KR102236617B1 - Method for measuring lung cancer tumor size and apparatus for executint the method - Google Patents

Abstract

A method for measuring the size of a lung cancer tumor and an apparatus for performing the same are disclosed. An apparatus according to the disclosed embodiment is a computing device having one or more processors and a memory for storing one or more programs executed by the one or more processors, and obtains a tumor photographing image of a tumor portion of a patient. The first boundary detection unit that detects the boundary of the GGO (Ground Glass Opacity) part in the tumor photographed image through the first deep learning module, and the solid part in the tumor photographed image through the second deep learning module. A second boundary detection unit for detecting a boundary, and a size measuring unit for measuring the size of the GGO portion based on the boundary of the detected GGO portion, and measuring the size of the solid portion based on the detected boundary of the solid portion.

Description

A method for measuring lung cancer tumor size and an apparatus for performing it {METHOD FOR MEASURING LUNG CANCER TUMOR SIZE AND APPARATUS FOR EXECUTINT THE METHOD}

An embodiment of the present invention relates to a technique for measuring lung cancer tumor size.

In general, cancer staging is measured to predict the prognosis of cancer patients. Cancer staging is determined by TNM (Tumor-Node-Metastasis), where T (Tumor) represents the size of the tumor, N (Node) represents the extent of spread to the lymph nodes, and M (Metastasis) represents whether metastases to other organs. Show. Among them, the size of the tumor becomes an important judgment factor in predicting the prognosis of a cancer patient, so it is important to accurately measure the size of the tumor.

On the other hand, in the case of lung cancer, if the CT (Computed Tomography) image contains GGO (Ground Glass Opacity), it is necessary to separately measure the volume of the solid part (solid part) and the volume of the GGO part to determine cancer staging. It becomes an important factor.

Korean Registered Patent Publication No. 10-1664159 (2016.10.12)

An embodiment of the present invention is to provide a method for measuring the size of a lung cancer tumor capable of measuring the size of a GGO portion and a solid portion in a tumor imaging image, and an apparatus for performing the same.

An apparatus according to an embodiment disclosed is a computing device having one or more processors and a memory for storing one or more programs executed by the one or more processors, wherein a tumor photographing image of a tumor portion of a patient is captured. An image acquisition unit to acquire; A first boundary detection unit configured to detect a boundary of a GGO (Ground Glass Opacity) portion in the tumor photographed image through a first deep learning module; A second boundary detector configured to detect a boundary of a solid portion in the tumor photographed image through a second deep learning module; And a size measuring unit that measures the size of the GGO portion based on the detected boundary of the GGO portion and measures the size of the solid portion based on the detected boundary of the solid portion.

The first boundary detection unit may include: a first image conversion unit configured to convert the tumor photographed image through a first window having a predetermined first Hounsfield unit range to generate a GGO portion image; And a first deep learning unit that receives the GGO portion image and detects a boundary of a GGO portion in the GGO portion image.

The first window may be provided to have a preset first window width based on a preset first window level in units of a hounsfield.

The computing device may further include a first image resizing unit for resizing the generated GGO portion image to a preset size and transmitting the resized GGO portion image to the first deep learning unit.

When a GGO portion image is input, the first deep learning unit may include a convolution neural network (CNN) trained to detect and output a boundary of a GGO portion in the input GGO portion image.

The second boundary detection unit may include a second image conversion unit configured to convert the tumor photographed image through a second window having a preset second Hounsfield unit range to generate a solid portion image; And a second deep learning unit that receives the solid portion image and boundary information of the GGO portion and detects a boundary of the solid portion in the solid portion image.

The second window may be provided to have a second window width narrower than the first window width based on a second window level having a higher Hounsfield unit value than the first window level.

The computing device may further include a second image resizing unit for resizing the generated solid portion image to a preset size and transmitting the resized solid portion image to the second deep learning unit.

The second deep learning unit includes a convolution neural network (CNN) trained to detect and output the boundary of the solid portion in the input solid portion image when boundary information of the solid portion image and the GGO portion is input. can do.

A method according to the disclosed embodiment is a method performed in a computing device having one or more processors and a memory storing one or more programs executed by the one or more processors, wherein a tumor portion of a patient is photographed. Obtaining a tumor imaging image; Detecting a boundary of a GGO (Ground Glass Opacity) portion in the tumor photographed image through a first deep learning module; Detecting a boundary of a solid portion in the tumor photographed image through a second deep learning module; Measuring the size of the GGO portion based on the detected boundary of the GGO portion; And measuring the size of the solid portion based on the detected boundary of the solid portion.

