Method for planning dental treatment on basis of lesion severity, apparatus therefor, and recording medium

Abstract

The present invention relates to a method for planning a dental treatment on the basis of lesion severity, an apparatus therefor, and a recording medium. According to the method for planning a dental treatment according to the present invention, lesion information and severity information of a lesion area detected from a medical image are visualized and provided, and a treatment plan for treating the lesion area is automatically established by determining a treatment priority for multiple lesion areas on the basis of the severity information. According to the present invention, the state of a patient can be accurately and intuitively identified through visualization of lesion information, and a more reasonable and optimized treatment solution can be provided to the patient by automatically establishing a treatment plan on the basis of lesion severity.

Description

Method for planning dental treatment based on lesion severity, device therefor, and recording medium

[0001] The present invention relates to a dental treatment planning method, a device therefor, and a recording medium, and more particularly, to a technology for automatically establishing and providing a dental treatment plan according to the oral condition of a patient based on a medical image of the patient.

[0002]

[0003] Unlike other medical fields, such as internal medicine and otolaryngology, dentistry offers a variety of treatment options for specific conditions. For example, when visiting an internal medicine or otolaryngology clinic for a cold, patients are rarely given a choice regarding their prescription. In contrast, when a patient visits the dentist for severe toothache, pain relief is the top priority. However, once the pain is resolved, the patient has a variety of treatment options for follow-up care, such as prosthetic treatment. For example, root canal treatment and tooth extraction may be options for a patient experiencing pain due to deep caries. Furthermore, abutment buildup or prosthetic treatment after root canal treatment may be optional and voluntary for the patient.

[0004] In this way, the dental field offers a variety of treatment options for patients to choose from, but patients often lack knowledge or information about these diverse options. Furthermore, dental treatments are often expensive, so patients expect accurate and detailed information based on their current oral condition to help them choose the most appropriate and reasonable treatment option at the time. While many dental clinics offer consultations to assist patients in their choices, these often remain limited to verbal explanations and presentations of medical images, falling far short of the level of information patients require. Consequently, patients frequently visit multiple dental clinics for reconfirmation, or are pressured into making treatment choices only to later regret it.

[0005] Therefore, there is a need for a method that can provide objective and reliable information that meets the level of information required by patients to assist them in their treatment choices and to establish a more optimized treatment plan for the patient based on the above information.

[0006]

[0007] The present invention has been proposed to solve the problems of the prior art as described above, and its purpose is to provide a dental treatment planning method and device that visualizes and provides information on lesion severity and treatment priority identified by analyzing a patient's medical image, and establishes a dental treatment plan optimized for the patient based on the information.

[0008]

[0009] The above object can be achieved by a dental treatment planning method according to one aspect of the present invention, wherein each step is performed through a computing device, the method comprising: a step of loading a medical image of a patient; a step of detecting at least one lesion area from the medical image; a step of calculating the severity of each lesion area based on the lesion progression status of each lesion area; and a step of visualizing and displaying the lesion information and the severity information of each lesion area.

[0010] Furthermore, the method may further include a step of determining treatment priorities for a plurality of lesion areas based on the severity of each lesion area; and a step of determining a treatment plan for treating the lesion area by reflecting the treatment priorities.

[0011] In addition, a step of calculating the expected treatment cost by reflecting the determined treatment plan may be further included.

[0012] In addition, the method further includes a step of receiving user input regarding the patient's available treatment cost, and the step of calculating the expected treatment cost may calculate the expected treatment cost by selecting a treatment item that can satisfy the range of the available treatment cost among a plurality of treatment items according to the treatment plan.

[0013] In addition, the method may further include a step of extracting consultation data regarding the treatment plan from among the stored consultation data and configuring a consultation screen for patient consultation.

[0014] Meanwhile, the step of detecting a restored object as a result of a previous treatment in the medical image is further included, and the step of calculating the severity of each lesion area can calculate the severity by reflecting whether the lesion area corresponds to a secondary lesion associated with the restored object.

[0015] In addition, the lesion area includes a tooth loss area where teeth are lost, and the step of calculating the severity of each lesion area can calculate the severity by reflecting information on the state in which the opposing teeth of the tooth loss area are protruded relative to the occlusal plane and the state in which the adjacent teeth of the tooth loss area are tilted.

[0016] In addition, the step of calculating the severity of each lesion area may calculate the severity by reflecting information about the patient's main complaint.

[0017] And, the step of calculating the severity of each lesion area can calculate the severity by reflecting information about the influence received by adjacent anatomical structures by the lesion area.

[0018] Furthermore, the step of calculating the severity of each lesion area may calculate the severity by reflecting the patient's age information, information on the tooth in which the lesion area exists among the baby teeth or permanent teeth, and whether there is a permanent tooth succeeding the baby teeth.

[0019] Meanwhile, the method may further include a step of displaying the lesion information and the severity information on the patient's electronic chart.

[0020] Here, the lesion information may include at least one of the tooth number where the lesion area exists, the lesion detailed location, the lesion type, the lesion progression stage, or the detection reliability of the lesion area.

[0021] In addition, the step of detecting the lesion area can be performed by segmenting a plurality of anatomical structures for each tooth in the medical image and applying an artificial intelligence model learned to identify the location and lesion type of the lesion area existing in each tooth.

[0022] At this time, the artificial intelligence model may include a first artificial neural network trained to segment the plurality of anatomical structures in the medical image; a second artificial neural network trained to identify the lesion area existing in each tooth based on the result of segmentation processing through the first artificial neural network; and a third artificial neural network trained to output data regarding the location, lesion type, and lesion detection reliability of the lesion area identified through the second artificial neural network.

[0023] In addition, the above-described object can be achieved by a dental treatment planning device that performs dental treatment planning according to another aspect of the present invention, comprising a processor, wherein the processor loads a medical image of a patient, detects at least one lesion area from the medical image, calculates the severity of each lesion area based on the lesion progression status of each lesion area, and performs processing to visualize and display the lesion information and the severity information of each lesion area.

