Program and index value calculation device

Abstract

To provide novel technology for analyzing an intraoral state using three-dimensional data representing the oral cavity.SOLUTION: A program according to the present disclosure causes a computer to execute: an acquisition step for acquiring intraoral three-dimensional data in which the teeth of a subject and a peripheral region of the teeth are included; and a calculation step for calculating a gum state index value which is an index value related to the gum state of the subject and an alveolar bone absorption index value which is an index value related to the absorption of the alveolar bone of the subject, or both of these values, as a state index value which is an index value related to the state of teeth of the subject, by using the three-dimensional data.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a program and an index value calculation device. [Background technology]

[0002] A technique for predicting and diagnosing periodontal-related diseases using data representing information in the oral cavity has been disclosed. For example, Patent Document 1 discloses a technique for testing gingivitis by comparing periodontal image data obtained by capturing an image of the oral cavity with reference color data, which is gingival color data indicating the state of gingivitis. Patent Document 2 discloses a technique for generating an oral photograph by synthesizing a plurality of partial images of a part of the oral cavity, and judging the presence and state of a predetermined disease in the oral cavity based on the image characteristics of a predetermined judgment target region. Patent Document 3 discloses a technique for estimating the state of an oral cavity target region from image data of the oral cavity region captured by a camera, and judging the state of the oral cavity target region based on the estimation result and predetermined reference information. Patent Document 4 discloses a technique for predicting oral health by comprehensively analyzing the presence or absence of orthodontic treatment, the state of caries, the state of periodontitis, the state of prosthetic appliances, and a questionnaire by analyzing the oral photograph using a machine learning algorithm. Patent Document 5 discloses a technique for estimating the presence or absence of periodontal disease by inputting an image of the periodontal region of an intraoral image into a model. Patent Document 6 discloses a technology for identifying areas in the oral cavity that are likely to cause disease by applying three-dimensional data of the shape of the oral cavity to an estimation model that includes a neural network. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2019-030587 A [Patent Document 2] JP 2019-155027 A [Patent Document 3] Patent Publication No. 2021-053175 [Patent Document 4] Special Publication No. 2022-508923 [Patent Document 5] Patent Publication No. 2022-073148 [Patent Document 6] Patent Publication No. 2022-012199 Summary of the Invention [Problem to be solved by the invention]

[0004] The information on the inside of the oral cavity includes information on the state of the gums and the resorption of the alveolar bone. None of the above-mentioned patent documents mentions the use of three-dimensional data representing the inside of the oral cavity to analyze the state of the gums or the resorption of the alveolar bone. The present disclosure has been made in view of the above-mentioned problems, and its purpose is to provide a new technique for analyzing the state of the oral cavity using three-dimensional data representing the inside of the oral cavity. [Means for solving the problem]

[0005] The program disclosed herein causes a computer to execute an acquisition step of acquiring an oral cavity image including the subject's teeth and the areas surrounding the teeth, and a calculation step of using the oral cavity image to calculate a gingival condition index value that is an index value related to the gingival condition of the subject, an alveolar bone resorption index value that is an index value related to the resorption of the alveolar bone of the subject, or both of these.

[0006] The index value calculation device disclosed herein includes an acquisition unit that acquires an oral cavity image including the subject's teeth and the areas surrounding the teeth, and a calculation unit that uses the oral cavity image to calculate a gingival condition index value that is an index value related to the gingival condition of the subject, an alveolar bone resorption index value that is an index value related to the resorption of the alveolar bone of the subject, or both. Effect of the Invention

[0007] According to the present invention, a new technique is provided for analyzing the condition inside the oral cavity using three-dimensional data representing the inside of the oral cavity. [Brief description of the drawings]

[0008] [Figure 1]FIG. 2 is a diagram illustrating an example of an outline of the operation of the index value calculation device. [Diagram 2] FIG. 1 illustrates the position of various parts of a tooth in relation to condition index values. [Diagram 3] 2 is a block diagram illustrating a functional configuration of an index value calculation device. FIG. [Figure 4] FIG. 2 is a block diagram illustrating a hardware configuration of a computer that realizes the index value calculation device. [Diagram 5] 1 is a flowchart illustrating a flow of a process executed by an index value calculation device. [Figure 6] 10 is a flowchart illustrating an example of a process for calculating a gingival condition index value. [Figure 7] 11 is a diagram illustrating an example of a case in which a gingival condition index value is calculated using a gingival condition index value calculation model. FIG. [Figure 8] FIG. 11 is a diagram illustrating a case in which PPD is calculated using a first prediction model. [Figure 9] FIG. 13 is a diagram illustrating a case in which CAL is calculated using a first prediction model. [Figure 10] 11 is a diagram illustrating an example of a case in which a gingival condition index value is calculated using a first feature amount calculation model and a gingival condition index value calculation model. FIG. [Figure 11] FIG. 13 is a diagram illustrating a case in which PPD is calculated using a first feature amount calculation model and a first prediction model. [Figure 12] 13 is a diagram illustrating an example of a case in which a first feature amount calculated without using a first feature amount calculation model is used by a gum condition index value calculation model. FIG. [Figure 13] 13 is a diagram illustrating a case in which a first feature amount calculated without using a first feature amount calculation model is used by a first prediction model. FIG. [Figure 14] 13 is a diagram showing a case where a part of a tooth of interest is included in both the region of interest and the related region. FIG. [Figure 15] 1 is a diagram illustrating an example of a region of interest that includes only a portion of a tooth of interest, and two related regions. [Figure 16]FIG. 13 is a diagram illustrating an example of a case in which a gingival condition index value 20 is calculated by a gingival condition index value calculation model to which a region of interest and a related region are input. [Figure 17] FIG. 13 is a diagram illustrating a case in which a PPD is calculated using a first prediction model to which a region of interest and a related region are input. [Figure 18] 11 is a flowchart illustrating an example of a process for calculating an alveolar bone resorption index value. [Figure 19] 11 is a diagram illustrating a case in which an alveolar bone resorption index value is calculated using an alveolar bone resorption index value calculation model. FIG. [Figure 20] FIG. 13 is a diagram illustrating an example of a case in which the alveolar bone resorption degree is calculated using the second prediction model. [Figure 21] 13 is a diagram illustrating a case in which an alveolar bone resorption index value is calculated using a third feature amount calculation model and an alveolar bone resorption index value calculation model. FIG. [Figure 22] 13 is a diagram illustrating an example of a case in which the alveolar bone resorption level is calculated using the third feature amount calculation model and the second prediction model. FIG. [Figure 23] 13 is a diagram illustrating a case in which a third feature amount calculated without using the third feature amount calculation model is used by the alveolar bone resorption index value calculation model. FIG. [Figure 24] 13 is a diagram illustrating a case in which a third feature amount calculated without using a third feature amount calculation model is used by a second prediction model. FIG. [Diagram 25] 13 is a diagram illustrating an example of a case in which an alveolar bone resorption index value is calculated by an alveolar bone resorption index value calculation model to which a region of interest and a related region are input. FIG. [Figure 26] FIG. 13 is a diagram illustrating an example of a case in which the degree of alveolar bone resorption is calculated using a second prediction model to which a region of interest and a related region are input. [Figure 27] 1 is a block diagram illustrating a functional configuration of an index value calculation device that performs a discrimination process using an index value. [Figure 28] FIG. 13 is a diagram illustrating training of a gingival condition index value calculation model. [Figure 29] FIG. 1 illustrates the training of a first prediction model. [Diagram 30] 11 is a diagram illustrating an example of training of a first feature amount calculation model and a gingival condition index value calculation model. FIG. [Diagram 31] FIG. 13 is a diagram illustrating an example of training of a gingival condition index value calculation model, in which a first feature amount calculated by a predetermined algorithm is input. [Diagram 32] FIG. 13 illustrates training of a gingival condition index calculation model in which relevant regions are used. [Diagram 33] FIG. 1 illustrates the training of a first predictive model in which relevant regions are used. [Diagram 34] FIG. 13 is a diagram illustrating training of an alveolar bone resorption index value calculation model. [Diagram 35] FIG. 13 illustrates the training of a second predictive model. [Diagram 36] 13 is a diagram illustrating an example of training of a third feature amount calculation model and an alveolar bone resorption index value calculation model. FIG. [Figure 37] FIG. 13 is a diagram illustrating an example of training of an alveolar bone resorption index value calculation model, in which a third feature value calculated by a predetermined algorithm is input. [Figure 38] FIG. 13 illustrates an example of training of an alveolar bone resorption index value calculation model in which relevant regions are used. [Figure 39] FIG. 13 illustrates the training of a second predictive model in which relevant regions are used. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, an embodiment of the present disclosure will be described in detail with reference to the drawings. In each drawing, the same or corresponding elements are given the same reference numerals, and duplicated explanations are omitted as necessary for clarity of explanation. Furthermore, unless otherwise specified, predetermined values ​​such as predetermined values ​​and threshold values ​​are stored in advance in a storage device accessible from a device that uses the values. Furthermore, unless otherwise specified, the storage unit is composed of one or more arbitrary number of storage devices.

[0010] Fig. 1 is a diagram illustrating an example of an outline of the operation of the index value calculation device 2000. Here, Fig. 1 is a diagram for facilitating understanding of the outline of the index value calculation device 2000, and the operation of the index value calculation device 2000 is not limited to the operation shown in Fig. 1.

[0011] The index value calculation device 2000 acquires oral cavity three-dimensional data 10. The oral cavity three-dimensional data 10 is data representing the three-dimensional shape of the teeth present in the oral cavity of the subject and the surface of the periodontal region of the teeth. Here, any data capable of representing the three-dimensional shape of an object can be used for the oral cavity three-dimensional data 10. The data representing the three-dimensional shape of an object is, for example, a three-dimensional model (for example, voxel data or mesh data) or point cloud data. The oral cavity three-dimensional data 10 may represent the distribution of colors in the oral cavity of the subject in addition to the three-dimensional shape in the oral cavity of the subject. For example, when the oral cavity three-dimensional data 10 is mesh data, the oral cavity three-dimensional data 10 represents the three-dimensional shape in the oral cavity of the subject by a plurality of meshes, and further represents the distribution of colors in the oral cavity of the subject by indicating the color of each mesh. Note that the oral cavity three-dimensional data 10 may include one or more teeth.

[0012] The index value calculation device 2000 calculates a condition index value, which is an index value related to the condition of the teeth of a subject, using the three-dimensional oral cavity data 10. As the condition index value, a gingival condition index value 20, an alveolar bone resorption index value 30, or both of them are calculated. The gingival condition index value 20 is an index value representing the condition of the gingiva. The index value representing the condition of the gingiva is, for example, a value representing a clinical attachment level (CAL), a probing pocket depth (PPD), the presence or absence of bleeding on probing (BOP), or a class of furcation lesion. The gingival condition index value 20 calculated by the index value calculation device 2000 is a predicted value of these various index values.

[0013] CAL is the distance from the cemento-enamel junction (CEJ) to the bottom of the gingival sulcus or pocket measured with a periodontal probe, and is an index of the state of attachment of the gingiva to the tooth surface. PPD is the distance from the gingival margin to the tip of the probe when the periodontal probe is inserted, and like CAL, is an index of the state of attachment of the gingiva to the tooth surface. Generally, the tip of the probe is the bottom of the pocket. CAL and PPD are expressed as numbers in units of 1 mm or 0.5 mm, for example.

[0014] Fig. 2 is a diagram illustrating the positions of various tooth parts related to the condition index value. Fig. 2 shows the gingival margin, CEJ, pocket bottom, alveolar crest, and root apex as tooth parts related to the index value. In Fig. 2, the position of each part is expressed as a relative position with the position of the gingival margin as the reference (i.e., zero).

[0015] For example, CAL can be expressed as the distance from the CEJ to the bottom of the pocket. In Figure 2, the CEJ and the bottom of the pocket are located at -1mm and 5mm, respectively. Therefore, CAL is 5-(-1)=6mm.

[0016] BOP is the occurrence of bleeding from the bottom of the pocket upon probing, and the presence or absence of this bleeding can be used to assess the resistance of the bottom of the pocket and the presence of inflammation.

[0017] Furcation disease is the extension of periodontal and / or pulp disease to the interradicular space of a multi-rooted tooth. When the Lindhe and Hyman furcation disease classification is used, furcation disease is classified into three classes: 1st, 2nd, and 3rd. When the Glickman furcation disease classification is used, furcation disease is classified into four classes: 1st, 2nd, 3rd, and 4th.

[0018] The index value calculation device 2000 may calculate multiple types of gingival condition index values ​​20, such as CAL and PPD.

[0019] The alveolar bone resorption index value 30 is an index value representing the degree of resorption of the alveolar bone. The index value relating to alveolar bone resorption is, for example, a value representing the degree of alveolar bone resorption (Bone Level: BL) or the alveolar bone resorption ratio (ABR ratio). The degree of alveolar bone resorption is calculated from the ratio of the resorbed alveolar bone distance to the tooth root length. Specifically, the degree of alveolar bone resorption is calculated as B / A based on the distance A from the CEJ to the tooth root apex and the distance B from the CEJ to the alveolar bone crest. The degree of alveolar bone resorption is expressed, for example, as a real number between 0 and 1. The alveolar bone resorption ratio is calculated as B / A*100 [%] using the above-mentioned A and B. The alveolar bone resorption index value 30 calculated by the index value calculation device 2000 is a predicted value of these various index values.

[0020] For example, in Figure 2, the positions of the CEJ, the apex, and the alveolar bone crest are -1mm, 15mm, and 7mm, respectively. Therefore, in calculating the degree of alveolar bone resorption, A = 15-(-1) = 16[mm] and B = 7-(-1) = 8[mm]. Therefore, the degree of alveolar bone resorption is 8 / 16 = 0.5.

