Neurosurgery brain puncture path accurate planning method and system

Abstract

The invention relates to the technical field of neurosurgery navigation, in particular to a neurosurgery brain puncture path accurate planning method and system.The neurosurgery brain puncture path accurate planning method comprises the following steps that by obtaining a preoperative brain function area image and nerve fiber bundle data, a function area range and sensitivity are extracted, a focus and cranial entry point path set is constructed, and a path set is established; the method comprises the steps of analyzing the distance between a path and a functional area and a fiber bundle included angle, calculating a functional interaction risk, dividing path sections and calculating the distances to a blood vessel, a ventricle and the functional area, generating obstacle field intensity and calculating a gradient difference, constructing a risk map, scoring candidate paths, and screening an optimal path to generate a puncture result. The method comprises the following steps: extracting a brain function region range and a fiber bundle direction, introducing a function sensitivity parameter, improving tiny damage identification, calculating obstacle field intensity in a path segmentation manner, forming a gradient, dynamically identifying a high-risk region, grading and screening candidate paths by integrating risk factors, performing function protection and risk reduction, and improving path planning refinement and safety.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of neurosurgery navigation, in particular to a neurosurgery brain puncture path accurate planning method and system. BACKGROUND

[0002] The technical field of neurosurgery navigation includes medical engineering technology for assisting neurosurgeons in positioning and path guidance during craniocerebral surgery. The core content is to obtain the structural information of the patient's craniocerebral through medical imaging, combined with spatial coordinate conversion, stereoscopic positioning and preoperative planning, to provide spatial reference for intraoperative operation. Its overall system covers medical image acquisition, three-dimensional reconstruction, spatial registration, surgical path design and navigation feedback, forming a clinical auxiliary technology framework for fine brain operation.

[0003] Among them, a neurosurgery brain puncture path accurate planning method and system refers to a technical scheme that analyzes the patient's craniocerebral image before brain puncture operation and generates a puncture path suitable for surgery. The technical matters include medical image segmentation to identify brain structures, feature point extraction to determine anatomical references, three-dimensional coordinate establishment to realize spatial positioning, and path planning calculation to generate feasible puncture trajectories. Its implementation is usually based on three-dimensional medical images to construct a patient's brain model, determine the angle and depth of the puncture path using mathematical relationships, and generate path data corresponding to the surgical positioning device through the planning system, thereby completing the accurate planning of the brain puncture path.

[0004] In the prior art, path planning mainly relies on geometric structure analysis after image reconstruction, generating paths by establishing three-dimensional models and combining mathematical formula calculations of angles and depths. This approach has strong intuitiveness in spatial positioning and trajectory generation, but has the problem of insufficient sensitivity to brain function areas and neural fiber bundles. In actual operation, path selection often pays more attention to the geometric connection between lesions and intracranial points, and the consideration of function area protection is relatively rough, which may lead to a situation where the path is feasible in space but has potential damage in function protection. At the same time, the existing technology mainly evaluates the risk of the path as a whole, lacks detailed description of the risk distribution along the path, and may encounter high-risk segments that are not pre-judged during the operation. In addition, the assessment of obstacles mainly stays in static distance calculation, and fails to reflect the potential impact of local changes. Including in some cases, the path distance from the function area boundary may seem safe, but when entering the critical section, the distribution of blood vessels or fiber bundles may increase the risk of surgery. The above shortcomings may cause a gap between the surgical planning result and the actual operation, reducing the reliability and safety of the puncture path in fine surgery. SUMMARY

[0005] In order to solve the technical problems existing in the prior art, the embodiment of the present application provides a neurosurgical brain puncture path accurate planning method, comprising the following steps: In order to achieve the above purpose, the present application adopts the following technical scheme: a neurosurgical brain puncture path accurate planning method, comprising the following steps: S1: obtaining preoperative brain function area image and neural fiber bundle data, extracting the volume of brain function area and the spatial coverage ratio of neural fiber bundle dominant direction, calculating the sensitivity of multiple brain function areas, and generating a function sensitivity parameter set; S2: obtaining the lesion and intracranial entry point based on the brain image and constructing a candidate path set, analyzing the optimal distance of the path to the brain function area boundary, if the distance threshold is exceeded, analyzing the included angle between the candidate path advancing direction and the neural fiber bundle dominant direction, and calculating the function interaction risk in combination with the function sensitivity parameter set; S3: dividing each candidate path based on the function interaction risk, calculating the optimal distance of the segmented path point to the brain blood vessel, brain ventricle and function area and converting it into obstacle field strength, and generating segmented obstacle field strength; S4: based on the segmented obstacle field strength, calculating the difference between adjacent segmented obstacle strengths as a gradient, if the gradient threshold is exceeded, converting the difference into a penalty value and marking a risk area, and constructing an obstacle gradient risk map; S5: based on the obstacle gradient risk map and function interaction risk, scoring all candidate paths, and selecting the path with the optimal score as the puncture planning path to generate a brain puncture path planning result.

[0006] As a further scheme of the present application, the function sensitivity parameter set includes volume parameters, spatial coverage ratio parameters and sensitivity coefficients, the function interaction risk specifically refers to optimal distance, included angle degree and function risk amount, the segmented obstacle field strength includes blood vessel distance strength, brain ventricle distance strength and function area distance strength, the obstacle gradient risk map specifically refers to gradient difference, penalty value and risk area coordinates, and the brain puncture path planning result specifically refers to path space coordinates, risk score and planning path sequence.

[0007] As a further scheme of the present application, the specific steps of S1 are: S101: obtaining preoperative brain function area image and neural fiber bundle data, segmenting and measuring the volume values of multiple partitions of brain function area image, while extracting the dominant direction data of neural fiber bundle, calling the partition volume values and dominant direction data to compare and calculate the spatial coverage ratio, and generating the spatial coverage ratio; S102: based on the spatial coverage ratio, combining the spatial coverage ratio of multiple brain function areas with the corresponding partition volume values, calculating the sensitivity degree of multiple brain function areas under neural fiber distribution by using proportional conversion, and obtaining the function area sensitivity coefficient. S103: According to the functional area sensitivity coefficient, the coefficients of all brain functional areas are integrated to establish a functional sensitivity parameter set. The functional sensitivity is the functional importance or vulnerability of the brain functional area under the influence of the neural fiber bundle distribution.

[0008] As a further scheme of the present application, the specific steps of S2 are: S201: Obtain brain image data, extract lesion location and intracranial entry point, connect the two as path start and end points as a candidate path set, simultaneously collect three-dimensional coordinate data of brain tissue, real-time monitor and update the positions of lesions, brain functional areas and intracranial points, and correct the path deviation caused by brain drift to generate a candidate path set; S202: Based on the candidate path set, call the brain functional area boundary coordinate data, calculate the distance from multiple candidate paths to the nearest brain functional area boundary, and compare the distance result with the set distance threshold, analyze the difference between the candidate path and the distance threshold, and obtain the boundary difference distance; S203: According to the boundary difference distance, extract the candidate path advancing direction vector and the neural fiber bundle dominant direction vector, calculate the included angle degree, and analyze the spatial risk level of the candidate path combined with the functional sensitivity parameter set to establish a functional interaction risk.

