Oral panoramic film decayed tooth depth recognition method based on deep learning

A deep learning and stomatological technology, applied in the field of artificial intelligence medical image processing, can solve the problems of not meeting the recognition requirements of caries depth, not finding caries deep learning recognition methods, reducing the accuracy of diagnosis results, etc., to improve diagnosis and treatment capabilities and work Efficiency, convenient retrieval and push, and the effect of reducing missed diagnosis and misdiagnosis rate

Inactive Publication Date: 2020-10-16
ZHEJIANG UNIV
View PDF5 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Long-term viewing of image data by medical staff can easily lead to visual fatigue, which reduces the accuracy of diagnostic results to a certain extent
At present, a variety of systems have been developed, such as auxiliary image reading robots, tumor diagnosis robots, etc., but because dental caries is a relatively special type of disease, its concept and method of diagnosis are somewhat different from those of ordinary diseases. There are a large number of dental caries, and the range of dental caries is small. The current commonly used machine learning methods cannot meet the identification requirements of dental caries depth. There is no deep learning identification method specifically for dental caries.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Oral panoramic film decayed tooth depth recognition method based on deep learning
  • Oral panoramic film decayed tooth depth recognition method based on deep learning
  • Oral panoramic film decayed tooth depth recognition method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0046] see figure 1 , the caries depth recognition method of oral panorama film based on deep learning proposed by the present invention is realized by the following steps:

[0047] 1. First collect image data, perform data preprocessing, uniformly number according to a certain order, use MicroDicom to adjust image brightness and contrast, convert to BMP format, and uniformly normalize, divide into training set and training set according to the ratio of 7:3 test set;

[0048]2. Secondly, segment the preprocessed image and extract the ROI (Region Of Interest, ROI) region of interest, such as image 3 and Figure 4 as shown, image 3 The part in the box has shallow caries (A), Figure 4 Middle B box is medium caries, C box is deep caries. Image segmentation adopts the method of maximum inter-class variance (OTSU algorithm). According to the gray feature of the image, the image is divided into two parts, the background and the target. The difference is also obvious. The OT...

Embodiment 2

[0052] (1) Data acquisition module

[0053] The data were derived from oral panoramas in DICOM format exported from the radiology database of the hospital. In order to ensure the accuracy of model training, image data that are too blurry and low in saturation are preliminarily screened out.

[0054] (2) Data preprocessing module

[0055] Adjust the brightness and contrast of the collected images, convert the DICOM format images into BMP format through MicroDicom, normalize them to the same size, and number them one by one in order. The normalization formula is:

[0056] where x min and x max represent the minimum and maximum values, respectively. The Lableme professional image labeling tool is used to uniformly label the caries area and its depth on the converted image. The labeling principle is based on the clinical manifestations of the caries depth ( figure 1 ), divide the data into training set and test set according to the ratio of 7:3.

[0057] (3) ROI target ar...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses an oral panoramic film decayed tooth depth recognition method based on deep learning. The method comprises the following 5 steps: preprocessing data; segmenting the image; constructing a model; performing feature extraction; and performing classification and identification. According to the method, the convolutional neural network is adopted to identify the decayed teeth, automatic identification of the decayed teeth in the oral cavity panoramic film is realized, and result export of automatic decayed tooth depth identification can be realized. Visual fatigue easily caused by long-term viewing of image data by medical staff is solved, diagnosis and treatment capacity and working efficiency are improved in an assisted mode, medical resource requirements are reduced,and medical efficiency is improved.

Description

technical field [0001] The invention belongs to the field of artificial intelligence medical image processing, and more specifically relates to a caries depth recognition method for oral panorama films based on deep learning. Background technique [0002] Dental caries is a progressive lesion of dental hard tissue caused by the compound action of various factors in the oral cavity. It is the main common disease of the oral cavity and one of the most common diseases of human beings. The imaging recognition of dental caries is mainly oral panorama. Due to the increasing demand for oral health, more and more people go to hospitals or clinics for oral consultation or treatment. However, due to the large number of patients and the limited time for diagnosis and treatment, sometimes doctors have to give priority to symptomatic teeth while ignoring other potential dental caries. Moreover, it is difficult for medical staff to judge the degree of dental caries during oral examination...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06T7/00G06T7/136G06K9/32G06N3/04G06N3/08A61B5/00
CPCG06T7/0012G06T7/136G06N3/08A61B5/4547A61B5/7203A61B5/7267G06T2207/30036G06V10/25G06N3/045
Inventor 朱海华朱赴东梁蒙蒙黄超强连璐雅陈庆光黄俊超
Owner ZHEJIANG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap