KR102321349B1 - Lesion Detection Apparatus and Method - Google Patents

Abstract

The present invention relates to an apparatus and method for detecting a lesion, wherein a predicted lesion detected in a 2D medical image is detected in 3D coordinates located in the 3D medical image and then applied to a 3D medical image by using the detected 3D coordinates. An object of the present invention is to provide a lesion detection apparatus and method capable of not only significantly reducing the time required for lesion detection, but also improving the accuracy of lesion detection by detecting a lesion by setting a three-dimensional region of interest (ROI).
To this end, the present invention provides an apparatus for detecting a lesion, comprising: a two-dimensional lesion detector for detecting a predicted lesion in a two-dimensional medical image using a two-dimensional lesion detection algorithm; a three-dimensional coordinate detector for detecting three-dimensional coordinates in which the predicted lesion detected by the two-dimensional lesion detector is located in a three-dimensional medical image; a 3D region of interest setting unit for setting a 3D region of interest (ROI) in the 3D medical image by using the 3D coordinates detected by the 3D coordinate detection unit; and a 3D lesion detector for detecting a lesion in a 3D region of interest (ROI) set in the 3D medical image by using a 3D lesion detection algorithm.

Description

Lesion Detection Apparatus and Method

The present invention relates to an apparatus and a method for detecting a lesion in a medical image, and more particularly, detecting a three-dimensional coordinate in which an expected lesion detected in a two-dimensional medical image (eg, a panoramic image) is located in a three-dimensional medical image. After the lesion is detected by setting a 3D region of interest (ROI) on a 3D medical image (eg, computed tomography image) after The present invention relates to a lesion detection device and method capable of improving the accuracy of the lesion.

In the following embodiments of the present invention, a panoramic image will be described as an example of a two-dimensional medical image, and a computed tomography (CT) image will be described as an example of a three-dimensional medical image. As an example, an oral lesion detection algorithm (Oral Lesion Detection Algorithm) will be described as an example, but it is to be noted in advance that the present invention is not limited thereto.

Recently, due to the development of surgical techniques, various methods of minimally invasive surgery have emerged. The minimally invasive method uses surgical tools such as syringes or catheters to access the lesion instead of incising the skin and muscle for access to the lesion and injecting drugs, removing the lesion, and inserting a prosthesis. . For this procedure, the doctor must accurately determine the location of the lesion. In addition, in order for a doctor to accurately identify the disease name, the size, shape, location, etc. of the lesion in the human body must be accurately identified.

To this end, various medical imaging devices capable of accurately detecting the size, shape, location, etc. of a lesion are being developed. For example, a computed tomography (CT) device, a magnetic resonance imaging (MRI) device, a positron emission tomography (PET) device, and a single photon emission tomography (SPECT: Single) device Photon Emission Computed Tomography) devices.

Meanwhile, in general, for accurate lesion detection, a lesion is detected using a panoramic image or a lesion is detected using a computed tomography (CT) image.

However, since a computed tomography image requires a lot of computation compared to a panoramic image, when a three-dimensional lesion detection algorithm (eg, a three-dimensional oral lesion detection algorithm) is directly applied to a computed tomography image, the lesion Not only does the detection take a lot of time, but the accuracy of lesion detection decreases.

As described above, the prior art has a problem in that it takes a lot of time to detect a lesion, and the accuracy of detecting a lesion is low, and it is an object of the present invention to solve this problem.

Therefore, the present invention sets a 3D region of interest (ROI) in the 3D medical image by using the 3D coordinates detected after detecting the 3D coordinates where the predicted lesion detected in the 2D medical image is located in the 3D medical image. Accordingly, an object of the present invention is to provide a lesion detection apparatus and method capable of remarkably reducing the time required for lesion detection and improving the accuracy of lesion detection by detecting the lesion.

The objects of the present invention are not limited to the above-mentioned objects, and other objects and advantages of the present invention not mentioned may be understood by the following description, and will be more clearly known by the embodiments of the present invention. It will also be readily apparent that the objects and advantages of the present invention may be realized by the means and combinations thereof indicated in the appended claims.

