CN111914931A - Method for establishing oral diagnosis neural network model and oral diagnosis method - Google Patents

Method for establishing oral diagnosis neural network model and oral diagnosis method Download PDF

Info

Publication number
CN111914931A
CN111914931A CN202010760132.XA CN202010760132A CN111914931A CN 111914931 A CN111914931 A CN 111914931A CN 202010760132 A CN202010760132 A CN 202010760132A CN 111914931 A CN111914931 A CN 111914931A
Authority
CN
China
Prior art keywords
oral
neural network
network model
image
diagnosis
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202010760132.XA
Other languages
Chinese (zh)
Inventor
宋锦璘
柴召午
张超
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Stomatological Hospital of Chongqing Medical University
Original Assignee
Stomatological Hospital of Chongqing Medical University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Stomatological Hospital of Chongqing Medical University filed Critical Stomatological Hospital of Chongqing Medical University
Priority to CN202010760132.XA priority Critical patent/CN111914931A/en
Publication of CN111914931A publication Critical patent/CN111914931A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61C—DENTISTRY; APPARATUS OR METHODS FOR ORAL OR DENTAL HYGIENE
    • A61C19/00—Dental auxiliary appliances
    • A61C19/04—Measuring instruments specially adapted for dentistry
    • A61C19/05—Measuring instruments specially adapted for dentistry for determining occlusion
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Biomedical Technology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Artificial Intelligence (AREA)
  • Biophysics (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computational Linguistics (AREA)
  • Software Systems (AREA)
  • Medical Informatics (AREA)
  • Epidemiology (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Dentistry (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Animal Behavior & Ethology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Veterinary Medicine (AREA)
  • Databases & Information Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Pathology (AREA)
  • Primary Health Care (AREA)
  • Image Analysis (AREA)

Abstract

The invention is suitable for the technical field of oral medicine, and provides a method for establishing an oral diagnosis neural network model and an oral diagnosis method, wherein the method for establishing the oral diagnosis neural network model comprises the following steps: acquiring a training set containing classified patient oral cavity images; acquiring a preset oral defect image from a training set to construct a sample set; applying a loss function and an activation function on the basis of the structure of the VGG16 convolutional neural network model; and (5) putting the training set and the sample set into a VGG16 neural network for training to generate an oral diagnosis neural network model. The invention can realize the image recognition and the preliminary diagnosis of the oral cavity defects such as malocclusion and the like, thereby simplifying the examination steps, shortening the examination time and improving the overall diagnosis efficiency of the patient for treatment.

