Method for calculating quantitative lesion information using picture archiving and communication system, and lesion quantification system

Abstract

A method for calculating quantitative lesion information using a picture archiving and communication system comprises the steps of: capturing, by a user terminal, a specific medical image that is currently output on a screen among medical images; acquiring, by the user terminal, a lesion region extracted from the captured specific medical image; acquiring, by the user terminal, positions of a plurality of points in the lesion region; transmitting, by the user terminal, the positions of the plurality of points to the picture archiving and communication system; and receiving, by the user terminal, quantitative information of the lesion region from the picture archiving and communication system.

Description

Method for producing quantitative information on lesions using a medical image storage and transmission system and a lesion quantitative system

[0001] The technology described below relates to a technique for automatically producing quantitative information of lesions in medical images.

[0002] Picture Archiving Communication Systems (PACS) store medical images and provide medical images and related information requested by user terminals. Currently, PACS provides integrated image processing, storage, and transmission. PACS stores images at a fixed size and can objectively calculate the length and size of each part or region within the image.

[0003] Interpreting medical images involves observing lesion characteristics, such as size, attenuation, and signal intensity. Users can select lesion boundaries on the image screen provided by the PACS to view information such as the length of the selected points.

[0004] Currently, lesion interpretation using PACS is performed by medical professionals. Because lesion area identification itself is performed manually, estimating lesion length and size is not objective and can be inaccurate.

[0005] Furthermore, 3D (dimensional) medical images are provided in sagittal, coronal, and axial planes, and if the direction of the lesion does not match a specific plane, the measurement of the long diameter of the lesion is inaccurate.

[0006] To address this issue, techniques exist that use medical image data to reconstruct lesion areas in 3D. However, this technique requires significant time for 3D modeling, making it difficult to apply clinically.

[0007] The technology described below aims to automatically derive quantitative information, such as the diameter of a lesion, from medical images. The technology uses a segmentation model to extract lesion boundaries and utilizes PACS to automatically derive quantitative information about the lesion.

[0008] A method for producing quantitative information on a lesion using a medical image storage and transmission system includes a step of a user terminal capturing a specific medical image currently displayed on a screen among medical images, a step of the user terminal obtaining a lesion area extracted from the captured specific medical image, a step of the user terminal obtaining positions of a plurality of points in the lesion area, a step of the user terminal transmitting the positions of the plurality of points to the medical image storage and transmission system, and a step of the user terminal receiving quantitative information on the lesion area from the medical image storage and transmission system.

[0009] The technology described below extracts lesion regions using a separate segmentation model to produce consistent measurement results. This technology can be implemented independently of the PACS system type by transmitting lesion region information extracted using the segmentation model to the PACS.

[0010] Figure 1 is an example of a PACS-based lesion information production system.

[0011] Figure 2 is another example of a PACS-based lesion information generation system.

[0012] Figure 3 is another example of a PACS-based lesion information generation system.

[0013] Figure 4 is an example of a process for generating lesion information using PACS.

[0014] Figure 5 is an example of a screen and interface for calculating lesion length on a user terminal.

[0015] Figure 6 is an example of a user terminal that produces lesion information.

[0016] The technology described below is susceptible to various modifications and embodiments. Specific embodiments are illustrated and described in detail in the drawings. However, this does not limit the technology described below to specific embodiments, and it should be understood that all modifications, equivalents, and alternatives fall within the spirit and scope of the technology described below.

[0017] Terms such as first, second, A, and B may be used to describe various components, but these components are not limited by these terms and are used solely to distinguish one component from another. For example, without departing from the scope of the technology described below, a first component may be referred to as a second component, and similarly, a second component may also be referred to as a first component. The term "and / or" includes a combination of a plurality of related described items or any of a plurality of related described items.

[0018] As used herein, the singular expressions should be understood to include the plural expressions unless the context clearly dictates otherwise, and the term "comprises" and the like should be understood to mean the presence of a described feature, number, step, operation, component, part, or combination thereof, but not to exclude the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0019] Before going into a detailed description of the drawings, it should be made clear that the division of components in this specification is merely a division based on the main function of each component. In other words, two or more components described below may be combined into a single component, or a single component may be further subdivided into two or more components with more detailed functions. In addition to its own main function, each component described below may additionally perform some or all of the functions of other components, and of course, some of the main functions of each component may be exclusively performed by other components.