[0009] As a further scheme of the present application, the distance threshold is set by statistically analyzing the brain functional area boundary coordinates and the spatial distribution characteristics of the candidate path, centrally calculating the minimum distance value from the path to the boundary, and setting according to the average distance range between the lesions and the brain functional area boundary in the same image sample.

[0010] As a further scheme of the present application, the specific steps of S3 are: S301: Based on the functional interaction risk, each candidate path is divided into equal lengths, and for each segmented path point, the coordinate position of the path point in three-dimensional space is obtained to generate path segmentation point data; S302: Based on the path segmentation point data, real-time three-dimensional coordinate data of brain blood vessels, ventricles and brain functional areas are obtained using intraoperative real-time images, the distance from each segmented path point to the brain blood vessels, ventricles and functional areas is calculated and the optimal distance is selected to generate optimal path distance; S303: According to the optimal path distance, the shortest distance is converted into obstacle field intensity value, and the obstacle field intensity of multiple segmented path points is summarized and evaluated to generate segmented obstacle field intensity.

[0011] As a further scheme of the present application, the specific steps of S4 are: S401: sequentially extracting intensity data of adjacent segments based on the segmented obstacle field intensity, calculating the difference between adjacent segments and recording the corresponding segment index, and generating an obstacle intensity difference sequence; S402: according to the obstacle intensity difference sequence, calling a set gradient threshold, comparing the difference value by item, screening the difference value exceeding the gradient threshold and recording the corresponding spatial coordinate range, and obtaining a super-threshold gradient difference value set; S403: calling the super-threshold gradient difference value set, converting the difference value into a penalty value and superimposing it in the corresponding spatial coordinate range, continuously tracking the changes of brain tissue morphology and position through real-time image, marking the risk area, and establishing an obstacle gradient risk map.

[0012] As a further scheme of the present application, the gradient threshold is set by statistically analyzing the segmented obstacle field intensity difference value, selecting the mean and standard deviation of the difference value between adjacent segments as the measurement basis, and combining the central interval of the difference value distribution in the same sample.

[0013] As a further scheme of the present application, the specific steps of S5 are: S501: based on the obstacle gradient risk map and functional interaction risk, extracting the corresponding risk value in the spatial coordinate range passed by each path, and performing weighted cumulative calculation according to the path order, to generate a path risk total amount; S502: according to the path risk total amount, scoring all candidate paths, comparing the risk total amount value of multiple paths with a set scoring reference value, and sorting the paths according to the comparison result to obtain a candidate path scoring sequence; S503: calling the candidate path scoring sequence, screening the path with the optimal scoring result, integrating the spatial coordinate range of the path with the risk marking result, real-time monitoring the deviation of the surgical instrument and the planned path and triggering an alarm when deviating, and establishing a brain puncture path planning result; The scoring reference value is set by analyzing the risk distribution of all candidate paths.

[0014] A neurosurgical brain puncture path precise planning system, comprising: A functional sensitivity module acquires preoperative brain functional area image and neural fiber bundle data, extracts the volume of brain functional area and the spatial coverage ratio of neural fiber bundle dominant direction, calculates the sensitivity of multiple brain functional areas, generates a functional sensitivity parameter set and transmits it to the path analysis module; A path analysis module acquires lesions and intracranial entry points based on brain images and constructs a candidate path set, analyzes the optimal distance of the path to the brain functional area boundary, if the distance exceeds the threshold, analyzes the angle between the candidate path advancing direction and the neural fiber bundle dominant direction, and combines the functional sensitivity parameter set to calculate the functional interaction risk and transmit it to the path segmentation module. a path segmenting module, dividing each candidate path based on the functional interaction risk, calculating the optimal distance from the segmented path point to the cerebral blood vessels, cerebral ventricles and functional areas and converting it into obstacle field strength, generating segmented obstacle field strength and transferring it to the risk gradient module; a risk gradient module, calculating the difference between adjacent segmented obstacle strength as a gradient based on the segmented obstacle field strength, converting the difference into a penalty value and marking the risk area if the gradient threshold is exceeded, constructing an obstacle gradient risk map and transferring it to the path scoring module; a path scoring module, scoring all candidate paths based on the obstacle gradient risk map and functional interaction risk, and screening the optimal path as the puncture planning path to generate the brain puncture path planning result.

[0015] Compared with the prior art, the advantages and positive effects of the present application are that: In the present application, by extracting the spatial range and volume of the brain functional area and combining the running direction of the neural fiber bundle, the functional sensitivity parameter can be introduced in the path generation stage, so that the candidate path has the consideration of protecting the functional area at an early stage. The analysis of the relationship between the path and the functional area and the neural fiber bundle is no longer limited to simple geometric distance, but combines the calculation of angle and functional interaction risk to improve the recognition ability of the possibility of minor damage. Further, the path is divided into different paragraphs, and the optimal distance from each paragraph to the blood vessels, cerebral ventricles and functional area boundary is calculated, converted into obstacle field strength, and then the gradient is formed by the difference between the segmented strength to dynamically identify the local risk of the path. This risk gradient-based marking method can reveal the high-risk areas hidden in the path, so that the path optimization not only stays in the evaluation of the overall distance and angle, but also includes the continuity analysis of the risk distribution. The candidate path considers the functional interaction risk and the obstacle gradient risk map in the final screening, which can make more balanced path scoring, so that the output result is more refined and operable in avoiding the functional area, protecting the fiber bundle and reducing the puncture risk. This overall logic improves the path planning from single geometric matching to a multi-dimensional functional and anatomical protection system, which can significantly enhance the safety and practical value of the path in clinical application. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0017] Figure 1 The step flowchart of the present application; Figure 2S1 is a detailed schematic diagram of the present application; Figure 3 S2 is a detailed schematic diagram of the present application; Figure 4 S3 is a detailed schematic diagram of the present application; Figure 5 S4 is a detailed schematic diagram of the present application; Figure 6 S5 is a detailed schematic diagram of the present application; Figure 7 S6 is a system module diagram of the present application. DETAILED DESCRIPTION

[0018] The technical solutions in the present application will be described below with reference to the drawings.

[0019] In the embodiments of the present application, the words such as "for example", "including", etc. are used to mean as an example, illustration or description. Any embodiment or design scheme described as "for example" in the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. Rather, the word "for example" is intended to present the concept in a specific manner. In addition, in the embodiments of the present application, the meaning expressed by "and / or" can be both, or can be one of the two.