Description

Method for establishing oral diagnosis neural network model and oral diagnosis method
Technical Field
The invention relates to the technical field of oral medicine, in particular to a method for establishing an oral diagnosis neural network model and an oral diagnosis method.
Background
The cardinality of the malocclusion malformation patients in China is large, and increasingly situations exist, wherein the malocclusion malformation of children is prevented and corrected by an oral treatment method in an early stage according to the causes, mechanisms, occurrence and development processes of malocclusion, the correction difficulty is small at the moment, and the beauty and the perfect coordinated development of functions of craniofacial malocclusion can be achieved after correction.
In the existing oral cavity diagnosis process, the oral cavity of a patient is mainly explored into the oral cavity of the patient through a stomatoscope to detect the oral cavity condition of the patient, wherein the stomatoscope is divided into a plurality of types, each stomatoscope has only a single function, if a plurality of tools are alternately used, the examination steps are obviously complicated, the examination time is prolonged, and the diagnosis efficiency for the patient is low.
Disclosure of Invention
The invention mainly aims to provide a method for establishing an oral diagnosis neural network model and an oral diagnosis method, and aims to solve the problems that in the prior art, in the oral diagnosis process, the number of detection steps is large, the detection time is long, and the diagnosis efficiency for patients is low.
In order to achieve the above object, a first aspect of embodiments of the present invention provides a method for establishing an oral diagnostic neural network model, including:
acquiring a training set containing classified patient oral cavity images;
acquiring a preset oral defect image construction sample set from the training set;
applying a loss function and an activation function on the basis of the structure of the VGG16 convolutional neural network model;
and putting the training set and the sample set into the VGG16 neural network for training to generate an oral cavity diagnosis neural network model.
With reference to the first aspect of the present invention, in the first embodiment of the present invention, before applying the loss function and the activation function based on the VGG16 convolutional neural network model structure, the method includes:
and performing incremental processing on the classified patient oral cavity image and the preset oral cavity defect image.
With reference to the first embodiment of the first aspect of the present invention, in a second embodiment of the present invention, performing incremental processing on the classified patient oral cavity image and the preset oral cavity defect image includes:
and executing at least one processing mode of rotating a preset angle, translating a preset amplitude, vertically mirroring and horizontally mirroring on the classified patient oral cavity image and the preset defect image.
With reference to the first aspect of the present invention, in a third embodiment of the present invention, the calculation formula of the loss function is:
Figure BDA0002612839310000021
wherein y represents the true probability of the occurrence of the preset oral defect image,
Figure BDA0002612839310000022
representing the predicted probability of occurrence of the preset oral defect image, M representing the number of classified patient oral cavity images, N representing the number of classification labels in each classified patient oral cavity image, i representing the ith classification label, j representing the jth classified patient oral cavity image, yijAnd
Figure BDA0002612839310000023
an ith label representing a jth classified patient oral image.
With reference to the first aspect of the present invention, in a fourth embodiment of the present invention, the training set and the sample set are put into the VGG16 convolutional neural network model for training to generate an oral diagnosis neural network model, and the method includes:
putting the training set into a VGG16 convolutional neural network model for K times of iterative training;
the difference between the nth iteration training result and the sample set is obtained through the loss function degree;
wherein K is a positive integer, and n is a positive integer less than or equal to K;
calculating a minimized loss function of the nth iterative training result, and adjusting the learning rate of the (n + 1) th iterative training according to the minimized loss function;
and generating an oral diagnosis neural network model according to the training results of the K times of iterative training.
A second aspect of the embodiments of the present invention provides an oral cavity diagnosis method, including:
acquiring an oral cavity image of a patient to be diagnosed;
and carrying out preliminary diagnosis of the preset oral defect on the oral cavity image of the patient to be diagnosed by using the oral cavity diagnosis neural network model.
In a third aspect, an embodiment of the present invention provides an apparatus for building an oral diagnostic neural network model, including:
the training set acquisition module is used for acquiring a training set containing classified patient oral cavity images;
the sample set acquisition module is used for acquiring a preset oral defect image from the training set to construct a sample set;
the basic neural network model setting module is used for applying a loss function and an activation function on the basis of the structure of the VGG16 convolutional neural network model;
and the oral cavity diagnosis neural network model establishing module is used for putting the training set and the sample set into the VGG16 neural network for training to generate an oral cavity diagnosis neural network model.
A fourth aspect of the embodiments of the present invention provides an oral diagnosis apparatus, including:
the image acquisition module is used for acquiring an oral image of a patient to be diagnosed;
a diagnosis module, configured to perform a preliminary diagnosis of a preset oral defect on the oral cavity image of the patient to be diagnosed by using the oral cavity diagnosis neural network model as described in any one of the above.
A fifth aspect of embodiments of the present invention provides a terminal device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the method provided in the first aspect when executing the computer program.
A sixth aspect of embodiments of the present invention provides a computer-readable storage medium storing a computer program which, when executed by a processor, implements the steps of the method as provided in the first aspect above.
The embodiment of the invention provides a method for establishing an oral diagnosis neural network model, which is characterized in that a training set of classified patient oral images and a sample set constructed by preset oral defect images are used for generating the oral diagnosis neural network model by applying a loss function and an activation function on the basis of the structure of a VGG16 convolutional neural network model. After the patient uploads the dental photo, the image recognition and the preliminary diagnosis of the oral defects such as malocclusion and the like can be realized through the oral diagnosis neural network model, so that the examination steps are simplified, the examination time is shortened, and the overall diagnosis efficiency of the patient for seeing a doctor is improved.
Drawings
Fig. 1 is a schematic flow chart illustrating an implementation of a method for establishing an oral diagnostic neural network model according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of an apparatus for building an oral diagnostic neural network model according to a second embodiment of the present invention;
fig. 3 is a schematic flow chart of an implementation of the oral diagnosis method according to the third embodiment of the present invention;
fig. 4 is a schematic structural diagram of an oral diagnostic apparatus according to a fourth embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
Suffixes such as "module", "part", or "unit" used to denote elements are used herein only for the convenience of description of the present invention, and have no specific meaning in themselves. Thus, "module" and "component" may be used in a mixture.
In the following description, the serial numbers of the embodiments of the invention are merely for description and do not represent the merits of the embodiments.
Example one
As shown in fig. 1, an embodiment of the present invention provides a method for establishing an oral diagnosis neural network model, which is intended to establish an oral diagnosis neural network model for obtaining a preliminary oral diagnosis result, saving examination time of a patient, and improving diagnosis efficiency. The method includes but is not limited to the following steps:
s101, acquiring a training set containing classified patient oral cavity images.
In step S101, the classified patient image is taken from an intraoral picture of a patient in clinical diagnosis, and classified to obtain image data after classification processing.
In specific application, the pictures in the mouth of a patient taken in clinical diagnosis can be classified through rapid modeling and testing, and in order to improve the classification accuracy, images with preset oral defects can be screened, marked and added with noise, and finally a training set of classified patient oral images is obtained. The preset oral defect can be tooth deformity such as too small front tooth crown, sparse front tooth, tiger tooth, incomplete front tooth crown, broken front tooth, front tooth reverse occlusion, mandibular protrusion, and mesiodism.
S102, obtaining a preset oral defect image from the training set to construct a sample set.
In the above steps S101 and S102, the classified patient oral cavity images may be divided into a training set, a sample set including preset oral defect images, and a test set for testing at a ratio of 70%, 20%, and 10%.
In one embodiment, the labels of the samples including the preset oral defect image are also unified, that is, if the preset oral defect image has a plurality of types of classification labels, only the classification labels about the preset oral defect are retained.
In the embodiment of the present invention, the neural network model for oral cavity diagnosis to be established is used for diagnosing an oral cavity defect condition, and therefore, in the above steps S101 and S102, the oral cavity defect image is preset to be one of the above-mentioned tooth malformation conditions.
S103, applying a loss function and an activation function on the basis of the structure of the VGG16 convolutional neural network model.
In one embodiment, the calculation formula of the loss function in step S103 is as follows:
Figure BDA0002612839310000061
wherein y represents the true probability of the occurrence of the preset oral defect image,
Figure BDA0002612839310000064
representing the predicted probability of occurrence of the preset oral defect image, M representing the number of classified patient oral cavity images, N representing the number of classification labels in each classified patient oral cavity image, i representing the ith classification label, j representing the jth classified patient oral cavity image, yijAnd
Figure BDA0002612839310000062
an ith label representing a jth classified patient oral image.
In one embodiment, if the preset oral defect image only retains classification labels for the preset oral defects, then
Figure BDA0002612839310000063
The first label representing the jth classified patient oral image will be presented.
And S104, putting the training set and the sample set into the VGG16 convolutional neural network model for training to generate an oral diagnosis neural network model.
In a specific application, the neural network training process of step S104 may be:
and S1041, putting the training set into a VGG16 convolutional neural network model for K times of iterative training.
S1042, obtaining the difference between the nth iteration training result and the sample set through the loss function degree.
S1043, calculating a minimized loss function of the nth iteration training result, and adjusting the learning rate of the (n + 1) th iteration training according to the minimized loss function.
And S1044, generating an oral diagnosis neural network model according to the training results of the K times of iterative training.
In order to judge whether the images in the training set have preset oral defect characteristics, the used VGG16 convolutional neural network model consists of a plurality of convolutional layers and a top full-connection layer, wherein the full-connection layer also comprises an associated weight and a pooling layer, and the full-connection layer is provided with two neurons.
In the embodiment of the present invention, in the model training process, the detailed training parameters may be set as:
(1) epochs-50, meaning that the dataset will be iteratively trained 50 times;
(2) BatchSize ═ 32, which represents the batch size of the samples input to the neural network at one time;
(3) a Loss Function, which is "catalytic _ cross", indicates the use of a Loss Function;
(4) the method comprises the steps of optimizing an optimal classifier, wherein the optimal classifier is 'Adadelta', and the learning rate of the neural network model training is restrained.
