KR-20260062882-A - METHOD AND ORAL SCANNER APPARATUS FOR ANALYZING INTERNAL TOOTH DEFECTS USING A VISIBLE-LIGHT STRUCTURED-LIGHT-BASED AI MODEL

KR20260062882AKR 20260062882 AKR20260062882 AKR 20260062882AKR-20260062882-A

Abstract

The present invention relates to a method for analyzing internal tooth defects based on visible structured light, performed by an oral scanner and a display device communicating with the oral scanner. The present invention may include the steps of: projecting a visible light pattern having different spatial frequencies onto a tooth surface; acquiring a plurality of images of the visible light pattern reflected from the tooth; calculating a plurality of phase maps corresponding to each combination of wavelength and spatial frequency based on the plurality of images; analyzing contrast features on the frequency axis and wavelength axis between the plurality of phase maps to separate surface components and deep estimated components; processing a plurality of input data including at least one of the separated component data, a color map, a color difference map, an elevation map, a gradient map, a frequency band energy map, and a reliability map derived in the phase map calculation step as input to an artificial intelligence model; calculating a defect probability or defect distribution of internal tooth defects through the artificial intelligence model; and aligning the calculated defect probability or defect distribution of internal tooth defects with a three-dimensional oral model and visualizing the calculated defect probability or defect distribution of internal tooth defects by overlaying it on the three-dimensional oral model.

Inventors

전승현
김경국
최윤수

Assignees

아크리얼 주식회사

Dates

Publication Date: 20260507
Application Date: 20251029
Priority Date: 20241029

Claims (12)

A method for analyzing internal tooth defects based on visible light structured light, performed by an oral scanner and a display device communicating with said oral scanner, wherein A step of projecting a visible light pattern having different spatial frequencies onto the tooth surface; A step of acquiring a plurality of images of the above visible light pattern reflected from the teeth; A step of calculating a plurality of phase maps corresponding to each combination of wavelength and spatial frequency based on the plurality of images above; A step of analyzing contrast features in the frequency and wavelength axes between the plurality of phase maps to separate surface components and deep estimation components; A step of processing a plurality of input data, including at least one of the separated component data, color map, color difference map, elevation map, gradient map, frequency band energy map, and reliability map derived in the phase map calculation step, as input to an artificial intelligence model; A step of calculating the defect probability or defect distribution of internal tooth defects through the above artificial intelligence model; and The method comprises the step of aligning the defect probability or defect distribution of the internal tooth defect calculated above with a three-dimensional oral model, and visualizing the defect probability or defect distribution of the internal tooth defect calculated above by overlaying it on the three-dimensional oral model. Method for analyzing internal tooth defects using a visible light structured light-based oral scanner.
In claim 1, The above visible light pattern is Comprising at least one of a fringe pattern, a speckle pattern, a checkerboard pattern, a random pattern, a polarized modulated pattern, and a Gray code pattern, Method for analyzing internal tooth defects using a visible light structured light-based oral scanner.
In claim 2, The above visible light pattern is, It includes different spatial frequencies of 0.01 cycles/mm or more and 2.0 cycles/mm or less, and Projected using a phase shift method of multiple stages having different phase shift amounts, Characterized by being controlled according to a schedule table that operates the visible light wavelength range using a time division method or a light intensity control method, Method for analyzing internal tooth defects using a visible light structured light-based oral scanner.
In claim 1, The above contrast features are, Characterized by including at least one of the standard deviation of the phase difference between the plurality of phase maps, the directional component of the phase slope, the ratio of low-frequency band energy to high-frequency band energy, and the attenuation rate between wavelengths. Method for analyzing internal tooth defects using a visible light structured light-based oral scanner.
In claim 1, The above artificial intelligence model is, Learning input feature data of multiple channels extracted from a plurality of input maps including the phase map, the color map, the color difference map, the elevation map, the gradient map, the frequency band energy map, and the confidence map, and It is trained using a primary task, defect segmentation or defect probabilistic regression, and a secondary task, a combination of losses including surface reconstruction loss, phase consistency loss, and chromatic consistency loss, and Characterized by being learned by utilizing together weak maps and synthetic data generated from clinical reading results or cross-match rates between multiple cycles, Method for analyzing internal tooth defects using a visible light structured light-based oral scanner.
In claim 1, The step of calculating the probability of internal tooth defects or the defect distribution described above is, It includes an additional step of entering a separate sub-mode, In the above diagnostic submode, The inference of the above artificial intelligence model is performed within a time less than or equal to a preset inference delay limit, and The result of the inference by the above artificial intelligence model visualizes the risk level through a color heatmap, and Characterized by providing a user interface (UI) that displays at least one of the above-mentioned color heatmap, reliability indicator, and additional shooting recommendation information together. Method for analyzing internal tooth defects using a visible light structured light-based oral scanner.
A system comprising an oral scanner and a display device communicating with said oral scanner, wherein the oral scanner and the display device each comprise at least one processor and a memory storing at least one instruction executed by the processor, and said at least one instruction Projecting visible light patterns having different spatial frequencies onto the tooth surface; Acquiring a plurality of images of the above visible light pattern reflected from the teeth; Based on the above plurality of images, a plurality of phase maps corresponding to each combination of wavelength and spatial frequency are calculated; By analyzing the contrast features on the frequency and wavelength axes between the above plurality of phase maps, surface components and deep estimation components are separated; A plurality of input data, including at least one of the separated component data, color map, color difference map, elevation map, gradient map, frequency band energy map, and reliability map derived in the phase map calculation step, are processed as inputs to an artificial intelligence model; Calculate the defect probability or defect distribution of internal tooth defects through the above artificial intelligence model; and Set to align the defect probability or defect distribution of the internal tooth defect calculated above with a 3D oral model, and to visualize the defect probability or defect distribution of the internal tooth defect calculated above by overlaying it on the 3D oral model, Visible light structured light-based internal tooth defect analysis system.
In claim 7, The above visible light pattern is Comprising at least one of a fringe pattern, a speckle pattern, a checkerboard pattern, a random pattern, a polarized modulated pattern, and a Gray code pattern, Visible light structured light-based internal tooth defect analysis system.
In claim 8, The above visible light pattern is, It includes different spatial frequencies of 0.01 cycles/mm or more and 2.0 cycles/mm or less, and Projected using a phase shift method of multiple stages having different phase shift amounts, Characterized by being controlled according to a schedule table that operates the visible light wavelength range using a time division method or a light intensity control method, Visible light structured light-based internal tooth defect analysis system.
In claim 7, The above contrast features are, Characterized by including at least one of the standard deviation of the phase difference between the plurality of phase maps, the directional component of the phase slope, the ratio of low-frequency band energy to high-frequency band energy, and the attenuation rate between wavelengths. Visible light structured light-based internal tooth defect analysis system.
In claim 7, The above artificial intelligence model is, Learning input feature data of multiple channels extracted from a plurality of input maps including the phase map, the color contrast map, the color difference map, the elevation map, the gradient map, the frequency band energy map, and the confidence map, and It is trained using a primary task, defect segmentation or defect probabilistic regression, and a secondary task, a combination of losses including surface reconstruction loss, phase consistency loss, and chromatic consistency loss, and Characterized by being learned by utilizing together weak maps and synthetic data generated from clinical reading results or cross-match rates between multiple cycles, Visible light structured light-based internal tooth defect analysis system.
In claim 7, The step of calculating the probability of internal tooth defects or the defect distribution described above is, It includes an additional step of entering a separate sub-mode, In the above diagnostic submode, The inference of the above artificial intelligence model is performed within a time less than or equal to a preset inference delay limit, and The result of the inference by the above artificial intelligence model visualizes the risk level through a color heatmap, and Characterized by providing a user interface (UI) that displays at least one of the above-mentioned color heatmap, reliability indicator, and additional shooting recommendation information together. Visible light structured light-based internal tooth defect analysis system.