[0024]

[0025] As described above, according to the present invention, by providing visualized information such as severity information and lesion type of each lesion area detected in a medical image, it is possible to accurately and intuitively understand the patient's condition.

[0026] In addition, according to the present invention, a treatment plan is automatically established based on the severity of the lesion, and a treatment cost plan is established and provided within the patient's available treatment cost, thereby providing a more reasonable and optimized treatment solution to the patient.

[0027]

[0028] FIG. 1 is a block diagram showing the configuration of a dental treatment planning device according to one embodiment of the present invention;

[0029] FIG. 2 is a flowchart illustrating a dental treatment planning method according to one embodiment of the present invention;

[0030] Figures 3a to 3d are examples of segmentation performed on each anatomical structure in a medical image;

[0031] FIG. 4 is a reference diagram for explaining a learning method of an artificial intelligence model for lesion area detection according to one embodiment of the present invention;

[0032] FIG. 5 is a reference diagram for explaining a method for identifying periodontal disease according to one embodiment of the present invention;

[0033] FIG. 6a and FIG. 6b are examples of screens in which detection results of a lesion area and a restored object are provided by a dental treatment planning device according to an embodiment of the present invention;

[0034] FIG. 7 is a reference diagram for explaining an example of calculating the severity of a lesion area by reflecting whether there is a secondary lesion according to one embodiment of the present invention;

[0035] FIG. 8 is a reference diagram for explaining an example of calculating the severity of a lesion area by reflecting status information of adjacent anatomical structures in a tooth loss area according to one embodiment of the present invention;

[0036] FIG. 9 is a reference diagram for explaining an example of calculating the severity of a lesion area by reflecting influence information on adjacent anatomical structures according to one embodiment of the present invention;

[0037] FIG. 10A and FIG. 10B are reference diagrams for explaining an example of calculating the severity of a lesion area by reflecting information on a tooth where a lesion area exists and information on a patient's age according to one embodiment of the present invention;

[0038] FIG. 11 is an example of a screen in which information regarding a lesion area detected through a dental treatment planning device according to an embodiment of the present invention is visualized and provided on a medical image;

[0039] FIG. 12 is an example of a screen in which information regarding a lesion area detected through a dental treatment planning device according to an embodiment of the present invention is provided as visualization on an electronic chart;

[0040] FIG. 13 is an example of an electronic chart screen displaying a treatment plan determined through a dental treatment planning device according to an embodiment of the present invention;

[0041] FIG. 14 is an example of a screen in which the expected treatment cost according to the treatment plan is calculated and displayed through a dental treatment planning device according to an embodiment of the present invention; and

[0042] Fig. 15 is an example of a patient consultation screen configured through a dental treatment planning device according to an embodiment of the present invention.

[0043]

[0044] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, detailed descriptions of well-known functions or components that may obscure the gist of the present invention will be omitted in the following description and the accompanying drawings. It should also be noted that, where possible, identical components are indicated with the same reference numerals throughout the drawings.

[0045] The terms and words used in this specification and claims described below should not be interpreted as limited to their conventional or dictionary meanings, but should be interpreted with meanings and concepts that conform to the technical idea of ​​the present invention based on the principle that the inventor can appropriately define the concept of the term to best describe his or her invention. Therefore, the embodiments described in this specification and the configurations illustrated in the drawings are merely the most preferred embodiments of the present invention and do not represent all of the technical idea of ​​the present invention. Therefore, it should be understood that there may be various equivalents and modified examples that can replace them at the time of filing this application.

[0046] In the flowcharts described with reference to the drawings of this specification, the order of operations may be changed, several operations may be merged, some operations may be split, and certain operations may not be performed.

[0047] The dental treatment planning device according to the present invention provides visualized lesion information detected through analysis of a patient's medical image, and supports automatic establishment of a treatment plan by determining treatment priorities for multiple lesions.

[0048] Fig. 1 is a block diagram showing the configuration of a dental treatment planning device according to one embodiment of the present invention. The dental treatment planning device (1) is an electronic device that can execute software that analyzes a patient's medical image and establishes a treatment plan for treating a lesion detected in the medical image, and can be implemented through a computing device such as a computer, a notebook computer, a laptop computer, a tablet PC, a smartphone, a mobile phone, a PMP (Personal Media Player), a PDA (Personal Digital Assistants).

[0049] Referring to FIG. 1, a dental treatment planning device (1) according to one embodiment of the present invention includes a user input unit (10), a display unit (20), a memory (30), and a processor (40).

[0050] The user input unit (10) is a module for receiving various inputs from the user in the process of establishing a dental treatment plan through analysis of a patient's medical image, and can be implemented with various input devices such as a mouse or keyboard.

[0051] The display unit (20) is configured to display various information including text, graphics, etc. on the screen, and displays various data, images, GUI, and processing outputs provided in the process of processing through the dental treatment planning device (1), including the patient's medical images, visualized lesion information and lesion severity information, patient electronic chart, and consultation screen for patient consultation.

[0052] Memory (30) is a readable recording medium in a computing device, and at least one computer program code executed by the processor (40) can be stored therein. The computer program code can also be loaded into the memory (30) from a floppy drive, disk, tape, DVD / CD-ROM drive, memory card, etc. separate from the memory (30).

[0053] The memory (30) can store software for visualizing lesion information detected in medical images and establishing a treatment plan, patient data including patient medical image data, and consultation data such as videos, images, and text data produced for use in patient consultation regarding the treatment process or treatment content.

[0054] The processor (40) is for executing and processing computer program commands by performing basic logic, calculations, operations, etc., and the processor (40) detects at least one lesion area from a patient's medical image according to the execution of a program stored or loaded in the memory (30), provides visualized lesion information and lesion severity information, and determines treatment priorities for multiple lesion areas to automatically establish a treatment plan for treating the lesion areas.

