Pulmonary vessel tree reconstruction method based on multi-scale watershed segmentation

Abstract

The invention relates to the technical field of medical image processing, in particular to a lung vessel tree reconstruction method based on multi-scale watershed segmentation, which comprises the following steps: acquiring lung CT image data; based on the lung CT image data, executing image preprocessing operation to obtain preprocessed image data; based on the preprocessed image data, executing multi-scale analysis to obtain multi-scale feature data; based on the multi-scale feature data, executing a drainage basin segmentation algorithm to obtain a blood vessel segmentation result; based on the blood vessel segmentation result, executing a blood vessel tree reconstruction operation to obtain reconstructed lung blood vessel tree data; according to the method, through innovative combination of multi-scale analysis and drainage basin segmentation, comprehensive processing of blood vessels with different thicknesses is realized. Multi-scale analysis can capture blood vessel characteristics under different scales, and drainage basin segmentation can accurately position blood vessel boundaries, so that the method can process thick and large blood vessels such as main pulmonary artery and the like and capillary vessels around tiny pulmonary alveoli at the same time.

Description

Technical Field

[0001] The present invention relates to the technical field of medical image processing, and more specifically, to a pulmonary vascular tree reconstruction method based on multi-scale watershed segmentation. Background Art

[0002] In the field of modern medical imaging, accurate reconstruction of the pulmonary vascular tree is of great significance for the diagnosis, treatment planning and prognosis assessment of lung diseases. With the continuous advancement of computed tomography (CT) technology, high-resolution CT images have made it possible to accurately reconstruct the pulmonary vascular tree. However, due to the complexity and diversity of the pulmonary vascular structure, achieving high-precision vascular tree reconstruction still faces many challenges.

[0003] Traditional pulmonary vascular tree reconstruction methods mainly include threshold segmentation, region growing, and model matching. These methods perform well when dealing with simple vascular structures, but often have many shortcomings when facing complex pulmonary vascular networks. For example, the threshold segmentation method has difficulty in dealing with noise and tissue heterogeneity in the image, and is prone to over-segmentation or under-segmentation; the region growing method is prone to leakage or interruption when dealing with vascular bifurcations and small blood vessels; the model matching method is not very accurate when dealing with blood vessels with large morphological variations.

[0004] In recent years, with the development of machine learning and deep learning technologies, blood vessel segmentation methods based on convolutional neural networks have made some progress. However, such methods often require a large amount of labeled data for training, and their performance is not stable enough when dealing with blood vessels of different thicknesses. In addition, methods that rely purely on data-driven methods are difficult to fully utilize prior knowledge of human vascular structure, and in some complex cases may produce reconstruction results that do not conform to anatomical laws.

[0005] The closest method in the prior art is the vascular segmentation algorithm based on multi-scale analysis. This type of method can better handle blood vessels of different thicknesses by analyzing image features at different scales. However, these methods still have difficulties in dealing with complex situations such as vascular bifurcations and vascular overlaps. Especially when dealing with small pulmonary blood vessels, since their diameters are close to the resolution limit of CT images, problems such as discontinuous segmentation or missed detections are prone to occur. In addition, existing multi-scale methods often find it difficult to effectively balance computational efficiency and reconstruction accuracy, and face the challenge of insufficient real-time performance in clinical applications.

[0006] In summary, the existing pulmonary vascular tree reconstruction methods still have significant deficiencies in dealing with complex vascular structures, balancing the reconstruction accuracy of vessels of different scales, and ensuring the anatomical rationality of the reconstruction results. Therefore, there is an urgent need for a new pulmonary vascular tree reconstruction method that can comprehensively consider multi-scale features, ensure the continuity and integrity of the reconstruction results, and take into account computational efficiency. Summary of the invention

[0007] The present invention aims to solve the above technical problems and proposes a pulmonary vascular tree reconstruction method based on multi-scale watershed segmentation. This method realizes the accurate segmentation and reconstruction of blood vessels of different thicknesses by innovatively combining multi-scale analysis and watershed segmentation algorithms. At the same time, by introducing an adaptive optimization mechanism, this method can effectively improve the computational efficiency while ensuring the reconstruction accuracy, thus meeting the needs of clinical applications.