[0020] In the embodiments of the present application, "image" and "picture" can be used interchangeably at times. It should be pointed out that when the distinction is not emphasized, the meanings expressed are consistent. The words "of", "corresponding" and "corresponding" can be used interchangeably at times. It should be pointed out that when the distinction is not emphasized, the meanings expressed are consistent.

[0021] In the embodiments of the present application, sometimes the subscript such as W1 can be written in the form of non-subscript such as W1. When the distinction is not emphasized, the meanings expressed are consistent.

[0022] In order to make the technical problems, technical solutions and advantages of the present application more clear, the following will be described in detail with reference to the drawings and specific embodiments.

[0023] Please refer to Figure 1 The embodiment of the present application provides a neurosurgical brain puncture path accurate planning method, which comprises the following steps: S1: acquiring preoperative brain function area image and neural fiber bundle data, extracting the volume of the brain function area and the spatial coverage ratio of the neural fiber bundle dominant direction, calculating the sensitivity of multiple brain function areas, and generating a function sensitivity parameter set; S2: obtaining the lesion and the intracranial entry point based on the brain image, constructing a candidate path set, analyzing the optimal distance of the path to the brain function area boundary, if the distance exceeds the threshold, analyzing the angle between the candidate path advancing direction and the dominant direction of the neural fiber bundle, and calculating the functional interaction risk combined with the functional sensitivity parameter set; S3: dividing each candidate path based on the functional interaction risk, calculating the optimal distance of the segmented path point to the brain blood vessels, brain ventricles and functional areas and converting it into obstacle field strength, generating segmented obstacle field strength; S4: based on the segmented obstacle field strength, calculating the difference between adjacent segmented obstacle strengths as a gradient, if the gradient exceeds the threshold, converting the difference into a penalty value and marking the risk area, constructing an obstacle gradient risk map; S5: based on the obstacle gradient risk map and the functional interaction risk, scoring all candidate paths, and selecting the optimal path as the puncture planning path, generating the brain puncture path planning result.

[0024] The functional sensitivity parameter set includes volume parameters, spatial coverage ratio parameters, sensitivity coefficients, the functional interaction risk is specifically the optimal distance, the angle degree, and the functional risk amount, the segmented obstacle field strength includes the blood vessel distance strength, the brain ventricle distance strength, and the functional area distance strength, the obstacle gradient risk map specifically refers to the gradient difference, the penalty value, and the risk area coordinates, and the brain puncture path planning result specifically refers to the path space coordinates, the risk score, and the planning path sequence.

[0025] Please refer to Figure 2 , the specific steps of S1 are: S101: obtaining preoperative brain function area image and neural fiber bundle data, segmenting and measuring the volume value of multiple partitions of the brain function area image, extracting the dominant direction data of the neural fiber bundle, calling the partition volume value and the dominant direction data to compare and calculate the spatial coverage ratio, and generating the spatial coverage ratio; Based on the acquisition of preoperative brain function area image and neural fiber bundle data, the brain T1 weighted image of the patient is collected by the magnetic resonance imaging device, the spatial resolution is set to 1 millimeter voxel, and the data is stored in DICOM format. The image processing software is used to import the data, and the brain function area is manually segmented. Referring to the standard brain atlas, Broca area, Wernicke area, Motor area and Sensory area are identified. The number of voxels in each partition is counted, the voxel volume is set to 1 cubic millimeter, and the partition volume value is calculated. The Broca area voxel number is 15000, the volume is 15 cubic centimeters, the Wernicke area voxel number is 20000, the volume is 20 cubic centimeters, the Motor area voxel number is 25000, the volume is 25 cubic centimeters, and the Sensory area voxel number is 18000, the volume is 18 cubic centimeters. At the same time, the DTI data is collected by the diffusion tensor imaging device, the b value is set to 1000 seconds per square millimeter, the direction number is 30, the fiber tracking software is used to process the data, the main neural fiber bundle such as arcuate bundle is extracted, and the dominant direction data is calculated. The average fiber direction vector is obtained, including the Broca area dominant direction vector (0.8, 0.2, 0.1), the Wernicke area (0.7, 0.3, 0.2), the Motor area (0.9, 0.1, 0.1), and the Sensory area (0.6, 0.4, 0.1). The partition volume value and the dominant direction data are called to calculate the spatial coverage ratio. The coverage ratio is defined as the proportion of voxels in the partition and the proportion of voxels aligned with the dominant direction of the neural fiber bundle. The alignment judgment standard is that the cosine value of the voxel direction vector and the dominant direction vector is greater than 0.866 (corresponding to the included angle is less than 30 degrees), and then the coverage voxel is calculated. Including the Broca area total voxel number 15000, the coverage voxel number 10500, the coverage ratio calculation is 10500 divided by 15000 multiplied by 100 equal to 70 percent, the Wernicke area total voxel number 20000, the coverage voxel number 13000, the coverage ratio 65 percent, the Motor area total voxel number 25000, the coverage voxel number 20000, the coverage ratio 80 percent, the Sensory area total voxel number 18000, the coverage voxel number 10800, the coverage ratio 60 percent, and the spatial coverage ratio value of each partition is generated.

[0026] Table 1: Brain function area volume and spatial coverage ratio example table Functional area Volume (cm3) Spatial coverage ratio (%) Broca 15.0 70.0 Wernicke 20.0 65.0 Motor 25.0 80.0 Sensory 18.0 60.0 As shown in Table 1, the table lists the volume and spatial coverage ratio data of four brain function areas for subsequent calculation to generate the spatial coverage ratio.

[0027] S102: Based on the spatial coverage ratio, the spatial coverage ratio of multiple brain function areas is combined with the corresponding partition volume value, the sensitivity of the multiple brain function areas under the neural fiber distribution is calculated by using the proportional conversion, and the function area sensitivity coefficient is obtained. Based on the spatial coverage ratio, the spatial coverage ratio values of multiple brain function areas are combined with the corresponding partition volume values 15 cubic centimeters, 20 cubic centimeters, 25 cubic centimeters, and 18 cubic centimeters, and the sensitivity of multiple brain function areas under the distribution of nerve fibers is calculated by using proportional conversion. The sensitivity is defined as the ratio of the spatial coverage ratio to the partition volume value, and the coverage intensity per unit volume is obtained as the sensitivity coefficient, including the Broca area sensitivity coefficient calculated as 70 divided by 15 equal to 4.67, the Wernicke area as 65 divided by 20 equal to 3.25, the Motor area as 80 divided by 25 equal to 3.20, and the Sensory area as 60 divided by 18 equal to 3.33. The sensitivity coefficient determination interval is set as follows: a coefficient greater than 5.0 is high sensitivity, indicating that the distribution of nerve fibers is dense, a coefficient between 3.0 and 5.0 is medium sensitivity, indicating that the distribution is moderate, and a coefficient less than 3.0 is low sensitivity, indicating that the distribution is sparse. The Broca area coefficient 4.67 belongs to medium sensitivity, the Wernicke area coefficient 3.25 belongs to medium sensitivity, the Motor area coefficient 3.20 belongs to medium sensitivity, and the Sensory area coefficient 3.33 belongs to medium sensitivity. The sensitivity coefficient value of each function area is obtained.

