Cone-beam oral CBCT image reconstruction system and method

Abstract

The present invention discloses a cone-beam oral CBCT image reconstruction system and method; the method comprises: calibrating a detector; collecting projection data of a patient's oral tissue preliminarily collected at different angles based on an X-ray emission device, a calibrated detector and an auxiliary device; obtaining preprocessed projection data through adaptive filtering and artifact correction; obtaining compensated projection data through correction and detector response compensation; reconstructing an oral CBCT image based on a hybrid reconstruction method in combination with the compensated projection data to obtain a first oral CBCT image; and automatically adjusting the grayscale distribution of the first oral CBCT image by using an enhancement technology based on histogram equalization to obtain a reconstructed image; the present invention reconstructs an image by using a hybrid reconstruction method in combination with the compensated projection data, and automatically adjusting the grayscale distribution of the reconstructed image by using an enhancement technology based on histogram equalization, thereby improving the quality of the reconstructed image.

Description

Technical Field

[0001] The present invention relates to the technical field of image reconstruction, and more specifically, to a system and method for cone-beam oral CBCT image reconstruction. Background Art

[0002] With the continuous development of stomatology, the requirements for accurate diagnosis of the internal structure of the oral cavity are increasing. Traditional two-dimensional oral imaging technology (such as oral X-rays) has limitations in showing the complex three-dimensional structure of the oral cavity. Oral CBCT (cone beam computed tomography) technology came into being. It can provide three-dimensional images of the oral cavity, which is of great significance for diagnosis and treatment planning in many fields such as orthodontics, implants, and maxillofacial surgery.

[0003] The Chinese patent application with the publication number CN108182665A discloses a CT system image reconstruction method based on a filtered back-projection-iterative algorithm, comprising: the first step: using the CT system to measure the material to be measured to obtain the projection value data of the object to be measured; the second step: using the filter function to filter the acquired data, and using the back-projection algorithm to solve the attenuation coefficient matrix of the object to be measured; the third step: using the attenuation coefficient matrix obtained by the back-projection algorithm as the initial iteration value of the iterative algorithm to obtain the corrected attenuation coefficient matrix, and completing the image reconstruction of the object to be measured. The invention introduces a filter function, combines the back-projection algorithm with the iterative algorithm, and becomes a filtered back-projection-iterative algorithm. First, the environmental noise is filtered out by using the filter function, and the solution result of the back-projection algorithm is used as the initial iteration value of the iterative algorithm, which can effectively improve the convergence speed of the algorithm. The use of the filtered back-projection-iterative algorithm makes up for the shortcomings of the iterative algorithm, such as poor anti-noise performance and slow imaging speed, and can quickly and stably obtain high-resolution object reconstruction images in a noisy environment, and can be widely used in medical, engineering and other fields.

[0004] Although the above method can meet most scenarios, research and practical application of the above method and existing technology have found that the above method and existing technology have at least the following defects:

[0005] Lack of detector calibration may cause detector-related errors to be directly transmitted to the projection data; no processing steps are performed on the projection data; the attenuation coefficient has a low accuracy, which causes deviations in the reconstruction of the internal structure and fails to truly reflect the density changes inside the oral tissue.

[0006] In view of this, the present invention proposes a cone-beam oral CBCT image reconstruction system and method to solve the above problems. Summary of the invention

[0007] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: a system and method for reconstructing an image based on a cone beam oral CBCT, comprising the following steps:

[0008] Calibrate the detector;

[0009] Based on X-ray emitting equipment, calibrated detectors and auxiliary devices, the patient's oral tissue is collected at different angles The projection data under the condition of The value range is ;

[0010] Performing adaptive filtering and artifact correction processing on the initially collected projection data to obtain pre-processed projection data;

[0011] Correcting the preprocessed projection data to obtain corrected projection data, and performing detector response compensation on the corrected projection data to obtain compensated projection data;

[0012] Based on the hybrid reconstruction method, the oral CBCT image is reconstructed in combination with the compensated projection data to obtain a first oral CBCT image;

[0013] The enhancement technology based on histogram equalization is used to automatically adjust the grayscale distribution of the first oral CBCT image to obtain a reconstructed oral CBCT image.

[0014] Furthermore, the method for obtaining the corrected projection data includes:

[0015] Step 1: Analyze the patient's oral tissue at different angles The pre-processed projection data is logarithmically transformed based on the transformation formula; and the transformed projection data is obtained;

[0016] Step 2, based on the method of scanning the patient's oral cavity with rays, the transformed projection data is corrected by the attenuation change of the rays in the patient's oral tissue to obtain corrected projection data;

[0017] The method for obtaining the compensated projection data includes:

[0018] Compensated projection data are obtained based on the ratio of the corrected projection data to the detector response function.

[0019] Further, the obtaining Methods for calculating the corrected attenuation coefficient for each oral tissue include:

[0020] Step 2.1, calculate the total linear attenuation coefficient;

[0021] Step 2.2: Based on Monte Carlo simulation and total linear attenuation coefficient calculation, the The corrected attenuation coefficients corresponding to the oral tissues.

[0022] Furthermore, the method for calculating the total linear attenuation coefficient includes:

[0023] Step 2.1.1: Based on the principle of photoelectric effect, The density of oral tissue, The molar mass of the oral tissue In oral tissue The atomic number fraction and energy of the element are The photon and atomic number are No. Calculation of the cross section of the photoelectric effect of an element Photoelectric effect attenuation coefficient of oral tissue ;

[0024] Step 2.1.2: Based on Compton scattering, combined with The density of oral tissue, The molar mass of the oral tissue In oral tissue The atomic number fraction and The energy of oral tissue is Compton scattering cross section calculation Compton scattering attenuation coefficient of oral tissue ;

[0025] Step 2.1.3: Calculate the first The total linear attenuation coefficient of the oral tissue; The total linear attenuation coefficient of the oral tissue is Photoelectric effect attenuation coefficient of oral tissue With Compton scattering attenuation coefficient of oral tissue The sum.

[0026] Further, the Monte Carlo simulation is combined with the total linear attenuation coefficient to obtain the first Methods for calculating the corrected attenuation coefficient for each oral tissue include:

[0027] Step 2.2.1: Initialize the energy of the photons emitted by the ray source so that the energy of the photons emitted by the ray source is Normal distribution ,in, is the average energy of photons emitted by the ray source, is the standard deviation of the photon energy emitted by the ray source;

[0028] Step 2.2.2: Calculate the probability of the occurrence of photons emitted by the ray source in the first step based on the Monte Carlo simulation method of simulating the probability of events in the physical process by random numbers. The probability of transmission and interaction in oral tissues;

[0029] Step 2.2.3: When the ray source emits a photon When the ray source interacts with the first oral tissue, the interaction type is determined. The interaction types include photoelectric effect and Compton scattering. When it is determined to be the photoelectric effect, the photons emitted by the ray source are absorbed, photoelectrons are generated, and energy is transferred to the photoelectrons. When it is determined to be the Compton scattering, the photons emitted by the ray source interact with the first oral tissue. The electrons in the atoms in the oral tissue collide, and the radiation source emits photons to change the propagation direction and transfer energy to the electrons. The energy of the photons emitted by the radiation source after scattering is calculated based on the Compton scattering formula, combined with the electron rest energy and the scattering angle;

[0030] Step 2.2.4: Record and track the photons emitted by the ray source. The propagation path in the oral tissue is When the energy of the photons emitted by the oral tissue or the scattered radiation source is less than the preset threshold, the tracking of the photons emitted by the radiation source is stopped; and the number of photons emitted by the radiation source received by the detector is counted. ;

[0031] Step 2.2.5: Calculate The corrected attenuation coefficient corresponding to each oral tissue;

[0032] Calculate the Methods for calculating the corrected attenuation coefficient for each oral tissue include:

[0033] Based on the Beer-Lambert law, the expected number of photons emitted by the source received by the detector without oral tissue attenuation and the number of photons emitted by the source are calculated. The total path length propagated in the oral tissue is calculated The corrected attenuation coefficients corresponding to the oral tissues.

[0034] Furthermore, the method for obtaining the first oral CBCT image includes:

[0035] Step A, performing a back-projection operation on the corrected projection data based on a back-projection algorithm to obtain a first reconstructed image;

[0036] The method of obtaining a first reconstructed image comprises:

[0037] Step A.1: Select the impulse response function as The filter compensates the projection data Performing filtering processing to obtain projection data after filtering processing;

[0038] Step A.2, obtaining a first reconstructed image by performing an integral back-projection operation on the filtered projection data;

[0039] Step B: performing iterative reconstruction optimization on the first reconstructed image based on an iterative reconstruction method to obtain a first oral CBCT image.

[0040] Furthermore, the method of iteratively reconstructing and optimizing the first reconstructed image based on the iterative reconstruction method to obtain the first oral CBCT image includes:

[0041] Step B.1: Reconstruct the first image Defined as the initial image estimate , based on the initial image estimate Constructing a forward projection model based on a ray physics model , used to estimate based on the current image Calculate the predicted projection data; among them, the forward projection model Calculate the forward projection model as the sum of the projection data and the attenuated ray intensity The difference between the projection data and the actual projection data is used to obtain the projection data residual. The method for obtaining the projection data residual includes: ;in, For the Iterative forward projection model The residual error with the actual projection data; is the transformed projection data; is the number of iterations;

[0042] Step B.2: Based on the projection data residual Use gradient descent method for optimization and introduce step size parameter , update the image estimate;

[0043] Step B.3: Set the iteration termination condition, stop the iteration when the termination condition is reached, and output the reconstructed image corresponding to the minimum residual. As the first oral CBCT image.

[0044] Furthermore, the method for obtaining the reconstructed oral CBCT image includes:

[0045] Step E: By analyzing the first oral CBCT image The corresponding grayscale histogram function is obtained by integrating in different directions , ,in, is the gray value; is the total number of gray levels;

[0046] Step F, obtaining a probability density function of the gray level by calculating the gray level histogram function, the number of pixel rows and the number of pixel columns of the first oral CBCT image;

[0047] Step G, obtaining a cumulative distribution function by integrating the probability density function of the gray level;

[0048] Step H: Set the grayscale mapping function , the grayscale mapping function The first oral CBCT image is enhanced by rounding off and combining the cumulative distribution function; the enhanced first oral CBCT image As the reconstructed oral CBCT image.

[0049] Furthermore, the method of performing adaptive filtering on the initially collected projection data includes:

[0050] Step a: The initially collected projection data is divided into The window is set according to the preset step size Traverse and divide the initially collected projection data into The projection data contained in each local area;

[0051] Step b: Calculate The mean and standard deviation of the projected data points in each local area;

[0052] Step c, performing adaptive Gaussian filtering on the initially collected projection data based on the standard deviation combined with the adjustment coefficient and Gaussian filtering;

[0053] Step d, set the number of iterations, repeat steps a to c during each iteration, calculate the mean square error between the filtered projection data and the initially collected projection data after each iteration, and compare it with a preset error threshold, when the mean square error is less than the preset error threshold or reaches the maximum number of iterations, stop the iteration, and mark the initially collected projection data corresponding to the minimum mean square error as the projection data after adaptive filtering.

[0054] Furthermore, the method for performing artifact correction on the projection data after the adaptive filtering process includes:

[0055] A three-dimensional rectangular coordinate system is established with the detector position as the origin, based on the initial velocity of the patient in three axes collected by the auxiliary device. , and and initial displacement , and , and at the time Acceleration , and ;

[0056] According to the initial speed of the three axes and at time Acceleration calculation time Three axial speeds;

[0057] Calculate the time according to the velocity and initial displacement of the three axes Displacement in three axes;

[0058] Based on the calculated displacements in the three axes, a compensation operation is performed on the projection data to obtain pre-processed projection data.

[0059] Furthermore, the method for calibrating the detector includes:

[0060] The original output signal of the detector is corrected by the gain coefficient to obtain a corrected signal.

[0061] Based on the cone beam oral CBCT image reconstruction system, a cone beam oral CBCT image reconstruction method is implemented, including:

[0062] Equipment calibration module: used to calibrate the detector;

[0063] Data acquisition module: used to collect data of the patient's oral tissue at different angles based on X-ray emission equipment, calibrated detectors and auxiliary devices. The projection data under the condition of The value range is ;

[0064] Data processing module: used to perform adaptive filtering and artifact correction processing on the initially collected projection data to obtain pre-processed projection data;

[0065] Data correction module: used for correcting the preprocessed projection data to obtain corrected projection data, performing detector response compensation on the corrected projection data to obtain compensated projection data;

[0066] Image reconstruction module: used to reconstruct the oral CBCT image based on the hybrid reconstruction method and the compensated projection data to obtain the first oral CBCT image;

[0067] Image enhancement module: used to automatically adjust the grayscale distribution of the first oral CBCT image by adopting an enhancement technology based on histogram equalization to obtain a reconstructed oral CBCT image.

[0068] The technical effects and advantages of the cone-beam oral CBCT image reconstruction system and method of the present invention are as follows:

[0069] The present invention can effectively reduce the error of the detector itself by calibrating the detector, and can more realistically reflect the actual attenuation of the ray, thereby providing a more reliable data basis for subsequent image reconstruction; the pre-processing of the initially collected projection data such as adaptive filtering and artifact correction can significantly improve the quality of the projection data; the pre-processed projection data is corrected and the detector response is compensated, which can further optimize the projection data; the attenuation coefficient is solved by using the ray physics model built by different oral tissue components in the oral cavity and Monte Carlo simulation. Monte Carlo simulation can accurately calculate the attenuation coefficient corresponding to different oral tissues in the oral cavity by simulating the interaction process between a large number of rays and oral tissue components, so as to more accurately present the boundaries and internal structures of these oral tissues in the reconstructed image. The hybrid reconstruction method is used to reconstruct the oral CBCT image in combination with the compensated projection data, which combines the advantages of different reconstruction algorithms. The grayscale distribution of the reconstructed oral CBCT image is automatically adjusted by the enhancement technology based on histogram equalization, which enhances the contrast of the image and further improves the quality of the reconstructed image. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 It is a schematic diagram of the process of the cone-beam oral CBCT image reconstruction method of the present invention;

[0071] Figure 2 It is a schematic flow chart of a method for obtaining corrected projection data according to the present invention;

[0072] Figure 3 This is a block diagram of the cone-beam oral CBCT image reconstruction system of the present invention. DETAILED DESCRIPTION

[0073] 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.

[0074] Embodiment 1:

[0075] See also Figure 1 As shown, the cone beam oral CBCT image reconstruction method described in this embodiment includes the following steps:

[0076] Calibrate the detector.

[0077] Methods for calibrating the detector include:

[0078] Assume the original output signal of the detector is , the dark current signal is , the gain factor is , the corrected signal Calculated as:

[0079]

[0080] in, is the time sampling point.

[0081] The detector's response signal to X-rays directly affects the grayscale value of the reconstructed oral CBCT image. By calibrating the detector, it can be ensured that its response to X-rays of different intensities is linear and accurate, thereby ensuring the image quality of the reconstructed oral CBCT image.

[0082] Used to collect patient oral tissue at different angles based on X-ray emitting equipment, calibrated detectors and auxiliary devices The projection data under the condition of The value range is Specifically, the patient is seated on the scanning chair of the oral CBCT device and the patient's head is fixed with a fixing device so that the patient's head is in a stable and repeatable position to collect projection data. By fixing the patient's head to collect projection data, motion artifacts can be reduced, the consistency and repeatability of projection data can be ensured, the accuracy and efficiency of the reconstruction algorithm can be improved, and the quality of the later reconstructed oral CBCT images can be improved.

[0083] The initially collected projection data are subjected to adaptive filtering processing and artifact correction processing to obtain pre-processed projection data.

[0084] The method of performing adaptive filtering on the initially collected projection data includes:

[0085] Step a: The initially collected projection data is divided into The window is set according to the preset step size Traverse and divide the initially collected projection data into The local area contains the projection data.

[0086] Step b: Calculate The mean and standard deviation of the projected data points in each local area.

[0087] Step c: performing adaptive Gaussian filtering on the initially collected projection data based on the standard deviation combined with the adjustment coefficient and Gaussian filtering.

[0088] The method of performing adaptive Gaussian filtering on the initially collected projection data includes: ;in, For the Local area The standard deviation of the initial collected projection data; is the adjustment factor; is the preset standard deviation threshold; is the initial Gaussian filter standard deviation; After adjustment Local area The Gaussian standard deviation of the initial collected projection data;

[0089] Compute the Gaussian filter function:

[0090] ;

[0091] in, is the Gaussian filter function, is the pixel coordinate of the projection data, Represents an exponential function with base e.

[0092] ;

[0093] in, For local areas Inner Projection Data Points The result after Gaussian filtering is: For local area The median coordinate is The projection data pixel position, Represented as a variable From 1 to Perform a sum operation, Represented as a variable From 1 to Perform a sum operation, is the pixel coordinate of the projection data; is the pixel coordinates of the projection data and The difference The result obtained by Gaussian filtering.

[0094] Step d, set the number of iterations, repeat steps a to c during each iteration, calculate the mean square error between the filtered projection data and the initially collected projection data after each iteration, and compare it with a preset error threshold, when the mean square error is less than the preset error threshold or reaches the maximum number of iterations, stop the iteration, and mark the initially collected projection data corresponding to the minimum mean square error as the projection data after adaptive filtering.

[0095] Performing the above-mentioned adaptive filtering processing on the initially collected projection data can effectively reduce the influence of noise, improve the quality of projection data, and improve the quality of the later reconstructed oral CBCT images.

[0096] The method for performing artifact correction on the projection data after adaptive filtering processing includes:

[0097] A three-dimensional rectangular coordinate system is established with the detector position as the origin, based on the initial velocity of the patient in three axes collected by the auxiliary device. , and and initial displacement , and , and at the time Acceleration , and .

[0098] Calculation time Three axial speeds:

[0099] ;

[0100] ;

[0101] ;

[0102] in, , and Separately for the moment Patients in axis, Axis and The speed of the axis; , and Separately for the moment patient axis, Axis and The acceleration of the axis; The integral variable is , The value range is ;

[0103] Calculation time Displacement in three axes:

[0104] ;

[0105] ;

[0106] ;

[0107] in, , and Separately for the moment Patients in axis, Axis and Displacement of the axis; , and Separately for the moment Patients in axis, Axis and The speed of the axis;

[0108] Based on the calculated displacements in the three axes, a compensation operation is performed on the projection data to obtain pre-processed projection data.

[0109] For example, during the projection data acquisition process, a certain pixel During the scan, Axial displacement , then its original position in the projection data Adjust to , to offset the projection deviation caused by movement.

[0110] Correcting the preprocessed projection data to obtain corrected projection data, and performing detector response compensation on the corrected projection data to obtain compensated projection data;

[0111] Reference Figure 2 , the method of obtaining the corrected projection data includes:

[0112] Step 1: Analyze the patient's oral tissue at different angles The preprocessed projection data is logarithmically transformed based on the transformation formula to convert the multiplication effect of ray intensity into an additive effect for subsequent processing.

[0113] The transformation formula is as follows:

[0114] ;

[0115] in, is the transformed projection data; is the sampling location; is the preprocessed projection data.

[0116] Step 2: Correct the transformed projection data to obtain corrected projection data.

[0117] Methods for obtaining corrected projection data include:

[0118] ;

[0119] in, is the corrected projection data; The number of different oral tissues included in the path of the projection ray during the process of scanning the patient's oral cavity; For the The thickness of oral tissue; For the The corrected attenuation coefficients corresponding to the oral tissues.

[0120] Oral CBCT image reconstruction relies on accurate estimation of the oral tissue attenuation coefficient, which reflects the attenuation of X-rays when they pass through oral tissue and is a key parameter in the reconstructed image that reflects the characteristics of oral tissue. By correcting the attenuation coefficient, the subsequent reconstruction of oral CBCT images can more accurately process projection data based on the corrected attenuation coefficient. This makes the reconstructed oral CBCT images more accurate in terms of oral tissue boundaries, internal structures, etc.

[0121] Get the Methods for calculating the corrected attenuation coefficient for each oral tissue include:

[0122] Step 2.1. Calculate the total linear attenuation coefficient.

