An artificial intelligence three-dimensional modeling system and method based on CT big data

Abstract

The present invention provides an artificial intelligence three-dimensional modeling system and method based on CT big data, which relates to the field of data processing. Among them, the method includes: establishing a plurality of three-dimensional reconstruction models based on CT images and three-dimensional CT images corresponding to the optimal CT scanning parameters of a plurality of test samples; determining a plurality of reference test samples from the plurality of test samples based on the tissue structure information of the target to be reconstructed and the tissue structure information of the plurality of test samples; determining the optimal CT scanning parameters of the target to be reconstructed based on the optimal CT scanning parameters of the plurality of reference test samples; performing CT scanning on the target to be reconstructed based on the optimal CT scanning parameters of the target to be reconstructed to obtain a CT image of the target to be reconstructed; and generating a three-dimensional CT image of the target to be reconstructed based on the CT image of the target to be reconstructed through the plurality of three-dimensional reconstruction models, which has the advantage of improving the quality of three-dimensional reconstruction of CT images.

Description

Technical Field

[0001] The present invention relates to the field of medical image processing, and particularly relates to an artificial intelligence three-dimensional modeling system and method based on CT big data. Background Art

[0002] Three-dimensional reconstruction refers to converting two-dimensional images or projection data into spatial information of three-dimensional objects. The reconstructed model is convenient for computer display and further processing, and has wide applications in multiple fields such as medicine, biology, engineering, and computer vision, and is of great significance for studying and analyzing the structure and form of objects. With the continuous progress of imaging technology, high-resolution three-dimensional reconstruction has become an important research direction. In recent years, with the rapid development of artificial intelligence technology, the research on key technologies of intelligent medical auxiliary diagnosis has great significance in modern medical clinics. At present, in the research of three-dimensional reconstruction technology of CT (Computed Tomography) images, due to the objective fact that medical images have less texture and more noise, there are certain difficulties in the research of current CT image three-dimensional reconstruction technology, which brings complexity to the three-dimensional reconstruction of models and is not conducive to the development of clinical medical auxiliary diagnosis technology.

[0003] Therefore, it is necessary to provide an artificial intelligence three-dimensional modeling system and method based on CT big data to improve the quality of three-dimensional reconstruction of CT images. Summary of the Invention

[0004] The present invention provides an artificial intelligence three-dimensional modeling method based on CT big data, including: obtaining the tissue structure information, optimal CT scanning parameters, CT images corresponding to the optimal CT scanning parameters, and three-dimensional CT images of multiple test samples; establishing multiple three-dimensional reconstruction models based on the CT images and three-dimensional CT images corresponding to the optimal CT scanning parameters of the multiple test samples; obtaining the tissue structure information of the target to be reconstructed; determining multiple reference test samples from the multiple test samples based on the tissue structure information of the target to be reconstructed and the tissue structure information of the multiple test samples; determining the optimal CT scanning parameters of the target to be reconstructed based on the optimal CT scanning parameters of the multiple reference test samples; performing CT scanning on the target to be reconstructed based on the optimal CT scanning parameters of the target to be reconstructed to obtain the CT image of the target to be reconstructed; and generating the three-dimensional CT image of the target to be reconstructed through the multiple three-dimensional reconstruction models based on the CT image of the target to be reconstructed.

[0005] Further, obtaining the optimal CT scanning parameters of the test sample includes: generating multiple groups of test CT scanning parameters, where the test CT scanning parameters at least include multiple CT scanning angles; obtaining the CT images corresponding to each group of the test CT scanning parameters; for each group of the test CT scanning parameters, generating the CT reconstruction result corresponding to the test CT scanning parameters according to the CT image corresponding to the test CT scanning parameters through a three-dimensional reconstruction algorithm; and based on the CT reconstruction results corresponding to each group of the test CT scanning parameters, screening the multiple groups of test CT scanning parameters to determine the optimal CT scanning parameters of the test sample.

[0006] Further, based on the CT images and three-dimensional CT images corresponding to the optimal CT scanning parameters of the multiple test samples, establishing multiple three-dimensional reconstruction models includes: clustering the multiple test samples through a clustering algorithm based on the tissue structure information of the multiple test samples to determine multiple first sample clusters; for each of the first sample clusters, clustering the multiple test samples included in the first sample cluster through a clustering algorithm based on the optimal CT scanning parameters of the multiple test samples included in the first sample cluster to determine multiple second sample clusters included in the first sample cluster; for each of the second sample clusters, clustering the multiple test samples included in the second sample cluster through a clustering algorithm based on the CT images and three-dimensional CT images corresponding to the optimal CT scanning parameters of the multiple test samples included in the second sample cluster to determine multiple third sample clusters included in the second sample cluster; for each of the third sample clusters, establishing multiple three-dimensional reconstruction models corresponding to the third sample cluster, and jointly training the multiple three-dimensional reconstruction models corresponding to the third sample cluster through the CT images and three-dimensional CT images corresponding to the optimal CT scanning parameters of the multiple test samples included in the third sample cluster, where the structures of any two three-dimensional reconstruction models corresponding to the third sample cluster are different.

