Inner-mouth state image diagnosis method

Abstract

To provide an inner-mouth state image diagnosis method that is high in identification of a progress state of periodontitis by using artificial intelligence.SOLUTION: An inner-mouth state image diagnosis method is composed of the steps of: preparing a tooth row-photographed image having a tooth row photographed from an oral cavity front face part; carrying out image processing editing the tooth row-photographed image to prepare a post-edition tooth row-photographed image; determining whether what is reflected in the post-edition tooth row-photographed image is a tooth or not; diagnosing a risk level serving as a degree of a periodontitis progress of a user from the post-edition tooth row-photographed image, using an image artificial intelligence tool causing a periodontitis tooth row-photographed image group having tooth rows in various progress states of periodontitis photographed to learn as big data; and preparing an advice text suitable for teeth with respect to respective users on the basis of a diagnosis result of the risk level and the post-edition tooth row-photographed image, using a generative artificial intelligence tool.SELECTED DRAWING: Figure 11

Description

[Technical Field]

[0001] The present invention relates to a method for imaging and diagnosing intra-oral conditions using artificial intelligence tools. [Background technology]

[0002] In recent years, the general public has become increasingly concerned about the health of their oral cavity (teeth and gums). Specifically, tooth color, for example, is an important factor in a person's appearance. This has led to excessive tooth brushing and excessive whitening, especially among young people.

[0003] Periodontal disease is also a widespread and common problem among modern people. Periodontal disease is caused by bacterial infection, which dissolves the gums and the bones that support the teeth, causing teeth to become loose and eventually fall out. Periodontal disease of the gums is a greater cause of tooth loss than tooth decay, and can also cause bad breath, so care is needed.

[0004] A common method for testing for periodontal disease is the probing method, which measures the depth of periodontal pockets. In this method, a pocket probe, a rod-shaped metal test tool bent at an approximately 90° angle near its tip, is gradually inserted along the root of the user's tooth toward the lower part of the gingiva adjacent to the tooth. The depth of the periodontal pocket is calculated based on the distance it takes for the tip of the pocket probe to reach the junctional epithelium of the tooth, and the progression of the user's periodontal disease is then assessed.

[0005] In recent years, for diseases whose progression can be determined by appearance, there has been technology that takes a photo of the affected area using an electronic device such as a smartphone or tablet, sends the image to a specialist, who then diagnoses and determines the condition of the disease online, as well as technology related to an oral condition assessment device that identifies teeth based on images of the affected area of periodontal disease and diagnoses the progression of periodontal disease (Patent Technical Document 1).

[0006] Furthermore, there are technologies for inspecting plaque, which induces periodontal disease, such as using a plaque-colored liquid to visualize it, or using a plaque inspection light (Patent Technology Document 2). [Prior art documents] [Patent documents]

[0007] [Patent Document 1] Japanese Patent Publication No. 2020-54646 [Patent Document 2] Patent Application No. 2023-191084 Summary of the Invention [Problem to be solved by the invention]

[0008] To quickly resolve the aforementioned dental problems, it is desirable to receive a diagnosis from a specialist on a regular basis by visiting a dentist, etc. However, regular visits to a dentist require a certain amount of time and money, and such a system is not easily accessible to users.

[0009] In addition, although dedicated devices for detecting oral health conditions (such as devices for testing bad breath) have been developed, they are not easily accessible to users due to the special methods of use and the high cost of the devices.

[0010] Furthermore, the conventional periodontal disease probing method described above requires the direct insertion of a pocket probe into the gums of the patient, which is painful for the patient and requires the patient to visit a clinic in person and for the dentist to perform the examination.

[0011] Furthermore, with regard to the diagnosis of periodontal disease based on photographed images described in Patent Document 1, there are cases where an image of the mouth being closed is recognized as a tooth, and areas that are not teeth are recognized as teeth, so the accuracy of diagnosis is insufficient. Therefore, there is room for improvement in this image diagnosis method as well.

[0012] In view of the above problems, the present invention aims to provide an oral examination system that takes images of the user's oral cavity (teeth and gums) without using a dedicated examination device using electronic devices such as smartphones with camera functions that are widely used in society, analyzes the captured images using artificial intelligence, and then diagnoses the health of the user's teeth and gums, allowing the user to easily understand the diagnosis results. [Means for solving the problem]

[0013] The present invention provides an image diagnosis method for intraoral conditions as described in claim 1, which comprises the steps of: capturing an image of a row of teeth including one or more teeth from the front of the oral cavity to create a row of teeth photographed image; image processing and editing the row of teeth photographed image to create an edited row of teeth photographed image; diagnosing a risk level, which represents the degree of progression of periodontal disease of the user, based on the edited row of teeth photographed image; and using an AI generation tool to create optimal dental advice text for each user based on the diagnosis results of the risk level diagnosing step and the edited row of teeth photographed image, wherein the risk level diagnosing step is configured to take images of rows of teeth affected with periodontal disease at various stages of progression, or rows of teeth in a healthy state without periodontal disease, from the front of the oral cavity, and diagnose the risk level of the degree of progression of periodontal disease using an image artificial intelligence tool that has been trained in advance on software that has learned as big data a group of periodontal disease row images obtained by capturing images of rows of teeth affected with periodontal disease at various stages of progression.

