Lock hole operation path planning method under rigid-flexible mixed constraint
Through the keyhole surgical path planning method under rigid-flexible hybrid constraints, the multi-constraint optimization problem in the skull base keyhole surgical path planning is solved, minimizing skull trauma and efficient tumor resection are achieved, and it is suitable for the auxiliary field of neurosurgery.
Patent Information
- Application Number
- CN202510454004.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-11
- Publication Date
- 2025-08-12
AI Technical Summary
The prior art has difficulty in building surgical path optimization models in the planning of skull base keyhole surgical paths, and it is difficult to achieve minimally invasive and precise surgery while avoiding important tissues.
The keyhole surgical path planning method under rigid-flexible hybrid constraints is adopted. By establishing a knowledge base for surgical approaches, the approach and initial bone window are automatically selected, combined with flexible constraints such as cranial nerves and blood vessels, the objective function is optimized, the tumor sub-regions are divided and the bone window is adjusted, and the optimal surgical path is designed.
While avoiding important tissues, it is designed to design a surgical path with minimal skull trauma, improve the minimally invasiveness and accuracy of the surgery, and reduce damage to normal tissues. It is suitable for young doctors with insufficient experience.
Smart Images
Figure CN120473110A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the fields of medical imaging and neurosurgery assistance under computer graphics, and in particular to a keyhole surgery path planning method under rigid-flexible hybrid constraints. Background Art
[0002] Skull base neurosurgery is a very challenging problem in the field of surgery due to the complex anatomical structure of the skull base, the abundance of important neural tissue and blood vessels, and the limited operating space. It is known as the "crown jewel" of surgery. In this field, keyhole minimally invasive surgery has become an important surgical method due to its advantages such as small surgical incisions and less retraction on brain tissue. However, its narrow surgical field space places extremely high demands on the doctor's professional knowledge and experience. Although image-guided surgery can assist in surgical planning through preoperative three-dimensional reconstruction of the skull base, existing technologies still have significant defects in surgical path planning.
[0003] Automatic surgical path planning algorithms aim to transform clinical needs into optimization constraints by constructing surgical path optimization models. These algorithms then calculate the surgical trajectory that best meets all clinical requirements, thereby reducing surgical risk and achieving minimally invasive and precise surgery. In actual surgical procedures, surgical path planning must consider numerous factors, such as the distance between the path and critical tissue, the angle of the surgical approach, and the coverage of the target area. Therefore, existing research generally transforms the planning problem into a multi-objective optimization (MOO) problem under constraints. The concept of Pareto optimality, derived from the definition of the optimal solution in the MOO problem, is derived. A Pareto optimal solution represents all solution vectors in the feasible space Z that cannot be further optimized without degrading the values of other objective functions (i.e., being on the Pareto frontier). To explore the feasible space and identify the optimal solution, various methods have been developed in various engineering fields, primarily categorized as scalarization methods and vector optimization methods. Scalar methods simplify the problem into a single equation consisting of weighted objective functions; in vector optimization methods, each objective function is independent of the others, aiming to find the optimal solution set located on the Pareto front. Research has shown that vector methods can discover more surgical plans that scalar methods cannot, and these plans often align with the choices made by experts.
[0004] In the field of neurosurgery, automated planning for keyhole surgery has been applied in interventions such as deep brain stimulation (DBS), stereotactic surgery, biopsy, and laser ablation. However, research on methods for skull base keyhole surgery planning is extremely scarce. Only Aghdasi et al. designed an automated planning method for a transcranial approach for tumor resection, while Rajesh et al. and Gao et al. used semi-automatic methods to plan skull base surgery. Currently, automated planning for skull base surgery faces significant challenges. The surgical path must circumvent critical structures such as the brainstem, cranial nerves, and blood vessels, while also addressing the limited surgical field of view and limited instrument access space caused by the complex skull base structure. This makes the construction of surgical path planning optimization models difficult. Therefore, it is of great significance to develop multi-constrained skull base keyhole surgery path planning methods based on precise reconstruction of nerves, blood vessels, and skull tissue. Summary of the Invention
[0005] To overcome the shortcomings of existing technologies, this paper develops a keyhole surgical path planning method for complex minimally invasive keyhole skull base tumor resection surgery under rigid-flexible hybrid constraints. This method automatically designs the bone window and surgical path for keyhole surgery. Based on a database of expert experience, this method automatically selects candidate surgical approaches, determines the initial bone window and the skull structure within the skull, and thus establishes a rigid skull space. A flexible constraint model and target optimization model are then constructed, incorporating neural and vascular constraints. The NSGAIII algorithm is then used to search for the optimal solution set to each tumor subregion. Finally, the bone window size is adjusted based on the solution set for each tumor subregion.