[0008] In order to solve the above technical problems, the present invention adopts the following technical solutions: Pulmonary vascular tree reconstruction method based on multi-scale watershed segmentation, The acquisition steps include: Acquire lung CT image data; Processing steps include: Based on the lung CT image data, performing an image preprocessing operation to obtain preprocessed image data; Based on the preprocessed image data, performing multi-scale analysis to obtain multi-scale feature data; Based on the multi-scale feature data, executing a watershed segmentation algorithm to obtain a blood vessel segmentation result; Based on the blood vessel segmentation result, a blood vessel tree reconstruction operation is performed to obtain reconstructed pulmonary blood vessel tree data; Output steps include: The reconstructed pulmonary vascular tree data is output.

[0009] Preferably, the image preprocessing operation specifically includes: Based on the lung CT image data, a 3DSobel edge extraction operation is performed to obtain edge-enhanced image data; Based on the edge-enhanced image data, performing gradient amplitude calculation to obtain a gradient image; Based on the gradient image, adaptive threshold processing is performed to obtain a binary image.

[0010] Preferably, the multi-scale analysis specifically includes: Calculate the thinness ratio and roundness ratio of the blood vessel shape; Based on the Gaussian kernel function, the scale space is constructed; In the scale space, multi-scale feature extraction is performed.

[0011] Preferably, the watershed segmentation algorithm specifically includes: Divide the initial blood vessel edge into multiple regions; Construct the evolution function of the gradient in scale space; Calculating the integral of the evolution function over a specified interval; Based on the integration result, a watershed point is determined.

[0012] Preferably, the specified interval is [0,1]; The evolution function is p(u); The integration result is I(u).

[0013] Preferably, the vascular tree reconstruction operation specifically includes: Perform morphological processing to remove noise and small areas; Extract the blood vessel centerline; Constructing the topological structure of the vascular tree; The blood vessel diameter information is restored through the interpolation algorithm.

[0014] As a preference, a multi-scale fusion step is also included: Perform segmentation and reconstruction operations at different scales to obtain reconstruction results at multiple scales; A weighted fusion strategy is adopted to merge the reconstruction results of the multiple scales to obtain fused vascular tree data.

[0015] Preferably, the evaluation step is also included: Based on a preset evaluation index, evaluating the reconstructed pulmonary vascular tree data; The evaluation indicators include segmentation accuracy and reconstruction completeness.

[0016] As a preferred embodiment, the optimization step is also included: Based on the evaluation results, adjusting algorithm parameters; Using the adjusted algorithm parameters, the processing steps are re-executed.

[0017] Preferably, the reconstructed pulmonary vascular tree data includes: Vascular spatial location information; Blood vessel diameter information; Vascular branching structure information.

[0018] The method of the present invention has the following significant technical effects: First, the present invention achieves comprehensive processing of blood vessels of different thicknesses through the innovative combination of multi-scale analysis and watershed segmentation. Multi-scale analysis can capture the characteristics of blood vessels at different scales, while watershed segmentation can accurately locate the boundaries of blood vessels. The synergy of these two technologies enables this method to simultaneously process large blood vessels such as the main pulmonary artery and tiny peri-alveolar capillaries, significantly improving the integrity and accuracy of the reconstruction results.

[0019] Secondly, the adaptive threshold processing and gradient evolution function introduced in the present invention effectively solve the difficulties of traditional methods in dealing with complex vascular structures. Especially in areas such as vascular bifurcation and overlap, the present method can accurately identify vascular boundaries and avoid the problem of over-segmentation or under-segmentation. This feature is of great significance for accurately reconstructing the complex vascular network between lung lobes.

[0020] Furthermore, the method of the present invention effectively ensures the continuity and anatomical rationality of the reconstruction results by introducing steps such as morphological processing and vascular centerline extraction. This not only improves the reliability of the reconstruction results, but also provides a more accurate data basis for subsequent hemodynamic analysis. For example, when evaluating pulmonary hypertension, accurate vascular diameter and branch structure information are crucial.

[0021] In addition, the multi-scale fusion strategy and adaptive optimization mechanism of the present invention achieve a good balance between reconstruction accuracy and computational efficiency, which enables the method to meet the needs of clinical real-time diagnosis and provide timely imaging support for diseases that require rapid diagnosis, such as acute pulmonary embolism.

[0022] Finally, the comprehensive vascular tree data output by the method of the present invention provides rich information for subsequent medical research and clinical applications. For example, accurate vascular spatial position and diameter information can be used for lung cancer surgery planning to help doctors formulate the best resection plan; detailed branch structure information can be used to evaluate pulmonary vascular malformations and provide guidance for interventional treatment.

[0023] In summary, the pulmonary vascular tree reconstruction method based on multi-scale watershed segmentation proposed in this invention effectively solves the problems existing in the prior art through the organic combination of multiple innovative technologies, and significantly improves the accuracy, integrity and efficiency of pulmonary vascular tree reconstruction. This not only provides a powerful tool for related medical research, but also makes an important contribution to improving the diagnosis and treatment of lung diseases. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 The figure is an overall flow chart of the method of the present invention.

[0025] Figure 2 The figure is a flow chart of the image preprocessing of the present invention.

[0026] Figure 3 Flow chart of the multi-scale analysis of the present invention.

[0027] Figure 4 This is a flow chart of the watershed segmentation of the present invention.

[0028] Figure 5 This is a flow chart of the vessel tree reconstruction of the present invention.

[0029] Figure 6Flow chart of the evaluation and optimization process of the present invention. DETAILED DESCRIPTION

[0030] like Figure 1-6 As shown, the invention provides a method for reconstructing a pulmonary vascular tree based on multi-scale watershed segmentation. The method achieves accurate extraction and reconstruction of vascular structures in pulmonary CT images by innovatively combining multi-scale analysis and watershed segmentation algorithms. The technical solution of the invention will be described in detail below.

[0031] First, the method of the present invention includes an acquisition step, a processing step and an output step. In the acquisition step, the method acquires lung CT image data. These CT image data are usually acquired by CT scanning equipment of a medical institution and contain detailed structural information of the patient's lungs.

[0032] Next, in the processing step, the method first performs an image preprocessing operation based on the acquired lung CT image data to obtain preprocessed image data. The image preprocessing operation is a very critical step, which lays the foundation for subsequent analysis and processing. Preferably, in one embodiment of the present invention, the image preprocessing operation specifically includes the following steps: First, based on the lung CT image data, a 3D Sobel edge extraction operation is performed to obtain edge-enhanced image data. The 3D Sobel operator is an extension of the traditional 2D Sobel operator in three-dimensional space, which can better capture the three-dimensional structural information in the CT image. The mathematical expression of the 3D Sobel operator is as follows: , , , in, is the input 3D image data, Represents a convolution operation.

[0033] Then, based on the edge enhanced image data, gradient amplitude calculation is performed to obtain a gradient image. The calculation formula of the gradient amplitude is: , Finally, based on the gradient image, adaptive threshold processing is performed to obtain a binary image. The selection of the adaptive threshold is usually based on the histogram distribution of the image. A commonly used method is the Otsu method, which determines the optimal threshold by maximizing the inter-class variance.

[0034] After the preprocessing is completed, the method performs multi-scale analysis based on the preprocessed image data to obtain multi-scale feature data. Multi-scale analysis is an innovation of the present invention, which can simultaneously process vascular structures of different thicknesses. In a preferred embodiment of the present invention, the multi-scale analysis specifically includes the following steps: First, the thinness ratio and roundness ratio of the blood vessel shape are calculated. and roundness ratio The calculation formula is as follows: , , Among them, L is the length of the blood vessel, W is the width of the blood vessel, A is the cross-sectional area of ​​the blood vessel, and P is the circumference of the blood vessel.

[0035] Next, the scale space is constructed based on the Gaussian kernel function. The expression of the Gaussian kernel function is: , in, is the scale parameter that controls the width of the Gaussian kernel.