[0020] Additionally, in performing a method or method of operation, each process constituting the method may occur in a different order than the stated order, unless the context clearly indicates a specific order. That is, each process may occur in the same order as the stated order, may be performed substantially simultaneously, or may be performed in the opposite order.

[0021] The technology described below assumes a conventional PACS. The PACS can be any system built by any manufacturer. The technology described below can calculate lesion size, length, or diameter regardless of the type of PACS.

[0022] PACS essentially stores medical images in a fixed size. PACS can provide various information about medical images. For example, PACS can provide information such as the length and diameter of specific areas or points selected by the user. As mentioned above, PACS can calculate the physical distance between points selected by the user in the medical image. This allows medical professionals to obtain quantitative information, such as the length of a specific lesion.

[0023] Quantitative information of a lesion is information related to the length or size of the lesion. The quantitative information of the lesion may include at least one of various indices. For example, the quantitative information of the lesion may include at least one of statistical values ​​(average, standard deviation, etc.) for the unidirectional length, unidirectional short axis, general long axis, and overall unidirectional length in the lesion region. In this case, the unidirectional length can be set by the user viewing the medical image on the screen interface. Furthermore, the quantitative information of the lesion may further include length information based on the vertical direction of the general direction. Furthermore, the quantitative information of the lesion may include the planar area of ​​the entire lesion region.

[0024] However, for the convenience of the following explanation, the explanation will focus on the length or diameter of the lesion.

[0025] A medical image may be at least one of a variety of image types. Medical images may include images such as x-ray images, computed tomography (CT), magnetic resonance imaging (MRI), positron emission tomography (PET), and ultrasound images.

[0026] The technique described below utilizes a segmentation model to identify specific regions of interest in medical images. Various segmentation models have been studied. Examples include U-net, SegNet, and FCN (Fully Convolutional Network). In the following description, the segmentation model can be any of these various types of models.

[0027] The following describes a data processing device that performs segmentation on medical images. The data processing device can extract specific lesion areas using a segmentation model. The data processing device can be implemented as a mobile terminal such as a smart device, a PC, a network server, or a chipset embedded with a dedicated program.

[0028] The user terminal, as described below, is a device that outputs medical images provided via a PACS. The user terminal can receive specific commands or requests from the user. The user terminal can also receive a request from the user to obtain quantitative information about lesions in the currently output medical image. In some cases, the user terminal may extract lesions from the current medical image using a segmentation model. Of course, the user terminal and the data processing device that extracts lesions may be physically separate devices.

[0029] Fig. 1 is an example of a PACS-based lesion information generation system (100). The lesion information generation system (100) of Fig. 1 includes a PACS (110) and a user terminal (120).

[0030] PACS (110) can store various types of medical images. PACS (110) stores patient identification information, medical image capture times, and medical images. PACS (110) stores medical images for multiple patients. PACS (110) provides a function for measuring the actual length of two points input through a user terminal (120).

[0031] A user terminal (120) is a device used by users (e.g., medical staff) utilizing medical images. The user terminal (120) corresponds to a computer device that outputs medical images, processes images, and receives user commands. The user terminal (120) may be implemented in the form of a PC, laptop, smartphone, or dedicated terminal.

[0032] A user terminal (120) requests a specific medical image for a specific patient from the PACS (110). The patient may be identified by identification information (patient number, personal identification number, etc.). The medical image may be identified by information such as patient identification information, shooting date, shooting time, shooting area, etc.

[0033] The user terminal (120) receives the requested medical image from the PACS (110). The medical image may be composed of multiple slices. The user terminal (120) may display a specific slide of the image or a specific region of the entire frame on the screen according to a user interface command.

[0034] The user terminal (120) can receive a lesion extraction command from the medical image currently displayed on the screen among the received image data. The user terminal (120) can extract the lesion area using a pre-trained segmentation model. Therefore, in the system (100) of FIG. 1, the user terminal (120) corresponds to a data processing device.

[0035] The user terminal (120) can extract a lesion area based on a captured image of the medical image currently displayed on the screen. The captured image may differ in resolution or size from the medical image transmitted from the PACS (110).

