Program, computer and information processing method

Abstract

A program, a computer 3, and an information processing method are provided that can efficiently and accurately assist dentists in determining whether or not orthodontic treatment is appropriate. [Solution] The system includes a receiving means 32, an estimation means 38, and a determination means 40. The receiving means 32 receives a plurality of images of a subject's dentition. The estimation means uses a trained model 200 generated by machine learning using training data 210 including images of the dentition and information on whether the images include dentition that is suitable for orthodontic treatment, to estimate whether each of the plurality of dentition images includes dentition that is suitable for orthodontic treatment. The determination means determines whether orthodontic treatment is suitable for the subject based on the estimation result by the estimation means.

Description

[Technical Field]

[0001] The present invention relates to a program, a computer, and an information processing method. [Background technology]

[0002] In recent years, a technique for orthodontic treatment using a mouthpiece (aligner) attached to the teeth by a user in daily life has become known. For example, Patent Document 1 discloses a technique for remotely performing orthodontic treatment using a mouthpiece to reduce the burden on the user. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 7594333 Summary of the Invention [Problem to be solved by the invention]

[0004] Traditionally, the decision on whether orthodontic treatment is appropriate and the selection of an orthodontic plan have been made manually based on the dentist's experience. This not only makes the assessment process time-consuming and laborious, but also leads to variations in the accuracy of orthodontic treatment assessments between dentists. In particular, there are limitations to the number of teeth that can be corrected and the ability to handle complex cases, so efficient and accurate diagnosis is required.

[0005] In consideration of these points, the present disclosure aims to provide a program, a computer, and an information processing method that can efficiently and accurately assist dentists in determining whether or not orthodontic treatment is appropriate. [Means for solving the problem]

[0006] The program of the present disclosure is A program that causes a computer to function as a receiving means, an estimating means, and a determining means, The receiving means receives a plurality of images of the row of teeth of the subject, the estimation means estimates whether or not each of the plurality of dental row images includes a dental row that can be orthodontally aligned, using a trained model generated by machine learning using training data including an image of a dental row and information on whether the image includes a dental row that can be orthodontally aligned, and The determining means determines whether orthodontic treatment is appropriate for the subject based on the estimation result by the estimating means.

[0007] In the program of the present disclosure, The determining means can determine whether or not a predetermined mouthpiece orthodontic plan is applicable based on the estimation result by the estimating means.

[0008] In the program of the present disclosure, The computer may further function as an image processing unit, The receiving means receives dentition scan data of the subject, The image processing means can process the row of teeth scan data received by the receiving means to generate a plurality of row of teeth images.

[0009] In the program of the present disclosure, The image processing means may process the dentition scan data received by the receiving means to generate a dentition image in which the upper and lower dentition of the subject are captured at an angle from diagonally below.

[0010] In the program of the present disclosure, The estimation means can estimate whether or not an integrated dentition image obtained by combining a plurality of the dentition images includes any dentition that can be corrected.

[0011] In the program of the present disclosure, The computer may further function as an arch drawing means, the arch drawing means uses an object detection model generated by machine learning using training data including an image of an upper dentition and an image of a lower dentition to detect and classify teeth in each of the images of the upper dentition and the lower dentition of the subject as part of the plurality of dentition images, and draws an arch representing the alignment of the teeth; The judgment means can judge whether orthodontic treatment is possible for the subject or whether a predetermined mouthpiece orthodontic plan can be applied based on the shape of the arch.

[0012] The computer of the present disclosure includes: A computer that functions as a receiving means, an estimating means, and a determining means by executing a program, The receiving means receives a plurality of images of the row of teeth of the subject, the estimation means estimates whether or not each of the plurality of dental row images includes a dental row that can be orthodontally aligned, using a trained model generated by machine learning using training data including an image of a dental row and information on whether the image includes a dental row that can be orthodontally aligned, and The determining means determines whether orthodontic treatment is appropriate for the subject based on the estimation result by the estimating means.