[0007] Further, obtaining the tissue structure information of the target to be reconstructed includes: S11, obtaining ultrasonic detection data at multiple positions of the target to be reconstructed; S12, calculating the ultrasonic fluctuation value of the target to be reconstructed based on the ultrasonic detection data at multiple positions of the target to be reconstructed in each round; S13, judging whether the detection condition is satisfied according to the ultrasonic fluctuation value of the target to be reconstructed, if not, executing S14, if so, executing S15; S14, performing the next round of ultrasonic detection to obtain the ultrasonic detection data at multiple positions of the target to be reconstructed, where the multiple positions in any two rounds are different, and executing S12; S15, generating the tissue structure information of the target to be reconstructed, where the tissue structure information of the target to be reconstructed at least includes the ultrasonic detection data at multiple positions of the target to be reconstructed in each round.

[0008] Further, based on the organizational structure information of the target to be reconstructed and the organizational structure information of the multiple test samples, multiple reference test samples are determined from the multiple test samples, including: based on the organizational structure information of the target to be reconstructed and the organizational structure information of the test sample corresponding to the cluster center of each of the first sample clusters, a reference first sample cluster is determined from the multiple first sample clusters; based on the organizational structure information of the target to be reconstructed and the organizational structure information of each test sample included in the reference first sample cluster, multiple reference test samples are determined from the multiple test samples included in the reference first sample cluster.

[0009] Further, based on the optimal CT scan parameters of the multiple reference test samples, the optimal CT scan parameters of the target to be reconstructed are determined, including: the optimal CT scan parameters of the target to be reconstructed are determined by a parameter determination model based on the optimal CT scan parameters of the multiple reference test samples.

[0010] Further, based on the optimal CT scan parameters of the target to be reconstructed, the target to be reconstructed is CT scanned, including: the target to be reconstructed is CT scanned based on the optimal CT scan parameters of the target to be reconstructed; during the CT scanning process, the scan status information of the target to be reconstructed is obtained; based on the scan status information of the target to be reconstructed, it is determined whether to perform a secondary CT scan, and if so, the target CT scan angle is determined, and a secondary CT scan is performed according to the target CT scan angle.

[0011] Further, obtaining the scan status information of the target to be reconstructed includes: multiple markers are set on the target to be reconstructed according to the optimal CT scan parameters and the organizational structure information of the target to be reconstructed; the scan status image of the target to be reconstructed is collected; the scan status information of the target to be reconstructed is determined according to the scan status image of the target to be reconstructed.

[0012] Further, the three-dimensional CT image of the target to be reconstructed is generated by the multiple three-dimensional reconstruction models based on the CT image of the target to be reconstructed, including: based on the optimal CT scan parameters of the target to be reconstructed, a target second sample cluster is determined from the multiple second sample clusters; based on the CT image of the target to be reconstructed, a target third sample cluster is determined from the multiple third sample clusters included in the target second sample cluster; the three-dimensional CT image of the target to be reconstructed is generated by the multiple three-dimensional reconstruction models corresponding to the target third sample cluster based on the CT image of the target to be reconstructed.

[0013] The present invention provides an artificial intelligence three-dimensional modeling system based on CT big data for implementing the above-mentioned artificial intelligence three-dimensional modeling method based on CT big data, including: a data acquisition module for acquiring the tissue structure information, optimal CT scanning parameters, CT images corresponding to the optimal CT scanning parameters, and three-dimensional CT images of a plurality of test samples; a model establishment module for establishing a plurality of three-dimensional reconstruction models based on the CT images and three-dimensional CT images corresponding to the optimal CT scanning parameters of the plurality of test samples; an information acquisition module for acquiring the tissue structure information of the target to be reconstructed; a sample determination module for determining a plurality of reference test samples from the plurality of test samples based on the tissue structure information of the target to be reconstructed and the tissue structure information of the plurality of test samples; a parameter determination module for determining the optimal CT scanning parameters of the target to be reconstructed based on the optimal CT scanning parameters of the plurality of reference test samples; a CT scanning module for performing a CT scan on the target to be reconstructed based on the optimal CT scanning parameters of the target to be reconstructed to obtain a CT image of the target to be reconstructed; and a three-dimensional reconstruction module for generating a three-dimensional CT image of the target to be reconstructed based on the CT image of the target to be reconstructed through the plurality of three-dimensional reconstruction models.