[0021] The index value calculation device 2000 may calculate a plurality of types of alveolar bone resorption index values ​​30, such as the alveolar bone resorption degree and the alveolar bone resorption rate.

[0022] <Examples of effects> Periodontal disease, also called periodontal disease, is broadly divided into gingival lesions and periodontitis. Gingival lesions, especially plaque-induced gingivitis, are gingival inflammation caused by bacteria present at the gingival margin. Periodontitis, especially chronic periodontitis, is a chronic inflammatory disease accompanied by attachment loss and alveolar bone resorption caused by periodontal pathogenic bacteria. For this reason, understanding the condition of the gingiva and the resorption of alveolar bone is useful for diagnosing periodontal-related diseases.

[0023] For example, according to the American Academy of Periodontology / European Federation of Periodontology's 2018 New Classification of Periodontitis, the classification of stage I to IV, which indicates the severity of periodontitis, is based on CAL, the degree of bone resorption, and the number of teeth lost due to periodontitis. On the other hand, according to the 2022 Guidelines for Periodontal Treatment of the Japanese Society of Periodontology, the diagnosis of plaque-induced gingivitis and periodontitis is made according to the classification of periodontal disease based on periodontal tissue examinations, etc. The classification of periodontitis according to the degree of tissue destruction in the diagnosis of a single tooth is based on the presence or absence of BL, CAL, or furcation lesions. The classification of periodontitis according to the degree of inflammation is based on PPD. In addition, according to a report by the Periodontal Medicine Committee of the Japanese Society of Periodontology, the committee has proposed a classification of severity recommended by the committee, with an alveolar bone resorption rate of 25% or less as clinically mild, 25% to 35% as clinically moderate, and 35% or more as clinically severe.

[0024] Here, the CAL, PPD, BOP, or furcation lesion, which are indicators of gingival condition, can be identified by measurement using a probe. In addition, the alveolar bone resorption degree and alveolar bone resorption rate, which are indicators of alveolar bone resorption, can be identified by measurement using X-ray images. However, identification using these methods has a problem in that it is difficult for anyone other than a person with specialized knowledge and skills (such as a dentist) to perform. In addition, measurement using a probe has a problem in that the measurement is complicated and time-consuming. Furthermore, measurement using X-ray images has a problem in that expensive and special devices and equipment such as an X-ray device and an X-ray room are required, and there is a problem in that the subject is exposed to radiation.

[0025] In this regard, according to the index value calculation device 2000, the gingival condition index value 20, which is a predicted value of an index representing the gingival condition, the alveolar bone resorption index value 30, which is a predicted value of an index representing the degree of resorption of the alveolar bone, or both of them are calculated using the oral cavity three-dimensional data 10. When the gingival condition index value 20 is calculated by the index value calculation device 2000, the index value representing the gingival condition can be calculated by using the index value calculation device 2000 without actually measuring the subject using a probe. Therefore, by using the index value calculation device 2000, even a person without specialized knowledge or skills can easily calculate the gingival condition index value 20. In addition, by using the index value calculation device 2000, the gingival condition index value 20 can be easily calculated without performing a cumbersome and time-consuming measurement using a probe.

[0026] Furthermore, when the alveolar bone resorption index value 30 is calculated by the index value calculation device 2000, the index value calculation device 2000 can be used to calculate an index value representing resorption of the alveolar bone without taking an X-ray of the subject's oral cavity. Therefore, even a person without specialized knowledge or skills can easily calculate the alveolar bone resorption index value 30 by using the index value calculation device 2000. Furthermore, the index value calculation device 2000 can be used to calculate the alveolar bone resorption index value 30 without using expensive and special devices and equipment such as an X-ray imaging device or X-ray imaging equipment. Furthermore, the index value calculation device 2000 can be used to calculate the alveolar bone resorption index value 30 without the risk of radiation exposure.

[0027] The index value calculation device 2000 of this embodiment will be described in more detail below.

[0028] <Example of functional configuration> 3 is a block diagram illustrating a functional configuration of the index value calculation device 2000. The index value calculation device 2000 has an acquisition unit 2020 and a calculation unit 2040. The acquisition unit 2020 acquires three-dimensional oral cavity data 10. The calculation unit 2040 uses the three-dimensional oral cavity data 10 to calculate a gingival condition index value 20, an alveolar bone resorption index value 30, or both of them.

[0029] <Example of hardware configuration> Each functional component of the index value calculation device 2000 may be realized by hardware that realizes each functional component (e.g., a hardwired electronic circuit, etc.), or may be realized by a combination of hardware and software (e.g., a combination of an electronic circuit and a program that controls it, etc.). Below, a further explanation will be given of the case where each functional component of the index value calculation device 2000 is realized by a combination of hardware and software.

[0030] 4 is a block diagram illustrating an example of a hardware configuration of a computer 1000 that realizes the index value calculation device 2000. The computer 1000 is any computer. For example, the computer 1000 is a stationary computer such as a PC (Personal Computer) or a server machine. In addition, for example, the computer 1000 is a portable computer such as a smartphone or a tablet terminal. The computer 1000 may be a dedicated computer designed to realize the index value calculation device 2000, or may be a general-purpose computer.

[0031] For example, by installing a specific application in the computer 1000, each function of the index value calculation device 2000 is realized in the computer 1000. The application is constituted by a program for realizing each functional component of the index value calculation device 2000.

[0032] The above-mentioned program may be acquired by any method. For example, the program may be acquired from a storage medium (such as a DVD (Digital Versatile Disk) or a USB (Universal Serial Bus) memory) in which the program is stored. Alternatively, the program may be acquired by downloading the program from a server device that manages the storage device in which the program is stored.

[0033] The computer 1000 has a bus 1020, a processor 1040, a memory 1060, a storage device 1080, an input / output interface 1100, and a network interface 1120. The bus 1020 is a data transmission path for the processor 1040, the memory 1060, the storage device 1080, the input / output interface 1100, and the network interface 1120 to transmit and receive data to and from each other. However, the method of connecting the processor 1040 and the like to each other is not limited to a bus connection.

[0034] The processor 1040 is a processor such as a central processing unit (CPU), a graphics processing unit (GPU), a field-programmable gate array (FPGA), or a digital signal processor (DSP). The memory 1060 is a main storage device realized using a random access memory (RAM) or the like. The storage device 1080 is an auxiliary storage device realized using a hard disk, a solid state drive (SSD), a memory card, or a read only memory (ROM) or the like.

[0035] The input / output interface 1100 is an interface for connecting the computer 1000 to an input / output device. For example, the input / output interface 1100 is connected to an input device such as a keyboard and an output device such as a display device.

[0036] The network interface 1120 is an interface for connecting the computer 1000 to a network. This network may be a LAN (Local Area Network) or a WAN (Wide Area Network).

[0037] The storage device 1080 stores a program (a program that realizes the above-mentioned application) that realizes each functional component of the index value calculation device 2000. The processor 1040 realizes each functional component of the index value calculation device 2000 by reading this program into the memory 1060 and executing it.

[0038] The index value calculation device 2000 may be realized by one computer 1000, or may be realized by multiple computers 1000. In the latter case, the configurations of the computers 1000 do not need to be the same, and can be different from each other.

[0039] The index value calculation device 2000 may be installed in a place where the subject's oral cavity is imaged, such as a medical examination facility, or may be installed in a place other than the place where the subject's oral cavity is imaged. In the latter case, for example, the index value calculation device 2000 can be realized as a server device configured to receive the oral cavity three-dimensional data 10 from a terminal used to image the subject's oral cavity.

[0040] <Processing flow> 5 is a flowchart illustrating a process executed by the index value calculation device 2000. The acquisition unit 2020 acquires the oral cavity three-dimensional data 10 (S102). The calculation unit 2040 uses the oral cavity three-dimensional data 10 to calculate the gingival condition index value 20, the alveolar bone resorption index value 30, or both of them (S104).

[0041] Here, when the calculation unit 2040 calculates both the gingival condition index value 20 and the alveolar bone resorption index value 30, the calculation of the gingival condition index value 20 and the calculation of the alveolar bone resorption index value 30 may be performed in any order. Moreover, the calculation of the gingival condition index value 20 and the calculation of the alveolar bone resorption index value 30 may be performed in parallel.

[0042] <How to generate oral 3D data 10> The three-dimensional oral cavity data 10 is generated, for example, by capturing images of the subject's teeth and the periodontal region of the teeth using an arbitrary imaging device. Note that existing technology can be used to generate data representing the three-dimensional shape and color distribution of an object using an imaging device.

[0043] As the imaging device used to generate the three-dimensional oral cavity data 10, various imaging devices can be used, such as a smartphone camera, a general-purpose digital camera, a dental digital camera, a dental camera for intraoral photography, a three-dimensional scanner, an intraoral three-dimensional scanner, or a dental three-dimensional scanner. The imaging device is not limited to a device that captures visible light, and may be a device that captures near-infrared light, or a device that captures reflected light consisting of a single or multiple wavelengths when a single wavelength from visible to near-infrared is irradiated.

[0044] The light source used for imaging the subject's teeth may be any light source such as sunlight, incandescent light, fluorescent light, LED, laser light, near-infrared light source, etc. Here, the amount of light irradiated into the oral cavity may be adjusted using a grating or a wavelength selection filter.

[0045] The oral cavity three-dimensional data 10 may be generated using a ranging device such as LiDAR (light detection and ranging). For example, the oral cavity of the subject may be scanned using a ranging device to obtain point cloud data representing a three-dimensional shape of the oral cavity of the subject. The thus obtained point cloud data may be used to generate a three-dimensional model such as mesh data.

[0046] <Acquisition of oral 3D data 10: S102> The acquiring unit 2020 acquires the three-dimensional oral cavity data 10 (S102). There are various methods for the acquiring unit 2020 to acquire the three-dimensional oral cavity data 10. For example, the three-dimensional oral cavity data 10 is stored in advance in a storage unit accessible from the index value calculation device 2000. In this case, the acquiring unit 2020 acquires the three-dimensional oral cavity data 10 from the storage unit.

[0047] When a plurality of pieces of oral cavity three-dimensional data 10 are stored in the storage unit, for example, the acquisition unit 2020 receives a designation of the oral cavity three-dimensional data 10 to be used from a user of the index value calculation device 2000. The acquisition unit 2020 acquires the oral cavity three-dimensional data 10 designated by the user from the plurality of oral cavity three-dimensional data 10 stored in the storage unit.

[0048] Here, the storage unit in which the oral cavity three-dimensional data 10 is stored may be a storage unit (e.g., the storage device 1080) provided inside the index value calculation device 2000, or may be a storage unit provided outside the index value calculation device 2000. For example, when an imaging device that generates the oral cavity three-dimensional data 10 is provided in the index value calculation device 2000, the oral cavity three-dimensional data 10 generated by the imaging device may be stored in a storage unit inside the index value calculation device 2000. Then, the index value calculation device 2000 acquires the oral cavity three-dimensional data 10 stored in the storage unit inside the index value calculation device 2000.

[0049] The acquisition unit 2020 may acquire the oral cavity three-dimensional data 10 by receiving the oral cavity three-dimensional data 10 transmitted from another device. In this case, for example, the oral cavity three-dimensional data 10 is transmitted to the index value calculation device 2000 from a terminal operated by a user of the index value calculation device 2000 (hereinafter, a user terminal).

[0050] Alternatively, for example, the acquisition unit 2020 may acquire one or more images of the subject's oral cavity from an imaging device, and generate the oral cavity three-dimensional data 10 using the acquired images. Here, the imaging device may be provided in a user terminal. In this case, for example, a user operates the user terminal to cause the imaging device to capture an image of the subject's oral cavity, thereby causing the imaging device to generate an image of the subject's oral cavity. After that, the user further operates the user terminal to transmit the above-mentioned image from the user terminal to the index value calculation device 2000.

[0051] The acquisition unit 2020 may apply a predetermined image processing to an image obtained from the imaging device, and generate the three-dimensional oral cavity data 10 using the image to which the image processing has been applied. For example, the acquisition unit 2020 acquires an image generated by capturing an image of the subject's oral cavity without using a filter, and applies a predetermined filter processing to this image. Then, the acquisition unit 2020 generates the three-dimensional oral cavity data 10 using the image to which the filter processing has been applied. As the filter processing, for example, a wavelength selection filter application process for obtaining an image of only a specific wavelength region, a polarizing filter application process for reducing the influence of reflection, etc., can be adopted.

[0052] <Calculation of state index value: S104> The calculation unit 2040 calculates the gingival condition index value 20, the alveolar bone resorption index value 30, or both of them (S104) using the oral cavity three-dimensional data 10. Hereinafter, a method for calculating the gingival condition index value 20 and a method for calculating the alveolar bone resorption index value 30 will be described respectively.

[0053] <<How to calculate the gingival condition index value 20>> For example, the calculation unit 2040 extracts a region of interest, which is a three-dimensional region including a tooth and the periodontal region of the tooth, from the oral cavity three-dimensional data 10, and uses this region of interest to calculate the gingival condition index value 20. Hereinafter, the tooth and the periodontal region of the tooth that are the targets for predicting the condition index value are referred to as the tooth of interest and the periodontal region of interest, respectively.

[0054] The region of interest includes a portion that is a target for predicting a condition index value. When the entire tooth of interest is a target for predicting a condition index value, the region of interest includes the tooth of interest and the periodontal region of interest. When a part of the tooth of interest is a target for predicting a condition index value, the region of interest includes the part of the tooth of interest and the periodontal region of the part. Note that when the entire oral cavity three-dimensional data 10 represents one tooth and the periodontal region of the tooth (in other words, the oral cavity three-dimensional data 10 includes one tooth) and the entire one tooth is a target for predicting a condition index value, the calculation unit 2040 may treat the entire oral cavity three-dimensional data 10 as one region of interest.

