System and method for reconstructing an anatomical structure model - Patents.com

Abstract

The heart valve model includes a parametric leaflet model representing a geometry of the heart valve constructed based on at least one landmark point associated with the heart valve. The heart valve model includes a ray-cast anatomical structure constructed based on an automatic aortic root reconstruction algorithm to extract a shape of the aorta. The heart valve also includes an additional anatomical structure, and the parametric leaflet model, the ray-cast anatomical structure, and the additional anatomical structure are combined to obtain the heart valve model.

Description

[Technical field]

[0001] This application claims priority to U.S. Provisional Patent Application No. 63 / 187,052, filed May 11, 2021, which is incorporated by reference in its entirety. [Background technology]

[0002] In recent years, transcatheter aortic valve replacement (TAVR) has become more popular as an alternative approach to highly invasive surgical aortic valve replacement (SAVR). Unlike SAVR, whose size and implantation location can be precisely determined during surgery, TAVR relies on a more thorough preoperative evaluation to determine the valve type, size, and deployment strategy with the aid of advanced medical imaging techniques. Suboptimal valve deployment has been associated with various adverse effects, including elevated transvalvular pressure gradients, paravalvular leakage, aortic root rupture during implantation, coronary artery occlusion, etc. Summary of the Invention [Problem to be solved by the invention]

[0003] To minimize these adverse effects, pre-TAVR evaluation may involve accurately reconstructing the patient's anatomy, generating a computational mesh, and using it for Finite Element Analysis (FEA) or / and Computational Fluid Dynamics (CFD) simulations. These modern approaches based on patient-specific geometry have proven to have a high predictive ability for TAVR outcomes and prevent the occurrence of severe adverse events in high-risk patients. As TAVR technology evolves, it is gradually being extended to low-risk patients. Since the long-term durability of the treatment is a major concern, a comprehensive pre-operative evaluation becomes more important for this cohort. The rate of adverse events is inversely correlated with durability. However, performing these evaluations requires the reconstruction of the patient's anatomy from CT or MRI data, which is a tedious and time-consuming manual task (3–6 hours per case). Subsequent in silico studies require even more expertise in solid / fluid mechanics and numerical methods, preventing them from being routinely performed in a typical clinical environment. [Means for solving the problem]

[0004] Exemplary systems and methods are disclosed that can be used to (1) generate a parametric heart valve model (e.g., an aortic valve model or a mitral valve model) using a few manually selected landmarks, (2) rapidly and automatically reconstruct the geometry of the patient's aorta and aortic valve cusps (e.g., for clinical pre-TAVR evaluation or other pre-procedure planning evaluation), and (3) output a readily usable computational mesh (e.g., for pre-TAVR in silico studies such as evaluation for coronary artery occlusion, root rupture, and prediction of valve hemodynamics). The exemplary systems and methods may be used for patient-specific computational or 3D printing evaluations that require input of the patient's geometry.

[0005] In some aspects, the exemplary systems and methods are configured to provide an automated method to reconstruct patient-specific geometry with only minor manual input (approximately 5-10 minutes per case) and generate a mesh for in silico studies. To start mesh generation from CT data, the user selects several landmark points related to the aortic valve. A parametric valve leaflet model is used to represent the valve geometry, and then an automatic aortic root reconstruction algorithm is used that can extract the shape of the aorta. Then, the coronary arteries and calcification deposits are automatically obtained using a region growing algorithm. Finally, the exemplary systems and methods combine all mesh components to generate a mesh for computational studies. Visual inspection and possibly manual cleanup are performed to ensure accuracy and quality of the mesh.

[0006] In one aspect, the heart valve model includes a parametric leaflet model representing a geometry of the heart valve constructed based on at least one landmark point associated with the heart valve. The heart valve model includes a ray-cast anatomical structure constructed based on an automatic aortic root reconstruction algorithm to extract a shape of the aorta. The heart valve also includes an additional anatomical structure, and the parametric leaflet model, the ray-cast anatomical structure, and the additional anatomical structure are combined to obtain the heart valve model.

[0007] In one embodiment, the heart valve model includes multi-point ray casting of the aorta.

[0008] In one aspect, the anatomical structure model includes a first anatomical structure, the first anatomical structure including a parametric valve leaflet model representing the first anatomical structure constructed based on at least one landmark point associated with the first anatomical structure. The anatomical structure model includes a second anatomical structure, the second anatomical structure including a ray casted second anatomical structure constructed based on an automatic anatomical structure reconstruction algorithm to extract a geometry of the second anatomical structure. The anatomical structure model also includes a third anatomical structure. The first, second, and third anatomical structures are combined to generate the anatomical structure model. [Brief description of the drawings]

[0009] [Figure 1] Flow chart of the semi-automated patient-specific mesh generation process. [Figure 2A] An example of the automated segmentation process showing the locations of the 13 landmark points. [Figure 2B] FIG. 11 is an example of the automated segmentation process, showing a parametric valve leaflet model with landmarks. [Figure 2C] An example of the automated segmentation process, showing the aortic root with parametric valve leaflets. [Figure 2D] An example of the automated segmentation process, showing coronary arteries and calcification deposits. [Figure 2E] An example of the automated segmentation process, showing the final mesh output. [Figure 3A] FIG. 13 is an example of the parametric leaflet generation process, showing a skeleton that uses a second order polynomial to connect landmarks. [Figure 3B] FIG. 13 is an illustration of the parametric leaflet generation process showing multiple second order polynomials fitted to represent the leaflet surface. [Figure 3C] FIG. 13 is an illustration of the parametric leaflet generation process showing the final triangle mesh. [Figure 4A]FIG. 1 illustrates a sample parametric leaflet showing one of three individual leaflets. [Figure 4B] FIG. 1 illustrates a sample parametric leaflet showing one of three individual leaflets. [Figure 4C] FIG. 1 illustrates a sample parametric leaflet showing one of three individual leaflets. [Figure 4D] FIG. 1 illustrates a sample parametric valve leaflet showing a complete aortic valve. [Figure 4E] FIG. 1 illustrates a sample parametric valve leaflet showing a complete aortic valve. [Figure 4F] FIG. 1 illustrates a sample parametric valve leaflet showing a complete aortic valve. [Figure 5A] FIG. 13 shows a sample comparison between the parametric leaflets (light grey) and the original image slices. [Figure 5B] FIG. 13 shows a sample comparison between the parametric leaflets (light grey) and the original image slices. [Figure 5C] FIG. 13 shows a sample comparison between the parametric leaflets (light grey) and the original image slices. [Figure 6A] An example diagram of the aortic reconstruction process, showing the aortic wall detected using a single-point ray-casting algorithm. [Figure 6B] An example of the aortic reconstruction process showing the aortic wall detected using a multi-point ray casting algorithm. The edge detection accuracy has improved dramatically compared to Figure 6A. [Figure 6C] FIG. 13 is an example diagram of the aortic reconstruction process showing sample aortic wall detection results. [Figure 6D] FIG. 13 is an illustration of the aortic reconstruction process showing the final aortic model and mesh. [Figure 7A] A comparison between the results of automatic segmentation (left: FIG. 7A) and manual segmentation (right: FIG. 7B). [Figure 7B]A comparison between the results of automatic segmentation (left: FIG. 7A) and manual segmentation (right: FIG. 7B). [Figure 7C] A comparison between the results of automatic segmentation (left: FIG. 7C) and manual segmentation (right: FIG. 7D). [Figure 7D] A comparison between the results of automatic segmentation (left: FIG. 7C) and manual segmentation (right: FIG. 7D). [Figure 8] 8 illustrates an exemplary computer architecture of a computer system 200 capable of executing software components that can use the output of the exemplary methods described herein. The computer architecture illustrated in FIG. 8 illustrates an exemplary computer system configuration in which computer 200 may be utilized to execute any aspect of the components and / or modules presented herein that are described as executing on an analysis system or any component in communication therewith. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] A flow chart of the semi-automatic mesh generation process is shown in Figure 1. The algorithm consists of three main functions: (1) a parametric valve leaflet model, (2) a method for accurately determining the aortic root / aorta geometry, and (3) a method for assembling the different components and outputting the final mesh for computational simulation.

[0011] In FIG. 1, to initiate mesh generation from scanned data (e.g., CT data), a user (e.g., a physician or technician) picks several landmark points related to the aortic valve. The exemplary system and method employs a parametric valve leaflet model to represent the valve geometry, which is then processed by an automatic aortic root reconstruction algorithm that can extract the shape of the aorta. The exemplary system and method is then configured to automatically determine the coronary arteries and calcification deposits using a region growing algorithm. Finally, the exemplary system and method combines all mesh components to generate a mesh for computational studies. Visual inspection and manual cleanup of surfaces are performed to ensure mesh accuracy and quality.

