Personalized surgical pathway planning system based on digital models of brain functional areas and lesions
Patent Information
- Application Number
- CN202610106676.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-01-27
- Publication Date
- 2026-05-26
Smart Images
Figure CN122091089A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of surgical assistance technology, and more specifically, to an individualized surgical path planning system based on a digital model of brain functional areas and lesions. Background Technology
[0002] In neurosurgery, especially in the resection of brain lesions, surgical pathway planning is the cornerstone of the entire treatment strategy, playing a decisive role in the patient's postoperative functional recovery and long-term quality of life. The core of surgical pathway planning is designing a surgical channel from the skull surface to the lesion area. This channel must provide surgical instruments with a clear field of vision and ample operating space, and more importantly, it must precisely avoid key cortical areas of the brain responsible for language, motor, sensory, and memory functions, as well as the core white matter nerve tracts connecting these functional areas, while removing as much diseased tissue as possible. Any unintentional damage to these critical structures can lead to permanent neurological dysfunction in the patient.
[0003] Currently, in practical decision-making, surgeons often face several conflicting goals: pursuing larger incisions and wider access routes, while providing excellent surgical visibility and instrument manipulation space, requires cutting or retracting more normal cortical tissue, increasing the risk of postoperative complications and neurological deficits; choosing the shortest geometric path can shorten surgical time and reduce disturbance to tissues along the way, but the path may need to pass through important functional areas or core blood vessels, leading to disastrous consequences; avoiding all known important functional areas ensures high safety, but may result in an abnormally tortuous and deep path, greatly increasing the difficulty and time of the operation. Balancing these factors relies entirely on the surgeon's personal experience and subjective judgment, without a unified, reproducible, quantitative evaluation standard. Therefore, how to provide a more objective and quantifiable system to assist surgeons in planning surgical pathways is a current technical challenge in the field of surgical assistance technology. Summary of the Invention
[0004] In view of the aforementioned existing problems, the present invention is proposed.
[0005] Therefore, this invention provides an individualized surgical path planning system based on digital models of brain functional areas and lesions, which solves the technical problem that existing surgical path planning processes rely on subjective experience and lack quantitative evaluation, leading to suboptimal path solutions.
[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:
[0007] This invention provides a personalized surgical pathway planning system based on a digital model of brain functional areas and lesions, consisting of a set of highly coordinated functional modules, specifically including:
[0008] Image analysis module: Receives and analyzes imported patient brain tissue image data;
[0009] Feature map generation module: Receives the analysis results from the image analysis module and generates an individualized brain tissue feature map accordingly;
[0010] Model instantiation module: receives the individualized brain tissue feature map generated by the feature map generation module and obtains brain tissue model data, and assigns the individualized brain tissue feature map to the brain tissue model;
[0011] Functional risk assessment module: Receives the brain tissue model from the model instantiation module, and with reference to an external brain function database, divides the brain tissue model into functional areas and assigns corresponding functional impairment risk indices to complete the construction of an individualized brain tissue model;
[0012] Path planning module: Receives the individualized brain tissue model output by the functional risk assessment module, and generates a set of candidate surgical paths on the model.
[0013] As a preferred embodiment of the individualized surgical path planning system based on digital models of brain functional areas and lesions described in this invention, the system further includes the following modules:
[0014] Human-computer interaction module: used to receive clinical preference weight parameters from the surgeon and pass the parameters to the path suggestion generation module;
[0015] The path suggestion generation module receives the set of candidate surgical paths generated by the path planning module, and performs quantitative evaluation, screening and weighted decision-making on the accessibility of the surgical approach, the functional retention rate and the instrument operation space, and finally outputs the optimal surgical path scheme suggestion.
[0016] Visualization module: Used to receive the optimal surgical path suggestion output by the path suggestion generation module and present it to the surgeon.
[0017] Verification and recording module: Receives intraoperative monitoring data, standardizes it, compares and verifies it against the individualized brain tissue model constructed preoperatively, and records the deviation data of functional area division;
[0018] As a preferred embodiment of the individualized surgical path planning system based on digital models of brain functional areas and lesions described in this invention, the specific process of the image analysis module analyzing brain tissue image data includes:
[0019] Imaging feature parameters of each region are extracted from the brain tissue imaging data, and the imaging feature parameters are compared with standard brain tissue imaging feature parameters stored in a preset image analysis database to determine the tissue affiliation of each region in the brain tissue imaging data. Regions in the imaging features that are significantly different from the standard brain tissue imaging features are marked as potential lesion areas.
[0020] As a preferred embodiment of the individualized surgical path planning system based on digital models of brain functional areas and lesions described in this invention, the specific process by which the model instantiation module assigns the individualized brain tissue feature map includes:
[0021] The imaging feature parameters in the individualized brain tissue feature atlas are used as physical attributes to the brain tissue model. Then, the tissue distribution information is mapped to the corresponding position of the brain tissue model to divide the region. At the same time, the adjacency relationship between each region is established to define the physical integrity of the brain tissue model.
[0022] As a preferred embodiment of the individualized surgical pathway planning system based on a digital model of brain functional areas and lesions as described in this invention, the specific process of the functional risk assessment module allocating the functional impairment risk index includes:
[0023] The brain tissue model is aligned with a standard brain functional zoning map in an external brain function database to classify the functional areas of each region in the brain tissue model. Then, based on the functional area classification, the corresponding risk level is queried from the brain function database, and the risk level is assigned to the corresponding region of the brain tissue model to form a regional functional impairment risk index.
[0024] As a preferred embodiment of the individualized surgical path planning system based on digital models of brain functional areas and lesions described in this invention, the specific process by which the path planning module generates a set of candidate surgical paths includes:
[0025] The individualized brain tissue model is transformed into a decision-making environment that includes state, action, and expected mobility associated with a functional impairment risk index. Starting from the skull surface and ending at the lesion, and with minimizing the functional impairment risk as a constraint, the optimal path is planned through continuous simulation and exploration in the environment. Then, the planning method is applied to generate multiple paths from multiple different starting points. Finally, the generated multiple paths are sorted according to their cumulative functional impairment risk, and a specified number of paths with the lowest risk are selected to form the candidate surgical path set.
[0026] As a preferred embodiment of the individualized surgical path planning system based on digital models of brain functional areas and lesions described in this invention, the specific process of quantitative evaluation and screening by the path suggestion generation module includes:
[0027] For all paths in the candidate surgical path set, a weighted calculation is performed based on the quantitative evaluation results of surgical approach accessibility, functional preservation rate, and instrument operation space, combined with the clinical preference weights; the path with the best evaluation is determined as the optimal surgical path recommendation.
[0028] As a preferred embodiment of the individualized surgical path planning system based on digital models of brain functional areas and lesions described in this invention, the path suggestion generation module determines the quantitative evaluation results in the following specific ways:
[0029] A shorter path indicates less brain tissue damage, shorter surgical time, and a higher surgical accessibility score. A lower risk of functional impairment along the path indicates less impact on functional areas and a higher functional preservation rate score. A larger semi-apex angle of the safe operation cone indicates better instrument operability and a higher instrument operation space score. For the surgical path, the functional impairment risk of the area traversed by the path is weighted and calculated based on the path length to obtain the final path functional impairment risk. A safe operation cone is constructed with the area where the path connects to the lesion as the vertex, and the semi-apex angle of the safe operation cone is obtained.
[0030] As a preferred embodiment of the individualized surgical path planning system based on a digital model of brain functional areas and lesions as described in this invention, the specific process by which the path suggestion generation module calculates the optimal path in conjunction with the clinical preference weights includes:
[0031] The accessibility of surgical approach, functional preservation rate and instrument operation space of all paths in the candidate surgical path set are quantitatively evaluated and weighted with the clinical preference weights to obtain a final comprehensive score that reflects the overall performance of the path. Finally, the final comprehensive scores of all paths are sorted and the path with the highest score is selected. At the same time, the three paths with the highest scores in different individual directions are retained as alternative surgical path suggestions.
[0032] As a preferred embodiment of the individualized surgical path planning system based on digital models of brain functional areas and lesions described in this invention, the specific process of the verification and recording module performing comparative verification includes:
[0033] Intraoperative monitoring data containing the functional response type and its three-dimensional spatial coordinates are acquired, and the three-dimensional spatial coordinates are mapped to the individualized brain tissue model to locate specific tissue regions. The functional response type in the monitoring data is then compared with the original functional area attribution of the located tissue region in the preoperative model to identify discrepancies between the two. Finally, the accuracy of the surgical path planning is evaluated based on the proportion of the discrepancy regions.
[0034] As a preferred embodiment of the individualized surgical path planning system based on digital models of brain functional areas and lesions described in this invention, the process of analyzing brain tissue imaging data specifically includes: extracting imaging feature parameters of each region from the brain tissue imaging data, comparing the imaging feature parameters with standard brain tissue imaging feature parameters stored in a preset image analysis database, determining the tissue affiliation of each region in the brain tissue imaging data, and marking regions in the imaging features that are significantly different from the standard brain tissue imaging features as potential lesion areas.
[0035] The beneficial effects of this invention are as follows: It constructs an individualized brain tissue model incorporating physical attributes and functional risk indices by analyzing patient imaging data; based on this model, candidate surgical pathways are generated, and these pathways undergo multi-dimensional quantitative evaluation and screening, combined with clinical preference weights to determine the optimal surgical pathway recommendation; simultaneously, by comparing and verifying intraoperative monitoring data with preoperative planning, deviation information is recorded to optimize subsequent planning. This invention provides a systematic quantitative decision-making basis for surgical pathway selection, offers balanced decision support reflecting surgical strategies, and establishes a postoperative verification and recording mechanism. Attached Figure Description
[0036] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0037] Figure 1 This is a framework diagram of the individualized surgical path planning system based on digital models of brain functional areas and lesions, as described in this invention.
[0038] Figure 2 The flowchart for generating personalized brain tissue feature maps for this invention is shown below.
[0039] Figure 3 A flowchart for generating surgical path suggestions is provided for this invention.
[0040] Figure 4 This is a flowchart of the individualized surgical path planning system based on digital models of brain functional areas and lesions, as described in this invention. Detailed Implementation
[0041] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
[0042] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.
[0043] Secondly, the term "one embodiment" or "example" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the invention. The appearance of an embodiment in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that mutually excludes other embodiments.
[0044] Example 1
[0045] Reference Figures 1-4 This is one embodiment of the present invention, which provides an individualized surgical pathway planning system based on a digital model of brain functional areas and lesions, consisting of a set of highly coordinated functional modules, specifically including:
[0046] Image analysis module: Receives and analyzes imported patient brain tissue image data;
[0047] The specific process of the image analysis module in analyzing brain tissue image data includes: extracting imaging feature parameters of each region from the brain tissue image data, comparing the imaging feature parameters with the standard brain tissue imaging feature parameters stored in the preset image analysis database, determining the tissue attribution of each region in the brain tissue image data, and marking regions with significant differences from the standard brain tissue imaging features as potential lesion areas.
[0048] Feature map generation module: Receives the analysis results from the image analysis module and generates an individualized brain tissue feature map accordingly;
[0049] The process involves using a high-field magnetic resonance imaging (MRI) device to execute a series of pre-defined scanning sequences to obtain brain tissue imaging data from patients. These sequences reflect the state of brain tissue at different physical contrast levels. To ensure data standardization and interoperability, the acquired data adheres to standard medical digital imaging and communication formats, including a two-dimensional point matrix and metadata describing slice thickness, location, orientation, and patient information. This metadata is parsed, and all two-dimensional slice images are reconstructed and aligned in a virtual three-dimensional space to form a three-dimensional data grid composed of a large number of spatial data points, which serves as the spatial basis for subsequent analysis.
[0050] The analysis utilizes a pre-defined image analysis database to analyze patient brain tissue imaging data. This database is a structured knowledge collection that stores a large number of standardized brain tissue imaging features from healthy individuals, statistically derived and annotated by experts, across different sequences. These features are organized into a multi-dimensional parameter space, including not only first-order statistical features, such as the average grayscale range and standard deviation of specific tissues (gray matter, white matter, cerebrospinal fluid), but also higher-order texture features that describe local complexity. In the analysis, each spatial data point in the three-dimensional data grid is used as the basic unit to calculate its imaging feature vector and that of its neighborhood.
[0051] The radiographic feature vector of each spatial data point is compared with the radiographic feature vectors of various standard tissues stored in the image analysis database. Distance is calculated in a multidimensional feature space composed of various feature coordinate axes (closer distances indicate higher similarity). Based on this calculation, a preliminary tissue affiliation is determined for each spatial data point. For example, if the feature vector of a spatial data point is closest to the cluster center of the standard gray matter features in the feature space, then the spatial data point is labeled as gray matter. This process covers all known normal brain tissue types.
[0052] After determining the attribution of normal brain tissue, the next step is to identify and segment the lesions. In the feature space, isolated points whose feature vectors are far from the cluster center of normal tissue, or spatial data points whose feature parameters differ significantly from the standard values, are identified and marked as potential lesion points. Next, in three-dimensional space, the spatial data points marked as potential lesion points are connected and aggregated to integrate them into lesion entities with a clear morphology. This process continues until the vast majority of neighboring spatial data points that meet the criteria are included in the same connected region, thus forming one or more complete lesion entities. Based on this, edge refinement processing is performed to accurately delineate the boundary between the lesion entity and the surrounding normal tissue, correcting any rough or stepped boundaries that may appear after the initial segmentation. This process examines each spatial data point located on the boundary of the lesion entity and analyzes the imaging features of its immediate neighbors on the normal tissue side. If a point on the boundary has imaging features closer to its external normal tissue neighbors but significantly different from its internal lesion neighbors, its attribution is corrected from lesion to normal tissue. Conversely, if the features of a normal tissue point outside the boundary are more similar to those of an internal lesion, it will be reclassified as a lesion. This fine-tuning process based on local neighborhood features continues until the entire lesion boundary reaches a stable and clearly defined state. All the analysis and processing results are then integrated into a structured dataset, namely, a personalized brain tissue feature atlas.
[0053] Model instantiation module: receives the individualized brain tissue feature map generated by the feature map generation module and obtains brain tissue model data, and assigns the individualized brain tissue feature map to the brain tissue model;
[0054] The specific process of assigning individualized brain tissue feature maps to the model instantiation module includes: assigning the imaging feature parameters in the individualized brain tissue feature maps as physical attributes to the brain tissue model; then mapping the tissue distribution information to the corresponding positions of the brain tissue model to divide the regions; and at the same time establishing the adjacency relationship between each region to define the physical integrity of the brain tissue model.
[0055] Functional risk assessment module: Receives the brain tissue model from the model instantiation module, and with reference to an external brain function database, divides the brain tissue model into functional areas and assigns corresponding functional impairment risk indices to complete the construction of an individualized brain tissue model;
[0056] The specific process of assigning functional impairment risk index in the functional risk calibration module includes: aligning the brain tissue model with the standard brain functional zoning map in an external brain function database to classify the functional areas of each region in the brain tissue model; then querying the corresponding risk level from the brain function database according to the functional area classification, and assigning the risk level to the corresponding region of the brain tissue model to form the regional functional impairment risk index.
[0057] The acquisition of brain tissue model data relies on modern medical imaging equipment (MRI scanner). While generating raw image data, its accompanying post-processing software automatically completes the brain tissue model and outputs the brain tissue model data. This brain tissue model is a digital shell that only contains the external outline of the brain tissue and surface information of the main internal cavities (such as the ventricles), without involving internal tissue details and physical properties.
[0058] A spatial correspondence was established between the 3D data grid of the individualized brain tissue feature atlas and the surface grid of the brain tissue model, ensuring that each internal data point in the atlas could find its unique spatial location within the geometric model. Based on this correspondence, the imaging feature parameters associated with each spatial data point in the individualized brain tissue feature atlas were added as physical attributes to the brain tissue model. Then, the tissue distribution information in the individualized brain tissue feature atlas was precisely mapped to the corresponding geometric coordinates of the initial brain tissue model, clearly defining the tissue affiliation of each spatial location in the brain tissue model. This, in turn, completed the regional division of lesion areas and normal tissue areas within the macroscopic geometric contour.
[0059] To ensure that the brain tissue model reflects the physical characteristics of real brain tissue as a continuous whole, it is necessary to construct the adjacency relationships between the divided tissue regions. Specifically, by analyzing the boundaries of each divided region in three-dimensional space, the physical contact and connection between them are clarified and recorded. By defining these adjacency relationships, the entire brain tissue model is given physical integrity, avoiding becoming merely a collection of independent digital fragments.
[0060] Next, to obtain the functional impairment risk index, a pre-defined external brain function database needs to be accessed. This database stores recognized standard brain function partition maps, such as the Broadman Partition Map or more modern functional connectivity maps, where each standard partition is associated with one or more known human cognitive or motor functions.
[0061] Multiple anatomically identical landmark structures were identified on both the brain tissue model and a standard atlas in a brain function database. Based on these landmark structures, a spatial transformation function was calculated to map points on the patient's brain tissue model to their corresponding points on the standard brain function atlas. This transformation function was then applied to the entire brain tissue model to spatially align it with the standard brain function atlas. The spatial transformation function ensured that while aligning the landmark structures, it also created smooth and biomechanically consistent distortions and deformations between the landmarks. Furthermore, the function could identify and match anatomically homologous structures between the two models, and based on these matching points, the patient's brain tissue model was subjected to virtual nonlinear stretching or compression to achieve optimal topological correspondence with the standard brain function atlas. After alignment, the functional areas corresponding to the anatomical regions in the patient's brain tissue model could be assigned to the standard brain function atlas. For brain tissue regions around lesions that were significantly affected by compression or deformation, an inference method based on relative topological relationships was used to determine their functional area assignments. By referencing the functional zone division information of areas far from the lesion and not significantly affected, and utilizing the inherent and stable arrangement and neighborhood relationships between functional zones, the true functional zone affiliation of the area squeezed by the lesion can be inferred in reverse.
[0062] Subsequently, based on the functional area attribution, a query is performed in the brain function database to extract the risk level associated with that functional area, pre-defined by neurosurgeons. These risk levels are typically qualitative, such as high risk, moderate risk, and low risk. Specifically, for those core areas defined by the brain function database as directly maintaining vital signs, such as the respiratory and cardiac centers in the brainstem, the highest level of risk is assigned, and they are directly marked as forbidden zones. When dealing with the very few specific areas where their anatomical structure deviates significantly from standard brain function maps due to individual differences or lesion influence, functionally undetermined areas whose functional attribution cannot be clearly defined are not simply considered safe zones, but are assigned a higher initial risk index. The purpose of this is to prioritize the safety of these small, uncertain areas in the absence of conclusive evidence, thereby providing a more cautious and comprehensive decision-making basis for subsequent pathway planning. For other non-forbidden zones, the qualitative risk level is quantified into a specific numerical value, namely, a functional impairment risk index.
[0063] Through the above steps, the simple geometric numerical shell is successfully transformed into an individualized brain tissue model containing personalized anatomical structure, tissue distribution, physical information, and functional impairment risk index.
[0064] Path planning module: Receives the individualized brain tissue model output by the functional risk assessment module, and generates a set of candidate surgical paths on the model.
[0065] The specific process of generating a candidate surgical path set by the path planning module includes: converting the individualized brain tissue model into a decision-making environment that includes state, action, and expected movement value associated with the functional impairment risk index; planning the optimal path by continuously simulating and exploring in the environment, starting from the skull surface and ending at the lesion, and under the constraint of minimizing the functional impairment risk; then applying the planning method to generate multiple paths from multiple different starting points; finally, sorting the generated multiple paths according to their cumulative functional impairment risk, selecting a specified number of paths with the lowest risk, and forming a candidate surgical path set.
[0066] Human-computer interaction module: used to receive clinical preference weight parameters from the surgeon and pass the parameters to the path suggestion generation module;
[0067] The path suggestion generation module receives the set of candidate surgical paths generated by the path planning module, and performs quantitative evaluation, screening and weighted decision-making on the accessibility of the surgical approach, the functional retention rate and the instrument operation space, and finally outputs the optimal surgical path scheme suggestion.
[0068] The specific process of quantitative evaluation and screening in the path suggestion generation module includes: for all paths in the candidate surgical path set, a weighted calculation is performed based on the quantitative evaluation results of surgical approach accessibility, functional preservation rate and instrument operation space, combined with clinical preference weights; the best-evaluated path is determined as the optimal surgical path suggestion.
[0069] The path suggestion generation module determines the specific methods for quantitative assessment results, including: shorter paths represent less damaged brain tissue, shorter surgical time, and higher surgical accessibility scores; lower path functional impairment risk represents less impact on functional areas and higher functional preservation scores; a larger semi-apex angle of the safe operation cone represents better instrument operability and higher instrument operation space scores; for surgical paths, the functional impairment risk of the areas traversed by the path is weighted and calculated based on the path length to obtain the final path functional impairment risk; a safe operation cone is constructed with the area where the path connects to the lesion as the vertex, and the semi-apex angle of the safe operation cone is obtained.
[0070] The specific process of the path suggestion generation module in calculating the optimal path in combination with clinical preference weights includes: quantifying the accessibility of surgical approaches, functional preservation rate, and instrument operation space of all paths in the candidate surgical path set, and weighting them with clinical preference weights to obtain a final comprehensive score that reflects the overall performance of the path; finally, sorting the final comprehensive scores of all paths and selecting the path with the highest score; and retaining the three paths with the highest scores in different individual directions as candidate surgical path suggestions.
[0071] Visualization module: Used to receive the optimal surgical path suggestion output by the path suggestion generation module and present it to the surgeon.
[0072] Based on a personalized brain tissue model, with minimizing the risk of functional area damage as the primary constraint, the lesion location is first determined, and the nearest opening location is calculated. A circular region centered on the opening location is delineated as a candidate opening area. Feasible paths are explored by constructing a sequence of potential paths from the skull surface to the candidate opening area, and a set of candidate surgical paths is obtained. This process transforms the personalized brain tissue model into a decision-making environment that includes state, action, and expected movement value associated with a functional impairment risk index. Movement costs are set for the virtual instrument based on the functional impairment risk index, with movement costs in restricted areas being significantly higher than others. Through continuous simulation exploration, multiple feasible paths with lower total movement costs are identified in space, and a specified number of paths with the lowest risk are selected.
[0073] Subsequently, for each path in the candidate surgical approach set, a quantitative evaluation of surgical accessibility, functional preservation rate, and instrument operation space was performed. The calculation method is as follows:
[0074] Surgical accessibility assessment By the total length of the current path The longest path length in the candidate surgical path set Minimum path length The result is obtained through calculation and falls within the range of 0 to 1 and varies with... Incremental. Surgical accessibility assessment The calculation formula is as follows:
[0075]
[0076] Functionality retention rate assessment By assessing the functional impairment risk value of the current path and the highest risk of functional impairment in the candidate surgical pathway set Minimum risk value for functional impairment The result is obtained through calculation and falls within the range of 0 to 1 and varies with... Increasing. The functional impairment risk value of the current path. It is calculated by combining the weights derived from the length of the current path in each functional zone with the functional impairment risk index of each functional zone. Functional retention rate assessment The calculation formula is as follows:
[0077]
[0078] Instrument operating space assessment This is achieved by constructing a safe operating cone at the end of the current path that does not collide with surrounding tissues, and then determining and utilizing its maximum safe semi-apex angle. The calculated result is within the interval of 0 to 1 and varies with... Incremental. Instrument operating space assessment The calculation formula is as follows:
[0079]
[0080] The above calculations normalized the three assessments. The numerical ranges or magnitudes of the three original assessment indicators may differ significantly. Normalization maps the scores of all indicators to the same numerical range, ensuring the balance of the weighted calculation. Without normalization, indicators with larger numerical ranges would dominate the weighted calculation, leading to an imbalance in weight allocation.
[0081] Finally, based on the clinical preference weights specified by the surgeon, the paths in the candidate path set are weighted and scored, and the path with the best overall performance is selected as the optimal surgical path recommendation. Additionally, the three paths with the highest individual quantitative assessment scores can be used as alternative surgical path recommendations for the surgeon's reference and comparison. The final comprehensive score for a path is obtained by multiplying the scores of the three assessment indicators of the path by their corresponding clinical preference weights, and then summing the three products. The calculation follows the formula:
[0082]
[0083] in, The final overall score representing path P; The clinical preference weights representing surgical accessibility, functional preservation rate, and instrument operation space are respectively summed to 1. This ensures that the final comprehensive score remains within a standardized and comparable numerical range, while the value of each weight can intuitively reflect its importance.
[0084] Verification and recording module: Receives intraoperative monitoring data, standardizes it, compares and verifies it against the individualized brain tissue model constructed preoperatively, and records the deviation data of functional area division;
[0085] The specific process of the verification recording module for comparison and verification includes: acquiring intraoperative monitoring data containing the functional response type and its three-dimensional spatial coordinates, and mapping the three-dimensional spatial coordinates to an individualized brain tissue model to locate specific tissue regions; then comparing the functional response type in the monitoring data with the original functional area attribution of the located tissue region in the preoperative model to identify discrepancies between the two; finally, evaluating the accuracy of the surgical path planning based on the proportion of the discrepancy regions.
[0086] The focus of validation is on the accuracy of functional area delineation because this delineation largely relies on aligning individualized models with standard brain tissue atlases, a process inherently susceptible to biases due to individual differences or lesion influence. In contrast, the model's geometry and physical properties are constructed based on the patient's precise imaging data, resulting in relatively high accuracy. Therefore, using intraoperative electrophysiological monitoring data to directly validate and correct functional area delineation is the most effective method to improve the overall accuracy of pathway planning.
[0087] Using a standardized data interface, data is continuously received from intraoperative electrophysiological monitoring equipment. This data includes physical location and a clearly defined functional response type. The data format is standardized during reception for ease of subsequent use. After reception, all recorded intraoperative monitoring data is traversed. For each data point, the specific stimulated tissue region is precisely located within the preoperatively constructed individualized brain tissue model using its three-dimensional spatial coordinates. Then, the actual functional response type recorded in this monitoring data point is directly and logically compared with the functional area assigned to that location in the preoperative model. Discrepancies are identified and recorded to optimize subsequent surgical path planning.
[0088] Finally, the ratio of the total number of areas effectively detected by intraoperative electrophysiological monitoring equipment to the number of discrepancies between the preoperative functional area division and the actual intraoperative monitoring results is used to provide an objective evaluation basis for the quality of this surgical path planning.
[0089] 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, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.
Claims
1. A personalized surgical pathway planning system based on a digital model of brain functional areas and lesions, characterized in that, It consists of a set of highly collaborative functional modules, specifically including: Image analysis module: Receives and analyzes imported patient brain tissue image data; Feature map generation module: Receives the analysis results from the image analysis module and generates an individualized brain tissue feature map accordingly; Model instantiation module: receives the individualized brain tissue feature map generated by the feature map generation module and obtains brain tissue model data, and assigns the individualized brain tissue feature map to the brain tissue model; Functional risk assessment module: Receives the brain tissue model from the model instantiation module, and with reference to an external brain function database, divides the brain tissue model into functional areas and assigns corresponding functional impairment risk indices to complete the construction of an individualized brain tissue model; Path planning module: Receives the individualized brain tissue model output by the functional risk assessment module, and generates a set of candidate surgical paths on the model.
2. The individualized surgical path planning system based on a digital model of brain functional areas and lesions according to claim 1, characterized in that, The system also includes the following modules: Human-computer interaction module: used to receive clinical preference weight parameters from the surgeon and pass the parameters to the path suggestion generation module; The path suggestion generation module receives the set of candidate surgical paths generated by the path planning module, and performs quantitative evaluation, screening and weighted decision-making on the accessibility of the surgical approach, the functional retention rate and the instrument operation space, and finally outputs the optimal surgical path scheme suggestion. Visualization module: Used to receive the optimal surgical path suggestion output by the path suggestion generation module and present it to the surgeon; Verification and recording module: Receives intraoperative monitoring data, standardizes it, compares and verifies it against the individualized brain tissue model constructed before surgery, and records the deviation data of functional area division.
3. The individualized surgical path planning system based on a digital model of brain functional areas and lesions according to claim 1, characterized in that, The specific process by which the image analysis module analyzes brain tissue image data includes: Imaging feature parameters of each region are extracted from the brain tissue imaging data, and the imaging feature parameters are compared with standard brain tissue imaging feature parameters stored in a preset image analysis database to determine the tissue affiliation of each region in the brain tissue imaging data. Regions in the imaging features that are significantly different from the standard brain tissue imaging features are marked as potential lesion areas.