[0055] When the oral cavity three-dimensional data 10 includes a plurality of teeth, for example, the calculation unit 2040 extracts an attention region for each of all teeth included in the oral cavity three-dimensional data 10. Alternatively, for example, the calculation unit 2040 may extract an attention region only for each of one or more specific teeth among the plurality of teeth included in the oral cavity three-dimensional data 10. For example, the calculation unit 2040 extracts an attention region only for a tooth that is closest to a specific position (for example, a central position) of the oral cavity three-dimensional data 10 among the plurality of teeth included in the oral cavity three-dimensional data 10. Alternatively, for example, the calculation unit 2040 extracts an attention region only for a tooth designated by a user among the plurality of teeth included in the oral cavity three-dimensional data 10.

[0056] As described above, the condition index value prediction target may be a part of a tooth. In this case, the calculation unit 2040 may extract multiple attention regions for one tooth. For example, the calculation unit 2040 detects a three-dimensional region representing a tooth and its periodontal region from the oral cavity three-dimensional data 10, and divides the detected three-dimensional region according to a predetermined division rule to extract multiple attention regions for one tooth.

[0057] Here, various rules can be adopted as the division rule. For example, the division rule is "divide into a buccal region and a lingual region." Another example of the division rule is "divide into a mesial portion and a distal portion." Here, the mesial portion refers to the portion of the tooth and its periodontal region that is closer to the midline (the side away from the molars). The distal portion refers to the portion of the tooth and its periodontal region that is closer to the midline (the side closer to the molars).

[0058] The division rule may be a rule for dividing a three-dimensional region of a tooth and the gums around the tooth into three or more (e.g., four or six) regions of interest. For example, the division rule may be a rule such as "divide into a predetermined number of regions in the horizontal direction" or "divide into a predetermined number of regions in the vertical direction." In addition, a rule for extracting one or more three-dimensional regions of a predetermined shape (such as a rectangular parallelepiped, cube, sphere, or oval sphere) may be used as the division rule.

[0059] The calculation unit 2040 may use a plurality of division rules. For example, the calculation unit 2040 uses two division rules, namely, "divide into a buccal region and a lingual region" and "divide into a mesial portion and a distal portion." In this case, the calculation unit 2040 performs both a process of dividing a three-dimensional region representing a tooth and its periodontal region into a buccal portion and a lingual portion, and a process of dividing into a mesial portion and a distal portion. As a result, four regions of interest are obtained for one tooth.

[0060] The extraction of the region of interest is performed, for example, by utilizing a machine learning model trained to extract the region of interest from the three-dimensional oral cavity data 10. Hereinafter, the model that extracts the region of interest from the three-dimensional oral cavity data 10 is referred to as a region of interest extraction model. As the region of interest extraction model, any machine learning model (for example, a neural network such as a Convolutional Neural Network (CNN)) that can extract a predetermined three-dimensional region from three-dimensional data can be used.

[0061] The region-of-interest extraction model is configured to output one or more three-dimensional regions to be treated as regions of interest in response to input of the three-dimensional oral cavity data 10, for example.

[0062] The extraction of the region of interest does not necessarily require the use of a model. For example, the calculation unit 2040 may extract the region of interest from the three-dimensional oral cavity data 10 by analyzing the three-dimensional oral cavity data 10 using a predetermined algorithm.

[0063] 6 is a flowchart illustrating a process for calculating the gingival condition index value 20. The calculation unit 2040 extracts one or more regions of interest from the three-dimensional oral cavity data 10 (S202). Steps S204 to S208 constitute a loop process L1 that is executed for each region of interest. In S204, the calculation unit 2040 determines whether or not the loop process L1 has been executed for all regions of interest. If the loop process L1 has already been executed for all regions of interest, the process in FIG. 6 ends.

[0064] If there is a region of interest that has not yet been the target of the loop process L1, the calculation unit 2040 selects one of the regions of interest that has not yet been the target of the loop process L1. The region of interest selected here is denoted as region of interest i.

[0065] The calculation unit 2040 calculates the gum condition index value 20 for the attention region i (S206). Since S208 is the end of the loop process L1, S204 is executed again.

[0066] The process of calculating the gingival condition index value 20 from the region of interest (S206) is performed, for example, by using a trained machine learning model. Hereinafter, the model for calculating the gingival condition index value 20 is referred to as a gingival condition index value calculation model.

[0067] The gingival condition index value calculation model may be any model, such as a logistic regression model, a multiple regression model, a multilayer perceptron, a neural network such as a CNN or an RNN (Recurrent Neural Network), a support vector machine, a random forest modeled as a regression tree, or a hidden Markov model, etc. Also, the gingival condition index value calculation model may be a model that combines various models to make a comprehensive judgment.

[0068] Fig. 7 is a diagram illustrating an example of a case in which a gingival condition index value 20 is calculated using a gingival condition index value calculation model 50. In Fig. 7, the gingival condition index value calculation model 50 is configured to output the gingival condition index value 20 in response to an input of a region of interest 12.

[0069] The calculation unit 2040 inputs the region of interest 12 extracted from the three-dimensional intraoral data 10 to the gum condition index calculation model 50. As a result, the calculation unit 2040 obtains the gum condition index value 20 for the region of interest 12 from the gum condition index calculation model 50.

[0070] Here, when the calculation unit 2040 calculates a plurality of types of gingival condition index values ​​20, the calculation unit 2040 has a gingival condition index value calculation model 50 for each type of gingival condition index value 20. For example, assume that CAL and PPD are calculated as the gingival condition index value 20. In this case, the calculation unit 2040 has a gingival condition index value calculation model 50 trained to calculate CAL and a gingival condition index value calculation model 50 trained to calculate PPD.

[0071] The calculation unit 2040 may also have a gingival condition index value calculation model 50 for each tooth or each part of the tooth. Here, it is assumed that there are N teeth. When the gingival condition index value 20 is calculated for each tooth, the calculation unit 2040 has N gingival condition index value calculation models 50 for each type of index value calculated as the gingival condition index value 20. For example, when PPD and CAL are calculated as the gingival condition index value 20, the calculation unit 2040 has 2*N gingival condition index value calculation models 50.

[0072] Furthermore, when index values ​​are calculated for M sites for each tooth, the calculation unit 2040 has M*N gingival condition index value calculation models 50 for each type of index value calculated as the gingival condition index value 20. For example, when PPD and CAL are calculated as the gingival condition index value 20, the calculation unit 2040 has 2*M*N gingival condition index value calculation models 50.

[0073] In addition, taking into consideration the left-right symmetry with respect to the oral cavity midline as a reference, the number of gingival condition index value calculation models 50 may be half of the above-mentioned number. In this case, the same gingival condition index value calculation models 50 are used for two attention areas 12 that are located at left-right symmetrical positions with respect to the oral cavity midline as a reference.

[0074] In this way, by preparing a gingival condition index value calculation model 50 for each tooth or each part of the tooth, there is an advantage that the gingival condition index value 20 can be calculated with higher accuracy compared to the case where a gingival condition index value calculation model 50 common to all teeth and tooth parts is used.

[0075] The gingival condition index value 20 may not be calculated using the gingival condition index value calculation model 50. For example, the calculation unit 2040 may be configured to predict the values ​​of parameters necessary for calculating the gingival condition index value 20 using a machine learning model. Hereinafter, the model that predicts the values ​​of parameters necessary for calculating the gingival condition index value 20 is referred to as a first prediction model. The types of machine learning models that can be used as the first prediction model are the same as the types of machine learning models that can be used as the gingival condition index value calculation model 50.

[0076] The first prediction model is trained in advance to output a specific parameter value in response to, for example, input of the region of interest 12. The calculation unit 2040 inputs the region of interest 12 to the first prediction model, and calculates the gingival condition index value 20 using the parameter value obtained from the first prediction model.

[0077] The parameters required for calculating the gingival condition index value 20 vary depending on the type of index value calculated as the gingival condition index value 20. For example, assume that PPD is calculated as the gingival condition index value 20. PPD is the distance between the gingival margin and the pocket bottom. Therefore, in order to calculate PPD, it is necessary to identify the position of the gingival margin and the position of the pocket bottom.

[0078] Therefore, for example, the calculation unit 2040 has a first prediction model trained to predict the position of the gingival margin for the region of interest 12, and a first prediction model trained to predict the position of the pocket bottom for the region of interest 12. The calculation unit 2040 calculates PPD using these first prediction models.

[0079] 8 is a diagram illustrating an example of a case in which PPD is calculated using a first prediction model. A first prediction model 170-1 is a first prediction model that predicts the position of the gingival margin. On the other hand, a first prediction model 170-2 is a first prediction model that predicts the position of the pocket bottom.

[0080] The calculation unit 2040 obtains the position of the gingival margin by inputting the region of interest 12 to the first prediction model 170-1. The calculation unit 2040 also obtains the position of the pocket bottom by inputting the region of interest 12 to the first prediction model 170-2. The calculation unit 2040 then calculates the difference between the position of the gingival margin obtained from the first prediction model 170-1 and the position of the pocket bottom obtained from the first prediction model 170-2 to calculate PPD.

[0081] It is assumed that CAL is calculated as the gingival condition index value 20. CAL is the distance between the CEJ and the bottom of the pocket. Therefore, in order to calculate CAL, it is necessary to identify the position of the CEJ and the position of the bottom of the pocket.

[0082] Therefore, for example, the calculation unit 2040 has a first prediction model 170 trained to predict the position of the CEJ for the region of interest 12, and a first prediction model 170 trained to predict the position of the pocket bottom for the region of interest 12. The calculation unit 2040 uses these first prediction models 170 to calculate CAL.

[0083] 9 is a diagram illustrating a case in which CAL is calculated using the first prediction model. The first prediction model 170-3 is a first prediction model that predicts the position of the CEJ. On the other hand, the first prediction model 170-2 is a first prediction model that predicts the position of the pocket bottom.

[0084] The calculation unit 2040 obtains the position of the CEJ by inputting the region of interest 12 to the first prediction model 170-3. The calculation unit 2040 also obtains the position of the pocket bottom by inputting the region of interest 12 to the first prediction model 170-2. Then, the calculation unit 2040 calculates CAL by calculating the difference between the position of the CEJ obtained from the first prediction model 170-3 and the position of the pocket bottom obtained from the first prediction model 170-2.

[0085] Here, the bottom of the pocket is hidden by the gums (see FIG. 2). The calculation unit 2040 can predict the position of the bottom of the pocket hidden by the gums by using the first prediction model 170. Also, the CEJ may be hidden by the gums. The calculation unit 2040 can predict the position of the CEJ by using the first prediction model 170 even when the CEJ is hidden by the gums. Therefore, by using the first prediction model 170, an index value that requires the position of the part hidden by the gums can be easily calculated.

[0086] The calculation unit 2040 has, for example, a first prediction model 170 for each parameter used to calculate the gingival condition index value 20. For example, when PPD and CAL are calculated as the gingival condition index value 20, the calculation unit 2040 has a first prediction model 170 for predicting the position of the gingival margin, a first prediction model 170 for predicting the position of the pocket bottom, and a first prediction model 170 for predicting the position of the CEJ.

[0087] The first prediction model 170 may be prepared for each tooth or each part of the tooth. For example, when the number of teeth is N, the calculation unit 2040 has N first prediction models 170 for each parameter. For example, when PPD and CAL are calculated as the gingival condition index value 20, the calculation unit 2040 has N first prediction models 170 for predicting the position of the gingival margin, N first prediction models 170 for predicting the position of the pocket bottom, and N first prediction models 170 for predicting the position of the CEJ.

[0088] Furthermore, when index values ​​are calculated for M sites for each tooth, the calculation unit 2040 has M*N first prediction models 170 for each parameter. For example, when PPD and CAL are calculated as the gingival condition index value 20, the calculation unit 2040 has N*M first prediction models 170 for predicting the position of the gingival margin, the first prediction model 170 for predicting the position of the pocket bottom, and the first prediction model 170 for predicting the position of the CEJ.

[0089] In addition, taking into consideration the left-right symmetry with respect to the oral cavity midline as a reference, the number of gingival condition index value calculation models 50 may be half of the above-mentioned number. In this case, the same first prediction model 170 is used for two attention areas 12 that are located at left-right symmetrical positions with respect to the oral cavity midline as a reference.

[0090] In this way, by preparing a first prediction model 170 for each tooth or each tooth part, there is an advantage that the parameter values ​​can be predicted with higher accuracy compared to the case where a first prediction model 170 common to all teeth or tooth parts is used.

[0091] Data input to the gingival condition index calculation model 50 or the first prediction model 170 may not be the region of interest 12, but may be a feature that can be calculated from the region of interest 12. Hereinafter, the feature that is calculated from the region of interest 12 and is used to calculate the gingival condition index 20 is referred to as a first feature.

[0092] The first feature amount is, for example, a value representing a predetermined type of feature related to the tooth of interest, or a value representing a predetermined type of feature related to the periodontal region of interest. Features related to the tooth include, for example, the shape of the tooth, the size of the tooth, the color tone of the tooth (change in the density of the color of the tooth depending on the position), the smoothness of the tooth, the distance between the tooth and the adjacent tooth (size of the gap), and whether or not the CEJ is exposed in the tooth. Here, the feature of the size of the object is represented, for example, by the vertical length of the object, the horizontal length of the object, or the thickness of the object. Also, the feature of the shape of the object is represented, for example, by the ratio of the vertical length of the object to the horizontal length of the object, or the ratio of the width of the upper part of the object to the width of the lower part of the object.