[0014] Compared with the prior art, the artificial intelligence three-dimensional modeling system and method provided in this specification have at least the following beneficial effects:

[0015] 1. By determining the optimal CT scanning parameters for each test sample, it is possible to ensure that the scanning process is optimized for a specific tissue structure, thereby improving the accuracy of scanning and the image quality. Optimizing the scanning parameters also helps to reduce unnecessary radiation exposure, which is safer for both patients and operators. The CT images obtained based on the optimal CT scanning parameters can generate more accurate three-dimensional CT images.

[0016] 2. Through multiple clustering operations (forming the first sample cluster, the second sample cluster, and the third sample cluster), the test samples are finely divided according to the similarity of their tissue structure information and the characteristics of CT scan parameters. This ensures that the samples within each sample cluster are highly similar in multiple dimensions, thereby improving the pertinence and accuracy of the subsequent 3D reconstruction model. Within each sample cluster, further clustering based on the optimal CT scan parameters can more precisely identify the range of scan parameters applicable to the samples within that cluster. These parameters will be directly used for 3D reconstruction, contributing to improving the quality and accuracy of the reconstructed images. For each third sample cluster, multiple 3D reconstruction models with different structures are established and jointly trained using the CT images and 3D CT images corresponding to the optimal CT scan parameters of multiple test samples within that cluster. This training method can make full use of the diversity of samples within the cluster, improve the generalization ability of the model, and enable it to better adapt to the reconstruction requirements of different samples. By clustering the samples into smaller clusters and independently establishing and training 3D reconstruction models within each cluster, the overall computational cost can be significantly reduced. This is because the number of samples within each cluster is relatively small and they have high similarity, so data processing and model training can be carried out more efficiently. The multi-layer clustering algorithm makes the entire modeling process exhibit modular characteristics. Each sample cluster and sub-cluster can be regarded as an independent module, facilitating subsequent management and optimization of the model. When the model needs to be updated or adjusted, operations can be performed on specific sample clusters without having to reprocess the entire dataset. Since multiple 3D reconstruction models with different structures are established for each third sample cluster, in practical applications, the most suitable model can be selected according to specific requirements. This flexibility makes this method more adaptable to complex and changing application scenarios.

[0017] 3. Ultrasonic detection is a non-invasive imaging technique that does not require operations such as cutting or injecting contrast agents on the target to be reconstructed, thus greatly reducing the risk of harm to patients or samples. By performing multiple rounds of detection and calculating the ultrasonic fluctuation values, detailed information on the target to be reconstructed at different positions and angles can be obtained, thereby constructing more comprehensive and accurate tissue structure information. By continuously iterating the detection process, the detection data can be gradually optimized to ensure that the finally generated tissue structure information meets the detection conditions and improve the reliability and accuracy of the data. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] This specification will be further described by way of exemplary embodiments, which will be described in detail through the accompanying drawings. These embodiments are not restrictive. In these embodiments, the same numbers represent the same structures, where:

[0019] Figure 1It is a flowchart of an artificial intelligence three-dimensional modeling method based on CT big data shown in an embodiment of the present application;

[0020] Figure 2 It is a module diagram of an artificial intelligence three-dimensional modeling system based on CT big data shown in an embodiment of the present application. Detailed implementation manners

[0021] To more clearly illustrate the technical solutions of the embodiments of this specification, the following will briefly introduce the accompanying drawings required for the description of the embodiments.

[0022] Figure 1 It is a schematic flowchart of an artificial intelligence three-dimensional modeling method based on CT big data shown in some embodiments of this specification. As Figure 1 shown, an artificial intelligence three-dimensional modeling method based on CT big data may include the following steps.

[0023] Step 110, obtain the tissue structure information, optimal CT scan parameters, CT images corresponding to the optimal CT scan parameters, and three-dimensional CT images of multiple test samples.

[0024] Specifically, the tissue structure information of the test samples may include names (for example, head, chest, abdomen and pelvis, upper limbs and lower limbs, etc.) and ultrasonic detection data at multiple positions.

[0025] The optimal CT scan parameters may include the scan angle of the CT scan device, etc.

[0026] The three-dimensional CT image may be a three-dimensional CT image of the test sample reconstructed according to the CT image corresponding to the optimal CT scan parameters through a reconstruction algorithm, etc.