[0014] The present invention is also an image diagnostic method according to claim 1, which includes a step of using an average number of pixels in the maxillary dentition length corresponding to the vertical length of the maxillary dentition and an average number of pixels in the mandibular dentition length corresponding to the vertical length of the mandibular dentition in the edited dentition photographed image, and determining the area of the dentition in the edited dentition photographed image from the ratio of the average number of pixels in the maxillary dentition to the average number of pixels in the mandibular dentition length.

[0015] The present invention also provides an image diagnostic method according to claim 1, further comprising the step of analyzing the edited photographed image of the row of teeth using a function to determine a region of the row of teeth in the edited photographed image of the row of teeth.

[0016] Furthermore, the present invention provides an image diagnostic method as described in claim 1, in which the image processing and editing of the dentition photographed image is configured to create exposure-adjusted dentition photographed images by adjusting the exposure to various levels of brightness based on the same dentition photographed image, and then overlay and combine the multiple exposure-adjusted dentition photographed images for image processing.

[0017] The present invention also provides an imaging diagnostic method according to claim 1, characterized in that the photographing device used in the step of creating the dentition image is configured to capture images using a short-focus lens that enables clear photography even at a focal length of 100 mm or less and to receive light using a proximity sensor, thereby creating the dentition image. [Effects of the Invention]

[0018] By using the image diagnostic technique according to the present invention, it is possible to perform image diagnosis of periodontal disease with higher accuracy than with conventional techniques. [Brief explanation of the drawings]

[0019] [Figure 1] FIG. 1 is a flowchart showing the flow of image diagnosis according to an embodiment of the present invention. [Figure 2] 1 is a correlation diagram showing the configuration of an oral cavity examination system according to an embodiment of the present invention. [Figure 3] 1 is a correlation diagram showing the electrical configuration of a network system server of an oral cavity examination system according to an embodiment of the present invention. [Figure 4] FIG. 2 is a diagram illustrating a mechanism of information processing in a network system server according to an embodiment of the present invention. [Figure 5] FIG. 1 is a diagram showing diagnostic information learned by a generation AI according to one embodiment of the present invention. [Figure 6] FIG. 10 is a diagram showing a start screen in an image diagnosis application according to an embodiment of the present invention. [Figure 7] FIG. 10 is a diagram showing an imaging screen in an image diagnosis application according to an embodiment of the present invention. [Figure 8]FIG. 10 is a diagram showing a screen during imaging in an image diagnosis application according to an embodiment of the present invention. [Figure 9] FIG. 10 is a diagram showing a capture image determination screen in an image diagnosis application according to one embodiment of the present invention. [Figure 10] FIG. 10 is a diagram showing a diagnosis result in an image diagnosis application according to an embodiment of the present invention. [Figure 11] 10A and 10B are diagrams showing diagnostic results and advice sentences in an image diagnostic application according to one embodiment of the present invention. [Figure 12] 12A and 12B are diagrams showing a photography support member attached to a smartphone or the like for use in image capture according to one embodiment of the present invention, in which Fig. 12A is a front view of the photography support member, and Fig. 12B is a right side view of the photography support member attached to a smartphone or the like. [Figure 13] FIG. 1 is a diagram showing one aspect of image capture during image diagnosis according to one embodiment of the present invention. [Figure 14] 14A and 14B are diagrams showing a camera-like imaging device used in image diagnosis according to one embodiment of the present invention, in which Fig. 14A is a front view of the imaging device and Fig. 14B is a right side view of the imaging device. DETAILED DESCRIPTION OF THE INVENTION

[0020] The present invention provides an image diagnosis method for intraoral conditions as described in claim 1, which comprises the steps of: capturing an image of a row of teeth including one or more teeth from the front of the oral cavity to create a row of teeth photographed image; image processing and editing the row of teeth photographed image to create an edited row of teeth photographed image; diagnosing a risk level, which represents the degree of progression of periodontal disease of the user, based on the edited row of teeth photographed image; and using an AI generation tool to create optimal dental advice text for each user based on the diagnosis results of the risk level diagnosing step and the edited row of teeth photographed image, wherein the risk level diagnosing step is configured to take images of rows of teeth affected with periodontal disease at various stages of progression, or rows of teeth in a healthy state without periodontal disease, from the front of the oral cavity, and diagnose the risk level of the degree of progression of periodontal disease using an image artificial intelligence tool that has been trained in advance on software that has learned as big data a group of periodontal disease row images obtained by capturing images of rows of teeth affected with periodontal disease at various stages of progression.

[0021] The image diagnosis according to this embodiment is basically performed via an Android application, but may also be performed in the form of other applications such as a web application or an iOS application.

[0022] [1. Flow of image diagnosis according to this embodiment] First, an overview of the flow of image diagnosis according to this embodiment will be described with reference to Figure 1. As shown in Figure 1, the flow of image diagnosis according to one embodiment of the present invention begins with imaging the dentition of a person to be diagnosed (step S1). To image the dentition, the mouth is opened laterally and vertically, for example, like when pronouncing the letter "i," while the teeth are kept as still as possible with the upper and lower teeth overlapping. In this state, a camera or imaging device of an electronic device such as a smartphone or tablet is placed slightly away from the front of the mouth, and the shutter is pressed to take a picture.

[0023] When photographing the dentition, a dedicated mouth opener may be used. That is, the dedicated mouth opener is attached to both the left and right ends of the mouth of the user 7, and after the mouth is fixed in the open state as described above, the dentition is photographed.

