Ultrasonic image processing system based on three-dimensional reconstruction

Abstract

The invention relates to the technical field of ultrasonic image processing systems, and provides an ultrasonic image processing system based on three-dimensional reconstruction, which comprises an ultrasonic data acquisition terminal, an image processing terminal, a three-dimensional reconstruction terminal, a quality evaluation terminal and a doctor interaction terminal, the ultrasonic data acquisition terminal is used for acquiring ultrasonic original radio frequency data and a B mode image and synchronously acquiring probe position information of the ultrasonic probe; the image processing terminal is used for performing image denoising, edge enhancement and tissue feature labeling according to the ultrasonic original radio frequency data and the B mode image, and extracting key section information for subsequent reconstruction; the three-dimensional reconstruction terminal is used for performing spatial reconstruction based on the key section data and the probe position information to form a three-dimensional ultrasonic image body; the quality evaluation terminal is used for scoring the three-dimensional reconstruction effect of the three-dimensional ultrasonic image body and generating three-dimensional reconstruction quality evaluation information; and the doctor interaction terminal is used for a doctor to check the three-dimensional ultrasonic image body and the three-dimensional reconstruction quality evaluation information in real time and generate and display optimization suggestion information. The method has the effect of improving the accuracy of three-dimensional reconstruction quality evaluation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of an ultrasonic image processing system, and in particular to an ultrasonic image processing system based on three-dimensional reconstruction. BACKGROUND

[0002] Ultrasound imaging is widely used in medical diagnosis due to its non-invasive, real-time, low-cost advantages. However, traditional two-dimensional ultrasound images have low spatial resolution, lack of structural information, and high dependence on operators, which limits their clinical application depth. With the development of three-dimensional reconstruction technology, three-dimensional visualization based on ultrasound images has become a hot topic, which can improve the accuracy of lesion positioning, assist in complex structure recognition and surgical navigation. However, existing three-dimensional ultrasound reconstruction systems generally have high computational complexity, fragmented processing procedures, and non-uniform quality evaluation standards, lacking an integrated, intelligent, and convenient comprehensive processing solution. Therefore, there is an urgent need for a three-dimensional ultrasound image processing system that integrates acquisition, processing, reconstruction, and evaluation.

[0003] Many ultrasound image processing systems have been developed, and after our extensive search and reference, we found that the prior art ultrasound image processing systems such as those disclosed in CN118196074A and CN112107332A generally include an ultrasound image acquisition terminal and a quality evaluation terminal. The ultrasound image acquisition terminal is used to acquire and process ultrasound images of a patient, and the quality evaluation terminal is used to evaluate the quality of the processed ultrasound images and output the quality evaluation results. The above-mentioned ultrasound image processing system lacks quality evaluation based on probe position during quality evaluation, but the probe position directly determines the quality of ultrasound signal acquisition. Improper probe position may cause ultrasound waves to be disturbed by the abdominal wall, intestinal gas, and other factors, affecting the imaging quality. Therefore, the prior art ultrasound image processing system has the problem of poor quality evaluation accuracy. SUMMARY

[0004] The present application aims to address the deficiencies of the above-mentioned ultrasound image processing system and proposes an ultrasound image processing system based on three-dimensional reconstruction.

[0005] The present application adopts the following technical solutions:

[0006] An ultrasound image processing system based on three-dimensional reconstruction includes an ultrasound data acquisition terminal, an image processing terminal, a three-dimensional reconstruction terminal, a quality evaluation terminal and a doctor interaction terminal; the ultrasound data acquisition terminal is used for acquiring ultrasound raw radio frequency data and B-mode images, and synchronously acquiring probe position information of an ultrasound probe; the image processing terminal is used for carrying out image denoising, edge enhancement and tissue feature labeling according to the ultrasound raw radio frequency data and the B-mode images, and extracting key section information for subsequent reconstruction; the three-dimensional reconstruction terminal is used for carrying out spatial reconstruction based on the key section data and the probe position information, forming a three-dimensional ultrasound image volume; the quality evaluation terminal is used for scoring the three-dimensional reconstruction effect of the three-dimensional ultrasound image volume, generating three-dimensional reconstruction quality evaluation information; and the doctor interaction terminal is used for enabling a doctor to view the three-dimensional ultrasound image volume and the three-dimensional reconstruction quality evaluation information in real time, generating and displaying optimization suggestion information.

[0007] Optionally, the ultrasound data acquisition terminal includes an ultrasound data acquisition module, a probe position information acquisition module and a data transmission module; the ultrasound data acquisition module is used for acquiring ultrasound raw radio frequency data and B-mode images; the probe position information acquisition module is used for synchronously acquiring probe position information of an ultrasound probe; and the data transmission module is used for transmitting the acquired ultrasound raw radio frequency data, B-mode images and probe position information to the image processing terminal in real time.

[0008] Optionally, the image processing terminal includes an image preprocessing module, a feature extraction module and a key section extraction module; the image preprocessing module is used for receiving ultrasound raw radio frequency data, performing envelope detection, gray mapping and denoising processing on the ultrasound raw radio frequency data, and converting the ultrasound raw radio frequency data into an enhanced gray image; the feature extraction module is used for carrying out edge enhancement and tissue structure labeling based on the B-mode images and the enhanced gray image; and the key section extraction module is used for extracting key section information for subsequent reconstruction from the image processed by the feature extraction module.