4. The individualized surgical path planning system based on a digital model of brain functional areas and lesions according to claim 1, characterized in that, The specific process by which the model instantiation module assigns the individualized brain tissue feature map includes: The imaging feature parameters in the individualized brain tissue feature atlas are used as physical attributes to the brain tissue model. Then, the tissue distribution information is mapped to the corresponding position of the brain tissue model to divide the region. At the same time, the adjacency relationship between each region is established to define the physical integrity of the brain tissue model.
5. The individualized surgical path planning system based on a digital model of brain functional areas and lesions according to claim 1, characterized in that, The specific process by which the functional risk assessment module assigns the functional impairment risk index includes: The brain tissue model is aligned with a standard brain functional zoning map in an external brain function database to classify the functional areas of each region in the brain tissue model. Then, based on the functional area classification, the corresponding risk level is queried from the brain function database, and the risk level is assigned to the corresponding region of the brain tissue model to form a regional functional impairment risk index.
6. The individualized surgical path planning system based on a digital model of brain functional areas and lesions according to claim 1, characterized in that, The specific process by which the path planning module generates a set of candidate surgical paths includes: The individualized brain tissue model is transformed into a decision-making environment that includes state, action, and expected mobility associated with a functional impairment risk index. Starting from the skull surface and ending at the lesion, and with minimizing the functional impairment risk as a constraint, the optimal path is planned through continuous simulation and exploration in the environment. Then, the planning method is applied to generate multiple paths from multiple different starting points. Finally, the generated multiple paths are sorted according to their cumulative functional impairment risk, and a specified number of paths with the lowest risk are selected to form the candidate surgical path set.
7. The individualized surgical path planning system based on a digital model of brain functional areas and lesions according to claim 1, characterized in that, The specific process of quantitative evaluation and screening by the path suggestion generation module includes: For all paths in the candidate surgical path set, a weighted calculation is performed based on the quantitative evaluation results of surgical approach accessibility, functional preservation rate, and instrument operation space, combined with the clinical preference weights; the path with the best evaluation is determined as the optimal surgical path recommendation.
8. The individualized surgical path planning system based on a digital model of brain functional areas and lesions according to claim 7, characterized in that, The specific methods by which the path suggestion generation module determines the quantitative evaluation result include: A shorter path indicates less brain tissue damage, shorter surgical time, and a higher surgical accessibility score. A lower risk of functional impairment along the path indicates less impact on functional areas and a higher functional preservation rate score. A larger semi-apex angle of the safe operation cone indicates better instrument operability and a higher instrument operation space score. For the surgical path, the functional impairment risk of the area traversed by the path is weighted and calculated based on the path length to obtain the final path functional impairment risk. A safe operation cone is constructed with the area where the path connects to the lesion as the vertex, and the semi-apex angle of the safe operation cone is obtained.
9. The individualized surgical path planning system based on a digital model of brain functional areas and lesions according to claim 7, characterized in that, The specific process by which the path suggestion generation module calculates the optimal path in conjunction with the clinical preference weights includes: The accessibility of surgical approach, functional preservation rate and instrument operation space of all paths in the candidate surgical path set are quantitatively evaluated and weighted with the clinical preference weights to obtain a final comprehensive score that reflects the overall performance of the path. Finally, the final comprehensive scores of all paths are sorted and the path with the highest score is selected. At the same time, the three paths with the highest scores in different individual directions are retained as alternative surgical path suggestions.
10. The individualized surgical path planning system based on a digital model of brain functional areas and lesions according to claim 2, characterized in that, The specific process of comparison verification performed by the verification record module includes: Intraoperative monitoring data containing the functional response type and its three-dimensional spatial coordinates are acquired, and the three-dimensional spatial coordinates are mapped to the individualized brain tissue model to locate specific tissue regions. The functional response type in the monitoring data is then compared with the original functional area attribution of the located tissue region in the preoperative model to identify discrepancies between the two. Finally, the accuracy of the surgical path planning is evaluated based on the proportion of the discrepancy regions.