[0024] One aspect of the display screen of this application during such imaging will be described. On the right side of the display screen, a captured image display screen is provided, which displays the captured image captured by the camera in real time in video format. A judgment end button and a capture button are provided at the bottom of the captured image display screen. Clicking the judgment end button or the capture button ends the assessment or captures still dentition image information 38 to be used for the assessment, respectively. A transition screen that changes every second for video capture is provided at the top left of the display screen. The user 7 performing the imaging can carefully review the transition screen to understand the imaging status. The still dentition image information 38 to be used for diagnosis is displayed at the bottom left of the display screen. Below the dentition image is a confirm button. Clicking the confirm button uses the captured dentition image information 38 and proceeds to the next step S2. Details of the display screen will be described later.

[0025] The imaging device used for image capture may be a smartphone or other electronic device, or a camera-like imaging device equipped with a short focal length lens. Using a camera equipped with such a short focal length lens enables better image quality with less distortion and aberration, resulting in more accurate image diagnosis.

[0026] Next, the captured dentition image information 38 is subjected to image processing (step S2). In this image processing, the exposure of the dentition image information 38 is corrected, and three or more dentition image information 38 with different exposures are created from one dentition image information 38. Then, these three or more exposure-adjusted dentition image information 38 are combined to create one edited dentition image information 39. By subjecting the captured image to this image processing, it is possible to significantly eliminate the influence of ambient brightness, enabling more accurate image diagnosis.

[0027] Next, it is determined whether or not what is captured in the image is a tooth (step S3). This process is an extension of image processing, and determines whether or not it is a tooth based on the degree of shading of the edges around the teeth.

[0028] As an additional condition for determining whether a tooth exists, the average number of pixels in the vertical length of the maxillary teeth in the captured image may be a, and the average number of pixels in the vertical length of the mandibular teeth may be b, with the ratio a / b being between 0.5 and 1.5. Specifically, the specific procedure is as follows: First, in the edited dentition image information 39, for maxillary teeth, a line is drawn connecting the cervical regions of adjacent teeth in the left and right direction at the estimated cervical region, which is the boundary between the crown and root. A line is also drawn connecting the tips of the crowns of adjacent teeth in the left and right direction. The same cervical and crown tangents are then drawn for the mandibular teeth. Then, the number of pixels between the maxillary cervical tangent and the crown tangent is defined as a, and the number of pixels between the mandibular cervical tangent and the crown tangent is defined as b. If the value of a / b is between 0.5 and 1.5, the tooth is determined to be non-tooth; otherwise, the tooth is determined to be non-tooth. If it is determined that the image captured is not a tooth, the dentition is imaged again using the electronic device (step S1).

[0029] In healthy dentition, the average number of pixels along the vertical length of the upper teeth is roughly the same as the average number of pixels along the vertical length of the lower teeth, or the average number of pixels along the vertical length of the upper teeth is slightly greater than the average number of pixels along the vertical length of the lower teeth. Therefore, if the condition for determining a tooth is that "a / b is in the range of 0.5 to 1.5," it is possible to identify the area of the dentition within the photographed image of the dentition.

[0030] Furthermore, as a means for determining whether or not what is captured in an image is a tooth, an object detection model using a function may be used to determine the region of the row of teeth in the row of teeth photographed image information 38, i.e., whether or not what is captured in the row of teeth photographed image information 38 is a tooth. Specifically, for example, as one mode of determining whether or not what is a tooth, an algorithm such as YOLO (You Only Look Once) is used to recognize the region of the row of teeth among what is captured in the row of teeth photographed image information 38.

[0031] Next, using the edited dentition image information 39, symptom classification is performed using image AI (artificial intelligence) (step S4). This step is executed by clicking the "OK" button on the operation screen. Specifically, if the tooth image captured in the edited dentition image information 39 determines that the periodontal pocket depth is less than 4 mm, the periodontal disease progression level is classified as Low (healthy); if the periodontal pocket depth is determined to be 4 mm or more but less than 6 mm, the periodontal disease progression level is classified as Medium (caution required); and if the periodontal pocket depth is determined to be 6 mm or more, the periodontal disease progression level is Heavy (visit recommended). This diagnosis is performed for each of the 12 teeth, and the most common diagnosis result among the 12 teeth is used to diagnose the overall severity of the oral condition. Note that if the symptom classification using image AI is not successful, click the "OK" button again to re-execute the process.

[0032] Furthermore, the symptom classification by the image AI in step S4 may be performed based on a group of periodontal disease images, which is a collection of photographed images of teeth affected by periodontal disease in various stages of progression, or of teeth in a healthy state without periodontal disease. Specifically, the image AI is trained in advance as big data on the group of periodontal disease images of teeth in various stages of progression, or of teeth in a healthy state, and the group of periodontal disease images is compared with the edited teeth image information 39 to have the image AI diagnose the stage of periodontal disease in the edited teeth image information 39 based on the condition and color of the gums, the presence or absence of tartar, etc.

[0033] Next, writing to the database is performed (step S5), and information such as the dentition image information 38, edited dentition image information 39, and image diagnosis results is stored in a HDD or the like.

[0034] Next, based on the edited dentition image information 39, the risk level is displayed as a symptom classification result (step S4) by the image AI (step S6). When the risk level is displayed, the risk level result is displayed on the screen of the user terminal 3. A list of options for each item in the user-selected item information 36 is displayed near the result display, such as on the page following the result display or at the bottom of the same screen. Specific examples of the user-selected item information 36 are included in the user-selected item information 36 of the diagnostic information 51 learned by the generation AI in Figure 5: "Age 56," "Periodontal disease prevention 57," "Cavity prevention 58," "Hypersensitivity 59," "Yellowing teeth 60," "Bad breath treatment 61," "Toothbrush 62," and "Toothpaste 63." These options were carefully selected based on the items desired by a large number of users, as determined by a prior survey, and also because they do not violate laws such as the Dentist Act, i.e., do not constitute medical procedures that can only be performed by qualified personnel. Therefore, this example of user selection item information 36 is merely one embodiment, and other items such as dental matters and lifestyle matters may be included as long as they fall within the scope of laws such as the Dentist Act.

[0035] Then, the user selects one or more items that interest him or her from the user selection item information 36 (it is also possible to select multiple items).

[0036] Next, information is provided to the generation AI (step S7). Examples of information provided to the generation AI include the risk level result in step S6, user selection item information 36, edited dentition image information 39, etc.

[0037] Finally, the AI creates an advice sentence and displays the final diagnosis result (step S8). In step S8, the AI determines the progression of periodontal disease, the state of oral health, and provides advice on future measures to combat periodontal disease.

[0038] More specifically, the generation AI creates written advice and displays the final diagnosis result based on the diagnosis results of the image AI diagnostic means, the edited dentition image information 39, and the content selected by the user 7 from the user selection item information 36. The final diagnosis result includes calculating an evaluation score for the health of the teeth and gums, and creating written advice. The written advice is approximately 1,000 characters long, and the generation AI creates the optimal advice for each user 7 based on the user selection item information 36 entered by the user 7 and big data such as prior art related to conventional periodontal disease treatments that the generation AI has previously learned. The data used by the generation AI includes prior art data, such as papers published up to one year ago.

[0039] [2. Configuration of the Oral Inspection System of the Present Embodiment] Next, the detailed configuration of the oral examination system of this embodiment will be described with reference to Figure 2. The oral examination system of this embodiment comprises a network system server 4 managed by an oral examination system administrator, a user terminal 3 used by a user 7 when using an application executed by the network system server 4 (hereinafter referred to as the application), and an administrator terminal 1 used by the oral examination system administrator to manage the application. The network system server 4, administrator terminal 1, and user terminal 3 are interconnected via a public communication network 2. Devices such as the administrator terminal 1 and user terminal 3 are, for example, personal computers or iPhone (registered trademark), but may also be other electronic devices such as personal computers, tablet terminals, smartphones, etc.

[0040] Next, the flow of information in the oral examination system of this embodiment will be described. As described above, the oral examination system of this embodiment comprises a network system server 4 for image analysis and a user terminal 3, such as a mobile device with a camera function, connected via a public communication network 2. The oral examination system works as follows: a user 7 uses his or her user terminal 3 to capture an image of his or her own dentition and transmits the captured dentition image information 38 to the network system server 4. When the network system server 4 receives the dentition image information 38, the network system server 4 analyzes the dentition image information 38 using tools such as generative artificial intelligence and transmits a diagnosis of the user 7's dental health status as the analysis result to the user terminal 3. The user 7 can then view the diagnosis on the user terminal 3 that received the diagnosis.

[0041] The application used in this embodiment may be a web application, a native application, or a hybrid application of a web application and a native application, regardless of its form.

[0042] [3. Electrical Configuration of the Network System Server of the Oral Examination System of the Present Embodiment] Next, the electrical configuration of the network system server 4 of the oral examination system of this embodiment will be described with reference to Fig. 3. The network system server 4 of the oral examination system of this embodiment is composed of a memory unit 11, an input / output control unit 12, a control unit 13, and an external communication I / F (Interface) control unit 14.

[0043] Of these, the storage unit 11 is configured with a large-capacity storage device such as an HDD (Hard Disk Drive), flash memory, etc. This storage unit stores the image AI, the generation AI, the personal user ID of the user 7, the user's personal identification number, the diagnosis date and time of the image diagnosis, dentition image information 38, edited dentition image information 39, symptom classification by the image AI, etc.

[0044] The input / output control unit 12 also controls communication with external output devices such as a liquid crystal display unit connected to the system server, a keyboard, and a mouse.

[0045] The control unit 13 is configured with a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), a flash memory, etc. The CPU of the control unit 13 functions as various processes of this embodiment by executing various programs stored in advance in the ROM or the storage unit 11.

[0046] The external communication I / F control unit 14 is connected to the external public communication network 2 and controls communication between the terminals based on instructions from the control unit 13 .

[0047] [4. Information Processing Mechanism of the Application of the Present Embodiment] Next, the mechanism of information processing in this embodiment will be described with reference to Fig. 4. Information processing in the network system server 4 in this embodiment is executed by the network system server 4, which is composed of an application server 5 and a database server 6. The application server 5 processes information related to the execution of this application, including diagnosis by image AI and advice creation by generation AI, and the database server 6 stores information related to this application in this embodiment.

[0048] Each mechanism will be described below. As shown in Fig. 4, the application server 5 has a captured image importing means 22, an image editing means 23, a tooth or not determining means 24, an image AI-based diagnosis means 25, an advice creating means 26 using a generation AI, a data output means 27, and a result display means 28. The database server 6 also has user ID information 32, user identification number information 33, diagnosis date and time information 34, diagnosis classification information 35, user selection item information 36, captured image editing function information 37, dentition image information 38, and edited dentition image information 39.