[0036] Finally, multi-scale feature extraction is performed in the scale space. This step can effectively capture vascular structures of different sizes by analyzing image features at different scales.

[0037] After the multi-scale analysis is completed, the method executes the watershed segmentation algorithm based on the multi-scale feature data to obtain the blood vessel segmentation result. The watershed segmentation algorithm is an image segmentation method based on topological theory. It regards the image as a topographic map and implements segmentation by simulating water flow. In the present invention, the specific implementation steps of the watershed segmentation algorithm are as follows: First, the initial blood vessel edge is divided into multiple regions. This step is achieved by detecting local minimum points.

[0038] Then, we construct the evolution function of the gradient in the scale space. It describes the law of gradient changing with scale, and its general form can be expressed as: , in, is the image gradient, is the scale parameter.

[0039] Next, the integral of the evolution function on the specified interval is calculated. In the present invention, the specified interval is usually selected as [0,1]. The calculation formula is: , Finally, based on the integration results, the watershed point is determined. The determination of the watershed point is usually based on The local maximum of .

[0040] Through the above steps, this method can accurately segment the vascular structure in the lung CT image. This method based on multi-scale watershed segmentation has higher accuracy and robustness than traditional methods, especially when dealing with complex vascular branch structures.

[0041] Finally, in the output step, the method outputs the reconstructed pulmonary vascular tree data, which contains information such as the spatial position, diameter, and branch structure of the blood vessels, providing an important basis for subsequent medical diagnosis and analysis.

[0042] Through the above detailed description, it can be seen that the method of the present invention has significant technical advantages and innovations in the field of pulmonary vascular tree reconstruction. It not only improves the accuracy of vascular segmentation and reconstruction, but also can effectively process vascular structures of different scales, providing strong support for related medical research and clinical applications. On the basis of the aforementioned technical solutions, the method of the present invention also includes some key steps and technical features to further improve the accuracy and effect of pulmonary vascular tree reconstruction.

[0043] First, the present invention provides a more detailed implementation method for the watershed segmentation algorithm. In a preferred embodiment of the present invention, the watershed segmentation algorithm specifically includes the following steps: First, the initial blood vessel edge is divided into multiple regions. The purpose of this step is to decompose the complex blood vessel network into relatively simple sub-regions to facilitate subsequent refinement. The division method is usually based on local features of the image, such as grayscale value, gradient, etc. Preferably, corrosion or dilation operations in morphological operations can be used to assist in region division.

[0044] Next, the evolution function of the gradient in the scale space is constructed. This step is the core of the watershed segmentation algorithm, which describes the law of image gradient changes with scale. In the embodiment of the present invention, the evolution function The specific form can be expressed as: , in, represents the image gradient, The scale is The Gaussian kernel function, Represents the convolution operation. This evolution function can effectively capture the gradient change characteristics of the image at different scales.

[0045] Then, the integral of the evolution function on the specified interval is calculated. In the present invention, the specified interval is selected as [0,1]. This selection is based on a large number of experiments and experience, and it can achieve a good balance between computational efficiency and accuracy. Integration results The calculation formula is: , This integral result reflects the cumulative effect of the gradient changes of the image in the entire scale space.

[0046] Finally, based on the integration result, the watershed point is determined. The determination of the watershed point is a key step in the watershed segmentation algorithm, which directly affects the final segmentation effect. In the method of the present invention, the determination of the watershed point is based on Specifically, if a point The value is greater than all points in its neighborhood value, the point is marked as a watershed point.

[0047] In order to further improve the robustness of the algorithm, the present invention also introduces some parameters to control the selection of watershed points. For example, a threshold T can be set, and only when When the local maximum of exceeds T, the point is marked as a watershed point. The choice of this threshold is usually determined according to the characteristics of the image and the specific application scenario. In the application of pulmonary vascular tree reconstruction, according to a large amount of experimental data, it is found that setting T to Around 70% of the maximum value can achieve good results.

