FIGS. 3(a) to (c), when the automatically determined region of the predetermined structure is different from the correct one, one of the positions to be corrected on the contour of the determined region of the predetermined structure is When the operator specifies the area of a predetermined structure by moving a point, the area of the predetermined structure is optimized based on the position after the movement and edge information, and the positions of other movable points are corrected in conjunction. Therefore, the time and effort required by an operator such as a doctor or engineer to specify a region can be greatly reduced, and diagnostic efficiency can be improved.

Here, as described above, after the movement of one movable point is completed by the operation of the operation unit 12, the area of the predetermined structure is determined based on the position of the movable point after the movement and the edge information of the medical image. Although optimization may be performed, each time one movable point moves a predetermined distance during a movement operation (dragging), the position and direction of the movable point after movement and edge information in that direction on the medical image are calculated. Based on this, candidates for the region of the predetermined structure may be extracted, and surrounding movable points may be moved and displayed based on the position of the extracted candidate region of the predetermined structure. Thereby, it is possible to perform a display that does not give an unnatural feeling even while the movable point is moving.

Further, when optimizing the region of a predetermined structure, it may not be possible to perform the optimization accurately using only the position and edge information of the movable point due to the influence of shadows of other structures drawn on the medical image. Therefore, in addition to the position of the movable point after movement and the edge information in the medical image, information regarding the predetermined structure is also taken into account, thereby improving the accuracy of optimization. For example, since the approximate position and shape of each predetermined structure in the entire medical image is determined, the area of the predetermined structure is optimized according to the approximate position and shape of the predetermined structure. Furthermore, compared to tissues such as organs, bone parts are hard, straight, and angular. On the other hand, the organs are soft and rounded, with no straight lines or corners. Therefore, when the predetermined structure is a bone, a straight or angular edge is selected, and when the predetermined structure is an organ, a rounded edge is selected, thereby improving the accuracy of optimization.

On the other hand, in step S3, if one movable point has not been moved by the operation of the operation unit 12 (step S3; NO), the control unit 11 moves to step S6.

In step S6, the control unit 11 determines whether confirmation has been instructed by the operation unit 12 (step S6).
If it is determined that confirmation has not been instructed (step S6; NO), the control unit 11 returns to step S2 and accepts position adjustment of the movable point again.
If it is determined that confirmation has been instructed (step S6; YES), the control unit 11 determines the area optimized in step S4 as the area of the predetermined structure (step S7). Then, the area of the determined predetermined structure in the displayed medical image is displayed in an identifiable manner by enclosing it (step S8), and the area optimization process A is completed.

Note that if measurement of the area, width (distance), etc. of the area of the predetermined structure is instructed, the control unit 11 measures the area of the determined predetermined structure and displays the measurement result on the display unit 13. .

In this way, in the area optimization process A of the present embodiment, when the operator corrects a mistake made in the automatic determination of the area of a predetermined structure, which relies entirely on a computer algorithm, and specifies the correct area, the specified area is corrected. Since the region of a predetermined structure can be optimized based on the position and edge information in the medical image, it is possible to prevent mistakes caused by automatic discrimination and support highly reliable diagnosis. In addition, when the operator specifies the area of the predetermined structure by moving one point on the contour of the automatically determined area of the predetermined structure to be corrected, the system can The area of the specified structure is optimized and the positions of other movable points are corrected in conjunction with each other, which greatly reduces the time and effort required for operators such as doctors and technicians to specify areas, and improves diagnostic efficiency. be able to. As a result, the diagnostic efficiency per person improves, making it possible to shorten diagnostic time and patient waiting time.

Note that the medical image to be diagnosed selected from among the medical images stored in the image DB 141 is a moving image consisting of a plurality of frame images obtained by video-recording the subject (capturing the subject multiple times in succession in a short period of time). In this case, the control unit 11 causes the display unit 13 to display, for example, a diagnostic screen 131 on which a representative image of the selected moving image is displayed. For example, the first frame image may be automatically set as the representative image, or each frame image of the moving image may be displayed on the display unit 13 side by side or sequentially to provide a frame image in which the outline of the predetermined structure is easy to see. The operator may be able to select the representative image by operating the operating unit 12.
When a predetermined structure to be diagnosed or measured (for example, a lung field, a heart, etc.) is selected by the operation unit 12 on the diagnostic screen 131, the control unit 11 performs the following for each frame image according to a predetermined algorithm: For example, edges are extracted, and the area of a predetermined structure is automatically determined based on information about the extracted edges. Next, the control unit 11 displays the outline of the region of the predetermined structure on the representative image in a discernible manner. In this state, when the operation unit 12 instructs to optimize the area of the predetermined structure, the control unit 11 executes steps S1 to S8 of the area optimization process A described above for the representative image. Once the region of the predetermined structure in the representative image is determined, the control unit 11 develops the determined region of the predetermined structure into other frame images.