[0027] Further, obtaining the optimal CT scan parameters of the test sample includes:

[0028] Generate multiple groups of test CT scan parameters, where the test CT scan parameters at least include multiple CT scan angles, and the test CT scan parameters may also include X-ray tube voltage, tube current, etc.;

[0029] Obtain the CT images corresponding to each group of test CT scan parameters;

[0030] For each group of test CT scan parameters, generate a CT reconstruction result corresponding to the test CT scan parameters according to the CT image corresponding to the test CT scan parameters through a three-dimensional reconstruction algorithm;

[0031] Based on the CT reconstruction results corresponding to each group of test CT scan parameters, screen the multiple groups of test CT scan parameters to determine the optimal CT scan parameters of the test sample.

[0032] Specifically, for each set of test CT scan parameters, determine the scores of the test CT scan parameters in multiple evaluation metrics. For example, the clarity, contrast, noise level, etc. of the CT reconstruction results corresponding to the test CT scan parameters can be evaluated by manual or machine learning model methods to determine the scores of the test CT scan parameters in the CT reconstruction quality evaluation metrics. The radiation dose corresponding to the test CT scan parameters is evaluated by manual or machine learning model methods to determine the scores of the test CT scan parameters in the radiation dose evaluation metrics. For the scores of the test CT scan parameters in multiple evaluation metrics, calculate the screening scores of the test CT scan parameters, and take the test CT scan parameter with the largest screening score as the optimal CT scan parameter.

[0033] Step 120: Based on the CT images and three-dimensional CT images corresponding to the optimal CT scan parameters of multiple test samples, establish multiple three-dimensional reconstruction models.

[0034] Further, step 120 specifically includes:

[0035] Based on the tissue structure information of multiple test samples, cluster the multiple test samples through a clustering algorithm (such as K-means clustering algorithm, hierarchical clustering algorithm, etc.) to determine multiple first sample clusters;

[0036] For each first sample cluster, based on the optimal CT scan parameters of the multiple test samples included in the first sample cluster, cluster the multiple test samples included in the first sample cluster through a clustering algorithm (such as K-means clustering algorithm, hierarchical clustering algorithm, etc.) to determine multiple second sample clusters included in the first sample cluster;

[0037] For each second sample cluster, based on the CT images and three-dimensional CT images corresponding to the optimal CT scan parameters of the multiple test samples included in the second sample cluster, cluster the multiple test samples included in the second sample cluster through a clustering algorithm to determine multiple third sample clusters included in the second sample cluster. Specifically, preprocess the CT images and three-dimensional CT images of the multiple test samples in the second sample cluster, including operations such as image denoising, contrast enhancement, normalization, etc., to eliminate the differences between images and improve the clustering effect. Extract effective feature information from the preprocessed CT images and three-dimensional CT images. These features can include gray values, texture features, shape features, etc. Through a clustering algorithm (such as K-means clustering algorithm, hierarchical clustering algorithm, etc.), according to the feature similarity between the feature information of the preprocessed CT images and three-dimensional CT images of any two test samples included in the second sample cluster, cluster the multiple test samples included in the second sample cluster to determine multiple third sample clusters included in the second sample cluster;

[0038] For each third sample cluster, multiple three-dimensional reconstruction models corresponding to the third sample cluster are established, and the multiple three-dimensional reconstruction models corresponding to the third sample cluster are jointly trained by using the CT images corresponding to the optimal CT scanning parameters of the multiple test samples included in the third sample cluster and the three-dimensional CT images, wherein the structures of any two three-dimensional reconstruction models corresponding to the third sample cluster are different. For example, one three-dimensional reconstruction model can be a Generative Adversarial Networks (GAN) model, and another three-dimensional reconstruction model can be a Convolutional Neural Networks (CNN) model, etc.

[0039] Specifically, the multiple three-dimensional reconstruction models corresponding to the third sample cluster can be jointly trained according to the following process:

[0040] S21. Initialize each three-dimensional reconstruction model corresponding to the third sample cluster, including setting parameters such as initial weights, bias terms, learning rates, etc.;

[0041] S22. Assign the CT images corresponding to the optimal CT scanning parameters of the multiple test samples included in the preprocessed third sample cluster and the three-dimensional CT images to each three-dimensional reconstruction model;

[0042] S23. Inside each three-dimensional reconstruction model, the input data is propagated forward through the layers of the network to calculate the three-dimensional CT image;

[0043] S24. Calculate the value of the joint loss function according to the three-dimensional CT image output by each three-dimensional reconstruction model and the three-dimensional CT image corresponding to the optimal CT scanning parameter through the joint loss function;

[0044] S25. Update the parameters of each three-dimensional reconstruction model through the backpropagation algorithm according to the value of the joint loss function;

[0045] S26. Repeat the above steps S22 - S25 until the stop condition is met (for example, reaching a preset number of iterations, the loss value is less than a certain threshold, etc.).