According to an aspect of the present invention, there is provided an apparatus for detecting a lesion, comprising: a two-dimensional lesion detector for detecting a predicted lesion in a two-dimensional medical image using a two-dimensional lesion detection algorithm; a three-dimensional coordinate detector for detecting three-dimensional coordinates in which the predicted lesion detected by the two-dimensional lesion detector is located in a three-dimensional medical image; a 3D region of interest setting unit for setting a 3D region of interest (ROI) in the 3D medical image by using the 3D coordinates detected by the 3D coordinate detection unit; and a 3D lesion detector for detecting a lesion in a 3D region of interest (ROI) set in the 3D medical image by using a 3D lesion detection algorithm.

On the other hand, the method of the present invention for achieving the above object, in a lesion detection method in a lesion detection apparatus, comprising the steps of: (a) detecting a predicted lesion in a two-dimensional medical image using a two-dimensional lesion detection algorithm; (b) detecting three-dimensional coordinates in which the detected predicted lesion is located in a three-dimensional medical image; (c) setting a 3D region of interest (ROI) in the 3D medical image by using the detected 3D coordinates; and (d) detecting a lesion in a 3D region of interest (ROI) set in the 3D medical image using a 3D lesion detection algorithm.

As described above, the present invention provides a 3D region of interest (ROI) in a 3D medical image by using the 3D coordinates detected after detecting the 3D coordinates in which the predicted lesion detected in the 2D medical image is located in the 3D medical image. ) to detect the lesion, it is possible to dramatically reduce the time required for lesion detection and to improve the accuracy of lesion detection.

1 is a block diagram of an apparatus for detecting a lesion according to an embodiment of the present invention;
2 is a view showing a panoramic image with a lesion;
3 is a view showing a panoramic trajectory by way of example;
4 is a view showing an example of setting a three-dimensional region of interest (ROI);
5 is a flowchart of a lesion detection method according to an embodiment of the present invention.

The above-described objects, features, and advantages will become more clear through the detailed description described below in detail with reference to the accompanying drawings, and accordingly, those of ordinary skill in the art to which the present invention pertains can understand the technical spirit of the present invention. can be easily implemented. In addition, in the description of the present invention, if it is determined that a detailed description of a known technology related to the present invention may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted.

And throughout the specification, when a part is "connected" with another part, it includes not only the case where it is "directly connected" but also the case where it is "electrically connected" with another element interposed therebetween. In addition, when a part "includes" or "includes" a certain component, it means that other components may be further included or provided without excluding other components unless otherwise stated. . In addition, it will be understood that even if some components are described in the singular in the description of the entire specification, the present invention is not limited thereto, and the corresponding components may be formed in plurality.

First, to summarize the technical gist according to an embodiment of the present invention, a two-dimensional lesion detection algorithm (eg, a two-dimensional oral lesion detection algorithm) is applied to a panoramic image to detect a predicted lesion. Then, the three-dimensional coordinates at which the detected predicted lesion is located in the computed tomography image are detected based on the panorama trajectory or the panorama calculation area. Based on the three-dimensional coordinates detected in this way, a three-dimensional region of interest (ROI) is set on a computed tomography image and a three-dimensional lesion detection algorithm (eg, three-dimensional oral lesion detection algorithm) is applied to detect the lesion. detect

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram of an apparatus for detecting a lesion according to an embodiment of the present invention, FIG. 2 is a diagram showing a panoramic image with a lesion, FIG. 3 is a diagram illustrating a panoramic trajectory, and FIG. 4 is a three-dimensional diagram It is a diagram showing an example of setting a region of interest (ROI).

As shown in FIG. 1 , the lesion detection apparatus according to an embodiment of the present invention includes a two-dimensional lesion detector 110 and a two-dimensional lesion detector for detecting a predicted lesion in a panoramic image using a two-dimensional lesion detection algorithm. Computed tomography using the 3D coordinate detection unit 120 for detecting 3D coordinates in which the predicted lesion detected in 110 is located in the computed tomography image, and the 3D coordinates detected by the 3D coordinate detection unit 120 In the 3D region of interest (ROI) set in the computed tomography image using the 3D region of interest setting unit 130 for setting a 3D region of interest (ROI) in the image, and a 3D lesion detection algorithm and a three-dimensional lesion detector 140 for detecting a lesion.

Next, with reference to FIGS. 1 to 4, each of the components will be described in more detail as follows.

First, the two-dimensional lesion detector 110 detects a predicted lesion in a panoramic image using a two-dimensional lesion detection algorithm. In this case, for example, when the panoramic image is a panoramic image of an oral, a known two-dimensional lesion detection algorithm may be used as the two-dimensional lesion detection algorithm. Here, the predicted lesion is a lesion expected to be a lesion, and means a lesion with a high probability of a lesion.

