Novel three-dimensional virtual model assisted partial renal resection artery occlusion strategy optimization method

Abstract

The invention provides a novel three-dimensional virtual model-assisted partial renal resection artery occlusion strategy optimization method, and relates to the technical field of artificial intelligence, medical image processing and application, in particular to a novel three-dimensional virtual model-assisted partial renal resection artery occlusion strategy optimization method. The method comprises the following steps of: firstly, processing original image data scanned by CT (Computed Tomography) by utilizing M3DICS three-dimensional reconstruction software, and obtaining a corresponding digital kidney three-dimensional model; secondly, the kidney three-dimensional model is combined with a mathematical algorithm, the kidney parenchyma volume is divided into blood vessel perfusion areas of the kidney parenchyma volume, a new three-dimensional virtual kidney model with different perfusion areas is generated, artery branches needing to be clamped when selective occlusion is possibly carried out are recorded based on the modeling result, selective occlusion simulation is carried out, and the blood vessel perfusion areas of the kidney are simulated; and an accurate and benefit-maximized treatment scheme is formulated for the object.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, medical image processing and application technology, and in particular to a novel three-dimensional virtual model-assisted partial nephrectomy artery occlusion strategy optimization method. Background Art

[0002] Traditional partial nephrectomy often adopts the method of blocking the main renal artery, which has a more accurate blocking effect, but it can also cause damage to kidney function due to ischemia-reperfusion. In recent years, with the development of minimally invasive technology and the implementation of the "zero ischemia" concept, selective renal artery blocking strategies are more commonly used in laparoscopic partial nephrectomy. The principle is to reduce the volume of renal parenchyma exposed to ischemic damage to optimize postoperative functional outcomes. Compared with blocking the main renal artery, selective blocking can better preserve renal function and has comparable safety results.

[0003] Selective blockage of the renal artery branches of the tumor requires the clear identification of the blood supply arteries of the renal tumor and its distribution and course before surgery so that doctors can make a good selective blockage plan. In the past, when standard two-dimensional images (magnetic resonance angiography (MRA), CT angiography (CTA), etc.) were used for selective blockage planning, the accuracy of the blockage plan could be determined by intraoperative injection of indocyanine green (ICG) (Diana P, Buffi NM, Lughezzani G, et al. The role of intraoperative indocyanine green in robot-assisted partial nephrectomy: results from a large, multi-institutional series. Eur Urol 2020; 78: 743–9). However, even under the guidance of ICG, selective blockage is still a critical step in partial nephrectomy because there may be a risk of bleeding if the vascular management strategy is inappropriate.

[0004] In recent years, the introduction of high-precision three-dimensional virtual models has solved the above problems to a certain extent. It assists surgeons by increasing the preoperative understanding of vascular anatomy and directly overlaying images into the surgical field during surgery through augmented reality (AR) technology (Wake N, Bjurlin MA, Rostami P, Chandarana H, Huang WC. Three-dimensional printing and augmented reality: enhanced precision for robotic assisted partial nephrectomy. Urology 2018; 116: 227–8). However, this technology may not be sufficient to ensure an appropriate bloodless resection plane, and surgeons may have to turn to overall blockade. Currently, high-precision kidney three-dimensional model reconstruction based on CT scan image data mostly defines the intrarenal vascular boundaries empirically based on the direction of the arterial branches entering the renal sinus, rather than based on the perfusion area, which may lead to failure during surgery.

[0005] In summary, regarding the problem of precise and selective vascular occlusion during partial nephrectomy, the current preoperative occlusion planning method still has a certain degree of empiricism in managing vascular veins, and the occlusion strategy needs to be adjusted according to the bleeding situation during the operation. Summary of the invention

[0006] The purpose of the present invention is to provide a novel three-dimensional virtual model-assisted partial nephrectomy artery occlusion strategy optimization method, aiming to solve the problems mentioned in the above background technology.