[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:
[0007] A keyhole surgical path planning method under rigid-flexible hybrid constraints was proposed. First, a surgical approach knowledge base was established to automatically select the surgical approach based on the tumor condition, clarify the initial bone window and skull structure along the approach, and construct the rigid constraint of the path using the impenetrable skull area and the bone window. Then, flexible tissue categories were classified according to the importance of cranial nerves, blood vessels, soft tissues, etc., and flexible constraints were defined. Then, considering the basic rules of neurosurgery, a surgical path planning objective function was established, which included target coverage, minimizing skull removal, minimizing structure overlap, and maximizing the distance to dangerous structures. NSGA III was used to obtain the Pareto frontier and the optimal solution set. Finally, the tumor was divided into multiple sub-regions, and the path was planned independently for each sub-region. The surgical approach bone window was adjusted based on the optimal path area of each sub-region.
[0008] Furthermore, the keyhole surgery path planning method under rigid-flexible hybrid constraints includes the following steps:
[0009] 1) Skull rigidity constraint, the process is as follows:
[0010] Establish a surgical approach knowledge base, automatically select the corresponding surgical approach according to the tumor condition and clinical conditions, determine the initial bone window and skull structure on the approach; divide the skull into removable skull S according to the importance of the skull bone-e and non-penetrable skull area S bone-ne , where the rigid constraints of the impenetrable area and bone window construction path must satisfy,
[0011]
[0012] Among them, P is the surgical path to be sought, S bone-e For the skull area that can be removed, S bone-ne For non-penetrable skull area, S win The pterional region is defined; the initial bone window is determined based on bony landmarks. For the small pterional keyhole approach, the pterional region is located and a predetermined opening length is set as the bone window size. The pterional region is defined by the junction of the frontal, sphenoid, temporal, and parietal bones.
[0013] 2) Flexible constraint, the process is as follows:
[0014] According to the importance of cranial nerves, blood vessels, and soft tissues, flexible tissues are divided into the following three categories, and flexible constraints are defined:
[0015] a) Removable structure S1: including soft tissue, which has little impact on patient function after removal;
[0016] b) Dangerous structures S2: including cranial nerves or important blood vessels. Removal of the structure will cause serious complications to the patient. The surgical path should not pass through these structures and the distance should be at least 2 mm;
[0017] c) Critical but movable or deformable tissues S3: including the cerebellum, which can be retracted to a certain extent, but the retraction range should not be excessive;
[0018] The S3 structure is divided into two areas: the area that can be safely passed through by safe displacement is considered S1, and the area that cannot be avoided by displacement is defined as S3′. The maximum deformation or movement degree is set according to anatomy and expert experience. The surgical path must meet
[0019] dist min (P,S2)>d (3)
[0020]
[0021] Where S2 is the dangerous structure, d is the distance between P and S2, and S3′ is the area that cannot be avoided even by displacement;
[0022] 3) Objective function, the process is as follows:
[0023] According to the basic rules of neurosurgery, a single keyhole path is considered and the objective function of surgical path planning is established.
[0024] min{f n (x)}n=1,2,3 (5)
[0025] in,
[0026]
[0027] Consider n keyhole path planning, the objective function is,
[0028] min{f j (x)}j=1,2,…,3n (7)
[0029] in,
[0030]
[0031] Among them, P i represents the i-th path, θ(P1,P2,···,P n )>θ m Indicates that the angle between any two paths must be greater than the set value. For the constructed multi-objective optimization problem, NSGA III is used to obtain the Pareto frontier and the Pareto optimal solution set;
[0032] 4) Adjustment of the surgical approach bone window. The process is as follows:
[0033] First, the tumor is divided into multiple sub-areas, and then path planning is performed independently for each sub-area. Finally, the optimal path areas of each sub-area are combined to find the path area with the smallest area, and the optimal path area of each sub-area intersects with it.
[0034] The technical concept of this invention is to integrate rigid constraints (constraints on the skull base and key bony landmarks) and flexible constraints (the need to protect cranial nerves and blood vessels) into the surgical path optimization model based on the precise identification of cranial nerves and key bony landmarks at the skull base. Using a multi-objective optimization algorithm, factors such as the distance between the path and key tissues, approach angle, and target area coverage are comprehensively considered. While ensuring instrument operation and effective tumor resection, an effective surgical path with minimal skull trauma is designed, thus realizing a keyhole surgical path planning algorithm under hybrid rigid-flexible constraints.
[0035] The beneficial effects of this invention are primarily manifested in: a keyhole surgical path planning algorithm under hybrid rigid-flexible constraints fully considers multiple clinical factors, designing a surgical path that minimizes skull trauma while avoiding critical neurovascular tissue, thus achieving minimally invasive surgery. This allows for maximum tumor resection while minimizing damage to normal tissue, thereby improving surgical outcomes.