Here, adjacent frame images of a moving image are temporally continuous images, and the positions of edges hardly change. Therefore, the control unit 11 searches for an edge in the same direction and strength as the edge in the vicinity of the edge position of the area of the predetermined structure optimized in the representative image in the frame image adjacent to the representative image, and automatically If the determined edge position of the region of the predetermined structure is different from the searched edge position, the edge position of the predetermined structure region is set as the searched position. This processing is sequentially extended to adjacent frame images, and the position of the area of the predetermined structure determined in the representative image is expanded to other frame images. That is, based on the edge information of the region of the predetermined structure that has been optimized in the representative image and the edge information of the other frame images, the region of the predetermined structure in the other frame images is automatically optimized.

In this way, in the case of moving images, if you specify (adjust) and optimize the area of a predetermined structure in one frame image, the area of the predetermined structure in other frame images is automatically optimized. It is no longer necessary for the operator to specify the area of the predetermined structure for each frame image of the image, and the labor and effort of the operator can be significantly reduced, and diagnostic efficiency can be improved. Furthermore, by specifying an area in a frame image where the outline of the predetermined structure is easy to see, it is possible to easily optimize areas of other frame images where the outline of the predetermined structure is difficult to see.

<Second embodiment>
A second embodiment of the present invention will be described below. In the second embodiment, a case will be described in which an image diagnosis support apparatus optimizes a region of a predetermined structure based on a region of a predetermined structure manually designated by an operator from a medical image.

The configuration of the image diagnosis support device in the second embodiment is the same as that described using FIG. 1, so the same reference numerals are used to refer to the description, and the operation of the second embodiment will be described below. .

In the image diagnosis support device 1, when a medical image (still image) to be diagnosed is selected by the operation unit 12 from among the medical images stored in the image DB 141, the control unit 11, for example, An image diagnostic screen 131 is displayed on the display unit 13. On the diagnostic screen 131, information on a predetermined structure to be diagnosed or measured (for example, lung field, heart, bone, etc.) is specified by the operation unit 12, and when an instruction to specify a region is given by the operation unit 12. , the control unit 11 executes area optimization processing B.

FIG. 4 is a flowchart showing area optimization processing B. The area optimization process B is executed in cooperation with the control unit 11 and a program stored in the storage unit 14.

First, the control unit 11 receives designation of the position of a predetermined structure region by the operation unit 12 on the displayed medical image (step S11).
For example, an operator uses the operation unit 12 to designate a region position by drawing a line or plotting a point on the outline of a predetermined structure on a medical image.

Next, the control unit 11 optimizes the area of the predetermined structure based on the area of the predetermined structure specified by the operation unit 12 and the edge information of the medical image (step S12).
In step S12, the control unit 11 performs edge extraction, compares the position of the area specified in step S11 with edges near the specified position on the medical image, and determines whether an edge exists at the specified position. For locations, the edge is the outline of the area of the predetermined structure, and the edge does not exist at the specified position (where the operator specified it incorrectly), or between the specified position and the specified position. and an edge that is continuous with the edge existing at the specified position in the surrounding area and has the same strength as the edge at the specified position (for example, the difference from the strength of that edge is within a predetermined range) is optimized as the contour of the region of the predetermined structure.

When optimizing the area of a given structure, it is difficult to accurately optimize the area using only the position and edge information of the specified structure due to the influence of shadows of other structures drawn on medical images. It may not be possible to do so. Therefore, in addition to the position of the designated region of the predetermined structure and the edge information in the medical image, information regarding the predetermined structure is also taken into consideration, as explained in the first embodiment, to improve the accuracy of optimization. can be done.

Then, the control unit 11 displays the optimized outline of the region of the predetermined structure on the medical image (step S13).

For example, as shown in FIG. 5(a), when the operator intermittently specifies areas of a predetermined structure, as shown in FIG. 5(b), the contours in between are and is displayed after being optimized based on the edge information of the medical image.

