Dental medical record device and dental medical record method thereof

Patent metadata

Field	Value
Publication number	US-12205689-B2
Application number	US-202117792271-A
Country	US
Kind code	B2
Filing date	Jan 7, 2021
Priority date	Jan 13, 2020
Publication date	Jan 21, 2025
Grant date	Jan 21, 2025

Abstract

A dental medical record device and a dental medical record method, in which: an image, such as a panoramic photo, a scan image, and a camera image of a patient's oral cavity, is received via artificial intelligence, and charting is performed using the artificial intelligence; and medical records for a treatment area can be read in association with a chart by clicking the treatment area in the image.

First claim

The invention claimed is:

1. A dental medical record method comprising:
receiving with a dental medical record device an image of a patient's oral cavity;
reading with a controller of the dental medical record device the image for a patient's oral cavity to mark the patient's oral cavity condition on a prognosis chart, using a text format, symbols, or both text and symbols; and
displaying the image and the prognosis chart on an output unit of the dental medical record device;
wherein the controller reads the image through a deep learning model including a feature extraction part that extracts features for the image by performing convolution and subsampling on the image and an output part that marks the prognosis chart with the patient's oral cavity condition with at least one of text and symbols using the extracted features, and
wherein in the displaying the image and the prognosis chart, the deep learning model is continuously trained with continuously collected training images and accumulated prognosis chart data to read the image wherein the collected training images include a guide chip with a different gray scale or a guide sample including a color number and a color corresponding to the color number, and the oral condition of the patient is indicated on the prognosis chart in text format or by combining the text and the symbol;
wherein in the displaying the image and the prognosis chart, when a specific tooth is clicked on the image, the color of the specific tooth presented by the deep learning model learned on the image is displayed as the color number and color of the guide sample.

2. The dental medical record method of claim 1 , wherein the image is one of a panoramic photo, a scan image, or a camera image obtained by one or two or more of radiography, oral camera, scanning device, and occlusion check data for the patient's oral cavity, and
wherein the patient's oral cavity condition is at least one of a dental formula, a dental condition, or a surrounding structure of teeth.

3. The dental medical record method of claim 1 , wherein the deep learning model automatically learns features for input values by being trained with a large amount of data from a deep neural network composed of a multi-layered network and train the multi-layered network to minimize errors in prediction accuracy and read the image.

4. The dental medical record method of claim 1 , wherein in marking the prognosis chart, if a specific treatment site of the image is selected, information about the corresponding treatment site is displayed, and if the image of the corresponding treatment site is selected, a treatment summary memo window is displayed, and if the treatment summary memo window is selected, the chart for the corresponding treatment site is output.

5. A dental medical record device, comprising:
an input unit receiving an image for a patient's oral cavity;
a controller reading the image for the patient's oral cavity to determine the patient's oral cavity condition and marking the patient's oral cavity condition on a prognosis chart, using a text format, symbols, or both text and symbols; and
an output unit displaying the image and the prognosis chart;
wherein the controller reads the image through a deep learning model including a feature extraction part that extracts features for the image by performing convolution and subsampling on the image and an output part that marks the prognosis chart with the patient's oral cavity condition with at least one of text and symbols using the extracted features, and
wherein in the displaying the image and the prognosis chart, the deep learning model is continuously trained with continuously collected training images and accumulated prognosis chart data to read the image wherein the collected training images include a guide chip with a different gray scale or a guide sample including a color number and a color corresponding to the color number, and the oral condition of the patient is indicated on the prognosis chart by combining the text and the symbol;
wherein when a specific tooth is clicked on the image, the color of the specific tooth presented by the deep learning model learned on the image is displayed as the color number and color of the guide sample.

6. The dental medical record device of claim 5 , wherein the image is one of a panoramic photo, a scan image, or a camera image obtained by one or two or more of radiography, oral camera, scanning device, and occlusion check data for the patient's oral cavity, and
wherein the patient's oral cavity condition is at least one of a dental formula, a dental condition, or a surrounding structure of teeth.

7. The dental medical record device of claim 5 , wherein the deep learning model automatically learns features for input values by being trained with a large amount of data from a deep neural network composed of a multi-layered network and train the multi-layered network to minimize errors in prediction accuracy and read the image.

8. The dental medical record device of claim 5 , wherein if a specific treatment site of the image is selected, the display unit displays information about the corresponding treatment site, and if the image of the corresponding treatment site is selected, displays a treatment summary memo window, and if the treatment summary memo window is selected, outputs the chart for the corresponding treatment site.

9. A non-transitory computer-readable storage medium storing instructions that, when executed by one or more controllers, enable the one or more controllers to perform a dental medical record method, the instructions comprising:
receiving with a dental medical record device an image of a patient's oral cavity;
reading with the one or more controllers of the dental medical record device the image for a patient's oral cavity to mark the patient's oral cavity condition on a prognosis chart, using a text format, symbols, or both text and symbols; and
displaying the image and the prognosis chart on an output unit of the dental medical record device;
wherein the instructions operate the one or more controllers to read the image through a deep learning model including a feature extraction part that extracts features for the image by performing convolution and subsampling on the image and an output part that marks the prognosis chart with the patient's oral cavity condition with text and symbols using the extracted features;
wherein when a specific tooth is clicked on the image, the color of the specific tooth presented by the deep learning model learned on the image is displayed as the color number and color of the guide sample.

Assignees

• Catholic Univ Korea Ind Academic Cooperation Foundation
• The Catholic Univ Of Korea Industry—Academic Cooperation Foundation

Inventors

• Lee Sang Hwa

Classifications

• G06T2207/30036

  Dental; Teeth · CPC title

• G06T2207/20084

  Artificial neural networks [ANN] · CPC title

• G06T2207/20081

  Training; Learning · CPC title

• G06T7/0012

  Biomedical image inspection · CPC title

• G06V10/40

  Extraction of image or video features · CPC title