The 3D coordinate detection unit 120 detects (acquires) 3D coordinates in which the predicted lesion detected by the 2D lesion detection unit 110 in the panoramic image is located in a computed tomography image. That is, the 3D coordinate detection unit 120 shows the 3D coordinates at which the predicted lesion detected from the panoramic image by the 2D lesion detection unit 110 is located in the computed tomography image in the panorama calculation area shown in FIG. 2 and in FIG. It is detected (acquired) using the obtained panoramic trajectory.

Looking at this in more detail, the three-dimensional coordinate detection unit 120 calculates the axial coordinates in which the predicted lesion detected in the panoramic image by the two-dimensional lesion detection unit 110 is located in the computed tomography image in the panorama calculation area (Fig. 2 in the Y-axis direction). In addition, the 3D coordinate detection unit 120 calculates the coronal direction coordinates in which the predicted lesion detected by the 2D lesion detection unit 110 in the panoramic image is located in the computed tomography image in the panorama calculation area (X-axis in FIG. 2 ). direction) is used to detect In addition, the 3D coordinate detection unit 120 uses the panoramic trajectory (trajectory in FIG. 3) for the Sagittal direction coordinates where the predicted lesion detected by the 2D lesion detection unit 110 in the panoramic image is located in the computed tomography image. to detect

In addition, the 3D region of interest setting unit 130 sets a 3D region of interest (ROI) in the computed tomography image by using the 3D coordinates detected by the 3D coordinate detection unit 120 . An example of the set 3D region of interest (ROI) is illustrated in FIG. 4 . In this case, the panoramic image and the computed tomography image are medical images obtained by photographing the same site for the same person based on the same starting point.

In addition, the 3D lesion detector 140 detects a lesion in a 3D region of interest (ROI) set in a computed tomography image by using a 3D lesion detection algorithm. In this case, when the computed tomography image is a computed tomography image of an oral, a well-known three-dimensional lesion detection algorithm may be used as the three-dimensional lesion detection algorithm.

In addition, although not shown in the drawing, the lesion detecting apparatus of FIG. 1 may be connected to a panoramic and computed tomography apparatus through a network.

On the other hand, there are mainly two methods for acquiring a panoramic image: a method of acquiring a 3D image by cross-sectional transformation, and a method of acquiring a panoramic image separately.

Here, the method of acquiring through the separate panoramic photographing can be divided into two cases again. One is a case where the photograph is taken with the same equipment for both panorama and computed tomography (CT), and the other is a separate panoramic equipment and Each may have been taken on a computed tomography (CT) machine.

In an embodiment of the present invention, the panoramic image obtained by any of the above methods may be used. However, in the above-described embodiment of the present invention, mainly the case of photographing with the same equipment for both panoramic and computed tomography (CT) is described as an example, and the panoramic trajectory is based on basic photographing information such as sensor size, magnification, and photographing trajectory. The axial, coronal, and sagittal direction coordinates of computed tomography (CT) images can be found through the and panorama calculation area.

And in the case of a method of acquiring a 3D image by cross-sectional transformation, it is possible to find the axial, coronal, and sagittal direction coordinates of a computed tomography (CT) image without any special difficulty because the coordinates and directions of the two images are substantially the same. can

However, when the images are taken by separate panoramic equipment and computed tomography (CT) equipment, the three-dimensional coordinate detection unit 120 must perform a matching process of matching the directions, coordinates, and scales of the two images before detecting the three-dimensional coordinates.

Such a registration process may be performed in a manner of matching a plurality of feature points that can be easily found in each image.

5 is a flowchart of a lesion detection method according to an embodiment of the present invention. Since a specific embodiment has been described in detail in the description of the lesion detection apparatus of FIG. 1 , the operation process will be briefly described here. do it with

First, the two-dimensional lesion detection unit 110 detects an expected lesion in the panoramic image using the two-dimensional lesion detection algorithm (S510).

Thereafter, the 3D coordinate detection unit 120 detects (acquires) 3D coordinates in which the predicted lesion detected in the panoramic image by the 2D lesion detection unit 110 is located in a computed tomography image (S520). That is, the 3D coordinate detection unit 120 shows the 3D coordinates at which the predicted lesion detected from the panoramic image by the 2D lesion detection unit 110 is located in the computed tomography image in the panorama calculation area shown in FIG. 2 and in FIG. It is detected (acquired) using the obtained panoramic trajectory.