[0036] Segmentation models can be pre-trained models based on the type of medical image, body part, and disease. (1) Multiple segmentation models can be built based on the type of medical image (e.g., CT, MRI). Multiple segmentation models can be built based on the dimensionality of the image (2D or 3D). (2) Different segmentation models can be built based on the type of disease (e.g., brain lesion, pulmonary nodule).

[0037] Meanwhile, the user terminal (120) can visually display the extracted lesion area by overlaying it on the current screen.

[0038] The user terminal (120) can determine points to be subject to size or length measurement in the extracted lesion area.

[0039] The length to be measured may be a straight line or a curved line. In this case, the straight line may be at least one of the longest straight line (major axis straight line) in the lesion area (lesion mask), a straight line from the center of the lesion area to the border of the lesion area, and a straight line defined by any two points within the lesion area. The curved line may be at least a portion of the border of the lesion area.

[0040] The user terminal (120) can input the positions of points defining a straight line or curve to be measured. The user terminal (120) can input the positions of points defining a straight line or curve to be measured through an input of an interface device (e.g., a mouse). For example, the user terminal (120) can input the positions of two points defining a straight line to be measured, the positions of end points of a curve, or consecutive points defining a curve. The user terminal (120) can also input positions defining multiple straight lines in a lesion area.

[0041] Furthermore, in a lesion region, the measurement target may be the area or volume of a specific region. In this case, the measurement target may also be identified by multiple points. For example, a specific region may be identified by multiple points located at the region's borders. Alternatively, a specific region may be identified by a straight line along the region's major axis or a straight line from the center to the border.

[0042] Alternatively, the user terminal (120) may automatically determine a specific straight line in the lesion area (lesion mask). The user terminal (120) may determine a major axis straight line, a straight line from the center to the boundary, etc. in the lesion area through an image processing algorithm. For example, the user terminal (120) may determine a major axis straight line for all possible pixel pairs (two pixels) among the pixels constituting the boundary of the lesion in a 2D plane or 3D space. The user terminal (120) may determine the positions of the two pixels (major axis points) constituting the major axis straight line.

[0043] The locations of points that define (identify) the straight line, curve, or area to be measured are referred to as point of interest information.

[0044] At this time, the point of interest information is a point determined based on the captured image. The point of interest information does not refer to a location in the original image domain stored in the PACS (110), but rather to a location in the frame domain output from the user terminal (120).

[0045] PACS (110) receives point of interest information from a user terminal (120). Based on the point of interest information, PACS (110) can determine the length, width, etc. of actual lesions determined by the points. Based on the point of interest information, PACS (110) can calculate the major axis length, radius length, boundary length, width, or volume of a physical lesion in a stored medical image space.

[0046] The resolution of the original image of the PACS (110) and the resolution of the output image of the user terminal (120) may be different. Therefore, the PACS (110) must map the received point of interest information (location of points) to a location in the original medical image space. Thereafter, the PACS (110) can estimate the length, area, or volume of the region of interest using a built-in algorithm or program. The PACS (110) can also produce additional quantitative information about the region of interest (minimum length, maximum length, average length, standard deviation of length, etc.).

[0047] The user terminal (120) can receive quantitative information (length, width, volume, etc.) about the lesion area from the PACS (110). The user terminal (120) can output the quantitative information about the lesion area together with the current medical image.

[0048] Fig. 2 is another example of a PACS-based lesion information generation system (200). The lesion information generation system (200) of Fig. 2 includes a PACS (210), a user terminal (220), and a service server (230).

[0049] PACS (210) can store various types of medical images. PACS (210) stores patient identification information, medical image capture times, and medical images. PACS (210) stores medical images for multiple patients. PACS (210) provides a function for measuring the actual length of two points input through a user terminal (220).

[0050] A user terminal (220) is a device used by users (e.g., medical staff) utilizing medical images. The user terminal (220) corresponds to a computer device that outputs medical images, processes images, and receives user commands. The user terminal (220) may be implemented in the form of a PC, laptop, smartphone, or dedicated terminal.

[0051] A user terminal (220) requests a specific medical image for a specific patient from the PACS (210). The patient may be identified by identification information (patient number, personal identification number, etc.). The medical image may be identified by information such as patient identification information, date of capture, time of capture, and area of ​​capture.

[0052] The user terminal (220) receives the requested medical image from the PACS (210). The medical image may be composed of multiple slices. The user terminal (220) may display a specific slide of the image or a specific region of the entire frame on the screen according to a user interface command.

