Multi-channel cerebral hemorrhage minimally invasive drainage method and system

Abstract

The invention relates to the technical field of intelligent medical treatment, and discloses a multi-channel cerebral hemorrhage minimally invasive drainage method and system.The method comprises the steps that multi-modal medical image data of the brain of a patient is obtained, and an image data set of the brain of the patient is generated; performing structure enhancement processing on the image data set to obtain a structure enhanced image set of the brain of the patient; based on the structure enhancement image set, constructing a spatial relation matrix of a hematoma area and a peripheral key anatomical structure; according to the spatial relation matrix, constructing a hematoma three-dimensional structure diagram of the brain of the patient; performing spatial distribution characteristic analysis on the hematoma three-dimensional structure diagram to obtain a preliminary drainage path scheme of the brain of the patient; performing multi-channel collaborative arrangement optimization on the preliminary drainage path scheme to obtain a final drainage path scheme of the brain of the patient; according to the invention, the accuracy of multi-channel cerebral hemorrhage minimally invasive drainage can be improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent medical technology, and in particular to a multi-channel minimally invasive drainage method and system for cerebral hemorrhage. Background Technology

[0002] Current technologies have significant shortcomings in the image data processing stage of minimally invasive multi-channel cerebral hemorrhage drainage. After acquiring multimodal medical imaging data of the patient's brain, systematic format standardization processing is not performed on the original images from different sources and in different formats, nor is data integration carried out according to the correspondence of anatomical structures. This results in chaotic image data formats, missing anatomical structure correlations, and an inability to form a well-organized image dataset. At the same time, multimodal registration and targeted noise suppression are not performed on the image data, and key anatomical features such as vascular structures and ventricular boundaries are not enhanced. This makes it difficult to distinguish the details of the hematoma area and surrounding important structures in the images, making it difficult to accurately distinguish the hematoma range from key functional areas such as motor function areas and language centers. Consequently, the quality of the basic image data provided for subsequent drainage path planning is low.

[0003] Existing technologies have significant shortcomings in the core aspect of drainage path planning. They fail to construct a spatial relationship matrix between the hematoma area and surrounding key anatomical structures based on processed image data, relying solely on two-dimensional images or simple three-dimensional models for path design. This fails to accurately represent the spatial relationship between the hematoma and important blood vessels and functional areas, leading to the initial drainage path planning easily neglecting the need to avoid anatomical structures. Furthermore, when arranging multiple channels in the initial drainage path, the spatial interference between channels is not analyzed, nor are path parameters optimized through drainage effect simulation verification and safety performance assessment. Simply setting up a multi-channel layout according to a fixed pattern not only makes it difficult to guarantee drainage efficiency but may also increase surgical risks due to mutual interference between channels or contact with key functional areas. Overall, it fails to meet the precision and safety requirements of multi-channel minimally invasive drainage for cerebral hemorrhage. Summary of the Invention

[0004] This invention provides a multi-channel minimally invasive drainage method and system for cerebral hemorrhage to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a multi-channel minimally invasive drainage method for cerebral hemorrhage, comprising: S1. Acquire multimodal medical image data of the patient's brain and generate an image dataset of the patient's brain; S2. Perform structural enhancement processing on the image dataset to obtain a structurally enhanced image set of the patient's brain; S3. Based on the structure-enhanced image set, construct a spatial relationship matrix of the hematoma region and surrounding key anatomical structures; S4. Construct a three-dimensional structural map of the hematoma in the patient's brain based on the spatial relationship matrix; S5. Perform spatial distribution feature analysis on the three-dimensional structure map of the hematoma to obtain a preliminary drainage path plan for the patient's brain; S6. Optimize the preliminary drainage path scheme by multi-channel collaborative arrangement to obtain the final drainage path scheme for the patient's brain.

[0006] In a preferred embodiment, acquiring multimodal medical image data of the patient's brain and generating the image dataset of the patient's brain includes: Acquire raw, multimodal medical imaging data of the patient's brain; The original multimodal medical image data is subjected to data format standardization processing to obtain standardized image data of the patient's brain; The standardized image data is integrated according to the correspondence of anatomical structures to obtain the image dataset of the patient's brain.

[0007] In a preferred embodiment, the step of performing structural enhancement processing on the image dataset to obtain a structurally enhanced image set of the patient's brain includes: Multimodal image registration is performed on the image dataset to obtain a registered image set of the patient's brain; The registered image set is subjected to noise suppression processing to obtain a denoised image set of the patient's brain; Based on the denoised image set, feature enhancement processing is performed on the vascular structure and ventricular boundary to obtain enhanced feature images of the patient's brain; The enhanced feature images are fused with the denoised image set to form a structurally enhanced image set of the patient's brain.

[0008] In a preferred embodiment, the formula for calculating the enhanced feature image is as follows: ; In the formula, For the enhanced feature image, The denoised image set, The weighting coefficient for enhancing vascular features. This is a diagram showing the structural features of blood vessels. The maximum feature value in the vascular structure feature map. The enhancement weighting coefficient for ventricular features. This is a map showing the features of the ventricular boundaries. The maximum eigenvalue in the ventricle feature map. These are the weighting coefficients for enhancing edge features. This is the result after applying Laplacian edge enhancement to the original image.

[0009] In a preferred embodiment, constructing a spatial relationship matrix of the hematoma region and surrounding key anatomical structures based on the structure-enhanced image set includes: Based on the structure-enhanced image set, the hematoma region is segmented in three dimensions to obtain the spatial location and morphological feature data of the hematoma region. The enhanced image set was processed to identify key anatomical structures, and the spatial distribution information of motor functional areas, language centers and important vascular channels was extracted. By performing correlation analysis on the spatial location, the morphological feature data, and the spatial distribution information, a spatial relationship matrix of the hematoma region and surrounding key anatomical structures is established.

[0010] In a preferred embodiment, establishing a three-dimensional structural map of the hematoma in the patient's brain based on the spatial relationship matrix includes: Based on the spatial location and morphological characteristics of the hematoma region in the spatial relationship matrix, a three-dimensional mesh data of the hematoma surface in the patient's brain is constructed. Based on the spatial distribution information of key anatomical structures in the spatial relationship matrix, construct the spatial data of the anatomical structure of the patient's brain; The three-dimensional mesh data of the hematoma surface is spatially registered and fused with the spatial data of the anatomical structure to obtain a three-dimensional structural map of the hematoma in the patient's brain.

[0011] In a preferred embodiment, constructing three-dimensional mesh data of the hematoma surface in the patient's brain based on the spatial location and morphological feature data of the hematoma region in the spatial relationship matrix includes: Based on the spatial location data of the hematoma region in the spatial relationship matrix, a basic mesh structure of the hematoma surface in the patient's brain is generated; Based on the boundary curvature information in the morphological feature data, the basic mesh structure of the hematoma surface is adaptively refined to obtain the refined basic mesh of the hematoma surface. The refined hematoma surface base mesh is smoothed and optimized to obtain the optimized mesh structure of the patient's brain; The optimized mesh structure is registered and aligned with the anatomical markers in the spatial relationship matrix to generate three-dimensional mesh data of the hematoma surface of the patient's brain.