[0093] The smoothness includes concepts such as the degree of smoothness, the degree of roughness, or the degree of unevenness. The severity of a tooth can be expressed, for example, by the severity ratio. The severity ratio is expressed, for example, by approximating the tooth surface with an elliptical surface, and is expressed as the area ratio of the region that deviates from the approximated surface within a standard value.

[0094] Characteristics related to the periodontal region include, for example, gingival color, gingival tone (change in gingival color intensity depending on position), gingival shape, gingival surface smoothness, distance between the gingival surface and the tooth surface (degree of protrusion of the gingiva in relation to the tooth), gingival surface area, gingival volume, distance between the gingival-alveolar junction and the gingival margin, gingival papilla shape, gingival papilla surface area, gingival papilla volume, or gingival papilla height.

[0095] Here, the value represented by the first feature may be an absolute value or a relative value to a reference value. The reference value may be obtained from the oral cavity three-dimensional data 10 or may be defined in advance. When the reference value is obtained from the oral cavity three-dimensional data 10, for example, an object representing the reference is imaged by an imaging device together with the subject's teeth.

[0096] The calculation of the first feature amount is performed, for example, by using a pre-trained machine learning model. Hereinafter, the model used to calculate the first feature amount is called a first feature amount calculation model. The types of models that can be used as the first feature amount calculation model are the same as the types of models that can be used as the gingival condition index value calculation model.

[0097] 10 is a diagram illustrating a case in which a gingival condition index value 20 is calculated using a first feature amount calculation model and a gingival condition index value calculation model. The first feature amount calculation model 40 is trained in advance to output a first feature amount 100 in response to an input of an attention area 12. The gingival condition index value calculation model 50 is trained in advance to output a gingival condition index value 20 in response to an input of the first feature amount 100.

[0098] The calculation unit 2040 inputs the region of interest 12 extracted from the three-dimensional intraoral data 10 to the first feature amount calculation model 40. Furthermore, the calculation unit 2040 inputs the first feature amount 100 output from the first feature amount calculation model 40 to the gingival condition index value calculation model 50. As a result, the calculation unit 2040 obtains the gingival condition index value 20 for the region of interest 12 from the gingival condition index value calculation model 50.

[0099] Here, when the calculation unit 2040 has multiple gingival condition index value calculation models 50, the first feature calculation model 40 may be shared by all gingival condition index value calculation models 50, or a first feature calculation model 40 may be prepared for each gingival condition index value calculation model 50.

[0100] 11 is a diagram illustrating a case in which PPD is calculated using the first feature calculation model and the first prediction model. The first prediction model 170-1 is trained in advance to output the position of the gingival margin in response to the input of the first feature 100. The first prediction model 170-2 is trained in advance to output the position of the pocket bottom in response to the input of the first feature 100.

[0101] The calculation unit 2040 inputs the attention area 12 extracted from the intraoral three-dimensional data 10 to the first feature amount calculation model 40. Furthermore, the calculation unit 2040 inputs the first feature amount 100 output from the first feature amount calculation model 40 to each of the first prediction model 170-1 and the first prediction model 170-2. Then, the calculation unit 2040 calculates the PPD by calculating the difference between the position of the gingival margin obtained from the first prediction model 170-1 and the position of the pocket bottom obtained from the first prediction model 170-2.

[0102] Here, when the calculation unit 2040 has multiple first prediction models 170, the first feature calculation model 40 may be shared by all the first prediction models 170, or a first feature calculation model 40 may be prepared for each first prediction model 170.

[0103] The first feature amount calculation model 40 does not have to be used to calculate the first feature amount 100. In this case, the calculation unit 2040 calculates the first feature amount 100 from the attention area 12 by analyzing the attention area 12 according to a predetermined algorithm.

[0104] 12 is a diagram illustrating a case in which a first feature calculated without using the first feature calculation model 40 is used by the gum condition index calculation model 50. The calculation unit 2040 calculates the first feature 100 by analyzing the attention area 12. After that, the calculation unit 2040 inputs the calculated first feature 100 to the gum condition index calculation model 50. As a result, the gum condition index 20 is obtained from the gum condition index calculation model 50.

[0105] 13 is a diagram illustrating a case in which a first feature calculated without using the first feature calculation model 40 is used by the first prediction model 170. In the example of FIG. 13, PPD is calculated as the gingival condition index value 20.

[0106] The calculation unit 2040 calculates the first feature amount 100 by analyzing the attention area 12. After that, the calculation unit 2040 inputs the calculated first feature amount 100 to each of the first prediction model 170-1 and the first prediction model 170-2. Then, the calculation unit 2040 calculates the difference between the position of the gingival margin obtained from the first prediction model 170-1 and the position of the pocket bottom obtained from the first prediction model 170-2, thereby calculating PPD.

[0107] When calculating the gingival condition index value 20 from the region of interest 12, a three-dimensional region other than the region of interest 12 may be further used. Specifically, the calculation unit 2040 may further use another three-dimensional region (hereinafter, a related region) related to the region of interest 12 among the three-dimensional regions included in the oral cavity three-dimensional data 10. The related region is, for example, a region close to the region of interest. In addition, for example, the related region is a region including a tooth located symmetrically to the tooth of interest with respect to the oral cavity midline.

[0108] The region adjacent to the region of interest is, for example, a region that includes a tooth adjacent to the target tooth (hereinafter, an adjacent tooth). The adjacent tooth is, for example, a tooth adjacent to the tooth of interest. The tooth adjacent to the target tooth is a tooth located on the mesial side or distal side of the tooth of interest. The adjacent tooth may be a tooth located a predetermined number of teeth (for example, two teeth) away from the tooth of interest.

[0109] The region of interest may include only a portion of the tooth of interest, not the entire tooth of interest. In this case, the region of interest may be a region that includes a portion of the tooth of interest that is not included in the region of interest. For example, the region of interest is a region that is located at the target position and the target region with respect to the tooth center line of the target tooth.

[0110] Fig. 14 is a diagram showing a case where a part of the tooth of interest is included in both the region of interest and the related region. In Fig. 14, the region of interest 12 includes the tooth of interest 13 and its periodontal region. The related region 14 is an area located symmetrically to the region of interest 12 with respect to the tooth center line 160, and includes the tooth of interest 13 and its periodontal region.

[0111] The size of the related region 14 may be the same as the size of the attention region 12, or may be larger or smaller than the size of the attention region 12. When the size of the related region 14 is different from the size of the attention region 12, for example, the size of the related region 14 is set to a predetermined multiple (e.g., twice) the size of the attention region 12.

[0112] Here, the length of the associated region 14 may be set to a predetermined multiple of the size of the region of interest 12 for all axial directions, or only for some axial directions. In the latter case, for example, the size of the associated region 14 is set to a predetermined multiple of the size of the region of interest 12 for only the lateral length when the tooth is viewed from the front (the horizontal length in FIG. 14). Then, for the remaining axial lengths, the size of the associated region 14 and the size of the region of interest 12 are set to be the same.

[0113] The calculation unit 2040 may use two or more associated regions 14. For example, the calculation unit 2040 uses, as the associated regions 14, the regions of the adjacent teeth that are within a distance of two teeth from the tooth of interest.

[0114] Assume that the region of interest does not include the entire tooth of interest, but only a part of the tooth of interest. In this case, for example, a region of interest 14 adjacent to the region of interest from the mesial side and a region of interest 14 adjacent to the region of interest from the distal side may be used as the region of interest 14. FIG. 15 is a diagram illustrating a region of interest 12 including only a part of the tooth of interest and two regions of interest 14. In the example of FIG. 15, two regions of interest 14, that is, a region of interest 14-1 and a region of interest 14-2, are used for one region of interest 12. In FIG. 15, the size of the region of interest 14 is set to be twice the size of the region of interest 12.

[0115] The associated region 14 is used by the gingival condition index calculation model 50 and the first prediction model 170. It is assumed that the gingival condition index 20 is calculated using the gingival condition index calculation model 50. In this case, the gingival condition index calculation model 50 is trained in advance to output the gingival condition index 20 in response to input of both the attention region 12 and the associated region 14.

[0116] 16 is a diagram illustrating an example of a case in which the gingival condition index value 20 is calculated by the gingival condition index value calculation model 50 to which the attention region 12 and the associated region 14 are input. The calculation unit 2040 inputs both the attention region 12 and the associated region 14 to the gingival condition index value calculation model 50. In response to the input of the attention region 12 and the associated region 14, the gingival condition index value calculation model 50 outputs the gingival condition index value 20.

[0117] Assume that the gingival condition index value 20 is calculated using a first prediction model 170. In this case, the first prediction model 170 is pre-trained to output a prediction value of a specific parameter in response to input of both the region of interest 12 and the region of interest 14.

[0118] 17 is a diagram illustrating a case in which PPD is calculated using a first prediction model 170 to which the region of interest 12 and the related region 14 are input. The first prediction model 170-1 is trained to output the position of the gingival margin in response to the input of the region of interest 12 and the related region 14. The first prediction model 170-2 is trained to output the position of the pocket bottom in response to the input of the region of interest 12 and the related region 14.

[0119] The calculation unit 2040 inputs both the attention region 12 and the associated region 14 to the first prediction model 170-1 and the first prediction model 170-2. The first prediction model 170-1 outputs a predicted value of the position of the gingival margin in response to the input of the gingival condition index value 20 and the associated region 14. The first prediction model 170-2 outputs a predicted value of the position of the pocket bottom in response to the input of the gingival condition index value 20 and the associated region 14. The calculation unit 2040 calculates the PPD by calculating the difference between the position of the gingival margin obtained from the first prediction model 170-1 and the position of the pocket bottom obtained from the first prediction model 170-2.

[0120] Instead of the related region 14, a feature amount calculated from the related region 14 may be input to the gum condition index value calculation model 50 and the first prediction model 170. The feature amount calculated from the related region 14 is called a second feature amount.

[0121] The second feature amount can be any of various types of data that can be used as the first feature amount described above. In addition, the second feature amount can be calculated from the related region 14 using a method similar to the method for calculating the first feature amount 100 from the attention region 12.

[0122] Here, it is assumed that a machine learning model is used in the process of calculating the second feature amount from the related region 14. Hereinafter, the model that calculates the second feature amount from the related region 14 is referred to as the second feature amount calculation model. In this case, the first feature amount calculation model 40 may be used as the second feature amount calculation model. That is, in this case, data obtained by inputting the related region 14 to the first feature amount calculation model 40 is treated as the second feature amount.

[0123] Using the region of interest 12 as well as the region of interest 14 in the process of calculating the gingival condition index value 20 of the tooth of interest has the effect of allowing the gingival condition index value 20 of the tooth of interest to be calculated (predicted) with higher accuracy. The reason for this will be explained below.

[0124] Periodontitis is more likely to develop locally than uniformly throughout the oral cavity. Therefore, even if the tooth or periodontal region of interest has periodontitis, other teeth and periodontal regions may not have periodontitis. Thus, by comparing the tooth or periodontal region of interest with other teeth and periodontal regions, it may be possible to compare teeth and their periodontal regions with those with periodontitis and those without periodontitis. Thus, the gingival condition index value 20 for the tooth of interest can be calculated with higher accuracy.

[0125] Similarly, even if a part of the tooth of interest or a part of the periodontal region of interest has periodontitis, other parts of the tooth of interest or other parts of the periodontal region of interest may not have periodontitis. Therefore, by comparing the part of the tooth of interest or the part of the periodontal region of interest with the other parts of the tooth of interest or the other parts of the periodontal region of interest, it may be possible to compare the parts of the tooth of interest and its periodontal region that have periodontitis and the parts that do not have periodontitis. Therefore, the gingival condition index value 20 for the tooth of interest can be calculated with higher accuracy.

[0126] There are various methods for extracting a related region from the three-dimensional oral cavity data 10. For example, the calculation unit 2040 has a machine learning model trained to extract a related region from the three-dimensional oral cavity data 10. Hereinafter, the model for extracting a related region from the three-dimensional oral cavity data 10 is referred to as a related region extraction model. As the related region extraction model, any machine learning model (for example, a neural network such as CNN) capable of extracting a predetermined three-dimensional region from three-dimensional data can be used.

[0127] For example, the associated region extraction model receives the three-dimensional oral cavity data 10 and the position of the region of interest. The associated region extraction model is configured to output, from the three-dimensional oral cavity data 10, one or more associated regions corresponding to the region of interest at a specified position.

[0128] The extraction of the relevant region does not necessarily require the use of a model. For example, the calculation unit 2040 may extract the relevant region from the oral cavity three-dimensional data 10 by analyzing the oral cavity three-dimensional data 10 using a predetermined algorithm.

[0129] In addition to or instead of the relevant region, various attribute information regarding the subject may be used in calculating the gingival condition index value 20. The attribute information of the subject includes the subject's race, age, sex, medical history and pathology (e.g., diabetes, periodontitis, gingivitis, endodontic lesion, root fracture, cementum detachment, caries, subgingival caries, or occlusal trauma), treatment history, smoking history, or chief complaint.

[0130] By using information representing the attributes of the subject to calculate the gingival condition index value 20, the gingival condition index value 20 can be calculated (predicted) with higher accuracy, taking into account the attributes of the subject.

[0131] When the gingival condition index value calculation model 50 is used to calculate the gingival condition index value 20, the gingival condition index value calculation model 50 is configured to further input attribute information of the subject or a feature amount calculated from the attribute information of the subject. The gingival condition index value calculation model 50 further uses the attribute information of the subject or a feature amount calculated from the attribute information of the subject to calculate the gingival condition index value 20.

[0132] When the first prediction model 170 is used to calculate the gingival condition index value 20, the first prediction model 170 is configured to further input attribute information of the subject or a feature amount calculated from the attribute information of the subject. The first prediction model 170 further uses the attribute information of the subject or a feature amount calculated from the attribute information of the subject to calculate a predicted value of a specific parameter.