[0053] The user terminal (220) can receive a command to extract a lesion from a medical image currently displayed on the screen. The user terminal (120) can generate an image by capturing the medical image currently displayed on the screen. The captured image may differ in resolution or size from the medical image transmitted from the PACS (110).

[0054] The user terminal (220) can transmit the captured image to the service server (230). The service server (230) is a device that extracts a lesion area from the captured image using a segmentation model.

[0055] The service server (230) can extract a lesion area using a pre-trained segmentation model. Therefore, in the system (200) of FIG. 2, the service server (230) corresponds to a data processing device.

[0056] Segmentation models can be pre-trained models based on the type of medical image, body part, and disease. (1) Multiple segmentation models can be built based on the type of medical image (e.g., CT, MRI). Multiple segmentation models can be built based on the dimensionality of the image (2D or 3D). (2) Different segmentation models can be built based on the type of disease (e.g., brain lesion, pulmonary nodule).

[0057] The service server (230) can transmit information on the lesion area extracted from the captured image to the user terminal (220). In this case, the user terminal (220) can determine points in the lesion area that are subject to size or length measurement.

[0058] The user terminal (220) can visually display the extracted lesion area by overlaying it on the current screen.

[0059] At this time, the user terminal (120) can determine points to be subject to size or length measurement in the extracted lesion area.

[0060] The length to be measured may be a straight line or a curved line. In this case, the straight line may be at least one of the longest straight line (major axis straight line) in the lesion area (lesion mask), a straight line from the center of the lesion area to the border of the lesion area, and a straight line defined by any two points within the lesion area. The curved line may be at least a portion of the border of the lesion area.

[0061] The user terminal (120) can input the positions of points defining a straight line or curve to be measured. The user terminal (120) can input the positions of points defining a straight line or curve to be measured through an input of an interface device (e.g., a mouse). For example, the user terminal (120) can input the positions of two points defining a straight line to be measured, the positions of end points of a curve, or consecutive points defining a curve. The user terminal (120) can also input positions defining multiple straight lines in a lesion area.

[0062] Furthermore, in a lesion region, the measurement target may be the area or volume of a specific region. In this case, the measurement target may also be identified by multiple points. For example, a specific region may be identified by multiple points located at the region's borders. Alternatively, a specific region may be identified by a straight line along the region's major axis or a straight line from the center to the border.

[0063] Alternatively, the user terminal (120) may automatically determine a specific straight line in the lesion area (lesion mask). The user terminal (120) may determine a major axis straight line, a straight line from the center to the boundary, etc. in the lesion area through an image processing algorithm. For example, the user terminal (120) may determine a major axis straight line for all possible pixel pairs (two pixels) among the pixels constituting the boundary of the lesion in a 2D plane or 3D space. The user terminal (120) may determine the positions of the two pixels (major axis points) constituting the major axis straight line.

[0064] The user terminal (220) transmits point of interest information to the PACS (210). The point of interest information is composed of the locations of points that define (identify) a straight line, curve, or area to be measured.

[0065] Meanwhile, the service server (230) can determine points to be measured for size or length in the extracted lesion area. For example, the service server (230) can determine the longest straight line in the lesion area (lesion mask). The service server (230) can determine the major axis straight line in the lesion area through an image processing algorithm. For example, the service server (230) can determine the major axis straight line for all possible pixel pairs (two pixels) among the pixels constituting the boundary of the lesion in a 2D plane or 3D space. The service server (230) can determine the positions (major axis point information) of the two pixels constituting the major axis straight line. The service server (230) can transmit the major axis point information to the user terminal (220).

[0066] Furthermore, the service server (230) may determine points defining a straight line from the center of the lesion area to the boundary, points defining at least part of the boundary of the lesion area, and a plurality of points identifying the lesion area.

[0067] The service server (230) can transmit interest designation information for points defining various straight lines or areas to the user terminal (220).

[0068] At this time, the point of interest information is a point determined based on the captured image. The point of interest information does not refer to a location in the original image domain stored in the PACS (110), but rather to a location in the frame domain output from the user terminal (120).