[0048] Through the above steps, the watershed segmentation algorithm of the present invention can effectively segment the vascular structure in the lung CT image. Compared with the traditional single-scale method, this multi-scale watershed segmentation method can better handle blood vessels of different thicknesses, thereby improving the accuracy and completeness of segmentation.

[0049] After the watershed segmentation is completed, the method of the present invention further includes a blood vessel tree reconstruction operation. In one embodiment of the present invention, the blood vessel tree reconstruction operation specifically includes the following steps: First, morphological processing is performed to remove noise and small areas. This step aims to improve the quality and reliability of the reconstruction results. Commonly used morphological operations include opening and closing operations, which can effectively remove small noise areas while maintaining the integrity of the main vascular structure.

[0050] Then, the centerline of the blood vessel is extracted. The extraction of the centerline of the blood vessel is a key step in reconstructing the blood vessel tree, which provides a skeleton for the subsequent topological structure construction. The present invention adopts a centerline extraction algorithm based on distance transformation, which can accurately capture the center position of the blood vessel and maintain good results even in complex areas where the blood vessel is curved or bifurcated.

[0051] Next, the vascular tree topology is constructed. This step is based on the extracted centerline, and by analyzing the connectivity and branching relationship of the blood vessels, a complete vascular tree structure is constructed. During the construction process, the present invention adopts an adaptive connection strategy, which can effectively deal with the break problem caused by image noise or segmentation error.

[0052] Finally, the interpolation algorithm is used to restore the blood vessel diameter information. Blood vessel diameter information is crucial for subsequent hemodynamic analysis. The present invention uses an interpolation algorithm based on radial basis functions, which can smoothly restore the changes in blood vessel diameter and avoid the step effect that may be introduced by traditional linear interpolation methods.

[0053] In order to further improve the accuracy and completeness of the reconstruction results, the present invention also introduces a multi-scale fusion step. In a preferred embodiment of the present invention, the multi-scale fusion step specifically includes: First, segmentation and reconstruction operations are performed at different scales to obtain reconstruction results at multiple scales. The different scales here usually refer to different spatial resolutions or different characteristic scales. For example, 3 to 5 different scales can be selected, corresponding to the characteristic scales of large, medium, and small blood vessels, respectively.

[0054] Then, a weighted fusion strategy is adopted to merge the reconstruction results of the multiple scales to obtain the fused vascular tree data. The core of the weighted fusion strategy is to assign appropriate weights to the results of each scale. In one embodiment of the present invention, the weights are assigned based on the reliability and integrity of the results of each scale. Specifically, the following weighted formula can be used: , in, is the result of fusion. is the reconstruction result of the i-th scale, is the corresponding weight, and n is the number of scales. This can be determined by evaluating the quality of the results at each scale, for example, it can be calculated based on indicators such as vessel continuity and branch integrity.

[0055] Through this multi-scale fusion method, the present invention can effectively combine the reconstruction advantages at different scales, which not only ensures the overall structure of large blood vessels, but also accurately captures the details of small blood vessels, thereby obtaining a more complete and accurate pulmonary vascular tree reconstruction result.

[0056] The above describes in detail the specific implementation of the present invention in terms of watershed segmentation algorithm, vascular tree reconstruction and multi-scale fusion. The organic combination of these technical features makes the method of the present invention have significant advantages in the field of pulmonary vascular tree reconstruction, and can provide more reliable and accurate data support for related medical research and clinical applications.

[0057] On the basis of the above technical solutions, the present invention also introduces evaluation and optimization mechanisms to further improve the quality and reliability of pulmonary vascular tree reconstruction. These mechanisms can not only objectively evaluate the reconstruction results, but also adaptively adjust the algorithm parameters according to the evaluation results, thereby achieving continuous optimization of the method.

[0058] First, the method of the present invention includes an evaluation step. In a preferred embodiment of the present invention, the evaluation step specifically includes: evaluating the reconstructed pulmonary vascular tree data based on preset evaluation indicators. The evaluation indicators include segmentation accuracy and reconstruction completeness. The selection of these evaluation indicators is determined based on the special requirements of pulmonary vascular tree reconstruction and clinical application needs.