[0123] Methods for calculating the total linear attenuation factor include:

[0124] Step 2.1.1. According to the principle of photoelectric effect, calculate the Photoelectric effect attenuation coefficient of oral tissue :

[0125] ;

[0126] ;

[0127] in, For the The molar mass of oral tissue; For the Density of oral tissues; is Avogadro's constant; For the In oral tissue The atomic number fraction of the element; For the In oral tissue The atomic number of an element; For the The total number of atoms of all elements in oral tissues; is the cross section of the photoelectric effect; is the X-ray photon energy; For the The atomic number of the element; For energy The photon and atomic number are No. The cross section of an element undergoing the photoelectric effect.

[0128] Step 2.1.2: Calculate the Compton scattering attenuation coefficient of oral tissue :

[0129] ;

[0130] in, For the Compton scattering cross section of oral tissue; For the The energy of oral tissue is The cross section for Compton scattering occurs below.

[0131] Step 2.1.3: Calculate the first Total linear attenuation coefficient of oral tissues:

[0132] ;

[0133] in, For the The total linear attenuation coefficient of all oral tissues.

[0134] Step 2.2: Based on Monte Carlo simulation and total linear attenuation coefficient calculation, the The corrected attenuation coefficients corresponding to the oral tissues.

[0135] Get the Methods for calculating the corrected attenuation coefficient for each oral tissue include:

[0136] Step 2.2.1: Initialize the energy of the photons emitted by the ray source so that the ray source emits photon energy Normal distribution ,in, is the average energy of photons emitted by the ray source, is the standard deviation of the photon energy emitted by the ray source; the emission angles are evenly distributed within a certain range, for example, within a cone angle Evenly distributed inside, Determined by the geometry of the scanning device. For each emitted photon, its position is initialized ,energy and direction of propagation .

[0137] Step 2.2.2: Calculate the probability of photons in the first The probability of transmission and interaction in oral tissues.

[0138] Photons in the Distance traveled in oral tissue The probability density function of for:

[0139] ;

[0140] in, is a mathematical constant; For the The total linear attenuation coefficient of each oral tissue; For photons The distance traveled in the oral tissue.

[0141] Step 2.2.3: When the ray source emits a photon When the ray source interacts with the first oral tissue, the interaction type is determined. The interaction types include photoelectric effect and Compton scattering. When it is determined to be the photoelectric effect, the photons emitted by the ray source are absorbed, photoelectrons are generated, and energy is transferred to the photoelectrons. When it is determined to be the Compton scattering, the photons emitted by the ray source interact with the first oral tissue. The electrons in the atoms in the oral tissue collide, and the radiation source emits photons to change the propagation direction and transfer energy to the electrons. The energy of the scattered photons emitted by the radiation source is calculated according to the Compton scattering formula. , the Compton scattering formula is: ;in, is the electron rest energy; is the scattering angle, and the distribution of the scattering angle obeys the Klein-Nishner formula , For the direction of propagation, is the electron radius.

[0142] Step 2.2.4: Record and track the photons emitted by the ray source. The propagation path in the oral tissue is The oral tissue or scattered radiation source emits photon energy When the number of photons emitted by the ray source is less than the preset threshold, the tracking of the photons emitted by the ray source is stopped; and the number of photons emitted by the ray source received by the detector is counted. .

[0143] Step 2.2.5: Calculate The corrected attenuation coefficients corresponding to the oral tissues.

[0144] Calculate the Methods for calculating the corrected attenuation coefficient for each oral tissue include:

[0145] Based on the Beer-Lambert law, the Corrected attenuation coefficient corresponding to oral tissue :

[0146] ;

[0147] ;

[0148] in, is the number of photons emitted by the expected radiation source received by the detector when there is no oral tissue attenuation; The ray source emits photons at Path length of propagation in oral tissues; The ray source emits photons in the propagation direction The photon on the Distance traveled in oral tissue Perform integral operation.

[0149] Methods for obtaining compensated projection data include:

[0150] The detector response compensation is performed on the corrected projection data based on the detector response function, and the compensated projection data is calculated as follows:

[0151] ;

[0152] in, is the projection data after compensation; is the detector response function, which is obtained through actual measurement. Combining the detector response function with the corrected projection data for detector response compensation can further improve the accuracy of the projection data and provide an accurate data basis for subsequent more accurate oral CBCT image reconstruction.

[0153] Based on the hybrid reconstruction method, the oral CBCT image is reconstructed in combination with the compensated projection data to obtain a first oral CBCT image;

[0154] The method of obtaining the first oral CBCT image includes:

[0155] Step A: performing a back-projection operation on the corrected projection data based on a back-projection algorithm to obtain a first reconstructed image.

[0156] The method of obtaining a first reconstructed image comprises:

[0157] Step A.1: Select the impulse response function as The filter compensates the projection data Perform filtering processing to obtain projection data after filtering.

[0158] Methods for obtaining filtered projection data include:

[0159] ;

[0160] in, is the projection data after filtering; is the convolution operation;

[0161] Step A.2: Perform a back-projection operation on the filtered projection data to obtain a first reconstructed image.