[0055] Hereinafter, each step of the dental treatment planning method performed by the dental treatment planning device (1) will be described with reference to FIGS. 2 to 15.

[0056] Figure 2 is a flowchart illustrating a dental treatment planning method according to one embodiment of the present invention.

[0057] Referring to Fig. 2, a patient's medical image is loaded into a memory (30) (S100). Here, the medical image includes images according to various modalities, such as images acquired through an intraoral camera, a digital camera, CT images, panoramic images, cephalometric images, intraoral sensor images, radiographic images, and intraoral scan images, and may also be an image obtained by matching two or more medical images according to the same or different modalities.

[0058] The processor (40) detects at least one lesion area from the loaded patient medical image (S110). Here, the lesion area refers to an area where various oral lesions such as dental caries, periodontal disease, and periapical inflammation exist, and a tooth loss area where a tooth is lost is also included in the lesion area. The processor (40) can detect the lesion area by applying an artificial intelligence model trained to segment a plurality of anatomical structures for each tooth in the medical image and identify the location and lesion type of the lesion area existing in each tooth. For reference, the learning can be performed using various machine learning or deep learning networks such as U-net, CNN (Convolutional Neural Network), RNN (Recurrent Neural Network), Deep Belief Network, and Restricted Boltzman Machine, and is not limited to a specific network.

[0059] Here is an example of an artificial intelligence model configuration for detecting a lesion area. The artificial intelligence model may include a first artificial neural network trained to segment multiple anatomical structures for each tooth in a medical image, a second artificial neural network trained to identify a lesion area existing in each tooth based on the segmentation results processed by the first artificial neural network, and a third artificial neural network trained to output data regarding the specific location of the lesion area identified by the second artificial neural network, the lesion type, and the detection confidence score of the lesion area. The detection confidence score may be expressed in percentage units as a probability value indicating the accuracy of the result predicted by the artificial neural network. For reference, when training the first to third artificial neural networks, not only the original medical image, but also an augmented image such as by reversing the color of the original medical image, rotating it at an angle, reversing the image direction, or adjusting the window level or window width may be applied as training data. Here, the window level refers to the median value of the gray scale, and the window width refers to the range of HU (HoUnsfield) values ​​that can be expressed in gray scale.

[0060] Below, regarding the first to third artificial neural networks, the first artificial neural network first segments the tooth region, crown, dental pulp, and bone level in the input medical image.

[0061] Figures 3a to 3d illustrate examples of segmentation performed on each anatomical structure in a medical image. Note that while Figures 3a to 3d illustrate an intraoral panoramic image as an example of a medical image, it is clear that processing for lesion area detection can be performed on various modality images, such as intraoral sensor images and CT images, in addition to intraoral panoramic images.

[0062] Figure 3a shows the tooth region, Figure 3b shows the crown, Figure 3c shows the pulp, and Figure 3d shows the bone level, each masked. The first artificial neural network can be created by training using medical images with each region masked, as shown in Figures 3a to 3d, as learning data.

[0063] Next, the second artificial neural network receives an image in which the anatomical structure of each tooth has been segmented through the first artificial neural network, identifies the lesion area existing in each tooth, and outputs the output. The second artificial neural network can be trained by applying training data in which each lesion area is labeled, and at this time, training can be performed by dividing it into multiple parts according to the specific lesion type. For example, training can be performed individually for each lesion such as dental caries, periapical inflammation, and periodontal disease. After training individually according to the lesion type as above, all lesions can be integrated and ultimately trained as a single second artificial neural network.

[0064] In addition, considering that the specific location of the lesion in the tooth area differs depending on the type of lesion, the tooth area is divided into multiple detailed areas and a region of interest (ROI) corresponding to each lesion is set so that intensive learning can be conducted for each ROI.

[0065] FIG. 4 is a reference diagram for explaining a learning method of an artificial intelligence model for detecting a lesion area according to one embodiment of the present invention, and shows an example of dividing a tooth area into multiple detailed areas.

[0066] Referring to Fig. 4, the tooth region can be sequentially divided into three in the direction from the crown to the root: the uppermost region (401) of the crown region, the lowermost region (403) of the root region, and the middle region (402) between the two regions (401, 403). In the tooth region divided as above, the uppermost region (401) is set as the ROI for the tooth caries region, the middle region (402) is set as the ROI for periodontal disease, and the lowermost region (403) is set as the ROI for periapical inflammation, and a second artificial neural network for identifying each lesion region can be trained for the set ROIs. In this way, by intensively performing training on the detailed region where the lesion actually exists, the accuracy of lesion detection can be effectively improved.

[0067] Furthermore, learning can be performed by further subdividing each region (401, 402, 403) of Fig. 4. Fig. 4 shows an example in which each region (401, 402, 403) is further divided into three regions in a direction parallel to the tooth axis, thereby dividing the tooth region into a total of nine subregions. Through this, it is possible to identify lesion regions by subregion by dividing them into mesial, distal, and central planes (occlusal planes in the case of the uppermost region).

[0068] Additionally, the second artificial neural network can be trained to calculate a quantitative value for each identified lesion area based on the lesion's progression status, or a specific level based on preset level classifications for each lesion. To this end, criteria for calculating the level classification or quantitative value for determining the degree of progression for each lesion can be pre-established.

[0069] For example, in the case of dental caries, it can be classified into levels C0 to C4 depending on the progression such as the area and depth of occurrence. Here, level C0 is an early caries state with a low degree of caries, level C1 is a state in which caries has developed up to the enamel layer, level C2 is a state in which caries has developed up to the dentin, level C3 is a state in which caries has progressed up to the nerve, and level C4 is a severe state in which caries has progressed up to the periapical lesion or the residual root, such as an abscess formed at the root tip.