[0069] The PACS (210) receives point-of-interest information from the user terminal (220). The PACS (210) can determine the length of an actual lesion determined by the points based on the point-of-interest information. The PACS (210) can measure the length (major axis) of a physical lesion based on the point-of-interest information in the stored medical image space. The PACS (210) can calculate the major axis length, radius length, boundary length, area, or volume of a physical lesion based on the point-of-interest information in the stored medical image space.

[0070] The resolution of the original image of the PACS (210) and the resolution of the output image of the user terminal (220) may be different. Therefore, the PACS (210) must map the received point of interest information (location of points) to a location in the original medical image space. Thereafter, the PACS (210) can estimate the length, area, or volume of the region of interest using a built-in algorithm or program. The PACS (110) can also produce additional quantitative information about the region of interest (minimum length, maximum length, average length, standard deviation of length, etc.).

[0071] The user terminal (220) can receive quantitative information (length, width, volume, etc.) about the lesion area from the PACS (110). The user terminal (220) can output the quantitative information about the lesion area along with the current medical image.

[0072] Fig. 3 is another example of a PACS-based lesion information generation system (300). The lesion information generation system (300) of Fig. 3 includes a PACS (310) and a user terminal (320).

[0073] The PACS (310) can store various types of medical images. It stores patient identification information, medical image capture times, and medical images. It stores medical images for multiple patients. The PACS (310) provides a function for measuring the actual lengths of two points input through a user terminal (320).

[0074] A user terminal (320) is a device used by users (e.g., medical staff) utilizing medical images. The user terminal (320) corresponds to a computer device that outputs medical images, processes images, and receives user commands. The user terminal (320) may be implemented in the form of a PC, laptop, smartphone, or dedicated terminal.

[0075] A user terminal (320) requests a specific medical image for a specific patient from the PACS (310). The patient may be identified by identification information (patient number, personal identification number, etc.). The medical image may be identified by information such as patient identification information, date of capture, time of capture, and area of ​​capture.

[0076] The user terminal (320) receives the requested medical image from the PACS (310). The medical image may be composed of multiple slices. The user terminal (320) may display an image of a specific point or slice on the screen according to a user interface command.

[0077] The user terminal (320) can receive a command to extract a lesion from a currently output medical image. The user terminal (320) can generate an image by capturing the currently output medical image. The captured image may differ in resolution or size from the medical image transmitted from the PACS (310).

[0078] The user terminal (320) can transmit the captured image to the PACS (310). The PACS (310) is a device that extracts a lesion area from the captured image using a segmentation model.

[0079] The control server (315) of the PACS (310) can extract a lesion area using a pre-learned segmentation model. Therefore, in the system (300) of FIG. 3, the control server (315) corresponds to a data processing device.

[0080] Segmentation models can be pre-trained models based on the type of medical image, body part, and disease. (1) Multiple segmentation models can be built based on the type of medical image (e.g., CT, MRI). Multiple segmentation models can be built based on the dimensionality of the image (2D or 3D). (2) Different segmentation models can be built based on the type of disease (e.g., brain lesion, pulmonary nodule).

[0081] The control server (315) can determine points in the extracted lesion area that are subject to size or length measurement.

[0082] The length to be measured may be a straight line or a curved line. In this case, the straight line may be at least one of the longest straight line (major axis straight line) in the lesion area (lesion mask), a straight line from the center of the lesion area to the border of the lesion area, and a straight line defined by any two points within the lesion area. The curved line may be at least a portion of the border of the lesion area.

[0083] For example, the control server (315) can determine the longest straight line in the lesion area (lesion mask). The control server (315) can determine the major axis straight line in the lesion area through an image processing algorithm. For example, the control server (315) can determine the major axis straight line for all possible pixel pairs (two pixels) among the pixels constituting the boundary of the lesion in a 2D plane or 3D space. The control server (315) can determine the positions (major axis point information) of the two pixels constituting the major axis straight line.

[0084] Furthermore, the control server (315) may determine points defining a straight line from the center of the lesion area to the boundary, points defining at least part of the boundary of the lesion area, and a plurality of points identifying the lesion area.

[0085] The PACS (310) can determine the length of an actual lesion determined by points of interest based on information about the points of interest determined by the control server (315) based on the captured image. The PACS (310) can measure the length (major axis) of a physical lesion based on information about the point of interest in the stored medical image space. The PACS (210) can calculate the major axis length, radius length, boundary length, area, or volume of a physical lesion based on information about the point of interest in the stored medical image space.