[0059] The segmentation accuracy is usually calculated by comparing with the gold standard of manual annotation. In an embodiment of the present invention, the Dice coefficient can be used to quantify the accuracy of segmentation. The calculation formula of the Dice coefficient is as follows: , Where X represents the algorithm segmentation result, and Y represents the gold standard of manual annotation. The Dice coefficient ranges from 0 to 1, and the larger the value, the more accurate the segmentation result. According to a large amount of experimental data, in the segmentation of pulmonary vascular trees, a Dice coefficient of 0.85 or above is generally considered a good result.

[0060] Reconstruction integrity mainly focuses on the integrity and continuity of the vascular tree structure. In one embodiment of the present invention, reconstruction integrity can be evaluated through the following aspects: 1. Number of vascular branches: Compared with anatomical knowledge, it is evaluated whether the reconstruction results capture enough vascular branches.

[0061] 2. Vascular continuity: Detect whether there are unreasonable breaks or gaps in the reconstruction results.

[0062] 3. Vascular diameter consistency: Evaluate whether the changes in vascular diameter conform to physiological laws.

[0063] Preferably, the present invention also introduces a comprehensive scoring mechanism to combine the above indicators to obtain an overall score. This comprehensive score can more comprehensively reflect the quality of the reconstruction result. For example, the following weighted summation formula can be used: , in, are the weights of various indicators, which can be adjusted according to specific application scenarios. In one embodiment of the present invention, these weights can be set as: =0.4, =0.2, =0.2, =0.2. This weight distribution takes into account both the importance of segmentation accuracy and the influence of other structural features.

[0064] After the evaluation step, the method of the present invention further comprises an optimization step. In one embodiment of the present invention, the optimization step specifically comprises: adjusting algorithm parameters based on the evaluation result; and re-executing the processing step using the adjusted algorithm parameters.

[0065] This optimization step actually constitutes a closed-loop feedback mechanism that can continuously improve the quality of the reconstruction results. Specifically, the following key parameters can be automatically adjusted based on the evaluation results: 1. Edge enhancement intensity in preprocessing: If it is found that the detection effect of small blood vessels is not good, the intensity of edge enhancement can be appropriately increased.

[0066] 2. Scale range in multiscale analysis: Adjust the scale range of multiscale analysis according to the integrity score of vascular branches.

[0067] 3. Threshold parameters in watershed segmentation: Fine-tune the threshold parameters in the watershed segmentation algorithm based on the segmentation accuracy.

[0068] 4. Connection strategy parameters in vascular tree reconstruction: According to the vascular continuity score, the connection strategy parameters are adjusted.

[0069] Preferably, the present invention adopts a parameter optimization algorithm based on gradient descent. The algorithm gradually adjusts the parameters in an iterative manner so that the evaluation score is continuously improved. The iterative process can be expressed as: , in, Represents the parameter set for the tth iteration is the learning rate, is the negative value of the evaluation score, represents the gradient of the evaluation score with respect to the parameters.

[0070] In this way, the method of the present invention can adaptively optimize the algorithm performance and continuously improve the quality of the reconstruction results. This self-optimization mechanism makes the method highly adaptable and can process various CT images generated by different patients and different scanning equipment.

[0071] Finally, the present invention clearly defines the content of the reconstructed pulmonary vascular tree data. In one embodiment of the present invention, the reconstructed pulmonary vascular tree data includes: vascular spatial position information, vascular diameter information and vascular branch structure information.

[0072] The spatial position information of blood vessels is usually expressed in the form of three-dimensional coordinates, which accurately describes the distribution of blood vessels in the lung space. Preferably, these coordinates can adopt the patient coordinate system defined in the DICOM standard to facilitate registration and fusion with other medical imaging data.

[0073] The blood vessel diameter information reflects the changes in the thickness of the blood vessel. In the embodiment of the present invention, the blood vessel diameter is usually recorded once every certain distance (for example, 1 mm) along the center line of the blood vessel. This representation method can accurately reflect the morphological characteristics of the blood vessel and maintain the compactness of the data.