[0070] Meanwhile, periodontal disease can be identified by analyzing the amount of subsidence at the bone level based on the segmented crown through the first artificial neural network.

[0071] FIG. 5 is a reference diagram for explaining a method for identifying periodontal disease according to one embodiment of the present invention.

[0072] Referring to Fig. 5, if there is a subsidence amount of the bone level greater than a predetermined reference value, for example, greater than 2 mm, from any point on the crown, it can be identified as periodontal disease. At this time, even in the case of periodontal disease, it can be learned to calculate the lesion progression level according to the specific subsidence distance (501) from any point on the crown to the bone level. That is, the lesion progression level can be set corresponding to each subsidence distance section, and the lesion progression level corresponding to the subsidence distance (501) measured through medical imaging can be calculated.

[0073] Next, in the case of periapical inflammation, it can be identified based on the HU value and the shape of the inflammation area. The periapical inflammation area can be identified by applying the characteristics that it is adjacent to the tooth based on the lowermost region (403) of Fig. 4, does not have the HU value of the bone and tooth, and has a roughly circular shape. In the case of periapical inflammation, it can also be learned to derive a specific lesion progression level based on the size of the inflammation area, etc.

[0074] Next, the third artificial neural network is trained to output data regarding the specific location of the lesion area identified by the second artificial neural network, the lesion type, and the lesion detection reliability. As illustrated in Fig. 4, the third artificial neural network can detect the exact location of the lesion area by training it by dividing the tooth area into nine sub-regions, thereby further improving the detection reliability. The third artificial neural network can be created by training it using training data labeled with the lesion location and lesion type in the nine sub-regions.

[0075] Referring back to FIG. 2, detection of not only the lesion region described above in the medical image but also the restored object as a result of previous treatment is performed simultaneously (S120). Note that the flowchart of FIG. 2 depicts steps S110 for detecting the lesion region and S120 for detecting the restored object as being performed sequentially; however, the order of these two steps may be changed and may be performed simultaneously.

[0076] A restored object refers to an object already existing in the patient's oral cavity as a result of treatment for a previously occurring lesion, including artificial crowns, implants, root canal restorations, and filling restorations. For reference, an artificial crown is an object that covers the entire tooth crown, while a filling restoration refers to an object that partially covers the tooth crown after caries has been removed using resin, amalgam, etc. The information on the detected restored object can be utilized to calculate the severity of the lesion area, as described below.

[0077] Areas where restorations are installed through treatment often exhibit higher pixel values ​​than the average pixel value in medical images due to metal artifacts. This characteristic can be exploited to detect restored objects by extracting areas with pixel values ​​above a certain brightness level. Detection of restored objects can also be achieved using an AI model. The AI ​​model can be generated by supervised learning using labeled medical images of restored objects as training data.

[0078] The results detected through steps S110 and S120 can be displayed and provided to the user through the display unit (20).

[0079] FIG. 6a and FIG. 6b show an example of a screen in which the detection results of a lesion area and a restored object are provided by a dental treatment planning device (1) according to an embodiment of the present invention.

[0080] First, referring to Fig. 6a, it can be confirmed that the location and type of each lesion area and the restored object detected in the medical image, and the detection confidence (%) of the lesion area are visualized and displayed in the medical image.

[0081] Referring to Fig. 6b, it can be seen that the detection results are provided in a visualized form as a list showing the type of lesion or restored object detected in the tooth corresponding to each tooth number and the detection reliability.

[0082] In this way, the detection results of the lesion area and the restored object can be visualized and provided in various ways, such as being overlaid on a medical image, or the detection results for each tooth number being provided in the form of a list or icon (not shown).

[0083] As described above, once the detection of lesion areas in a medical image is completed, a process of calculating the severity of each detected lesion area follows (S130). The severity of a lesion area is an indicator that serves as a criterion for determining treatment priorities, and as described below, it is calculated by comprehensively reflecting various factors such as the lesion progression status of the lesion area, the patient's main complaint, whether there are secondary lesions related to the restored object, the impact on adjacent anatomical structures of the lesion area or the status of the adjacent anatomical structures, and information on the teeth where the lesion area exists.

[0084] Hereinafter, regarding the multiple items considered for calculating the severity of the lesion area, the severity of the lesion area can basically be calculated by reflecting the lesion progression status of the lesion area. At this time, the lesion progression status can be identified through the lesion progression level or quantitative numerical value of the lesion area calculated through the second artificial neural network as explained above. That is, in the case of dental caries, the lesion progression status can be identified based on the area of ​​caries occurrence, the depth of caries, etc., and in the case of periodontal disease, the lesion progression status can be identified from the subsidence of the bone level, which is identified through the distance from an arbitrary point of the crown to the bone level, i.e., the downward movement of the bone level toward the root. In addition, in the case of periapical inflammation, a certain lesion level can be assigned in itself, but the specific lesion progression status can also be identified based on the size of the inflamed area, etc.

[0085] Depending on the degree of lesion progression as described above, a severity score can be differentially assigned. For example, assuming that the dental caries progression level is divided into C0 to C4 as described above, C0 can be given 1 point, C4 can be given 5 points, and so on, and a score can be differentially assigned for each stage. In addition, periodontal disease can be classified into gingivitis if bone loss is less than 4 mm from the segmented crown through the first artificial neural network, early periodontitis if bone loss is 4 to 6 mm, intermediate periodontitis if bone loss is 6 to 8 mm, and late periodontitis if bone loss is 8 to 12 mm, and a differential score can be assigned for each level by classifying the lesion progression level. Meanwhile, since periapical inflammation is a lesion caused by dental caries or irreversible pulpitis at the dental caries level C3 or C4, a severity score equivalent to the C3 or C4 score can be given by taking into account the size of the inflammatory area, etc.