[0162] The method of obtaining a first reconstructed image comprises:

[0163] ;

[0164] in, is the first reconstructed image; Represents the projection data after filtering Different angles exist Perform integral operation on The value range corresponds to The value of The image space coordinates are integrated back-projected on the filtered data of all projection angles to reconstruct the complete structure of the oral tissue inside the mouth, including the morphology of teeth, jaws, soft oral tissues, etc. This facilitates further iteration and enhancement of the image in the later stage, and more accurate structural reconstruction.

[0165] Step B: performing iterative reconstruction optimization on the first reconstructed image based on an iterative reconstruction method to obtain a first oral CBCT image.

[0166] The method for iteratively reconstructing and optimizing the first reconstructed image based on the iterative reconstruction method to obtain the first oral CBCT image includes:

[0167] Step B.1: Reconstruct the first image Defined as the initial image estimate , based on the initial image estimate Constructing a forward projection model based on a ray physics model , used to estimate based on the current image Calculate the predicted projection data; among them, the forward projection model Calculate the forward projection model as the sum of the projection data and the attenuated ray intensity The difference between the actual projection data and the projection data residual is obtained.

[0168] Methods for obtaining projection data residuals include:

[0169] ;

[0170] in, For the Iterative forward projection model The residual with the actual projection data; is the transformed projection data; is the number of iterations.

[0171] Step B.2: Based on the projection data residual Update the image estimate, optimize it using the gradient descent method, and introduce the step size parameter , the update formula is:

[0172] ;

[0173] in, For the Image estimation for iterations; For the Image estimation for iterations; is the forward projection model for the Iteration image estimation The derivative of is obtained by taking the derivative of the forward projection model. In the iterative process, the image estimation is continuously adjusted to make the predicted projection data closer and closer to the actual projection data. At the same time, the physical propagation characteristics of the rays in different oral tissues are considered to optimize the reconstruction results.

[0174] Step B.3: Set the iteration termination condition, such as reaching the predetermined number of iterations , or when the RMS value of the residual is less than the preset iteration threshold or other termination conditions, the iteration is stopped and the reconstructed image corresponding to the minimum residual is output As the first oral CBCT image.

[0175] The enhancement technology based on histogram equalization is used to automatically adjust the grayscale distribution of the first oral CBCT image to obtain a reconstructed oral CBCT image.

[0176] The method of obtaining the reconstructed oral CBCT image includes:

[0177] Step E: Based on the first oral CBCT image acquired , the gray value range of the image is , get the grayscale histogram function of the first oral CBCT image , ,in, is the gray value; is the total number of gray levels; grayscale histogram function Indicates that the gray value in the image is The number of pixels,

[0178] Get grayscale histogram function The methods include:

[0179] ;

[0180] in, is the Kronecker function, when hour, ;otherwise .

[0181] Step F, calculating the probability density function of the gray level;

[0182] Methods for calculating the probability density function of gray levels include:

[0183] ;

[0184] in, is the probability density function of the gray level, indicating that the gray value is The probability of a pixel appearing in the image; and are the number of pixel rows and pixel columns of the first oral CBCT image, respectively.

[0185] Step G, calculating the cumulative distribution function;

[0186] Methods for calculating the cumulative distribution function include:

[0187] ;

[0188] in, is the cumulative distribution function, indicating that the gray value is less than or equal to Pixels The proportion of the first oral CBCT image, specifically, all gray values ​​in the gray level probability density function of the first oral CBCT image are less than or equal to the gray value Pixels The probability density function of the gray level exist The summed value on .

[0189] Step H: Set the grayscale mapping function , the first oral CBCT image is enhanced.

[0190] The method for enhancing the first oral cavity CBCT image includes:

[0191] Set the grayscale mapping function as follows:

[0192] ;

[0193] in, Is the rounding function. Grayscale mapping function Can be used to convert the original grayscale value Map to new grayscale values , making the grayscale distribution of the first oral CBCT image more uniform.

[0194] According to the grayscale mapping function Enhance the first oral CBCT image:

[0195] ;

[0196] in, is the enhanced first oral CBCT image. As the reconstructed oral CBCT image.

[0197] In this way, the grayscale distribution of the image is automatically adjusted, making the details of the originally darker or brighter areas clearer and the contrast improved. For example, in the case of a small grayscale difference between teeth and surrounding oral tissues in oral CBCT images, histogram equalization can enhance this difference, making details such as tooth structure and soft oral tissue more obvious, which is beneficial for the diagnosis of oral diseases.

[0198] Embodiment 2:

[0199] See also Figure 3 As shown, the cone beam oral CBCT image reconstruction system described in this embodiment includes:

[0200] Equipment calibration module: used to calibrate the detector;

[0201] Data acquisition module: used to collect data of the patient's oral tissue at different angles based on X-ray emission equipment, calibrated detectors and auxiliary devices. The projection data under the condition of The value range is ;

[0202] Data processing module: used to perform adaptive filtering and artifact correction processing on the initially collected projection data to obtain pre-processed projection data;

[0203] Data correction module: used to correct the pre-processed projection data to obtain corrected projection data, and also to perform detector response compensation on the corrected projection data to obtain compensated projection data;

[0204] Image reconstruction module: used to reconstruct the oral CBCT image based on the hybrid reconstruction method and the compensated projection data to obtain the first oral CBCT image;

[0205] Image enhancement module: used to automatically adjust the grayscale distribution of the first oral CBCT image by adopting an enhancement technology based on histogram equalization to obtain a reconstructed oral CBCT image.