[0028] S103: According to the function area sensitivity coefficient, the coefficients of all brain function areas are integrated to establish a function sensitivity parameter set; The function sensitivity is the functional importance or vulnerability of the brain function area under the influence of the distribution of nerve fiber bundles; According to the function area sensitivity coefficient, including Broca area 4.67, Wernicke area 3.25, Motor area 3.20, and Sensory area 3.33, the coefficients of all brain function areas are integrated by calculating the average sensitivity coefficient as the integration method. The sum of all coefficient values 4.67 plus 3.25 plus 3.20 plus 3.33 is equal to 14.45, and the average sensitivity coefficient 3.6125 is obtained by dividing the number of function areas 4. The function sensitivity parameter set is established, which includes the set of all sensitivity coefficients and average values, including the parameter set [Broca: 4.67, Wernicke: 3.25, Motor: 3.20, Sensory: 3.33, Average: 3.61].

[0029] Please refer to Figure 3 , the specific steps of S2 are as follows: S201: Obtain brain image data, extract lesion position and intracranial entry point, connect the two as the starting point and ending point of the path as the candidate path set, simultaneously collect three-dimensional coordinate data of brain tissue, real-time monitor and update the position of lesion, brain function area and intracranial point, and correct the path deviation caused by brain drift, generate candidate path set; Based on the acquisition of brain image data, the patient's brain T1 weighted sequence image is collected by a magnetic resonance imaging scanner, the spatial resolution is set to 1 millimeter isotropic, stored in NIfTI format, imported into the data using medical image processing software, the lesion area boundary is manually outlined, the lesion position coordinates are obtained by calculating the area centroid, including the lesion center point coordinates (10.0, 20.0, 30.0) millimeters, the intracranial entry point is preset by a neurosurgeon based on preoperative planning, usually selecting the frontal bone window position, including the entry point coordinates (5.0, 5.0, 5.0) millimeters, the lesion position is taken as the path endpoint, the intracranial entry is taken as the path starting point, the two points are connected using the path generation method, a plurality of candidate paths are generated to increase diversity, including path 1 using straight line connection, calculating the direction vector (5, 15, 25) from the starting point (5, 5, 5) to the end point (10, 20, 30), after normalization, it is (0.1826, 0.5477, 0.9129), the path points are sampled at 1 millimeter step, the starting point (5, 5, 5), the next point (5+0.1826, 5+0.5477, 5+0.9129)≈(5.1826, 5.5477, 5.9129), and so on until the end point, the path point sequence is [(5, 5, 5), (5.1826, 5.5477, 5.9129), …, (10, 20, 30)], path 2 uses curve connection to avoid known blood vessel regions, using Bezier curve fitting, control points include (5, 5, 5), (7, 10, 15), (10, 20, 30), generating point sequence [(5, 5, 5), (5.5, 6.5, 7.5), …, (10, 20, 30)], to overcome the precision decline problem caused by the above preoperative static data planning, a key intraoperative dynamic monitoring and correction mechanism is further introduced to cope with the "brain shift" phenomenon. The core is to use advanced intraoperative real-time imaging technology, such as intraoperative magnetic resonance imaging (iMRI) or intraoperative ultrasound (iUS), to periodically collect three-dimensional image data of the current brain tissue at key stages of the operation. The system performs high-precision non-rigid image registration on the real-time image and the preoperative reference image, calculates a displacement vector field that quantifies and describes the three-dimensional spatial displacement and deformation of the brain tissue. The displacement field is applied to all pre-defined key coordinate points, and the lesion centroid, intracranial point and boundary coordinates of multiple important brain function areas are updated from the static position before the operation to the actual position during the operation in real time and dynamically. Based on these new starting points and new endpoints after dynamic correction, the path planning algorithm is called again to iteratively optimize and recalculate all candidate paths. Finally, the continuous coordinate points of each candidate path are summarized and stored in the form of a three-dimensional array.For example, after modification, the point set of path 1 can contain 50 points, whose coordinate sequence is updated as (5.0, 5.0, 5.0), (5.2, 5.5, 5.8), …, (10.0, 20.0, 30.0); the point set of path 2 can contain 60 points, whose coordinate sequence is updated as (5.0, 5.0, 5.0), (5.3, 6.0, 6.5), …, (10.0, 20.0, 30.0). Thus, a candidate path set capable of reflecting the real state of brain tissue in real time is generated, which contains the point set data of multiple paths.

[0030] S202: Based on the candidate path set, the brain function area boundary coordinate data is called, the distance from the multiple candidate paths to the nearest brain function area boundary is calculated, the distance result is compared with the set distance threshold, the difference between the candidate path and the distance threshold is analyzed, and the boundary difference distance is obtained; Based on the candidate path set, including the point set (5.0, 5.0, 5.0), (5.2, 5.5, 5.8), …, (10.0, 20.0, 30.0) of path 1, the brain function area boundary coordinate data is called, the Broca area boundary point list (15.0, 20.0, 25.0), (16.0, 21.0, 26.0), …, (20.0, 25.0, 30.0) and the Wernicke area boundary point list (25.0, 30.0, 35.0), …, (30.0, 35.0, 40.0) are obtained from the foregoing, the distance from each point of each candidate path to the nearest brain function area boundary is calculated, the Euclidean distance formula is used, for the path point (12.0, 18.0, 25.0), the distance to all boundary points is calculated, including the distance to the Broca area point (15.0, 20.0, 30.0) mm; to the Wernicke area point (25.0, 30.0, 35.0), the Euclidean distance is calculated and compared with the distance threshold, mm, the minimum value 6.164 mm is taken as the nearest distance, the distance threshold is set based on the clinical safety standard, including the threshold 2.0 mm, which represents the minimum safety distance allowed, the distance result is compared with the threshold, for the point (12.0, 18.0, 25.0), the distance 6.164 mm is greater than the threshold 2.0 mm, the difference is the distance minus the threshold, 6.164-2.0=4.164 mm, the difference between the candidate path and the threshold is analyzed, the average difference of all points on the path is taken, including 50 points of path 1, the sum of the multiple point differences is divided by 50, the average difference is set as 3.0 mm, and the boundary difference distance is obtained as a numerical result.