[0133] The calculation unit 2040 may calculate the gingival condition index value 20 using time series data of the attention region 12, the related region 14, or the attribute information of the subject. A machine learning model capable of handling time series data, such as an RNN, may be used to calculate the gingival condition index value 20 using the time series data. By using the time series data to calculate the gingival condition index value 20, the gingival condition index value 20 can be calculated (predicted) with higher accuracy, taking into account changes over time in the teeth, periodontal region, attributes, etc. of the subject.

[0134] <<How to calculate alveolar bone resorption index value 30>> For example, the calculation unit 2040 extracts the region of interest 12 from the three-dimensional oral cavity data 10, and calculates the alveolar bone resorption index value 30 using the region of interest 12. FIG. 18 is a flowchart illustrating a process flow for calculating the alveolar bone resorption index value 30. The calculation unit 2040 extracts one or more regions of interest 12 from the three-dimensional oral cavity data 10 (S302). Steps S304 to S308 constitute a loop process L2 that is executed for each region of interest 12. In S304, the calculation unit 2040 determines whether the loop process L2 has been executed for all regions of interest 12. If the loop process L2 has already been executed for all regions of interest 12, the process of FIG. 18 ends.

[0135] If there are any attention regions 12 that have not yet been the target of the loop process L2, the calculation unit 2040 selects one of the attention regions 12 that have not yet been the target of the loop process L2. The attention region 12 selected here is denoted as attention region i.

[0136] The calculation unit 2040 calculates the alveolar bone resorption index value 30 for the region of interest i (S306). Since S308 is the end of the loop process L2, S304 is executed again.

[0137] The process (S306) of calculating the alveolar bone resorption index value 30 from the region of interest 12 is performed, for example, by using a trained machine learning model. Hereinafter, the model for calculating the alveolar bone resorption index value 30 is referred to as an alveolar bone resorption index value calculation model.

[0138] The alveolar bone resorption index value calculation model may employ, for example, various models such as those described above as examples of the gingival condition index value calculation model. In addition, the alveolar bone resorption index value calculation model may employ a model that combines various models to perform a comprehensive judgment.

[0139] Fig. 19 is a diagram illustrating a case in which an alveolar bone resorption index value 30 is calculated using an alveolar bone resorption index value calculation model 70. In Fig. 19, an alveolar bone resorption index value calculation model 70 is trained in advance to output an alveolar bone resorption index value 30 in response to an input of an attention area 12.

[0140] The calculation unit 2040 inputs the region of interest 12 extracted from the oral cavity three-dimensional data 10 to the alveolar bone resorption index value calculation model 70. As a result, the calculation unit 2040 obtains the alveolar bone resorption index value 30 for the region of interest 12 from the alveolar bone resorption index value calculation model 70.

[0141] Here, when the calculation unit 2040 calculates a plurality of types of alveolar bone resorption index values ​​30, the calculation unit 2040 may have an alveolar bone resorption index value calculation model 70 for each type of alveolar bone resorption index value 30. For example, it is assumed that an alveolar bone resorption degree and an alveolar bone resorption rate are calculated as the alveolar bone resorption index value 30. In this case, the calculation unit 2040 has an alveolar bone resorption index value calculation model 70 trained to calculate the alveolar bone resorption degree and an alveolar bone resorption index value calculation model 70 trained to calculate the alveolar bone resorption rate.

[0142] However, when dealing with two index values ​​that can be calculated from one another, such as the alveolar bone resorption degree and the alveolar bone resorption rate, the model may be used to calculate only one of the index values. For example, the calculation unit 2040 may calculate the alveolar bone resorption degree using the alveolar bone resorption index value calculation model 70, and calculate the alveolar bone resorption rate from the alveolar bone resorption degree.

[0143] The calculation unit 2040 may have an alveolar bone resorption index value calculation model 70 for each tooth or each tooth part. Here, it is assumed that there are N teeth. When the alveolar bone resorption index value 30 is calculated for each tooth, the calculation unit 2040 has N alveolar bone resorption index value calculation models 70 for each type of index value calculated as the alveolar bone resorption index value 30.

[0144] Furthermore, when index values ​​are calculated for M sites for each tooth, the calculation unit 2040 has M*N alveolar bone resorption index value calculation models 70 for each type of index value calculated as the alveolar bone resorption index value 30.

[0145] In addition, taking into consideration the left-right symmetry with respect to the oral cavity midline as a reference, the number of alveolar bone resorption index value calculation models 70 may be half of the above-mentioned number. In this case, the same alveolar bone resorption index value calculation models 70 are used for two attention areas 12 that are located at left-right symmetrical positions with respect to the oral cavity midline as a reference.

[0146] In this way, by preparing an alveolar bone resorption index value calculation model 70 for each tooth or each tooth part, there is an advantage that the alveolar bone resorption index value 30 can be calculated with higher accuracy compared to the case where a common alveolar bone resorption index value calculation model 70 is used for all teeth or tooth parts.

[0147] The alveolar bone resorption index value calculation model 70 may not be used to calculate the alveolar bone resorption index value 30. For example, the calculation unit 2040 may be configured to predict the values ​​of parameters necessary for calculating the alveolar bone resorption index value 30 using a machine learning model. Hereinafter, the model predicting the values ​​of parameters necessary for calculating the alveolar bone resorption index value 30 is referred to as a second prediction model. The types of machine learning models that can be used as the second prediction model are the same as the types of machine learning models that can be used as the first prediction model.

[0148] The second prediction model is trained in advance to output a specific parameter value in response to input of the region of interest 12. The calculation unit 2040 inputs the region of interest 12 to the second prediction model, and calculates the alveolar bone resorption index value 30 using the parameter value obtained from the second prediction model.

[0149] The parameters required for calculating the alveolar bone resorption index value 30 vary depending on the type of index value calculated as the alveolar bone resorption index value 30. For example, assume that the degree of alveolar bone resorption is calculated as the alveolar bone resorption index value 30. The degree of alveolar bone resorption is calculated as B / A, based on the distance A from the CEJ to the root apex and the distance B from the CEJ to the alveolar crest. Therefore, in order to calculate the degree of alveolar bone resorption, it is necessary to identify the positions of the CEJ, the root apex, and the alveolar crest.

[0150] Therefore, for example, the calculation unit 2040 has a second prediction model trained to predict the position of the CEJ for the region of interest 12, a second prediction model trained to predict the position of the apex for the region of interest 12, and a second prediction model trained to predict the alveolar bone crest for the region of interest 12. The calculation unit 2040 uses these second prediction models to calculate the degree of alveolar bone resorption.

[0151] FIG. 20 is a diagram illustrating a case in which the degree of alveolar bone resorption is calculated using the second prediction model. The second prediction model 180-1 is a second prediction model trained to predict the position of the CEJ in response to input of the region of interest 12. The second prediction model 180-2 is a second prediction model trained to predict the position of the apex in response to input of the region of interest 12. The second prediction model 180-3 is a second prediction model trained to predict the alveolar bone crest in response to input of the region of interest 12.

[0152] The calculation unit 2040 obtains the position of the CEJ by inputting the region of interest 12 to the second prediction model 180-1. The calculation unit 2040 also obtains the position of the apex by inputting the region of interest 12 to the second prediction model 180-2. The calculation unit 2040 further obtains the position of the alveolar crest by inputting the region of interest 12 to the second prediction model 180-3. The calculation unit 2040 then calculates the degree of alveolar bone resorption using the position of the CEJ obtained from the second prediction model 180-1, the position of the apex obtained from the second prediction model 180-2, and the position of the alveolar crest obtained from the second prediction model 180-3.

[0153] The alveolar bone resorption rate can be calculated by multiplying the alveolar bone resorption degree by 100. Therefore, even when calculating the alveolar bone resorption rate as an alveolar bone resorption index value of 30, the second prediction models 180-1 to 180-3 shown in FIG. 20 can be used.

[0154] Here, both the root apex and the alveolar bone crest are hidden by the gums (see FIG. 2). The calculation unit 2040 can predict the positions of the root apex and the alveolar bone crest hidden by the gums by using the second prediction model 180. Furthermore, the CEJ may be hidden by the gums. The calculation unit 2040 can predict the position of the CEJ by using the second prediction model 180 even when the CEJ is hidden by the gums. Therefore, by using the second prediction model 180, an index value that requires the position of the part hidden by the gums can be easily calculated.

[0155] The calculation unit 2040 has, for example, a second prediction model 180 for each parameter used to calculate the alveolar bone resorption index value 30. For example, when the alveolar bone resorption degree and the alveolar bone resorption rate are calculated as the alveolar bone resorption index value 30, the calculation unit 2040 has a second prediction model 180 for predicting the position of the CEJ, a second prediction model 180 for predicting the position of the tooth apex, and a second prediction model 180 for predicting the position of the alveolar bone crest.

[0156] Here, there are parameters, such as the position of the CEJ, that can be used for both the calculation of the gingival condition index value 20 and the calculation of the alveolar bone resorption index value 30. For such parameters, the calculation unit 2040 may use the values ​​output from the first prediction model 170 for both the calculation of the gingival condition index value 20 and the calculation of the alveolar bone resorption index value 30.

[0157] For example, the calculation unit 2040 calculates CAL as the gingival condition index value 20, and calculates the degree of alveolar bone resorption as the alveolar bone resorption index value 30. In this case, the calculation unit 2040 inputs the region of interest 12 to a first prediction model 170 that predicts the position of the CEJ, and uses the value output from this first prediction model 170 to both calculate CAL and the degree of alveolar bone resorption. Therefore, in this case, the calculation unit 2040 does not need to include a second prediction model 180 that predicts the position of the CEJ.

[0158] The second prediction model 180 may be prepared for each tooth or each tooth part. For example, when the number of teeth is N, the calculation unit 2040 has N second prediction models 180 for each parameter. When index values ​​are calculated for M parts for each tooth, the calculation unit 2040 has M*N second prediction models 180 for each parameter.

[0159] In addition, taking into consideration the left-right symmetry with respect to the oral cavity midline as a reference, the number of alveolar bone resorption index value calculation models 70 may be half of the above-mentioned number. In this case, the same second prediction model 180 is used for two attention areas 12 that are located at left-right symmetrical positions with respect to the oral cavity midline as a reference.

[0160] In this way, by preparing a second prediction model 180 for each tooth or each tooth part, there is an advantage that the parameter values ​​can be predicted with higher accuracy compared to the case where a second prediction model 180 common to all teeth or tooth parts is used.

[0161] Data input to the alveolar bone resorption index value calculation model 70 and the second prediction model 180 may not be the region of interest 12, but may be a feature that can be calculated from the region of interest 12. Hereinafter, the feature that is calculated from the region of interest 12 and is used to calculate the alveolar bone resorption index value 30 is called a third feature.

[0162] The third feature amount is, for example, a value representing a predetermined type of feature regarding the tooth of interest, or a value representing a predetermined type of feature regarding the periodontal region of interest, similar to the first feature amount. However, the type of feature used to calculate the gingival condition index value 20 and the type of feature used to calculate the alveolar bone resorption index value 30 may be different from each other.

[0163] The method for calculating the third feature amount from the attention area 12 can be the same as the method for calculating the first feature amount from the attention area 12. For example, the calculation of the third feature amount is performed using a pre-trained machine learning model. Hereinafter, the model used for calculating the third feature amount is referred to as a third feature amount calculation model. The types of models that can be used as the third feature amount calculation model are the same as the types of models that can be used as the first feature amount calculation model.

[0164] Fig. 21 is a diagram illustrating a case in which an alveolar bone resorption index value 30 is calculated using the third feature amount calculation model and an alveolar bone resorption index value calculation model 70. In Fig. 21, the third feature amount calculation model 60 is trained in advance to output the third feature amount 120 in response to input of the attention area 12. Moreover, the alveolar bone resorption index value calculation model 70 is trained in advance to output the alveolar bone resorption index value 30 in response to input of the third feature amount 120.

[0165] The calculation unit 2040 inputs the region of interest 12 extracted from the three-dimensional oral cavity data 10 to the third feature amount calculation model 60. Furthermore, the calculation unit 2040 inputs the third feature amount 120 output from the third feature amount calculation model 60 to the alveolar bone resorption index value calculation model 70. As a result, the calculation unit 2040 obtains the alveolar bone resorption index value 30 for the tooth of interest included in the region of interest 12 from the alveolar bone resorption index value calculation model 70.

[0166] Here, when the calculation unit 2040 has multiple alveolar bone resorption index value calculation models 70, the third feature calculation model 60 may be shared by all the alveolar bone resorption index value calculation models 70, or a third feature calculation model 60 may be prepared for each alveolar bone resorption index value calculation model 70.

[0167] FIG. 22 is a diagram illustrating a case in which the degree of alveolar bone resorption is calculated using the third feature calculation model 60 and the second prediction model 180. The second prediction model 180-1 is trained in advance to output the position of the CEJ in response to the input of the third feature 120. The second prediction model 180-2 is trained in advance to output the position of the tooth apex in response to the input of the third feature 120. The second prediction model 180-3 is trained in advance to output the position of the alveolar bone crest in response to the input of the third feature 120.

[0168] The calculation unit 2040 inputs the region of interest 12 extracted from the three-dimensional intraoral data 10 to the third feature amount calculation model 60. Furthermore, the calculation unit 2040 inputs the third feature amount 120 output from the third feature amount calculation model 60 to each of the second prediction models 180-1, 180-2, and 180-3. Then, the calculation unit 2040 calculates the degree of alveolar bone resorption by calculating the difference between the position of the CEJ obtained from the second prediction model 180-1, the position of the root apex obtained from the second prediction model 180-2, and the position of the alveolar bone crest obtained from the second prediction model 180-3.