[0086] In addition, the severity of the lesion area can be calculated by reflecting whether the lesion corresponds to a secondary lesion related to a restored object and / or the type of secondary lesion. Here, a secondary lesion refers to a new lesion that has occurred in or around a tooth (in the case of an implant, an artificial tooth root) where a restored object exists. For example, if inflammation occurs around an existing restored object, such as an implant, or a secondary lesion is detected in or around a tooth restored with a filling restoration or a crown, a relatively high severity score is assigned because measures such as removal or removal of the existing restored object and subsequent retreatment or tooth extraction are required.

[0087] FIG. 7 is a reference diagram for explaining an example of calculating the severity of a lesion area by reflecting whether there is a secondary lesion according to one embodiment of the present invention.

[0088] Referring to Fig. 7, it can be confirmed that an implant exists in the tooth area (701) number 46 and a peri-implantitis lesion exists. For reference, peri-implantitis refers to inflammation that occurs around an implant, and periodontitis refers to inflammation that occurs around a tooth. Accordingly, for the lesion in the tooth area (701) number 46, a severity score is primarily assigned according to the progression of peri-implantitis, and since it corresponds to a secondary lesion related to an existing restoration object as bone resorption has occurred based on the implant, an additional severity score can be assigned. For example, according to the progression of the lesion, 3 points corresponding to mid-stage periodontitis and an additional severity score of 8 points according to the secondary lesion can be assigned, resulting in a total of 11 points as the severity score.

[0089] In addition, looking at tooth area 45 (703) in FIG. 7, it can be confirmed that a crown, which is an existing restoration object, exists and a periapical inflammatory lesion exists. Therefore, a severity score (e.g., 3 points) is initially assigned according to the progression of periapical disease, and an additional severity score (e.g., 8 points) is assigned according to the secondary lesion, so that a total of 11 points can be calculated as a severity score.

[0090] Next, let's look at another item for calculating the severity of the lesion area. Among the detected lesion areas, a preset severity score is initially assigned to the tooth loss area, and an additional severity score can be assigned by reflecting information on the state of the opposing tooth in the tooth loss area being extruded compared to the occlusal plane and the state of the adjacent teeth in the tooth loss area being tilted. For example, if the opposing tooth in the tooth loss area is extruded by a preset value, for example, 2 mm or more, compared to the occlusal plane, an additional severity score can be assigned. This is to consider that patients requiring root canal treatment can be distinguished when the extruded antagonist tooth is removed in accordance with the occlusal plane. In addition, if the adjacent teeth in the tooth loss area are tilted, an additional severity score can be assigned to the tooth loss area to distinguish patients who have difficulty securing prosthetic space and poor alveolar bone condition. For reference, whether a tooth is tilted or not is determined by creating a tooth axis based on the bounding box of the segmented tooth area through the first artificial neural network, and determining whether the tooth axis is tilted by a preset angle or more than the tooth axis of the average tooth.

[0091] FIG. 8 is a reference diagram for explaining an example of calculating the severity of a lesion area by reflecting status information of adjacent anatomical structures in a tooth loss area according to one embodiment of the present invention.

[0092] Referring to Fig. 8, since the tooth area of ​​No. 46 is detected as a tooth loss area (801), a predetermined severity score (e.g., 9 points) can be initially assigned to the tooth loss area (801). Next, since the teeth No. 15, 16, and 17 (803, 804, and 805) corresponding to the opposing teeth of the tooth loss area (801) are in a state of being extruded compared to the occlusal plane (807), and at the same time, the tooth No. 47 (806), which is an adjacent tooth of the tooth loss area (801), is in a state of being tilted forward, an additional severity score (e.g., 2 points) can be assigned accordingly, so that a total of 11 points can be calculated as a severity score for the tooth loss area (801).

[0093] For reference, the periodontal lesion areas present in teeth 15, 16, and 17 (803, 804, and 805) are assigned a primary severity score based on the progression level of the periodontal lesion. For example, 2 points corresponding to early periodontitis can be assigned as a severity score to the lesion areas of each tooth (803, 804, and 805).

[0094] Looking at another item, the severity of a lesion area can be calculated by reflecting information about the patient's chief complaint. Here, the chief complaint (CC) is a medical term that refers to the most important symptom that caused the patient to visit the hospital. The chief complaint may include information about the location and degree of pain felt by the patient, and may be received from a patient management system (PMS) linked to the dental treatment planning device (1) or input through a user input unit (10). When the patient complains of pain at the time of registration for treatment according to the chief complaint among multiple lesion areas, a relatively high severity score can be assigned to the pain area among the multiple lesion areas. For example, a score set initially is assigned according to the lesion progression level of a specific lesion area, and if the lesion area corresponds to the chief complaint, a preset score (e.g., 10 points) can be added as an additional severity score.

[0095] Another consideration is that the severity of a lesion region can be calculated by reflecting information about the impact of the lesion on adjacent anatomical structures. That is, if there are adjacent anatomical structures affected by the lesion, the lesion region can be given a relatively high severity score.

[0096] FIG. 9 is a reference diagram for explaining an example of calculating the severity of a lesion area by reflecting influence information on adjacent anatomical structures according to one embodiment of the present invention.

[0097] First, looking at an example of calculating the lesion severity score for the periapical inflammation area (903) of teeth 33 and 34 (901, 902) in FIG. 9, a severity score (e.g., 3 points) is primarily assigned according to the level of periapical inflammation progression. Here, since the range of periapical inflammation is wide and includes two or more teeth, affecting adjacent teeth, an additional severity score (e.g., 1 point per tooth) is assigned to the periapical inflammation area (903) to reflect this, and as a result, a total of 4 points can be assigned to the periapical inflammation area (903) of each tooth (901, 902).