[0074] The vascular branch structure information describes the topological structure of the vascular tree. Preferably, it can be represented by a tree data structure, in which each node contains information such as the location of the vascular branch point, the number of the connected sub-branch, etc. This representation method facilitates subsequent hemodynamic analysis and lesion location.

[0075] By providing this rich information, the pulmonary vascular tree data reconstructed by the present invention provides a solid foundation for various medical applications. For example, automatic detection of pulmonary embolism, assessment of pulmonary hypertension, surgical planning, etc. can be performed based on these data.

[0076] In summary, the present invention further improves the quality and practicality of pulmonary vascular tree reconstruction by introducing an evaluation and optimization mechanism and clearly defining the content of reconstruction data. The combination of these technical features makes the method of the present invention have significant advantages in accuracy, reliability and adaptability, providing strong support for related medical research and clinical applications.

[0077] The above description is only a preferred specific implementation manner of the present invention; however, the protection scope of the present invention is not limited thereto; any person familiar with the art who, within the scope disclosed by the present invention, makes equivalent replacements or changes based on the scheme and improved concepts of the present invention shall be covered within the protection scope of the present invention.

Claims

1. A pulmonary vascular tree reconstruction method based on multi-scale watershed segmentation, characterized in that: The acquisition steps include: Acquire lung CT image data; Processing steps include: Based on the lung CT image data, performing an image preprocessing operation to obtain preprocessed image data; Based on the preprocessed image data, performing multi-scale analysis to obtain multi-scale feature data; Based on the multi-scale feature data, executing a watershed segmentation algorithm to obtain a blood vessel segmentation result; Based on the blood vessel segmentation result, a blood vessel tree reconstruction operation is performed to obtain reconstructed pulmonary blood vessel tree data; Output steps include: The reconstructed pulmonary vascular tree data is output.

2. The method according to claim 1, characterized in that The image preprocessing operation specifically includes: Based on the lung CT image data, a 3DSobel edge extraction operation is performed to obtain edge-enhanced image data; Based on the edge-enhanced image data, performing gradient amplitude calculation to obtain a gradient image; Based on the gradient image, adaptive threshold processing is performed to obtain a binary image.

3. The method according to claim 1, characterized in that: The multi-scale analysis specifically includes: Calculate the thinness ratio and roundness ratio of the blood vessel shape; Based on the Gaussian kernel function, the scale space is constructed; In the scale space, multi-scale feature extraction is performed.

4. The method according to claim 1, characterized in that: The watershed segmentation algorithm specifically includes: Divide the initial blood vessel edge into multiple regions; Construct the evolution function of the gradient in scale space; Calculating the integral of the evolution function over a specified interval; Based on the integration result, a watershed point is determined.

5. The method according to claim 4, characterized in that: The specified interval is [0,1]; The evolution function is p(u); The integration result is I(u).

6. The method according to claim 1, characterized in that The vascular tree reconstruction operation specifically includes: Perform morphological processing to remove noise and small areas; Extract the blood vessel centerline; Constructing the topological structure of the vascular tree; The blood vessel diameter information is restored through the interpolation algorithm.

7. The method according to claim 1, characterized in that It also includes a multi-scale fusion step: Perform segmentation and reconstruction operations at different scales to obtain reconstruction results at multiple scales; A weighted fusion strategy is adopted to merge the reconstruction results of the multiple scales to obtain fused vascular tree data.

8. The method according to claim 1, characterized in that It also includes the evaluation step: Based on a preset evaluation index, evaluating the reconstructed pulmonary vascular tree data; The evaluation indicators include segmentation accuracy and reconstruction completeness.

9. The method according to claim 8, characterized in that Also includes optimization steps: Based on the evaluation results, adjusting algorithm parameters; Using the adjusted algorithm parameters, the processing steps are re-executed.

10. The method according to claim 1, characterized in that The reconstructed pulmonary vascular tree data includes: Vascular spatial location information; Blood vessel diameter information; Vascular branching structure information.

Images