[0046] Further, the joint loss function is:

[0047]

[0048] where L joint is the joint loss, b 11 , b 12 and b 13 are preset weights, b 11 , b 12 and b 13 are greater than 0, b 11 +b12 +b 13 = 1, a k is the loss weight corresponding to the k-th 3D reconstruction model of the third sample cluster, L k is the loss value of the k-th 3D reconstruction model of the third sample cluster, K is the total number of 3D reconstruction models corresponding to the third sample cluster, N is the total number of training samples used in one round of training of the k-th 3D reconstruction model corresponding to the third sample cluster, Y (n,k) is the 3D CT image corresponding to the n-th test sample output by the k-th 3D reconstruction model of the third sample cluster, Y (n,true) is the 3D CT image corresponding to the optimal CT scanning parameters of the n-th test sample for training the k-th 3D reconstruction model of the third sample cluster; Calculated the weighted average of the losses of all 3D reconstruction models. This weighted average method ensures that each model contributes to its total loss and the weights of each model can be adjusted according to actual needs;. Calculated the average value of the losses of all models without considering the weights of each model. It provides an indicator of the overall performance of all models, which helps to monitor the overall progress during training. By incorporating this average loss into the joint loss function, all models can be encouraged to maintain a consistent performance level as a whole and avoid some models deviating too much; Calculated the average value of the squares of the differences between the loss values of each model and its average loss value, that is, the variance of the loss, which reflects the degree of dispersion of the performance among models. By minimizing this variance term, each model can be encouraged to gradually converge during training and reduce the performance differences between them. This is crucial for enabling the collaborative work of multiple models and improving the stability and robustness of the overall system.

[0049] It can be understood that due to the different structures of each 3D reconstruction model, they can extract data features from different angles and levels, thereby improving the overall reconstruction effect. By jointly training, the advantages of different machine learning models can be utilized to achieve complementary advantages. Since multiple models are used for joint training, even if a certain model has problems or its performance deteriorates during training, other models can still work properly, thus improving the robustness.

[0050] Step 130, obtain the organizational structure information of the target to be reconstructed.

[0051] Furthermore, step 130 specifically includes:

[0052] S11. Obtain ultrasonic detection data at multiple positions of the target to be reconstructed;

[0053] S12. Calculate the ultrasonic fluctuation value of the target to be reconstructed based on the ultrasonic detection data at multiple positions of the target to be reconstructed in each round.

[0054] S13. Determine whether the detection condition is met according to the ultrasonic fluctuation value of the target to be reconstructed. If not, execute S14; if so, execute S15.

[0055] S14. Perform the next round of ultrasonic detection to obtain the ultrasonic detection data at multiple positions of the target to be reconstructed, where the multiple positions in any two rounds are different, and then execute S12.

[0056] S15. Generate the tissue structure information of the target to be reconstructed, where the tissue structure information of the target to be reconstructed at least includes the ultrasonic detection data at multiple positions of the target to be reconstructed in each round.

[0057] Specifically, for each position, perform variational mode decomposition on the ultrasonic detection data at this position to determine multiple modal components corresponding to the ultrasonic detection data at this position, extract the features of each modal component (such as center frequency, bandwidth, mean, standard deviation, etc.), and generate the feature matrix of the ultrasonic detection data at this position, where a row vector of the feature matrix includes the features of one modal component.

[0058] Calculate the ultrasonic fluctuation value of the target to be reconstructed according to the feature matrix of the ultrasonic detection data at each position.

[0059] Further, the ultrasonic fluctuation value of the target to be reconstructed can be calculated according to the following formula:

[0060]

[0061] where σ t is the ultrasonic fluctuation value of the target to be reconstructed after t rounds of ultrasonic detection, S (i,j) is the matrix similarity between the feature matrix of the i-th position and the feature matrix of the j-th position where the ultrasonic detection is performed, M is the total number of positions where the ultrasonic detection is performed after t rounds of ultrasonic detection, η is a preset parameter, η is greater than 0, V ((e,f),i) is the value of the element in the e-th row and j-th column of the feature matrix of the i-th position where the ultrasonic detection is performed, V ((e,f),j) is the value of the element in the e-th row and j-th column of the feature matrix of the j-th position where the ultrasonic detection is performed, E is the total number of rows of the feature matrix, and F is the total number of columns of the feature matrix.

[0062] When the ultrasonic fluctuation value of the target to be reconstructed is less than the ultrasonic fluctuation value threshold, it is determined that the detection condition is met.

[0063] Step 140: Determine multiple reference test samples from multiple test samples based on the organizational structure information of the target to be reconstructed and the organizational structure information of multiple test samples.