[0098] Meanwhile, looking at the tooth caries area (904) that occurred in tooth 47, a score is primarily given according to the tooth caries progression level (e.g., 2 points), and the tooth caries area (904) above is an adjacent surface caries that occurred on the surface that comes into contact with tooth 48 (905), which corresponds to the lower right third molar, and is the cause of adjacent tooth caries, so an additional severity score (e.g., 1 point) is given, so that a total of 3 severity scores can be given to the tooth caries area (904) in tooth 47.

[0099] For reference, for each periodontitis area detected in teeth 24 to 27 (906-909), a score (e.g., 3 points) is initially assigned according to the periodontitis progression level as intermediate periodontitis. However, since the periodontitis existing in teeth 26 and 27 (908, 909) corresponds to a secondary lesion as these teeth were previously restored with crown prostheses, an additional severity score (e.g., 8 points) may be assigned to the periodontitis areas of teeth 26 and 27 (908, 909) as described above with reference to FIG. 7. As a result, 3 points may be assigned to the periodontitis areas of teeth 24 and 25 (906, 907), and 11 points may be assigned to the periodontitis areas of teeth 26 and 27 (908, 909) as lesion severity scores, respectively.

[0100] Next, looking at another item, the severity of the lesion area can be calculated by reflecting the patient's age information, the tooth information where the lesion area exists, whether it is a primary or permanent tooth, and the presence of a permanent tooth following the primary tooth. Here, the patient's age is a variable for distinguishing between lesions on primary teeth and lesions on permanent teeth, and the age information can be received from a patient management system (PMS) linked to the dental treatment planning device (1), or can be input through a user input unit (10).

[0101] In general, because the lifespan of deciduous teeth is significantly shorter, they can be assigned a relatively lower severity score than permanent teeth even if the same lesion exists. However, if there are no subsequent permanent teeth, a higher severity score is assigned even for deciduous teeth.

[0102] FIG. 10A and FIG. 10B are reference diagrams for explaining an example of calculating the severity of a lesion area by reflecting information on the teeth where the lesion area exists and information on the patient's age according to one embodiment of the present invention.

[0103] First, referring to Figure 10a, a tooth with a carious area (1001) corresponds to a baby tooth, with a subsequent permanent tooth (1002) located below it. The presence of a subsequent permanent tooth can be determined through segmentation processing using the first artificial neural network.

[0104] Therefore, when the patient's age according to the patient information is 9 years old, the patient is in the mixed dentition period and is a case in which a caries area (1001) is detected in a primary tooth with subsequent permanent teeth, so a relatively low severity score can be assigned to reflect this. For example, a lesion severity score according to the dental caries status is primarily assigned, and a preset score can be deducted from the assigned score considering that the tooth in which the dental caries occurred is a primary tooth. For example, a preset point of 1 can be deducted for the lesion area in the primary tooth from 2 points corresponding to the C2 tooth caries level, so that a total of 1 point can be assigned as the lesion severity of the dental caries area (1001).

[0105] On the other hand, Fig. 10b shows an example of a medical image of a 15-year-old patient whose permanent teeth have all erupted and whose mixed dentition has passed. Referring to Fig. 10b, it can be confirmed that although the mixed dentition has passed, a primary tooth (1004) remains and a dental caries area (1005) exists in the primary tooth (1004), but there is no subsequent permanent tooth below it. For reference, the distinction between primary and permanent teeth can be identified through the size and / or root length of the corresponding tooth. In this way, even if a lesion area exists in a primary tooth without a subsequent permanent tooth, a relatively higher severity score can be assigned than a lesion area in a permanent tooth. That is, a lesion severity score is first assigned according to the dental caries status, and a preset score can be added to this score. For example, a total of 3 points can be assigned as a lesion severity score for the dental caries area (1004) by adding a preset score of 1 point to a 2-point score corresponding to a C2 dental caries level for a lesion area in a primary tooth without a subsequent permanent tooth.

[0106] Above, we have looked at an example of calculating a lesion severity score by reflecting the lesion progression status of the lesion area, whether it is a secondary lesion, the status of the anatomical structures around the tooth loss area, the patient's main complaint, the impact on adjacent anatomical structures, and the information on the tooth where the lesion area is located among the deciduous and permanent teeth.

[0107] However, in the example explained above, the severity score was given primarily according to the lesion progression status, and then the additional score set for each item was added or subtracted in consideration of the remaining items to give the severity score. However, unlike this, the weight for each item can be set in advance, and the severity score can be calculated by applying the weight for each item to the primary severity score according to the lesion progression status. In other words, the severity score is calculated primarily according to the lesion progression status, but the severity score is calculated by multiplying this by the weight set for each item. For example, when 3 points are calculated according to the lesion progression level, if the lesion area corresponds to the patient's main complaint, it can be multiplied by the preset weight of 2, and at the same time, if it corresponds to the secondary lesion, it can be multiplied by the weight of 1.3 to give a total of 3 X 2 X 1.3 = 7.8 points.

[0108] In this way, the lesion severity score for the lesion area can be calculated by adding or subtracting a preset additional score for each item to the primary severity score according to the lesion progression status or by multiplying the weight for each item.

[0109] Referring back to FIG. 2, the dental treatment planning device (1) determines treatment priorities for multiple lesion areas based on the severity of each lesion area calculated as described above (S140). Treatment priorities may be assigned in ascending order of the calculated lesion severity scores. Note that the lesion severity scores may be displayed via the display unit (20), but they may also be internally utilized solely for determining treatment priorities.

[0110] A dental treatment planning device (1) visualizes information about each lesion area detected in a medical image, such as lesion information, lesion severity information, and treatment priority information, and provides the information to a user through a display unit (20) (S150). Here, the lesion information may include at least one of the tooth number where the lesion area exists, the lesion detailed location, the lesion type, the lesion progression stage, and the lesion area detection reliability.

[0111] Fig. 11 shows an example of a screen in which information regarding a lesion area detected through a dental treatment planning device (1) according to an embodiment of the present invention is visualized and provided on a medical image.

[0112] Referring to Fig. 11, for the caries area of ​​tooth 44 (1101), 3 points are given according to the C3 level, and an additional 10 points are given as the patient's main complaint, resulting in a total of 13 points calculated as severity, and for the periodontitis area of ​​tooth 46 (1103) replaced with an implant, 3 points are given primarily according to the lesion progression due to severe periodontitis, and an additional 8 points are given as a secondary lesion related to the implant, which is the existing restoration object, resulting in a total of 11 points calculated as lesion severity. In addition, it is assumed that the periapical inflammation area of ​​tooth 45 (1105) is given a score of 3 points primarily based on the lesion progression status, and an additional score of 8 points is given as a secondary lesion related to the crown, which is the existing restoration object, for a total of 11 points, the periapical inflammation area of ​​tooth 34 (1107) is given a score of 3 points based on the lesion progression status, and the caries area of ​​tooth 13 (1109) is given a score of 2 points based on the caries progression level C2, as a lesion severity score.

[0113] Accordingly, tooth number 44 (1101), which has the highest lesion severity score, has the highest treatment priority, and tooth number 13 (1109) has the lowest treatment priority.

[0114] Referring to Figure 11, it can be seen that the type and detection confidence (%) of each lesion area are overlaid on the medical image and the treatment priority determined according to the lesion severity is displayed as TOP1, TOP2, and TOP3.

[0115] Meanwhile, the dental treatment planning device (1) sets multiple sections according to the lesion severity score, and displays an indicator (1111) having a predetermined color or pattern for each severity score section as a mark indicating the lesion severity, thereby allowing the user to intuitively identify and compare the lesion severity. In Fig. 11, it can be seen that an indicator (1111) having a diagonal grid pattern is displayed for the lesion area of ​​teeth No. 44, No. 45, No. 46, and No. 34 (1101, 1103, 1105, and 1107) having relatively high lesion severity scores according to the lesion severity score, and an indicator (1111) having a checkerboard pattern is displayed for the lesion area of ​​tooth No. 13 (1109) having relatively low lesion severity. This is an example of an indicator (1111), and of course, it can be displayed in different colors, such as red when the lesion severity is high, yellow, green, etc., as the severity decreases.

[0116] In addition, the dental treatment planning device (1) can display lesion information and severity information by overlaying them on a medical image, as shown in Fig. 11, but can also display the above information by inserting it on a patient's electronic chart.

[0117] Fig. 12 shows an example of a screen in which information regarding a lesion area detected through a dental treatment planning device (1) according to an embodiment of the present invention is visualized and provided on an electronic chart.

[0118] Referring to Fig. 12, it can be confirmed that the tooth number where the lesion area was detected, the lesion type, the lesion site, the lesion progression stage, an indicator according to the lesion severity, and the lesion detection reliability are displayed in the medical history area (1201) of the electronic chart.

[0119] In addition, the dental treatment planning device (1) determines a treatment plan for treatment of each lesion area by reflecting the treatment priority determined based on the lesion severity (S160).

[0120] To this end, the dental treatment planning device (1) stores information on treatment items required for each lesion type and, based on this, can determine a treatment plan by applying treatment items corresponding to each detected lesion type. The treatment plan can be displayed on the electronic chart of the patient management system.

[0121] Fig. 13 shows an example of an electronic chart screen displaying a treatment plan determined through a dental treatment planning device (1) according to an embodiment of the present invention.

[0122] Referring to Fig. 13, it can be confirmed that the treatment plan is displayed in a list format with the necessary treatment items corresponding to each lesion area in the left area (1301) of the electronic chart. At this time, as shown in Fig. 13, the treatment plans can be listed in order of treatment priority. For reference, the user can add a new treatment plan or modify a treatment plan automatically determined through the dental treatment planning device (1) by selecting the treatment item button displayed in the right area (1303) of the electronic chart.

[0123] Next, the dental treatment planning device (1) calculates and provides the expected treatment cost by reflecting the determined treatment plan (S170). The dental treatment planning device (1) stores information on categories for treatment items and the cost of each treatment item included in the category, and can calculate the expected treatment cost based on the cost of the treatment item corresponding to the treatment plan.

[0124] However, a single treatment item may include multiple sub-treatment items. For example, a crown treatment item may be categorized into sub-items such as zirconia, gold, metal, PFM, and ceramic crowns depending on the crown material. Since the treatment cost varies for each of the above sub-items, the treatment cost may vary depending on the sub-item selected when calculating the expected treatment cost. Taking this into account, the dental treatment planning device (1) can calculate and provide an expected treatment cost that the patient can actually afford within the range of available treatment costs.

[0125] To this end, the dental treatment planning device (1) receives input regarding the patient's available treatment cost through the user input unit (10), selects a specific detailed treatment item that can satisfy the range of the patient's available treatment cost among multiple detailed treatment items belonging to each treatment item according to the treatment plan, and calculates the treatment cost.

[0126] Fig. 14 shows an example of a screen where the expected treatment cost according to the treatment plan is calculated and displayed through a dental treatment planning device (1) according to an embodiment of the present invention.

[0127] Referring to FIG. 14, a treatment plan is displayed on the left area (1401) of the screen, a treatment item list listing multiple treatment items and cost information according to treatment item category is displayed on the central area (1403) of the screen, and an example is shown in which the estimated treatment cost calculated by selecting specific detailed treatment items according to the treatment plan is displayed on the right area (1405) of the screen. As shown in FIG. 14, the treatment cost can be displayed by calculating the amount for each treatment item and the total sum.

[0128] Additionally, the dental treatment planning device (1) can configure and provide a consultation screen so that the user can conduct patient consultation regarding the treatment plan determined based on the severity of the lesion (S180). For reference, although step S180 is depicted in FIG. 2 as being performed after step S170, this is merely an example, and the order of the two steps may be interchanged.