[0206] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0207] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A cone-beam oral CBCT image reconstruction method, characterized in that: The steps include: Calibrate the detector; Based on X-ray emitting equipment, calibrated detectors and auxiliary devices, the patient's oral tissue is collected at different angles The projection data under the condition of The value range is ; Performing adaptive filtering and artifact correction processing on the initially collected projection data to obtain pre-processed projection data; Correcting the preprocessed projection data to obtain corrected projection data, and performing detector response compensation on the corrected projection data to obtain compensated projection data; Based on the hybrid reconstruction method, the oral CBCT image is reconstructed in combination with the compensated projection data to obtain a first oral CBCT image; The enhancement technology based on histogram equalization is used to automatically adjust the grayscale distribution of the first oral CBCT image to obtain a reconstructed oral CBCT image; The method for calculating the total linear attenuation coefficient in the corrected projection data includes: According to the principle of photoelectric effect, combined with the The density of oral tissue, The molar mass of the oral tissue, In oral tissue The atomic number fraction and energy of the element are The photon and atomic number are No. Calculation of the cross section of the photoelectric effect of an element Photoelectric effect attenuation coefficient of oral tissue ; According to Compton scattering, combined with the The density of oral tissue, The molar mass of the oral tissue In oral tissue The atomic number fraction and The energy of oral tissue is Compton scattering cross section calculation Compton scattering attenuation coefficient of oral tissue ; According to the photoelectric effect attenuation coefficient and Compton scattering attenuation coefficient, the The total linear attenuation coefficient of all oral tissues.

2. The method for cone-beam oral CBCT image reconstruction according to claim 1, characterized in that: The method for obtaining the corrected projection data comprises: Step 1: Analyze the patient's oral tissue at different angles Performing logarithmic transformation on the preprocessed projection data under the condition; obtaining transformed projection data; Step 2, based on the method of scanning the patient's oral cavity with rays, the transformed projection data is corrected by the attenuation change of the rays in the patient's oral tissue to obtain corrected projection data; The method for obtaining the compensated projection data includes: Compensated projection data are obtained based on the ratio of the corrected projection data to the detector response function.

3. The method for cone-beam oral CBCT image reconstruction according to claim 2, characterized in that: Get the The method for calibrating the attenuation coefficient corresponding to the oral tissue comprises: Step 2.1, calculate the total linear attenuation coefficient; Step 2.2: Based on Monte Carlo simulation and total linear attenuation coefficient calculation, the The corrected attenuation coefficients corresponding to the oral tissues.

4. The method for cone-beam oral CBCT image reconstruction according to claim 1, characterized in that: The Monte Carlo simulation combined with the total linear attenuation coefficient calculation obtains the Methods for calculating the corrected attenuation coefficient for each oral tissue include: Step 2.2.1: Initialize the energy of the photons emitted by the ray source so that the ray source emits photon energy Normal distribution ,in, is the average energy of photons emitted by the ray source, is the standard deviation of the photon energy emitted by the ray source; Step 2.2.2: Calculate the probability of the occurrence of photons emitted by the ray source in the first step based on the Monte Carlo simulation method of simulating the probability of events in the physical process by random numbers. The probability of transmission and interaction in oral tissues; Step 2.2.3: When the ray source emits a photon When the ray source interacts with the first oral tissue, the interaction type is determined. The interaction types include photoelectric effect and Compton scattering. When it is determined to be the photoelectric effect, the photons emitted by the ray source are absorbed, photoelectrons are generated, and energy is transferred to the photoelectrons. When it is determined to be the Compton scattering, the photons emitted by the ray source interact with the first oral tissue. The electrons in the atoms in the oral tissue collide, and the radiation source emits photons to change the propagation direction and transfer energy to the electrons. The energy of the photons emitted by the radiation source after scattering is calculated based on the Compton scattering formula, combined with the electron rest energy and the scattering angle; Step 2.2.4: Record and track the photons emitted by the ray source. The propagation path in the oral tissue is When the energy of the photons emitted by the oral tissue or the scattered radiation source is less than the preset threshold, the tracking of the photons emitted by the radiation source is stopped; and the number of photons emitted by the radiation source received by the detector is counted. ; Step 2.2.5: Calculate The corrected attenuation coefficient corresponding to each oral tissue; Calculate the Methods for calculating the corrected attenuation coefficient for each oral tissue include: Based on the Beer-Lambert law, the number of photons emitted by the radiation source received by the detector is considered to be the same as the number of photons emitted by the radiation source in the first The total path length propagated in the oral tissue is calculated The corrected attenuation coefficients corresponding to the oral tissues.

5. The method for cone-beam oral CBCT image reconstruction according to claim 4, characterized in that: The method for obtaining the first oral CBCT image comprises: Step A, performing a back-projection operation on the corrected projection data based on a back-projection algorithm to obtain a first reconstructed image; The method of obtaining a first reconstructed image comprises: Step A.1: Select the impulse response function as The filter on the compensated projection data Performing filtering processing to obtain projection data after filtering processing; Step A.2, obtaining a first reconstructed image by performing an integral back-projection operation on the filtered projection data; Step B: performing iterative reconstruction optimization on the first reconstructed image based on an iterative reconstruction method to obtain a first oral CBCT image.