[0031] S203: According to the boundary difference distance, the candidate path advancing direction vector and the neural fiber bundle dominant direction vector are extracted, the included angle is calculated, and the spatial risk level of the candidate path is analyzed in combination with the functional sensitivity parameter set to establish the functional interaction risk; According to the boundary difference distance, the average difference is 3.0 mm, the candidate path advancing direction vector is extracted, the path point sequence is calculated, including the path point from (5.0, 5.0, 5.0) to (5.2, 5.5, 5.8), the vector (0.2, 0.5, 0.8), the normalized unit vector (0.2182, 0.5455, 0.8727), the neural fiber bundle dominant direction vector is obtained from the foregoing, including the Broca area dominant direction vector (0.8, 0.2, 0.1), the included angle is calculated, the dot product , the length of v1 (unit vector modulus should be 1, verified by calculation here), v2 , the cosine value , the included angle , in combination with the functional sensitivity parameter set, including the Broca area sensitivity coefficient 4.67, the risk level of the candidate path in space is analyzed, the risk value is defined as the difference distance multiplied by the included angle multiplied by the sensitivity coefficient, 3.0x64.8x4.67, 3.0x64.8=194.4, 194.4x4.67≈908.448, the functional interaction risk value 908.448 is established.

[0032] Please refer to Figure 4 , the specific steps of S3 are as follows: S301: Based on the functional interaction risk, each candidate path is divided into equal lengths, and for each segmented path point, the coordinate position of the path point in three-dimensional space is obtained to generate path segmented point data; Based on the functional interaction risk value 908.448, the risk judgment interval is set: the risk value greater than 500 is high risk, the path is divided into 10 segments, the risk value between 100 and 500 is medium risk, the segment number is 5, the risk value less than 100 is low risk, and the segment number is 3. Here, the risk 908.448 is greater than 500, so the path is divided into 10 equal lengths, including the path from the starting point (5.0, 5.0, 5.0) mm to the end point (10.0, 20.0, 30.0) mm, and the path length is calculated using the Euclidean distance formula mm, the third point (5.540+0.540, 6.620+1.620, 7.700+2.700) ≈ (6.080, 8.240, 10.400) mm, and so on until the tenth point (10.0, 20.0, 30.0), for each segmented path point, obtain its coordinate position in three-dimensional space, stored in array form, including the point set [(5.0, 5.0, 5.0), (5.540, 6.620, 7.700), (6.080, 8.240, 10.400), …, (10.0, 20.0, 30.0)], generate path segment point data.

[0033] S302: Based on the path segment point data, obtain the three-dimensional coordinate data of the brain blood vessels, ventricles and brain function areas in real time using intraoperative real-time images, calculate the distance of each segmented path point to the brain blood vessels, ventricles and function areas and select the optimal distance, and generate optimal path distance. Based on the path segment point data, periodically collect three-dimensional images of the current brain tissue through intraoperative magnetic resonance (iMRI) or ultrasound (iUS) and the like, including the second segmented path segment point (5.540, 6.620, 7.700) mm, call the three-dimensional coordinate data of the brain blood vessels, such as the blood vessel point list (8.0, 10.0, 12.0), (9.0, 11.0, 13.0), …, (15.0, 18.0, 20.0), the three-dimensional coordinate data of the brain ventricle, such as the ventricle point list [(12.0, 15.0, 18.0), (13.0, 16.0, 19.0), …, (20.0, 25.0, 30.0)], and the three-dimensional coordinate data of the brain function area, such as the Broca area point list [(15.0, 20.0, 25.0), (16.0, 21.0, 26.0), …, (20.0, 25.0, 30.0)], calculate the optimal distance of each segmented path point to the brain blood vessels, ventricles and function areas, and use the Euclidean distance formula, for the point (5.540, 6.620, 7.700), calculate the distance to the brain blood vessel point (8.0, 10.0, 12.0) mm, calculate the distance to the ventricle point (12.0, 15.0, 18.0) mm, and the distance to the Broca area point (15.0, 20.0, 25.0) needs to be converted into obstacle field intensity values, first calculate the distance mm, and the minimum value of 6.00 mm is taken as the optimal path distance.

[0034] S303: According to the optimal path distance, convert the shortest distance into obstacle field intensity values, and aggregate and evaluate the obstacle field intensity of the multi-section path points to generate the segmented obstacle field intensity; Convert the optimal path distance of 6.00 mm into obstacle field intensity values, use the conversion formula intensity = 1 / distance, when the distance is less than the safety threshold of 2.0 mm, the intensity is set to the maximum value of 100 (to avoid division by zero), the distance here is 6.00 mm which is greater than 2.0 mm, intensity = 1 / 6.00 ≈ 0.1667, similarly calculate 10 points [0.1667, 0.1818, 0.1950, 0.2000, 0.1900, 0.1850, 0.1800, 0.1750, 0.1700, 0.1650], aggregate and evaluate the obstacle field intensity of the multi-section path points, take the average intensity as the aggregation method, 0.1667+0.1818+0.1950+0.2000+0.1900+0.1850+0.1800+0.1750+0.1700+0.1650, the total is about 1.8000, divided by the number of points 10, the average intensity is 0.1800, evaluate the intensity level, set the intensity judgment interval: average intensity greater than 0.5 is high obstacle, indicating high path risk, intensity between 0.1 and 0.5 is moderate obstacle, intensity less than 0.1 is low obstacle, the average intensity here is 0.1800 which belongs to moderate obstacle, generate the segmented obstacle field intensity value 0.1800.

[0035] Please refer to Figure 5 , the specific steps of S4 are: S401: Based on the segmented obstacle field intensity, extract the intensity data of adjacent sections in turn, calculate the difference between adjacent sections and record the corresponding section index, generate the obstacle intensity difference sequence; Based on the segmented obstacle field intensity sequence, the obtained 10 segmented point intensity value array [0.1667, 0.1818, 0.1950, 0.2000, 0.1900, 0.1850, 0.1800, 0.1750, 0.1700, 0.1650], the intensity data of adjacent segments are extracted in turn, for index 1 and 2, the intensity values 0.1667 and 0.1818 are extracted, the absolute difference value |0.1818-0.1667|=0.0151 is calculated, and the corresponding segmented index pair (1, 2) is recorded, for index 2 and 3, the intensity values 0.1818 and 0.1950 are extracted, the difference value |0.1950-0.1818|=0.0132 is calculated, and the index pair (2, 3) is recorded, for index 3 and 4, the intensity values 0.1950 and 0.2000 are extracted, the difference value |0.2000-0.1950|=0.0050 is calculated, and the index pair (3, 4) is recorded, for index 4 and 5, the intensity values 0.2000 and 0.1900 are extracted, the difference value |0.1900-0.2000|=0.0100 is calculated, and the index pair (4, 5) is recorded, for index 5 and 6, the intensity values 0.1900 and 0.1850 are extracted, the difference value |0.1850-0.1900|=0.0050 is calculated, and the index pair (5, 6) is recorded, for index 6 and 7, the intensity values 0.1850 and 0.1800 are extracted, the difference value |0.1800-0.1850|=0.0050 is calculated, and the index pair (6, 7) is recorded, for index 7 and 8, the intensity values 0.1800 and 0.1750 are extracted, the difference value |0.1750-0.1800|=0.0050 is calculated, and the index pair (7, 8) is recorded, for index 8 and 9, the intensity values 0.1750 and 0.1700 are extracted, the difference value |0.1700-0.1750|=0.0050 is calculated, and the index pair (8, 9) is recorded, for index 9 and 10, the intensity values 0.1700 and 0.1650 are extracted, the difference value |0.1650-0.1700|=0.0050 is calculated, and the index pair (9, 10) is recorded, and the obstacle intensity difference value sequence is generated, including the difference value list [0.0151, 0.0132, 0.0050, 0.0100, 0.0050, 0.0050, 0.0050, 0.0050, 0.0050] and the index list (1, 2), (2, 3), (3, 4), (4, 5), (5, 6), (6, 7), (7, 8), (8, 9), (9, 10), and the obstacle intensity difference value sequence is generated.