[0007] To achieve the above object, the present invention adopts the following technical solution: the novel three-dimensional virtual model-assisted partial nephrectomy artery occlusion strategy optimization method comprises the following steps:

[0008] (1) Image acquisition: Obtaining the subject's CT scan image data from the preoperative abdominal enhanced CT examination image database;

[0009] (2) High-precision 3D model rendering: CT scan data is imported into the dedicated software M3DICS in DICOM format for segmentation and 3D kidney model reconstruction;

[0010] (3) Estimating the renal vascular perfusion area: using the three-dimensional virtual model in step (2) in combination with corresponding mathematical tools, the vascular perfusion area of ​​the kidney is calculated to generate a three-dimensional virtual kidney model with different perfusion areas;

[0011] (4) Planning the surgical plan: Observe the three-dimensional virtual kidney model in step (3), predefine the renal pedicle management, and record the number and level / sequence of arterial branches that need to be clamped for selective occlusion;

[0012] (5) Preoperative simulation: selective blockade simulation based on the three-dimensional virtual model of the perfusion area, simulation of the ischemic area and the preserved vascularized area.

[0013] Furthermore, the CT image in step (2) is three-dimensionally reconstructed, including reconstruction of the renal vascular system, collecting system, renal shape and tumor characteristics; specifically, as follows:

[0014] a. Use dynamic region growing to reconstruct the course of the renal pedicle, tumor feeding artery, extrarenal artery and intrarenal artery, up to the segmental artery;

[0015] b. The renal parenchyma is segmented using a selective threshold segmentation method to separate different voxels and group them according to their grayscale.

[0016] Furthermore, the Voronoi diagram software is applied to the three-dimensional kidney model in step (3) to calculate the vascular perfusion area. The specific steps are as follows:

[0017] a. Extract the centerline from the arterial model visualization in the arterial phase of the CT scan and represent it as a series of seed points located on consecutive 2D planes to adapt the 3D kidney model to the algorithm;

[0018] b. Each renal parenchymal voxel is associated with a centerline point based on a proximity criterion, i.e., using the mathematical Euclidean distance from a seed point on the arterial centerline to define each perfusion region; the algorithm assigns each individual voxel to the specific arterial branch with the highest probability of supplying that voxel, thereby defining a specific perfusion region;

[0019] Assume B i ∈B, (i∈R) is a branch of the renal artery, and the Voronoi cell C is defined i function:

[0020] C i ={x∈X V |d(x,B i )≤d(x,B j )}i≠j;

[0021] Among them, the image voxel x is located in the extracted kidney region X V Internal; d(x,B j ) is the point x to the blood vessel branch B j The minimum Euclidean distance, d(x,B i ) is the point x to the blood vessel branch B i The minimum Euclidean distance of

[0022] c. The data of all 2D planes and their relative centerline points are continuously merged to generate a new generation of 3D kidney virtual model consisting of different vascularized areas. Each area is presented in a corresponding color and finally output in STL format.

[0023] Furthermore, in (4), before performing surgical simulation, the perfusion area of ​​the three-dimensional virtual kidney model obtained in step (3) is observed to predefine the renal pedicle management, taking into account the relationship between the position of the tumor in the kidney and the vascular perfusion area in contact with its surface, and planning to block the blood vessels supplying these areas to achieve bloodless resection limited to the tissue surrounding the tumor.

[0024] By adopting the above technology, compared with the prior art, the beneficial effects of the present invention are as follows:

[0025] The method of the present invention utilizes M3DICS three-dimensional reconstruction software to process the original image data of CT scans and obtain a corresponding digital three-dimensional kidney model; the three-dimensional kidney model is used in combination with a mathematical algorithm to segment the kidney parenchyma volume into its vascular perfusion areas, generating a new three-dimensional virtual kidney model with different perfusion areas; and based on the modeling results, the arterial branches that may need to be clamped when selective blocking is performed are recorded, and a selective blocking simulation is performed to formulate an accurate and benefit-maximizing treatment plan for the subject; the method can accurately display the kidney vascular irrigation area and provide a highly accurate optimal selective renal artery blocking position, thereby improving the success rate of intraoperative blocking of kidney tumor-related renal artery branches and minimizing kidney damage caused by ischemia. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A schematic diagram of the general process of making a three-dimensional virtual model of a kidney with different perfusion areas according to the present invention;

[0027] Figure 2 This is a flow chart of the novel three-dimensional virtual model-assisted partial nephrectomy artery occlusion strategy optimization method of the present invention;

[0028] Figure 3 is CT imaging data about a particular patient according to an exemplary embodiment;

[0029] Figure 4 is used Figure 3 An exemplary three-dimensional reconstructed model generated from imaging data;

[0030] Figure 5 is through Figure 4 The three-dimensional model of the renal vascular perfusion area was generated;

[0031] Figure 6 is the ratio of the vascular perfusion region volume to the adjacent tumor area according to an exemplary embodiment. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments of the specification. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0033] On the contrary, the present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention as defined by the claims. Further, in order to make the public have a better understanding of the present invention, some specific details are described in detail in the detailed description of the present invention below. Those skilled in the art can fully understand the present invention without the description of these details.

[0034] See also Figure 1 and Figure 2 The novel three-dimensional virtual model-assisted partial nephrectomy artery occlusion strategy optimization method comprises the following steps:

[0035] (1) Image acquisition: Obtain the CT scan image data of the subject from the preoperative abdominal enhanced CT examination image database, such as Figure 3 As shown, the image data is in DICOM format;

[0036] (2) High-precision 3D model rendering: The dedicated software M3DICS is used to perform 3D reconstruction of CT images in DICOM format. Figure 4 As shown in Figure 2, high-precision 3D reconstruction focuses on the renal vascular system (including arteries and veins), collecting system, renal shape, and tumor characteristics.

[0037] As shown below:

[0038] a. Dynamic region growing method was used to reconstruct the renal pedicle, tumor feeding artery, extrarenal artery and intrarenal artery

[0039] The course of the artery, up to the segmental artery;

[0040] b. The renal parenchyma is segmented using a selective threshold segmentation method to separate different voxels and group them according to their grayscale.

[0041] (3) Estimating the renal vascular perfusion area: Apply the Voronoi diagram software to the three-dimensional renal model in step (2) to calculate the vascular perfusion area. The specific steps are as follows:

[0042] Step 1: From the arterial model visualization in the arterial phase of the CT scan, the centerline is extracted and represented as a series of seed points located on consecutive 2D planes to fit the 3D kidney model to the algorithm.

[0043] Step 2: Each renal parenchymal voxel is associated with a centerline point (i.e., seed point) based on the proximity criterion, that is, using the mathematical Euclidean distance from a seed point on the arterial centerline to define each perfusion area, the algorithm assigns each individual voxel to the specific arterial branch with the highest probability of supplying the voxel, thereby defining a specific perfusion area.

[0044] Assume B i ∈B, (i∈R) is a branch of the renal artery, and the Voronoi cell C is defined i function:

[0045] C i ={x∈X V |d(x,B i )≤d(x,B j )}i≠j;

[0046] Among them, the image voxel x is located in the extracted kidney region X V Internal; d(x,B j ) is the point x to the blood vessel branch B j The minimum Euclidean distance, d(x,B i ) is the point x to the blood vessel branch B i The minimum Euclidean distance of

[0047] Step 3: Continuously merge the data of all 2D planes and their relative centerline points to generate a new generation of 3D kidney virtual model consisting of different vascularized areas, each of which is presented in a specific color ( Figure 5 ), and finally output in STL format.

[0048] (4) Planning the surgical plan: Before the operation, the doctor needs to observe the perfusion area of ​​the new 3D virtual kidney model obtained in step (3) to predefine the renal pedicle management, consider the relationship between the position of the tumor in the kidney and the vascular perfusion area in contact with its surface, and plan to block the blood vessels supplying these areas to achieve bloodless resection limited to the tissue around the tumor as much as possible.

[0049] For example, for surgical safety, Figure 5 The shaded area shows the 5 mm margin of the tumor. It can be seen that the light blue, yellow, green and purple areas are adjacent to the tumor, and at most 4 vascular branches are clamped. Figure 6 The study showed the volume ratio of the four regions (Vol) and the ratio of the adjacent area (Area) of the tumor. The light blue blood vessels were clamped to prevent bleeding, and the purple blood vessels were permanently cut off because they were the main vascular supply of tumor nutrition. During the intervention, the renal pedicle was managed according to the plan, and the strategy from selective clamping to comprehensive clamping was implemented.

[0050] (5) Preoperative simulation: Selective blockade simulation, ischemic area simulation and vascularization preservation area simulation based on the three-dimensional virtual model of the perfusion area are performed. The specific details are as follows:

[0051] a. Use first-person perspective technology and virtual surgical tools to simulate surgery;

[0052] b. Perform surgical simulation on the kidney model according to the principles set in the surgical planning stage, using methods such as simulating the surgical approach, selective blocking sequence, cutting route, and management of the renal artery, and alerting for reasonable handling methods for unplanned situations.

[0053] The present invention can be implemented in other specific embodiments within the scope of the idea and basic features of the present invention. Therefore, the above detailed description is only for illustration and does not limit the present invention at all levels. The scope of the present invention should be determined based on the reasonable interpretation of the additional invention claims, and all changes within the equivalent scope of the present invention belong to the scope of the present invention. The present invention is not limited to the embodiments described herein, but is given the widest scope that is the same as the principles and new features disclosed herein. In addition, an embodiment can be constructed by combining the invention claims that do not have a clear reference relationship in the invention claims, or by modifying the new invention claims after the application.

Claims

1. A novel three-dimensional virtual model-assisted partial nephrectomy artery occlusion strategy optimization method, characterized by: The following steps are involved: (1) Image acquisition: extracting the CT scan image data of the subject from the preoperative enhanced CT examination image database; (2) High-precision 3D model rendering: The extracted CT scan data were imported into the dedicated software M3DICS in DICOM format for segmentation and 3D kidney model reconstruction; (3) Estimating the renal vascular perfusion area: using the three-dimensional virtual model in step (2) in combination with corresponding mathematical tools, the vascular perfusion area of ​​the kidney is calculated to generate a three-dimensional virtual kidney model with different perfusion areas; (4) Planning the surgical plan: Observe the three-dimensional virtual kidney model in step (3), predefine the renal pedicle management, and record the number and level / sequence of arterial branches that need to be clamped for selective occlusion; (5) Preoperative simulation: selective blockade simulation based on the three-dimensional virtual model of the perfusion area, simulation of the ischemic area and the preserved vascularized area.

2. The novel three-dimensional virtual model-assisted partial nephrectomy artery occlusion strategy optimization method according to claim 1 is characterized in that: The CT image in step (2) is three-dimensionally reconstructed, including reconstruction of the renal vascular system, collecting system, renal shape and tumor characteristics, as follows: a. Use dynamic region growing to reconstruct the course of the renal pedicle, tumor feeding artery, extrarenal artery, and intrarenal artery, up to the segmental artery; b. The renal parenchyma is segmented using a selective threshold segmentation method to separate different voxels and group them according to their grayscale.

3. The novel three-dimensional virtual model-assisted partial nephrectomy artery occlusion strategy optimization method according to claim 2 is characterized in that: The Voronoi diagram software is applied to the three-dimensional kidney model in step (3) to calculate the vascular perfusion area. The specific steps are as follows: a. Extract the centerline from the arterial model visualization of the arterial phase of the CT scan and represent it as a series of seed points located on continuous two-dimensional planes; b. According to the proximity criterion, each renal parenchymal voxel is associated with a centerline point (i.e., a seed point), that is, a Voronoi diagram is constructed based on the principle of mathematical Euclidean distance. Each voxel is divided into the Voronoi unit corresponding to its nearest seed point, and each voxel is assigned to a perfusion area by comparing the distance from each voxel to all seed points; Assume B i ∈B, (i∈R) is a branch of the renal artery, and the Voronoi cell C is defined i function; C i ={x∈X V |d(x,B i )≤d(x,B j )}i≠j; Among them, the image voxel x is located in the extracted kidney region X V Internal; d(x,B j ) is the point x to the blood vessel branch B j The minimum Euclidean distance, d(x,B i ) is the point x to the blood vessel branch B i The minimum Euclidean distance of c. The data of all 2D planes and their relative centerline points are continuously merged to generate a new generation of 3D kidney virtual model consisting of different vascularized areas. Each area is presented in a corresponding color and finally output in STL format.

4. The novel three-dimensional virtual model-assisted partial nephrectomy artery occlusion strategy optimization method according to claim 1 is characterized in that: In (4), before performing surgical simulation, the renal pedicle management is predefined by observing the perfusion area of ​​the three-dimensional virtual kidney model obtained in step (3), considering the relationship between the position of the tumor in the kidney and the vascular perfusion area in contact with its surface, and planning to block the blood vessels supplying these areas to achieve bloodless resection limited to the tissue surrounding the tumor.

Images