CN113487611B - Dental film image processing method and system based on artificial intelligence - Google Patents
Dental film image processing method and system based on artificial intelligence Download PDFInfo
- Publication number
- CN113487611B CN113487611B CN202111045542.7A CN202111045542A CN113487611B CN 113487611 B CN113487611 B CN 113487611B CN 202111045542 A CN202111045542 A CN 202111045542A CN 113487611 B CN113487611 B CN 113487611B
- Authority
- CN
- China
- Prior art keywords
- tooth
- incisor
- standard
- inclination
- teeth
- 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.)
- Active
Links
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 24
- 238000003672 processing method Methods 0.000 title claims abstract description 18
- 230000002159 abnormal effect Effects 0.000 claims abstract description 109
- 238000000034 method Methods 0.000 claims abstract description 31
- 230000002776 aggregation Effects 0.000 claims abstract description 30
- 238000004220 aggregation Methods 0.000 claims abstract description 30
- 238000012545 processing Methods 0.000 claims abstract description 12
- 210000004283 incisor Anatomy 0.000 claims description 112
- 238000004458 analytical method Methods 0.000 claims description 29
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 15
- 230000008569 process Effects 0.000 description 7
- 230000005856 abnormality Effects 0.000 description 5
- 230000008447 perception Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 230000011218 segmentation Effects 0.000 description 4
- 210000000214 mouth Anatomy 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 2
- 230000000877 morphologic effect Effects 0.000 description 2
- 208000014151 Stomatognathic disease Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
- A61B6/5205—Devices using data or image processing specially adapted for radiation diagnosis involving processing of raw data to produce diagnostic data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10116—X-ray image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30036—Dental; Teeth
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Radiology & Medical Imaging (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Biomedical Technology (AREA)
- Animal Behavior & Ethology (AREA)
- Optics & Photonics (AREA)
- High Energy & Nuclear Physics (AREA)
- Heart & Thoracic Surgery (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Pathology (AREA)
- Biophysics (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Quality & Reliability (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of image processing, in particular to a dental film image processing method and system based on artificial intelligence. The method comprises the following steps: acquiring an X-ray dental film image; obtaining tooth characteristics according to the dental film image, wherein the tooth characteristics comprise tooth number, tooth length standard deviation, tooth inclination index, tooth inclination aggregation degree and hole tooth proportion; calculating the abnormal degree corresponding to the dental film image according to the tooth characteristics and the standard tooth characteristics; according to the method and the device, whether the dental film images need to be carefully checked is judged according to the abnormal degree corresponding to each dental film image obtained by the tooth characteristics, so that the workload of a dentist can be effectively reduced, and meanwhile, the detection efficiency of the dentist on the dental film images can be improved.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a dental film image processing method and system based on artificial intelligence.
Background
In the field of modern dental medicine, when a dentist performs tooth detection on a patient and the like, the dentist usually performs examination and analysis on teeth by taking X-ray dental films to identify abnormal conditions of the teeth and give corresponding treatment to the abnormal teeth.
The current technology is that the doctor directly carries out analysis and identification to patient's X-ray dental film image to judge the particular case of patient's tooth, this process is comparatively loaded down with trivial details. To like children's tooth regular physical examination or human detection like in the situation that has a large amount of crowds to carry out tooth detection, can produce a large amount of X-ray dental film images, but if only there is the dentist to detect the X-ray dental film that produces, the work load can be very big, also the condition of false retrieval and missed retrieval appears easily in the testing process, leads to detection efficiency to be lower.
Disclosure of Invention
In order to solve the problem of low detection efficiency when the existing dentist detects the X-ray dental film image, the invention aims to provide a dental film image processing method and a dental film image processing system based on artificial intelligence, and the adopted technical scheme is as follows:
in a first aspect, an embodiment of the present invention provides a method for processing dental film images based on artificial intelligence, the method including the following steps:
acquiring an X-ray dental film image;
obtaining tooth characteristics according to the dental film image, wherein the tooth characteristics comprise tooth number, tooth length standard deviation, tooth inclination index, tooth inclination aggregation degree and hole tooth proportion;
calculating the abnormal degree corresponding to the dental film image according to the tooth characteristics and the standard tooth characteristics;
the method for obtaining the tooth inclination index comprises the following steps:
obtaining the inclination of each tooth in the dental film image according to a set judgment standard, and endowing each tooth with a corresponding inclination grade according to a preset number of inclination grade intervals;
and calculating the tooth inclination index according to the inclination grade corresponding to each tooth and the constructed tooth inclination index analysis model.
In a second aspect, another embodiment of the present invention provides an artificial intelligence-based dental image processing system, which includes a memory and a processor, wherein the processor executes a computer program stored in the memory to implement the artificial intelligence-based dental image processing method.
Preferably, the method for obtaining the tooth inclination aggregation degree comprises the following steps:
constructing abnormal subsequences according to the inclination levels corresponding to the teeth, and obtaining the length of each abnormal subsequence;
calculating the tooth inclination aggregation degree according to the number of the constructed abnormal subsequences, the length of each abnormal subsequence and the constructed tooth inclination aggregation degree analysis model;
the method for constructing the abnormal subsequence comprises the following steps:
setting the inclination grade greater than or equal to the inclination grade threshold value as an abnormal inclination grade;
and when the number of the abnormal inclination grades continuously appearing in the dental film image is more than or equal to the preset number, taking the continuously appearing abnormal inclination grades as an abnormal subsequence.
Preferably, the method for obtaining the inclination of each tooth in the dental film image according to the set determination criteria includes:
dividing the lower teeth in the dental film image into three sets, namely a lower incisor set, a lower incisor set and a lower posterior set, wherein the lower incisor set comprises a left lower incisor set and a right lower incisor set, and the lower posterior set comprises a left lower posterior set and a right lower posterior set;
calculating the deviation angle of each lower tooth from the corresponding judgment standard according to the judgment standard corresponding to the set to which each lower tooth belongs, and recording the calculated deviation angle corresponding to each lower tooth as the inclination corresponding to each lower tooth;
wherein the lower incisor set takes the lower incisor standard direction as a judgment standard;
the left lower incisor set takes a left lower incisor standard direction obtained by rotating a first angle in a clockwise direction from a lower incisor standard direction as a judgment standard, and the left lower posterior incisor set takes a left lower posterior incisor standard direction obtained by rotating a second angle in a clockwise direction from the left lower incisor standard direction as a judgment standard;
and the lower right incisor set takes a lower right incisor standard direction obtained by rotating the lower incisor standard direction by a first angle in the anticlockwise direction as a judgment standard, and the lower right posterior incisor set takes a lower right posterior incisor standard direction obtained by rotating the lower right incisor standard direction by a second angle in the anticlockwise direction as a judgment standard.
Preferably, the method for obtaining the inclination of each tooth in the dental film image according to the set determination criteria includes:
dividing upper teeth in the dental film image into three sets, namely an upper incisor set, an upper incisor set and an upper posterior set, wherein the upper incisor set comprises a left upper incisor set and a right upper incisor set, and the upper posterior set comprises a left upper posterior set and a right upper posterior set;
calculating the deviation angle of each upper row of teeth and the corresponding judgment standard according to the judgment standard corresponding to the set to which each upper row of teeth belongs, and recording the calculated deviation angle corresponding to each upper row of teeth as the inclination corresponding to each upper row of teeth;
wherein the upper incisor set takes the upper incisor standard direction as a judgment standard;
the left upper incisor set takes a left upper incisor standard direction obtained by rotating the upper incisor standard direction by a third angle in the clockwise direction as a judgment standard, and the left upper posterior incisor set takes a left upper posterior incisor standard direction obtained by rotating the left upper incisor standard direction by a fourth angle in the clockwise direction as a judgment standard;
and the right upper incisor set takes a right upper incisor standard direction obtained by rotating the upper incisor standard direction by a third angle in the anticlockwise direction as a judgment standard, and the right upper posterior incisor set takes a right upper posterior incisor standard direction obtained by rotating the right upper incisor standard direction by a fourth angle in the anticlockwise direction as a judgment standard.
Preferably, the expression of the tooth inclination index analysis model is as follows:
wherein,
in order to be of the grade of the inclination,
to grade of inclination
The frequency of occurrence of the (co) signal,
is an index of tooth inclination.
Preferably, the following formula is adopted to calculate the inclined concentration of the upper tooth row and the lower tooth row:
wherein,
the number of abnormal subsequences of the upper teeth row,
the number of abnormal subsequences in the lower teeth,
the longest abnormal subsequence length in the upper teeth row,
the longest abnormal subsequence length in the lower teeth,
in order to obtain the inclined concentration degree of the upper row of teeth,
the degree of the inclined convergence of the lower teeth,
is a natural constant;
the expression of the tooth inclination aggregation degree analysis model is as follows:
wherein,
in order to provide a degree of angular concentration of the teeth,
the weights corresponding to the oblique concentration of the upper row teeth,
the weight value is corresponding to the inclined concentration of the lower tooth-arrangement.
Preferably, the method for calculating the degree of abnormality corresponding to the dental film image includes:
and (3) subtracting the tooth characteristics corresponding to the dental film image from the standard tooth characteristics to construct a tooth characteristic abnormal degree analysis model, wherein the expression of the tooth characteristic abnormal degree analysis model is as follows:
wherein,
the corresponding abnormal degree of the dental film image,
the number of teeth corresponding to the dental film image,
the tooth length standard deviation corresponding to the dental film image,
for the corresponding degree of tilt of the dental film image,
for the corresponding degree of oblique concentration of the dental film image,
is the hole tooth proportion corresponding to the dental film image,
the number of teeth is the standard number of teeth,
is the standard deviation of the standard tooth length,
is the degree of inclination of the standard tooth,
for a standard degree of tooth tilt concentration,
is the proportion of the teeth in the holes of the standard teeth.
Preferably, the tooth length standard deviation is calculated by the following formula:
wherein,
is as follows
The length of each of the teeth is such that,
is the average of the lengths of all the teeth,
as to the number of teeth,
is the standard deviation of tooth length.
The embodiment of the invention has the following beneficial effects:
according to the X-ray dental film image, the tooth characteristics are obtained, the abnormal degree corresponding to the dental film image is calculated according to the obtained tooth characteristics and the standard tooth characteristics, the abnormal degree is obtained by combining the number of teeth, the tooth length standard deviation, the tooth inclination index, the tooth inclination aggregation degree and the hole tooth ratio in the dental film image, and the abnormal degree of the teeth in the dental film image is reflected. If the abnormal degree is larger, the corresponding dental film image is more abnormal, and the problems of the teeth of the patient are more serious; if the abnormal degree is smaller, the corresponding dental film image is more normal, and the problems of the teeth of the patient are less. The invention realizes the automatic processing of the dental film image, can effectively reduce the workload of dentists and improve the detection efficiency of the dental film image.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions and advantages of the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a flowchart of a method for artificial intelligence based dental image processing according to an embodiment of the present invention;
FIG. 2 is an X-ray dental image provided by one embodiment of the present invention.
Detailed Description
In order to further explain the technical means and functional effects of the present invention adopted to achieve the predetermined invention purpose, the following detailed description will be made of a dental film image processing method and system based on artificial intelligence according to the present invention with reference to the accompanying drawings and preferred embodiments.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The invention considers that when a large number of persons carry out tooth detection, a large number of X-ray dental film images are generated, and if only a dentist carries out detection on the generated X-ray dental films, the workload is very large and the working efficiency is very low. The main conception of the invention is as follows: obtaining tooth characteristics according to the obtained X-ray dental film image, wherein the dental film characteristics comprise: the method comprises the steps of calculating the number of teeth, the standard deviation of the tooth length, the tooth inclination index, the tooth inclination aggregation degree and the hole tooth proportion, and calculating the abnormal degree corresponding to the dental film image according to the obtained tooth characteristics and the standard tooth characteristics, wherein the abnormal degree can reflect the number of problems of the patient teeth corresponding to the dental film image.
The following describes a specific scheme of a dental film image processing method and system based on artificial intelligence in detail with reference to the accompanying drawings.
The embodiment of the dental film image processing method based on artificial intelligence comprises the following steps:
as shown in fig. 1, the artificial intelligence based dental film image processing method of the present embodiment includes the following steps:
step S1, an X-ray dental image is acquired.
In the embodiment, the X-ray dental film machine is used for collecting human dental film images, the tooth health condition is detected and judged based on the collected dental film images, and the obtained X-ray dental film images are as shown in FIG. 2.
And step S2, obtaining tooth characteristics according to the X-ray dental film image, wherein the tooth characteristics comprise tooth number, tooth length standard deviation, tooth inclination indexes, tooth inclination aggregation degree and hole tooth proportion.
Step S2 of this embodiment is specifically implemented by the following sub-steps:
and 2-1, identifying the dental film image.
In this embodiment, a trained semantic segmentation network (e.g., a DNN network) is used to perform tooth recognition and segmentation on an acquired X-ray dental image, so as to obtain a semantic perception effect graph of teeth and obtain number information corresponding to each tooth, where the training mode of the semantic segmentation network is as follows:
firstly, a network label is manufactured, the embodiment classifies image data based on pixels, wherein a tooth pixel is marked as 1, a tooth hole region pixel is marked as 2, other region pixels are marked as 0, and in order to accurately analyze the condition of each tooth subsequently, the embodiment also marks a serial number for each tooth.
And then inputting the marked X-ray dental film image into a network for training, after the dental film image is input into the network, firstly, an encoder performs encoding operation on the image to obtain a characteristic vector, and then, the obtained characteristic vector is input into a tooth perception decoder to obtain a final tooth semantic perception image.
The semantic segmentation network adopts a cross entropy loss function to carry out iterative training to continuously update network parameters.
And step S2-2, detecting and describing tooth forms based on tooth characteristics.
The present embodiment describes the tooth morphology in detail according to tooth feature data, which includes: tooth number, tooth length standard deviation, tooth inclination index, tooth inclination aggregation degree and hole tooth proportion. This process provides a basis for subsequent dental data structuring.
The process of detecting and describing the tooth features in the embodiment comprises the following steps:
firstly, the number of teeth is counted in the segmented image according to the tooth semantic perception effect graph.
Secondly, calculating the number of pixels contained in each line segment according to a connecting line from the top point of the crown to the end point of the longest tooth root of each tooth in the dental film image, and recording the number as the tooth length, wherein the standard deviation of the tooth length is as follows:
wherein,
is as follows
The length of each of the teeth is such that,
for all teethThe average value of the length of the teeth,
as to the number of teeth,
is the standard deviation of tooth length.
Thirdly, the tooth inclination index is analyzed by the following specific steps:
as shown in fig. 2, in the present embodiment, in order to accurately analyze the inclination of the teeth, the upper teeth and the lower teeth are divided into three sets, and then, for each set, the inclination of the teeth in the set is analyzed in a targeted manner. In the present embodiment, the vector from the midpoint of the crown to the midpoint of the tooth root is taken as the direction of the tooth, and in other embodiments, the vector from the root to the crown may be taken as the direction of the tooth. Embodiments wherein determining the inclination of the teeth comprise:
1) in the embodiment, the lower teeth in the dental film image are divided into three sets, namely a lower incisor set, a lower incisor set and a lower posterior set, wherein the lower incisor set comprises a left lower incisor set and a right lower incisor set, and the lower posterior set comprises a left lower posterior set and a right lower posterior set.
And calculating the deviation angle of each lower tooth from the corresponding judgment standard according to the judgment standard corresponding to the set to which each lower tooth belongs, and recording the calculated deviation angle corresponding to each lower tooth as the inclination corresponding to each lower tooth.
In the present embodiment, the lower incisor standard direction is set as a vertically upward direction for the lower incisor set as a determination standard.
And the left lower incisor set takes a left lower incisor standard direction obtained by rotating the lower incisor standard direction by a first angle in the clockwise direction as a judgment standard, and the left lower posterior incisor set takes a left lower posterior incisor standard direction obtained by rotating the left lower incisor standard direction by a second angle in the clockwise direction as a judgment standard.
And the lower right incisor set takes a lower right incisor standard direction obtained by rotating the lower incisor standard direction by a first angle in the anticlockwise direction as a judgment standard, and the lower right posterior incisor set takes a lower right posterior incisor standard direction obtained by rotating the lower right incisor standard direction by a second angle in the anticlockwise direction as a judgment standard.
In this embodiment, the first angle and the second angle are set to 6 ° and 9 °, respectively, and can be set as needed in practical application.
2) This embodiment will last tooth in the dental film image divide into three set, is last incisor set, upside incisor set and last back tooth set respectively, wherein upside incisor set includes upper left side incisor set and upper right side incisor set, it includes upper left back tooth set and upper right back tooth set to go up the back tooth set.
And calculating the deviation angle of each upper row tooth and the corresponding judgment standard according to the judgment standard corresponding to the set to which each upper row tooth belongs, and recording the calculated deviation angle corresponding to each upper row tooth as the inclination corresponding to each upper row tooth.
The upper incisor set takes the upper incisor standard direction as a determination standard, and the upper incisor standard direction is set to be a vertical downward direction in the embodiment.
The upper left incisor set takes the upper left incisor standard direction obtained by rotating the upper incisor standard direction by a third angle in the clockwise direction as a judgment standard, and the upper left posterior set takes the upper left posterior standard direction obtained by rotating the upper left incisor standard direction by a fourth angle in the clockwise direction as a judgment standard.
And the right upper incisor set takes a right upper incisor standard direction obtained by rotating the upper incisor standard direction by a third angle in the anticlockwise direction as a judgment standard, and the right upper posterior incisor set takes a right upper posterior incisor standard direction obtained by rotating the right upper incisor standard direction by a fourth angle in the anticlockwise direction as a judgment standard.
In this embodiment, the third angle and the fourth angle are respectively set to 6 ° and 9 °, and can be set as required in practical application.
The embodiment performs diversity combining processing on the teeth, and performs detection in a self-adaptive manner through the corresponding standard direction, so as to reduce the error of judging the tooth inclination, avoid the false detection of the teeth at two sides due to the imaging reason, and improve the accuracy of the tooth inclination.
After obtaining the inclination of each tooth, each tooth is assigned with a corresponding inclination grade according to a preset number of inclination grade intervals, the preset number is set by the actual requirement, in the embodiment, the tooth inclination is divided into 10 inclination grade intervals, the inclination grades of the teeth are represented by 0-9, wherein the inclination grade corresponding to the interval [0 degrees, 9 degrees ] is 0, the inclination grade corresponding to the interval (9 degrees, 18 degrees ] is 1, the inclination grade corresponding to the interval (18 degrees, 27 degrees ] is 2, the inclination grade corresponding to the interval (27 degrees, 36 degrees ] is 3, the inclination grade corresponding to the interval (36 degrees, 45 degrees ] is 4, the inclination grade corresponding to the interval (45 degrees, 54 degrees ] is 5, the inclination grade corresponding to the interval (54 degrees, 63 degrees ] is 6, the inclination grade corresponding to the interval (63 degrees, 72 degrees ] is 7, the inclination grade corresponding to the interval (72 degrees, 81 degrees ] is 8, the interval (81 degrees, 90 degrees) corresponds to an inclination grade of 9, and the grade interval can be set according to actual conditions.
The larger the inclination level, the higher the degree of inclination of the tooth, and this embodiment sets an inclination level equal to or greater than an inclination level threshold as an abnormal inclination level, the inclination level threshold being set according to the actual situation, and this embodiment sets an inclination level threshold to 2, and therefore an inclination level equal to or greater than 2 is set as an abnormal inclination level.
And after the inclination grade corresponding to each tooth in the dental film image is obtained, calculating the tooth inclination index according to the inclination grade corresponding to each tooth and the constructed tooth inclination index analysis model. In the embodiment, a mathematical modeling method is adopted to fit a functional relation between the inclination levels and the occurrence frequency of the inclination levels, so as to construct a tooth inclination index analysis model, wherein the tooth inclination index analysis model is used for detecting the inclination degrees of all teeth in the whole oral cavity, and the higher the inclination degree is, the more abnormal the teeth are. Wherein the inclination grade is in positive correlation with the tooth inclination index, and when the occurrence frequency of the teeth with large inclination grade is high, the tooth inclination index is higher.
The expression of the tooth inclination index analysis model is as follows:
wherein,
in order to be of the grade of the inclination,
to grade of inclination
The frequency of occurrence of the (co) signal,
is an index of the inclination of the teeth,
to grade of tilting
The weight occupied by the time.
Fourthly, the degree of the oblique tooth aggregation, the present invention considers that the high degree of the oblique tooth aggregation causes crowding of teeth, thus increasing the difficulty of daily care of teeth, and further causing an increased risk of dental diseases. The invention analyzes and detects the tooth inclination aggregation degree to further judge the abnormal condition of the tooth.
In this embodiment, an abnormal subsequence is first constructed according to the inclination level corresponding to each tooth, and the length of each abnormal subsequence is obtained. The method for constructing the abnormal subsequence comprises the following steps:
the tooth inclination index analysis model obtains the inclination grade corresponding to each tooth in the dental film image, two groups of inclination grade sequences corresponding to the upper tooth row and the lower tooth row can be obtained, and the two groups of sequences are analyzed. In this embodiment, when the number of the abnormal inclination levels continuously appearing in the dental film image is equal to or greater than 3, the abnormal inclination level sequence continuously appearing is taken as an abnormal subsequence, and the number of the inclination levels included in the abnormal subsequence is taken as the length of the abnormal subsequence, and the continuous number can be set according to the actual situation.
In order to analyze the tooth inclination aggregation degree more comprehensively, the invention adjusts the length of the abnormal subsequence according to the number of the interval inclination grades between the adjacent abnormal subsequences. The specific process comprises the following steps: and analyzing the number of the gradient levels of the middle intervals of the adjacent abnormal subsequences, and merging the two adjacent abnormal subsequences to be used as a new abnormal subsequence when the number of the gradient levels of the middle intervals of the adjacent abnormal subsequences is 1.
The number of the abnormal subsequences and the length of each abnormal subsequence are obtained by analyzing the abnormal subsequences constructed by the method, and the tooth inclination aggregation degree is calculated by combining the obtained data and the constructed tooth inclination aggregation degree analysis model, wherein the specific implementation mode is as follows:
the method comprises the steps of firstly constructing the inclined aggregation degrees of upper teeth and lower teeth, wherein the inclined aggregation degrees of the upper teeth or the lower teeth can reflect the aggregation degree of abnormal teeth in the upper teeth or the lower teeth, the higher the inclined aggregation degree is, the more abnormal the teeth are, the more problems exist in the teeth of a patient corresponding to a dental film image, and the upper teeth and the lower teeth are different from each other, so that the upper teeth and the lower teeth are calculated separately. The relationship between the number of abnormal subsequences and the inclination degree is in positive correlation, and the longest abnormal subsequence length and the inclination degree are also in positive correlation, so that the larger the two are, the larger the inclination degree of the tooth is. Because the relationship between the two parameters and the inclination concentration is complex, the mathematical modeling method is used for fitting the functional relationship among the number of the abnormal subsequences, the length of the longest abnormal subsequence and the inclination concentration to obtain the inclination concentration of the upper tooth row and the lower tooth row, wherein the calculation formula of the inclination concentration of the upper tooth row is as follows:
the calculation formula of the inclined concentration degree of the lower teeth is as follows:
wherein,
the number of abnormal subsequences of the upper teeth row,
the number of abnormal subsequences in the lower teeth,
the longest abnormal subsequence length in the upper teeth row,
the longest abnormal subsequence length in the lower teeth,
in order to obtain the inclined concentration degree of the upper row of teeth,
the degree of the inclined convergence of the lower teeth,
is a natural constant.
Combining the oblique aggregation degrees of the upper teeth and the lower teeth, and constructing a tooth oblique aggregation degree analysis model through a certain functional relation, wherein the tooth oblique aggregation degree represents the oblique aggregation degree of teeth in the whole dental film image, and the expression is as follows:
wherein,
the degree of tooth inclination gathering is the same as that of the tooth,
the weights corresponding to the oblique concentration of the upper row teeth,
the weight value is corresponding to the inclined concentration of the lower tooth-arrangement. In this embodiment is provided with
=0.5。
Fifth, the proportion of the teeth with holes, which are the decayed teeth in the mouth. In this embodiment, the number of the hole teeth is obtained by performing statistical analysis on the semantic perception effect graph of the dental film image, and the hole tooth ratio is calculated according to the following calculation formula:
wherein,
the number of the teeth with the holes is,
is the proportion of the teeth with holes. The larger the proportion of the holes and the teeth is, the more the decayed teeth are in the oral cavity, and the more abnormal the corresponding dental film picture is.
And step S2-3, realizing the structuring of dental film image data.
Constructing a morphological code corresponding to the dental film image based on the dental characteristic information corresponding to the dental film image obtained in step S2-2, wherein the morphological code is sequence data consisting of the dental characteristic information and is designated as [, ]
,
,
,
,
]And each dental film image corresponds to a form code and is used for describing information of tooth characteristics. The process of constructing the morphology code in this embodiment is a process of structuring the dental film image data.
And 3, calculating the abnormal degree corresponding to the dental film image according to the various tooth characteristics and the standard tooth characteristics.
And converting the dental film image data into structured data according to the information of the tooth characteristics, and constructing a tooth characteristic abnormal degree analysis model so as to detect the abnormal degree of the tooth characteristics. The specific implementation mode is as follows:
firstly, analysis statistics is carried out based on a large amount of data, thereby obtaining the standard characteristic of the tooth, and the corresponding standard form code is [ 2 ]
,
,
,
,
]And the standard tooth characteristics are used for the subsequent analysis of the abnormal degree of the tooth.
In this embodiment, the form codes corresponding to all dental film images obtained by the above method are recorded as
Wherein
Representing the number of teeth detected. Then, a tooth characteristic abnormal degree analysis model is constructed based on the form code corresponding to the dental film image and the standard form, namely, the tooth characteristic corresponding to the dental film image is differentiated from the standard tooth characteristic to construct the tooth characteristic abnormal degree analysis model, the abnormal degree is the basis of whether the dental film image is carefully detected, and the expression of the tooth characteristic abnormal degree analysis model is as follows:
wherein,
is as follows
The degree of abnormality corresponding to the image of the individual dental film,
is as follows
The number of teeth corresponding to the individual dental image,
is as follows
The standard deviation of the tooth length corresponding to the individual dental film image,
is as follows
The corresponding degree of tilt of the image of the individual dental film,
first, the
The degree of oblique concentration to which the image of the individual dental film corresponds,
first, the
The ratio of the hole teeth corresponding to the personal dental film image,
the number of teeth is the standard number of teeth,
is the standard deviation of the standard tooth length,
is the degree of inclination of the standard tooth,
for a standard degree of tooth tilt concentration,
is the proportion of the teeth in the holes of the standard teeth.
In this embodiment, the model is normalized by a normalization method, so as to ensure that the function value of the model is (0, 1). The larger the abnormal degree corresponding to the dental film image calculated by the tooth characteristic abnormal degree analysis model is, the more problems exist in the teeth of the patient corresponding to the dental film image, and the more careful examination is needed; if the smaller the degree of abnormality, the more normal the dental image is, then only a rough inspection of the dental image may be performed.
In order to reduce the workload of the dentist to a greater extent and improve the overall detection speed of the system, the embodiment performs filtering on the dental film images according to the calculated abnormal degree corresponding to each dental film image, and the specific implementation manner is as follows:
the abnormal degree of the dental film image is firstly classified into an abnormal grade, and the abnormal degree is classified into a grade interval of 4 according to the embodiment, wherein the interval can be set according to the actual situation. In this embodiment, when the abnormal degree corresponding to the dental film image is (0, 0.25), the abnormal level of the dental film image of the subject is considered to be 1, the tooth condition is good, and the tooth physical examination is qualified, when the abnormal degree corresponding to the dental film image is (0.25, 0.5), the abnormal level of the dental film image of the subject is considered to be 2, the tooth condition is general, when the abnormal degree corresponding to the dental film image is (0.5, 0.75), the abnormal level of the dental film image of the subject is considered to be 3, which represents that the tooth is slightly abnormal, the system prompts the dentist to quickly check the dental film image to be detected, and when the abnormal degree corresponding to the dental film image is (0.75, 1), the abnormal level of the dental film image is considered to be 4, the tooth is severely abnormal, and the system prompts the dentist to spend a lot of time for fine detection and analysis.
In this embodiment, the dental film images with the abnormality levels of 1 and 2 corresponding to the dental film images are filtered, and only the dental film images with the abnormality levels of 3 and 4 may be analyzed by the subsequent dentist during dental film detection.
According to the X-ray dental film image, the tooth characteristics are obtained, the abnormal degree corresponding to the dental film image is calculated according to the obtained tooth characteristics and the standard tooth characteristics, the abnormal degree is obtained by combining the number of teeth, the tooth length standard deviation, the tooth inclination index, the tooth inclination aggregation degree and the hole tooth ratio in the dental film image, and the abnormal degree of the teeth in the dental film image is reflected. If the abnormal degree is larger, the corresponding dental film image is more abnormal, and the problems of the teeth of the patient are more serious; if the abnormal degree is smaller, the corresponding dental film image is more normal, and the problems of the teeth of the patient are less. The invention realizes the automatic processing of the dental film image, can effectively reduce the workload of dentists and improve the detection efficiency of the dental film image.
An embodiment of an artificial intelligence based dental film image processing system:
the artificial intelligence based dental film image processing system comprises a memory and a processor, wherein the processor executes a computer program stored in the memory to realize the artificial intelligence based dental film image processing method.
Since the artificial intelligence based dental film image processing method has been described in the embodiment of the artificial intelligence based dental film image processing method, the embodiment does not describe the artificial intelligence based dental film image processing method again.
It should be noted that: the above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (4)
1. A dental film image processing method based on artificial intelligence is characterized by comprising the following steps:
acquiring an X-ray dental film image;
obtaining tooth characteristics according to the dental film image, wherein the tooth characteristics comprise tooth number, tooth length standard deviation, tooth inclination index, tooth inclination aggregation degree and hole tooth proportion;
calculating the abnormal degree corresponding to the dental film image according to the tooth characteristics and the standard tooth characteristics;
the method for obtaining the tooth inclination index comprises the following steps:
obtaining the inclination of each tooth in the dental film image according to a set judgment standard, and endowing each tooth with a corresponding inclination grade according to a preset number of inclination grade intervals;
calculating a tooth inclination index according to the inclination grade corresponding to each tooth and the constructed tooth inclination index analysis model;
the method for obtaining the tooth inclined aggregation degree comprises the following steps:
constructing abnormal subsequences according to the inclination levels corresponding to the teeth, and obtaining the length of each abnormal subsequence;
calculating the tooth inclination aggregation degree according to the number of the constructed abnormal subsequences, the length of each abnormal subsequence and the constructed tooth inclination aggregation degree analysis model;
the method for constructing the abnormal subsequence comprises the following steps:
setting the inclination grade greater than or equal to the inclination grade threshold value as an abnormal inclination grade;
when the number of the abnormal inclination grades which continuously appear in the dental film image is more than or equal to the preset number, taking the continuously appearing abnormal inclination grades as an abnormal subsequence;
calculating the inclined concentration of the upper tooth row and the lower tooth row by adopting the following formula:
wherein,
the number of abnormal subsequences of the upper teeth row,
the number of abnormal subsequences in the lower teeth,
the longest abnormal subsequence length in the upper teeth row,
the longest abnormal subsequence length in the lower teeth,
in order to obtain the inclined concentration degree of the upper row of teeth,
the degree of the inclined convergence of the lower teeth,
is a natural constant;
the tooth inclination aggregation degree analysis model expression is as follows:
wherein,
in order to provide a degree of angular concentration of the teeth,
the weights corresponding to the oblique concentration of the upper row teeth,
the weight value is corresponding to the inclined concentration degree of the lower tooth arrangement;
the method for obtaining the inclination of each tooth in the dental film image according to the set judgment standard comprises the following steps:
dividing the lower teeth in the dental film image into three sets, namely a lower incisor set, a lower incisor set and a lower posterior set, wherein the lower incisor set comprises a left lower incisor set and a right lower incisor set, and the lower posterior set comprises a left lower posterior set and a right lower posterior set;
calculating the deviation angle of each lower tooth from the corresponding judgment standard according to the judgment standard corresponding to the set to which each lower tooth belongs, and recording the calculated deviation angle corresponding to each lower tooth as the inclination corresponding to each lower tooth;
wherein the lower incisor set takes the lower incisor standard direction as a judgment standard;
the left lower incisor set takes a left lower incisor standard direction obtained by rotating a first angle in a clockwise direction from a lower incisor standard direction as a judgment standard, and the left lower posterior incisor set takes a left lower posterior incisor standard direction obtained by rotating a second angle in a clockwise direction from the left lower incisor standard direction as a judgment standard;
the lower right incisor set takes a lower right incisor standard direction obtained by rotating a lower incisor standard direction by a first angle in the counterclockwise direction as a judgment standard, and the lower right posterior incisor set takes a lower right posterior incisor standard direction obtained by rotating a lower right incisor standard direction by a second angle in the counterclockwise direction as a judgment standard;
the method for obtaining the inclination of each tooth in the dental film image according to the set judgment standard comprises the following steps:
dividing upper teeth in the dental film image into three sets, namely an upper incisor set, an upper incisor set and an upper posterior set, wherein the upper incisor set comprises a left upper incisor set and a right upper incisor set, and the upper posterior set comprises a left upper posterior set and a right upper posterior set;
calculating the deviation angle of each upper row of teeth and the corresponding judgment standard according to the judgment standard corresponding to the set to which each upper row of teeth belongs, and recording the calculated deviation angle corresponding to each upper row of teeth as the inclination corresponding to each upper row of teeth;
wherein the upper incisor set takes the upper incisor standard direction as a judgment standard;
the left upper incisor set takes a left upper incisor standard direction obtained by rotating the upper incisor standard direction by a third angle in the clockwise direction as a judgment standard, and the left upper posterior incisor set takes a left upper posterior incisor standard direction obtained by rotating the left upper incisor standard direction by a fourth angle in the clockwise direction as a judgment standard;
the upper right incisor set takes a standard direction of upper right incisor obtained by rotating the standard direction of upper incisor by a third angle in the anticlockwise direction as a judgment standard, and the upper right posterior set takes a standard direction of upper right incisor obtained by rotating the standard direction of upper right incisor by a fourth angle in the anticlockwise direction as a judgment standard;
the method for calculating the corresponding abnormal degree of the dental film image comprises the following steps:
and (3) subtracting the tooth characteristics corresponding to the dental film image from the standard tooth characteristics to construct a tooth characteristic abnormal degree analysis model, wherein the expression of the tooth characteristic abnormal degree analysis model is as follows:
wherein,
the corresponding abnormal degree of the dental film image,
the number of teeth corresponding to the dental film image,
the tooth length standard deviation corresponding to the dental film image,
for the corresponding degree of tilt of the dental film image,
for the corresponding degree of oblique concentration of the dental film image,
is the hole tooth proportion corresponding to the dental film image,
the number of teeth is the standard number of teeth,
is the standard deviation of the standard tooth length,
is the degree of inclination of the standard tooth,
for a standard degree of tooth tilt concentration,
is the proportion of the teeth in the holes of the standard teeth.
2. The artificial intelligence based dental film image processing method according to claim 1, wherein the expression of the tooth inclination index analysis model is:
wherein,
in order to be of the grade of the inclination,
to grade of inclination
The frequency of occurrence of the (co) signal,
is an index of tooth inclination.
3. The artificial intelligence based dental film image processing method according to claim 1, wherein the calculation formula of the tooth length standard deviation is:
wherein,
is as follows
The length of each of the teeth is such that,
is the average of the lengths of all the teeth,
as to the number of teeth,
is the standard deviation of tooth length.
4. An artificial intelligence based dental image processing system comprising a memory and a processor, wherein the processor executes a computer program stored in the memory to implement the artificial intelligence based dental image processing method according to any one of claims 1-3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111045542.7A CN113487611B (en) | 2021-09-07 | 2021-09-07 | Dental film image processing method and system based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111045542.7A CN113487611B (en) | 2021-09-07 | 2021-09-07 | Dental film image processing method and system based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113487611A CN113487611A (en) | 2021-10-08 |
| CN113487611B true CN113487611B (en) | 2021-11-26 |
Family
ID=77947265
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111045542.7A Active CN113487611B (en) | 2021-09-07 | 2021-09-07 | Dental film image processing method and system based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113487611B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115035025A (en) * | 2022-04-28 | 2022-09-09 | 南通林德安全设备科技有限公司 | Tooth correction auxiliary method based on dental film processing |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3572030B1 (en) * | 2018-05-22 | 2022-11-02 | Dental Monitoring | Method for analysing a dental situation |
| CN112837279A (en) * | 2021-01-25 | 2021-05-25 | 孙猛猛 | Tooth distortion detection pre-diagnosis system based on artificial intelligence |
-
2021
- 2021-09-07 CN CN202111045542.7A patent/CN113487611B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113487611A (en) | 2021-10-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11790643B2 (en) | Deep learning for tooth detection and evaluation | |
| CN111083922B (en) | Dental image analysis method for correction diagnosis and apparatus using the same | |
| US20210118132A1 (en) | Artificial Intelligence System For Orthodontic Measurement, Treatment Planning, And Risk Assessment | |
| US20210118129A1 (en) | Dental Image Synthesis using Generative Adversarial Networks with Semantic Activation Blocks | |
| CN109543526B (en) | True and false facial paralysis recognition system based on depth difference characteristics | |
| TW202101480A (en) | Dehydration amount prediction method for hemodialysis and electronic device using the same | |
| EP3453321B1 (en) | Non-invasive method and system for estimating blood pressure from photoplethysmogram using statistical post-processing | |
| KR102462185B1 (en) | Method for providing information about orthodontics and device for providing information about orthodontics using deep learning ai algorithm | |
| CN109243620B (en) | Drug effect optimization method and device based on blood concentration monitoring | |
| Kong et al. | Automated maxillofacial segmentation in panoramic dental x-ray images using an efficient encoder-decoder network | |
| CN115547500B (en) | Hemodialysis patient's health monitoring equipment and system | |
| Lin et al. | Tooth numbering and condition recognition on dental panoramic radiograph images using CNNs | |
| CN113487611B (en) | Dental film image processing method and system based on artificial intelligence | |
| CN116246779B (en) | Dental diagnosis and treatment scheme generation method and system based on user image data | |
| CN116092680A (en) | Abdominal aortic aneurysm early prediction method and system based on random forest algorithm | |
| CN114566293B (en) | CRRT (China railway terminal) shutdown decision-making auxiliary system and method | |
| CN110223743A (en) | A kind of structuring processing method and system for pulmonary cancer diagnosis record | |
| CN114743271A (en) | Method and system for assessing abnormal swallowing | |
| CN116152271A (en) | CBCT (Cone-based computed tomography) tooth example segmentation method based on boundary supervision and multiple attentions | |
| CN112150422B (en) | Modeling method of oral health self-detection model based on multitask learning | |
| Carneiro | Enhanced tooth segmentation algorithm for panoramic radiographs | |
| Erkan et al. | Objective characterization of dental occlusal and fissure morphologies: Method development and exploratory analysis | |
| Fitria et al. | Development of Intraoral Clinical Image Dataset for Deep Learning Caries Detection | |
| CN113792773B (en) | Dental root canal treatment history detection model training method, image processing method and device | |
| CN118247219B (en) | Tooth root external absorption detection system based on deep 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220310 Address after: 226100 No. 690, Xiushan Road, Haimen City, Nantong City, Jiangsu Province Patentee after: Haimen Art Design Co.,Ltd. Address before: 226100 No. 599, Wangjiang South Road, Sanchang street, Haimen District, Nantong City, Jiangsu Province Patentee before: Nantong Linde safety equipment Technology Co.,Ltd. |
|
| TR01 | Transfer of patent right |