[0169] Here, when the calculation unit 2040 has multiple second prediction models 180, the third feature calculation model 60 may be shared by all the second prediction models 180, or a third feature calculation model 60 may be prepared for each second prediction model 180.

[0170] The first feature amount 100 may be used as the third feature amount 120. In this case, the same feature amount is used in the process (S206) for calculating the gingival condition index value 20 and the process (S306) for calculating the alveolar bone resorption index value 30. In this case, the calculation unit 2040 does not need to have the third feature amount calculation model 60.

[0171] The calculation of the third feature amount 120 may be performed without using a machine learning model. In this case, the calculation unit 2040 calculates, as the third feature amount 120, a value representing a predetermined type of feature related to the tooth of interest or a value representing a predetermined type of feature related to the periodontal region of the tooth of interest by analyzing the region of interest 12 using a predetermined algorithm.

[0172] 23 is a diagram illustrating a case in which a third feature calculated without using the third feature calculation model 60 is used by the alveolar bone resorption index value calculation model 70. The calculation unit 2040 calculates the third feature 120 by analyzing the attention area 12. Thereafter, the calculation unit 2040 inputs the calculated third feature 120 to the alveolar bone resorption index value calculation model 70. As a result, the alveolar bone resorption index value 30 is obtained from the alveolar bone resorption index value calculation model 70.

[0173] 24 is a diagram illustrating a case in which the third feature amount calculated without using the third feature amount calculation model 60 is used by the second prediction model 180. In the example of FIG. 24, the alveolar bone resorption degree is calculated as the alveolar bone resorption index value 30.

[0174] The calculation unit 2040 calculates the third feature amount 120 by analyzing the attention area 12. Thereafter, the calculation unit 2040 inputs the calculated third feature amount 120 to each of the second prediction model 180-1, the second prediction model 180-2, and the second prediction model 180-3. Then, the calculation unit 2040 calculates the degree of alveolar bone resorption based on the position of the CEJ obtained from the second prediction model 180-1, the position of the root apex obtained from the second prediction model 180-2, and the position of the alveolar bone crest obtained from the second prediction model 180-3.

[0175] When calculating the alveolar bone resorption index value 30 from the attention region 12, the above-mentioned related region 14 may be further used. The alveolar bone resorption index value 30 is calculated using an alveolar bone resorption index value calculation model 70. In this case, the alveolar bone resorption index value calculation model 70 is trained in advance to output the alveolar bone resorption index value 30 in response to input of both the attention region 12 and the related region 14.

[0176] 25 is a diagram illustrating a case in which an alveolar bone resorption index value 30 is calculated by an alveolar bone resorption index value calculation model 70 to which an attention region 12 and an associated region 14 are input. The calculation unit 2040 inputs both the attention region 12 and the associated region 14 to the alveolar bone resorption index value calculation model 70. In response to the input of the attention region 12 and the associated region 14, the alveolar bone resorption index value calculation model 70 outputs the alveolar bone resorption index value 30.

[0177] Assume that the alveolar bone resorption index value 30 is calculated using the second prediction model 180. In this case, the second prediction model 180 is trained in advance to output a predicted value of a specific parameter in response to input of both the region of interest 12 and the related region 14.

[0178] FIG. 26 is a diagram illustrating a case in which the degree of alveolar bone resorption is calculated using the second prediction model 180 to which the region of interest 12 and the related region 14 are input. The second prediction model 180-1 is trained in advance to output the position of the CEJ in response to the input of the region of interest 12 and the related region 14. The second prediction model 180-2 is trained in advance to output the position of the apex in response to the input of the region of interest 12 and the related region 14. The second prediction model 180-3 is trained in advance to output the position of the alveolar bone crest in response to the input of the region of interest 12 and the related region 14.

[0179] The calculation unit 2040 inputs both the region of interest 12 and the associated region 14 to each of the second prediction model 180-1, the second prediction model 180-2, and the second prediction model 180-3. The second prediction model 180-1 outputs a predicted value of the position of the CEJ in response to the input of the alveolar bone resorption index value 30 and the associated region 14. The second prediction model 180-2 outputs a predicted value of the position of the apex in response to the input of the alveolar bone resorption index value 30 and the associated region 14. The second prediction model 180-3 outputs a predicted value of the position of the alveolar bone crest in response to the input of the alveolar bone resorption index value 30 and the associated region 14. The calculation unit 2040 calculates the degree of alveolar bone resorption using the position of the CEJ obtained from the second prediction model 180-1, the position of the apex obtained from the second prediction model 180-2, and the position of the alveolar bone crest obtained from the second prediction model 180-3.

[0180] Instead of the related region 14, a feature calculated from the related region 14 may be input to the alveolar bone resorption index value calculation model 70 and the second prediction model 180. The feature calculated from the related region 14 and used to calculate the alveolar bone resorption index value 30 is called a fourth feature.

[0181] The fourth feature amount can be any of various types of data that can be used as the first feature amount described above. In addition, the fourth feature amount can be calculated from the related region 14 using the same method as the method for calculating the first feature amount 100 from the attention region 12.

[0182] Here, it is assumed that a machine learning model is used in the process of calculating the fourth feature amount from the related region 14. Hereinafter, the model that calculates the fourth feature amount from the related region 14 is referred to as the fourth feature amount calculation model. In this case, the third feature amount calculation model 60 may be used as the fourth feature amount calculation model. That is, in this case, data obtained by inputting the related region 14 to the third feature amount calculation model 60 is treated as the fourth feature amount.

[0183] Using the related region 14 in addition to the region of interest 12 in the process of calculating the alveolar bone resorption index value 30 of the tooth of interest has the effect of allowing the alveolar bone resorption index value 30 of the tooth of interest to be calculated (predicted) with higher accuracy. The reason for this is the same as the reason why using the related region 14 in addition to the region of interest 12 in the process of calculating the gingival condition index value 20 of the tooth of interest can be calculated with higher accuracy.

[0184] In addition to or instead of the relevant region, various attribute information on the subject may be used in calculating the alveolar bone resorption index value 30. The types of subject attribute information are as described above. By using information representing the subject's attributes in calculating the alveolar bone resorption index value 30, the alveolar bone resorption index value 30 can be calculated (predicted) with higher accuracy, taking into account the subject's attributes.

[0185] When the alveolar bone resorption index value calculation model 70 is used to calculate the alveolar bone resorption index value 30, the alveolar bone resorption index value calculation model 70 is configured to further input attribute information of the subject or a feature amount calculated from the attribute information of the subject. The alveolar bone resorption index value calculation model 70 further uses the attribute information of the subject or a feature amount calculated from the attribute information of the subject to calculate the alveolar bone resorption index value 30.

[0186] When the second prediction model 180 is used to calculate the alveolar bone resorption index value 30, the second prediction model 180 is configured to further input attribute information of the subject or a feature amount calculated from the attribute information of the subject. The second prediction model 180 further uses the attribute information of the subject or a feature amount calculated from the attribute information of the subject to calculate a predicted value of a specific parameter.

[0187] The calculation unit 2040 may calculate the alveolar bone resorption index value 30 using time-series data of the attention region 12, the related region 14, or the subject's attribute information. A machine learning model capable of handling time-series data, such as an RNN, may be used to calculate the alveolar bone resorption index value 30 using the time-series data. By using the time-series data to calculate the alveolar bone resorption index value 30, the alveolar bone resorption index value 30 can be calculated (predicted) with higher accuracy, taking into account changes over time in the subject's teeth, periodontal region, attributes, and the like.

[0188] <Result output> The index value calculation device 2000 can output, in any manner, the gingival condition index value 20, the alveolar bone resorption index value 30, or both of them calculated by the calculation unit 2040. Hereinafter, information output by the index value calculation device 2000 will be referred to as output information.

[0189] The output information indicates the index value calculated by the index value calculation device 2000. Here, when an index value is calculated for each of the teeth included in the oral cavity three-dimensional data 10, the output information preferably indicates the calculated index value for the tooth together with information that can identify the tooth. For example, the index value calculation device 2000 assigns an identification number to each of the teeth included in the oral cavity three-dimensional data 10 according to a predetermined rule. The index value calculation device 2000 generates output information that indicates the tooth identification number and the calculated index value for that tooth in association with each other.

[0190] Here, when multiple attention areas are extracted for one tooth, multiple index values ​​are calculated for one tooth. Therefore, in this case, the index value calculation device 2000 assigns an identification number to each part for which an index value is calculated.

[0191] The predetermined rule for assigning identification numbers to teeth or tooth parts can be any rule, for example, any identification number used by those skilled in the art.

[0192] The output information may be output in any manner. For example, the index value calculation device 2000 stores the output information in any storage unit. Alternatively, for example, the index value calculation device 2000 displays the output information on a display device. Alternatively, for example, the index value calculation device 2000 transmits the output information to another device (for example, a user terminal).

[0193] <How to use the index values> The gingival condition index value 20 and the alveolar bone resorption index value 30 can be used in any manner. For example, the gingival condition index value 20 and the alveolar bone resorption index value 30 can be used to determine the presence or absence of a periodontal-related disease (e.g., gingivitis), to determine the state of a periodontal-related disease, or to determine whether or not a visit to a dentist is recommended. These various determinations may be performed manually or by a computer. In the latter case, the determination process may be performed by the index value calculation device 2000 or by a device other than the index value calculation device 2000.

[0194] Fig. 27 is a block diagram illustrating a functional configuration of an index value calculation device 2000 that performs a discrimination process using an index value. In Fig. 27, the index value calculation device 2000 further includes a discrimination unit 2060. The discrimination unit 2060 performs a discrimination process using the gingival condition index value 20, the alveolar bone resorption index value 30, or both of them calculated by the calculation unit 2040. The discrimination process is, for example, any one or more of the above-mentioned processes of discriminating the presence or absence of a periodontal-related disease, discriminating the state of a periodontal-related disease, and discriminating whether or not to recommend visiting a dentist.

[0195] The discrimination process by the discrimination unit 2060 using the gingival condition index value 20 and the alveolar bone resorption index value 30 can be implemented in accordance with, for example, guidelines that are widely recognized and used by those skilled in the art, such as the 2022 Guidelines for Periodontal Treatment by the Japanese Society of Periodontology, the 2018 New Classification of Periodontitis by the American Academy of Periodontology / European Federation of Periodontology, the Community Periodontal Disease Index Classification by the World Health Organization (WHO), or the Periodontal Disease Check-up Manual by the Ministry of Health, Labor and Welfare.

[0196] The index value calculation device 2000 may further have another function. For example, the index value calculation device 2000 may have a function of calculating a gingival index (GI) from the oral cavity three-dimensional data 10, and a function of predicting a plaque control state. The index value calculation device 2000 may also have a function of diagnosing or predicting a dental caries state, a dental prosthetic state, and an implant tooth state from the oral cavity three-dimensional data 10. Furthermore, the index value calculation device 2000 may further have a function of introducing a dentist or making a dentist appointment, etc., in response to the determination by the discrimination unit 2060 that a visit to a dentist is recommended.

[0197] <About model training> As described above, one or more models may be used in the index value calculation device 2000. Each model is trained in advance before being used by the index value calculation device 2000.

[0198] A device used to train a model is called a training device. The training device may be the index value calculation device 2000 or a device other than the index value calculation device 2000.

[0199] The training device generates a model with initial values ​​set for the trainable parameters (in other words, generates and initializes the model). The training device then trains the model by repeatedly updating the trainable parameters of the model using multiple training data. For example, if the model is a neural network, the trainable parameters are the bias and the weights assigned to each edge.

[0200] Below, we explain in more detail how each model is trained.

[0201] <<Training the model used to calculate the gingival condition index value 20>> The gum condition index value calculation model 50 or the first prediction model 170 may be used to calculate the gum condition index value 20. When the gum condition index value calculation model 50 is used, the training device trains the gum condition index value calculation model 50.

[0202] FIG. 28 is a diagram illustrating training of the gingival condition index calculation model 50. The training data 140 includes a region of interest 142 and a gingival condition index value 144. The region of interest 142 is a three-dimensional region including a tooth of interest and the periodontal region of the tooth of interest, similar to the region of interest 12. The gingival condition index value 144 is a ground truth gingival condition index value to be calculated from the region of interest 142 by the gingival condition index calculation model 50. The gingival condition index value 144 is calculated, for example, by performing actual measurement using a probe on the tooth of interest and its periodontal region included in the region of interest 142, or by performing actual measurement using a probe and an X-ray image.

[0203] The training device obtains the gingival condition index value 20 by inputting the attention area 142 into the gingival condition index value calculation model 50. The training device calculates a loss based on the gingival condition index value 20 and the gingival condition index value 144, and updates each parameter of the gingival condition index value calculation model 50 based on the calculated loss.

[0204] The training device repeatedly updates the gingival condition index value calculation model 50 using a plurality of training data 140. In this way, the training device obtains a trained gingival condition index value calculation model 50. The trained gingival condition index value calculation model 50 is then used by the index value calculation device 2000.

[0205] As described above, the index value calculation device 2000 may have a gingival condition index value calculation model 50 for each tooth or each tooth part. In this case, the training device performs training for each gingival condition index value calculation model 50 prepared for each tooth or each tooth part.

[0206] When the first prediction model 170 is used to calculate the gingival condition index value 20, the training device trains the first prediction model 170. FIG. 29 is a diagram illustrating training of the first prediction model 170. The training data 140 includes an attention region 142 and a parameter prediction value 145. The parameter prediction value 145 indicates a predicted value of the ground truth to be output by the first prediction model 170. For example, it is assumed that the first prediction model 170 is a model that predicts the position of a pocket bottom. In this case, the parameter prediction value 145 indicates an actual measurement value of the position of the pocket bottom in the attention region 142.

