WO2019027166A1

WO2019027166A1 - Data generation system for reading dental image and read data generation system for dental image

Info

Publication number: WO2019027166A1
Application number: PCT/KR2018/008260
Authority: WO
Inventors: 송인석; 정규환
Original assignee: 주식회사 뷰노; 고려대학교 산학협력단
Priority date: 2017-08-01
Filing date: 2018-07-23
Publication date: 2019-02-07

Abstract

A data generation system for reading a dental image according to an embodiment can comprise: a storage unit in which a dental image, diagnosis information, and result data are stored; an input unit in which a first input comprising selection of an area of the dental image and a second input comprising selection of a diagnosis name corresponding to the selected area; a processing unit which generates the result data; and a display unit which displays the dental image and the result data. A read data generation system for dental image according to another embodiment can comprise: a storage unit in which an input dental image and a reference dental image are stored; a processing unit which detects an area with a lesion from the input dental image on the basis of the reference dental image, recognizing a position of the area with the lesion, and generates diagnostic information or an output dental image for the area with the lesion; and a display unit which displays the diagnostic information or the output dental image.

Description

Data generation system for reading dental images and read data generation system for dental images

Embodiments relate to a data generation system for reading a dental image and a read data generation system for a dental image.

Reading images in the medical field is important for diagnosis and treatment. Particularly, unlike general medical images, the anatomical structure of the oral cavity in which the multi-disease is mixed is often photographed in the dental image, so that a complex judgment of the dentist is often required. Tools for generating user-friendly and high-quality result data are being developed when users read an image of a maxilla (mandilla) and a mandible (mandible), which is an anatomical structure of the oral cavity, that is, a dental image.

For example, Korean Patent Laid-Open No. 10-2007-0083645 discloses a user interface for CT scan analysis.

In addition, systems have been developed to assist in the diagnosis of a dentist by discovering a disease from a dental image and producing medically meaningful reading data for the dentist.

For example, Korean Patent Laid-Open Publication No. 10-2000-0049915 discloses a medical image reading system and method through the Internet.

An object according to an embodiment is to provide a data generation system for dental image reading that produces result data used as data for machine learning.

An object of an embodiment of the present invention is to provide a data generation system for reading a dental image that is easy to use and produces high-quality result data for a user.

An object according to an exemplary embodiment is to learn a series of reading patterns through machine learning from data including dental images read by a user, automatically analyze a dental image not read by a user, And a system for generating a read data of a dental image that generates information in the form of a report.

The data generation system for reading a dental image according to an embodiment includes diagnostic information including a dental image of anatomical structure of the oral cavity, an anatomical identification number of the anatomical tissue and a plurality of diagnosis names, A storage unit for storing the resultant data; An input for receiving a second input comprising a first input comprising selection of an area corresponding to the anatomical structure of the dental image and a selection of a diagnosis name corresponding to the selected area; A processing unit for generating result data for the dental image based on the first input and the second input; And a display unit for displaying the dental image and the resultant data.

The processing unit recognizes a region selected from the anatomical structure of the dental image when a first input is received from the user to the input unit, and the display unit displays the marker in the selected region together with the dental image.

The processing unit may match the anatomical identification number with the recognized area, and the display unit may further display the anatomical identification number.

The processing unit may match the diagnosis name to the recognized area when the second input is received from the user to the input unit, and the display unit may display the matched diagnosis name.

The processing unit counts the number of times the diagnosis name is selected from the second input received from the user to the input unit, and stores the counted number of times for the selected diagnosis name in the storage unit.

The processing unit sets the priority of the plurality of diagnosis names according to the counted number of times for the selected diagnosis name, and the display unit may further display a plurality of diagnosis names listed in accordance with the set priority.

Wherein the first input comprises a plurality of first sub icons each corresponding to the anatomical identification number and the second input comprises a plurality of second sub icons respectively corresponding to the plurality of diagnosis names have.

The processing unit can determine whether or not an additional dental image is required based on the generated result data.

The processing unit counts the number of times the result data has been generated, generates a pattern of the number of times the result data has been generated over time based on the number of times of generating the counted result data and the time of generating the result data.

According to an embodiment of the present invention, there is provided a system for generating a dental image read data including a storage unit storing an input dental image for an anatomical structure of the oral cavity and a reference dental image for analyzing the input dental image; Detecting an area having a lesion from the input dental image based on the reference dental image, recognizing the position of the area having the detected lesion, and diagnosing the area having the detected lesion, A processor for generating an output dental image for the subject; And a display unit for displaying the diagnostic information or the output dental image.

The processing unit may generate the diagnostic information or the output dental image by learning the neural network based on the input dental image and the reference dental image.

The neural network comprising: a generated neural network for generating an output dental image read from the input dental image; And comparing the generated output dental image with the reference dental image stored in the storage unit to determine whether the generated output dental image is an output dental image generated through the generated neural network or a reference dental image stored in the storage unit And the like.

The generated neural network may regenerate an output dental image read from the input dental image when the classified neural network discriminates the generated output dental image as the reference dental image stored in the storage unit.

The processing unit generates lesion finding information for the lesion area based on the reference dental image, and the display unit can display the lesion finding information together with the output dental image.

The processing unit may match the figure information including the point, the line, or the face to the lesion area, and the display unit may display the lesion area with matched figure information.

The display unit may display the diagnosis information as an identification number for the anatomy of the oral cavity, a diagnosis name for the lesion area, or a diagnostic code for the lesion area.

The display unit may further display a plurality of icons corresponding to anatomical structures of the oral cavity, and the display unit may display at least one or more of the plurality of icons as an identification number for the region having the lesion.

The data generation system for reading dental images according to an exemplary embodiment of the present invention is easy to use for a user and can generate a high quality reading result.

A data generation system for dental image reading according to an embodiment may generate a reading result used as data for machine learning.

The effects of the data generation system for dental image reading according to one embodiment are not limited to those mentioned above, and other effects not mentioned may be clearly understood by those skilled in the art from the following description.

The read data generation system of a dental image according to an embodiment assists the diagnosis of the dentist, thereby reducing the error rate of the dentist and shortening the diagnosis time.

The effects of the system for generating a read data of a dental image according to an embodiment are not limited to those mentioned above, and other effects not mentioned can be clearly understood by those skilled in the art from the following description.

FIG. 1 is a block diagram schematically illustrating a data generation system for reading a dental image according to an exemplary embodiment of the present invention.

2 is a flowchart schematically illustrating a process of generating result data by a data generation system for reading a dental image according to an exemplary embodiment.

FIGS. 3 and 4 are views schematically showing the operation state of a data generation system for reading a dental image according to an exemplary embodiment.

FIG. 5 is a flowchart schematically illustrating a method of arranging diagnosis names according to priorities by a data generation system for reading dental images according to an exemplary embodiment.

6 is a schematic view of a main screen of a data generation system for reading a dental image according to an embodiment.

7 and 8 are views schematically showing a feedback providing screen of a data generating system for reading a dental image according to an embodiment.

FIG. 9 is a flowchart schematically illustrating a method of changing an image display value of a dental image by a data generation system for reading a dental image according to an exemplary embodiment.

10 is a view schematically showing a state in which an image display value of a dental image is changed by a data generation system for reading a dental image according to an embodiment.

11 is a flowchart schematically illustrating a secure connection method of a data generation system for reading a dental image according to an exemplary embodiment.

FIG. 12 is a view schematically illustrating a data access system for reading a dental image according to an embodiment of the present invention.

13 is a block diagram schematically showing a read data generation system for a dental image according to an embodiment.

14 is a flowchart schematically illustrating a process of learning a method of reading a dental image in a read data generating system for a dental image according to an embodiment.

15 is a diagram schematically illustrating a model for reading a dental image in a read data generation system for a dental image according to an embodiment.

16 is a flowchart schematically illustrating a process of generating diagnosis information or an output dental image by the read data generation system of a dental image according to an embodiment.

17 is a view showing read data generated by a read data generating system of a dental image according to an embodiment.

18 is a flowchart schematically illustrating a process of generating diagnostic information or output dental images by a read data generation system for dental images according to an embodiment.

19 is a diagram showing read data generated by a read data generation system of a dental image according to an embodiment.

Hereinafter, embodiments will be described in detail with reference to exemplary drawings. It should be noted that, in adding reference numerals to the constituent elements of the drawings, the same constituent elements are denoted by the same reference numerals even though they are shown in different drawings. In the following description of the embodiments, detailed description of known functions and configurations incorporated herein will be omitted when it may make the best of an understanding clear.

In describing the components of the embodiment, terms such as first, second, A, B, (a), and (b) may be used. These terms are intended to distinguish the constituent elements from other constituent elements, and the terms do not limit the nature, order or order of the constituent elements. When a component is described as being "connected", "coupled", or "connected" to another component, the component may be directly connected or connected to the other component, Quot; may be " connected, " " coupled, " or " connected. &Quot;

The components included in any one embodiment and the components including common functions will be described using the same names in other embodiments. Unless otherwise stated, the description of any one embodiment may be applied to other embodiments, and a detailed description thereof will be omitted in the overlapping scope.

Referring to FIG. 1, a data generation system 100 for reading dental images according to an embodiment is for generating learning data for machine learning. The data generation system 100 includes an anatomical anatomy A region corresponding to the anatomical structure of the oral cavity may be selected from the dental image in which the structure is photographed to match the diagnosis name to the selected region to generate the result data for the dental image and display the result data together with the dental image. Here, the displayed result data is displayed for the user's operation, and the generated result data can be used as learning data for machine learning.

The dental image herein may refer to multi-dimensional data consisting of discrete image elements (e.g., pixels in a two-dimensional image and voxels in a three-dimensional image). For example, the image may include Xray, Computed Tomography (CT), Magnetic Resonance Imaging (MRI), ultrasound, and medical images of objects obtained by other medical imaging systems.

The subject may be a person or animal, or a part of a person or an animal, and the user may be a medical professional such as a doctor, a nurse, a clinical pathologist, a medical imaging expert, etc., But is not limited thereto.

The data generation system 100 for reading a dental image may include a storage unit 110, an input unit 120, a processing unit 130, and a display unit 140.

Diagnostic information including anatomical identification number of the anatomical structure, a plurality of diagnosis names, and result data for the dental image can be stored in the storage unit 110, respectively .

The anatomical structure of the oral cavity herein may indicate, but is not limited to, the structure including the maxillary and mandibular structures of the human body.

Here, the anatomical identification number of the oral cavity is given according to the ISO standard (ISO 3950) produced by the World Health Organization (WHO) or given according to the American System (or universal numbering system) But are not limited thereto. In the case of the identification numbering method according to the ISO standard, a tooth number is assigned to 1 to 8 divided into four quadrants based on the center line of the mouth. In the case of the identification number assigning method according to the American system, It rotates clockwise to the bottom, and the teeth number (including English) is given. For convenience of explanation, the description will be made using anatomical identification numbers (teeth) of the mouth according to the ISO standard (ISO 3950).

The diagnosis name herein may be, but not limited to, a diagnosis name according to the Korean standard disease / sign classification chart, a diagnosis name according to the international diagnosis standard, and the like.

The input unit 120 may receive input from a user. Wherein the input includes a first input comprising a selection of an area corresponding to an anatomical structure of the oral cavity of the dental image, a second input comprising a selection of a diagnosis name corresponding to the selected area, A fourth input for changing the image display value of the dental image, a fifth input for inputting a set code and a secret code for comparison, and the like, but the present invention is not limited thereto And the necessary number of inputs can be received from the user as required. The description of each input will be described later in detail.

The processing unit 130 may generate result data for the dental image based on the input of the user. Here, the resultant data may be expressed in a manner that graphic information, diagnostic information, and the like are displayed together with the dental image, or an identification number, a diagnosis name, a diagnostic code, and the like of the anatomical structure of the oral cavity are displayed as a table . The format of the result data will be described in detail with reference to FIG. 3 and FIG.

The display unit 140 may display the dental image stored in the storage unit 110 and the resultant data generated by the processing unit 130. The display unit 140 may output result data generated by a user using a data generation system 100 for reading a dental image, such as a user interface (UI), user information, or object information. For example, the display may include a CRT display, an LCD display, a PDP display, an OLED display, a FED display, an LED display, a VFD display, a DLP display, a PFD display, a 3D display, a transparent display, And may include various display devices within an apparent range. The various forms displayed by the display unit 140 will be described with reference to FIGS. 3 and 4, which will be described later.

Although not shown in FIG. 1, the data generation system 100 for reading a dental image may further include a communication unit. The communication unit can be connected to the network by wire or wireless and can communicate with an external device or an external server. For example, the communication unit may be a server operating on a cloud basis. When the communication unit is a server, the data generation system 100 for reading a dental image can generate result data based on a user's input on a web-based basis, have.

Referring to FIG. 2, a data generation system for reading a dental image may load (210) a dental image.

Thereafter, the data generation system for dental image reading may recognize 220 the area corresponding to the anatomical structure based on the first input. As described above, the first input may include selection of an area corresponding to the anatomical structure of the oral cavity in the dental image. For example, if a user views a dental image and selects the third tooth on the upper left side of the virtual center line of the anatomical structure of the oral cavity displayed on the dental image, The region corresponding to the third tooth on the upper left side selected by the user can be recognized.

Thereafter, the data generation system for dental image reading may match the diagnosis name to the area corresponding to the anatomical structure based on the second input (230). As described above, the second input may include selection of a diagnostic name corresponding to the selected area. For example, if the user selects a diagnosis name of " vertical bone loss " after determining that the selected region after the first input is " vertical bone loss ", the data generation system for reading the dental image, Can be matched with the diagnosis name " vertical bone loss ".

Thereafter, the data generation system for dental image reading may display the dental image, the area corresponding to the anatomical structure, and the diagnosis name (240). The manner in which the data generating system for reading dental images is displayed will be described in detail with reference to FIGS. 3 and 4. FIG.

FIGS. 3 and 4 are views schematically showing the operation state of a data generation system for reading a dental image according to an exemplary embodiment. In FIGS. 3 and 4, a case will be exemplified in which a part of the anatomical structure of the oral cavity is selected and a diagnosis is given.

Referring to FIGS. 3 and 4, the data generation system 300 for reading a dental image may display a menu for working with at least one dental image and at least one dental image. For example, the data generation system 300 for reading a dental image may display at least one dental image in a first region 310, and an anatomical identification number of an oral cavity according to a specific rule may be displayed in the second region 320 A plurality of icons to be input are displayed, a plurality of icons for inputting a diagnosis name according to a specific rule are displayed in the third area 330, and an anatomical Display icons in the fifth area 350 and at least one dental image in the sixth area 360 are displayed in the fifth area 350, A plurality of edit icons (for example, color assigning, deleting, enlarging, reverting, etc.) for editing can be displayed.

When the data generation system 300 for reading a dental image loads a dental image in the first area 310, the user can read the dental image shown in the anatomical structure of the oral cavity and determine the anatomy of the oral cavity corresponding to the lesioned area At least some of the plurality of icons displayed in the second area 320 may be selected to match the enemy identification number. For example, when the anatomical structure of the oral cavity is virtually divided into quadrants, the upper left third region (tooth number 13) can be selected based on the center.

Then, in order to match the figure information including the point, line or face to the upper left third region (tooth no. 13), an icon for performing a drawing function among a plurality of edit icons displayed in the sixth region 360 You can choose. For example, the user can draw the area with the face of a specific color in the upper left third area (tooth number 13).

In addition, the user can select any one of the plurality of icons displayed in the third area 330 to match the diagnosis name to the third upper left area (tooth # 13). For example, if the user determines that the diagnosis name for the upper left third region (tooth number 13) is " occlusal wear, " You can select the icon labeled "Occlusal wear".

The user selects the anatomical identification number of the oral cavity displayed in the second area 320 and the diagnosis name displayed in the third area 330 and if necessary uses the edit icon displayed in the sixth area 360 to display the dental image The data generation system 300 for reading a dental image displays an anatomical identification number (grooming), a diagnosis name (Present illness), a diagnostic code (not shown), etc. of the oral cavity selected by the user in the fourth area 340 Can be displayed. The user examines the anatomical identification number and the diagnosis name of the oral cavity displayed in the fourth area to confirm whether or not the report has been correctly written, and if necessary, uses an additional icon, a correction icon, or a deletion icon to check the anatomy of the oral cavity Identification number, diagnosis name, and the like can be added, modified or deleted.

Thereafter, the user can temporarily save the edited dental image using the job storage icon displayed in the fifth area 350 to temporarily store the job for the dental image. Alternatively, the user may store the edited dental image using the job completion icon displayed in the fifth area 350 to finally store the operation on the dental image. In this case, the data generation system 300 for reading a dental image may store the anatomical identification number of the oral cavity, the diagnosis name, the figure information, and the like in the header of the dental image together with the edited dental image.

As described above, as another operation for one dental image, the user selects an anatomical identification number in the first lower right region (# 41 tooth), gives a diagnosis name, and matches the figure information as needed . The description thereof will be obvious to those skilled in the art upon reference to the foregoing description, and a detailed description thereof will be omitted.

Referring to FIG. 5, the data generation system for dental image reading can firstly count the number of selection of the diagnosis name based on the second input (510). Here, as described above, the second input may include selection of a diagnosis name corresponding to the selected area. When the data generation system for reading the dental image receives the second input from the user, the data generation system for reading the dental image may count and store the number of selection of the diagnosis name selected by the user.

Thereafter, the data generation system for reading the dental image can set the priority of the diagnosis names based on the number of selection of the diagnosis name (520). For example, if the user selects a first number of " implants " as the diagnosis name and selects a second number of " implants " greater than the first number, the data generation system for reading the dental images will determine the priority of " implants " Can be set to be higher than the priority of " implant ".

Thereafter, the data generation system for reading the dental image may sequentially display the diagnosis names according to the priority (530). In the above example, the " implants " selected a second number of times may be preferentially displayed to the user than the " implants " selected a first number of times. Referring back to FIG. 3, it can be seen that among the plurality of icons for the diagnosis name displayed in the third area 330, the icon indicated by the "ambush" is arranged above and to the left of the icon marked "Implant".

6, a data generation system 600 for reading a dental image can display the status of at least one or more dental images unprocessed by the user, the status of the dental image being operated on, and the status of the generated result data have. For example, the data generation system 600 for reading a dental image displays the status of at least one or more dental images unprocessed in the first area 610 and the status of the dental image in operation in the second area 620 Displays the status of the result data generated in the third area 630, displays an icon (HOME) for returning to the main screen and an icon (DASHBOARD) for providing feedback in the fourth area 640 . The result of the selection of the icon DASHBOARD for providing the feedback displayed in the fourth area 640 will be described in detail with reference to FIGS. 7 and 8. FIG.

Referring to FIG. 7, the data generation system 700 for reading a dental image graphically displays the number of times of generating result data for a dental image according to month, weekly, and daily in the first region And the number of times the result data for the user-specific dental image is generated in the second area can be graphically displayed. Accordingly, it is possible to confirm the number of times the individual user's result data is generated, thereby providing feedback to individual users.

Referring to FIG. 8, the data generation system 800 for reading a dental image two-dimensionally graphs and displays the number of times the result data for the dental image is generated in the first area, The number of times of generating the overall result data according to the first period (one year), the number of times of generating the overall result data according to the second period (one week), and the total result according to the third period The number of times data is generated can be displayed. The number of times of generating the resultant data for the dental image displayed in the first region can be two-dimensionally grained so that the degree of the shade changes according to the number of times and the number of times.

Referring to FIG. 9, the data generation system for reading a dental image may change the image display value of the dental image based on the fourth input (810). Here, the fourth input may mean an input for changing the image display value of the dental image. Thereafter, the data generation system for reading the dental image may display (display) the dental image having the changed image display value (820). Accordingly, unlike a general medical image composed of 12 bits, a dental image composed of 8 bits can be used as data for machine learning according to a predetermined standard by changing the image display value.

Referring to FIG. 10, a data generation system 1000 for reading a dental image can change an image display value of a dental image based on a fourth input of a user who changes an image display value of the dental image. For example, if the user moves the mouse up or down while holding the scrolling of the mouse, the data generation system 1000 for reading a dental image can be used to determine the window center value or slope of the dental image, The user can display the dental image having the changed image display value by changing the value.

Referring to FIG. 11, a data generation system for reading a dental image may generate a secret code input window (1110).

Thereafter, the data generating system for reading the dental image may determine 1120 that the entered code matches the set code based on the fifth input. Here, the fifth input may include an input of a secret code for comparison with the set code. When a user inputs a secret code of a user, which is a fifth input, into a secret code input window created by a data generation system for reading a dental image, a data generation system for reading a dental image reads the secret code of the input user, Can be compared to determine whether they match.

Thereafter, the data generation system for reading the dental image may determine whether to connect to the main screen (1130). When the secret code of the fifth input coincides with the setting code, the data generating system for reading the dental image connects to the main screen (for example, the main screen of Fig. 6), and the secret code and the setting code of the fifth input If they do not match, the data generation system for reading the dental image may block the connection to the main screen. Accordingly, only a user who is authorized to use the data generation system for reading dental images can use a data generation system for reading dental images, thereby preventing the acquisition of various kinds of information of the object from the outside.

Referring to FIG. 12, a data generation system 1200 for dental image reading may generate a secret code registration window to store a secret code in a data generation system 1200 for reading a dental image. The data generation system 1200 for reading dental images can receive a user's name, ID, password, e-mail, and the user's own secret code from the user have.

Referring to FIG. 13, a dental image read data generation system 1300 according to an exemplary embodiment may automatically read a dental image of an anatomical structure of a subject's mouth and provide read data to a user. Here, the dental image read data generation system 1300 may be implemented not only as a fixed terminal but also as a mobile terminal. Examples of the mobile terminal include a smart phone, a laptop computer, a PDA, a tablet PC, and the like.

The form of the read data herein includes form information, lesion finding information, and the like to be described later on the dental image, medical information such as an identification number (i.e., a gavel), diagnosis name, diagnostic code, etc. for the anatomical structure of the oral cavity But not limited to, a variety of forms that can provide meaningful read data to a user or dental patient.

The dental image read data generation system 1300 may include a storage unit 1310, a processing unit 1320, and a display unit 1330.

An input dental image for the anatomical structure of the oral cavity and a reference dental image for analyzing the input dental image may be stored in the storage unit 1310, respectively.

The input dental image herein may be a dental image that is processed by the processing unit 1320 and is not read by the processing unit 1320 and the reference dental image may be a user other than the processing unit 1320 , A dentist) may be a dental image in which diagnostic information about a lesioned area is expressed. For example, in a reference dental image, a region having lesions may be displayed as graphic information such as a point, a line, or a face, or information about an area having a lesion in a header of the reference dental image may be described.

The processing unit 1320 may load the input dental image and the reference dental image from the storage unit 1310 and pre-process the input dental image. For example, the preprocessing process in this embodiment is a process of matching the image size of the input dental image with the image size of the reference dental image, etc., in order for the processing unit 1320 to post-process the input dental image based on the reference dental image Lt; / RTI >

The processing unit 1320 can detect an area having a lesion from the input dental image based on the reference dental image and recognize the position of the area having the detected lesion. For example, lesions may indicate dental disease such as implants, tooth loss, periodontal disease, vertical bone loss, horizontal bone loss, fracture, malocclusion, and wear. The processing unit 1320 learns how to read the input dental image based on the diagnostic information about the lesioned area represented in the reference dental image stored in the storage unit 1310, It is possible to detect an area having a lesion from the lesion, and to recognize the position of the lesioned area. The learning method of the detailed processing unit 1320 will be described with reference to FIG.

The processing unit 1320 may generate diagnostic information for the detected lesion area or an output dental image for the lesioned area. In order to generate the diagnostic information or the output dental image for the lesion region in which the processing unit 1320 is detected, the storage unit 1310 may include an identification number (diagnosis), a diagnosis name, a diagnostic code, and the like for the anatomical structure of the oral cavity And shape information including a point, a line, or a plane for displaying the diagnostic reference data or the lesioned area on the output dental image may be stored. The processor 1320 may generate diagnostic information or an output dental image for the lesion region detected from the input dental image using the diagnostic reference data or the graphic information described above. Here, the diagnostic reference data means information that can be used by a user of the read data generation system 1300 of the dental image to diagnose anatomical structures of the oral cavity.

The display unit 1330 can display the diagnostic information or the output dental image generated by the processing unit 1320. The display unit 1330 can output the read data generated by the read data generation system 1300 of the dental image, such as a user interface (UI), user information, or object information. Various forms displayed by the display unit 1330 will be described with reference to FIG. 17 or FIG. 19 to be described later.

Although not shown in FIG. 13, the dental image read data generation system 1300 may further include an image acquisition unit (not shown) and a communication unit.

The image acquiring unit may be an image photographing apparatus for photographing the anatomical structure of the oral cavity of the object. The image acquiring unit may transmit the dental image of the anatomical structure of the oral cavity to the storage unit 1310.

The communication unit may be a server in which data is stored. For example, the server may be a Picture Archiving and Communication System (PACS). If the communication unit is a server, the communication unit may acquire and store a dental image of the anatomical structure of the oral cavity in a digital state through the image acquisition unit, and may transmit the dental image to the storage unit 1310 or the processing unit 1320.

Referring to FIG. 14, first, a read data generating system of a dental image may load an input dental image (1410). For example, an input dental image may be a dental image that displays the anatomical structure of the oral cavity and may be a dental image that has not been processed by the read data generation system of the dental image. In this process, the read data generation system of the dental image can preprocess the input dental image according to the set standard.

Thereafter, the read data generation system of the dental image may read the input dental image (1420). The read data generation system of the dental image can read the loaded dental image loaded on the basis of the set diagnostic reference data. Here, the diagnostic reference data may be data set by the user in the state before the reading is performed, i.e., in the initialized state, and may include information necessary for diagnosing the object. For example, the diagnostic criteria data may include content such as diagnosis name, diagnostic code, lesion, and the like. The diagnostic reference data can be variably reset according to the repetitive processing on the input dental image of the read data generation system of the dental image.

Thereafter, the read data generation system of the dental image may generate the read output dental image (1430). The read data generation system of the dental image can generate the readout dental image in various forms based on the diagnostic reference data. For example, the output dental image may be generated to match geometric information such as points, lines, or faces where the lesion is present, or to provide medical findings about lesions at or near the lesion area May be generated to match. In addition, the read data generation system of the dental image can determine whether an additional input dental image is required in addition to the readout dental image based on the diagnostic reference data.

Thereafter, the read data generation system of the dental image may compare the generated output dental image with the reference dental image (1440). The read data generation system of the dental image can compare the output dental image with the reference dental image to determine how similar the output dental image is to the reference dental image. For example, the read data generation system of the dental image compares the image display value of the reference dental image with the image display value of the output dental image or includes the diagnosis information included in the header of the reference dental image and the header of the output dental image The diagnostic information can be compared with each other to determine whether the output dental image is close to the reference dental image.

Thereafter, the read data generation system of the dental image may classify the generated output dental image (1450). The read data generation system of the dental image can be distinguished from the reference dental image or the output dental image generated by the read data generation system of the dental image through the process of generating the output dental image. In other words, the output dental image generated by the reading data generation system of the dental image can be classified into the dental image read by the user or the dental image read by the read data generation system of the dental image according to the degree of similarity . If the output dental image read and generated by the read data generation system of the dental image is categorized into an output dental image generated by self-reading, the read data generation system of the dental image generates the readout dental image once again 1430). Accordingly, the output dental image read once again is generated, and the output dental image is generated so as to be the same as or similar to the quality of the dental image read by the user through the comparison step 1440 and the classification step 1450, The quality of the output dental image can be improved.

Referring to FIG. 15, a system 1500 for generating a dental image read data according to an embodiment learns a neural network based on an input dental image and a reference dental image, and generates diagnostic information or lesion Can output an output dental image for the region where the image is located.

Here, the neural network may include a production neural network 1510 and a classification neural network 1520.

The generated neural network 1510 may generate an output dental image read from the input dental image. The generated neural network 1510 may include a lesion reading model in which diagnostic reference data for reading an input dental image is stored as an initialized state. The lesion readout model can generate an output dental image by reading the input dental image based on the diagnostic reference data.

The classified neural network 1520 compares the generated output dental image with the reference dental image and classifies whether the generated output dental image is an output dental image generated through the generated neural network or a reference dental image not through the generated neural network have. The classified neural network 1520 may include a classification model in which classification criterion data for classifying the generated output dental image is stored. The classification model can determine whether the output dental image read according to the classification reference data is a dental image generated by the lesion reading model or a reference dental image.

In this process, the generated neural network 1510 and the classified neural network 1520 can achieve an adversarial relationship. The lesion reading model of the production neural network 1510 can generate an output dental image in a direction that maximizes the classification error so that the classification model can not distinguish the readout output dental image generated by the lesion reading model from the reference dental image. On the other hand, the classification model of the classified neural network 1520 can classify the output dental image in a direction that minimizes the classification error so that the output dental image generated by the lesion reading model is distinguished from the reference dental image. In this way, the output neural network 1510 and the classification neural network 1520 generate or classify the output dental images in the direction of maximizing the classification error or minimizing the classification error in the mutually opposing relationship, The quality of the final output dental image generated by the read data generation system 1500 can be improved similar to the quality of the output dental image read by the user.

Due to the mutual opposing relationship between the generated neural network 1510 and the classified neural network 1520, the output dental images can be repeatedly generated and classified. When the output dental image generated through the generated neural network 1510 is discriminated as the reference dental image by the classification neural network 1520, the lesion reading model of the generated neural network 1510 is diagnosed based on the diagnostic information included in the reference dental image Reset the reference data, and read the input dental image based on the reset diagnostic reference data to generate an output dental image. As described above, the output dental image repeatedly generated and classified can be displayed to the user in a state in which the quality is improved to such an extent that the dental image can not be distinguished from the reference dental image by the classification model.

Referring to FIG. 16, first, a read data generation system of a dental image may detect a lesion area based on a reference dental image (1610). As described above with reference to FIG. 15, the system for generating a read data of a dental image can detect a lesioned area based on a reference dental image through a series of learning processes.

Thereafter, the read data generation system of the dental image may match the figure information to the lesion area (1620). Here, the graphic information may include points, lines, or faces, but the present invention is not limited thereto, and may include a polygonal shape. In addition, the size of the point may vary depending on the type of the lesion, and the type of the line may be various forms such as a dashed line and a solid line, and the color of the surface may be changed.

For example, among the anatomical structures of the oral cavity of a dental image, nerves can be matched in the form of dashed lines. For example, if there are two regions in a dental image that have different lesions, the read data generation system of the dental image recognizes the two regions as regions with different lesions, Can be matched. Here, the shape of the surface can be determined according to the shape of the area.

Thereafter, the read data generation system of the dental image may generate diagnostic information for the lesioned area (1630). Here, the diagnostic information may be medically meaningful information to the user, such as a diagnosis name, an identification number (i.e., a grooming) for the anatomical structure of the oral cavity, a diagnostic code, and the like.

Thereafter, the read data generation system of the dental image may generate lesion finding information for the lesioned area (1640). Herein, the lesion finding information may include a medical judgment according to the diagnosis name of the lesion area. For example, the lesion finding information can be matched not only with character information but also with figure information. In order to express the height between the nerve and bone in the anatomical structure of the oral cavity of the output dental image, the height is indicated by a dashed line. The lesion finding information can be generated so as to be displayed by a character. In other words, the lesion finding information can be generated by matching with character information or graphic information.

Thereafter, the read data generating system of the dental image may display (display) the figure information, diagnostic information, or lesion information together with the output dental image (1650). Details will be described with reference to FIG.

Referring to Fig. 17, the read data generated by the read data generation system of the dental image according to an embodiment is shown. Here, the read data is shown as a dental x-ray image as an example. The read data may be a dental image together with the shape information and the lesion finding information matched to the lesion area by the read data generation system of the dental image.

For example, if different lesions are present in the anatomical region of the oral cavity of the subject, the readout data generated from the dental image of the subject's mouth may have different color saturation to indicate that different lesions are present The face can be displayed in the corresponding area. For example, the nerve may be displayed in the form of a dotted line in the read data. For example, in the read data, information such as lesion finding information may be displayed as a letter, or information such as " bone height considering the distance from the nerve " indicating the meaning of the lesion finding information may be displayed as a dashed line. For example, in the read data, the diagnosis name may be displayed in Korean or English in the vicinity of the lesion area.

Referring to FIG. 18, a read data generation system of a dental image may first detect a lesion area based on a reference dental image (1810). As described above with reference to FIG. 15, the system for generating a read data of a dental image can detect a lesioned area based on a reference dental image through a series of learning processes.

Thereafter, the read data generation system of the dental image may generate diagnostic information for the lesioned area (1820). As described above, the diagnostic information may be medically meaningful information to the user, such as a diagnosis name, an identification number (i.e., a grooming) for the anatomy of the oral cavity, a diagnostic code, and the like. In this case, the diagnostic information may be generated in the form of a report so that it is displayed to the user in a medically meaningful form. The specific form of the report will be described in detail with reference to FIG.

For example, the diagnosis name and the diagnosis code may be, but not limited to, the diagnosis name / diagnosis code according to the Korean standard disease / sign classification table, the diagnosis name / diagnosis code according to the international diagnosis standard, and the like. For example, the way in which the anatomical structure of the oral cavity is represented is based on the identification numbering scheme according to the ISO standard (ISO 3950) produced by the World Health Organization (WHO), the American System (or Universal Numbering System (universal numbering system), but it is not limited to any one scheme. In the case of the identification numbering method according to the ISO standard, a tooth number is assigned to 1 to 8 divided into four quadrants based on the center line of the mouth. In the case of the identification number assigning method according to the American system, It rotates clockwise to the bottom, and the teeth number (including English) is given.

Thereafter, the read data generation system of the dental image may display the diagnostic information as an identification number for the anatomical structure of the oral cavity, a diagnostic name for the lesioned area, or a diagnostic code for the lesioned area (1830) . Details will be described with reference to FIG.

19 is a view showing read data generated by a read data generation system 1900 of a dental image according to an embodiment.

Referring to Fig. 19, the read data generated by the read data generation system 1900 of the dental image according to an embodiment is shown. Here, the read data is shown as a readout of a table shape as an example. The readout data may be represented by an identification number (grooming), diagnosis name (Present illness), etc. for the anatomical structure of the oral cavity.

For example, the identification number (grooming) for the anatomical structure of the mouth may be indicated by an identification numbering scheme according to the ISO standard. If the position of the lesion is recognized by the dental image read data generation system 1900, the identification number may be matched to the recognized lesion area. The matched identification number can be displayed in the form of a quadrant showing the location of the lesion area. It should be noted, however, that the present invention is not intended to be limited to the ISO standard. For example, a diagnosis may be displayed together with an identification number (gag) for the anatomical structure of the mouth.

In addition, although not shown in FIG. 19, a diagnosis code may be displayed together with the diagnosis name. The diagnostic code is displayed to assist the user's judgment. If the diagnosis name is an implant, a normal finding, etc., there is no diagnostic code, so the diagnostic code may not be displayed.

An erroneously determined diagnosis name or identification number may be displayed in the read data generated by the read data generation system 1900 of the dental image. Therefore, in order to correct this, an icon for erasing or modifying the read data may be displayed have. When the user inputs an icon, the read data can be modified or deleted.

The method according to an embodiment 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, and the like, alone or in combination. The program instructions to be recorded on the medium may be those specially designed and configured for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI > or equivalents, even if it is replaced or replaced.

Claims

Diagnostic information including an anatomical identification number of the anatomical structure and a plurality of diagnosis names, and a resultant data for the dental image are stored;

An input for receiving a second input comprising a first input comprising selection of an area corresponding to the anatomical structure of the dental image and a selection of a diagnosis name corresponding to the selected area;

A processing unit for generating result data for the dental image based on the first input and the second input; And

A display unit for displaying the dental image and the result data;

Wherein the data acquisition system comprises:
The method according to claim 1,

Wherein the processing unit recognizes a region selected from the anatomical organization of the dental image when a first input is received from the user to the input unit,

Wherein the display unit displays a marker in the selected area together with the dental image.
3. The method of claim 2,

Wherein the processing unit matches the anatomical identification number to the recognized region,

Wherein the display unit further displays the anatomical identification number.
3. The method of claim 2,

Wherein the processing unit matches a diagnosis name to a recognized area when a second input is received from the user to the input unit,

Wherein the display unit further displays a matched diagnosis name.
5. The method of claim 4,

Wherein the processing unit counts the number of times the diagnosis name is selected from the second input received from the user to the input unit and stores the counted number of times for the selected diagnosis name in the storage unit.
6. The method of claim 5,

Wherein the processing unit sets the priority of the plurality of diagnosis names according to the counted number of times for the selected diagnosis name,

Wherein the display unit further displays a plurality of diagnosis names listed in accordance with the set priorities.
The method according to claim 1,

Wherein the first input comprises a plurality of first sub icons each corresponding to the anatomical identification number,

Wherein the second input comprises a plurality of second sub-icons each corresponding to the plurality of diagnostics names.
The method according to claim 1,

The processing unit counts the number of times of generating the result data, reads out the dental image that generates a pattern of the number of times the result data is generated over time based on the number of times of generating the counted result data and the time of generating the result data A data generation system for a computer.
A storage unit for storing an input dental image for the anatomical structure of the oral cavity and a reference dental image for analyzing the input dental image;

Detecting an area having a lesion from the input dental image based on the reference dental image, recognizing the position of the area having the detected lesion, and diagnosing the area having the detected lesion, A processor for generating an output dental image for the subject; And

A display unit for displaying the diagnostic information or the output dental image;

And a readout data generating system for generating a readout data of the dental image.
10. The method of claim 9,

Wherein the processing unit learns the neural network based on the input dental image and the reference dental image to generate the diagnostic information or the output dental image.
11. The method of claim 10,

The neural network,

A generating neural network for generating an output dental image read from the input dental image; And

And compares the generated output dental image with the reference dental image stored in the storage unit and determines whether the generated output dental image is an output dental image generated through the generated neural network or a reference dental image stored in the storage unit Classified Neural Network;

The system comprising:
12. The method of claim 11,

Wherein the generated neural network regenerates an output dental image read from the input dental image when the classified neural network discriminates the generated output dental image as a reference dental image stored in the storage unit.
10. The method of claim 9,

Wherein the processing unit generates lesion finding information for the lesion area based on the reference dental image,

Wherein the display unit displays the lesion finding information together with the output dental image.
10. The method of claim 9,

Wherein the processing unit matches graphic information including points, lines, or faces to the lesion area,

Wherein the display unit displays the lesioned area as matched figure information.
10. The method of claim 9,

Wherein the display unit displays the diagnosis information as an identification number for the anatomy of the oral cavity, a diagnosis name for the lesion area, or a diagnosis code for the lesion area.
10. The method of claim 9,

Wherein the display unit further displays a plurality of icons corresponding to anatomical structures of the oral cavity,

Wherein the display unit displays at least one of the plurality of icons as an identification number for the region having the lesion.