Next, the control unit 11 determines whether confirmation has been instructed by the operation unit 12 (step S14).
If it is determined that confirmation has not been instructed (step S14; NO), the control unit 11 returns to step S11 and receives designation of the position (position adjustment) of the predetermined structure. For example, a point once plotted can be moved and adjusted by dragging or the like using the operation unit 12.

If it is determined that confirmation has been instructed (step S14; YES), the control unit 11 confirms the optimized area as the area of the predetermined structure in the displayed medical image (step S15), and The area of the determined predetermined structure above is displayed in an identifiable manner by enclosing it (step S16), and area optimization processing B is ended.

In this way, in the area optimization process B of the second embodiment, mistakes in area specification by the operator's visual inspection can be corrected by optimization, so mistakes in area specification by the operator can be prevented, and the reliability can be improved. can support high diagnosis. Furthermore, since it is not necessary for a doctor, engineer, etc. to accurately specify the entire outline of a predetermined structure, the time and effort required for the operator to specify an area is greatly reduced, and diagnostic efficiency can be improved. As a result, the diagnostic efficiency per person improves, making it possible to shorten diagnostic time and patient waiting time.

Note that when the medical image to be diagnosed selected from among the medical images stored in the image DB 141 is a moving image, the control unit 11 transfers the region of the specified structure specified in the representative image to another frame image. Expand to. The method and effect of expanding the region of the predetermined structure designated in the representative image into other frame images are the same as those described in the first embodiment, so the description will be referred to.

Although the embodiment of the image diagnosis support device 1 according to the present invention has been described above, the description in the above embodiment is an example of a suitable image diagnosis support device according to the present invention, and is not limited thereto.

Further, for example, in the above description, an example is disclosed in which a hard disk, a semiconductor nonvolatile memory, etc. are used as a computer-readable medium for the program according to the present invention, but the present invention is not limited to this example. As other computer-readable media, it is possible to apply a portable recording medium such as a CD-ROM. Furthermore, a carrier wave is also applied as a medium for providing data of the program according to the present invention via a communication line.

In addition, the detailed configuration and detailed operation of the image diagnosis support device can be changed as appropriate without departing from the spirit of the present invention.

1 Image diagnosis support device 11 Control section 12 Operation section 13 Display section 14 Storage section 141 Image DB
15 Communication Department 16 Bus

Claims (5)

an operation unit that performs input according to the operator's operations;
a display unit that displays medical images;
Discriminating a region of a predetermined structure from the medical image displayed on the display section using a predetermined algorithm, displaying the discrimination result on the display section, and discriminating the region of the predetermined structure on the medical image displayed on the display section. When the region of the predetermined structure is specified by moving a point on the outline of the region of the predetermined structure using the operation unit, the edge where the point after the movement is located is set to the edge of the region of the predetermined structure. a control unit that regards the edge as an outline of the region of the predetermined structure and corrects the determination result of the region of the predetermined structure;
An image diagnosis support device comprising: The image diagnosis support apparatus according to claim 1, wherein the control unit further modifies the determination result of the region of the predetermined structure in the medical image based on information regarding the predetermined structure. When displaying the determination result on the display unit, the control unit displays movable points movable by the operation unit on the contour of the determined predetermined structure, and one of the movable points When the region of the predetermined structure is specified by moving with the operation unit, the edge where the point after the movement is located is taken as the edge of the region of the predetermined structure, and the edge is designated as the region of the predetermined structure. The image diagnosis support device according to claim 1 or 2, wherein the determination result of the region of the predetermined structure is corrected as the contour of the region. When the medical image is a moving image consisting of a plurality of frame images, when the determination result of the area of the predetermined structure is corrected in one frame image among the plurality of frame images, the control unit corrects the determination result of the predetermined structure area. 4. The determination result of the area of the predetermined structure in the other frame image is automatically corrected based on the edge information of the area in which the structure has been determined and the edge information of the other frame image. The image diagnosis support device described in . The computer of the image diagnosis support device is equipped with an operation section that performs input according to the operator's operations, and a display section that displays medical images.
Discriminating a region of a predetermined structure from the medical image displayed on the display section using a predetermined algorithm, displaying the discrimination result on the display section, and discriminating the region of the predetermined structure on the medical image displayed on the display section. When the region of the predetermined structure is specified by moving one point on the outline of the region of the predetermined structure using the operation unit, the edge where the point after the movement is located is set to the edge of the region of the predetermined structure. a control unit that regards the edge as an outline of the region of the predetermined structure and corrects the determination result of the region of the predetermined structure;
A program to function as

Images