In a specific application, to avoid overfitting during training of the neural network model, before step S103, the method may further include:
and performing incremental processing on the classified patient oral cavity image and the preset oral cavity defect image.
By increasing the data volume of the training set and the sample set, errors can be reduced, and the over-fitting phenomenon in the training of the neural network model can be avoided.
In one embodiment, the data volume of the training set and the sample set may be increased according to the original classified patient oral cavity image and the preset oral defect image, which includes:
and executing at least one processing mode of rotating a preset angle, translating a preset amplitude, vertically mirroring and horizontally mirroring on the classified patient oral cavity image and the preset defect image.
In detail, the above-mentioned rotation of the classified patient oral cavity image and the preset defect image by the preset angle can be expressed as: rotating the classified oral cavity image and the preset defect image of the patient by 10 percent;
the above-mentioned shifting the classified patient oral cavity image and the preset defect image by the preset amplitude can be expressed as: transversely translating the classified oral cavity images of the patients and the preset defect images by 10% and longitudinally translating by 10%;
the vertical mirror processing and the horizontal mirror processing of the classified patient oral cavity image and the preset defect image can be represented as follows: and horizontally and vertically overturning the classified oral cavity image and the preset defect image of the patient.
In another embodiment, the original classified patient oral cavity image and the preset oral defect image may be subjected to multi-view sampling, and the data amount of the training set and the sample set is increased, which includes:
sampling the classified patient oral cavity image and the preset defect image from H different visual angles to obtain sampling images;
and extracting the characteristics of the sampling image according to the color characteristics, wherein the sampling image after the characteristics are extracted is the added training set and sample set.
The visual angle is an included angle between the lens direction and the image.
In a specific application, the sampling image is subjected to feature extraction according to color features, and a detailed implementation manner of obtaining the sampling image after feature extraction may be as follows:
acquiring the color characteristics of a sampling image, wherein the formula is as follows:
Figure BDA0002612839310000081
Figure BDA0002612839310000082
Figure BDA0002612839310000083
wherein, muxOverall color information representing the sampled image; sigmaxDetail color information representing a sampled image; sxRepresenting color information of a sampled image with a viewing angle within 0-45 DEG, wherein PxtAnd expressing the x color component of the t pixel in the sampling image, and H expressing the number of pixel points of the sampling image.
And calculating the parameter value of each pixel point of the sampling image according to the color characteristics, and selecting the pixels with the parameter values ranging from (255, 0) to (255 ) through a filter to obtain the sampling image after characteristic extraction.
Example two
As shown in fig. 2, an apparatus 20 for modeling an oral diagnostic neural network according to an embodiment of the present invention includes:
a training set obtaining module 21, configured to obtain a training set including classified patient oral cavity images;
a sample set obtaining module 22, configured to obtain a preset oral defect image construction sample set from the training set;
a basic neural network model setting module 23, configured to apply a loss function and an activation function based on the structure of the VGG16 convolutional neural network model;
and the oral cavity diagnosis neural network model establishing module 24 is used for putting the training set and the sample set into the VGG16 neural network for training to generate an oral cavity diagnosis neural network model.
EXAMPLE III
As shown in fig. 3, an embodiment of the present invention further provides an oral cavity diagnosis method, including:
s301, obtaining an oral cavity image of a patient to be diagnosed;
and S302, performing preliminary diagnosis of the preset oral defect on the oral cavity image of the patient to be diagnosed by using the oral cavity diagnosis neural network model in the figure 1.
In practical applications, the step S301 may be to obtain an image of the oral cavity of the patient to be diagnosed through the terminal.
It is conceivable that the oral diagnosis neural network model obtained according to fig. 1 is stored in the cloud server, the terminal acquires an oral image of the patient to be diagnosed and uploads the oral image to the cloud server, and the oral diagnosis neural network module stored in the cloud server determines whether the patient has the preset oral defect according to the oral image of the patient to be diagnosed, so that diagnosis of the preset oral defect is completed based on the patient.
As shown in fig. 4, an embodiment of the present invention further provides an oral cavity diagnosis apparatus 40, including:
an image acquisition module 41, configured to acquire an image of an oral cavity of a patient to be diagnosed;
and a diagnosis module 42, configured to perform a preliminary diagnosis of the preset oral defect on the oral cavity image of the patient to be diagnosed by using the oral cavity diagnosis neural network model as described in fig. 1.
The embodiment of the present invention further provides a terminal device, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the computer program, the processor implements the steps of the method for establishing an oral diagnostic neural network model as described in the first embodiment or the steps of the oral diagnostic method as described in the second embodiment.
Embodiments of the present invention further provide a storage medium, which is a computer readable storage medium, and a computer program is stored thereon, and when being executed by a processor, the computer program implements the steps of the method for establishing an oral diagnostic neural network model according to embodiment one or the steps of the oral diagnostic method according to embodiment two.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the foregoing embodiments illustrate the present invention in detail, those of ordinary skill in the art will understand that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present invention, and are intended to be included within the scope of the present invention.