[0036] S402: According to the obstacle intensity difference value sequence, the set gradient threshold value is called, the difference values are compared one by one, the difference values exceeding the threshold value are screened and the corresponding spatial coordinate range is recorded, and the super-threshold gradient difference value set is obtained; According to the obstacle intensity difference sequence, including the difference list [0.0151, 0.0132, 0.0050, 0.0100, 0.0050, 0.0050, 0.0050, 0.0050, 0.0050] and the index list [(1, 2), (2, 3), (3, 4), (4, 5), (5, 6), (6, 7), (7, 8), (8, 9), (9, 10)], call the set gradient threshold value, the gradient threshold value is based on the brain imaging typical spatial resolution 1 millimeter and the intensity change sensitivity, set the threshold value = 0.0100, indicating that the intensity change exceeding this value is considered as a significant gradient, compare the difference one by one, for the difference 0.0151, compare 0.0151> 0.0100, true, filter this difference, record the corresponding spatial coordinate range, the index pair (1, 2) corresponds to the segmented point coordinates obtained from the foregoing S301, point 1 (5.0, 5.0, 5.0) millimeter, point 2 (5.540, 6.620, 7.700) millimeter, the coordinate range is the line segment between the two points, approximately as the cuboid range x [5.0, 5.540], y [5.0, 6.620], z [5.0, 7.700], for the difference 0.0132, compare 0.0132> 0.0100, true, filter, index pair (2, 3) point 2 (5.540, 6.620, 7.700), point 3 (6.080, 8.240, 10.400), coordinate range x [5.540, 6.080], y [6.620, 8.240], z [7.700, 10.400], for the difference 0.0050, compare 0.0050> 0.0100, false, do not filter, similar processing all differences, filter out the threshold difference 0.0151 and 0.0132, record the coordinate range, get the threshold gradient difference set, including the difference [0.0151, 0.0132] and the coordinate range [range 1, range 2].

[0037] S403: Call the threshold gradient difference set, convert the difference into a penalty value and superimpose it in the corresponding spatial coordinate range, track the changes of brain tissue morphology and position through real-time images, mark the risk area, and establish an obstacle gradient risk map; Call the super-threshold gradient difference set, including the difference [0.0151, 0.0132] and the coordinate range [range 1, range 2], wherein range 1 corresponds to the index (1, 2) coordinates x [5.0, 5.540], y [5.0, 6.620], z [5.0, 7.700], and range 2 corresponds to the index (2, 3) coordinates x [5.540, 6.080], y [6.620, 8.240], z [7.700, 10.400], convert the difference into a penalty value, the penalty value is set based on the difference amplitude, define the penalty value = difference x 1000 to amplify to the operable range, for the difference 0.0151, the penalty value = 0.0151 x 1000 = 15.1, for the difference 0.0132, the penalty value = 0.0132 x 1000 = 13.2, superimpose the corresponding space coordinate range, for range 1, superimpose the penalty value 15.1, for range 2, superimpose the penalty value 13.2, mark the risk area, the risk area is determined based on the penalty value, the penalty value greater than 10 is a high-risk area, finally, through real-time image continuous tracking of the morphological and positional changes of brain tissue due to "brain drift", and taking this as the trigger condition, periodically and cyclically recalculate the steps S3 to S402, so as to establish and dynamically refresh an obstacle gradient risk map, the map contains coordinate ranges and corresponding penalty values, including map entries: range 1 penalty 15.1, range 2 penalty 13.2.

[0038] Please refer to Figure 6 , the specific steps of S5 are: S501: Based on the obstacle gradient risk map and the functional interaction risk, extract the corresponding risk value in the space coordinate range passed by each path, and perform weighted cumulative calculation according to the path order to generate the total amount of path risk; Based on the hazard gradient risk map and the functional interaction risk, the hazard gradient risk map is obtained from S403, containing the coordinate range and the penalty value, including the map entry of path A: range 1 penalty 15.1 (coordinates x[5.0, 5.540], y[5.0, 6.620], z[5.0, 7.700]), range 2 penalty 13.2 (coordinates x[5.540, 6.080], y[6.620, 8.240], z[7.700, 10.400]), the functional interaction risk is obtained from S203, the value is 908.448, for each path, the corresponding risk value is extracted in the space coordinate range passed, including path A has 10 segment points, the coordinate sequence is obtained from S301: point 1 (5.0, 5.0, 5.0), point 2 (5.540, 6.620, 7.700), point 3 (6.080, 8.240, 10.400),..., point 10 (10.0, 20.0, 30.0), for point 1 (5.0, 5.0, 5.0), check whether it is in the risk map range, point 1 is in range 1 (x=5.0 is in [5.0, 5.540], y=5.0 is in [5.0, 6.620], z=5.0 is in [5.0, 7.700]), the penalty value 15.1 is extracted, for point 2 (5.540, 6.620, 7.700), at the junction of range 1 and range 2, the average penalty value (15.1+13.2) / 2=14.15 is taken, for point 3 (6.080, 8.240, 10.400), in range 2, the penalty value 13.2 is extracted, for other points not in the risk range, the penalty value is 0, set the path A point risk value list as [15.1, 14.15, 13.2, 0, 0, 0, 0, 0, 0, 0], according to the path order, the weighted cumulative calculation is carried out, the weight is based on the functional interaction risk, the weight factor is defined as the functional interaction risk divided by 1000 for normalization, 908.448 / 1000≈0.9084, the weighted risk accumulation is calculated, for point 1 risk 15.1, the weighted value 15.1x0.9084≈13.718, the cumulative value is 13.718, for point 2 risk 14.15, the weighted value 14.15x0.9084≈12.857, the cumulative value is 13.718+12.857=26.575, for point 3 risk 13.2, the weighted value 13.2x0.9084≈11.991, the cumulative value is 26.575+11.991=38.566, for point 4 to point 10 risk 0, the weighted value 0, the cumulative value is unchanged, the final path risk total amount is 38.566, similar processing other paths, including path B risk total amount 25.000, path C risk total amount 50.000, the path risk total amount value is generated.