[0012] In a preferred embodiment, the step of analyzing the spatial distribution characteristics of the three-dimensional structure of the hematoma to obtain a preliminary drainage path plan for the patient's brain includes: Analyze the density distribution characteristics of the hematoma region in the three-dimensional structure diagram of the hematoma to determine the drainage priority order of the hematoma region; Assess the spatial relationship between the hematoma region and the surrounding critical anatomical structures to determine the avoidance range of the hematoma region; Based on the drainage priority order and the avoidance range, a potential drainage path for the hematoma area is constructed; A safety assessment of the potential drainage pathway was conducted to obtain a preliminary drainage pathway plan for the patient's brain.

[0013] In a preferred embodiment, the step of optimizing the preliminary drainage path scheme through multi-channel coordinated arrangement to obtain the final drainage path scheme for the patient's brain includes: The spatial relationship of the channels in the preliminary drainage path plan is analyzed and evaluated to obtain the evaluation results of the mutual interference between the channels in the patient's brain; Based on the evaluation results of the mutual interference between channels, the channel parameters in the preliminary diversion path scheme are adjusted to obtain the channel layout scheme of the preliminary diversion path scheme; Based on the three-dimensional structure diagram of the hematoma, the drainage effect of the channel layout scheme was simulated and verified to obtain drainage efficiency evaluation data of the channel layout scheme. Based on the preset important functional area avoidance requirements, a security analysis is performed on the diversion efficiency evaluation data to obtain the security verification results of the diversion efficiency evaluation data. Based on the safety verification results, the channel layout scheme is adjusted to obtain the final drainage path scheme for the patient's brain.

[0014] To address the aforementioned problems, the present invention also provides a multi-channel minimally invasive drainage system for cerebral hemorrhage, the system comprising: A multimodal image acquisition module is used to acquire multimodal medical image data of the patient's brain and generate an image dataset of the patient's brain. The structural enhancement processing module is used to perform structural enhancement processing on the image dataset to obtain a structurally enhanced image set of the patient's brain; The spatial relationship matrix construction module is used to construct a spatial relationship matrix of the hematoma region and surrounding key anatomical structures based on the structure-enhanced image set. A three-dimensional structure reconstruction module is used to construct a three-dimensional structural map of the hematoma in the patient's brain based on the spatial relationship matrix. The preliminary drainage path planning module is used to analyze the spatial distribution characteristics of the three-dimensional structure map of the hematoma to obtain a preliminary drainage path plan for the patient's brain. The multi-channel collaborative optimization module is used to optimize the preliminary drainage path plan through multi-channel collaborative arrangement to obtain the final drainage path plan for the patient's brain.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention acquires multimodal medical image data of the patient's brain through a multimodal image acquisition module, and generates an image dataset through format standardization and anatomical structure integration, providing a regular and comprehensive image foundation for subsequent analysis; the structure enhancement processing module further performs multimodal registration, noise suppression, and vascular and ventricular feature enhancement on the image dataset, and generates a structure-enhanced image set by combining Laplacian edge enhancement, clearly presenting the details of the hematoma and surrounding anatomical structures; the spatial relationship matrix construction module accurately segments the hematoma region, identifies key anatomical structures, and establishes a spatial relationship matrix, providing accurate spatial correlation data for three-dimensional reconstruction, and greatly improving the reliability of the basic data for subsequent drainage path planning.

[0016] 2. The three-dimensional structure reconstruction module of this invention constructs a three-dimensional mesh of the hematoma surface and spatial data of anatomical structures based on a spatial relationship matrix. After registration and fusion, it forms an accurate three-dimensional structure map of the hematoma, intuitively presenting the spatial distribution of the hematoma and surrounding structures. The preliminary drainage path planning module determines the drainage priority based on the hematoma density distribution, avoids key anatomical structures, and constructs a safe preliminary drainage path. The multi-channel collaborative optimization module adjusts and optimizes the parameters of the preliminary path by analyzing channel interference, simulating drainage effects, and verifying safety performance. Finally, it generates a multi-channel drainage scheme that takes into account both drainage efficiency and safety, significantly improving the accuracy and reliability of multi-channel minimally invasive drainage for cerebral hemorrhage, and providing scientific and practical path guidance for clinical surgery. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a minimally invasive multi-channel cerebral hemorrhage drainage method according to an embodiment of the present invention. Figure 2 A functional module diagram of a multi-channel minimally invasive drainage system for cerebral hemorrhage provided in an embodiment of the present invention; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] This application provides a multi-channel minimally invasive drainage method for cerebral hemorrhage. The executing entity of this multi-channel minimally invasive drainage method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the multi-channel minimally invasive drainage method for cerebral hemorrhage can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0020] Reference Figure 1 The diagram shown is a flowchart illustrating a multi-channel minimally invasive drainage method for cerebral hemorrhage according to an embodiment of the present invention. In this embodiment, the multi-channel minimally invasive drainage method for cerebral hemorrhage includes: S1. Acquire multimodal medical image data of the patient's brain and generate an image dataset of the patient's brain; In this embodiment of the invention, acquiring multimodal medical image data of the patient's brain and generating the image dataset of the patient's brain includes: Acquire raw, multimodal medical imaging data of the patient's brain; The original multimodal medical image data is subjected to data format standardization processing to obtain standardized image data of the patient's brain; The standardized image data is integrated according to the correspondence of anatomical structures to obtain the image dataset of the patient's brain.

[0021] Specifically, various types of medical imaging data of the patient's brain are obtained through hospital imaging equipment, including brain tomographic images generated by computed tomography (CT) equipment, brain magnetic resonance imaging (MRI) images generated by magnetic resonance imaging (MRI) equipment, and brain positron emission tomography (PET) images generated by positron emission tomography (PET) equipment. These imaging data contain scan images of different sections of the patient's brain and corresponding scan parameters, as well as metadata such as the patient's basic information. All the image data generated directly by the equipment and not processed are summarized to form the patient's raw multimodal medical imaging data.

[0022] Furthermore, the raw multimodal medical image data is imported into medical image processing software. First, the original format of each image data is identified. Images generated by different devices often have different formats. A standard format commonly used in the medical field is selected, and the non-standard format image data is converted. During the conversion process, key information such as spatial resolution, pixel grayscale value, and tomographic thickness of the image remains unchanged. At the same time, the metadata fields of the image data are unified. The content of the metadata generated by different devices, such as patient identity information, examination date, and scan sequence, is adjusted to a unified field name and record format. After the format of all data is unified, a complete set of image data is formed, which is the standardized image data of the patient's brain.

[0023] Furthermore, using MRI images that clearly show the brain's anatomical structure from the standardized imaging data as the reference images, other types of images in the standardized imaging data are spatially aligned with the reference images using medical image registration tools, ensuring that the spatial coordinates of the same anatomical structure in different types of images remain consistent. Subsequently, according to the brain's anatomical structure, the tomographic images corresponding to the same anatomical region in all types of images are classified. For example, all types of image slices containing the frontal lobe region are classified into the frontal lobe group, those containing the temporal lobe region are classified into the temporal lobe group, and so on. The final complete dataset that integrates the anatomical structure correspondence and includes multiple types of images is the patient's brain imaging dataset.

[0024] In summary, acquiring raw multimodal medical imaging data of the patient's brain, covering various image types generated by different examination devices, and completely preserving the original imaging information of different anatomical structures of the patient's brain, provides rich and undisturbed basic data for subsequent accurate analysis, avoiding incomplete judgment of brain lesions and anatomical structures due to data loss.

[0025] In summary, standardizing the format of raw multimodal medical image data unifies the diverse formats generated by different devices into a common standard format in the medical field. At the same time, it standardizes the recording method of metadata fields, eliminates data reading errors and analysis biases caused by format confusion, and enables various types of image data to have a unified processing foundation, thereby improving the efficiency of subsequent data integration and analysis.

[0026] In summary, standardized image data is integrated according to the correspondence of anatomical structures. Spatial alignment enables precise matching of the same anatomical structures in images from different modalities. Relevant image slices are then categorized by anatomical region, forming a structurally well-structured image dataset. This dataset clearly presents the performance of each anatomical structure in different modalities, facilitating rapid localization of image information for specific anatomical regions and providing well-organized data support for subsequent steps such as hematoma region identification and key anatomical structure analysis.

[0027] S2. Perform structural enhancement processing on the image dataset to obtain a structurally enhanced image set of the patient's brain; In this embodiment of the invention, the step of performing structural enhancement processing on the image dataset to obtain a structurally enhanced image set of the patient's brain includes: Multimodal image registration is performed on the image dataset to obtain a registered image set of the patient's brain; The registered image set is subjected to noise suppression processing to obtain a denoised image set of the patient's brain; Based on the denoised image set, feature enhancement processing is performed on the vascular structure and ventricular boundary to obtain enhanced feature images of the patient's brain; The enhanced feature images are fused with the denoised image set to form a structurally enhanced image set of the patient's brain.

[0028] The formula for calculating the enhanced feature image is as follows: ; In the formula, For the enhanced feature image, The denoised image set, The weighting coefficient for enhancing vascular features. This is a diagram showing the structural features of blood vessels. The maximum feature value in the vascular structure feature map. The enhancement weighting coefficient for ventricular features. This is a map showing the features of the ventricular boundaries. The maximum eigenvalue in the ventricle feature map. These are the weighting coefficients for enhancing edge features. This is the result after applying Laplacian edge enhancement to the original image.

[0029] Specifically, magnetic resonance imaging (MRI) images that clearly show the overall anatomical structure of the brain are selected as the baseline modality. Other modalities, such as computed tomography (CT) images and positron emission tomography (PET) images, are then registered with the baseline modality. By identifying common anatomical landmarks such as the falx cerebri, pineal gland, and internal capsule in each modality, the spatial position, rotation angle, and scaling of the non-baseline modality images are adjusted to ensure that the spatial coordinates of the same anatomical structures in different modalities correspond completely. This ensures that the anatomical structures at the same location are precisely superimposed in all modalities. After completing the registration of all modalities, the resulting multimodal image set containing spatial alignment is the registered image set of the patient's brain.

[0030] Furthermore, spatial filtering is used to suppress noise in the registered image set. Each image in the registered image set is traversed, and for each pixel in the image, a sliding window is formed by selecting a certain range of neighboring pixels. The average gray value of all pixels in the window is calculated, and this average value is used to replace the gray value of the original pixel. For obvious edge regions in the image, such as the ventricular boundary and the junction of gray matter and white matter, the gray value difference of edge pixels is preserved during the calculation to avoid edge blurring caused by filtering. In this way, random noise and artifacts in the image are removed, while the edge and detail information of the proximal structure are preserved. The complete image set after processing is the denoised image set of the patient's brain.

[0031] Furthermore, for the vascular structures concentrated in the denoised images, an enhancement method based on gray-level differences is adopted. By comparing the gray-level differences between the vascular region and the surrounding brain tissue, the gray-level contrast of the vascular region is improved, making the course and branches of the blood vessels clearer and more distinguishable. For the ventricular boundaries, edge enhancement processing is used to identify the boundary pixels between the ventricles and the surrounding brain tissue, increase the gray-level difference between these pixels and adjacent non-boundary pixels, enhance the continuity and clarity of the ventricular boundaries, and ensure that the vascular structures and ventricular boundaries are more prominent in the images. The images obtained after the above processing, which specifically highlight the features of blood vessels and ventricles, are the enhanced feature images of the patient's brain.

[0032] Furthermore, the enhanced feature images and the denoised image set are fused pixel by pixel. For each pixel location in each image, if the location belongs to a vascular structure or ventricular boundary, the gray value of that pixel in the enhanced feature image is used to preserve the clarity of the enhanced features; if the location belongs to other brain tissue regions, the gray value of the corresponding pixel in the denoised image set is used to maintain the original information of the overall structure. During the fusion process, it is ensured that the spatial positions of the two images are completely matched without misalignment or overlap. The complete image set formed after fusion, which contains both a clear overall structure and highlights key features, is the structural enhanced image set of the patient's brain.

[0033] Specifically, among the parameters of the enhanced feature image, the enhanced feature image comes from the final image result calculated by the entire formula, and its generation depends on the synergistic effect of other parameters.

[0034] Furthermore, the denoised image set, derived from noise suppression processing of the registered image set, serves as the foundational image data for generating enhanced feature images. The enhancement weight coefficient for vascular features is set according to the needs of clinical image analysis; a larger value is set if vascular structures require focused observation, and a smaller value is set otherwise.

[0035] Furthermore, the vascular structure feature map is generated by extracting concentrated vascular regions from the denoised image. First, regions in the image with significantly different grayscale values ​​from the surrounding brain tissue are identified; these regions correspond to vascular structures. Then, the pixel information of these regions is integrated to form an image specifically reflecting the distribution and morphology of blood vessels. The maximum eigenvalue in the vascular structure feature map is obtained by traversing the feature values ​​of all pixels in the map and selecting the largest value.

[0036] Furthermore, the enhancement weighting coefficient for ventricular features is also set according to clinical needs. If a clear view of the ventricular boundaries is required, a larger value is set, and vice versa. The ventricular boundary feature map is generated by extracting the boundary region between the ventricles and surrounding brain tissue from the denoised image, identifying pixels at the ventricular boundaries, and integrating this pixel information to form an image reflecting the morphology of the ventricular boundaries.

[0037] Furthermore, the maximum eigenvalue in the ventricle feature map is determined by traversing all feature values ​​in the ventricle boundary feature map and selecting the maximum value. The enhancement weight coefficient for edge features is set according to the requirement for image edge sharpness; a larger value is set when the edges need to be emphasized.

[0038] Furthermore, the result of Laplacian edge enhancement processing on the original image is obtained through the Laplacian edge enhancement method. Specifically, it involves traversing each pixel of each image in the denoised image set, selecting the neighboring pixels around the pixel, calculating the difference in grayscale values ​​between the surrounding pixels and the center pixel, and adjusting the grayscale value of the center pixel according to the difference to make the grayscale contrast of the pixels at the edge more obvious, thereby strengthening the edge structure in the image. The image obtained after processing is the result.

[0039] Furthermore, the significance of this formula lies in using the denoised image set as a base, superimposing the enhancement effects of vascular structures and ventricular boundaries, and then fusing the results of edge enhancement processing to generate enhanced feature images that highlight key brain structures. Specifically, multiplying the denoised image set by 1 and summing it with the vascular enhancement term and the ventricular enhancement term, while preserving the original overall brain structural information of the denoised images, superimposes the enhancement effects of blood vessels and ventricles according to set weights, making the course, branches, and ventricular boundaries of blood vessels clearer. Adding the results of edge enhancement processing further strengthens the edges of all structures in the images, ensuring that both the overall anatomical structure of the brain and key local features are clearly presented, facilitating subsequent observation and analysis of brain structures.

[0040] Furthermore, the trend of this formula is that when the enhancement weight coefficient of vascular features increases, the enhancement effect of vascular structures in enhanced feature images will be more significant, and the course and branches of blood vessels will be more clearly distinguishable; if the coefficient decreases, the enhancement effect of blood vessels will weaken, and the contrast between blood vessels and surrounding brain tissue will decrease.

[0041] Furthermore, when the enhancement weighting coefficient of ventricular features increases, the ventricular boundaries become more prominent in the enhanced feature images, and the shape and extent of the ventricles become easier to identify; when the coefficient decreases, the enhancement effect of the ventricular boundaries weakens, and the boundary with the surrounding tissues becomes blurred.

[0042] Furthermore, when the enhancement weight coefficient of edge features increases, the edges of all structures in the enhanced feature image become clearer, and the overall anatomical structure has a stronger sense of hierarchy; when the coefficient decreases, the edge enhancement effect weakens, and the transition between structures becomes smoother.

[0043] Furthermore, when the maximum eigenvalue in the vascular structure feature map increases, if the enhancement weight coefficient of the vascular feature remains unchanged, the value of the vascular enhancement term will increase, and the vascular enhancement effect will be more obvious; conversely, when the maximum eigenvalue decreases, the value of the vascular enhancement term decreases, and the enhancement effect weakens.

[0044] Furthermore, when the maximum eigenvalue in the ventricular feature map increases, if the enhancement weight coefficient of the ventricular features remains unchanged, the value of the ventricular enhancement term will increase, and the enhancement of the ventricular boundary will be more prominent; when the maximum eigenvalue decreases, the value of the ventricular enhancement term decreases, and the enhancement effect weakens.

[0045] Furthermore, when the sharpness of the result after Laplacian edge enhancement is improved, i.e. the edges are more obvious, the overall edge sharpness of the enhanced feature image will also be improved simultaneously; if the result has blurry edges, the edge effect of the enhanced feature image will also deteriorate.

[0046] Furthermore, when the enhancement weighting coefficients of vascular features and ventricular features are increased simultaneously, the vascular structures and ventricular boundaries in the enhanced feature images are enhanced simultaneously. The enhancement effects of the two do not interfere with each other, and together they improve the visibility of these two key structures in the images. When the enhancement weighting coefficients of edge features and vascular and ventricular features are increased simultaneously, the enhanced feature images will highlight blood vessels and ventriculars while making the edges of all structures clearer, thus presenting more comprehensive details of brain structures.

[0047] In summary, multimodal image registration aligns the interpretation structures of different modalities, eliminates spatial misalignment, and provides a registered image set with a unified spatial reference for subsequent processing.

[0048] In summary, noise suppression removes image noise artifacts, preserves edge details, generates a high-quality denoised image set, and reduces the interference of invalid information on subsequent processing.

[0049] In summary, enhancing vascular and ventricular features and improving their contrast with brain tissue generates enhanced feature images, which helps to accurately locate key anatomical structures.

[0050] In summary, by fusing enhanced features and denoised images, a set of structure-enhanced images is formed that balances overall structural integrity with clear local features, supporting subsequent precise analysis.

[0051] In summary, the formula uses a set of denoised images as the base data. These images have been de-noised and retain key structural details, providing a high-quality original carrier for enhancement processing. This ensures that the enhanced images do not deviate from the real anatomical structure of the brain and avoids the impact of feature presentation due to distortion of the base data.

[0052] In summary, by adjusting the enhancement weighting coefficients for blood vessels, ventricles, and limbic features, the enhancement intensity can be flexibly adjusted according to the focus of clinical observation, allowing key structures to be clearly highlighted in images as needed, thereby improving the efficiency of identifying specific anatomical structures.

[0053] In summary, the maximum eigenvalue of the vascular and ventricular feature maps is introduced to normalize the feature maps, ensuring that the enhancement of the vascular and ventricular features remains within a reasonable range. This prevents excessive enhancement of local features from obscuring other important structures and ensures the overall balance of image information.

[0054] In summary, simultaneously enhancing the structural features of blood vessels and ventricles, as well as the overall edge features of the images, not only clearly presents key anatomical structures closely related to the drainage of cerebral hemorrhage, but also enhances the edge clarity of all brain structures, providing comprehensive and accurate imaging evidence for subsequent hematoma area identification and spatial relationship analysis.

[0055] S3. Based on the structure-enhanced image set, construct a spatial relationship matrix of the hematoma region and surrounding key anatomical structures; In this embodiment of the invention, constructing a spatial relationship matrix of the hematoma region and surrounding key anatomical structures based on the structure-enhanced image set includes: Based on the structure-enhanced image set, the hematoma region is segmented in three dimensions to obtain the spatial location and morphological feature data of the hematoma region. The enhanced image set was processed to identify key anatomical structures, and the spatial distribution information of motor functional areas, language centers and important vascular channels was extracted. By performing correlation analysis on the spatial location, the morphological feature data, and the spatial distribution information, a spatial relationship matrix of the hematoma region and surrounding key anatomical structures is established.

[0056] Specifically, the structure-enhanced image set is imported into medical imaging 3D processing software. Based on the grayscale difference between the hematoma region and the surrounding brain tissue in the images, multiple feature points on the edge of the hematoma region are manually selected. The software automatically generates closed contour lines based on these feature points, covering the range of the hematoma region on each image slice. Then, the hematoma contour lines on all slices are stacked in 3D to form a complete 3D model of the hematoma. Using the software's built-in measurement function, the coordinate range of the 3D model in the spatial coordinate system is extracted to determine the spatial location of the hematoma. At the same time, data such as the model's volume, surface area, longest diameter, and shortest diameter are obtained. These data together constitute the morphological feature data of the hematoma region. After the two are integrated, they are the spatial location and morphological feature data of the hematoma region.

[0057] Furthermore, a medical imaging anatomical structure recognition tool was used to process the structure-enhanced image set. Based on the typical morphology and grayscale characteristics of brain anatomical structures, the tool first identified motor function areas by locating specific brain regions in the cerebral cortex related to limb motor control, such as the precentral gyrus of the frontal lobe, to determine their specific range in the images. Then, the language center was identified by locking onto the areas in the dominant hemisphere of the brain responsible for language processing, such as the posterior part of the superior temporal gyrus and the posterior part of the inferior frontal gyrus, to clarify their spatial distribution.

[0058] Furthermore, important vascular pathways are identified. Based on the high grayscale features of blood vessels in enhanced images, the course and distribution range of major cerebral blood vessels such as the anterior cerebral artery, middle cerebral artery and their branches are traced. The spatial range, location coordinates and other information of the motor function area, language center and important vascular pathways are compiled and summarized, which is the spatial distribution information of the motor function area, language center and important vascular pathways.

[0059] Furthermore, using the standard anatomical coordinate system of the brain as a unified reference, the spatial location data of the hematoma region is compared with the spatial distribution information of the motor function area, language center, and important vascular channels. The shortest straight-line distance between the hematoma region and each key anatomical structure is calculated to determine whether the hematoma is compressing a certain structure. If the shortest distance is zero or negative, it indicates that the hematoma overlaps with or directly compresses the structure.

[0060] Furthermore, by combining the morphological characteristics of the hematoma, the influence of morphological parameters such as the volume and longest diameter of the hematoma on surrounding key structures is analyzed. For example, a large hematoma may compress multiple adjacent structures. These distance data, compression relationships, and degree of influence are arranged in matrix form. The rows of the matrix correspond to different characteristics of the hematoma region, such as spatial location, volume, and longest diameter, while the columns correspond to key anatomical structures such as motor functional areas, language centers, and important vascular channels. Each element in the matrix is ​​filled with the correlation information between the corresponding characteristics of the hematoma and the anatomical structure. The matrix established in this way is the spatial relationship matrix of the hematoma region and surrounding key anatomical structures.

[0061] In summary, by relying on the structure-enhanced image set to segment the hematoma in three dimensions, and by generating a three-dimensional model through grayscale differences to extract spatial location and morphological data, the problem of boundary ambiguity is solved, providing accurate and comprehensive core data of the hematoma for subsequent analysis.

[0062] In summary, by leveraging the clear details in structurally enhanced images, anatomical recognition tools can be used to locate key structures and extract their spatial distribution information, thus clarifying the focus of surgical avoidance and reducing surgical risks caused by unclear structural identification.

[0063] In summary, by using a unified coordinate system to correlate and analyze hematoma and key structural data, and by using matrices to quantify spatial relationships, we can provide a systematic basis for 3D modeling and path planning, thereby improving the accuracy and safety of drainage.

[0064] S4. Construct a three-dimensional structural map of the hematoma in the patient's brain based on the spatial relationship matrix; In this embodiment of the invention, establishing a three-dimensional structural map of the hematoma in the patient's brain based on the spatial relationship matrix includes: Based on the spatial location and morphological characteristics of the hematoma region in the spatial relationship matrix, a three-dimensional mesh data of the hematoma surface in the patient's brain is constructed. Based on the spatial distribution information of key anatomical structures in the spatial relationship matrix, construct the spatial data of the anatomical structure of the patient's brain; The three-dimensional mesh data of the hematoma surface is spatially registered and fused with the spatial data of the anatomical structure to obtain a three-dimensional structural map of the hematoma in the patient's brain.

[0065] The construction of three-dimensional mesh data of the hematoma surface in the patient's brain based on the spatial location and morphological features of the hematoma region in the spatial relationship matrix includes: Based on the spatial location data of the hematoma region in the spatial relationship matrix, a basic mesh structure of the hematoma surface in the patient's brain is generated; Based on the boundary curvature information in the morphological feature data, the basic mesh structure of the hematoma surface is adaptively refined to obtain the refined basic mesh of the hematoma surface. The refined hematoma surface base mesh is smoothed and optimized to obtain the optimized mesh structure of the patient's brain; The optimized mesh structure is registered and aligned with the anatomical markers in the spatial relationship matrix to generate three-dimensional mesh data of the hematoma surface of the patient's brain.

[0066] Specifically, spatial location and morphological feature data of the hematoma region are extracted from the spatial relationship matrix. This data is then imported into 3D modeling software. The software determines the specific location and extent of the hematoma in the 3D coordinate system based on the spatial location data. Using the surface area, longest diameter, and shortest diameter from the morphological feature data as a basis, uniformly distributed vertices are generated within this range. Each vertex corresponds to a point on the hematoma surface. Then, the software algorithm connects adjacent vertices to form triangular mesh units. All mesh units are interconnected to cover the entire hematoma surface, ensuring that the mesh density matches the morphological complexity of the hematoma. Mesh units are finer in areas with complex morphology. The final generated complete data set consisting of vertices and triangular meshes is the 3D mesh data of the hematoma surface in the patient's brain.

[0067] Furthermore, spatial distribution information of motor functional areas, language centers, and important vascular pathways is extracted from the spatial relationship matrix. This information is then input into a 3D modeling tool for anatomical structures. Based on the spatial range and morphological characteristics of each key anatomical structure, the tool constructs a three-dimensional model composed of multiple curved surfaces for the motor functional areas, with the curvature of the surfaces matching the natural shape of the cerebral cortex. For the language center, a polygonal three-dimensional structure conforming to its anatomical location is generated, clearly defining its boundaries with surrounding brain regions. For important vascular pathways, a tubular 3D model is created, with the diameter of the tube corresponding to the thickness of the blood vessel, and the direction of the tube following the natural course of the blood vessel. After summarizing the 3D model data of all key anatomical structures, a complete data set containing the spatial location, shape, and mutual positional relationships of each structure is formed, which is the spatial data of the patient's brain anatomical structure.

[0068] Furthermore, using the standard anatomical coordinate system of the brain as a unified spatial reference, the three-dimensional mesh data of the hematoma surface and the spatial data of the anatomical structure are imported into the same three-dimensional fusion platform. The platform automatically adjusts the spatial position of the three-dimensional mesh data of the hematoma surface by comparing the coordinate information of the two in the coordinate system, so that the position of the hematoma and the positional relationship of each key anatomical structure in the spatial data of the anatomical structure completely conform to the distance and compression relationship recorded in the spatial relationship matrix. After the adjustment is completed, the platform superimposes and fuses the three-dimensional data of the two. During the fusion process, the clarity of the hematoma surface mesh and the integrity of the anatomical structure are maintained, and no information loss is caused by mutual occlusion. The final three-dimensional image that clearly displays the surface morphology of the hematoma and its spatial relationship with the surrounding key anatomical structures is the three-dimensional structural map of the hematoma in the patient's brain.

[0069] Specifically, spatial location data of the hematoma region is extracted from the spatial relationship matrix. This data clarifies the coordinate range and specific location of the hematoma in the standard anatomical coordinate system of the brain. This data is then imported into 3D mesh modeling software. The software generates an initial mesh framework in the corresponding space, consisting of uniformly distributed vertices and edges connecting the vertices, with the coordinate range as the boundary. The vertices are arranged at fixed intervals, and the edges connect adjacent vertices to form polygonal mesh units. All mesh units together constitute the basic structure covering the approximate surface of the hematoma. This structure is the basic mesh structure of the hematoma surface in the patient's brain.

[0070] Furthermore, the boundary curvature information of the hematoma is extracted from the morphological feature data. This information reflects the degree of curvature in different regions of the hematoma surface. This information is then imported into a mesh processing tool. The tool traverses each mesh cell of the basic mesh structure of the hematoma surface, determining the magnitude of the boundary curvature at the location of each cell. For regions with larger curvature, i.e., more pronounced surface curvature, the original mesh cells are divided into smaller cells, increasing the number of vertices and mesh density in that region, so that the mesh can better fit the curved surface morphology of the hematoma. For regions with smaller curvature, i.e., relatively flat surfaces, the size and density of the original mesh cells remain unchanged, without additional subdivision. The mesh structure obtained after this targeted processing is the refined basic mesh of the hematoma surface.

[0071] Furthermore, the refined hematoma surface base mesh is imported into a mesh smoothing tool. The tool uses a moving vertex smoothing method to process the mesh, traversing each vertex in the mesh, calculating the average spatial coordinates of the surrounding adjacent vertices, and adjusting the coordinates of the original vertex to the position corresponding to this average value. At the same time, it is ensured that the movement range of each vertex during the adjustment process does not exceed the side length of its mesh cell to avoid distortion of the hematoma surface shape due to excessive movement. After adjusting the coordinates of all vertices once, the above process is repeated twice to further eliminate sharp protrusions and irregular depressions on the mesh surface, making the overall shape of the mesh more consistent with the natural surface of the hematoma. The processed mesh structure is the optimized mesh structure of the patient's brain.

[0072] Furthermore, anatomical markers are extracted from the spatial relationship matrix. These markers are the coordinates of known fixed anatomical locations of the brain in the coordinate system. The optimized mesh structure and these anatomical markers are then imported into the registration software. The software first determines the spatial position of each anatomical marker in the optimized mesh structure, then calculates the deviation between the current position of the optimized mesh structure and the corresponding position of the marker. Based on the deviation, the spatial orientation of the optimized mesh structure is adjusted, including translation, rotation, and fine-tuning, so that the parts of the mesh structure related to the anatomical markers are accurately aligned with the markers. This ensures that the position of the mesh structure in the brain coordinate system is completely consistent with the actual position of the hematoma. The mesh data generated after alignment, which contains accurate spatial position and fine surface morphology, is the three-dimensional mesh data of the hematoma surface of the patient's brain.

[0073] In summary, based on the spatial location and morphological characteristics of the hematoma in the spatial relationship matrix, a three-dimensional mesh data of the hematoma surface is constructed to accurately restore the three-dimensional morphology of the hematoma, providing fine hematoma data support for subsequent modeling.

[0074] In summary, based on the spatial distribution information of key anatomical structures in the spatial relationship matrix, spatial data of anatomical structures are constructed, clearly presenting the spatial morphology of key structures and clarifying their anatomical association with hematoma.

[0075] In summary, spatial registration and fusion of the three-dimensional mesh data of the hematoma surface and the spatial data of the anatomical structure generate a three-dimensional structural map of the hematoma that intuitively shows the spatial relationship between the two, providing a visual basis for drainage path planning.

[0076] In summary, a basic mesh structure is generated based on the spatial location data of hematoma in the spatial relationship matrix, clarifying the orientation range of the hematoma in the three-dimensional coordinate system, and providing a precise spatial reference for subsequent mesh optimization.

[0077] In summary, the basic mesh is refined by combining the boundary curvature information of morphological feature data, the mesh is densified in curved areas, and the density is maintained in flat areas, so that the mesh can better fit the actual shape of the hematoma.

[0078] In summary, smoothing optimization is performed on the refined mesh, and the vertex coordinates are adjusted to eliminate sharp protrusions, making the mesh shape closer to the natural surface of the hematoma and improving the realism of the mesh data.

[0079] In summary, the optimized mesh is registered and aligned with the anatomical markers to ensure the mesh is accurately positioned in the brain coordinate system, laying a reliable foundation for subsequent fusion with anatomical structures.

[0080] S5. Perform spatial distribution feature analysis on the three-dimensional structure map of the hematoma to obtain a preliminary drainage path plan for the patient's brain; In this embodiment of the invention, the step of performing spatial distribution feature analysis on the three-dimensional structural map of the hematoma to obtain a preliminary drainage path scheme for the patient's brain includes: Analyze the density distribution characteristics of the hematoma region in the three-dimensional structure diagram of the hematoma to determine the drainage priority order of the hematoma region; Assess the spatial relationship between the hematoma region and the surrounding critical anatomical structures to determine the avoidance range of the hematoma region; Based on the drainage priority order and the avoidance range, a potential drainage path for the hematoma area is constructed; A safety assessment of the potential drainage pathway was conducted to obtain a preliminary drainage pathway plan for the patient's brain.

[0081] Specifically, observe the grayscale distribution of the hematoma region in the three-dimensional structure image of the hematoma. The higher the grayscale value, the greater the hematoma density in that region. Use a three-dimensional analysis tool to detect the grayscale value of the hematoma region point by point, divide the regions with similar grayscale values ​​into the same sub-region, calculate the average grayscale value of each sub-region, and arrange all sub-regions in order of grayscale value from high to low. The sub-region with the highest average grayscale value needs to be drained first, followed by the next sub-region, and so on. The order of the sub-regions formed is the drainage priority order of the hematoma region.

[0082] Furthermore, the spatial distance between the hematoma region and surrounding key anatomical structures in the three-dimensional hematoma structure map is calculated using three-dimensional measurement tools. The shortest distance between each key anatomical structure and the edge of the hematoma is recorded. According to clinical safety standards, a certain range is extended outward from each key anatomical structure as the area where drainage paths are prohibited. The size of the extension range is determined according to the importance of the structure. The extension range for motor function areas and language centers is larger than that for important blood vessel channels to ensure that the drainage path will not damage these structures. The prohibited areas corresponding to all key anatomical structures are summarized to form the avoidance range of the hematoma region.

[0083] Furthermore, a suitable entry point is selected from the patient's head surface. The entry point should be located in an area of ​​the scalp without important blood vessels and nerves. The sub-region with the highest drainage priority is used as the target endpoint. A straight line from the entry point to the target endpoint is drawn in three-dimensional space. It is checked whether the straight line crosses the avoidance range. If it does not cross the avoidance range, the straight line is used as a potential drainage path. If it does cross the avoidance range, the path is adjusted to a broken line, and a turning point is set outside the avoidance range so that the path reaches the target endpoint from the entry point through the turning point without entering the avoidance range. The same method is used to design paths for the sub-regions with the next lower drainage priority. All paths that meet the conditions are summarized as potential drainage paths for the hematoma area.

[0084] Furthermore, a safety check is performed on each potential drainage path to ensure that the path completely avoids the avoidance area. The shortest distance between the path and each key anatomical structure within the avoidance area is measured to ensure that the distance meets clinical safety requirements. At the same time, the length of the path is assessed, and shorter paths are preferred to reduce trauma. Other potential risks are checked for in the scalp, skull, brain tissue, and other areas traversed by the path. Paths with risks are eliminated, and all paths that meet the safety requirements are retained. The path that covers the highest priority sub-region for drainage and has the best safety is selected as the core path, and the remaining paths are considered as alternatives. The set of paths together constitutes the preliminary drainage path plan for the patient's brain.

[0085] In summary, by analyzing the density distribution characteristics of the hematoma region in the three-dimensional structural image of the hematoma, and by using three-dimensional analysis tools to detect gray values ​​point by point and divide sub-regions with similar gray values, the drainage priority is determined by sorting according to the average gray value. This ensures that high-density hematomas that require more removal are treated first, greatly improving the targeting and efficiency of drainage.

[0086] In summary, assessing the spatial relationship between the hematoma area and surrounding key anatomical structures, calculating distances using three-dimensional measurement tools, and delineating avoidance zones based on clinical standards are crucial to prevent damage to important structures such as motor function areas and language centers via the drainage path, thus strengthening the surgical safety barrier.

[0087] In summary, potential traffic diversion paths are constructed based on traffic diversion priority and avoidance range. With high-priority hematomas as the target, entry points and path directions are planned within the avoidance range, taking into account both the need for efficient traffic diversion and the protection of key structures, thus providing a solid and reasonable path foundation for subsequent optimization.

[0088] In summary, a comprehensive safety assessment of potential traffic diversion paths is conducted to check whether the avoidance range is maintained, calculate the distance to key structures, investigate the risks along the path, eliminate potentially hazardous paths and retain safe and effective paths, and form a scientific and standardized preliminary traffic diversion path plan, which strongly supports subsequent multi-channel collaborative optimization.

[0089] S6. Optimize the preliminary drainage path scheme by multi-channel collaborative arrangement to obtain the final drainage path scheme for the patient's brain.

[0090] In this embodiment of the invention, the step of optimizing the preliminary drainage path scheme through multi-channel coordinated arrangement to obtain the final drainage path scheme for the patient's brain includes: The spatial relationship of the channels in the preliminary drainage path plan is analyzed and evaluated to obtain the evaluation results of the mutual interference between the channels in the patient's brain; Based on the evaluation results of the mutual interference between channels, the channel parameters in the preliminary diversion path scheme are adjusted to obtain the channel layout scheme of the preliminary diversion path scheme; Based on the three-dimensional structure diagram of the hematoma, the drainage effect of the channel layout scheme was simulated and verified to obtain drainage efficiency evaluation data of the channel layout scheme. Based on the preset important functional area avoidance requirements, a security analysis is performed on the diversion efficiency evaluation data to obtain the security verification results of the diversion efficiency evaluation data. Based on the safety verification results, the channel layout scheme is adjusted to obtain the final drainage path scheme for the patient's brain.

[0091] Specifically, all channels in the preliminary drainage path plan are imported into a three-dimensional spatial analysis tool. The entry point, path direction, and end point of each channel are marked one by one. The tool measures the shortest spatial distance between any two channels and observes whether the channels intersect, overlap, or are too close to each other in the brain. If the shortest distance between two channels is less than the clinical safe distance, or if the paths intersect or overlap, it is determined that there is mutual interference. The interference location, degree of interference, and type of interference between each channel and other channels are recorded. The interference situation of all channels is compiled into a complete report, which is the assessment result of the mutual interference between channels in the patient's brain.

[0092] Furthermore, based on the assessment results of mutual interference between channels, the channel parameters in the preliminary drainage path plan are adjusted. For channels with mutual interference, if the interference is caused by the entry points being too close, the entry point of one channel is shifted within the safe area of ​​the scalp to ensure that the distance between the entry points of the two channels meets the safety requirements. If the interference is caused by the intersection of the paths, the turning point of one channel is adjusted to ensure that the paths of the two channels maintain a safe distance within the skull and do not intersect. If the interference is caused by the distance between channels being too small, the spatial orientation of one channel is slightly adjusted to increase the distance between the two channels. For channels without interference, their original parameters remain unchanged. After all channel parameters are adjusted, the resulting complete plan, which includes the entry points, path orientations, endpoints, and mutual positional relationships of each channel, is the channel layout plan of the preliminary drainage path plan.

[0093] Furthermore, the channel layout scheme is imported into the hematoma drainage simulation system. Based on the three-dimensional structure diagram of the hematoma, the system simulates the working state of each channel during the actual drainage process. By simulating and observing the speed at which hematoma in different sub-regions of the hematoma area is drained through the channels, the range of hematoma covered by each channel, the total amount of hematoma drained through all channels per unit time, the proportion of hematoma area covered by drainage, and the location of hematoma areas not covered by drainage, the system also records the emptying time of areas with high hematoma density during the drainage process. These simulated data, such as speed, coverage ratio, and emptying time, are compiled and summarized to form the drainage efficiency evaluation data of the channel layout scheme.

[0094] Furthermore, the preset avoidance requirements for important functional areas are retrieved. These requirements specify the minimum safe distance between the channel and key structures such as the motor function area, language center, and important blood vessels. The channel layout scheme corresponding to the drainage efficiency assessment data is analyzed against these requirements. The actual distance between each channel and each important functional area is measured to determine whether the minimum safe distance requirements are met. The channel path is checked to see if it approaches or crosses important functional areas. If a channel has high drainage efficiency but is too close to an important functional area, it needs to be marked as having a safety hazard. If a channel meets both the drainage efficiency requirements and the avoidance requirements, it is marked as safe and effective. The safety inspection results of all channels are combined with the drainage efficiency assessment data to form a comprehensive analysis report, which is the safety verification result of the drainage efficiency assessment data.

[0095] Furthermore, the channel layout plan is adjusted based on the safety verification results. If the safety verification results show that some channels have safety hazards but high drainage efficiency, the path of these channels is adjusted first. Under the premise of ensuring that they are far away from important functional areas and meeting the avoidance requirements, their drainage efficiency is preserved as much as possible. This can be achieved by adding path turning points or fine-tuning the position of the entry point. If some channels have low drainage efficiency but are safe, their path is optimized so that the channels are closer to sub-regions with high hematoma density and high drainage priority, thereby improving drainage efficiency. If there is a channel that is both unsafe and inefficient, it is directly eliminated and a new channel is redesigned to supplement its drainage function. After all adjustments are completed, the channel layout plan that meets the avoidance requirements of important functional areas and has high drainage capacity is the final drainage path plan for the patient's brain.

[0096] In summary, the analysis and evaluation of the spatial relationships of the channels in the preliminary diversion path plan, and the identification of interference such as intersections and excessive proximity between channels, provide a precise basis for subsequent adjustments to address interference issues.

[0097] In summary, the channel parameters in the preliminary plan were adjusted based on the inter-channel interference assessment results to resolve the interference problem and form a channel layout plan that is reasonable and free from mutual interference.

[0098] In summary, by combining the three-dimensional structure diagram of the hematoma with simulation to verify the drainage effect of the channel layout scheme, efficiency data such as drainage speed and coverage were obtained, providing a reference for scheme optimization.

[0099] In summary, by comparing the proposed solutions with the pre-set requirements for avoiding key functional areas, analyzing the safety of the traffic diversion efficiency data, identifying the potential safety hazards and advantages of the solutions, and providing a basis for safety judgment.

[0100] In summary, the channel layout was adjusted based on the safety verification results, balancing drainage efficiency and safety, ultimately resulting in a precise and safe final drainage path solution that meets clinical needs.

[0101] like Figure 2The diagram shown is a functional block diagram of a multi-channel minimally invasive drainage system for cerebral hemorrhage provided in an embodiment of the present invention.

[0102] The multi-channel minimally invasive drainage system 100 for cerebral hemorrhage described in this invention can be installed in an electronic device. Depending on the functions implemented, the multi-channel minimally invasive drainage system 100 may include a multimodal image acquisition module 101, a structural enhancement processing module 102, a spatial relationship matrix construction module 103, a three-dimensional structural reconstruction module 104, a preliminary drainage path planning module 105, and a multi-channel collaborative optimization module 106. The modules described in this invention can also be referred to as units, which are a series of computer program segments that can be executed by the processor of an electronic device and perform a fixed function, stored in the memory of the electronic device.

[0103] In this embodiment, the functions of each module / unit are as follows: The multimodal image acquisition module 101 is used to acquire multimodal medical image data of the patient's brain and generate an image dataset of the patient's brain. The structural enhancement processing module 102 is used to perform structural enhancement processing on the image dataset to obtain a structurally enhanced image set of the patient's brain; The spatial relationship matrix construction module 103 is used to construct a spatial relationship matrix of the hematoma region and surrounding key anatomical structures based on the structure-enhanced image set. The three-dimensional structure reconstruction module 104 is used to construct a three-dimensional structural map of the hematoma in the patient's brain based on the spatial relationship matrix. The preliminary drainage path planning module 105 is used to analyze the spatial distribution characteristics of the three-dimensional structure map of the hematoma to obtain a preliminary drainage path plan for the patient's brain. The multi-channel collaborative optimization module 106 is used to optimize the preliminary drainage path scheme through multi-channel collaborative arrangement to obtain the final drainage path scheme for the patient's brain.