[0086] The resolution of the original image of the PACS (310) and the resolution of the output image of the user terminal (320) may be different. Therefore, the PACS (310) must map the received point of interest information (location of points) to a location in the original medical image space. Thereafter, the PACS (310) can estimate the length, area, or volume of the region of interest using a built-in algorithm or program. The PACS (310) can also produce additional quantitative information about the region of interest (minimum length, maximum length, average length, standard deviation of length, etc.).

[0087] The user terminal (320) can receive quantitative information (length, width, volume, etc.) about the lesion area from the PACS (310). The user terminal (320) can output the quantitative information about the lesion area along with the current medical image.

[0088] The user terminal (320) can visually display the extracted lesion area by overlaying it on the current screen. The user terminal (320) can output quantitative information about the lesion area along with the current medical image.

[0089] Figure 4 is an example of a process (400) for generating lesion information using PACS.

[0090] The user terminal requests a medical image of a specific patient from the PACS (410).

[0091] The user terminal receives medical image data from the PACS. The user terminal receives control commands for medical image output from the user. These control commands may include commands such as selecting a specific region, selecting a specific slice, and zooming in / out of the screen. The user terminal outputs a specific medical image to the screen according to the user control command (420).

[0092] The user terminal may receive a request for lesion detection or quantitative information on a medical image currently displayed on the screen. In this case, the user terminal may capture the medical image displayed on the screen (430).

[0093] The user terminal can extract the lesion area from the captured image using a segmentation model (440). Alternatively, the user terminal can request lesion area extraction while transmitting the captured image to another device. As described above, the service server or the PACS control server can extract the lesion area from the captured image using a segmentation model (440).

[0094] The user terminal can determine point-of-interest information, which is a measurement target, in the lesion area based on the extracted lesion area (450). For example, the user terminal can receive point-of-interest information from the user or automatically generate interest designation information.

[0095] Alternatively, the service server or control server may determine information on points of interest that are subject to measurement in the lesion area (450). In the latter case, the service server or control server may transmit directions or points for measuring length in the lesion area to the user terminal.

[0096] The user terminal can transmit point of interest information to the PACS (460).

[0097] The PACS can calculate quantitative information about the lesion based on the transmitted point of interest information (470). The quantitative information may be at least one of statistical values ​​for the short axis, long axis, and length in a specific direction. Alternatively, the quantitative information may be at least one of statistical values ​​for the area, volume, or area or volume of the region of interest. The PACS can provide quantitative information about the lesion to the user terminal (470).

[0098] Figure 5 is an example of a screen and interface for calculating lesion length on a user terminal.

[0099] Figure 5(A) illustrates an example of a process for selecting a lesion from a medical image. The user terminal receives a medical image of a specific patient from the PACS. The user terminal can receive a specific medical image from the PACS by entering specific patient information and medical image information in the "Image Request" or "Image Loading" menus.

[0100] The user terminal can input the current medical image or captured image into a segmentation model to automatically extract a specific lesion area. Alternatively, the user terminal can also pass the current medical image or captured image to another object that performs a segmentation function to extract a specific lesion area.

[0101] Meanwhile, the user terminal can input a region of interest (ROI) rather than the entire image displayed on the screen into the segmentation model. The user terminal can select a lesion selection command and set the region where the lesion is estimated to be located. The user terminal can set a rectangular ROI, as shown in Fig. 5(A). In this case, the segmentation model can only receive and process the set ROI. Alternatively, the segmentation model can extract the lesion region centered on the ROI in the medical image. In other words, the segmentation model can extract the lesion by providing attention to the ROI.

[0102] Figure 5(B) illustrates an example of the process for calculating longitudinal information from an extracted lesion region. Figure 5(B) assumes that a lesion region has been extracted from a medical image. It is assumed that the user terminal has acquired the extracted lesion region information. Furthermore, it is assumed that the user terminal has acquired the locations of points defining the longitudinal axis (lateral axis point information) in the lesion region through the aforementioned process.

[0103] The user terminal can request lesion information for the extracted lesion through the "Calibrate lesion" menu. In Figure 5(B), the lesion point information includes the coordinate locations of points a and b. The lesion point information is automatically determined within the extracted lesion area. In this case, the lesion point information is similar to information determined by the user selecting (e.g., clicking) points a and b on the screen.