[0013] The information processing method of the present disclosure includes: An information processing method executed by a computer having a control unit, a step of the control unit receiving a plurality of images of the row of teeth of a subject; a step in which the control unit estimates whether or not each of the plurality of dental row images includes a dental row that can be orthodontally aligned, using a trained model generated by machine learning using training data including an image of a dental row and information on whether the image includes a dental row that can be orthodontally aligned; a step of the control unit determining whether or not orthodontic treatment is possible for the subject based on the estimation result; The present invention is characterized by the following features. [Effects of the Invention]

[0014] The program, computer, and information processing method disclosed herein can efficiently and accurately assist dentists in determining whether orthodontic treatment is appropriate. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a diagram schematically illustrating an information processing system according to an embodiment of the present disclosure. [Figure 2] 1A and 1B are diagrams illustrating examples of row-of-teeth images that are estimation targets in an information processing system and an information processing method according to an embodiment of the present disclosure. [Figure 3] 10A and 10B are diagrams showing examples of images showing an arch indicating the alignment of teeth in an information processing system and an information processing method according to an embodiment of the present disclosure. [Figure 4] FIG. 1 is a diagram illustrating an exemplary information processing flow of an information processing method according to an embodiment of the present disclosure. [Figure 5] FIG. 10 is a diagram showing another exemplary flow of information processing of the information processing method according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0016] 1 to 5 are diagrams illustrating an information processing system 1 and an information processing method according to the present disclosure.

[0017] [Information Processing System (AI Orthodontic Treatment Support System) 1] The information processing system 1 according to this embodiment shown in Fig. 1 includes a 3D scanner 2 and a server 3. The 3D scanner 2 and the server 3 can transmit and receive information wirelessly or via a wire. For example, information may be transmitted and received via a communication network such as the Internet.

[0018] (3D Scanner 2) The 3D scanner 2 has an imaging unit 21 and a communication unit 22 (Fig. 1). The imaging unit 21 acquires scan data of the subject's dentition by scanning the inside of the oral cavity, including the subject's upper and lower dentition. From this dentition scan data, an image is generated that will be used to estimate whether orthodontic treatment is feasible, as described below. The imaging unit 21 (3D scanner 2) is not particularly limited as long as it can acquire such dentition scan data, and known dental 3D scanners, 3D optical cameras, etc. can be used. Note that the "subject" refers to a person for whom the feasibility of orthodontic treatment is to be determined.

[0019] The communication unit 22 includes a communication interface for transmitting and receiving information to and from an external device wirelessly or via a wire. The dentition scan data captured by the imaging unit 21 is transmitted to the server 3 by the communication unit 22.

[0020] <Server (computer) 3> The server 3 can be configured from an industrial computer, a personal computer, a tablet terminal, etc., or it can be a virtual server within a physical server accessible via the Internet. There can be multiple pieces of hardware or virtual servers, or any combination of these. The server 3 shown in FIG. 1 includes a control unit 30, a storage unit 48, a communication unit 50, a display unit 52, and an operation unit 54.

[0021] (control unit 30) The control unit 30 is configured with a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an AI inference device, etc., and controls the operation of the server 3. Specifically, the control unit 30 executes programs stored in a storage unit 48 (described later) to function as a reception means 32, an image processing means 34, an arch drawing means 36, an estimation means 38, a determination means 40, an output means 42, an object detection model generation means 44, a trained model generation means 46, etc. Note that these functions may be achieved by executing one or more independent programs or applications. Furthermore, these programs and applications may be provided in a single terminal (the server 3) or may be distributed across multiple terminals (including the server 3), and in the latter case, may be connected to each other via a wired cable or a communication network. Each means will be described later.

[0022] (Storage unit 48) The storage unit 48 is configured by, for example, an HDD (Hard Disk Drive), RAM (Random Access Memory), ROM (Read Only Memory), SSD (Solid State Drive), etc. Furthermore, the storage unit 48 is not limited to being built into the server 3, but may be a storage medium (for example, a USB memory) that can be detachably attached to the server 3. Furthermore, instead of providing the storage unit 48, various information may be stored in other storage means (such as a cloud server).

[0023] The information stored in the memory unit 48 includes the program executed by the control unit 30, criteria for determining whether orthodontic treatment is possible and whether orthodontic plans are appropriate, as well as subject information such as the subject's dentition scan data, dentition images, and judgment results, the trained model 200 and its training data 210 described below, and the object detection model 100 and its training data 110.

[0024] (Communication unit 50) The communication unit 50 includes a communication interface for transmitting and receiving information to and from an external device wirelessly or via a wire. The communication unit 50 allows the server 3 to transmit and receive information to and from the 3D scanner 2 (communication unit 22).

[0025] (Display section 52, operation section 54) The display unit 52 is, for example, a monitor or the like, and displays various screens by receiving a display command signal from the control unit 30. The operation unit 54 is, for example, a keyboard or the like, and can give (input) various commands to the control unit 30. Note that in this embodiment, a display operation unit such as a touch panel in which the display unit 52 and the operation unit 54 are integrated may be used.

[0026] (Details of the control unit 30) (Reception means 32) The receiving means 32 receives a plurality of row-of-teeth images of the subject. Each row-of-teeth image received by the receiving means 32 in this manner becomes a target for estimation of whether or not orthodontic treatment is possible by the estimation means 38, which will be described later. The receiving means 32 may also receive row-of-teeth scan data of the subject from the 3D scanner 2. The row-of-teeth scan data received by the receiving means 32 is processed by the image processing means 34, which will be described later, to become a row-of-teeth image to be estimated by the estimation means 38. That is, in the information processing system 1 according to the present disclosure, it is assumed that the row-of-teeth image received by the receiving means 32 and used as a target for estimation of whether orthodontic treatment is possible is the same as a substitute for the row-of-teeth image generated by the image processing means 34. For this reason, these row-of-teeth images will be described collectively in relation to the image processing means 34, which will be described below.

[0027] The receiving means 32 may further receive clinical data such as X-ray data of the subject's dentition. This data may also be used in the estimation process by the estimation means .

[0028] (Image processing means 34) The image processing means 34 processes (processes) the row of teeth scan data received by the receiving means 32 to generate multiple row of teeth images. FIG. 2 shows an example of row of teeth images to be estimated, but the number of row of teeth images to be estimated that can be generated by the image processing means 34 may not be limited. Six row of teeth images shown as (a) to (f) in FIG. 2 may be used. Furthermore, the more rows of teeth images there are, the more accurately it is possible to determine whether orthodontic treatment is possible, which is preferable. The number of row of teeth images can be set appropriately depending on the orthodontic plan to be determined by the determination means 40.

[0029] The image processing means 34 may process the dentition scan data received by the receiving means 32 to generate a dentition image (FIG. 2(e)) in which the subject's upper and lower dentition (specifically, the state in which the upper and lower dentition are bitten as shown in FIG. 2(b)) is captured at an angle from diagonally below. FIG. 2(e) "Downward Diagonal" is a dentition image showing the dentition of a subject viewed from the front (front) side, captured at an angle of 50° from diagonally below (the state in FIG. 2(a) is set to 0°) by processing the dentition scan data.

[0030] The angle at which the upper and lower teeth are imaged from a diagonal angle below is not limited to 50°, but may be within the range of 40° to 60°, for example, 55°. This range is preferable because it allows for more accurate estimation from a different perspective than the estimation results for other teeth images. This angle can be set appropriately depending on the orthodontic plan to be evaluated by the evaluation means 40. Furthermore, to improve the accuracy of evaluation, the image processing means 34 may generate teeth images captured from a diagonal angle below at multiple angles. In this way, malocclusions such as anterior crossbite, crossbite, deep bite, and protrusion can be estimated.