[0064] Further, step 140 specifically includes:

[0065] Determine a reference first sample cluster from multiple first sample clusters based on the organizational structure information of the target to be reconstructed and the organizational structure information of the test sample corresponding to the cluster center of each first sample cluster. For example, use the first sample cluster to which the test sample with an organizational structure information similarity greater than the first organizational structure information similarity threshold belongs as the reference first sample cluster.

[0066] Determine multiple reference test samples from the multiple test samples included in the reference first sample cluster based on the organizational structure information of the target to be reconstructed and the organizational structure information of each test sample included in the reference first sample cluster. For example, use the test sample with an organizational structure information similarity greater than the second organizational structure information similarity threshold included in the reference first sample cluster as the reference test sample.

[0067] Step 150: Determine the optimal CT scanning parameters of the target to be reconstructed based on the optimal CT scanning parameters of multiple reference test samples.

[0068] Further, step 150 specifically includes:

[0069] Determine the optimal CT scanning parameters of the target to be reconstructed through a parameter determination model based on the optimal CT scanning parameters of multiple reference test samples, where the parameter determination model can be a Convolutional Neural Networks (CNN) model.

[0070] Step 160: Perform CT scanning on the target to be reconstructed based on the optimal CT scanning parameters of the target to be reconstructed to obtain the CT image of the target to be reconstructed.

[0071] Further, step 160 specifically includes:

[0072] Perform CT scanning on the target to be reconstructed based on the optimal CT scanning parameters of the target to be reconstructed;

[0073] During the CT scanning process, obtain the scanning status information of the target to be reconstructed;

[0074] Based on the scanning status information of the target to be reconstructed, determine whether to perform secondary CT scanning. If so, determine the target CT scanning angle and perform secondary CT scanning according to the target CT scanning angle.

[0075] Further, obtaining the scanning status information of the target to be reconstructed includes:

[0076] Set a plurality of markers on the target to be reconstructed according to the optimal CT scanning parameters and tissue structure information of the target to be reconstructed;

[0077] Acquire the scanning status image of the target to be reconstructed;

[0078] Determine the scanning status information of the target to be reconstructed according to the scanning status image of the target to be reconstructed.

[0079] Specifically, a position determination model can determine a position that has less impact on CT scanning and is prone to displacement according to the optimal CT scanning parameters and tissue structure information of the target to be reconstructed, and set a marker at this position. Among them, the position determination model can be a convolutional neural network model.

[0080] Determine the relative position relationship of the plurality of markers according to the scanning status image of the target to be reconstructed, so as to determine the displacement that occurs to the target to be reconstructed during the scanning process. The scanning status information of the target to be reconstructed can include the displacement that occurs to the target to be reconstructed during the scanning process.

[0081] If the displacement that occurs to the target to be reconstructed during the scanning process is greater than the displacement threshold, it is determined to perform a second CT scan.

[0082] Step 170, generate a three-dimensional CT image of the target to be reconstructed based on the CT image of the target to be reconstructed through a plurality of three-dimensional reconstruction models.

[0083] Further, step 170 specifically includes:

[0084] Determine a target second sample cluster from a plurality of second sample clusters based on the optimal CT scanning parameters of the target to be reconstructed;

[0085] Determine a target third sample cluster from a plurality of third sample clusters included in the target second sample cluster based on the CT image of the target to be reconstructed;

[0086] Generate a three-dimensional CT image of the target to be reconstructed based on the CT image of the target to be reconstructed through a plurality of three-dimensional reconstruction models corresponding to the target third sample cluster.

[0087] Specifically, calculate the parameter difference between the optimal CT scanning parameters of the target to be reconstructed and the optimal CT scanning parameters of the test samples corresponding to the cluster centers of each second sample cluster, and use the second sample cluster to which the test samples with parameter differences less than the parameter difference threshold belong as the target second sample cluster.

[0088] Calculate the feature similarity between the CT image of the target to be reconstructed and the CT images of the test samples corresponding to the cluster centers of each third sample cluster included in the target second sample cluster, and use the third sample cluster to which the test samples with feature similarities less than the feature similarity threshold belong as the target third sample cluster.

[0089] Figure 2 It is a schematic diagram of the modules of an artificial intelligence three-dimensional modeling system based on CT big data shown in some embodiments of this specification. As Figure 2 shown, an artificial intelligence three-dimensional modeling system based on CT big data may include a data acquisition module, a model establishment module, an information acquisition module, a sample determination module, a parameter determination module, a CT scanning module, and a three-dimensional reconstruction module.