According to the disclosed embodiment, by measuring the sizes of the GGO portion and the solid portion, respectively, using deep learning technology in a tumor imaging image including GGO, the size of the GGO portion and the solid portion can be quickly and accurately measured. Therefore, in lung cancer staging, it is possible to accurately determine the T (Tumor) stage, which is an important determining factor of patient prognosis.

1 is a diagram showing a state in which the volume of a lung nodule is measured in the 7th plate lung cancer staging
2 is a view showing a state of measuring the volume of a lung nodule in the eighth plate lung cancer staging
3 is a view showing the configuration of a lung cancer tumor size measuring apparatus according to an embodiment of the present invention
4 is a view showing the configuration of a first boundary measurement unit and a second boundary measurement unit in the lung cancer tumor size measurement apparatus according to an embodiment of the present invention
5 is a flow chart showing a method for measuring lung cancer tumor size according to an embodiment of the present invention
6 is a view showing a state in which a GGO portion image and a solid portion image are respectively generated from a tumor photographing image according to an embodiment of the present invention
7 is a diagram comparing boundary detection by deep learning and boundary detection by an expert according to an embodiment of the present invention
8 is a block diagram illustrating and describing a computing environment including a computing device suitable for use in example embodiments.

Hereinafter, a specific embodiment of the present invention will be described with reference to the drawings. The following detailed description is provided to aid in a comprehensive understanding of the methods, devices, and/or systems described herein. However, this is only an example and the present invention is not limited thereto.

In describing the embodiments of the present invention, when it is determined that a detailed description of known technologies related to the present invention may unnecessarily obscure the subject matter of the present invention, a detailed description thereof will be omitted. In addition, terms to be described later are terms defined in consideration of functions in the present invention, which may vary according to the intention or custom of users or operators. Therefore, the definition should be made based on the contents throughout the present specification. The terms used in the detailed description are only for describing embodiments of the present invention, and should not be limiting. Unless explicitly used otherwise, expressions in the singular form include the meaning of the plural form. In this description, expressions such as "comprising" or "feature" are intended to indicate certain features, numbers, steps, actions, elements, some or combination thereof, and one or more It should not be construed to exclude the presence or possibility of other features, numbers, steps, actions, elements, any part or combination thereof.

In the following description, "transmission", "communication", "transmission", "reception" of signals or information, and other terms having a similar meaning are not only directly transmitted signals or information from one component to another component. It includes what is passed through other components. In particular, "transmitting" or "transmitting" a signal or information to a component indicates the final destination of the signal or information and does not mean a direct destination. The same is true for "reception" of signals or information. In addition, in the present specification, when two or more pieces of data or information are "related", it means that when one data (or information) is obtained, at least a part of other data (or information) can be obtained based thereon.

In addition, terms such as first and second may be used to describe various components, but the components should not be limited by the terms. The terms may be used for the purpose of distinguishing one component from another component. For example, without departing from the scope of the present invention, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element.

In the case of the previous plate (that is, the 7th plate) lung cancer staging, as shown in FIG. 1, when the CT (Computed Tomography) image contains GGO (Ground Glass Opacity), the GGO portion The volume of pulmonary nodules including the whole was measured and the T (Tumor) stage was determined accordingly.

However, the eighth edition published in 2017 (8 ^th edition), lung cancer staging (Lung Cancer Staging) In the case that contains the GGO a CT (Computed Tomography) image, the Solid part as shown in Figure 2 (solid portion) of It has been changed to determine the T (Tumor) stage by measuring only the volume. In other words, as it is known that GGO is contained outside the solid part (solid part) clinically and the patient's prognosis is known to be better, the T-stage is determined by measuring only the volume of the solid part, not the total volume including the GGO part. I decided to do it.

Accordingly, in the embodiment of the present invention, when GGO is included in the wound image (eg, CT, MRI, etc.), the volume of the GGO part and the solid part is measured using a deep learning technology. Includes technology.

3 is a diagram showing the configuration of a lung cancer tumor size measuring apparatus according to an embodiment of the present invention, Figure 4 is a first boundary measurement unit and a second boundary measurement in the lung cancer tumor size measuring apparatus according to an embodiment of the present invention It is a figure showing the structure of a part.

3 and 4, the lung cancer tumor size measurement apparatus 100 includes an image acquisition unit 102, a first boundary detection unit 104, a second boundary detection unit 106, and a size measurement unit 108. can do.

The image acquisition unit 102 may acquire a tumor photographing image of a patient's tumor portion. In an exemplary embodiment, the tumor imaging image may be a computed tomography (CT) image, but is not limited thereto, and may include a magnetic resonance imaging (MRI) image, an ultrasound image, and the like. Hereinafter, it will be described that the tumor imaging image is a CT image.

The first boundary detection unit 104 may detect the boundary of a GGO (Ground Glass Opacity) portion in the tumor photographed image. The first boundary detection unit 104 may include a first image conversion unit 111, a first image resizing unit 113, and a first deep learning unit 115.

The first image conversion unit 111 converts the tumor photographed image through a window having a preset first Hounsfield Unit (HU) range (hereinafter, may be referred to as a “first window”). Video can be converted. The first image conversion unit 111 may generate a GGO portion image by converting the tumor photographed image through the first window. That is, the first image conversion unit 111 may generate a GGO portion image in which the GGO portion is clearly visible among the tumor imaged images by converting the tumor imaged image through the first window.

Here, the first window may be provided to have a preset first window width (WW) based on a preset first window level (WL). The first window level WL and the first window width WW may be expressed in a Hounsfield unit (HU). In an exemplary embodiment, the first window may have a first window level WL of -700HU and a first window width WW of 1500HU. That is, the first window may be provided to filter a value of a hound field unit in a range of 1500HU based on -700HU.

The first image resizing unit 113 may resize the GGO portion image to a preset size. The first image resizing unit 113 may resize the GGO portion image to the size of images learned by the first deep learning unit 115. That is, since the GGO portion image is used as an input of the first deep learning unit 115, it can be resized to the same size as the images learned by the first deep learning unit 115. In an exemplary embodiment, the first image resizing unit 113 may resize the GGO portion image to a size of 128×128.

The first deep learning unit 115 may receive the resized GGO portion image and detect a boundary of a GGO (Ground Glass Opacity) portion of the GGO portion image. When the GGO portion image is input, the first deep learning unit 115 performs deep learning based on the already learned images (i.e., GGO portion images), and performs a GGO portion (Portion) in the input GGO portion image. ) Can be detected.

In an exemplary embodiment, the first deep learning unit 115 may use a convolution neural network (CNN) as a deep learning technology. In this case, when a GGO portion image is input, the first deep learning unit 115 may be a neural network trained to detect and output a boundary of a GGO portion from the input GGO portion image. Since the convolutional neural network is a known technology, a detailed description thereof will be omitted. The first deep learning unit 115 may transmit the boundary of the GGO portion to the second deep learning unit 125.

The second boundary detection unit 106 may detect the boundary of a solid portion (solid portion) in the tumor imaging image. The second boundary detection unit 106 may include a second image conversion unit 121, a second image resizing unit 123, and a second deep learning unit 125.

The second image conversion unit 121 converts the tumor photographed image through a window having a preset second Hounsfield Unit (HU) range (hereinafter, may be referred to as a “second window”). Video can be converted. The second image conversion unit 121 may generate a solid partial image by converting the tumor photographed image through the second window. That is, the second image conversion unit 121 may generate a solid portion image in which a solid portion of the tumor photographed image is clearly visible by converting the tumor photographed image through the second window.

Here, the second window may be provided to have a preset second window width (WW) based on a preset second window level (WL). The second window level WL and the second window width WW may be expressed in a Hounsfield unit (HU).

The second window level may have a Hounsfield unit value higher than the first window level. In addition, the second window width may have a narrower Hounsfield unit range than the first window width. In an exemplary embodiment, the second window may have a second window level WL of 20HU and a second window width WW of 400HU. That is, the second window may be provided to filter the value of the hound field unit in the range of 400HU based on 20HU.

The second image resizing unit 123 may resize the solid portion image to a preset size. The second image resizing unit 123 may resize the solid portion image to the size of images learned by the second deep learning unit 125. That is, since the solid portion image is used as an input of the second deep learning unit 125, it may be resized to the same size as the images learned by the second deep learning unit 125. In an exemplary embodiment, the second image resizing unit 123 may resize the solid portion image to a size of 128 × 128. The second image resizing unit 123 may resize the solid portion image to have the same size as the size at which the first image resizing unit 113 resizes the GGO portion image.

The second deep learning unit 125 may receive the resized solid portion image and detect a boundary of a solid portion in the solid portion image. In an exemplary embodiment, the second deep learning unit 125 may use boundary information of the GGO portion in the GGO portion image when detecting the boundary of the solid portion in the solid portion image.

That is, the second deep learning unit 125 receives the resized solid portion image from the second image resizing unit 123, and information output from the first deep learning unit 115 (that is, the GGO portion in the GGO portion image). Boundary information) can be input respectively.

When the boundary information of the GGO portion is input from the resized solid portion image and the GGO portion image, the second deep learning unit 125 has already learned data (i.e., boundary information of the GGO portion in the solid portion image and the GGO portion image). Based on these, deep learning may be performed to detect the boundary of the solid portion in the input solid portion image.

In an exemplary embodiment, the second deep learning unit 125 may use a convolution neural network (CNN) as a deep learning technology. In this case, the second deep learning unit 125 may be a neural network trained to detect and output the boundary of the solid portion from the solid portion image and the GGO portion image when boundary information of the GGO portion is input. have. Here, the second deep learning unit 125 may be a neural network that has been trained to detect a solid portion within the boundary of the GGO portion of the solid portion image.

The size measurement unit 108 may measure the size of the GGO portion in the tumor imaging image based on the boundary of the GGO portion detected by the first boundary detection unit 104. Also, the size measurement unit 108 may measure the size of the solid portion in the tumor photographing image based on the boundary of the solid portion detected by the second boundary detection unit 106.

Meanwhile, when the tumor imaging image is a CT image, a tumor imaging image may be obtained for each image layer, and a boundary of a GGO portion and a boundary of a solid portion may be detected for the tumor imaging image of each image layer. In addition, the size measuring unit 108 may measure the volume of the GGO portion by stacking the boundary of the GGO portion of each image layer, and measure the volume of the solid portion by stacking the boundary of the solid portion of each image layer.

According to the disclosed embodiment, by measuring the sizes of the GGO portion and the solid portion, respectively, using deep learning technology in a tumor imaging image including GGO, the size of the GGO portion and the solid portion can be quickly and accurately measured. Therefore, in lung cancer staging, it is possible to accurately determine the T (Tumor) stage, which is an important determining factor of the patient's prognosis.

5 is a flowchart illustrating a method of measuring the size of a lung cancer tumor according to an embodiment of the present invention. The method shown in FIG. 5 may be performed, for example, by the lung cancer tumor size measuring apparatus 100 described above. In the illustrated flowchart, the method is described by dividing the method into a plurality of steps, but at least some of the steps are performed in a different order, combined with other steps, performed together, omitted, divided into detailed steps, or not shown. One or more steps may be added and performed.

Referring to FIG. 5, the lung cancer tumor size measuring apparatus 100 acquires a tumor photographing image of a tumor portion of a patient (S 101 ). Here, the tumor imaging image may be an image including GGO (Ground Glass Opacity).

Next, the lung cancer tumor size measuring apparatus 100 converts the tumor photographed image through the first window to generate a GGO portion image (S 103), and converts the tumor photographed image through the second window to generate a solid portion image. Generates (S 105). 6 is a state in which a GGO portion image is generated by applying a first window to a tumor photographed image (Fig. 6 (a)) in an embodiment of the present invention (Fig. 6 (b)) and a second window on the tumor photographed image. It is a diagram showing a state in which a solid potion image is generated by applying (Fig. 6(c)).

Next, the lung cancer tumor size measuring apparatus 100 resizes the GGO portion image to a preset size (S 107), and resizes the solid portion image to a preset size (S 109). The lung cancer tumor size measuring apparatus 100 may resize the GGO portion image and the solid portion image to the same size.

Next, the lung cancer tumor size measuring apparatus 100 detects the boundary of the GGO portion in the GGO portion image by inputting the GGO portion image to the first deep learning module (S111). The first deep learning module may be implemented in the lung cancer tumor size measuring apparatus 100, but is not limited thereto, and may be implemented as an external device or system interlocked with the lung cancer tumor size measuring apparatus 100.