[0031] Figure 2 shows dental image (a) of the upper and lower teeth viewed from the right side of the subject, dental image (b) of the upper and lower teeth viewed from the front (front) of the subject, dental image (c) of the upper and lower teeth viewed from the left side of the subject, dental image (d) of the upper teeth viewed from below, dental image (e) of the upper and lower teeth viewed from diagonally below, dental image (f) of the lower teeth viewed from above, and an integrated dental image (g) combining dental images (a) to (f).

[0032] Furthermore, the angles (orientations) of the rows of teeth shown in Figure 2 are merely examples, and the rows of teeth included in the training data or the images of rows of teeth to be estimated may be identified at various angles, and the number of teeth in each row of teeth may be appropriately set and changed depending on the orthodontic plan. Furthermore, it is not necessary to estimate all parts of each tooth; only the crowns or roots may be estimated depending on the orthodontic plan. However, when determining the suitability of a mouthpiece orthodontic plan that covers all teeth, it is preferable to prepare an image of rows of teeth at an angle at which all teeth can be seen, as shown in Figure 2.

[0033] (Arch drawing means 36) The arch drawing means 36 uses an object detection model 100 generated by machine learning using training data 110 including images of upper and lower teeth to detect and classify teeth in each of the subject's upper and lower teeth images as part of multiple teeth images, and draws an arch showing the tooth alignment (see Figure 3).

[0034] Explaining this with reference to Figures 2 and 3, the object detection model 100 detects each tooth from images of the upper and lower dentition, as shown in Figures 2(d) and (f). Then, based on the position and shape in each image, each tooth is classified as a canine, molar, or the like. Next, the object detection model 100 determines the center of each tooth in the dentition image (the points in Figure 3), taking into account the characteristics of each tooth based on the classification results. The arch drawing means 36 (controller 30) can draw an arch by connecting these centers. Furthermore, the degree of distortion of the subject's upper and lower dentition can be determined from this arch.

[0035] Furthermore, the arch drawing means 36 may calculate information on whether there are excess or deficiency of teeth in the dentition image based on the detection and classification results by the object detection model 100. This information can also be used to determine whether or not orthodontic treatment is possible.

[0036] The object detection model 100 can be generated by the object detection model generation means 44 described later, but as long as it can draw an arch as described above, there are no particular limitations on the generation method or the training data 110. Various techniques such as deep learning and supervised learning can be used as "machine learning."

[0037] (Estimation means 38) The estimation means 38 estimates whether or not each of the multiple dentition images (for each dentition image) includes a dentition that can be orthodontally aligned, using a trained model 200 generated by machine learning using training data 210 including images of dentition and information on whether the images include dentition that can be orthodontally aligned. An image such as that shown in Figure 2 can be used as the "dentition image."

[0038] The "information on whether the image contains teeth that can be orthodontally treated" is preferably correct data that has been labeled by a dentist as indicating that orthodontic treatment is possible or that a mouthpiece orthodontic plan can be applied.

[0039] "Machine learning" is not particularly limited as long as it can accept a dentition image as "input" and "output" whether or not the dentition image contains dentition that can be corrected, and various techniques such as deep learning, supervised learning, and anomaly detection can be used.

[0040] The estimation means 38 can also estimate whether an integrated dentition image, which is a combination of multiple dentition images, includes any dentition that can be orthodontally corrected. Figure 2(g) is an example of an integrated dentition image, which is a combination of the dentition images (a) to (f) shown in Figure 2. However, it is not always necessary to combine six dentition images into an integrated dentition image; the number and type of dentition images to be combined can be appropriately set and changed depending on the orthodontic plan to be determined.