[0009] Optionally, the three-dimensional reconstruction terminal includes a data registration module, a spatial reconstruction module and a three-dimensional display module; the data registration module is used for spatially aligning the key section information according to the probe position information; the spatial reconstruction module is used for carrying out three-dimensional reconstruction using the aligned key section information, forming a three-dimensional ultrasound image volume; and the three-dimensional display module is used for displaying the reconstructed three-dimensional ultrasound image volume.

[0010] Optionally, the quality evaluation terminal includes an image scoring module and a quality evaluation output module; the image scoring module is used for scoring calculation on the three-dimensional ultrasound image volume; and the quality evaluation output module is used for generating three-dimensional reconstruction quality evaluation information according to the score.

[0011] Optionally, the doctor interaction terminal comprises a real-time display module, an interaction control module and a report generation module; the real-time display module is configured to display the three-dimensional ultrasound image and the three-dimensional reconstruction quality assessment information; the interaction control module is configured to support the doctor to perform real-time operations of zooming in, rotating and adjusting the image data; and the report generation module is configured to generate and display the optimization suggestion information.

[0012] Optionally, the image scoring module comprises a scoring parameter acquisition submodule and an image quality score calculation submodule; the scoring parameter acquisition submodule is configured to extract calculation parameters for calculating the image quality score from the probe position information, the key section information and the three-dimensional ultrasound image; and the image quality score calculation submodule is configured to calculate the image quality score according to the calculation parameters.

[0013] When the image quality score calculation submodule is working, the following formula is satisfied:

[0014]

[0015] wherein Q represents the image quality score of the three-dimensional ultrasound image; S c represents the structural clarity; V r represents the reconstruction completeness; A n represents the abnormal noise area ratio; D s represents the probe movement stability; C o represents the structural continuity; the greater the Q value, the better the image quality; in the scoring formula, each scoring parameter is integrated through a nonlinear transformation with clear physical meaning and evaluation logic, avoiding the problems of slow reaction to abnormal values and insufficient ability to distinguish key indicators in the traditional weighted summation method; the structural clarity score is enhanced by squaring the high-quality area score, the reconstruction completeness is improved by taking the square root to improve the distinction of the medium and low-quality sections, the abnormal noise is quickly punished by using an exponential function, the probe jitter is quickly suppressed by using a sine function, and the structural continuity is linearly involved in the stable total score; the change trend of each parameter is consistent with the actual medical image usability perception, ensuring that the scoring system has good sensitivity, robustness and discrimination accuracy; when Q ref < q ref , the quality assessment output module generates three-dimensional reconstruction quality assessment information indicating that the image quality meets the standard, and the optimization suggestion information generated by the report generation module at this time comprises: direct archiving, providing a 3D visual interaction interface for diagnosis, and enabling the export model for preoperative navigation.

[0016] A method for processing ultrasonic images based on three-dimensional reconstruction is applied to the above-mentioned ultrasonic image processing system based on three-dimensional reconstruction. The method for processing ultrasonic images based on three-dimensional reconstruction includes:

[0017] S1, collecting ultrasound raw radio frequency data and B-mode images, and synchronously obtaining the probe position information of the ultrasound probe;

[0018] S2, based on the original ultrasound RF data and B-mode images, performs image denoising, edge enhancement, tissue feature annotation, and extracts key section information for subsequent reconstruction;

[0019] S3, spatial reconstruction is performed based on key section data and probe position information to form a three-dimensional ultrasound image volume;

[0020] S4, scoring the three-dimensional reconstruction effect of the three-dimensional ultrasound image volume to generate three-dimensional reconstruction quality assessment information;

[0021] S5 allows doctors to view 3D ultrasound images and 3D reconstruction quality assessment information in real time, and generate and display optimization suggestion information.

[0022] The beneficial effects achieved by the present invention are:

[0023] 1. By setting up ultrasound data acquisition terminals, image processing terminals, 3D reconstruction terminals, quality assessment terminals, and doctor interaction terminals, the system can complete the entire closed-loop processing process from raw ultrasound data acquisition, image preprocessing, key section extraction, 3D reconstruction to quality assessment and doctor interaction. This is conducive to achieving structured modeling of ultrasound data, thereby improving the integrity and diagnostic applicability of 3D image reconstruction, and improving the accuracy of system quality assessment, thus facilitating the construction of an automated and intelligent ultrasound image reconstruction and analysis system.

[0024] 2. Through the configuration of the ultrasound data acquisition module, the probe position information acquisition module, and the data transmission module, the system can not only synchronously acquire images and spatial trajectory information, but also send the data to the image processing terminal in real time. This helps to ensure the synchronization and timeliness of image data and position information, thereby improving the registration accuracy of subsequent 3D reconstruction, thereby enhancing the structural consistency and spatial accuracy of 3D image reconstruction;

[0025] 3. By setting up an image preprocessing module, a feature extraction module, and a key section extraction module, the system can perform grayscale enhancement and denoising on the original ultrasound radiofrequency data. Based on image enhancement, it can also achieve automatic annotation of tissue structures and accurate extraction of key sections. This is conducive to improving the quality control capability of the image processing stage, thereby ensuring the stability of the reconstructed input image data, and thus helping to improve the usability and resolution of the final 3D reconstruction.

[0026] 4. By setting the data registration module, the spatial reconstruction module and the three-dimensional display module, the key section information can be registered with the probe track and reconstructed into a three-dimensional ultrasound image volume, and displayed in a visualized manner, which is beneficial to improve the spatial consistency and structural expression capability of the three-dimensional reconstruction result, and further enhance the three-dimensional perception of the doctor on the tissue lesion, thereby being beneficial to support the actual application requirements of clinical diagnosis and preoperative planning;

[0027] 5. By setting the image scoring module and the quality evaluation output module, the system can objectively evaluate the three-dimensional ultrasound image volume according to a preset scoring algorithm and output the scoring result, which is beneficial to establish a quantifiable image quality evaluation mechanism, and further assist the system to realize automatic judgment and strategy feedback, thereby being beneficial to improve the reliability and efficiency of the three-dimensional ultrasound image in the actual clinical process;

[0028] 6. By setting the real-time display module, the interactive control module and the report generation module, the doctor can view, operate and adjust the reconstructed image and the quality evaluation result in the visualized interface, which is beneficial to enhance the doctor's use experience and diagnosis control capability on the image data, and further improve the individualized image interpretation efficiency, thereby being beneficial to build an interactive ultrasound reconstruction system with clinical friendliness and operability;

[0029] 7. By setting the scoring value Q calculation algorithm including the structural clarity, the reconstruction completeness, the abnormal noise area ratio, the probe movement stability and the structural continuity, the system can perform nonlinear weighted evaluation on the three-dimensional reconstruction image from four dimensions of clarity, completeness, stability and continuity, which is beneficial to establish a multi-factor, segmented nonlinear fusion image quality judgment mechanism, and further realize high sensitivity identification and error suppression on the image quality, thereby being beneficial to improve the clinical interpretability and automatic adaptation capability of the image scoring;

[0030] 8. By setting the unfolding calculation formula of the structural continuity score, the system can perform pairwise correlation calculation based on the gray image matrix between the continuous key sections, and quantitatively analyze the structural consistency of the adjacent sections through the Pearson correlation coefficient, which is beneficial to capture the changes of the tissue texture continuity and spatial consistency in the image reconstruction sequence, and further identify abnormal conditions such as section misplacement, fault or jitter, thereby being beneficial to improve the continuity and stability of the three-dimensional reconstruction result in the spatial structure expression.

[0031] In order to further understand the features and technical contents of the present application, please refer to the following detailed description and drawings of the present application. However, the provided drawings are only used for reference and illustration, and are not used to limit the present application. BRIEF DESCRIPTION OF DRAWINGS

[0032] Figure 1It is a schematic diagram of the overall structure of the present application;

[0033] Figure 2 It is a schematic diagram of the structure of the image scoring module in the present application;

[0034] Figure 3 It is a schematic diagram of the method flow of an ultrasound image processing method based on three-dimensional reconstruction in the present application;

[0035] Figure 4 It is a schematic diagram of the structure of the image quality score calculation sub-module in another embodiment of the present application;

[0036] Figure 5 It is a schematic diagram of the statistical effect of the calculation results of Q values in another embodiment of the present application. DETAILED DESCRIPTION

[0037] The following describes the embodiments of the present application through specific specific embodiments, and those skilled in the art can understand the advantages and effects of the present application from the disclosed content. The present application can be implemented or applied through other different specific embodiments, and various modifications and changes can be made based on different viewpoints and applications without departing from the spirit of the present application. In addition, the drawings of the present application are only simple schematic illustrations, not actual size drawings, and it is declared in advance. The following embodiments will further illustrate the related technical content of the present application, but the disclosed content is not used to limit the protection scope of the present application.

[0038] Embodiment one: the present embodiment provides an ultrasound image processing system based on three-dimensional reconstruction. As shown in the combination Figure 1 As shown in the combination, an ultrasound image processing system based on three-dimensional reconstruction includes an ultrasound data acquisition terminal, an image processing terminal, a three-dimensional reconstruction terminal, a quality evaluation terminal, and a doctor interaction terminal; the ultrasound data acquisition terminal is used to acquire ultrasound raw radio frequency data and B-mode images, and synchronously acquire probe position information of an ultrasound probe; the image processing terminal is used to perform image denoising, edge enhancement, and tissue feature labeling according to the ultrasound raw radio frequency data and B-mode images, and extract key section information for subsequent reconstruction; the three-dimensional reconstruction terminal is used to perform spatial reconstruction based on the key section data and the probe position information, forming a three-dimensional ultrasound image volume; the quality evaluation terminal is used to score the three-dimensional reconstruction effect of the three-dimensional ultrasound image volume, generating three-dimensional reconstruction quality evaluation information; the doctor interaction terminal is used for doctors to perform real-time viewing of the three-dimensional ultrasound image volume, the three-dimensional reconstruction quality evaluation information, generate and display optimization suggestion information.