Next, the lung cancer tumor size measuring apparatus 100 detects the boundary of the solid portion in the solid portion image by inputting boundary information of the GGO portion from the solid portion image and the GGO portion image using a second deep learning module (S113). The second deep learning module performs deep learning independent of the first deep learning module, and may be implemented in the lung cancer tumor size measuring device 100, but is not limited thereto and interlocks with the lung cancer tumor size measuring device 100 It may be implemented as an external device or system.

Next, the lung cancer tumor size measuring apparatus 100 measures the size of the GGO portion in the tumor photographing image based on the boundary of the GGO portion (S 115), and the size of the solid portion in the tumor photographing image based on the boundary of the solid portion. It measures (S117).

In the present specification, a module may mean a functional and structural combination of hardware for performing the technical idea of the present invention and software for driving the hardware. Predictably, the "module" may mean a logical unit of a predetermined code and a hardware resource for executing the predetermined code, and does not necessarily mean a physically connected code or a single type of hardware. .

7 is a diagram comparing boundary detection by deep learning and boundary detection by an expert according to an embodiment of the present invention. In FIG. 7, Gold Standard represents boundary detection by a lung cancer specialist.

Referring to FIG. 7, the GGO partial boundary and solid partial boundary detected by a lung cancer specialist (Gold Standard) in the tumor imaging image (original image), and the GGO partial boundary and solid detected by deep learning according to an embodiment of the present invention. You can see that the partial boundaries are almost the same.

8 is a block diagram illustrating and describing a computing environment 10 including a computing device suitable for use in example embodiments. In the illustrated embodiment, each component may have different functions and capabilities in addition to those described below, and may include additional components in addition to those described below.

The illustrated computing environment 10 includes a computing device 12. In an embodiment, the computing device 12 may be a lung cancer tumor size measuring device 100.

The computing device 12 includes at least one processor 14, a computer-readable storage medium 16 and a communication bus 18. The processor 14 may cause the computing device 12 to operate in accordance with the aforementioned exemplary embodiments. For example, the processor 14 may execute one or more programs stored in the computer-readable storage medium 16. The one or more programs may include one or more computer-executable instructions, and the computer-executable instructions are configured to cause the computing device 12 to perform operations according to an exemplary embodiment when executed by the processor 14. Can be.

The computer-readable storage medium 16 is configured to store computer-executable instructions or program code, program data, and/or other suitable form of information. The program 20 stored in the computer-readable storage medium 16 includes a set of instructions executable by the processor 14. In one embodiment, the computer-readable storage medium 16 includes memory (volatile memory such as random access memory, nonvolatile memory, or a suitable combination thereof), one or more magnetic disk storage devices, optical disk storage devices, flash It may be memory devices, other types of storage media that can be accessed by the computing device 12 and store desired information, or a suitable combination thereof.

The communication bus 18 interconnects the various other components of the computing device 12, including the processor 14 and computer readable storage medium 16.

Computing device 12 may also include one or more input/output interfaces 22 and one or more network communication interfaces 26 that provide interfaces for one or more input/output devices 24. The input/output interface 22 and the network communication interface 26 are connected to the communication bus 18. The input/output device 24 may be connected to other components of the computing device 12 through the input/output interface 22. The exemplary input/output device 24 includes a pointing device (such as a mouse or trackpad), a keyboard, a touch input device (such as a touch pad or a touch screen), a voice or sound input device, and various types of sensor devices and/or a photographing device. Input devices, and/or output devices such as display devices, printers, speakers, and/or network cards. The exemplary input/output device 24 may be included in the computing device 12 as a component constituting the computing device 12, and may be connected to the computing device 12 as a separate device distinct from the computing device 12. May be.

Although the exemplary embodiments of the present invention have been described in detail above, those of ordinary skill in the art to which the present invention pertains will understand that various modifications can be made to the above-described embodiments without departing from the scope of the present invention. . Therefore, the scope of the present invention is limited to the described embodiments and should not be determined, and should not be determined by the claims to be described later, but also by those equivalents to the claims.