[0041] In addition, when a dental arch image in which an arch has been drawn by the arch drawing means 36 is input, the estimation means 38 may compare the degree of distortion with a predetermined threshold value to estimate whether or not the dental arch image of the subject includes dental arches that can be subjected to orthodontic treatment (whether orthodontic treatment is possible, and whether a predetermined mouthpiece orthodontic plan can be applied).

[0042] (Judgment means 40) The determining means 40 determines whether orthodontic treatment is appropriate for the subject based on the estimation result by the estimating means 38. "Orthodontic treatment" may be wire or mouthpiece orthodontic treatment.

[0043] The determination means 40 can also determine whether a predetermined mouthpiece orthodontic plan is applicable based on the estimation result by the estimation means 38. The "mouthpiece orthodontic plan" is not particularly limited, and various orthodontic plans may be set. For example, a plan targeting all teeth, including molars, and a partial orthodontic plan are examples. Partial orthodontic plans include a plan targeting only the upper dentition, a plan targeting only 2 to 6 front teeth in the upper dentition, a plan targeting only the upper and lower dentition (e.g., 4 to 32 teeth, 4 to 24 teeth, or 4 to 12 teeth), a plan targeting 12 front teeth in the upper and lower dentition, and a plan targeting 24 front teeth in the upper and lower dentition.

[0044] The determination means 40 may determine whether orthodontic treatment is possible for the subject or whether a predetermined mouthpiece orthodontic plan can be applied, based on the shape of the arch drawn by the arch drawing means 36. For example, the determination whether or not a mouthpiece orthodontic plan can be applied may be made based on a rule that compares the shape with a threshold value of the arch distortion set for each mouthpiece orthodontic plan.

[0045] (output means 42) The output means 42 transmits a display instruction signal to the display unit 52 to cause the display unit 52 to display the determination result by the determination means 40 (for example, step S50 in FIG. 4). By displaying such a determination result, it is possible to assist the dentist in determining whether or not orthodontic treatment is appropriate for the subject.

[0046] The output means 42 may further display the image of the subject's dentition and details of the judgment result together with the judgment result, such as the confidence level of the judgment made by the trained model 200 and the object detection model 100, and the reason or cause of the judgment.

[0047] Even if the judgment result is that orthodontic treatment is not possible, the output means 42 may further present conditions under which orthodontic treatment is "possible." For example, the output means 42 may display on the display unit 52 a suggestion to treat the missing part, silver filling, tooth decay, periodontal disease, or gingivitis that caused the judgment result.

[0048] The output means 42 may transmit an instruction signal to the display unit 52 to output a sound, light, or color according to the judgment result. For example, if orthodontic treatment is not possible, an instruction signal may be transmitted to the display unit 52 to highlight the image of the dentition or the tooth that caused the judgment result with a color or the like.

[0049] (Object detection model generation means 44, trained model generation means 46) The object detection model generation means 44 and the trained model generation means 46 generate the object detection model 100 and the trained model 200 by machine learning using the training data 110 and 210, respectively. As described above, the learning method is not limited as long as it is capable of "input" and "output (classification)." In this way, the trained model 200 and the object detection model 100 are not limited to those generated by an external device, but may also be generated in the server 3.

[0050] The trained model 200 and / or the object detection model 100 may function by the control unit 30 executing a program stored in the memory unit 48, or may function by a device other than the server 3 executing a predetermined program.

[0051] If a dental image that has been estimated by the estimation means 38 to be suitable for orthodontic treatment is determined by a dentist to be unsuitable for orthodontic treatment, the dental image may also be used to retrain the trained model 200 (fine tuning, RLHF, etc.).

[0052] R-CNN, SSD, YOLO, etc. can be used as the learning model that forms the basis of the trained model 200 and the object detection model 100, but the YOLO model is preferred in terms of accuracy and speed.