[0129] In order to configure the consultation screen, the dental treatment planning device (1) can extract consultation data regarding the treatment plan from a database where consultation data is stored and configure the consultation screen based on this.

[0130] Fig. 15 is an example of a patient consultation screen configured through a dental treatment planning device (1) according to an embodiment of the present invention.

[0131] Referring to Figure 15, the treatment plan determined for the patient is displayed in the upper left area (1501) of the screen, and consultation data related to the treatment plan is extracted and displayed in the right area (1503) of the screen. The user can proceed with patient consultation using the extracted consultation data.

[0132] So far, a method of planning dental treatment based on lesion severity has been described with reference to FIG. 2. However, it is to be understood that the order of the steps in FIG. 2 may be changed or new steps may be added depending on the situation, as described above.

[0133] As described above, the dental treatment planning device (1) and method according to the present invention can accurately and intuitively understand a patient's condition by providing visualized information such as the severity of each lesion area detected in a medical image and the type of lesion. Furthermore, by automatically establishing a treatment plan based on the lesion severity and establishing and providing a treatment cost plan within the patient's available treatment costs, a more reasonable and optimized treatment solution can be provided to the patient.

[0134] The present invention can also be implemented in various computer-readable recording media, such as a magnetic storage medium, an optical reading medium, or a digital storage medium, in which a computer program for performing a dental treatment planning method according to the present invention is stored when executed on a computer.

[0135] While several embodiments of the present invention have been described so far, those skilled in the art will appreciate that modifications or substitutions of certain embodiments may be made without departing from the technical spirit of the present invention. Therefore, the scope of protection of the present invention should be deemed to encompass the inventions described in the claims and their equivalents.

Claims

1. In a dental treatment planning method in which each step is performed through a computing device, Step of loading the patient's medical images; A step of detecting at least one lesion area in the above medical image; A step of calculating the severity of each lesion area based on the lesion progression status of each lesion area; and A dental treatment planning method characterized by including a step of visualizing and displaying lesion information and severity information of each lesion area.

2. In paragraph 1, A step of determining treatment priorities for a plurality of lesion areas based on the severity of each lesion area; and A dental treatment planning method characterized by further comprising a step of determining a treatment plan for treatment of the lesion area by reflecting the above treatment priority.

3. In paragraph 2, A dental treatment planning method characterized by further including a step of calculating an expected treatment cost by reflecting the determined treatment plan.

4. In paragraph 3, Further comprising the step of receiving user input regarding the patient's available treatment costs, The steps for calculating the above expected treatment costs are: A dental treatment planning method characterized in that the expected treatment cost is calculated by selecting a treatment item that can satisfy the range of available treatment costs among multiple treatment items according to the above treatment plan.

5. In paragraph 2, A dental treatment planning method characterized by further including a step of extracting consultation data regarding the treatment plan from among the stored consultation data and configuring a consultation screen for patient consultation.

6. In paragraph 1, Further comprising a step of detecting a restored object as a result of a previous treatment in the above medical image, The steps for calculating the severity of each lesion area above are: A dental treatment planning method characterized in that the severity is calculated by reflecting whether the lesion area corresponds to a secondary lesion associated with the restored object.

7. In paragraph 1, The above lesion area includes the tooth loss area where the tooth is lost, The steps for calculating the severity of each lesion area above are: A dental treatment planning method characterized in that the severity is calculated by reflecting information on the state in which the opposing teeth of the above tooth loss area are protruded relative to the occlusal plane and the state in which the adjacent teeth of the above tooth loss area are tilted.

8. In paragraph 1, The steps for calculating the severity of each lesion area above are: A dental treatment planning method characterized by calculating the severity level by reflecting information on the patient's main complaint.

9. In paragraph 1, The steps for calculating the severity of each lesion area above are: A dental treatment planning method characterized in that the severity is calculated by reflecting information on the influence of the lesion area on adjacent anatomical structures.

10. In paragraph 1, The steps for calculating the severity of each lesion area above are: A dental treatment planning method characterized in that the severity is calculated by reflecting the patient's age information, information on the tooth in which the lesion area exists among the baby teeth or permanent teeth, and the presence or absence of a permanent tooth succeeding the baby teeth.

11. In paragraph 1, A dental treatment planning method further comprising a step of displaying the lesion information and the severity information on a patient's electronic chart.

12. In paragraph 1, A dental treatment planning method, characterized in that the lesion information includes at least one of a tooth number where the lesion area exists, a lesion detailed location, a lesion type, a lesion progression stage, or a detection reliability of the lesion area.

13. In paragraph 1, The step of detecting the above lesion area is: A dental treatment planning method characterized in that it is performed by segmenting multiple anatomical structures for each tooth in the above medical image and applying an artificial intelligence model learned to identify the location and lesion type of the lesion area existing in each tooth.

14. In paragraph 13, The above artificial intelligence model is, A dental treatment planning method, characterized by comprising: a first artificial neural network trained to segment the plurality of anatomical structures in the medical image; a second artificial neural network trained to identify the lesion area existing in each tooth based on the result of segmentation processing through the first artificial neural network; and a third artificial neural network trained to output data regarding the location, lesion type, and lesion detection reliability of the lesion area identified through the second artificial neural network.

15. A computer-readable recording medium having recorded thereon a program for performing a dental treatment planning method according to any one of paragraphs 1 to 14.

16. In a dental treatment planning device that performs dental treatment planning, Contains a processor, The above processor, Loading the patient's medical images, Detecting at least one lesion area in the above medical image, Based on the lesion progression status of each lesion area above, the severity of each lesion area above is calculated, A dental treatment planning device characterized by performing processing that visualizes and displays lesion information and severity information of each lesion area.

Images