[0207] The training device obtains a parameter prediction value 172 by inputting the region of interest 142 into the first prediction model 170. The training device calculates a loss based on the parameter prediction value 172 and the parameter prediction value 145, and updates each parameter of the first prediction model 170 based on the calculated loss.

[0208] The training device repeatedly updates the first prediction model 170 by using a plurality of training data 140. In this way, the training device obtains a trained first prediction model 170. The trained first prediction model 170 is then used by the index value calculation device 2000.

[0209] As described above, a plurality of parameters, such as the position of the gingival margin, the position of the pocket bottom, and the position of the CEJ, may be used to calculate the gingival condition index value 20. The training device trains a first prediction model 170 for each parameter, which is used to predict the value of the parameter. For example, the first prediction model 170 for predicting the position of the gingival margin is trained using training data 140 in which the actual measured value of the position of the gingival margin is indicated as the parameter predicted value 145. The first prediction model 170 for predicting the position of the pocket bottom is trained using training data 140 in which the actual measured value of the position of the pocket bottom is indicated as the parameter predicted value 145.

[0210] Furthermore, the index value calculation device 2000 may have a first prediction model 170 for the same parameters for each tooth or each tooth part. In this case, the training device trains each of the first prediction models 170 prepared for each tooth or each tooth part.

[0211] The first feature amount calculation model 40 may be used to calculate the gingival condition index value 20. In this case, the training device also trains the first feature amount calculation model 40. The training of the first feature amount calculation model 40 may be performed together with the training of the gingival condition index value calculation model 50 and the first prediction model 170, or may be performed independently of the training of the gingival condition index value calculation model 50 and the first prediction model 170.

[0212] 30 is a diagram illustrating training of the first feature amount calculation model 40 and the gum condition index value calculation model 50. The training device obtains the first feature amount 100 by inputting the attention area 142 to the first feature amount calculation model 40. Furthermore, the training device obtains the gum condition index value 20 by inputting the first feature amount 100 output from the first feature amount calculation model 40 to the gum condition index value calculation model 50. The training device calculates a loss based on the gum condition index value 20 and the gum condition index value 144, and updates each parameter of the first feature amount calculation model 40 and the gum condition index value calculation model 50 based on the calculated loss.

[0213] The training device repeatedly updates the first feature amount calculation model 40 and the gingival condition index value calculation model 50 by using a plurality of training data 140. In this way, the training device obtains a trained first feature amount calculation model 40 and a trained gingival condition index value calculation model 50. Then, the trained first feature amount calculation model 40 and the trained gingival condition index value calculation model 50 are used by the index value calculation device 2000.

[0214] The method for training both the first feature amount calculation model 40 and the first prediction model 170 is the same as the method for training both the first feature amount calculation model 40 and the gingival condition index value calculation model 50. That is, the training device obtains the first feature amount 100 by inputting the attention area 142 to the first feature amount calculation model 40. Furthermore, the training device obtains the parameter predicted value 172 by inputting the first feature amount 100 output from the first feature amount calculation model 40 to the first prediction model 170. The training device calculates a loss based on the parameter predicted value 172 and the parameter predicted value 145, and updates each parameter of the first feature amount calculation model 40 and the first prediction model 170 based on the calculated loss.

[0215] When the training of the first feature amount calculation model 40 is performed independently, training data including a region of interest and a first feature amount of the ground truth is used. The training device calculates a loss based on the first feature amount 100 obtained by inputting the region of interest shown in the training data to the first feature amount calculation model 40 and the first feature amount of the ground truth shown in the training data. Then, the training device updates each parameter of the first feature amount calculation model 40 based on the calculated loss.

[0216] As described above, the first feature amount calculation model 40 may not be used to calculate the first feature amount 100. In this case, the first feature amount 100 calculated from the attention area 142 using a predetermined algorithm is used to train the gingival condition index value calculation model 50 and the first prediction model 170.

[0217] 31 is a diagram illustrating training of the gingival condition index value calculation model 50, to which a first feature amount 100 calculated by a predetermined algorithm is input. The training device calculates the first feature amount 100 by analyzing a region of interest 142 by a predetermined algorithm. The training device inputs the calculated first feature amount 100 to the gingival condition index value calculation model 50 to obtain the gingival condition index value 20. The training device calculates a loss based on the gingival condition index value 20 and the gingival condition index value 144, and updates each parameter of the gingival condition index value calculation model 50 based on the calculated loss.

[0218] Training of the first prediction model 170 can be performed in a similar manner. That is, the training device calculates the first feature amount 100 by analyzing the attention area 142 with a predetermined algorithm, and inputs the first feature amount 100 to the first prediction model 170. The training device calculates a loss based on the parameter prediction value 172 and the parameter prediction value 145 output from the first prediction model 170, and updates each parameter of the first prediction model 170 based on the calculated loss.

[0219] When the region of interest 14 is used to calculate the gingival condition index value 20, training data 140 including the region of interest is used to train the gingival condition index value calculation model 50 and the first prediction model 170.

[0220] 32 is a diagram illustrating training of the gingival condition index calculation model 50 in which a region of interest is used. Training data 140 includes a region of interest 142, a region of interest 146, and a gingival condition index value 144. The region of interest 146 is a region of interest that corresponds to the region of interest 142.

[0221] The training device obtains the gingival condition index value 20 by inputting the attention area 142 and the related area 146 into the gingival condition index value calculation model 50. The training device calculates a loss based on the gingival condition index value 20 and the gingival condition index value 144, and updates each parameter of the gingival condition index value calculation model 50 based on the calculated loss.

[0222] 33 is a diagram illustrating training of a first prediction model 170 in which an associated region is used. Training data 140 includes an attention region 142, an associated region 146, and a parameter prediction value 145. The training device obtains the parameter prediction value 172 by inputting the attention region 142 and the associated region 146 to the first prediction model 170. The training device calculates a loss based on the parameter prediction value 172 and the parameter prediction value 145, and updates each parameter of the first prediction model 170 based on the calculated loss.

[0223] A second feature calculated from the associated region 14 may be input to the gum condition index value calculation model 50 or the first prediction model 170. When the second feature is calculated using the second feature calculation model, the training device further trains the second feature calculation model.

[0224] The training of the second feature amount calculation model can be performed in the same manner as the training of the first feature amount calculation model 40. For example, in training of the gingival condition index value calculation model 50, instead of inputting the associated region 146 to the gingival condition index value calculation model 50, the training device inputs the second feature amount calculated from the associated region 146 by the second feature amount calculation model to the gingival condition index value calculation model 50. The training device calculates a loss based on the gingival condition index value 20 and the gingival condition index value 144 output from the gingival condition index value calculation model 50, and updates each parameter of the gingival condition index value calculation model 50 and the second feature amount calculation model based on the loss.

[0225] Similarly, for example, in training the first prediction model 170, instead of inputting the related region 146 to the first prediction model 170, the training device inputs a second feature calculated from the related region 146 by the second feature calculation model to the first prediction model 170. The training device calculates a loss based on the parameter prediction value 172 and the parameter prediction value 145 output from the first prediction model 170, and updates each parameter of the first prediction model 170 and the second feature calculation model based on the loss.

[0226] When the second feature calculation model is independently trained, the second feature calculation model is trained using training data including the related region and the second feature of the ground truth. The training device calculates a loss based on the second feature obtained by inputting the related region shown in the training data to the second feature calculation model and the second feature of the ground truth shown in the training data. Then, the training device updates the parameters of the second feature calculation model based on the calculated loss.

[0227] The calculation of the second feature amount does not necessarily require the use of the second feature amount calculation model. In this case, the second feature amount calculated from the related region 146 using a predetermined algorithm is used to train the gingival condition index value calculation model 50 and the first prediction model 170.

[0228] Attribute information of the subject may be used to calculate the gingival condition index value 20. When attribute information is used to calculate the gingival condition index value 20, the training device trains each model using training data 140 that includes the attribute information.

[0229] <<Training the model used to calculate the alveolar bone resorption index value 30>> The alveolar bone resorption index value calculation model 70 or the second prediction model 180 is used to calculate the alveolar bone resorption index value 30. When the alveolar bone resorption index value calculation model 70 is used, the training device trains the alveolar bone resorption index value calculation model 70.

[0230] 34 is a diagram illustrating training of the alveolar bone resorption index value calculation model 70. Training data 150 includes a region of interest 152 and an alveolar bone resorption index value 154. The region of interest 152 is a three-dimensional region including a tooth of interest and a periodontal region of the tooth of interest, similar to the region of interest 12. The alveolar bone resorption index value 154 is a ground truth alveolar bone resorption index value to be calculated from the region of interest 152 by the alveolar bone resorption index value calculation model 70.

[0231] The alveolar bone resorption index value 154 is calculated, for example, using an X-ray image of the tooth of interest and its periodontal region included in the region of interest 152. The alveolar bone resorption index value 154 may be a one-dimensional index calculated from one-dimensional data such as distance, a two-dimensional index calculated from two-dimensional data such as the area of ​​resorbed alveolar bone, or a three-dimensional index calculated from three-dimensional data such as the volume of resorbed alveolar bone.

[0232] The training device obtains the alveolar bone resorption index value 30 by inputting the attention area 152 into the alveolar bone resorption index value calculation model 70. The training device calculates a loss based on the alveolar bone resorption index value 30 and the alveolar bone resorption index value 154, and updates each parameter of the alveolar bone resorption index value calculation model 70 based on the calculated loss.

[0233] The training device repeatedly updates the alveolar bone resorption index value calculation model 70 by using a plurality of training data 150. In this way, the training device obtains a trained alveolar bone resorption index value calculation model 70. Then, the trained alveolar bone resorption index value calculation model 70 is used by the index value calculation device 2000.

[0234] As described above, the index value calculation device 2000 may include an alveolar bone resorption index value calculation model 70 for each tooth or each tooth part. In this case, the training device performs training for each alveolar bone resorption index value calculation model 70 prepared for each tooth or each tooth part.

[0235] When the second prediction model 180 is used to calculate the alveolar bone resorption index value 30, the training device trains the second prediction model 180. FIG. 35 is a diagram illustrating training of the second prediction model 180. The training data 150 includes an attention region 152 and a parameter prediction value 155. The parameter prediction value 155 indicates a predicted value of the ground truth to be output by the second prediction model 180. For example, it is assumed that the second prediction model 180 is a model that predicts the position of the tooth apex. In this case, the parameter prediction value 155 indicates an actual measurement value of the position of the tooth apex in the attention region 152.

[0236] The training device obtains a parameter prediction value 182 by inputting the region of interest 152 into the second prediction model 180. The training device calculates a loss based on the parameter prediction value 182 and the parameter prediction value 155, and updates each parameter of the second prediction model 180 based on the calculated loss.

[0237] The training device repeatedly updates the second prediction model 180 by using a plurality of training data 150. In this way, the training device obtains a trained second prediction model 180. The trained second prediction model 180 is then used by the index value calculation device 2000.

[0238] As described above, a plurality of parameters, such as the position of the CEJ, the position of the apex, and the position of the alveolar bone crest, may be used to calculate the alveolar bone resorption index value 30. The training device trains a second prediction model 180 for predicting the value of each parameter. For example, the second prediction model 180 for predicting the position of the CEJ is trained using training data 150 in which the actual measured value of the position of the CEJ is indicated as the parameter predicted value 155. The second prediction model 180 for predicting the position of the apex is trained using training data 150 in which the actual measured value of the position of the apex is indicated as the parameter predicted value 155.

[0239] Furthermore, the index value calculation device 2000 may have a second prediction model 180 for the same parameters for each tooth or each tooth part. In this case, the training device trains each second prediction model 180 prepared for each tooth or each tooth part.

[0240] The third feature amount calculation model 60 may be used to calculate the alveolar bone resorption index value 30. In this case, the training device also trains the third feature amount calculation model 60. The training of the third feature amount calculation model 60 may be performed together with the training of the alveolar bone resorption index value calculation model 70 and the second prediction model 180, or may be performed independently of the training of the alveolar bone resorption index value calculation model 70 and the second prediction model 180.

[0241] 36 is a diagram illustrating training of the third feature amount calculation model 60 and the alveolar bone resorption index value calculation model 70. The training device obtains the third feature amount 120 by inputting the attention area 152 to the third feature amount calculation model 60. Furthermore, the training device obtains the alveolar bone resorption index value 30 by inputting the third feature amount 120 output from the third feature amount calculation model 60 to the alveolar bone resorption index value calculation model 70. The training device calculates a loss based on the alveolar bone resorption index value 30 and the alveolar bone resorption index value 154, and updates each parameter of the third feature amount calculation model 60 and the alveolar bone resorption index value calculation model 70 based on the calculated loss.

[0242] The training device repeatedly updates the third feature amount calculation model 60 and the alveolar bone resorption index value calculation model 70 by using a plurality of training data 150. In this way, the training device obtains a trained third feature amount calculation model 60 and a trained alveolar bone resorption index value calculation model 70. Then, the trained third feature amount calculation model 60 and the trained alveolar bone resorption index value calculation model 70 are used by the index value calculation device 2000.

[0243] The method for training the third feature amount calculation model 60 and the second prediction model 180 together is the same as the method for training the third feature amount calculation model 60 and the alveolar bone resorption index value calculation model 70 together. That is, the training device obtains the third feature amount 120 by inputting the attention area 152 to the third feature amount calculation model 60. Furthermore, the training device obtains the parameter predicted value 182 by inputting the third feature amount 120 output from the third feature amount calculation model 60 to the second prediction model 180. The training device calculates a loss based on the parameter predicted value 182 and the parameter predicted value 155, and updates each parameter of the third feature amount calculation model 60 and the second prediction model 180 based on the calculated loss.