[0104] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0105] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0106] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0107] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0108] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0109] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A multi-channel minimally invasive drainage method for cerebral hemorrhage, characterized in that, The method comprises: S1. Obtain multi-modal medical image data of the patient's brain to generate an image data set of the patient's brain; S2, structure enhancement processing is carried out on the image data set to obtain a structure enhanced image set of the patient's brain; S3, based on the structure enhanced image set, the spatial relationship matrix of the hematoma region and the surrounding key anatomical structure is constructed; S4, according to the spatial relationship matrix, the hematoma three-dimensional structure diagram of the patient's brain is constructed; S5, the spatial distribution characteristic analysis is carried out on the hematoma three-dimensional structure diagram, and the preliminary drainage path scheme of the patient's brain is obtained; S6, the multi-channel collaborative arrangement optimization is carried out on the preliminary drainage path scheme, and the final drainage path scheme of the patient's brain is obtained.

2. A minimally invasive drainage method for multiple channel brain hemorrhage as claimed in claim 1, wherein, The multi-modal medical image data of the patient's brain is obtained, and the image data set of the patient's brain is generated, which comprises: Obtain the original multi-modal medical image data of the patient's brain; The data format standardization processing is carried out on the original multi-modal medical image data to obtain the standardized image data of the patient's brain; According to the anatomical structure corresponding relationship, the data integration is carried out on the standardized image data to obtain the image data set of the patient's brain.

3. A minimally invasive drainage method for brain hemorrhage in multiple channels as claimed in claim 1, wherein, The structure enhancement processing is carried out on the image data set to obtain the structure enhanced image set of the patient's brain, which comprises: Multi-modal image registration is carried out on the image data set to obtain the registered image set of the patient's brain; Noise suppression processing is carried out on the registered image set to obtain the de-noised image set of the patient's brain; Based on the de-noised image set, the feature enhancement processing is carried out on the blood vessel structure and the ventricle boundary to obtain the enhanced feature image of the patient's brain; The enhanced feature image and the de-noised image set are fused into the structure enhanced image set of the patient's brain.

4. A minimally invasive drainage method for brain hemorrhage in multiple channels as claimed in claim 3, wherein, The calculation formula of the enhanced feature image is as follows: ； wherein, is the enhanced feature image of the blood vessel structure, is the set of denoised images, is the enhancement weight coefficient of the blood vessel structure feature, is the blood vessel structure feature map, is the maximum feature value in the blood vessel structure feature map, is the enhancement weight coefficient of the ventricle feature, is the ventricle boundary feature map, is the maximum feature value in the ventricle feature map, is the enhancement weight coefficient of the edge feature, is the result of the Laplace edge enhancement processing on the original image.

5. A minimally invasive drainage method for brain hemorrhage in multiple channels as claimed in claim 1 wherein, Based on the structure enhanced image set, the spatial relationship matrix of the hematoma region and the surrounding key anatomical structure is constructed, which comprises: Based on the structure enhanced image set, three-dimensional segmentation processing is carried out on the hematoma region to obtain the spatial position and morphological feature data of the hematoma region; The key anatomical structure recognition processing is carried out on the structure enhanced image set, and the spatial distribution information of the motor function area, the language center and the important blood vessel channel is extracted; The spatial position, the morphological feature data and the spatial distribution information are associated and analyzed to establish the spatial relationship matrix of the hematoma region and the surrounding key anatomical structure.

6. A minimally invasive drainage method for brain hemorrhage in multiple channels as claimed in claim 1 wherein, According to the spatial relationship matrix, the hematoma three-dimensional structure diagram of the patient's brain is established, which comprises: According to the hematoma region spatial position and morphological feature data in the spatial relationship matrix, the hematoma surface three-dimensional grid data of the patient's brain is constructed; According to the key anatomical structure spatial distribution information in the spatial relationship matrix, the anatomical structure spatial data of the patient's brain is constructed; The hematoma surface three-dimensional grid data and the anatomical structure spatial data are spatially registered and fused to obtain the hematoma three-dimensional structure diagram of the patient's brain.

7. A minimally invasive drainage method for multiple channel brain hemorrhage as claimed in claim 6 wherein, According to the hematoma region spatial position and morphological feature data in the spatial relationship matrix, the hematoma surface three-dimensional grid data of the patient's brain is constructed, which comprises: generating a hematoma surface basic grid structure of the patient's brain based on the hematoma region spatial position data in the spatial relationship matrix; performing adaptive refinement processing on the hematoma surface basic grid structure based on the boundary curvature information in the morphological feature data to obtain a refined hematoma surface basic grid; performing smoothing optimization processing on the refined hematoma surface basic grid to obtain an optimized grid structure of the patient's brain; aligning the optimized grid structure with the anatomical marker points in the spatial relationship matrix to generate hematoma surface three-dimensional grid data of the patient's brain.

8. A minimally invasive drainage method for brain hemorrhage in multiple channels as claimed in claim 1 wherein, The spatial distribution characteristic analysis of the hematoma three-dimensional structure diagram includes: analyzing the density distribution characteristics of the hematoma region in the hematoma three-dimensional structure diagram to determine the drainage priority order of the hematoma region; evaluating the spatial position relationship between the hematoma region and the surrounding key anatomical structures to determine the avoidance range of the hematoma region; constructing potential drainage paths of the hematoma region according to the drainage priority order and the avoidance range; performing safety evaluation on the potential drainage paths to obtain a preliminary drainage path scheme of the patient's brain.

9. A minimally invasive drainage method for brain hemorrhage in multiple channels as claimed in claim 1, wherein, The multi-channel collaborative arrangement optimization of the preliminary drainage path scheme includes: analyzing and evaluating the spatial position relationship of the channels in the preliminary drainage path scheme to obtain an inter-channel mutual interference condition evaluation result of the patient's brain; performing parameter adjustment on the channel parameters in the preliminary drainage path scheme based on the inter-channel mutual interference condition evaluation result to obtain a channel layout scheme of the preliminary drainage path scheme; performing drainage effect simulation verification on the channel layout scheme based on the hematoma three-dimensional structure diagram to obtain drainage efficiency evaluation data of the channel layout scheme; performing safety analysis on the drainage efficiency evaluation data based on the preset important functional area avoidance requirement to obtain a safety verification result of the drainage efficiency evaluation data; adjusting the channel layout scheme according to the safety verification result to obtain a final drainage path scheme of the patient's brain.

10. A multi-channel brain hemorrhage minimally invasive drainage system, characterized in that, The system includes: a multi-modal image acquisition module for acquiring multi-modal medical image data of a patient's brain to generate an image data set of the patient's brain; a structure enhancement processing module for performing structure enhancement processing on the image data set to obtain a structure-enhanced image set of the patient's brain; a spatial relationship matrix construction module for constructing a spatial relationship matrix of a hematoma region and surrounding key anatomical structures based on the structure-enhanced image set; a three-dimensional structure reconstruction module for constructing a hematoma three-dimensional structure diagram of the patient's brain according to the spatial relationship matrix; a preliminary drainage path planning module for performing spatial distribution characteristic analysis on the hematoma three-dimensional structure diagram to obtain a preliminary drainage path scheme of the patient's brain; a multi-channel collaborative optimization module for performing multi-channel collaborative arrangement optimization on the preliminary drainage path scheme to obtain a final drainage path scheme of the patient's brain.

Images