The three-dimensional coordinate detection process ( S520 ) will be described in more detail with reference to processes S521 to S523 .

The three-dimensional coordinate detector 120 calculates the axial coordinates of the predicted lesion detected from the panoramic image by the two-dimensional lesion detector 110 in the computed tomography image in the panoramic calculation area (Y-axis direction in FIG. 2). is detected using (S521).

In addition, the 3D coordinate detection unit 120 calculates the coronal direction coordinates in which the predicted lesion detected by the 2D lesion detection unit 110 in the panoramic image is located in the computed tomography image in the panorama calculation area (X-axis in FIG. 2 ). direction) to detect (S522).

In addition, the 3D coordinate detection unit 120 uses the panoramic trajectory (trajectory in FIG. 3) for the Sagittal direction coordinates where the predicted lesion detected by the 2D lesion detection unit 110 in the panoramic image is located in the computed tomography image. to detect (S523).

Here, even if steps S521 to S523 are performed regardless of the order, the same result can be obtained.

Thereafter, the 3D region of interest setting unit 130 sets a 3D region of interest (ROI) in the computed tomography image by using the 3D coordinates detected by the 3D coordinate detection unit 120 ( S530 ). .

Thereafter, the 3D lesion detector 140 detects the lesion in the 3D region of interest (ROI) set in the computed tomography image by using the 3D lesion detection algorithm (S540).

As described above, in an embodiment of the present invention, the predicted lesion detected in the 2D medical image is 3D coordinates in the 3D medical image by using the detected 3D coordinates after detecting the 3D coordinates in the 3D medical image. By setting a dimensional region of interest (ROI) to detect a lesion, the time required for lesion detection can be reduced several times, and the lesion detection probability, that is, the accuracy of lesion detection can be improved by 3 to 4 times.

On the other hand, the lesion detection method according to the present invention as described above may be implemented in the form of a program command that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the present invention, or may be known and available to those skilled in the art of computer software. Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic such as floppy disks. - includes magneto-optical media, and hardware devices specially configured to store and carry out program instructions, such as ROM, RAM, flash memory, and the like. The medium may be a transmission medium such as an optical or metal wire or a waveguide including a carrier wave for transmitting a signal designating a program command, a data structure, and the like. Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware device may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

As described above, although the present invention has been described with reference to limited embodiments and drawings, the present invention is not limited to the above embodiments, and those of ordinary skill in the art to which the present invention pertains can learn from these descriptions of the present invention. Various substitutions, modifications, and changes are possible without departing from the technical spirit.

Therefore, the scope of the present invention should not be limited to the described embodiments, and should be defined by the following claims as well as the claims and equivalents.

110: two-dimensional lesion detection unit 120: three-dimensional coordinate detection unit
130: 3D region of interest setting unit 140: 3D lesion detection unit

Claims (6)

In the lesion detection device,
a two-dimensional lesion detector for detecting a predicted lesion in the two-dimensional first medical image;
a three-dimensional coordinate detector configured to detect a three-dimensional position of the predicted lesion in a three-dimensional second medical image acquired separately from the first medical image;
a 3D region of interest setting unit for setting a 3D region of interest (ROI) in the 3D second medical image by using the 3D position; and
and a three-dimensional lesion detector that finally detects a lesion in the three-dimensional region of interest (ROI);
The three-dimensional coordinate detection unit,
detecting a three-dimensional position of the expected lesion by using the two-dimensional position of the expected lesion in the two-dimensional first medical image and photographing information of the two-dimensional first medical image;
lesion detection device.
The method of claim 1,
The two-dimensional first medical image is a panoramic image,
The photographing information includes a photographing trajectory of a device that photographed the panoramic image,
The three-dimensional second medical image is a computed tomography image,
lesion detection device.
3. The method of claim 2,
The three-dimensional coordinate detection unit,
Detecting an axial cross-sectional position and a coronal direction cross-sectional position of the 3D second medical image in horizontal and vertical directions of the predicted lesion detected in the two-dimensional first medical image,
Detecting the position of the cross section in the Sagittal direction of the 3D second medical image by the shooting trajectory,
Detecting the three-dimensional position of the expected lesion with the axial, coronal and sagittal direction cross-sectional positions,
lesion detection device.
delete delete delete

Images