[0039] S502: According to the total risk of the path, score all candidate paths, compare the total risk value of the multi-path with the set score benchmark value, and sort the paths according to the comparison result to obtain a candidate path score sequence; According to the total risk of the path, including the total risk of path A 38.566, the total risk of path B 25.000, and the total risk of path C 50.000, score all candidate paths, the score benchmark value is set by analyzing the risk distribution of all candidate paths, calculate the average value of all path risk total (38.566+25.000+50.000) / 3≈37.855, set the score benchmark value as 37.855, the score formula is score=100-(risk total / benchmark value)×100, if risk total<benchmark value, otherwise score=50-(risk total-benchmark value) / benchmark value×50, for path A risk total 38.566>37.855, score=50-(38.566-37.855) / 37.855×50≈50-(0.711 / 37.855)×50≈50-(0.01878)×50≈50-0.939≈49.061, for path B risk total 25.000<37.855, score=100-(25.000 / 37.855)×100≈100-(0.660)×100≈100-66.0=34.0 (Correction: should be 100-66.0=34.0, but usually the higher the score, the better, so adjust the formula, set the score to be inversely proportional to the risk, define score=(benchmark value / risk total)×100 if risk total>0, otherwise 100, for path A score=(37.855 / 38.566)×100≈0.9816×100=98.16, for path B score=(37.855 / 25.000)×100≈1.5142×100=151.42, for path C score=(37.855 / 50.000)×100≈0.7571×100=75.71, compare the total risk value of the multi-path with the set score benchmark value, path A risk 38.566>37.855, score 98.16, path B risk 25.000<37.855, score 151.42, path C risk 50.000>37.855, score 75.71, and sort the paths according to the comparison result, score from high to low: path B score 151.42, path A score 98.16, path C score 75.71, obtain a candidate path score sequence(path B, 151.42), (path A, 98.16), (path C, 75.71).

[0040] S503: Call the candidate path scoring sequence, filter the path with the optimal score result, integrate the spatial coordinate range of the path with the risk marking result, monitor the deviation of the surgical instrument and the planned path in real time and trigger an alarm when the deviation occurs, and establish the brain puncture path planning result; The scoring reference value is set by analyzing the risk distribution of all candidate paths. Call the candidate path scoring sequence, including sequence (path B, 151.42), (path A, 98.16), (path C, 75.71), filter the path with the optimal score result, take the highest score 151.42 corresponding to path B, and integrate the spatial coordinate range of the path with the risk marking result. The coordinate range of path B is obtained from the previous text, including point sequence (5.0, 5.0, 5.0), (5.5, 6.5, 7.5),..., (10.0, 20.0, 30.0), the risk marking is obtained from the obstacle gradient risk map in the previous text, including a penalty of 15.1 in range 1 and a penalty of 13.2 in range 2, and the integration is the path planning result, including path coordinates and risk values, including the result: path B, risk marking [range 1: 15.1, range 2: 13.2], real-time monitoring of the deviation of the surgical instrument and the planned path B: through high-frequency sampling (such as 100 times per second) of the instrument position by optical or electromagnetic tracking system, calculate the shortest Euclidean distance from the instrument to path B, set the deviation threshold value to 2.0 millimeters (based on the scoring reference value and risk distribution analysis), when the deviation exceeds the deviation threshold value, immediately trigger a multi-modal alarm (including visual warning, audible beep and log recording), to ensure the safety and accuracy of the operation, and establish the brain puncture path planning result.

[0041] Please refer to Figure 7 A neurosurgical brain puncture path accurate planning system, comprising: A functional sensitivity module that acquires preoperative brain function area images and neural fiber bundle data, extracts the volume of the brain function area and the spatial coverage ratio of the neural fiber bundle dominant direction, calculates the sensitivity of multiple brain function areas, generates a functional sensitivity parameter set and transmits it to the path analysis module; A path analysis module that acquires lesions and intracranial entry points based on brain images and constructs a candidate path set, analyzes the optimal distance of the path to the brain function area boundary, and if the distance exceeds the threshold, analyzes the angle between the candidate path advancing direction and the neural fiber bundle dominant direction, and calculates the functional interaction risk in combination with the functional sensitivity parameter set and transmits it to the path segmentation module; A path segmentation module that divides each candidate path based on the functional interaction risk, calculates the optimal distance of the segmented path point to the brain blood vessels, ventricles and functional areas and converts it to obstacle field strength, generates segmented obstacle field strength and transmits it to the risk gradient module; The risk gradient module calculates the difference of adjacent segment obstacle intensity as a gradient based on the segment obstacle field intensity, converts the difference into a penalty value and marks a risk area if the difference exceeds a gradient threshold, constructs an obstacle gradient risk map and delivers the map to the path scoring module; The path scoring module scores all candidate paths based on the obstacle gradient risk map and functional interaction risk, and selects the path with the best score as the puncture planning path to generate a brain puncture path planning result.

[0042] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for precise planning of neurosurgical brain puncture pathways, characterized in that, Includes the following steps: S1: Acquire preoperative brain functional area images and nerve fiber bundle data, extract the volume of brain functional areas and the spatial coverage ratio of the dominant direction of nerve fiber bundles, calculate the sensitivity of multiple brain functional areas, and generate a set of functional sensitivity parameters. S2: Based on brain imaging, obtain the lesion and entry point location into the skull and construct a candidate path set. Analyze the optimal distance from the path to the boundary of the brain functional area. If the distance exceeds the threshold, analyze the angle between the direction of the candidate path and the dominant direction of the nerve fiber bundle, and calculate the functional interaction risk in combination with the functional sensitivity parameter set. S3: Based on the functional interaction risk, divide each candidate path into equal lengths, calculate the optimal distance from the segmented path points to cerebral blood vessels, ventricles and functional areas and convert it into obstacle field intensity, and generate segmented obstacle field intensity; S4: Based on the segmented obstacle field strength, calculate the difference in obstacle strength between adjacent segments as the gradient. If the difference exceeds the gradient threshold, convert the difference into a penalty value and mark the risk area to construct an obstacle gradient risk map. S5: Based on the obstacle gradient risk map and functional interaction risk, all candidate paths are scored, and the path with the best score is selected as the puncture planning path to generate brain puncture path planning results.

2. The method for precise planning of neurosurgical brain puncture pathways according to claim 1, characterized in that, The functional sensitivity parameter set includes volume parameters, spatial coverage ratio parameters, and sensitivity coefficients. The functional interaction risk specifically includes optimal distance, angle of inclination, and functional risk quantity. The segmented obstacle field intensity includes vascular distance intensity, ventricular distance intensity, and functional area distance intensity. The obstacle gradient risk map specifically refers to gradient difference, penalty value, and risk area coordinates. The brain puncture path planning result specifically includes path spatial coordinates, risk score, and planned path sequence.