Description

Method and Oral Scanner Apparatus for Analyzing Internal Tooth Defects Using a Visible-Light Structured-Light-Based AI Model This specification relates to a technology for diagnosing the condition of teeth by optically scanning intraoral structures, and more specifically, to a visible light structured light-based method and system for analyzing internal tooth defects that utilizes structured light in the visible light range to analyze images reflected from the tooth surface and calculates and visualizes the probability or distribution of defects within the tooth through an artificial intelligence model. Various optical 3D scanners are used in the dental treatment and prosthetic fabrication process to accurately measure the oral structure of patients. These oral scanners project structured or patterned light in the visible light band onto the tooth surface and analyze the reflected image to precisely reconstruct the shape of the surface, thereby enabling the rapid and non-contact acquisition of external tooth shape information. However, conventional oral scanners are primarily focused on measuring the external shape of teeth and have limitations in identifying deep defects such as microcracks within the enamel, early caries, and adhesive delamination. These internal defects are difficult to detect when relying solely on light reflected from the tooth surface, and methods intended to complement this, such as near-infrared transmission imaging, fluorescence analysis, or optical coherence tomography (OCT), are difficult to utilize in clinical settings due to the complexity of the equipment or slow imaging speeds. Accordingly, there is a demand for technology that can simultaneously analyze surface and deep information of teeth using image data acquired from existing visible light structured light-based oral scanners without additional expensive equipment, and efficiently identify internal defects through artificial intelligence models. FIG. 1 is a system diagram including an oral scanner device and a display device for analyzing internal tooth defects using a visible light structured light-based AI model according to one embodiment of the present specification. FIG. 2 is a flowchart illustrating a method for analyzing internal tooth defects using a visible light structured light-based AI model according to one embodiment of the present specification. FIG. 3 is a diagram showing a screen that displays the inference results of an artificial intelligence model according to one embodiment of the present specification by overlaying them as a color heatmap on a three-dimensional oral model. As the present specification is susceptible to various modifications and may have various embodiments, specific embodiments are illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present specification to specific embodiments, and it should be understood that it includes all modifications, equivalents, and substitutions that fall within the spirit and scope of the present specification. Similar reference numerals have been used for similar components in the description of each drawing. Terms such as first, second, A, B, etc., may be used to describe various components, but said components should not be limited by said terms. These terms are used solely for the purpose of distinguishing one component from another. For example, without departing from the scope of the rights of this specification, the first component may be named the second component, and similarly, the second component may be named the first component. The term "and/or" includes a combination of a plurality of related described items or any of a plurality of related described items. When it is stated that one component is "connected" or "connected" to another component, it should be understood that while it may be directly connected or connected to that other component, there may also be other components in between. On the other hand, when it is stated that one component is "directly connected" or "directly connected" to another component, it should be understood that there are no other components in between. The terms used in this application are used merely to describe specific embodiments and are not intended to limit this specification. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this application, terms such as "comprising" or "having" are intended to indicate the presence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof. Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which this sp