100: lung cancer tumor size measuring device
102: image acquisition unit
104: first boundary detection unit
106: second boundary detection unit
108: size measurement unit
111: first image conversion unit
113: first image resizing unit
115: first deep learning unit
121: second image conversion unit
123: second image resizing unit
125: 2nd deep learning unit

Claims (11)

One or more processors, and
A computing device having a memory storing one or more programs executed by the one or more processors,
An image acquisition unit for obtaining a tumor photographing image of a patient's tumor portion;
A first boundary detection unit configured to detect a boundary of a GGO (Ground Glass Opacity) portion in the tumor photographed image through a first deep learning module;
A second boundary detector configured to detect a boundary of a solid portion in the tumor photographed image through a second deep learning module; And
A size measuring unit that measures the size of the GGO portion based on the detected boundary of the GGO portion, and measures the size of the solid portion based on the detected boundary of the solid portion,
The first boundary detection unit,
A first image conversion unit for generating a GGO portion image by converting the tumor photographed image through a first window having a preset first Hounsfield unit range based on a preset first window level; And
A first deep learning unit for receiving the GGO portion image and detecting a boundary of a GGO portion in the GGO portion image,
The second boundary detection unit,
A second image conversion unit generating a solid portion image by converting the tumor photographed image through a second window having a second hounsfield unit range based on a preset second window level; And
Computing device for measuring lung cancer tumor size, including a second deep learning unit that receives the boundary information of the solid portion image and the GGO portion and detects the boundary of the solid portion in the solid portion image.
delete The method according to claim 1,
The first window,
A computing device for measuring lung cancer tumor size, which is provided to have a preset first window width based on a preset first window level in units of Hounsfield.
The method according to claim 1,
The computing device,
A computing device for measuring lung cancer tumor size, further comprising a first image resizing unit for resizing the generated GGO portion image to a preset size and transferring the resized GGO portion image to the first deep learning unit.
The method according to claim 1,
The first deep learning unit,
Computing device for measuring lung cancer tumor size, including a convolution neural network (CNN) trained to detect and output a boundary of a GGO portion in the input GGO portion image when a GGO portion image is input.
delete The method of claim 3,
The second window,
A computing device for measuring the size of a lung cancer tumor, which is provided to have a second window width in a range narrower than the first window width based on a second window level having a Hounsfield unit value higher than the first window level.
The method according to claim 1,
The computing device,
A computing device for measuring lung cancer tumor size, further comprising a second image resizing unit for resizing the generated solid portion image to a preset size and transmitting the resized solid portion image to the second deep learning unit.
The method according to claim 1,
The second deep learning unit,
Lung cancer tumor size measurement including a convolution neural network (CNN) trained to detect and output the boundary of the solid portion in the solid portion image and when the boundary information of the GGO portion is input For computing devices.
One or more processors, and
A method performed in a computing device having a memory storing one or more programs executed by the one or more processors,
Obtaining a tumor photographing image of the patient's tumor portion;
Generating a Ground Glass Opacity (GGO) portion image by converting the tumor photographed image based on a preset first window level through a first window having a preset first Hounsfield unit range;
Detecting a boundary of a GGO portion in the GGO portion image through a first deep learning module;
Generating a solid portion image by converting the tumor photographed image based on a preset second window level through a second window having a preset second Hounsfield unit range;
Receiving boundary information of the solid portion image and the GGO portion and detecting a boundary of the solid portion in the solid portion image through a second deep learning module;
Measuring the size of the GGO portion based on the detected boundary of the GGO portion; And
And measuring the size of the solid portion based on the detected boundary of the solid portion.
As a computer program stored in a non-transitory computer readable storage medium,
The computer program includes one or more instructions, and when the instructions are executed by a computing device having one or more processors, the computing device causes:
Acquire a tumor photographing image of the patient's tumor part,
Generates a GGO (Ground Glass Opacity) portion image by converting the tumor photographed image based on a preset first window level through a first window having a preset first Hounsfield unit range,
Detecting the boundary of the GGO portion in the GGO portion image through the first deep learning module,
A solid portion image is generated by converting the tumor photographed image based on a preset second window level through a second window having a preset second Hounsfield unit range,
Receives the solid portion image and boundary information of the GGO portion and detects the boundary of the solid portion in the solid portion image through a second deep learning module,
Measuring the size of the GGO portion based on the boundary of the detected GGO portion, and
A computer program for measuring the size of a lung cancer tumor stored in a non-transitory computer-readable storage medium to measure the size of the solid portion based on the detected boundary of the solid portion.

Images