[0104] The user terminal transmits the longitudinal point information to the PACS. The user terminal transmits the longitudinal point information through a user solution linked to the PACS. That is, the user terminal transmits the longitudinal point information, automatically determined, to the PACS, similar to the user's action of directly selecting a screen. The user terminal receives the length information between the selected points (a and b) from the PACS. The user terminal can output longitudinal information (e.g., longitudinal line and / or longitudinal length) along with the medical image, as shown in Fig. 5(B).

[0105] Figure 6 illustrates an example of a user terminal (500) that generates lesion information. The user terminal (500) corresponds to the user terminal (120, 220, or 320) described above. The user terminal (500) may be physically implemented in various forms. For example, the user terminal (500) may take the form of a PC, laptop, smartphone, VR device, smart glasses, a data processing-dedicated chipset, etc.

[0106] The user terminal (500) may include a storage device (510), a memory (520), a computing device (530), an interface device (540), a communication device (550), and an output device (560).

[0107] The storage device (510) can store specific medical images.

[0108] The storage device (510) can store a segmentation model that extracts a lesion area. The segmentation model may be any of various types of models. As described above, the segmentation model is a pre-trained model.

[0109] The storage device (510) can store a lesion area or mask extracted by the segmentation model.

[0110] The storage device (510) can store quantitative information (length, short axis, long axis, radius, area, volume, etc.) about the lesion area.

[0111] The memory (520) can store data and information generated during the process of the user terminal producing lesion information.

[0112] The interface device (540) is a device that receives certain commands and data from the outside.

[0113] The interface device (540) can receive medical images from a physically connected input device or an external storage device.

[0114] The interface device (540) can receive medical images from a medical image storage and transmission system.

[0115] The interface device (540) can also receive information on a specific region of interest set in a medical image displayed on the screen.

[0116] The interface device (540) may also receive a learned segmentation model.

[0117] The interface device (540) may also receive lesion area information (lesion mask) extracted from a medical image.

[0118] The interface device (540) may also receive the locations of locations or points that are targets of quantitative information measurement in the lesion area.

[0119] The interface device (540) can receive quantitative information (length, short axis, long axis, and statistical values ​​of length) produced in the lesion area.

[0120] Meanwhile, the interface device (540) may also receive data or information transmitted via the communication device (550) below.

[0121] A communication device (550) refers to a configuration that receives and transmits certain information through a wired or wireless network.

[0122] The communication device (550) can transmit a specific medical image request to the PACS.

[0123] The communication device (550) can receive medical images from an external object.

[0124] The communication device (550) can receive medical images from a medical image storage and transmission system.

[0125] The communication device (550) can receive the locations of points of interest in medical images.

[0126] The communication device (550) can receive a segmentation model constructed from an external object.

[0127] The communication device (550) can receive lesion area information (lesion mask) extracted from an external object (such as a separate service server).

[0128] The communication device (550) can transmit point of interest information for a quantitative information measurement target in the lesion area to the PACS.

[0129] The communication device (550) can receive quantitative information (length, short axis, long axis, radius, area, volume, or statistical values ​​of each item) of a lesion determined as point of interest information from PACS.

[0130] The output device (560) is a device that outputs certain information. The output device (560) can output interfaces required for the process of generating lesion information, lesion areas in medical images, specific areas in medical images, locations of points of interest, quantitative information, etc.

[0131] The computing device (530) can preprocess medical images of a specific patient on a regular basis.

[0132] The operation device (530) can capture a specific area to be output to the output device (560) and produce an image of the specific area.

[0133] The computational device (530) can extract a lesion region from a medical image or a medical image in which a region of interest is set. The computational device (530) can extract a lesion region from a captured specific region image using a segmentation model.

[0134] The computational device (530) can determine the positions of two points defining a one-way straight line based on the boundary of the lesion area.

[0135] The computational device (530) can determine a plurality of point pairs defining a plurality of lengths in one direction based on the boundary of the lesion area. For example, the computational device (530) can calculate the positions of the major axis points forming the major axis (major axis point information).

[0136] The computational device (530) can control the quantitative information of the lesion area to be output by overlaying it on the screen of the current medical image.

[0137] The computing device (530) may be a device such as a processor, AP, or chip embedded with a program that processes data and performs certain operations.