6. The method for cone-beam oral CBCT image reconstruction according to claim 5, characterized in that: The method for iteratively reconstructing and optimizing the first reconstructed image based on an iterative reconstruction method to obtain the first oral CBCT image includes: Step B.1: Reconstruct the first image Defined as the initial image estimate , based on the initial image estimate Constructing a forward projection model based on a ray physics model , used to estimate based on the current image Calculate the predicted projection data; among them, the forward projection model Calculate the forward projection model as the sum of the projection data and the attenuated ray intensity The difference between the actual projection data and the projection data residual is obtained; where, is the number of iterations; Step B.2: Based on the projection data residual Use gradient descent method for optimization and introduce step size parameter , update the image estimation; Step B.3: Set the iteration termination condition, stop the iteration when the termination condition is reached, and output the reconstructed image corresponding to the minimum residual. As the first oral CBCT image.

7. The method for cone-beam oral CBCT image reconstruction according to claim 6, characterized in that: The method for obtaining the reconstructed oral CBCT image comprises: Step E: By analyzing the first oral CBCT image The corresponding grayscale histogram function is obtained by integrating in different directions , ,in, is the gray value; is the total number of gray levels; Step F, obtaining a probability density function of the gray level by calculating the gray level histogram function, the number of pixel rows and the number of pixel columns of the first oral CBCT image; Step G, obtaining a cumulative distribution function by integrating the probability density function of the gray level; Step H: Set the grayscale mapping function , the grayscale mapping function The first oral CBCT image is enhanced by rounding off and combining the cumulative distribution function; the enhanced first oral CBCT image As the reconstructed oral CBCT image.

8. The method for cone-beam oral CBCT image reconstruction according to claim 1, characterized in that: The method of performing adaptive filtering on the initially collected projection data includes: Step a: The initially collected projection data is divided into The window is set according to the preset step size Traverse and divide the initially collected projection data into The projection data contained in each local area; Step b: Calculate The mean and standard deviation of the projected data points in each local area; Step c, performing adaptive Gaussian filtering on the initially collected projection data based on the standard deviation combined with the adjustment coefficient and Gaussian filtering; Step d, set the number of iterations, repeat steps a to c during each iteration, calculate the mean square error between the filtered projection data and the initially collected projection data after each iteration, and compare it with a preset error threshold, when the mean square error is less than the preset error threshold or reaches the maximum number of iterations, stop the iteration, and mark the initially collected projection data corresponding to the minimum mean square error as the projection data after adaptive filtering.

9. The method for cone-beam oral CBCT image reconstruction according to claim 8, characterized in that: The method for performing artifact correction on the projection data after the adaptive filtering process comprises: A three-dimensional rectangular coordinate system is established with the detector position as the origin, based on the initial velocity of the patient in three axes collected by the auxiliary device. , and and initial displacement , and , and at the time Acceleration , and ; According to the initial speed of the three axes and at time Acceleration calculation time Three axial speeds; Calculate the time according to the velocity and initial displacement of the three axes Displacement in three axes; Based on the calculated displacements in the three axes, a compensation operation is performed on the projection data to obtain pre-processed projection data.

10. The method for cone-beam oral CBCT image reconstruction according to claim 1, characterized in that: The method for calibrating the detector comprises: The original output signal of the detector is corrected by the gain coefficient to obtain a corrected signal.

11. A cone-beam oral CBCT image reconstruction system, implementing the cone-beam oral CBCT image reconstruction method according to any one of claims 1 to 10, characterized in that: include: Equipment calibration module: used to calibrate the detector; Data acquisition module: used to collect data of the patient's oral tissue at different angles based on X-ray emission equipment, calibrated detectors and auxiliary devices. The projection data under the condition of The value range is ; Data processing module: used to perform adaptive filtering and artifact correction processing on the initially collected projection data to obtain pre-processed projection data; Data correction module: used for correcting the preprocessed projection data to obtain corrected projection data, performing detector response compensation on the corrected projection data to obtain compensated projection data; Image reconstruction module: used to reconstruct the oral CBCT image based on the hybrid reconstruction method and the compensated projection data to obtain the first oral CBCT image; Image enhancement module: used for automatically adjusting the grayscale distribution of the first oral CBCT image by using an enhancement technology based on histogram equalization to obtain a reconstructed oral CBCT image; The method for calculating the total linear attenuation coefficient in the corrected projection data includes: According to the principle of photoelectric effect, combined with the The density of oral tissue, The molar mass of the oral tissue In oral tissue The atomic number fraction and energy of the element are The photon and atomic number are No. Calculation of the cross section of the photoelectric effect of an element Photoelectric effect attenuation coefficient of oral tissue ; According to Compton scattering, combined with the The density of oral tissue, The molar mass of the oral tissue, In oral tissue The atomic number fraction and The energy of oral tissue is Compton scattering cross section calculation Compton scattering attenuation coefficient of oral tissue ; According to the photoelectric effect attenuation coefficient and Compton scattering attenuation coefficient, the The total linear attenuation coefficient of all oral tissues.

Images