3. The method for precise planning of neurosurgical brain puncture pathways according to claim 1, characterized in that, The specific steps of S1 are as follows: S101: Obtain preoperative brain functional area images and nerve fiber bundle data, segment the brain functional area images and measure the volume values ​​of multiple regions, extract the dominant direction data of nerve fiber bundles, call the regional volume values ​​and dominant direction data to compare and calculate the spatial coverage ratio, and generate the spatial coverage ratio. S102: Based on the spatial coverage ratio, the spatial coverage ratio of multiple brain functional areas is combined with the corresponding partition volume value, and the sensitivity of multiple brain functional areas under the distribution of nerve fibers is calculated by proportional conversion to obtain the functional area sensitivity coefficient. S103: Based on the aforementioned functional area sensitivity coefficients, integrate the coefficients of all brain functional areas to establish a functional sensitivity parameter set; The functional sensitivity refers to the functional importance or vulnerability of brain functional areas under the influence of the distribution of nerve fiber bundles.

4. The method for precise planning of neurosurgical brain puncture pathways according to claim 1, characterized in that, The specific steps of S2 are as follows: S201: Acquire brain imaging data, extract the lesion location and the entry point into the skull, connect the two as the starting point and ending point of the path to form a candidate path set, and at the same time collect three-dimensional coordinate data of brain tissue, monitor and update the location of lesions, brain functional areas and entry points in real time, and correct path deviations caused by brain drift to generate a candidate path set. S202: Based on the candidate path set, call the brain functional area boundary coordinate data, calculate the distance from multiple candidate paths to the nearest brain functional area boundary, compare the distance results with the set distance threshold, analyze the difference between the candidate paths and the distance threshold, and obtain the boundary difference distance; S203: Based on the boundary difference distance, extract the candidate path advancement direction vector and the dominant direction vector of the nerve fiber bundle, calculate the angle between them, and combine the functional sensitivity parameter set to analyze the spatial risk level of the candidate path and establish functional interaction risk.

5. The method for precise planning of neurosurgical brain puncture pathways according to claim 4, characterized in that, The distance threshold is calculated by statistically analyzing the spatial distribution characteristics of the brain functional area boundary coordinates and candidate paths, centrally calculating the minimum distance value from the path to the boundary, and setting it according to the average distance range between the lesion and the brain functional area boundary in similar image samples.

6. The method for precise planning of neurosurgical brain puncture pathways according to claim 1, characterized in that, The specific steps for S3 are as follows: S301: Based on the aforementioned functional interaction risk, each candidate path is divided into equal length segments. For each segment path point, the coordinate position of the path point in three-dimensional space is obtained to generate path segment point data. S302: Based on the path segmentation point data, real-time path segmentation point data is obtained using intraoperative real-time images, and three-dimensional coordinate data of cerebral blood vessels, ventricles and functional areas of the brain are called to calculate the distance from each segmentation path point to cerebral blood vessels, ventricles and functional areas, and the optimal path distance is selected. S303: Based on the optimal path distance, convert the shortest distance into an obstacle field intensity value, and summarize and evaluate the obstacle field intensity of the multi-segment path points to generate segmented obstacle field intensity.

7. The method for precise planning of neurosurgical brain puncture pathways according to claim 1, characterized in that, The specific steps of S4 are as follows: S401: Based on the segmented obstacle field intensity, extract the intensity data of adjacent segments in sequence, calculate the difference between adjacent segments and record the corresponding segment index to generate an obstacle intensity difference sequence; S402: Based on the obstacle strength difference sequence, call the set gradient threshold, compare the differences one by one, filter the differences that exceed the gradient threshold and record the corresponding spatial coordinate range to obtain the over-threshold gradient difference set; S403: Call the set of superthreshold gradient differences, convert the differences into penalty values ​​and superimpose them in the corresponding spatial coordinate range, continuously track the changes in brain tissue morphology and position through real-time images, mark risk areas, and establish a barrier gradient risk map.

8. The method for precise planning of neurosurgical brain puncture pathways according to claim 7, characterized in that, The gradient threshold is set by statistically analyzing the difference in the intensity of the segmented obstacle field, selecting the mean and standard deviation of the difference between adjacent segments as the measurement basis, and combining the concentration interval of the difference distribution in similar samples.

9. The method for precise planning of neurosurgical brain puncture pathways according to claim 1, characterized in that, The specific steps of S5 are as follows: S501: Based on the obstacle gradient risk map and functional interaction risk, extract the corresponding risk value for each path within the spatial coordinate range it passes through, and perform weighted cumulative calculation according to the path order to generate the total path risk. S502: Based on the total path risk, score all candidate paths, compare the total risk value of multiple paths with the set scoring benchmark value, and sort the paths according to the comparison results to obtain a candidate path scoring sequence. S503: Call the candidate path scoring sequence, filter the path with the best scoring result, integrate the spatial coordinate range of the path with the risk marking result, monitor the deviation between the surgical instruments and the planned path in real time and trigger an alarm when the deviation occurs, and establish the brain puncture path planning result. The scoring benchmark value is set by analyzing the risk distribution of all candidate paths.

10. A precise planning system for neurosurgical brain puncture paths, characterized in that, The system is used to perform the precise planning method for neurosurgical brain puncture path as described in any one of claims 1-9, the system comprising: The functional sensitivity module acquires preoperative brain functional area images and nerve fiber bundle data, extracts the volume of brain functional areas and the spatial coverage ratio of the dominant direction of nerve fiber bundles, calculates the sensitivity of multiple brain functional areas, generates a functional sensitivity parameter set and transmits it to the path analysis module. The path analysis module acquires the lesion and entry point of the brain based on brain imaging and constructs a set of candidate paths. It analyzes the optimal distance from the path to the boundary of the brain functional area. If the distance exceeds the threshold, it analyzes the angle between the direction of the candidate path and the dominant direction of the nerve fiber bundle. It also calculates the functional interaction risk in combination with the functional sensitivity parameter set and transmits it to the path segmentation module. The path segmentation module divides each candidate path into equal lengths based on the functional interaction risk, calculates the optimal distance from the segmented path point to the cerebral blood vessels, ventricles and functional areas and converts it into obstacle field intensity, generates segmented obstacle field intensity and transmits it to the risk gradient module. The risk gradient module calculates the difference in obstacle intensity between adjacent segments as a gradient based on the segmented obstacle field intensity. If the difference exceeds the gradient threshold, it converts the difference into a penalty value and marks the risk area, constructs an obstacle gradient risk map, and transmits it to the path scoring module. The path scoring module scores all candidate paths based on the obstacle gradient risk map and functional interaction risk, and selects the path with the best score as the puncture planning path to generate brain puncture path planning results.

Images