[0090] The data acquisition module can be used to acquire the tissue structure information of multiple test samples, the optimal CT scanning parameters, the CT images corresponding to the optimal CT scanning parameters, and the three-dimensional CT images;

[0091] The model establishment module can be used to establish multiple three-dimensional reconstruction models based on the CT images and three-dimensional CT images corresponding to the optimal CT scanning parameters of multiple test samples;

[0092] The information acquisition module can be used to acquire the tissue structure information of the target to be reconstructed;

[0093] The sample determination module can be used to determine multiple reference test samples from multiple test samples based on the tissue structure information of the target to be reconstructed and the tissue structure information of multiple test samples;

[0094] The parameter determination module can be used to determine the optimal CT scanning parameters of the target to be reconstructed based on the optimal CT scanning parameters of multiple reference test samples;

[0095] The CT scanning module can be used to perform CT scanning on the target to be reconstructed based on the optimal CT scanning parameters of the target to be reconstructed to obtain the CT image of the target to be reconstructed;

[0096] The three-dimensional reconstruction module can be used to generate the three-dimensional CT image of the target to be reconstructed based on the CT image of the target to be reconstructed through multiple three-dimensional reconstruction models.

[0097] An artificial intelligence three-dimensional modeling system based on CT big data can be used to execute an artificial intelligence three-dimensional modeling method based on CT big data. For more descriptions of an artificial intelligence three-dimensional modeling system based on CT big data, reference can be made to the relevant descriptions of an artificial intelligence three-dimensional modeling method based on CT big data, which will not be elaborated here.

[0098] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other deformations may also fall within the scope of this specification. Therefore, as an example rather than a limitation, the alternative configurations of the embodiments of this specification can be regarded as consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments clearly introduced and described in this specification.

Claims

1. An artificial intelligence three-dimensional modeling method based on CT big data, characterized in that, Including: Obtaining the tissue structure information, optimal CT scanning parameters, CT images corresponding to the optimal CT scanning parameters, and three-dimensional CT images of multiple test samples; Based on the CT images and three-dimensional CT images corresponding to the optimal CT scanning parameters of the multiple test samples, establishing multiple three-dimensional reconstruction models; Obtaining the tissue structure information of the target to be reconstructed; Based on the tissue structure information of the target to be reconstructed and the tissue structure information of the multiple test samples, determining multiple reference test samples from the multiple test samples; Based on the optimal CT scanning parameters of the multiple reference test samples, determining the optimal CT scanning parameters of the target to be reconstructed; Based on the optimal CT scanning parameters of the target to be reconstructed, performing CT scanning on the target to be reconstructed to obtain the CT image of the target to be reconstructed; Based on the CT image of the target to be reconstructed through the multiple three-dimensional reconstruction models, generating the three-dimensional CT image of the target to be reconstructed; Among them, based on the CT images and three-dimensional CT images corresponding to the optimal CT scanning parameters of the multiple test samples, establishing multiple three-dimensional reconstruction models, including: Based on the tissue structure information of multiple test samples, clustering the multiple test samples through a clustering algorithm to determine multiple first sample clusters; For each of the first sample clusters, based on the optimal CT scanning parameters of the multiple test samples included in the first sample cluster, clustering the multiple test samples included in the first sample cluster through a clustering algorithm to determine multiple second sample clusters included in the first sample cluster; For each of the second sample clusters, based on the CT images and three-dimensional CT images corresponding to the optimal CT scanning parameters of the multiple test samples included in the second sample cluster, clustering the multiple test samples included in the second sample cluster through a clustering algorithm to determine multiple third sample clusters included in the second sample cluster; For each of the third sample clusters, establishing multiple three-dimensional reconstruction models corresponding to the third sample cluster, and jointly training the multiple three-dimensional reconstruction models corresponding to the third sample cluster through the CT images and three-dimensional CT images corresponding to the optimal CT scanning parameters of the multiple test samples included in the third sample cluster, where the structures of any two three-dimensional reconstruction models corresponding to the third sample cluster are different.

2. The artificial intelligence three-dimensional modeling method based on CT big data according to claim 1, wherein Obtaining the optimal CT scanning parameters of the test sample, including: Generating multiple groups of test CT scanning parameters, where the test CT scanning parameters at least include multiple CT scanning angles; Obtaining the CT images corresponding to each group of the test CT scanning parameters; For each group of the test CT scanning parameters, generating the CT reconstruction result corresponding to the test CT scanning parameters through a three-dimensional reconstruction algorithm according to the CT image corresponding to the test CT scanning parameters; Based on the CT reconstruction results corresponding to each group of the test CT scanning parameters, screening the multiple groups of test CT scanning parameters to determine the optimal CT scanning parameters of the test sample.