[0244] When the training of the third feature amount calculation model 60 is performed independently, training data including the region of interest and the third feature amount of the ground truth is used. The training device calculates a loss based on the third feature amount 120 obtained by inputting the region of interest 12 shown in the training data to the third feature amount calculation model 60 and the third feature amount of the ground truth shown in the training data. Then, the training device updates each parameter of the third feature amount calculation model 60 based on the calculated loss.

[0245] As described above, the third feature amount calculation model 60 may not be used to calculate the third feature amount 120. In this case, the third feature amount 120 calculated from the attention area 152 using a predetermined algorithm is used to train the alveolar bone resorption index value calculation model 70 and the second prediction model 180.

[0246] 37 is a diagram illustrating training of the alveolar bone resorption index value calculation model 70, to which a third feature amount 120 calculated by a predetermined algorithm is input. The training device calculates the third feature amount 120 by analyzing the attention area 152 by a predetermined algorithm. The training device inputs the calculated third feature amount 120 to the alveolar bone resorption index value calculation model 70 to obtain the alveolar bone resorption index value 30. The training device calculates a loss based on the alveolar bone resorption index value 30 and the alveolar bone resorption index value 154, and updates each parameter of the alveolar bone resorption index value calculation model 70 based on the calculated loss.

[0247] Training of the second prediction model 180 can be performed in a similar manner. That is, the training device calculates the third feature amount 120 by analyzing the attention area 152 with a predetermined algorithm, and inputs the third feature amount 120 to the second prediction model 180. The training device calculates a loss based on the parameter prediction value 182 and the parameter prediction value 155 output from the second prediction model 180, and updates the parameters of the second prediction model 180 based on the calculated loss.

[0248] When the relevant region 14 is used to calculate the alveolar bone resorption index value 30, training data 150 including the relevant region is used to train the alveolar bone resorption index value calculation model 70 and the second prediction model 180.

[0249] 38 is a diagram illustrating training of the alveolar bone resorption index value calculation model 70 in which a related region is used. Training data 150 has a region of interest 152, a related region 156, and an alveolar bone resorption index value 154. The related region 156 is a related region that corresponds to the region of interest 152.

[0250] The training device obtains the alveolar bone resorption index value 30 by inputting the attention area 152 and the related area 156 into the alveolar bone resorption index value calculation model 70. The training device calculates a loss based on the alveolar bone resorption index value 30 and the alveolar bone resorption index value 154, and updates each parameter of the alveolar bone resorption index value calculation model 70 based on the calculated loss.

[0251] 39 is a diagram illustrating training of a second prediction model 180 in which a related region is used. Training data 150 includes an attention region 152, an attention region 156, and a parameter prediction value 155. The training device obtains the parameter prediction value 182 by inputting the attention region 152 and the attention region 156 to the second prediction model 180. The training device calculates a loss based on the parameter prediction value 182 and the parameter prediction value 155, and updates each parameter of the second prediction model 180 based on the calculated loss.

[0252] A fourth feature calculated from the associated region 14 may be input to the alveolar bone resorption index value calculation model 70 or the second prediction model 180. When the fourth feature is calculated using the fourth feature calculation model, the training device further trains the fourth feature calculation model.

[0253] The training of the fourth feature amount calculation model can be performed in the same manner as the training of the third feature amount calculation model 60. For example, in training of the alveolar bone resorption index value calculation model 70, instead of inputting the related region 156 to the alveolar bone resorption index value calculation model 70, the training device inputs the fourth feature amount calculated from the related region 156 by the fourth feature amount calculation model to the alveolar bone resorption index value calculation model 70. The training device calculates a loss based on the alveolar bone resorption index value 30 and the alveolar bone resorption index value 154 output from the alveolar bone resorption index value calculation model 70, and updates each parameter of the alveolar bone resorption index value calculation model 70 and the fourth feature amount calculation model based on the loss.

[0254] Similarly, for example, in training the second prediction model 180, instead of inputting the related region 156 to the second prediction model 180, the training device inputs a fourth feature calculated from the related region 156 by the fourth feature calculation model to the second prediction model 180. The training device calculates a loss based on the parameter prediction value 182 and the parameter prediction value 155 output from the second prediction model 180, and updates each parameter of the second prediction model 180 and the fourth feature calculation model based on the loss.

[0255] When the fourth feature calculation model is independently trained, the fourth feature calculation model is trained using training data including the related region and the fourth feature of the ground truth. The training device calculates a loss based on the fourth feature obtained by inputting the related region shown in the training data to the fourth feature calculation model and the fourth feature of the ground truth shown in the training data. Then, the training device updates each parameter of the fourth feature calculation model based on the calculated loss.

[0256] The calculation of the fourth feature amount does not necessarily require the use of the fourth feature amount calculation model. In this case, the alveolar bone resorption index value calculation model 70 and the second prediction model 180 are trained using the fourth feature amount calculated from the related region 156 using a predetermined algorithm.

[0257] Attribute information of a subject may be used in calculating the alveolar bone resorption index value 30. When attribute information is used in calculating the alveolar bone resorption index value 30, the training device trains each model using training data 150 including the attribute information.

[0258] <<Verification of model accuracy>> The training device may verify the accuracy of the model using verification data having a similar structure to the training data. For example, verification data having a similar structure to the training data 140 is used to verify a model used to calculate the gingival condition index value 20.

[0259] The training device calculates the difference between the gingival condition index value 20 calculated using the verification data and the ground truth gingival condition index value shown in the verification data, and judges whether the accuracy of the model is sufficiently high based on the difference. For example, the training device judges that the prediction using the model is correct when the ground truth gingival condition index value is included in a predetermined numerical range (e.g., a range of ±10% of the gingival condition index value 20) centered on the gingival condition index value 20 calculated using the model. On the other hand, the training device judges that the prediction using the model is incorrect when the ground truth gingival condition index value is not included in the above-mentioned predetermined numerical range.

[0260] The training device, for example, performs a judgment for each of a plurality of validation data, and judges that the accuracy of the model is sufficiently high when the rate at which the prediction using the model is judged to be correct is equal to or higher than a threshold value. On the other hand, when the rate at which the prediction is judged to be correct is less than the threshold value, the training device judges that the accuracy of the model is not sufficiently high. When it is judged that the accuracy of the model is not sufficiently high, the training device further trains the model used to calculate the gingival condition index value 20, for example, using additional training data 140. In addition, for example, the training device may increase or decrease the types of feature amounts calculated from the attention region and the related region. In addition, for example, the training device may change the type of model.

[0261] Here, verification of the accuracy of the model for calculating the gingival condition index value 20 has been described, but verification of the accuracy of the model for calculating the alveolar bone resorption index value 30 can also be carried out in a similar manner.

[0262] Although the present invention has been described above with reference to the embodiment, the present invention is not limited to the above embodiment. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.

[0263] In the above example, the program includes a set of instructions (or software code) for making the computer perform one or more functions described in the embodiment when the program is loaded into the computer. The program may be stored in a non-transitory computer-readable medium or a tangible storage medium. By way of example and not limitation, the computer-readable medium or tangible storage medium includes random-access memory (RAM), read-only memory (ROM), flash memory, solid-state drive (SSD) or other memory technology, CD-ROM, digital versatile disc (DVD), Blu-ray (registered trademark) disk or other optical disk storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage device. The program may be transmitted on a transitory computer-readable medium or communication medium. By way of example and not limitation, the transitory computer-readable medium or communication medium includes electrical, optical, acoustic, or other forms of propagating signals. [Explanation of symbols]

[0264] 10 3D oral cavity data 12 Areas of Interest 13 Notable Teeth 14 Related Areas 20 Gingival condition index 30 Alveolar bone resorption index value 40 First feature calculation model 50 Gingival condition index calculation model 60 Third feature calculation model 70 Alveolar bone resorption index calculation model 100 First feature 120 Third feature 140 Training Data 142 Areas of Interest 144 Gingival condition index 145 Parameter Estimates 146 Related Areas 150 training data 152 Areas of Interest 154 Alveolar bone resorption index value 155 Parameter Estimates 156 Related Areas 160 Tooth centerline 170 First Prediction Model 172 Parameter Estimates 180 Second Prediction Model 182 Parameter Estimates 1000 Computers 1020 Bus 1040 Processor 1060 Memory 1080 Storage Device 1100 Input / Output Interface 1120 Network Interface 2000 Index value calculation device 2020 Acquisition Department 2040 Calculation Department 2060 Discrimination section

Claims

1. An acquisition step of acquiring three-dimensional data of a subject's oral cavity including the teeth and the surrounding areas of the teeth; and a calculation step of calculating, using the three-dimensional oral cavity data, a condition index value which is an index value related to the condition of the subject's teeth, such as a gingival condition index value which is an index value related to the gingival condition of the subject, an alveolar bone resorption index value which is an index value related to the resorption of the alveolar bone of the subject, or both of these.

2. In the calculation step, calculating predicted values ​​of one or more parameters required for calculating the condition index value using a three-dimensional region of interest that is included in the oral cavity three-dimensional data and that includes the whole or a part of the tooth of interest that is the subject of calculation of the condition index value and the periodontal region of the tooth of interest; The program according to claim 1 , further comprising: calculating the state index value based on the calculated predicted value.

3. a prediction model that is trained to output a predicted value of a parameter used to calculate the state index value in response to input of the region of interest; In the calculation step, Calculating predicted values ​​of the parameters by inputting the region of interest into the prediction model; The program according to claim 2 , further comprising: calculating the state index value using a predicted value of the calculated parameter.

4. a prediction model trained to output a predicted value of the parameter in response to input of a related region, which is another three-dimensional region related to the region of interest, included in the oral cavity three-dimensional data, and the region of interest; In the calculation step, extracting the related region corresponding to the region of interest from the three-dimensional oral cavity data; Calculating a predicted value of the parameter by inputting the region of interest and the extracted related region into the prediction model; The program according to claim 2 , further comprising: calculating the condition index value using a predicted value of the calculated parameter.

5. a prediction model trained to output a predicted value of a parameter used to calculate the index value in response to input of a feature amount related to the tooth of interest; In the calculation step, Calculating the feature amount from the region of interest; Calculating a predicted value of the parameter by inputting the calculated feature amount into the prediction model; The program according to claim 2 , further comprising: calculating the state index value using a predicted value of the calculated parameter.

6. an index value calculation model that is trained to output the condition index value in response to input of a three-dimensional region included in the oral cavity three-dimensional data, the three-dimensional region being a three-dimensional region including a tooth of interest that is a calculation target of the condition index value and a whole or a part of the periodontal region of the tooth of interest; The program according to claim 1 , wherein in said calculation step, said condition index value is calculated by inputting said attention area into said index value calculation model.

7. an index value calculation model trained to output the condition index value in response to input of a three-dimensional region included in the oral cavity three-dimensional data, the three-dimensional region being a three-dimensional region including a tooth of interest that is a calculation target of the condition index value and a whole or a part of the periodontal region of the tooth of interest, and an associated region included in the oral cavity three-dimensional data, the other three-dimensional region being associated with the area of ​​interest; In the calculation step, extracting the related region corresponding to the region of interest from the three-dimensional oral cavity data; The program according to claim 1 , further comprising: inputting the region of interest and the extracted associated region into the index value calculation model to calculate the state index value.

8. an index value calculation model trained to output the condition index value in response to input of a feature value related to a target tooth for calculating the condition index value; In the calculation step, Calculating the feature amount from a three-dimensional region of interest included in the oral cavity three-dimensional data, the three-dimensional region being a three-dimensional region including the tooth of interest and the whole or a part of the periodontal region of the tooth of interest; The program according to claim 1 , further comprising: calculating the state index value by inputting the calculated feature amount into the index value calculation model.

9. The program according to claim 4 or 7, wherein the related area is a three-dimensional area adjacent to the area of ​​interest, a three-dimensional area located symmetrically to the area of ​​interest with respect to the tooth center line of the tooth of interest, or a three-dimensional area located at an opposite position to the area of ​​interest with respect to the oral midline.

10. 9. The program according to claim 5 or 8, wherein the feature amount related to the tooth of interest represents a shape of the tooth of interest, a size of the tooth of interest, a color tone of the tooth of interest, a smoothness of the tooth of interest, a distance between the tooth of interest and an adjacent tooth, or presence or absence of exposure of the cement-enamel junction at the tooth of interest, a color of the gingiva around the tooth of interest, a change in the intensity of the color of the gingiva depending on the position, a shape of the gingiva, a surface smoothness of the gingiva, a distance between the surface of the gingiva and the surface of the tooth of interest, a surface area of ​​the gingiva, a volume of the gingiva, a distance between the gingival-alveolar junction of the gingiva and the gingival margin of the gingiva, a shape of the gingival papilla of the gingiva, a surface area of ​​the gingival papilla of the gingiva, a volume of the gingival papilla of the gingiva, or a height of the gingival papilla of the gingiva.

11. A program according to any one of claims 1 to 8, comprising a discrimination means for using the gingival condition index value, the alveolar bone resorption index value, or both, to determine the presence or absence of a periodontal-related disease, to determine the state of a periodontal-related disease, or to determine whether or not a visit to a dentist is recommended.

12. An acquisition unit that acquires three-dimensional data of a subject's oral cavity including the teeth and the surrounding areas of the teeth; and a calculation unit that uses the three-dimensional oral cavity data to calculate a condition index value that is an index value related to the condition of the subject's teeth, such as a gingival condition index value that is an index value related to the gingival condition of the subject, an alveolar bone resorption index value that is an index value related to the absorption of the alveolar bone of the subject, or both of these.

Images