[0036] This invention provides an automated and intelligent surgical planning method for skull base keyhole surgery, reducing excessive reliance on the doctor's rich clinical experience. It enables young doctors with relatively insufficient experience to formulate surgical plans more accurately, helping to promote the popularization and development of skull base neurosurgery techniques. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] Figure 1 This is a flow chart of the keyhole surgery path planning method under rigid-flexible mixed constraints. DETAILED DESCRIPTION
[0038] The present invention will be further described below.
[0039] Reference Figure 1 A keyhole surgical path planning method under rigid-flexible hybrid constraints is proposed. First, a surgical approach knowledge base is established to automatically select the surgical approach according to the tumor condition, clarify the initial bone window and skull structure on the approach, and construct the path rigid constraint with the impenetrable skull area and the bone window; then, flexible tissue categories are divided according to the importance of cranial nerves, blood vessels, soft tissues, etc., and flexible constraints are defined; then, considering the basic rules of neurosurgery, a surgical path planning objective function is established that covers target coverage, minimizes skull removal, minimizes structure overlap, and maximizes the distance to dangerous structures. NSGA III is used to obtain the Pareto frontier and the optimal solution set. Finally, the tumor is divided into multiple sub-areas, and the path is planned independently for each sub-area. The surgical approach bone window is adjusted based on the optimal path area of each sub-area.
[0040] The method comprises the following steps:
[0041] 1) Skull rigidity constraint, the process is as follows:
[0042] Establish a surgical approach knowledge base, automatically select the corresponding surgical approach according to the tumor condition and clinical conditions, determine the initial bone window and skull structure on the approach, and divide the skull into removable skull S according to the importance of the skull bone-e and non-penetrable skull area S bone-ne , where the rigid constraints of the impenetrable area and bone window construction path must satisfy,
[0043]
[0044] Among them, P is the surgical path to be sought, S bone-e For the skull area that can be removed, S bone-ne For non-penetrable skull area, S win The initial bone window can be determined based on bony landmarks. For example, in a small pterional keyhole approach, the pterional region is located and a 4-cm opening (slightly larger than a conventional keyhole) is set as the bone window size. The pterional region can be defined by the junction of the frontal, sphenoid, temporal, and parietal bones.
[0045] 2) Flexible constraint, the process is as follows:
[0046] According to the importance of cranial nerves, blood vessels, and soft tissues, flexible tissues are divided into the following three categories, and flexible constraints are defined:
[0047] a) Removable structures S1: such as soft tissue, which have little impact on patient function after removal;
[0048] b) Dangerous structures S2: such as cranial nerves, important blood vessels, etc. Removal of the structure will cause serious complications to the patient. The surgical path should not pass through these structures and the distance should be at least 2 mm;
[0049] c) Critical but movable or deformable tissues S3: such as the cerebellum, can be stretched to a certain extent, but the stretching range should not be too large.
[0050] The surgical path may intersect with the S3 structure, and the instrument can be moved by pulling. To simulate this feature, the S3 structure is divided into two areas: the area that can be safely passed through by safe displacement (considered as S1), and the area that cannot be avoided by displacement, defined as S3'. The maximum deformation or movement can be set according to anatomy and expert experience. The surgical path must meet the following requirements:
[0051] dist min (P,S2)>d (3)
[0052]
[0053] Where S2 is the dangerous structure, d is the distance between P and S2, and S3′ is the area that cannot be avoided even by displacement;
[0054] 3) Objective function, the process is as follows:
[0055] Automatic surgical path planning requires familiarity with the neurosurgeon's thought process during preoperative planning and conversion of that description into a mathematical model. The basic rules for neurosurgery are as follows:
[0056] a) Target coverage: The surgical path must reach the target area.
[0057] b) Minimized skull removal: Although surgeons try to avoid skull removal to minimize trauma, in some cases, skull removal is necessary to free up surgical space due to target location limitations.
[0058] c) Minimize structural overlap: Soft tissue resection and brain tissue displacement will cause certain damage to the patient, so the surgical path should pass through as few structures as possible.
[0059] d) Maximize the distance from critical structures: The surgical path should be as far away from critical structures as possible, such as the optic nerve, to prevent damage to these important structures during surgery. Therefore, the distance between the path and critical tissues should be as large as possible.
[0060] Therefore, considering a single keyhole path, the objective function of surgical path planning is established.