[0039] Optionally, the ultrasound data acquisition terminal comprises an ultrasound data acquisition module, a probe position information acquisition module, and a data transmission module; the ultrasound data acquisition module is configured to acquire ultrasound raw radio frequency data and B-mode images; the probe position information acquisition module is configured to synchronously acquire probe position information of an ultrasound probe; and the data transmission module is configured to transmit the acquired ultrasound raw radio frequency data, B-mode images, and probe position information to the image processing terminal in real time.

[0040] Optionally, the image processing terminal comprises an image preprocessing module, a feature extraction module, and a key slice extraction module; the image preprocessing module is configured to receive ultrasound raw radio frequency data, perform envelope detection, gray scale mapping, and denoising processing on the ultrasound raw radio frequency data, and convert the ultrasound raw radio frequency data into enhanced gray scale images; the feature extraction module is configured to perform edge enhancement and tissue structure labeling based on B-mode images and enhanced gray scale images; and the key slice extraction module is configured to extract key slice information for subsequent reconstruction from the images processed by the feature extraction module. When extracting key slice information based on the images processed by the feature extraction module, an edge detection algorithm, such as Sobel, is used to process B-mode images or enhanced gray scale images, the edge intensity distribution of each frame of image is calculated, and the slice with the richest edge information is selected as the key slice.

[0041] Optionally, the three-dimensional reconstruction terminal comprises a data registration module, a spatial reconstruction module, and a three-dimensional display module; the data registration module is configured to perform spatial alignment of key slice information according to probe position information; the spatial reconstruction module is configured to perform three-dimensional reconstruction using the aligned key slice information to form a three-dimensional ultrasound image volume; and the three-dimensional display module is configured to display the reconstructed three-dimensional ultrasound image volume.

[0042] Optionally, the quality assessment terminal comprises an image scoring module and a quality assessment output module; the image scoring module is configured to perform scoring calculation on the three-dimensional ultrasound image volume; and the quality assessment output module is configured to generate three-dimensional reconstruction quality assessment information according to the score.

[0043] Optionally, the doctor interaction terminal comprises a real-time display module, an interaction control module, and a report generation module; the real-time display module is configured to display the three-dimensional ultrasound image volume and the three-dimensional reconstruction quality assessment information; the interaction control module is configured to support real-time operations of the doctor on the image data, such as zooming in, rotating, and adjusting slices; and the report generation module is configured to generate and display optimization suggestion information.

[0044] Optionally, in combination with Figure 2As shown, the image scoring module includes a scoring parameter acquisition submodule and an image quality score value calculation submodule; the scoring parameter acquisition submodule is configured to extract calculation parameters for calculating the image quality score value from the probe position information, the key section information, and the three-dimensional ultrasound image volume; the image quality score value calculation submodule is configured to calculate the image quality score value according to the calculation parameters;

[0045] When the image quality score value calculation submodule is working, the following formula is satisfied:

[0046]

[0047] wherein Q represents the image quality score value of the three-dimensional ultrasound image volume; S c represents the structural definition; V r represents the reconstruction completeness; A n represents the abnormal noise area ratio; D s represents the probe movement stability; C o represents the structural continuity, which is set by an administrator according to experience; in this embodiment, the structural continuity C o is set by an administrator according to the actual image acquisition quality and the structure performance between sections: generally, when the texture of adjacent image sections changes smoothly and the structure is connected naturally, C o may be set to a higher value; if there are frame skipping, fault, texture mutation, etc. in the image sections, C o may be set to a lower value to reflect the poor structural continuity. The larger the Q value is, the better the image quality is; in the scoring formula, each scoring parameter is integrated through nonlinear transformation with clear physical meaning and evaluation logic, avoiding the problems of slow reaction to abnormal values and insufficient discrimination of key indicators in the traditional weighted summation method; the structural definition score is enhanced by squaring to increase the score of high-quality areas, the reconstruction completeness is improved by taking the square root to increase the discrimination of medium and low-quality sections, the abnormal noise is punished by using an exponential function, the probe jitter is quickly suppressed by using a sine function, and the structural continuity is linearly involved in the stable total score; the change trend of each parameter is consistent with the actual medical image usability perception, ensuring that the scoring system has good sensitivity, robustness, and discrimination accuracy; when Q < q ref , the quality evaluation output module generates three-dimensional reconstruction quality evaluation information indicating that the image quality is not up to standard; at this time, the optimization suggestion information generated by the report generation module includes: immediately determining as “reconstruction failure”, prohibiting the export of the model, the diagnosis graph, forcibly guiding the reacquisition of the full image, and recording the acquisition abnormal type for statistical analysis; when Q ≥ q refWhen the image quality is up to standard, the quality assessment output module generates three-dimensional reconstruction quality assessment information representing the image quality, and the report generation module generates optimization suggestion information including: direct archiving, providing a 3D visual interactive interface for diagnosis, and enabling export of the model for preoperative navigation. ref is an image quality score value threshold, which is set by an administrator according to experience.

[0048] In ultrasonic imaging, structure clarity, that is, the degree of edge clarity, is an important basis for judging image usability. If the boundary is blurred, it is difficult for a doctor to identify the structure. The physical meaning of structure clarity is that ultrasonic waves reflect at the interface of different media, such as muscle, fat, and bone, and the clearer the interface and the stronger the reflection, the greater the gradient of the image. Structure clarity amplifies the contribution of high-quality areas by squaring, so that the score of a clear image is obviously improved, and the score of a blurred image is reduced, which conforms to the experience rule that clear images are preferentially recognized in actual medical imaging.

[0049]

[0050] wherein I represents a grayscale matrix of a single ultrasonic slice image in the key section information; N e represents the total number of edge points; S c When S represents the grayscale gradient of the i th edge point in the X-axis direction; represents the grayscale gradient of the i th edge point in the Y-axis direction; represents the synthetic gradient amplitude of the i th edge point calculated using the Sobel operator; x i represents the horizontal coordinate of the i th edge point in the grayscale matrix; y i represents the vertical coordinate of the i th edge point in the grayscale matrix; I() represents the grayscale value of the edge point at the corresponding coordinate in the parentheses;

[0051] The key to three-dimensional reconstruction is the completeness of the coverage of the target structure. If the volume is defective, a lesion may be missed or the boundary of the lesion may be incorrectly judged. The physical meaning of reconstruction completeness is that the ratio between the theoretical scanning volume and the actual reconstructed volume can directly reflect whether the scanning is comprehensive. Taking the square root of the reconstruction completeness can enhance the influence of changes in the middle and low sections on the score (for example, when V r = 0.5, the score is lower), while the influence of changes in the high completeness interval (V r = 0.9 to V r = 1.0) is weakened, which conforms to the clinical logic that missing middle and low section content is more deadly than subtle completeness difference in actual medical judgment.

[0052]

[0053] V rec =N vox ·v unit ;

[0054] V theo =W s ·H s ·S s ;

[0055] Among them, V r Indicates the reconstruction completeness of the three-dimensional ultrasound image volume; V rec represents the actual reconstruction volume; V theo Indicates the theoretical scanning volume; N vox Indicates the number of voxel points actually involved in reconstruction; v unit Represents the physical volume of a single voxel, determined by the resolution; W s Indicates the width recorded during the probe scanning process; H s Indicates the height recorded during the probe scanning process; S s Indicates the depth recorded during the probe scanning process;

[0056] Abnormal noise in ultrasound images, such as random bright spots, can easily obscure important tissues and affect diagnosis. The physical significance of the abnormal noise area ratio: abnormally high grayscale noise points often arise from probe contact anomalies, coupling agent problems, electronic interference, and other factors, reflecting data quality defects. The reason for using the abnormal noise area ratio as an index is that when the noise is small, the impact on the image is acceptable, so the effect is set to increase slowly. Once the noise area reaches a certain threshold, the image is essentially unusable, so an exponential increase is used to rapidly lower the score, reflecting the "catastrophic effect" of noise on quality, which is consistent with actual medical perception.

[0057]

[0058] Among them, A noise The total area of ​​abnormally high grayscale regions in the representative image of a single ultrasound slice image in the key section information is defined as the sum of the areas of connected regions that exceed the noise judgment threshold. The noise judgment threshold is set by the administrator based on experience; total represents the total area of ​​the image representing the graph;

[0059] During the 3D scanning process, probe jitter can lead to spatial positioning errors, causing structural dislocation and reconstruction failure. The physical significance of probe movement stability is that the stability of the probe position directly determines the correct arrangement of the image sequence in space, which in turn affects the final 3D imaging quality. The reason for choosing sine as the mathematical processing method is that when the jitter amplitude is small and Ds is close to 0, Approximately 1, indicating no impact; the jitter amplitude is small Ds near 1, near 0; with sinusoidal nonlinearity, small jitter is tolerated, but large jitter is quickly punished, which is consistent with the actual acquisition experience;

[0060]

[0061] wherein N represents the number of continuous spatial position information points in the probe position information, that is, the total number of sampling moments, and three-dimensional coordinate values (x k ,y k ,z k ) corresponding to the probe are recorded in each frame; d k represents the spatial movement distance of the probe between the kth and k+1th frames; represents the average displacement of all adjacent frames; D max represents the maximum acceptable stability standard deviation, which is used to realize normalization processing, so that D s ∈[0,1], D max The specific value is set according to the actual situation, and the requirements are: D max ≥D s .

[0062] In summary, by setting the ultrasonic data acquisition terminal, image processing terminal, three-dimensional reconstruction terminal, quality evaluation terminal and doctor interaction terminal, the system can complete the whole process closed loop processing from original ultrasonic data acquisition, image preprocessing, key section extraction, three-dimensional reconstruction to quality evaluation and doctor interaction, which is conducive to realizing the structured modeling of ultrasonic data and further improving the integrity and diagnostic applicability of three-dimensional image reconstruction; through the setting of the ultrasonic data acquisition module, probe position information acquisition module and data transmission module, the system can not only synchronously acquire image and spatial trajectory information, but also send data to the image processing terminal in real time, which is conducive to ensuring the synchronicity and timeliness of image data and position information and further improving the registration accuracy of subsequent three-dimensional reconstruction; by setting the image preprocessing module, feature extraction module and key section extraction module, the system can perform gray scale enhancement and denoising processing on the original radio frequency data of ultrasonic, and realize automatic labeling of tissue structure and accurate extraction of key section on the basis of image enhancement, which is conducive to improving the quality control ability of the image processing stage and further ensuring the stability of the reconstructed input image data; by setting the data registration module, spatial reconstruction module and three-dimensional display module, the key section information and probe trajectory can be registered and reconstructed into a three-dimensional ultrasonic image body, which is displayed in a visual way, which is conducive to improving the spatial consistency and structure expression ability of the three-dimensional reconstruction result and further enhancing the stereoscopic perception of doctors to the tissue lesions; by setting the image scoring module and quality evaluation output module, the system can objectively evaluate the three-dimensional ultrasonic image body according to the preset scoring algorithm and output the scoring result, which is conducive to establishing a quantifiable image quality evaluation mechanism and further assisting the system to realize automatic judgment and strategy feedback; by setting the real-time display module, interactive control module and report generation module, the doctor can view, operate and adjust the reconstructed image and quality evaluation result in the visual interface, which is conducive to enhancing the doctor's use experience and diagnostic control ability of image data and further improving the individualized image interpretation efficiency; by setting the scoring value Q calculation algorithm including structure clarity, reconstruction integrity, abnormal noise area ratio, probe movement stability and structure continuity, the system can perform nonlinear weighted evaluation on the three-dimensional reconstructed image from four dimensions of clarity, integrity, stability and continuity, which is conducive to establishing a multi-factor, segmented nonlinear fusion image quality judgment mechanism and further realizing high sensitivity identification and error suppression of image quality, so as to be conducive to improving the clinical interpretability and automatic adaptability of image scoring.

[0063] An ultrasonic image processing method based on three-dimensional reconstruction, applied to the ultrasonic image processing system based on three-dimensional reconstruction as described above, in combination with Figure 3 As shown in the figure, the ultrasonic image processing method based on three-dimensional reconstruction includes:

[0064] S1, collect the ultrasound raw radio frequency data and B-mode image, and synchronously acquire the probe position information of the ultrasound probe;

[0065] S2, perform image denoising, edge enhancement, and tissue feature labeling according to the ultrasound raw radio frequency data and B-mode image, and extract key slice information for subsequent reconstruction;

[0066] S3, perform spatial reconstruction based on the key slice data and the probe position information to form a three-dimensional ultrasound image volume;

[0067] S4, score the three-dimensional reconstruction effect of the three-dimensional ultrasound image volume to generate three-dimensional reconstruction quality assessment information;

[0068] S5, provide the doctor with real-time viewing of the three-dimensional ultrasound image volume and the three-dimensional reconstruction quality assessment information, generate and display optimization suggestion information.

[0069] Embodiment two: this embodiment contains all the contents of embodiment one, in order to avoid the structural continuity of embodiment one o The subjective and intelligent degree of the administrator's experience selection is not high, and an ultrasound image processing system based on three-dimensional reconstruction is provided, which combines Figure 4 As shown, the image quality score value calculation submodule includes a structure continuity precalculation unit and an image quality score value calculation unit; the structure continuity precalculation unit is configured to precalculate structure continuity according to key slice information; and the image quality score value calculation unit is configured to calculate an image quality score value according to a calculation parameter.

[0070] The continuity of three-dimensional reconstruction depends on the consistency of adjacent slices, and if the adjacent slices change dramatically, the reconstruction will fail; the physical meaning of structure continuity is that the real tissue should change slowly and smoothly in continuous slices; texture, gray scale, and edge continuity reflect the real structure continuity of the tissue; the reason for the direct multiplication mathematical processing method is that structure continuity is a basic requirement, and without continuity, the reconstruction effect is directly denied; and directly as a product factor, the importance is completely retained without amplification or reduction processing.

[0071] When the structure continuity precalculation unit works, the following formula is satisfied:

[0072]

[0073] wherein, I j represents the gray scale image of the jth key slice in the key slice information; I j+1 represents the gray scale image of the j+1th key slice in the key slice information; M represents the number of key slices participating in the three-dimensional reconstruction process; and p(I j ,I j+1represents the Pearson gray scale correlation coefficient between two pictures; I j (u,v) represents the gray scale value at pixel coordinate (u,v) in the gray scale image of the jth key section; I j+1 (u,v) represents the gray scale value at pixel coordinate (u,v) in the gray scale image of the j+1th key section; represents the gray scale mean value of the gray scale image of the jth key section; represents the gray scale mean value of the gray scale image of the j+1th key section.

[0074] In order to facilitate the process of using the formula and conveniently use the formula, the following shows a calculation example, in which the units of the known conditions correspond to the input values of the parameters in the process of using the above formula. Input the parameter values in the corresponding units, which can improve the accuracy of formula use.

[0075] Calculation example:

[0076] Known conditions: N e = 1000, S c = 40, N vox = 120000, v unit = 0.125 mm 3 , W s = 60 mm, H s = 40 mm, D s = 30 mm, A noise = 120 mm 2 , A total = 6000 mm 2 , the number of frames of probe position information collected N = 6, the distances between each frame d1 to d5 are 2.1 mm, 1.9 mm, 2.3 mm, 2.0 mm, 2.2 mm, the total number of slices M = 5, the Pearson gray scale correlation coefficients between two pictures are 0.88, 0.90, 0.91, 0.89, q ref = 500. Substituting into the formula to calculate:

[0077]

[0078] Since Q ≈ 535.1 > q ref , the quality evaluation output module generates three-dimensional reconstruction quality evaluation information for indicating that the image quality meets the standard, and at this time the optimization suggestion information generated by the report generation module includes: direct archiving, providing a 3D visual interactive interface for diagnosis, enabling export of the model for preoperative navigation.

[0079] The program code of the above calculation example is as follows:

[0080] import math

[0081] #---Input section (known conditions)---

[0082] #1. Structural clarity parameter

[0083] N_e = 1000

[0084] avg_gradient = 40 # G(i)

[0085] S_c = avg_gradient # since each G(i) is 40

[0086] S_c_squared = S_c ** 2

[0087] #2. Reconstructed completeness parameter

[0088] N_vox = 120000

[0089] v_unit = 0.125 # mm 3

[0090] V_rec = N_vox * v_unit

[0091] W_s = 60 # mm

[0092] H_s = 40 # mm

[0093] D_s_depth = 30 # mm

[0094] V_theo = W_s * H_s * D_s_depth

[0095] V_r = V_rec / V_theo

[0096] sqrt_V_r = math.sqrt(V_r)

[0097] #3. Anomalous noise parameter A_noise = 120 # mm 2

[0098] A_total = 6000 # mm 2

[0099] A_n = A_noise / A_total

[0100] e_An = math.exp(A_n)

[0101] denominator_noise = 1 + e_An

[0102] #4. Probe movement stability parameter d_k = [2.1, 1.9, 2.3, 2.0, 2.2] # distances between adjacent frames, 5 segments in total

[0103] # Calculate average displacement

[0104] d_mean = sum(d_k) / len(d_k)

[0105] # Calculate standard deviation

[0106] variance = sum((dk - d_mean) ** 2 for dk in d_k) / (len(d_k)) D_s = math.sqrt(variance)

[0107] # Calculate sine penalty term sin_arg = (math.pi / 2) * D_s

[0108] sin_val = math.sin(sin_arg)

[0109] probe_stability_factor = 1 - sin_val

[0110] # 5. Structure continuity parameter rho = [0.88, 0.90, 0.91, 0.89]

[0111] C_o = sum(rho) / len(rho)

[0112] # --- Q value calculation ---

[0113] # Molecular part

[0114] numerator = S_c_squared + sqrt_V_r

[0115] # Overall Q value

[0116] Q = (numerator / denominator_noise) * probe_stability_factor * C_o

[0117] # --- Output process and results ---

[0118] print("====== Q value score calculation process ======")

[0119] print(f"Structure clarity S_c: {S_c}")

[0120] print(f"Structure clarity squared S_c^2: {S_c_squared}")

[0121] print(f"Reconstruction completeness V_r: {V_r:.4f}")

[0122] print(f"√V_r:{sqrt_V_r:.4f}")

[0123] print(f"Abnormal noise area ratio A_n:{A_n:.4f}")

[0124] print(f"1+e^A_n:{denominator_noise:.4f}")

[0125] print(f"Probe stability D_s:{D_s:.4f}")

[0126] print(f"Sine penalty term (1-sin): {probe_stability_factor: .4f}")

[0127] print(f"Structural continuity C_o:{C_o:.4f}")

[0128] print(f"\nFinal Q value: {Q:.2f}").

[0129] Combine Figure 5 As shown, Figure 5 This is a diagram showing the statistical effect of the Q-value calculation results, which shows the calculation results of the Q-value calculated by the system multiple times.

[0130] In summary, by setting the expanded calculation formula for the structural continuity score, the system can perform pairwise correlation calculations based on the grayscale image matrix between consecutive key sections, and quantitatively analyze the structural consistency of adjacent sections through the Pearson correlation coefficient. This is beneficial for capturing the continuity of tissue texture and changes in spatial consistency in the image reconstruction sequence, and then identifying abnormal conditions such as section dislocation, faults or jitter, which is beneficial for improving the coherence and stability of the three-dimensional reconstruction results in spatial structural expression.

[0131] The contents disclosed above are only preferred feasible embodiments of the present invention and do not limit the scope of protection of the present invention. Therefore, all equivalent technical changes made using the contents of the present invention description and drawings are included in the scope of protection of the present invention. In addition, the elements therein can be updated as technology develops.

Claims

1. An ultrasonic image processing system based on three-dimensional reconstruction, characterized in that: It includes an ultrasound data acquisition terminal, an image processing terminal, a three-dimensional reconstruction terminal, a quality assessment terminal and a doctor interaction terminal; the ultrasound data acquisition terminal is used to collect ultrasound original radio frequency data and B-mode images, and synchronously obtain the probe position information of the ultrasound probe; the image processing terminal is used to perform image denoising, edge enhancement, tissue feature annotation based on the ultrasound original radio frequency data and B-mode images, and extract key section information for subsequent reconstruction; the three-dimensional reconstruction terminal is used to perform spatial reconstruction based on key section data and probe position information to form a three-dimensional ultrasound image volume; the quality assessment terminal is used to score the three-dimensional reconstruction effect of the three-dimensional ultrasound image volume and generate three-dimensional reconstruction quality assessment information; the doctor interaction terminal is used for doctors to view the three-dimensional ultrasound image volume and three-dimensional reconstruction quality assessment information in real time, and generate and display optimization suggestion information.

2. The ultrasonic image processing system based on three-dimensional reconstruction according to claim 1, characterized in that: The ultrasound data acquisition terminal includes an ultrasound data acquisition module, a probe position information acquisition module and a data transmission module; the ultrasound data acquisition module is used to acquire ultrasound raw radio frequency data and B-mode images; the probe position information acquisition module is used to synchronously obtain the probe position information of the ultrasound probe; the data transmission module is used to transmit the collected ultrasound raw radio frequency data, B-mode images and probe position information to the image processing terminal in real time.

3. The ultrasonic image processing system based on three-dimensional reconstruction according to claim 1, characterized in that: The image processing terminal includes an image preprocessing module, a feature extraction module, and a key section extraction module; the image preprocessing module is used to receive raw ultrasound radio frequency data, perform envelope detection, grayscale mapping, and denoising on the raw ultrasound radio frequency data, and convert the raw ultrasound radio frequency data into an enhanced grayscale image; The feature extraction module is used to perform edge enhancement and tissue structure annotation based on the B-mode image and the enhanced grayscale image; the key section extraction module is used to extract key section information for subsequent reconstruction from the image processed by the feature extraction module.

4. The ultrasonic image processing system based on three-dimensional reconstruction according to claim 1, characterized in that: The three-dimensional reconstruction terminal includes a data alignment module, a spatial reconstruction module and a three-dimensional display module; the data alignment module is used to spatially align key section information according to probe position information; the spatial reconstruction module is used to use the aligned key section information to perform three-dimensional reconstruction to form a three-dimensional ultrasound image volume; the three-dimensional display module is used to display the reconstructed three-dimensional ultrasound image volume.

5. The ultrasonic image processing system based on three-dimensional reconstruction according to claim 1, characterized in that: The quality assessment terminal includes an image scoring module and a quality assessment output module; the image scoring module is used to calculate the score of the three-dimensional ultrasound image volume; and the quality assessment output module is used to generate three-dimensional reconstruction quality assessment information according to the score.

6. The ultrasonic image processing system based on three-dimensional reconstruction according to claim 5, characterized in that: The doctor interaction terminal includes a real-time display module, an interactive control module and a report generation module; the real-time display module is used to display three-dimensional ultrasound images and three-dimensional reconstruction quality assessment information; the interactive control module is used to support doctors in real-time operations of magnifying, rotating and slicing image data; the report generation module is used to generate and display optimization suggestion information.

7. The ultrasonic image processing system based on three-dimensional reconstruction according to claim 6, characterized in that: The image scoring module includes a scoring parameter acquisition submodule and an image quality score value calculation submodule; the scoring parameter acquisition submodule is used to extract calculation parameters for calculating the image quality score value from the probe position information, key section information and the three-dimensional ultrasound image volume; the image quality score value calculation submodule is used to calculate the image quality score value according to the calculation parameters; When the image quality score calculation submodule is working, the following formula is satisfied: Where Q represents the image quality score of the 3D ultrasound image; S c Indicates structural clarity; V r Indicates the reconstruction completeness; A n represents the abnormal noise area ratio; D s Indicates the probe movement stability; C o Indicates structural continuity; when Q ref When Q≥q ref When the image quality is evaluated, the quality assessment output module generates three-dimensional reconstruction quality assessment information indicating that the image quality meets the standard. At this time, the optimization suggestion information generated by the report generation module includes: direct archiving, providing a 3D visual interactive interface for diagnosis, and enabling exporting models for preoperative navigation.​ 8. A method for processing ultrasonic images based on three-dimensional reconstruction, applied to the ultrasonic image processing system based on three-dimensional reconstruction according to claim 7, characterized in that: The ultrasonic image processing method based on three-dimensional reconstruction includes: S1, collecting ultrasound raw radio frequency data and B-mode images, and synchronously obtaining the probe position information of the ultrasound probe; S2, based on the original ultrasound RF data and B-mode images, performs image denoising, edge enhancement, tissue feature annotation, and extracts key section information for subsequent reconstruction; S3, spatial reconstruction is performed based on key section data and probe position information to form a three-dimensional ultrasound image volume; S4, scoring the three-dimensional reconstruction effect of the three-dimensional ultrasound image volume to generate three-dimensional reconstruction quality assessment information; S5 allows doctors to view 3D ultrasound images and 3D reconstruction quality assessment information in real time, and generate and display optimization suggestion information.

Images