[0053] [Information processing method 1] Next, an information processing method 1 in the above-described information processing system 1 will be described with reference to Fig. 4. Here, an example is shown in which a program for determining whether orthodontic treatment is possible from a dentition image is executed in the server 3. In the following description, components with the same reference numerals are the same as the components described above, and redundant description will be omitted as appropriate. Note that the processing described below is performed by executing a program stored in the storage unit 48, but the information processing method according to the present disclosure is not limited to this.

[0054] First, the control unit 30 (accepting means 32) accepts a plurality of images of the row of teeth of a subject (see FIG. 2) (step S10).

[0055] Next, the control unit 30 (estimation means 38) uses the trained model 200 generated by machine learning using training data 210 including images of teeth and information on whether the images contain teeth that can be orthodontally aligned, to estimate whether each of the multiple teeth images contains teeth that can be orthodontally aligned (step S20).

[0056] Next, the control unit 30 (determination means 40) determines whether orthodontic treatment is appropriate for the subject based on the estimation result of step S20 (step S30).

[0057] In this embodiment, if it is estimated that all the dentition images include dentition that is suitable for orthodontic treatment (step S30, "YES"), the control unit 30 (determination means 40) determines that orthodontic treatment is possible for the subject (step S40). In this case, the control unit 30 (output means 42) outputs this determination result, all the dentition images, and the confidence level for each dentition image to the display unit 52 (step S50). This ends the process. That is, a final confirmation is made by the dentist.

[0058] On the other hand, if it is estimated that none of the multiple dentition images includes a dentition that can be orthodontally aligned (step S30, "NO"), the control unit 30 (determination means 40) determines that orthodontic treatment is not possible for the subject (step S60). In this case, the control unit 30 (output means 42) outputs this determination result, all dentition images, the confidence level for each dentition image, and the reason for the determination result to the display unit 52 (step S70). Then, the process ends.

[0059] [Information processing method 2] Next, an information processing method 2 in the information processing system 1 will be described with reference to Fig. 5. The information processing method 2 is an example in which a dentition image is acquired from dentition scan data in the server 3, and a program for determining the suitability of a mouthpiece orthodontic plan is executed. Duplicate descriptions will be omitted where appropriate.

[0060] First, the control unit 30 (accepting means 32) accepts scan data of the row of teeth of the subject from the 3D scanner 2 (step S100).

[0061] Next, the control unit 30 (image processing means 34) processes the received dentition scan data to generate multiple dentition images (Figures 2(a) to (f)) including a dentition image of the subject's upper dentition, a dentition image of the lower dentition, and dentition images of the upper and lower dentition captured at an angle from diagonally below (step S110).

[0062] Next, the control unit 30 (image processing means 34) generates an integrated row-of-teeth image (FIG. 2(g)) by combining the plurality of row-of-teeth images (step S120).

[0063] Next, the control unit 30 (arch drawing means 36) uses the object detection model 100 generated by machine learning using training data 110 including images of the upper and lower dentition to detect and classify the teeth in the images of the subject's upper and lower dentition, and draws an arch showing the alignment of the teeth (step S130).

[0064] Next, the control unit 30 (estimation means 38) uses the trained model 200 to estimate whether or not each of the multiple dentition images obtained in step S110 and the integrated dentition image obtained in step S120 contains dentition that can be orthodontally corrected (step S140).

[0065] Next, the control unit 30 (determination means 40) determines whether orthodontic treatment is appropriate for the subject based on the estimation result in step S140 and the arch shape acquired in step S130 (step S150).

[0066] In this embodiment, if it is estimated that all of the tooth alignment images contain tooth alignments that are suitable for orthodontic treatment (step S150, "YES"), the control unit 30 (determination means 40) determines that orthodontic treatment is possible for the subject (step S160).

[0067] Next, the control unit 30 (determination means 40) determines that a predetermined mouthpiece orthodontic plan can be applied based on the arch shape acquired in step S130 (step S170).

[0068] Next, the control unit 30 (output means 42) outputs the determination result (including the applicable orthodontic plan), all dentition images (including the arch), and the confidence level for each dentition image to the display unit 52 (step S180). Then, the process ends.

[0069] On the other hand, if it is estimated that none of the multiple tooth row images includes a tooth row that can be orthodontally treated (step S150, "NO"), the control unit 30 (determination means 40) determines that orthodontic treatment is not possible for the subject (step S190). Note that in another embodiment, if the arch shape acquired in step S130 indicates an abnormal value, it may be determined that orthodontic treatment is not possible based only on the information acquired in step S130.

[0070] Next, the control unit 30 (output means 42) causes the display unit 52 to output the determination result, all the row-of-teeth images, the confidence level for each row-of-teeth image, and the reason for the determination result (step S200). Then, the process ends.

[0071] The above-described information processing method is an example, and the processing flow is not limited to the above. For example, the step of drawing the arch (step S130) may be performed before the step of generating the integrated dentition image (step S120).

[0072] In another embodiment, instead of the step of determining whether orthodontic treatment is feasible for the subject (steps S150 to S160), a determination step of determining whether orthodontic treatment is feasible for a specific mouthpiece orthodontic plan may be performed. In this case, images of the dentition and / or arch shape (curvature) compatible with a specific mouthpiece orthodontic plan may be used as training data 210 used to generate the trained model 200. In addition, in one embodiment, the applicability of multiple mouthpiece orthodontic plans may be estimated and determined simultaneously or consecutively. In this case, a trained model 200 is created for each orthodontic plan. Furthermore, if it is determined that all orthodontic plans are inapplicable, a step of determining that orthodontic treatment is not feasible for the subject (step S190) may be performed.

[0073] The information processing system 1, computer (server 3), program, and information processing method according to the present disclosure, configured as described above, are provided with a receiving means 32, an estimating means 38, and a determining means 40. The receiving means 32 receives a plurality of images of a subject's dentition. The estimating means 38 uses a trained model 200 generated by machine learning using training data 210 including images of the dentition and information indicating whether the images include dentition that is suitable for orthodontic treatment. The determining means 40 determines whether orthodontic treatment is suitable for the subject based on the estimation result by the estimating means 38. The information processing system 1, computer (server 3), program, and information processing method according to the present disclosure can efficiently and accurately assist dentists in determining whether orthodontic treatment is suitable.

[0074] Furthermore, in the information processing system 1, the computer (server 3), the program, and the information processing method according to the present disclosure, the determination means 40 may determine whether or not a predetermined mouthpiece orthodontic plan is applicable, based on the estimation result by the estimation means 38. In this way, it is possible to reduce the amount of training data 210 prepared to generate a trained model 200 in accordance with a specific mouthpiece orthodontic plan.

[0075] Furthermore, the information processing system 1, computer (server 3), program, and information processing method according to the present disclosure may further include image processing means 34. In this case, the receiving means 32 receives dentition scan data of the subject, and the image processing means 34 processes the dentition scan data received by the receiving means 32 to generate and acquire multiple dentition images. In this way, a series of data processing operations can be completed on a single computer (server 3).

[0076] Furthermore, in the information processing system 1, computer (server 3), program, and information processing method according to the present disclosure, the image processing means 34 may process the dentition scan data received by the receiving means 32 to generate a dentition image in which the subject's upper and lower dentition are captured at an angle from diagonally below. This is preferable because processing the dentition scan data in this way increases the number of features for estimating orthodontic treatment using the trained model 200, enabling more accurate estimation processing.

[0077] Furthermore, in the information processing system 1, computer (server 3), program, and information processing method according to the present disclosure, the estimation means 38 can also estimate whether an integrated dental image, which is a combination of multiple dental images, includes any dental alignment that can be corrected. By providing an integrated dental image in this way, it is possible to comprehensively include areas that would be overlooked in a dental alignment image viewed from a specific direction as the target of estimation, thereby enabling more accurate estimation processing.

[0078] Furthermore, the information processing system 1, computer (server 3), program, and information processing method according to the present disclosure may further include arch drawing means 36 that uses an object detection model 100 generated by machine learning using training data 110 including images of the upper and lower teeth to detect and classify teeth in each of the subject's upper and lower teeth as part of multiple teeth images, and draws an arch indicating the tooth alignment. In this way, the determination means 40 can more accurately determine whether orthodontic treatment or a predetermined mouthpiece orthodontic plan is appropriate for the subject, based on the shape of the arch.

[0079] The information processing system 1, the computer (server 3), the program, and the information processing method according to the present disclosure are not limited to the above-described aspects and combinations, and various modifications can be made. [Explanation of symbols]

[0080] 1. Information processing system (AI orthodontic treatment feasibility assessment support system) 2. 3D scanner 21 Imaging unit 22 Communications Department 3 Server (computer) 30 Control Unit 32 Reception methods 34 Image processing means 36 Arch Drawing Tools 38 Estimation means 40 Judgment means 42 Output Method 44 Object detection model generation means 46 Trained model generation method 48 Memory section 50 Communications Department 52 Display section 54 Operation section 100 object detection models 110 Teacher Data 200 trained models 210 Teacher Data

Claims

1. A program that causes a computer to function as a receiving means, an estimating means, and a determining means, The receiving means receives a plurality of images of the row of teeth of the subject, the estimation means estimates whether or not each of the plurality of dental row images includes a dental row that can be orthodontally aligned, using a trained model generated by machine learning using training data including an image of a dental row and information on whether the image includes a dental row that can be orthodontally aligned, and The determination means determines whether orthodontic treatment is appropriate for the subject based on the estimation result by the estimation means.

2. The program according to claim 1, wherein the determining means determines whether or not a predetermined mouthpiece orthodontic plan is applicable based on the estimation result by the estimating means.

3. causing the computer to further function as an image processing unit; The receiving means receives dentition scan data of the subject, The program according to claim 1 , wherein the image processing means processes the row of teeth scan data received by the receiving means to generate a plurality of row of teeth images.

4. The program according to claim 3 , wherein the image processing means processes the dentition scan data received by the receiving means to generate a dentition image in which the subject's upper and lower dentition are captured at an angle from diagonally below.

5. 5. The program according to claim 1, wherein the estimation means estimates whether an integrated dental image obtained by combining a plurality of dental image data includes any dental alignment that can be orthodontally treated.

6. The computer further functions as an arch drawing means, the arch drawing means uses an object detection model generated by machine learning using training data including an image of an upper dentition and an image of a lower dentition to detect and classify teeth in each of the images of the upper dentition and the lower dentition of the subject as part of the plurality of dentition images, and draws an arch representing the alignment of the teeth; The program according to claim 1 , wherein the judgment means judges whether orthodontic treatment or a predetermined mouthpiece orthodontic plan can be applied to the subject based on the shape of the arch.

7. A computer that functions as a receiving means, an estimating means, and a determining means by executing a program, The receiving means receives a plurality of images of the row of teeth of the subject, the estimation means estimates whether or not each of the plurality of dental row images includes a dental row that can be orthodontally aligned, using a trained model generated by machine learning using training data including an image of a dental row and information on whether the image includes a dental row that can be orthodontally aligned, and The judgment means is a computer that judges whether orthodontic treatment is appropriate for the subject based on the estimation result by the estimation means.

8. An information processing method executed by a computer having a control unit, a step of the control unit receiving a plurality of images of the row of teeth of a subject; a step in which the control unit estimates whether or not each of the plurality of dental row images includes a dental row that can be orthodontally aligned, using a trained model generated by machine learning using training data including an image of a dental row and information on whether the image includes a dental row that can be orthodontally aligned; a step of the control unit determining whether or not orthodontic treatment is possible for the subject based on the estimation result; An information processing method comprising:

Images