3. An artificial intelligence three-dimensional modeling method based on CT big data according to claim 2, characterized in that Obtaining the tissue structure information of the target to be reconstructed, including: S11. Obtaining the ultrasonic detection data at multiple positions of the target to be reconstructed; S12. Calculate the ultrasonic fluctuation value of the target to be reconstructed based on the ultrasonic detection data at multiple positions of the target to be reconstructed in each round; S13. Determine whether the detection condition is satisfied according to the ultrasonic fluctuation value of the target to be reconstructed. If not, execute S14; if so, execute S15; S14. Perform the next round of ultrasonic detection to obtain the ultrasonic detection data at multiple positions of the target to be reconstructed, where the multiple positions in any two rounds are different, and then execute S12; S15. Generate the tissue structure information of the target to be reconstructed, where the tissue structure information of the target to be reconstructed at least includes the ultrasonic detection data at multiple positions of the target to be reconstructed in each round.

4. An artificial intelligence three-dimensional modeling method based on CT big data according to claim 3, characterized in that Based on the tissue structure information of the target to be reconstructed and the tissue structure information of the multiple test samples, determine multiple reference test samples from the multiple test samples, including: Based on the tissue structure information of the target to be reconstructed and the tissue structure information of the test sample corresponding to the cluster center of each first sample cluster, determine the reference first sample cluster from the multiple first sample clusters; Based on the tissue structure information of the target to be reconstructed and the tissue structure information of each test sample included in the reference first sample cluster, determine multiple reference test samples from the multiple test samples included in the reference first sample cluster.

5. An artificial intelligence three-dimensional modeling method based on CT big data according to any one of claims 1-4, characterized in that, Based on the optimal CT scanning parameters of the multiple reference test samples, determine the optimal CT scanning parameters of the target to be reconstructed, including: Determine the optimal CT scanning parameters of the target to be reconstructed through a parameter determination model based on the optimal CT scanning parameters of the multiple reference test samples.

6. A method for artificial intelligence three-dimensional modeling based on CT big data according to any one of claims 1-4, characterized in that, Based on the optimal CT scanning parameters of the target to be reconstructed, perform CT scanning on the target to be reconstructed, including: Based on the optimal CT scanning parameters of the target to be reconstructed, perform CT scanning on the target to be reconstructed; During the CT scanning process, obtain the scanning status information of the target to be reconstructed; Based on the scanning status information of the target to be reconstructed, determine whether to perform secondary CT scanning. If so, determine the target CT scanning angle and perform secondary CT scanning according to the target CT scanning angle.

7. An artificial intelligence three-dimensional modeling method based on CT big data according to claim 6, characterized in that, Obtain the scanning status information of the target to be reconstructed, including: Set multiple markers on the target to be reconstructed according to the optimal CT scanning parameters and tissue structure information of the target to be reconstructed; Collect the scanning status image of the target to be reconstructed; Determine the scanning status information of the target to be reconstructed according to the scanning status image of the target to be reconstructed.

8. A method for artificial intelligence three-dimensional modeling based on CT big data according to any one of claims 1-4, characterized in that, Generate the three-dimensional CT image of the target to be reconstructed through the multiple three-dimensional reconstruction models based on the CT image of the target to be reconstructed, including: Based on the optimal CT scanning parameters of the target to be reconstructed, determine the target second sample cluster from the multiple second sample clusters; Based on the CT image of the target to be reconstructed, determine the target third sample cluster from the multiple third sample clusters included in the target second sample cluster; Generate the three-dimensional CT image of the target to be reconstructed through the multiple three-dimensional reconstruction models corresponding to the target third sample cluster based on the CT image of the target to be reconstructed.

9. An artificial intelligence three-dimensional modeling system based on CT big data, characterized in that, A method for performing an artificial intelligence three-dimensional modeling based on CT big data according to any one of claims 1-8, comprising: A data acquisition module, configured to acquire the tissue structure information of a plurality of test samples, the optimal CT scanning parameters, the CT images corresponding to the optimal CT scanning parameters, and three-dimensional CT images; A model establishment module, configured to establish a plurality of three-dimensional reconstruction models based on the CT images and three-dimensional CT images corresponding to the optimal CT scanning parameters of the plurality of test samples; An information acquisition module, configured to acquire the tissue structure information of the target to be reconstructed; A sample determination module, configured to determine a plurality of reference test samples from the plurality of test samples based on the tissue structure information of the target to be reconstructed and the tissue structure information of the plurality of test samples; A parameter determination module, configured to determine the optimal CT scanning parameters of the target to be reconstructed based on the optimal CT scanning parameters of the plurality of reference test samples; A CT scanning module, configured to perform CT scanning on the target to be reconstructed based on the optimal CT scanning parameters of the target to be reconstructed to obtain the CT image of the target to be reconstructed; A three-dimensional reconstruction module, configured to generate the three-dimensional CT image of the target to be reconstructed based on the CT image of the target to be reconstructed through the plurality of three-dimensional reconstruction models.

Images