[0049] The above-mentioned means will be described in more detail below. First, the means possessed by the application server 5 will be described. First, the photographed image importing means 22 is a means for importing photographed images of the dentition captured using the camera function of an electronic device such as a smartphone into this application. Furthermore, the image editing means 23, as described above, is a means for adjusting the exposure of the photographed photographed image of the dentition, preparing multiple photographed images with adjusted exposure from the same photographed image, and combining them. This image editing means 23 can minimize the influence of ambient brightness, etc., thereby improving the accuracy of image diagnosis. This image editing means 23 is performed within this application. Furthermore, the tooth or not determining means 24 is a means for determining whether the object captured in the edited dentition image information 39 is a tooth or not, based on the edited dentition image information 39 that has been processed by the image editing means 23.

[0050] The image AI diagnostic means 25 is a means for performing a diagnosis using image AI. Examples of the subject of such a diagnosis include the color of the gums, the shape of the gums between the teeth, or the blood flow in the gums, and based on the results of such a diagnosis, the health condition of the gums (for example, periodontal disease or pyorrhea, which is a worsening of periodontal disease) is diagnosed. In addition, the subject of the diagnosis also includes the condition of tooth stains and tartar buildup, and based on such a diagnosis, the diagnosis result is notified to the user 7 as a guideline for oral care. Furthermore, with regard to tooth color, not only whiteness but also the normal color of teeth is diagnosed.

[0051] Furthermore, if there is a problem with the health of the gums, the future state of the gums and the progress of periodontal disease are predicted and notified to the user 7. Furthermore, if there is a problem with the health of the gums, a prediction is made on bad breath and the user 7 is notified.

[0052] In addition, the advice creation means 26 by the generation AI is a means by which the generation AI creates written advice based on the image diagnosis results of the user 7, the user selection item information 36 selected by the user 7 from the user selection item information 36, and big data such as prior art.

[0053] As an example of such a diagnosis using image AI, the oral condition of the user 7 is scored out of 100 points based on a photograph of the oral condition of the user 7. If the scoring result is in the range of 0 to 60 points, it is classified as "recommended to receive," if it is 60 to 80 points, it is classified as "needs caution," and if it is 80 to 100 points, it is classified as "healthy."

[0054] The data output means 27 is a means for outputting the diagnostic data to the database server and storing it therein. The result display means 28 is a means for displaying the above diagnostic results on the screen of the user terminal 3.

[0055] Next, one embodiment of a diagnostic system using an image AI and a generation AI will be described with reference to Figure 5. As shown in Figure 5, the diagnostic information 51 used by the generation AI includes edited dentition image information 39 from the image AI, symptom classification diagnostic results 54, and user-selected item information 36, including age 56, periodontal disease prevention 57, tooth decay prevention 58, hypersensitivity 59, yellowing of teeth 60, bad breath treatment 61, toothbrush 62, and toothpaste 63. These items are diagnostic information used by the generation AI, and in addition to these items, the generation AI has also learned in advance, as big data, prior art related to periodontal disease and teeth.

[0056] Then, based on two of these items, the image AI's edited dentition image information 39 and symptom classification diagnosis result 54, and the items selected by the user 7 from the user selection item information 36 (age 56 is a required entry), the generation AI calculates a teeth and gum health score, creates an advice sentence, and presents it to the user 7. In other words, it is possible to provide optimal advice to each individual user 7 based on the progression of the user's periodontal disease and the items of concern to the user 7.

[0057] The optimal advice created by the AI is divided into, for example, a general assessment and detailed advice. The general assessment provides a commentary on the current state of the user's 7's oral condition based on the captured image, such as, for example, "There are no obvious direct signs of serious cavities or gingivitis" or "The oral environment cannot be said to be perfect." The detailed advice suggests products appropriate to the user 7 based on the oral condition, such as, for example, "We recommend a head toothbrush from company XX, which is inexpensive and can remove plaque quickly," when the user 7 selects toothbrush 62 from the user selection item information 36.

[0058] Another example of such advice is to advise the amount of toothpaste to use based on the age 56 information entered by the user 7. Since the optimal amount of toothpaste to use varies depending on age, more accurate advice can be given based on the age 56 information.

[0059] [5. Operation flow in the application of the oral examination system of this embodiment] Next, the flow of the above-mentioned diagnostic process will be described in detail using the application screen images of FIGS.

[0060] First, FIG. 6 shows a start screen 70 of an application for performing image diagnosis in this embodiment. The start screen 70 includes a start screen center section 71 in which the logo of the application provider is located in the center of the start screen 70. A personal identification number input field 72 is located below the start screen center section 71. The personal identification number input field 72 is a field for entering the personal identification number assigned to each user 7 in order to identify the user 7 who will be using the diagnosis of this application when using this application for a corporate health checkup or the like. A terms of use agreement button 73 is also located below the personal identification number input field 72. By clicking the terms of use agreement button 73, the user 7 who will be using this application is deemed to have agreed to the terms of use of this application and can begin diagnosis using this application.

[0061] Next, we will discuss Figure 7. Figure 7 is the next display screen of this application after Figure 6. This display screen is a photography start screen 80 that is displayed when photographing and acquiring dentition image information 38 from the front of the oral cavity of a user 7 using a smartphone camera or the like. On this photography start screen 80, a photography screen 81 that displays the photographed image is disposed on the right side of the photography start screen 80. Below the photography screen 81, a photography start button 82 is disposed on the left side, and a photography button 83 is disposed on the right side. A user 7 using this application can press the photography start button 82 to start taking continuous videos, and can press the photography button 83 to take still images to be used for image diagnosis.