[0061] min{f n (x)}n=1,2,3 (5)
[0062] in,
[0063]
[0064] Consider n keyhole path planning, the objective function is,
[0065] min{f j (x)}j=1,2,…,3n (7)
[0066] in,
[0067]
[0068] Among them, P i represents the i-th path, θ(P1,P2,···,P n )>θ m Indicates that the angle between any two paths must be greater than a set value, usually set to 20°. For the constructed multi-objective optimization problem, the algorithm intends to use NSGA III to obtain the Pareto frontier and the Pareto optimal solution set;
[0069] 4) Adjustment of the surgical approach bone window. The process is as follows:
[0070] Since existing methods only plan the path for a certain point on the lesion, such as puncture and ablation, in order to achieve effective resection of the entire tumor, the algorithm first divides the tumor into multiple sub-areas, then independently plans the path for each sub-area, and finally combines the optimal path areas of each sub-area to find the path area with the smallest area, with which the optimal path area of each sub-area intersects.
[0071] The embodiments of this specification are merely examples of implementations of the invention and are provided for illustrative purposes only. The scope of protection of the present invention should not be considered limited to the specific embodiments described in these embodiments. The scope of protection of the present invention also extends to equivalent technical means that can be conceived by a person of ordinary skill in the art based on the invention.
Claims
1. A keyhole surgery path planning method under rigid-flexible hybrid constraints, characterized by: First, a surgical approach knowledge base was established to automatically select the surgical approach based on the tumor condition, identify the initial bone window and skull structure along the approach, and construct rigid constraints for the path using the impenetrable skull area and the bone window. Flexible tissue categories were then classified according to the importance of cranial nerves, blood vessels, soft tissues, etc., and flexible constraints were defined. Considering the basic rules of neurosurgery, a surgical path planning objective function was established that included target coverage, minimizing skull removal, minimizing structure overlap, and maximizing distance from dangerous structures. NSGAIII was used to obtain the Pareto frontier and optimal solution set. Finally, the tumor was divided into multiple sub-areas, and the path was independently planned for each sub-area. The surgical approach bone window was then adjusted based on the optimal path area of each sub-area.
2. The keyhole surgery path planning method under rigid-flexible hybrid constraints according to claim 1, characterized in that: The method comprises the following steps: 1) Skull rigidity constraint, the process is as follows: Establish a surgical approach knowledge base, automatically select the corresponding surgical approach according to the tumor condition and clinical conditions, determine the initial bone window and skull structure on the approach; divide the skull into removable skull S according to the importance of the skull bone-e and non-penetrable skull area S bone-ne , where the rigid constraints of the impenetrable area and bone window construction path must satisfy, Among them, P is the surgical path to be sought, S bone-e For the skull area that can be removed, S bone-ne For non-penetrable skull area, S win The pterional region is defined; the initial bone window is determined based on bony landmarks. For the small pterional keyhole approach, the pterional region is located and a predetermined opening length is set as the bone window size. The pterional region is defined by the junction of the frontal, sphenoid, temporal, and parietal bones. 2) Flexible constraint, the process is as follows: According to the importance of cranial nerves, blood vessels, and soft tissues, flexible tissues are divided into the following three categories, and flexible constraints are defined: a) Removable structure S1: including soft tissue, which has little impact on patient function after removal; b) Dangerous structures S2: including cranial nerves or important blood vessels. Removal of the structure will cause serious complications to the patient. The surgical path should not pass through these structures and the distance should be at least 2 mm; c) Critical but movable or deformable tissues S3: including the cerebellum, which can be retracted to a certain extent, but the retraction range should not be excessive; The S3 structure is divided into two areas: the area that can be safely passed through by safe displacement is considered S1, and the area that cannot be avoided by displacement is defined as S3′. The maximum deformation or movement degree is set according to anatomy and expert experience. The surgical path must meet dist min (P,S2)>d (3) Where S2 is the dangerous structure, d is the distance between P and S2, and S3′ is the area that cannot be avoided even by displacement; 3) Objective function, the process is as follows: According to the basic rules of neurosurgery, a single keyhole path is considered and the objective function of surgical path planning is established. min{f n (x)}n=1,2,3 (5) in, Consider n keyhole path planning, the objective function is, min{f j (x)} j=1,2,…,3n (7) in, Among them, P i represents the i-th path, θ(P1,P2,···,P n )>θ m Indicates that the angle between any two paths must be greater than the set value. For the constructed multi-objective optimization problem, NSGA III is used to obtain the Pareto frontier and the Pareto optimal solution set; 4) Adjustment of the surgical approach bone window. The process is as follows: First, the tumor is divided into multiple sub-areas, and then path planning is performed independently for each sub-area. Finally, the optimal path areas of each sub-area are combined to find the path area with the smallest area, and the optimal path area of each sub-area intersects with it.