Claims (10)

1. A method of modeling an oral diagnostic neural network, comprising:
acquiring a training set containing classified patient oral cavity images;
acquiring a preset oral defect image construction sample set from the training set;
applying a loss function and an activation function on the basis of the structure of the VGG16 convolutional neural network model;
and putting the training set and the sample set into the VGG16 neural network for training to generate an oral cavity diagnosis neural network model.
2. The method of modeling an oral diagnostic neural network of claim 1, wherein prior to applying the loss function and the activation function based on the VGG16 convolutional neural network model structure, comprising:
and performing incremental processing on the classified patient oral cavity image and the preset oral cavity defect image.
3. The method of building an oral diagnostic neural network model of claim 2, wherein incrementally processing the classified patient oral image and the pre-set oral defect image comprises:
and executing at least one processing mode of rotating a preset angle, translating a preset amplitude, vertically mirroring and horizontally mirroring on the classified patient oral cavity image and the preset defect image.
4. The method of modeling an oral diagnostic neural network of claim 1, wherein the loss function is calculated by the formula:
Figure FDA0002612839300000011
wherein y represents the true probability of the occurrence of the preset oral defect image,
Figure FDA0002612839300000012
representing the predicted probability of occurrence of the preset oral defect image, M representing the number of classified patient oral cavity images, N representing the number of classification labels in each classified patient oral cavity image, i representing the ith classification label, j representing the jth classified patient oral cavity image, yijAnd
Figure FDA0002612839300000013
an ith label representing a jth classified patient oral image.
5. The method of claim 1, wherein the training set and the sample set are subjected to the VGG16 convolutional neural network model for training to generate the oral diagnostic neural network model, comprising:
putting the training set into a VGG16 convolutional neural network model to perform K times of iterative training;
the difference between the nth iteration training result and the sample set is obtained through the loss function degree;
wherein K is a positive integer, and n is a positive integer less than or equal to K;
calculating a minimized loss function of the nth iterative training result, and adjusting the learning rate of the (n + 1) th iterative training according to the minimized loss function;
and generating an oral diagnosis neural network model according to the training results of the K times of iterative training.
6. An oral diagnostic method comprising:
acquiring an oral cavity image of a patient to be diagnosed;
preliminary diagnosis of a preset oral defect is performed on the patient's oral image to be diagnosed using the oral diagnostic neural network model of any one of claims 1 to 5.
7. An apparatus for modeling an oral diagnostic neural network, comprising:
the training set acquisition module is used for acquiring a training set containing classified patient oral cavity images;
the sample set acquisition module is used for acquiring a preset oral defect image from the training set to construct a sample set;
the basic neural network model setting module is used for applying a loss function and an activation function on the basis of the structure of the VGG16 convolutional neural network model;
and the oral cavity diagnosis neural network model establishing module is used for putting the training set and the sample set into the VGG16 neural network for training to generate an oral cavity diagnosis neural network model.
8. An oral diagnostic device, comprising:
the image acquisition module is used for acquiring an oral image of a patient to be diagnosed;
a diagnosis module for performing a preliminary diagnosis of a preset oral defect on the patient's oral image to be diagnosed using the oral diagnostic neural network model of any one of claims 1 to 5.
9. A terminal device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the method of establishing an oral diagnostic neural network model according to any one of claims 1 to 5 or the steps of the oral diagnostic method according to claim 6 when executing the computer program.
10. A storage medium being a computer readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method of establishing an oral diagnostic neural network model of any one of claims 1 to 5 or the steps of the method of oral diagnosis of claim 6.
CN202010760132.XA 2020-07-31 2020-07-31 Method for establishing oral diagnosis neural network model and oral diagnosis method Pending CN111914931A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010760132.XA CN111914931A (en) 2020-07-31 2020-07-31 Method for establishing oral diagnosis neural network model and oral diagnosis method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010760132.XA CN111914931A (en) 2020-07-31 2020-07-31 Method for establishing oral diagnosis neural network model and oral diagnosis method

Publications (1)

Publication Number Publication Date
CN111914931A true CN111914931A (en) 2020-11-10

Family

ID=73287490

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010760132.XA Pending CN111914931A (en) 2020-07-31 2020-07-31 Method for establishing oral diagnosis neural network model and oral diagnosis method

Country Status (1)

Country Link
CN (1) CN111914931A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112914774A (en) * 2021-01-26 2021-06-08 安徽中科本元信息科技有限公司 Digital oral occlusion analysis system and analysis method based on cloud platform
WO2023279201A1 (en) * 2021-07-06 2023-01-12 Orthodontia Vision Inc. System and method for determining an orthodontic occlusion class

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109859203A (en) * 2019-02-20 2019-06-07 福建医科大学附属口腔医院 Defect dental imaging recognition methods based on deep learning
CN110399899A (en) * 2019-06-21 2019-11-01 武汉大学 Uterine neck OCT image classification method based on capsule network

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109859203A (en) * 2019-02-20 2019-06-07 福建医科大学附属口腔医院 Defect dental imaging recognition methods based on deep learning
CN110399899A (en) * 2019-06-21 2019-11-01 武汉大学 Uterine neck OCT image classification method based on capsule network

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
WEIXIN_34175509: "多分类问题的交叉熵计算", Retrieved from the Internet <URL:https://blog.csdn.net/weixin_34175509/article/details/88769144> *
史丹利复合田: "一文搞懂交叉熵在机器学习中的使用，透彻理解交叉熵背后的直觉", Retrieved from the Internet <URL:https://blog.csdn.net/tsyccnh/article/details/79163834> *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112914774A (en) * 2021-01-26 2021-06-08 安徽中科本元信息科技有限公司 Digital oral occlusion analysis system and analysis method based on cloud platform
WO2023279201A1 (en) * 2021-07-06 2023-01-12 Orthodontia Vision Inc. System and method for determining an orthodontic occlusion class

Similar Documents

Publication Publication Date Title
US20220218449A1 (en) Dental cad automation using deep learning
US11238586B2 (en) Excess material removal using machine learning
CN109816742B (en) Cone beam CT geometric artifact removing method based on fully-connected convolutional neural network
CN111914931A (en) Method for establishing oral diagnosis neural network model and oral diagnosis method
CN109146897B (en) Oral cavity radiation image quality control method and device
CN110728666B (en) Typing method and system for chronic nasosinusitis based on digital pathological slide
WO2021135773A1 (en) Image reconstruction method, apparatus, device, and system, and computer readable storage medium
KR102224597B1 (en) CNN-based high resolution image generating apparatus for minimizing data acquisition time and method therefor
CN113361623B (en) Medical image classification method combining lightweight CNN with transfer learning
CN113298780B (en) Deep learning-based bone age assessment method and system for children
CN110555834B (en) CT bad channel real-time detection and reconstruction method based on deep learning network
WO2022267494A1 (en) Image data generation method and apparatus
WO2024046400A1 (en) Tooth model generation method and apparatus, and electronic device and storage medium
CN113516639B (en) Training method and device for oral cavity abnormality detection model based on panoramic X-ray film
CN110739051B (en) Method for establishing eosinophilic granulocyte proportion model by using nasal polyp pathological picture
JP7430232B2 (en) Sparse projection reconstruction method based on graph learning
CN109978832B (en) Twisted pair lay detection method based on edge reconstruction
CN112488125B (en) Reconstruction method and system based on high-speed visual diagnosis and BP neural network
CN114862982A (en) Hybrid domain unsupervised finite angle CT reconstruction method based on generation countermeasure network
CN114972065A (en) Training method and system of color difference correction model, electronic equipment and mobile equipment
CN111402242B (en) Medical image refinement processing system and method
Faure et al. A Convolutional Neural Network for Dental Panoramic Radiograph Classification
CN112150422A (en) Modeling method of oral health self-detection model based on multitask learning
TWI807809B (en) System, computer program and computer-readable medium for assisting in the recognition of periodontitis and dental caries by using the convolutional neural network of deep learning
US20240127428A1 (en) System and method for probabilistic detection of tooth cavities using machine learning

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination