An oral disease auxiliary diagnosis system based on big data

Abstract

The present invention relates to the technical field of auxiliary diagnosis of oral diseases, and discloses an auxiliary diagnosis system for oral diseases based on big data. The system collects oral images of patients through X-rays, and calculates oral density values, oral color depth values, oral structure characteristic values, area values ​​of oral lesions and boundary values ​​of oral lesions based on the oral image data, and trains a deep learning model using the above-mentioned calculated values ​​as input parameters, and calculates the degree value of the patient's oral lesion through the deep learning model. The auxiliary diagnosis module generates different diagnostic analysis reports based on the size of the patient's oral lesion degree value, and the result report module transmits the generated diagnostic analysis report to the doctor's work computer screen, so as to assist doctors in improving the accuracy of disease diagnosis through multi-parameter calculation of oral image pictures.

Description

Technical Field

[0001] The present invention relates to the technical field of oral disease auxiliary diagnosis, and in particular to an oral disease auxiliary diagnosis system based on big data. Background Art

[0002] Oral disease refers to various diseases or lesions occurring in the oral cavity, which can lead to problems such as oral pain, bad breath, emotional stress, and oral infection. Oral disease not only harms oral health, but also affects overall health and quality of life. Timely prevention and treatment of oral disease is crucial to maintaining oral health and overall health. Regular oral examinations, maintaining good oral hygiene habits, and timely medical treatment and treatment of oral problems are the key to preventing oral disease.

[0003] Usually, during the initial examination in the hospital's dental department, the doctor observes the patient's actual oral condition with the naked eye and makes a diagnosis based on his or her own clinical experience. Since young doctors do not have rich clinical experience, they may make misjudgments or missed diagnosis when diagnosing oral diseases due to lack of experience, resulting in insufficient accuracy in the diagnosis of oral diseases. Therefore, it is necessary to propose an oral disease auxiliary diagnosis system based on big data. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In view of the shortcomings of the prior art, the present invention provides an oral disease auxiliary diagnosis system based on big data, which is capable of collecting oral images of patients through X-rays, calculating oral density values, oral color depth values, oral structure characteristic values, area values ​​of oral lesions and boundary values ​​of oral lesions based on the oral images, and training a deep learning model with the above-mentioned calculated values ​​as input parameters, and calculating the degree of oral lesions of patients through the deep learning model. The auxiliary diagnosis module will generate different diagnostic analysis reports based on the size of the degree of oral lesions of patients, and the result report module will transmit the generated diagnostic analysis report to the doctor's work computer screen. The purpose of assisting doctors to improve the accuracy of disease diagnosis is achieved through multi-parameter calculation of oral images, thereby solving the above-mentioned problems.

[0006] (II) Technical solution

[0007] To achieve the above-mentioned object, the present invention provides the following technical solutions: an oral disease auxiliary diagnosis system based on big data, comprising a data acquisition module, a data analysis module, a deep learning module, an auxiliary diagnosis module and a result reporting module;

[0008] The data acquisition module is connected to the data analysis module through a network, the data analysis module is connected to the deep learning module through a network, the deep learning module is connected to the auxiliary diagnosis module through a network, and the auxiliary diagnosis module is connected to the result reporting module through a network;

[0009] The data acquisition module is used to collect the patient's oral image data Kqtp, and transmit the collected data to the data analysis module through the network;

[0010] The data analysis module calculates the oral density value Kqmd, the oral color depth value Kqys, the oral structure characteristic value Kqjg, the area value Bbdx of the oral lesion and the boundary value Bbbj of the oral lesion according to the patient's oral image data Kqtp transmitted by the data acquisition module;

[0011] The data analysis module transmits the above calculated values ​​to the deep learning module through the network;

[0012] The deep learning module uses the oral density value Kqmd, the oral color depth value Kqys, the oral structure characteristic value Kqjg, the area value Bbdx of the oral lesion, and the boundary value Bbj of the oral lesion as input parameters to train the deep learning model;

[0013] The deep learning module calculates the patient's oral lesion degree value Bbcd according to the training model, and transmits the calculated value to the auxiliary diagnosis module;

[0014] The auxiliary diagnosis module makes an oral condition diagnosis based on the patient's oral lesion degree value Bbcd, generates a detailed oral condition analysis report, and transmits the report to the result reporting module;

[0015] The result reporting module receives the oral condition analysis report and displays the report on the dentist's computer.

[0016] Preferably, the data acquisition module compiles the collected patient oral image data into a data set through X-ray, and the data set is {Kqtp n-9 ,Kqtp n-8 ,Kqtp n-7 ,...,Kqtp n}, where the convolution algorithm is used to process and calculate the oral image data. The calculation formula is as follows:

[0017] (f*g)(i,j)=∑ m ∑ n f(m,n)*g(im,jn)

[0018] In the formula, (f*g)(i,j) represents the i,jth element in the output matrix generated after the convolution operation, that is, an element in the convolution result matrix, ∑ m ∑ n f(m,n) represents the summation operation of sliding calculation of convolution kernel g on input matrix f. m and n represent the horizontal and vertical offsets of convolution kernel respectively, f(m,n) represents the mth and nth elements in input matrix f, i.e., the pixel values ​​of oral images at corresponding positions in input matrix, g(im,jn) represents the value of convolution kernel g at offset (im,jn), i.e., the weight value of convolution kernel at specified position.

[0019] Preferably, the data analysis module calculates the oral density value Kqmd according to the patient oral image data Kqtp transmitted by the data acquisition module, and the calculation formula is as follows:

[0020]

[0021] In the formula, Kqmd represents the oral density value, Kqtp r Represents the grayscale value of the oral image data, Kqhd represents the grayscale value of the reference oral tissue area. The above values ​​can be obtained by scanning through the data analysis module. The grayscale value of the above oral image data represents the average of the grayscale values ​​of the entire oral image data. The average brightness of the entire oral cavity can be obtained through this value for overall analysis.

[0022] Preferably, the data analysis module calculates the oral color depth value Kqys according to the patient oral image data Kqtp transmitted by the data acquisition module, and the calculation formula is as follows:

[0023]

[0024] In the formula, Kqys represents the oral color depth value, Kqtpc represents the color value of the oral image data, and Qtcy represents the standard color value of the reference oral tissue area. The above values ​​are automatically obtained by the system.

[0025] Preferably, the data analysis module calculates the oral structure characteristic value Kqjg according to the patient's oral image data Kqtp transmitted by the data acquisition module, and the calculation formula is as follows:

[0026] Kqjg=Tzcs1*Qz1+Tzcs2*Qz2+Tzcs3*Qz3+Tzcs4*Qz4

[0027] In the formula, Kqjg represents the characteristic value of the oral structure, Tzcs1, Tzcs2, Tzcs3 and Tzcs4 correspond to the shape of oral teeth, the size of teeth, the number of missing teeth and the distance between the upper and lower jaws, respectively; Qz1, Qz2, Qz3 and Qz4 correspond to the shape weight of oral teeth in oral imaging pictures, the parameter weight of tooth size, the parameter weight of the number of missing teeth and the parameter weight of the distance between the upper and lower jaws, respectively. The above weights can be extracted by combining the system with Matlab image processing software.

[0028] Preferably, the data analysis module calculates the area value Bbdx of the oral lesion according to the patient's oral image data Kqtp transmitted by the data acquisition module, and the calculation formula is as follows:

[0029]

[0030] In the formula, Bbdx represents the area value of the oral lesion, Bmqy represents the number of pixels in the oral lesion area, Xsbl represents the pixel area ratio of the oral image, and the above values ​​are automatically obtained by the system. Kqtp represents the patient's oral image data, and i=1 means that the first value of the patient's oral image data set is used as the initial position of the calculation, and the tenth value of the patient's oral image data set is used as the end position of the calculation.

[0031] Preferably, the data analysis module calculates the boundary value Bbbj of the oral lesion according to the patient's oral image data Kqtp transmitted by the data acquisition module, and the calculation formula is as follows:

[0032] Bbbj=Bjcs*Qxqz+Bjgz*Gzqz

[0033] In the formula, Bbbj represents the boundary value of the oral lesion, Bjcs represents the boundary clarity parameter, Qxqz represents the clarity weight, Bjgz represents the boundary regularity parameter, and Gzqz represents the regularity weight. The data parameters and regularity weight can be obtained through the Numpy library of Python software.

[0034] Preferably, the deep learning module uses the oral density value Kqmd, the oral color depth value Kqys, the oral structure characteristic value Kqjg, the area value Bbdx of the oral lesion, and the boundary value Bbbj of the oral lesion as input parameters to train the deep learning model, and calculates the patient's oral lesion degree value Bbcd through the training model. The calculation formula is as follows:

[0035]

[0036] In the formula, Bbcd represents the degree of oral lesions of the patient, Bbdx represents the area value of the oral lesions, Mj represents the area weight, Kqys represents the oral color depth value, Ys represents the color weight, Bbbj represents the boundary value of the oral lesions, Bj represents the boundary weight, Kqjg represents the oral structure characteristic value, Jg represents the oral structure weight, Kqmd represents the oral density value, and Md represents the density weight. The above weights are obtained by statistical analysis of the patient's oral image data using the statistical analysis method, and the number 5 represents five reference calculation parameters.

[0037] Preferably, the auxiliary diagnosis module generates different detailed oral disease analysis reports according to the patient's oral disease severity value Bbcd. The specific generation method is as follows:

[0038] When the patient's oral lesion severity value Bbcd is greater than or equal to 0.9, it means that the condition is serious and requires expert consultation to study the surgical plan. At this time, the system will remind the doctor that this analysis report cannot be shown to the patient to avoid increasing the patient's pressure;

[0039] When the patient's oral lesion severity value Bbcd is greater than or equal to 0.6 and less than 0.9, it means that the disease is relatively minor and only requires a minor surgery to recover. At this time, the system will prompt the doctor to invite the patient to view this analysis report together to reduce the patient's surgical pressure;

[0040] When the patient's oral lesion severity value Bbcd is less than 0.6, it means that the condition is very mild and does not meet the surgical criteria. The patient only needs to take medicine and rest to recover. At this time, the system will prompt the doctor to invite the patient to view this analysis report and inform the patient not to worry too much about the disease.

[0041] Preferably, the result reporting module receives the oral condition analysis report and displays the report on the dentist's computer, and the dentist informs the patient of the current condition of the disease based on the detailed report.

[0042] Compared with the prior art, the present invention provides an oral disease auxiliary diagnosis system based on big data, which has the following beneficial effects:

[0043] The present invention collects oral images of patients through X-rays, calculates oral density values, oral color depth values, oral structure characteristic values, area values ​​of oral lesions and boundary values ​​of oral lesions through the oral images, and uses the above-mentioned calculated values ​​as input parameters to train a deep learning model, and calculates the patient's oral lesion degree value through the deep learning model. The auxiliary diagnosis module will generate different diagnostic analysis reports based on the size of the patient's oral lesion degree value, and the result report module will transmit the generated diagnostic analysis report to the doctor's work computer screen. The purpose of assisting doctors in improving the accuracy of disease diagnosis is achieved through multi-parameter calculation of oral images. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a schematic diagram of the system flow of the present invention. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] Most of the oral disease auxiliary diagnosis systems on the market do not take advantage of big data technology to establish a deep learning model for oral disease examination, and do not accurately analyze the internal tissue structure of the oral cavity, resulting in a lack of accuracy in oral disease diagnosis. Therefore, a big data-based oral disease auxiliary diagnosis system is proposed. Figure 1 The system consists of a data acquisition module, a data analysis module, a deep learning module, an auxiliary diagnosis module, and a result reporting module, among which:

[0047] The data acquisition module compiles the collected patient oral image data into the data set Kqtp through X-rays, and transmits the collected data to the data analysis module through the network;

[0048] The convolution algorithm is used here to process and calculate the oral image data. The calculation formula is as follows:

[0049] (f*g)(i,j)=∑ m ∑ n f(m,n)*g(im,jn)

[0050] In the formula, the two-bit convolution formula is used to convert the oral image data set. (f*g)(i,j) represents the i,jth element in the output matrix generated after the convolution operation, that is, an element in the convolution result matrix. ∑ m ∑ nf(m,n) represents the summation operation of sliding calculation of convolution kernel g on input matrix f. Among them, m and n represent the horizontal and vertical offsets of convolution kernel respectively, f(m,n) represents the mth and nth elements in input matrix f, that is, the pixel value of the oral image at the corresponding position in the input matrix, g(im,jn) represents the value of convolution kernel g at the offset (im,jn), that is, the weight value of convolution kernel at the specified position. Through the two-dimensional convolution calculation formula, we can multiply the input matrix f and convolution kernel g element by element and sum them to obtain the value of each element in the output matrix, thereby realizing the convolution operation. This process can effectively extract the local features of the input image and help the neural network learn more abstract and complex feature representations, thereby realizing the processing and analysis of oral image data.

[0051] The data analysis module calculates the oral density value Kqmd, the oral color depth value Kqys, the oral structure characteristic value Kqjg, the area value Bbdx of the oral lesion and the boundary value Bddj of the oral lesion based on the patient's oral image data Kqtp transmitted by the data acquisition module, where:

[0052] The oral density value is a parameter of oral digital images. By calculating the density values ​​of teeth and bone tissues in X-ray images, it helps doctors diagnose oral diseases and formulate treatment plans more accurately. For this purpose, the oral density value Kqmd is calculated. The calculation formula is as follows:

[0053]

[0054] Oral density is important for assessing the extent and range of periodontal disease and helps determine the degree of inflammation and the progression of the disease. In the formula, Kqmd represents the oral density, Kqtp represents the oral density, r represents the grayscale value of the oral image data, Kqhd represents the grayscale value of the reference oral tissue area, the above values ​​can be obtained by scanning the data analysis module, the grayscale value of the above oral image data represents the average grayscale value of the entire oral image data, through which the average brightness of the entire oral cavity can be obtained for overall analysis, by calculating Kqtp r -The ratio of Kqhd to Kqhd indicates that all oral tissues in the area are covered, which helps to obtain accurate oral density. The oral density value can help doctors monitor the caries on the tooth surface and the degree of root absorption, which helps to detect and treat problem teeth early;

[0055] The oral color depth value is an important parameter in oral digital images. By calculating the color depth of oral soft tissue, it helps doctors better understand the health status and disease status of oral tissue. For this purpose, it is necessary to calculate the oral color depth value. The calculation formula is as follows:

[0056]

[0057] The oral color depth value can provide detailed information about oral soft tissues, which helps doctors determine the type and severity of oral diseases and develop corresponding treatment plans to improve treatment effects. In the formula, Kqys represents the oral color depth value, Kqtp c Indicates the color value of the oral image data, Qtcy indicates the standard color value of the reference oral tissue area. The above values ​​are automatically obtained by the system and calculated by The value of can clearly reflect the color in the oral inspection area, which helps doctors compare the color of normal oral images;

[0058] The oral structure characteristic value is a part of the oral digital image analysis, which can provide quantitative characteristic information of the oral structure, help doctors to have a more comprehensive understanding of the state of the oral structure, and guide the diagnosis and treatment of oral diseases. To this end, the oral structure characteristic value is calculated, and the calculation formula is as follows:

[0059] Kqjg=Tzcs1*Qz1+Tzcs2*Qz2+Tzcs3*Qz3+Tzcs4*Qz4

[0060] Oral structure characteristic values ​​can reflect the condition of periodontal tissues, help doctors detect the degree of inflammation and gingival recession of periodontal diseases such as periodontitis, and guide the formulation of treatment plans. In the formula, Kqjg represents the oral structure characteristic value, Tzcs1, Tzcs2, Tzcs3 and Tzcs4 respectively correspond to the shape of oral teeth, the size of teeth, the number of missing teeth and the distance between the upper and lower jaws, Qz1, Qz2, Qz3 and Qz4 respectively correspond to the shape weight of oral teeth in oral imaging pictures, the parameter weight of tooth size, the parameter weight of the number of missing teeth and the parameter weight of the distance between the upper and lower jaws. The above weights can be extracted by combining the system with matlab image processing software. According to the relevant parameter calculation of the oral characteristic value, the doctor can evaluate the position, arrangement and distance between adjacent teeth of the teeth, and help find abnormal arrangement or missing problems of the dentition;

[0061] The area value of oral lesions can be used in oral disease research and clinical trials, and provide important data support for scientific research in the field of oral medicine. For this purpose, the area value of oral lesions is calculated using the following formula:

[0062]

[0063] By calculating the area value of oral lesions, doctors can objectively evaluate the size, range and degree of lesions, help determine the severity of lesions, and provide a basis for treatment priority sorting. In the formula, Bbdx represents the area value of oral lesions, Bmqy represents the number of pixels in the oral lesion area, and Xsbl represents the pixel area ratio of the oral image. The above values ​​are automatically obtained by the system. Kqtp represents the patient's oral image data. i=1 means that the first value of the patient's oral image data set is used as the initial calculation position, and the tenth value of the patient's oral image data set is used as the end calculation position. Determine the size of the oral lesion and multiply that value by the original oral image data The sum of the oral lesions can further accurately determine the area of ​​the oral lesions. The area value of the oral lesions can provide a basis for dentists to guide the formulation of reasonable treatment plans, including surgical treatment, drug treatment, laser treatment, etc., to achieve the best treatment effect;

[0064] Analyzing the boundary values ​​of oral lesions can help avoid misdiagnosis and missed diagnosis, reduce the blindness and risk of treatment, and improve the accuracy and reliability of oral lesion diagnosis. To this end, it is necessary to calculate the boundary values ​​of oral lesions. The calculation formula is as follows:

[0065] Bbbj=Bjcs*Qxqz+Bjgz*Gzqz

[0066] Analyzing the boundary value of oral lesions can help doctors determine the nature of lesions, such as benign or malignant, inflammatory or tumorous, and help determine the clinical manifestations and biological characteristics of lesions. In the formula, Bbbj represents the boundary value of oral lesions, Bjcs represents the boundary clarity parameter, Qxqz represents the clarity weight, Bjgz represents the boundary regularity parameter, and Gzqz represents the regularity weight. The data parameters and regularity weight can be obtained through the Numpy library of Python software. By combining the oral-related boundary parameters and weights to calculate the value of Bjcs*Qxqz+Bjhz*Gzqz, the range of the oral lesion boundary can be accurately determined, and the boundary between the lesion and the surrounding normal tissue can be clearly displayed, which helps doctors to more accurately assess the size and location of the lesion;

[0067] By combining the above calculated values, the above values ​​are used as input parameters to train the deep learning model. Since the oral lesion degree value can accurately reflect the severity of the oral condition, the patient's oral lesion degree value Bbcd is calculated based on the model. The calculation formula is as follows:

[0068]

[0069] The degree of oral lesions in patients can help doctors distinguish different types and properties of oral lesions, such as leukoplakia, ulcers, inflammation, and tumors, and help doctors make corresponding diagnosis and treatment decisions. In the formula, Bbcd represents the degree of oral lesions in patients, Bbdx represents the area value of oral lesions, Mj represents the area weight, Kqys represents the oral color depth value, Ys represents the color weight, Bbbj represents the boundary value of oral lesions, Bh represents the boundary weight, Kqjg represents the oral structure characteristic value, Jg represents the oral structure weight, Kqmd represents the oral density value, and Md represents the density weight. The above weights are obtained by statistically analyzing the patient's oral image data using the statistical analysis method. The number 5 represents five reference calculation parameters. By calculating the five oral parameters Obtain accurate values ​​of the degree of oral lesions in patients. By monitoring the degree of oral lesions, doctors can understand the development trend and changes of lesions, assess the progression of the disease, predict follow-up conditions, and make timely treatment adjustments.

[0070] The auxiliary diagnosis module generates different detailed oral disease analysis reports based on the patient's oral disease severity value Bbcd calculated above. The specific generation method is as follows:

[0071] When the patient's oral lesion severity value Bbcd is greater than or equal to 0.9, it means that the condition is serious and requires expert consultation to study the surgical plan. At this time, the system will remind the doctor that this analysis report cannot be shown to the patient, and the doctor will inform the patient of the condition result in a tactful manner to avoid increasing the patient's pressure;

[0072] When the patient's oral lesion severity value Bbcd is greater than or equal to 0.6 and less than 0.9, it means that the disease is relatively minor and only requires a minor surgery to recover. At this time, the system will prompt the doctor to invite the patient to view this analysis report together to reduce the patient's surgical pressure;

[0073] When the patient's oral lesion severity value Bbcd is less than 0.6, it means that the disease is very mild and does not meet the surgical criteria. It can be cured by taking medicine and resting. At this time, the system will prompt the doctor to invite the patient to view this analysis report and tell the patient not to worry too much about the disease.

[0074] The result reporting module receives the oral condition analysis report and displays the report on the dentist's computer. The doctor then informs the patient of the current condition based on the detailed report.

[0075] By combining multi-parameter calculations of oral imaging images and using big data technology to establish a deep learning model, the degree of oral lesions in patients is calculated to reflect the severity of the patient's condition. This value is used to accurately analyze the patient's condition and produce a condition analysis report. Doctors make a diagnosis based on the condition analysis report, thereby improving the accuracy of oral disease diagnosis.

[0076] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An oral disease auxiliary diagnosis system based on big data, characterized by: It includes data acquisition module, data analysis module, deep learning module, auxiliary diagnosis module and result reporting module; The data acquisition module is connected to the data analysis module via a network; The data analysis module is connected to the deep learning module through a network; The deep learning module is connected to the auxiliary diagnosis module through a network; The auxiliary diagnosis module is connected to the result reporting module via a network; The data acquisition module is used to collect the patient's oral image data Kqtp, and transmit the collected data to the data analysis module through the network; The data analysis module calculates the oral density value Kqmd, the oral color depth value Kqys, the oral structure characteristic value Kqjg, the area value Bbdx of the oral lesion and the boundary value Bbbj of the oral lesion according to the patient's oral image data Kqtp transmitted by the data acquisition module; The data analysis module transmits the calculated values ​​to the deep learning module through the network; The deep learning module uses the oral density value Kqmd, the oral color depth value Kqys, the oral structure characteristic value Kqjg, the area value Bbdx of the oral lesion, and the boundary value Bbbj of the oral lesion as input parameters to train the deep learning model; The deep learning module calculates the patient's oral lesion degree value Bbcd according to the training model, and transmits the calculated value to the auxiliary diagnosis module; The auxiliary diagnosis module makes an oral condition diagnosis based on the patient's oral lesion degree value Bbcd, generates a detailed oral condition analysis report, and transmits the report to the result reporting module; The result reporting module receives the oral condition analysis report and displays the report on the dentist's computer.

2. The oral disease auxiliary diagnosis system based on big data according to claim 1, characterized in that: The data acquisition module compiles the collected patient oral image data into a data set through X-ray, and the data set is {Kqtp n-9 ,Kqtp n-8 ,Kqtp n-7 ,...,Kqtp n }, where the convolution algorithm is used to process and calculate the oral image data. The calculation formula is as follows: (f*g)(i,j)=∑ m ∑ n f(m,n)*g(i-m,j-n) In the formula, (f*g)(i,j) represents the i,jth element in the output matrix generated after the convolution operation, that is, an element in the convolution result matrix, ∑ m ∑ n f(m,n) represents the summation operation of sliding calculation on the convolution kernel g on the input matrix f; m and n represent the horizontal and vertical offsets of the convolution kernel respectively, f(m,n) represents the mth and nth elements in the input matrix f, that is, the pixel values ​​of the oral image at the corresponding positions in the input matrix, and g(im,jn) represents the value of the convolution kernel g at the offset (im,jb), that is, the weight value of the convolution kernel at the specified position.

3. The oral disease auxiliary diagnosis system based on big data according to claim 2, characterized in that: The data analysis module calculates the oral density value Kqmd according to the patient's oral image data Kqtp transmitted by the data acquisition module. The calculation formula is as follows: In the formula, Kqmd represents the oral density value, Kqtp r Represents the grayscale value of the oral image data, Kqhd represents the grayscale value of the reference oral tissue area. The above values ​​can be obtained by scanning the data analysis module. The grayscale value of the oral image data represents the average of the grayscale values ​​of the entire oral image data. The average brightness of the entire oral cavity can be obtained through this value for overall analysis.

4. The oral disease auxiliary diagnosis system based on big data according to claim 3 is characterized by: The data analysis module calculates the oral color depth value Kqys according to the patient oral image data Kqtp transmitted by the data acquisition module. The calculation formula is as follows: In the formula, Kqys represents the oral color depth value, Kqtp c It represents the color value of the oral image data, Qtcy represents the standard color value of the reference oral tissue area, and the above values ​​are automatically obtained by the system.

5. The oral disease auxiliary diagnosis system based on big data according to claim 4, characterized in that: The data analysis module calculates the oral structure characteristic value Kqjg according to the patient's oral image data Kqtp transmitted by the data acquisition module. The calculation formula is as follows: Kqjg=Tzcs1*Qz1+Tzcs2*Qz2+Tzcs3*Qz3+Tzcs4*Qz4 In the formula, Kqjq represents the characteristic value of the oral structure, Tzcs1, Tzcs2, Tzcs3 and Tzcs4 correspond to the shape of oral teeth, the size of teeth, the number of missing teeth and the distance between the upper and lower jaws, respectively; Qz1, Qz2, Qz3 and Qz4 correspond to the shape weight of oral teeth in oral imaging pictures, the parameter weight of tooth size, the parameter weight of the number of missing teeth and the parameter weight of the distance between the upper and lower jaws, respectively. The above weights can be extracted by combining the system with Matlab image processing software.

6. The oral disease auxiliary diagnosis system based on big data according to claim 5, characterized in that: The data analysis module calculates the area value Bbdx of the oral lesion according to the patient's oral image data Kqtp transmitted by the data acquisition module. The calculation formula is as follows: In the formula, Bbdx represents the area value of the oral lesion, Bmqy represents the number of pixels in the oral lesion area, Xsbl represents the pixel area ratio of the oral image, and the above values ​​are automatically obtained by the system. Kqtp represents the patient's oral image data, and i=1 means that the first value of the patient's oral image data set is used as the initial position of the calculation, and the tenth value of the patient's oral image data set is used as the end position of the calculation.

7. The oral disease auxiliary diagnosis system based on big data according to claim 6, characterized in that: The data analysis module calculates the boundary value Bbbj of the oral lesion according to the patient's oral image data Kqtp transmitted by the data acquisition module. The calculation formula is as follows: Bbbj=Bjcs*Qxqz+Bjgz*Gzqz In the formula, Bbbj represents the boundary value of the oral lesion, Bjcs represents the boundary clarity parameter, Qxqz represents the clarity weight, Bjgz represents the boundary regularity parameter, and Gzqz represents the regularity weight. The data parameters and regularity weight can be obtained through the Numpy library of Python software.

8. The oral disease auxiliary diagnosis system based on big data according to claim 7, characterized in that: The deep learning module uses the oral density value Kqmd, the oral color depth value Kqys, the oral structure characteristic value Kqjg, the area value Bbdx of the oral lesion, and the boundary value Bbbj of the oral lesion as input parameters to train the deep learning model, and calculates the patient's oral lesion degree value Bbcd through the training model. The calculation formula is as follows: In the formula, Bbcd represents the degree of oral lesions of the patient, Bbdx represents the area value of the oral lesions, Mj represents the area weight, Kqys represents the oral color depth value, Ys represents the color weight, Bbbj represents the boundary value of the oral lesions, Bj represents the boundary weight, Kqjg represents the oral structure characteristic value, Jg represents the oral structure weight, Kqmd represents the oral density value, and Md represents the density weight. The above weights are obtained by statistical analysis of the patient's oral image data using the statistical analysis method, and the number 5 represents five reference calculation parameters.

9. The oral disease auxiliary diagnosis system based on big data according to claim 8, characterized in that: The auxiliary diagnosis module generates different detailed oral disease analysis reports according to the patient's oral disease severity value Bdcd. The specific generation method is as follows: When the patient's oral lesion severity value Bdcd is greater than or equal to 0.9, it means that the condition is serious and requires expert consultation to study surgical plans; When the patient's oral lesion severity value Bdcd is greater than or equal to 0.6 and less than 0.9, it means that the disease is relatively mild and only requires a minor surgery to achieve recovery; When the patient's oral lesion severity value Bbcd is less than 0.6, it means that the condition is very mild and does not meet the surgical criteria. The patient only needs to take medicine and rest to recover.

10. The oral disease auxiliary diagnosis system based on big data according to claim 9, characterized in that: The result reporting module receives the oral condition analysis report, and the doctor makes a diagnosis and analysis of the patient's condition based on the oral condition analysis report, and informs the patient of the current condition in detail.

Images