[0062] Next, we will discuss Figure 8. Figure 8 shows a display screen 84 switched from Figure 7 by clicking the start shooting button 82 in Figure 7. On this display screen 84, a small screen 90 in which the captured image changes every second is displayed on the left side of the shooting screen 81. By switching every second in this manner, the photographer can accurately grasp the shooting situation. Furthermore, when the start shooting button 82 is clicked, the display changes from "start shooting" to "pause." If "pause" is displayed, shooting is in progress, and the captured image on the small screen 90 changes every second. If you want to temporarily stop shooting, click the start shooting button 82 that displays "pause."

[0063] A simple diagnosis result is also displayed at the bottom of the small screen 90. The message "NO DENTAL" may be displayed at the location, and this message is displayed when nothing resembling a row of teeth is recognized in the captured image. When "NO DENTAL" is displayed, the user should again use an electronic device such as a smartphone to capture an image of the row of teeth (step S3).

[0064] Next, we will discuss Figure 9. Figure 9 shows the display screen after pressing the capture button 83 on the display screen of Figure 8. In addition to the capture screen 81 and small screen 90, this screen also has a diagnostic image selection screen 101 below the small screen 90. On this diagnostic image selection screen 101, clicking the capture button 83 allows the user to select an image to be used for diagnosis from the images being captured. In addition, a decision button 102 is provided below the diagnostic image selection screen 101. By clicking the decision button 102, a user 7 using this application can decide to use the image displayed on the diagnostic image selection screen 101 for image diagnosis. When the user 7 clicks the decision button 102, the display screen switches to the next screen.

[0065] Next, we will discuss Figure 10. Figure 10 shows a diagnosis result display screen 115 that is displayed when the user 7 clicks the decision button 102 in Figure 9. In this diagnosis result display screen 115, a diagnostic image display section 111 in which edited dentition image information 39 is displayed is located on the upper left side of the display screen, and a diagnosis result display section 112 is located on the right side. The diagnosis result display section 112 evaluates the health state of the diagnosed teeth on a three-level scale, and is accompanied by illustrations to make the tooth health state easier to understand. A result display end button 113 is located below the diagnosis result display section 112, and the user 7 can end the result display by clicking this display end button 113.

[0066] Furthermore, user selection item information 36 is provided at the bottom of the display screen, and the user 7 selects the item he or she desires from among these items. The contents of the user selection item information 36 include age 56, toothbrush 62, periodontal disease prevention 57, toothpaste 63, caries prevention 58, tooth color 60, diet 64, bad breath countermeasures 61, tooth sensitivity 59, and assistive devices 65. Then, the generation AI creates an advice sentence based on the contents of the user selection item information 36 selected by the user 7.

[0067] In addition, an advice start button 114 is provided below the user selection item information 36. After the user 7 selects the desired item from the user selection item information 36, the user 7 clicks the advice start button 114, which causes the generation AI to start creating an advice sentence.

[0068] Next, we will discuss Figure 11. Figure 11 is a screen for creating advice text using the generation AI, and is displayed by clicking the advice start button 114 in Figure 10. On this screen, user-selected item information 36 is displayed at the top of the screen. The contents of this user-selected item information 36 are the same as those on the previously described screen in Figure 10: age 56, toothbrush 62, toothpaste 63, periodontal disease prevention 57, caries prevention 58, tooth color 60, diet 64, bad breath prevention 61, tooth sensitivity 59, and assistive devices 65. With regard to this user-selected item information 36, items selected by the user 7 are displayed in green, for example, while items not selected by the user 7 are displayed in gray, for example, so that the selected contents can be displayed appropriately.

[0069] Furthermore, a diagnostic image display unit 111 is disposed on the lower left side of the user selection item information 36, and a diagnostic result display unit 112 is disposed on the lower right side, and in addition to a three-level evaluation, the diagnostic result display unit 112 displays a health score indicating the dental health condition of the user 7 on a scale of 100 to the left of the three-level evaluation. The relationship between the health score and the three-level evaluation is as follows: if the health score is 60 points or less, reception is recommended; if the health score is 60 points to 80 points, caution is required; and if the health score is 80 points or more, the user is healthy.

[0070] Advice sentences created by the generation AI are arranged below the diagnostic image display unit 111 and the diagnostic result display unit 112. These advice sentences consist of an advice general evaluation unit 121 and an advice detail unit 122. In these units, the advice general evaluation unit 121 displays a general evaluation based on the diagnostic results of the diagnostic image display unit 111, and the advice detail unit 122 provides advice on toothpaste selection, diet, etc. based on the user selection item information 36.

[0071] [6. Imaging device used in the oral cavity examination system of this embodiment] Next, the photographing device used in the oral examination system of this embodiment will be described with reference to Fig. 12. One embodiment of the photographing device used in the oral examination system of this embodiment is a photographing device 138 in which a photographing auxiliary member 137 is attached to the center of the lens of a camera unit 136 of an electronic device 135 such as a smartphone, as shown in Fig. 12. In this photographing auxiliary member 137, a wide-angle lens 134 is disposed in the center, and a large number of LED light emitters 132 are disposed around this wide-angle lens 134. The LED light emitters 132 are disposed, for example, in a circular arrangement with 12 LED light emitters at equal intervals.

[0072] The photographing support member 137 is disposed behind a photographing unit 139 consisting of a wide-angle lens 134 and an LED light emitter 132, with a connection member 131 connected to the photographing unit 139. When photographing, as shown in FIG. 12(b), the photographing support member 137 is used in a state where the connection member 131 is placed over the top of an electronic device 135 such as a smartphone so as to cover it.

[0073] By using the imaging device 138 equipped with such imaging auxiliary member 137, it becomes possible to take a magnified image while illuminating the inside of the oral cavity of the user 7, which is the imaging target, and to take a high-resolution image. Furthermore, the LED light emitter 132 may be used not only to illuminate the inside of the oral cavity of the user 7, but also to highlight the color of the teeth and gums of the user 7, for example, by mixing in an ultraviolet light emitter. In this manner, the color of the teeth and gums of the user 7 stands out more, allowing for more accurate image diagnosis.

[0074] Another embodiment of the photographing support member 137 may be to place polarizing filters that transmit colors of a predetermined wavelength in front of or behind the wide-angle lens 134, thereby highlighting the teeth and gums of the user 7.

[0075] In another embodiment, a lens capable of capturing a spectral image may be provided instead of the wide-angle lens, and the image may be captured.

[0076] Furthermore, as one aspect of the photographing device according to one embodiment of the present invention, a camera-like photographing device including a short-focus lens and a proximity sensor may be used, which allows clear photographing even at a focal length of 100 mm or less. Specifically, as shown in FIG. 14, one aspect of such a camera-like photographing device 150 includes a short-focus lens 151, a handle 153, a cord 152, and the like. The short-focus lens 151 is a lens that forms a short focus that allows clear photographing even at a focal length of 100 mm or less. The device is used by connecting it to a personal computer or the like running the oral examination system of this embodiment via the cord 152.

[0077] FIG. 13 is a schematic diagram showing the actual imaging state of an oral examination system according to one embodiment of the present invention. As shown in this figure, this oral examination system is performed using an imaging device connected to a smartphone, personal computer, or the like. In this manner, the user 7 can carry the imaging device 140 such as a camera by himself, capture the dentition image information 38, and view the diagnosis results. The imaging device 140 such as a camera may be an imaging device 138 that is an electronic device such as a smartphone.

[0078] [7. Effects of the Oral Inspection System of the Present Embodiment] Next, the effects of the oral examination system of this embodiment will be described. First, the oral examination system of this embodiment does not require a doctor to directly diagnose the user 7. Therefore, the diagnosis can be completed online, which is very useful.

[0079] Furthermore, the oral examination system of this embodiment does not require the insertion of a pocket probe into the dentition of the user 7. This means that the user 7 does not experience negative emotions such as pain during the diagnosis, and can actively participate in the diagnosis. This is particularly important when making a diagnosis as a medical procedure.

[0080] Furthermore, the oral examination system of this embodiment uses a generation AI to create advice tailored to each individual user 7. This makes it highly convenient in that it can make an accurate diagnosis and provide optimal advice.

[0081] Furthermore, the oral cavity examination system of this embodiment performs image processing on the captured dentition image before determining whether or not an object is a tooth and performing image diagnosis, which results in higher diagnostic accuracy than conventional oral cavity examination systems.

[0082] Furthermore, the oral examination system of this embodiment allows self-examination to be performed by just one user 7. In this respect, the oral examination system of this embodiment is highly convenient.

[0083] Furthermore, the oral examination system of this embodiment allows the user 7 to perform a self-check by himself / herself. In this respect, the user 7 can understand the health condition of his / her own teeth and improve his / her awareness of dental health.

[0084] Furthermore, the oral cavity examination system of this embodiment allows the user 7 to understand the health condition of his or her own teeth, which can be expected to promote treatment of periodontal disease.

[0085] Furthermore, the oral examination system of this embodiment allows users 7 to be aware of the health of their own teeth and to seek early treatment, thereby reducing medical costs and ultimately reducing social insurance premiums.

[0086] Furthermore, the oral examination system of this embodiment makes it possible to raise awareness of voluntary care and early treatment for dental and oral diseases and their complications, thereby enabling users 7 to maintain a high level of health.

[0087] Furthermore, although the oral examination system according to this embodiment is used to diagnose periodontal disease in humans, it may also be used to diagnose periodontal disease in animals other than humans, such as pets such as dogs and cats, and livestock.

[0088] Furthermore, according to the oral examination system of this embodiment, an oral diagnosis system can be constructed using an existing public communication network, electronic devices such as smartphones with camera functions that can be connected to such a public communication network, and a network system server 4, making the initial construction of this oral examination system easy.

[0089] Furthermore, with regard to the imaging function of the electronic device 135 such as a smartphone, the electronic device 135 such as a smartphone that has become widespread in recent years is capable of capturing high-resolution images and is equipped with high-performance functions for capturing clearer images, such as an autofocus function and an image stabilization function, etc. Therefore, even with such commercially available electronic device 135 such as a smartphone, sufficiently accurate image diagnosis can be performed.

[0090] Furthermore, according to one aspect of the present invention, the mouth is held open with a dedicated mouth opener, and then the dentition image information 38 is captured. This allows for accurate diagnosis of the health condition of the oral cavity, including the teeth and gums, of the user 7.

[0091] According to one aspect of the present invention, the dentition image information 38 of the user 7 to be imaged is edited by adjusting brightness and darkness, and the edited dentition image information 39 is used for image diagnosis. Therefore, by using clearer dentition image information 38, more precise image diagnosis can be performed.

[0092] Furthermore, according to one aspect of the present invention, it is possible to arrange an emitter of light of a specific wavelength, such as ultraviolet light, which makes the color of the teeth and gums of the user 7 stand out when capturing an image, mixed in with part of the LED emitter 132.

[0093] Furthermore, according to one aspect of the present invention, the photographing device is configured such that polarizing filters that transmit colors of a predetermined wavelength are disposed in front of and behind the lens portion for photographing, thereby enabling the photographing of clearer dentition photographed image information 38, and as a result, more accurate image diagnosis can be performed.

[0094] Furthermore, when imaging is performed using a lens for capturing a spectral image of the gums according to an embodiment of the present invention, blood flow in the gums can be measured, which provides new insight into the condition of one's own gums and helps to more accurately understand the health of one's own gums.

[0095] Furthermore, in the image diagnosis according to one embodiment of the present invention, the health condition of the teeth is diagnosed from multiple perspectives, such as the color of the gums, the shape of the gums between the teeth, and the blood flow in the gums, etc. Therefore, the accuracy of the diagnosis is high.

[0096] Furthermore, in the image diagnosis according to one embodiment of the present invention, if the image diagnosis reveals a problem with the health of the gums, the user 7 is notified of the progress of periodontal disease or alveolar pyorrhea and a future prediction of such diseases. Therefore, the image diagnosis according to one embodiment of the present invention is highly convenient.

[0097] Furthermore, in the image diagnosis according to one embodiment of the present invention, diagnosis is made based on the normal color of teeth, so it is possible to sound the alarm about excessively white teeth caused by excessive brushing or excessive whitening.

[0098] Furthermore, in the image diagnosis according to one embodiment of the present invention, if there is a problem with the health of the gums, the user 7 is also notified of a future prediction of the gums, thereby raising the user 7's awareness of oral care.

[0099] The above description is merely one embodiment of the present invention. In other words, the present invention is not limited to the embodiments described in this specification. Furthermore, the present invention allows modifications or changes to the implementation without departing from the spirit and scope of the present invention as defined by the claims. [Explanation of symbols]

[0100] 1. Administrator terminal 2 Public communication network 3. User terminal 4 Network System Server 5 Application Server 6 Database Server 7 User 11 Storage section 12 Input / output control section 13 Control Unit 14 External communication I / F control section 36 User selection information 37 Image editing function information 38 Dental arch imaging information 39 Edited dental arch image information 51 Diagnostic information learned by generative AI 52 Image AI 54 Symptom classification diagnosis results 55 User Selection Items 70 Start screen 71 Center of start screen 72 Personal Number Entry Field 73 Terms of Use Agreement Button 80 Shooting startup screen 81 Shooting screen 82 Start shooting button 83 Shooting button 90 small screen 101 Diagnostic image selection screen 102 Decision button 111 Diagnostic image display unit 112 Diagnostic result display section 113 End display button 114 Advice Start Button 115 Display screen 121 General Affairs Section 121 122 Advice Details 131 Connecting member 132 LED light source 134 Wide-angle lens 135 Smartphones and other electronic devices 136 Camera Department 137 Photography Auxiliary Materials 138 Imaging Device 139 Photography Department 140 Cameras and other photographic equipment 150 Camera-like photographic device 151 Short focal length lens 152 Code section 153 Handle

Claims

1. A step of capturing an image of a dental row including one or more teeth from the front of the oral cavity to create a dental row photographed image; a step of image processing and editing the photographed image of the row of teeth to create an edited photographed image of the row of teeth; a step of diagnosing a risk level, which is the progression of periodontal disease of the user, based on the edited dentition photographed image; and a step of generating optimal advice text about teeth for each user based on the diagnosis result of the step of diagnosing the risk level and the edited dentition photographed image, using a generation AI tool; The step of diagnosing the risk level includes: The system is configured to take photographs of dentition affected by periodontal disease in various stages of periodontal disease progression, or healthy dentition without periodontal disease, from the front of the oral cavity, and diagnose the risk level of periodontal disease progression using an image artificial intelligence tool that has been trained in software in advance as big data from the group of periodontal disease dentition photographs taken of dentition in various stages of periodontal disease progression. The method for image diagnosis of oral conditions according to claim 1.

2. In the edited dentition photographed image, Using the average number of pixels in the maxillary dental row length corresponding to the vertical length of the maxillary dental row and the average number of pixels in the mandibular dental row length corresponding to the vertical length of the mandibular dental row, determining a region of the row of teeth in the edited photographed image of the row of teeth based on the ratio of the average number of pixels of the row of teeth in the upper jaw to the average number of pixels of the row of teeth in the lower jaw; The diagnostic imaging method according to claim 1 .

3. The edited dentition photographed image is analyzed using a function, determining a region of the row of teeth in the edited row of teeth photographed image; The diagnostic imaging method according to claim 1 .

4. The image processing and editing of the dentition photographed image includes: Based on the same dentition photographed image, exposure-adjusted dentition photographed images are created by adjusting the exposure to various levels of brightness; A configuration in which multiple exposure-adjusted dental arch photographed images are superimposed and synthesized to perform image processing. The diagnostic imaging method according to claim 1 .

5. An imaging device used in the step of creating the dentition image includes: The dental arch photographic image is created by photographing using an imaging device configured to receive light with a short focal length lens that enables clear photography even when the focal length is 100 mm or less and a proximity sensor. The imaging diagnostic method of claim 1.

Images