[0138] Additionally, the method for calculating lesion information and the method for calculating the length or diameter of a lesion, as described above, may be implemented as a program (or application) including an executable algorithm that can be executed on a computer. The program may be stored and provided on a temporary or non-transitory computer-readable medium.

[0139] A non-transitory readable medium refers to a medium that stores data semi-permanently and can be read by a device, rather than a medium that stores data for a short period of time, such as a register, cache, or memory. Specifically, the various applications or programs described above may be stored and provided on a non-transitory readable medium, such as a CD, DVD, hard disk, Blu-ray disk, USB, memory card, ROM (read-only memory), PROM (programmable read only memory), EPROM (Erasable PROM, EPROM), EEPROM (Electrically EPROM), or flash memory.

[0140] Temporarily readable media refers to various types of RAM, such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDR SDRAM), Enhanced SDRAM (ESDRAM), Synchronous DRAM (Synclink DRAM, SLDRAM), and Direct Rambus RAM (DRRAM).

[0141] The present embodiment and the drawings attached to the present specification only clearly illustrate a part of the technical idea included in the above-described technology, and it is obvious that all modified examples and specific embodiments that can be easily inferred by a person skilled in the art within the scope of the technical idea included in the specification and drawings of the above-described technology are included in the scope of the rights of the above-described technology.

Claims

1. A step in which a user terminal receives a medical image from a medical image storage and transmission system; A step in which the user terminal captures a specific medical image currently displayed on the screen among the medical images; A step in which the user terminal obtains a lesion area extracted from the captured specific medical image using a segmentation model; A step in which the user terminal acquires the positions of a plurality of points in the lesion area; A step in which the user terminal transmits the locations of the plurality of points to the medical image storage and transmission system; and A method for calculating quantitative information on a lesion using a medical image storage and transmission system, comprising a step of the user terminal receiving quantitative information on the lesion area from the medical image storage and transmission system.

2. In paragraph 1, A method for producing quantitative information on lesions using a medical image storage and transmission system, wherein the above segmentation model is a model learned in advance according to the type of medical image, the type of body part included in the medical image, or the type of disease.

3. In paragraph 1, The above lesion area is (i) A method for producing lesion quantitative information using a medical image storage and transmission system, wherein the information is produced using the segmentation model possessed by the user terminal, or (ii) a separate device that received the captured specific medical image or the medical image storage and transmission system produces the information using the segmentation model.

4. In paragraph 1, A method for producing quantitative information on a lesion using a medical image storage and transmission system, wherein the plurality of points include at least a portion of the major axis, minor axis, radius, and boundary border of the lesion area and at least one of a plurality of points located on the boundary.

5. In paragraph 1, A method for producing lesion quantitative information using a medical image storage and transmission system, wherein the above quantitative information includes at least one of statistical values ​​of length, short axis, long axis, radius, area, volume, and length value for the lesion area. 6.(i) Receive medical images from the medical image storage and transmission system, (ii) Transmitting the locations of points of the area of ​​interest that are the measurement target in the lesion area of ​​the medical image to the medical image storage and transmission system; (iii) a communication device that receives quantitative information on the lesion area from the medical image storage and transmission system; An output device that outputs a specific area of ​​the above medical image and a lesion area among the specific areas; A computing device that captures the specific area output from the output device and generates an image of the captured specific area; and Including an interface device that receives the positions of points of the area of ​​interest that are the measurement target in the lesion area output from the above output device, The above lesion area is a terminal device that provides lesion quantitative information calculated by (i) the computing device using a segmentation model from the specific area image, or (ii) a separate device that received the specific medical image or the medical image storage and transmission system using a segmentation model.

7. In paragraph 6, The above segmentation model is a terminal device that provides lesion quantitative information, which is a model learned in advance according to the type of medical image, the type of body part included in the medical image, or the type of disease.

8. In paragraph 6, A terminal device providing lesion quantitative information, wherein the above points include at least a portion of the major axis, minor axis, radius, and boundary border of the lesion area and at least one of a plurality of points located on the boundary.

9. In paragraph 6, A terminal device providing lesion quantitative information, wherein the quantitative information includes at least one of statistical values ​​of length, short axis, long axis, radius, area, volume, and length value for the lesion area.

Images