[0012] Since it is difficult to identify the boundary of the aortic valve in a CT scan, especially when there are few calcification deposits on the leaflets, a common method for manual segmentation starts by segmenting the blood volume on the ventricular side in diastole. The leaflets are then generated by extruding the border of the blood pool. In an exemplary automated segmentation process (Figures 2A-E), the method starts by selecting (e.g., by the user) landmark points in a 3D cardiac CT scan in diastole when the aortic valve is closed, and the algorithm takes over to output the final mesh.

[0013] 2A-2E provide detailed illustrations of an exemplary automated segmentation process employed in the method of FIG. 1.

[0014] A total of 13 different landmarks are selected for the parametric leaflet model, as shown in FIG. 2A. These points are P0, which is the central junction point of the three leaflets; P 1~3 and P 4~6are the six commissure points. Two sets of these points are intended to resolve the finite coaptation height between the leaflets at the commissure. For example, P1 and P4 are at the lower and upper ends of this finite commissure coaptation line (Figure 2B). 1~9 are the three surface points on the valve leaflet, P 10~12 are the three leaflet hinges. The landmarks may be curated by a clinician or obtained by artificial intelligence (AI) and / or machine learning approaches. The number and locations of the landmarks are not limited to those described.

[0015] Figure 2B shows the distribution of landmark points in the parametric aortic valve model. With these landmarks and the guide of the parametric valve leaflet model, the aortic root geometry is constructed in a slice-by-slice manner using the multi-point ray-casting method described below.

[0016] Figure 2C shows the reconstructed aortic root geometry with the parametric valve leaflets attached. Unlike many other parametric models, to ensure the quality of our output mesh, the leaflets are stitched to the aortic root to form a watertight connection. Two extra landmarks are required at the coronary ostia to automatically segment the coronary arteries by intensity-based region growing. Similarly, the calcifications in the aortic root region are obtained by the same intensity-based segmentation method.

[0017] Figure 2D shows the geometry of the coronary arteries and calcifications. The final step is to assemble the mesh to constitute the computational mesh used in the FEA / CFD simulations. As the parametric leaflets and the aortic root are all surface meshes, specific thickness values ​​are assigned to them based on physiological data. The three leaflets are separated by a finite distance corresponding to the leaflet thickness to ensure that they open during the numerical simulations. The final mesh is shown in Figure 2E.

[0018] Parametric valve leaflet model A parametric leaflet model may be constructed based on the selected landmarks (e.g., based on the 13 landmark points in the example described above). There may be seven points on each leaflet surface (one center point, four commissure points, one point on the surface, and one hinge point). Figures 3A-3C show an example parametric leaflet generation process of the method of Figure 1.

[0019] First, a leaflet skeleton is generated, in some embodiments using a quadratic polynomial that connects these points (FIG. 3A). Then, multiple quadratic polynomials are fitted to the skeleton to create a surface (FIG. 3B). Finally, a triangular mesh is used to represent the leaflets (FIG. 3C). Sample leaflet models are shown in FIGS. 4A-4F. The method of generating the leaflet surface is not limited to the polynomial fitting method described here. Other surface fitting methods, such as fitting with a non-uniform Rational B-spline (NURBS) surface, can be used with little or no modification to the procedure described here.

[0020] Figures 4A-4F show an example parametric leaflet model that may be generated with the method of Figure 1. Figures 4A-4C show three individual leaflet meshes, and the corresponding aortic valve model is shown in Figures 4D-4E. As demonstrated in Figures 4A-4C, the parametric leaflet model can handle complex surface geometries and capture the irregular shapes of the individual leaflets.

[0021] Comparing the reconstructed leaflets with the 3D CT images (Figures 5A-C) we see that the parametric model (light grey) captures the original leaflet geometry very well. Some deviation from the original geometry is inevitable, but future improved algorithms will be able to locally deform the leaflet mesh to better match the original geometry.

[0022] Aortic Root / Aorta Extraction Algorithm Extraction of the aortic root / aortic geometry may be based on an intensity-based, slice-wise, multi-point ray-casting edge detection algorithm (FIGS. 6A-6D). FIGS. 6A-6D show an example aortic reconstruction process of the method of FIG.

[0023] After constructing the parametric leaflet model, the aortic annulus can be easily defined by the three leaflet hinge points. A slice parallel to the aortic annulus is extracted from the original 3D CT data. Starting from a point in the blood region, a ray is cast in all directions, and the intensity variation along the ray can be extracted. The sudden jump from high intensity (blood) to low intensity (tissue) indicates the aortic wall (Figure 6A). The contour of the aortic root in this slice is generated by connecting all the detected edge points. However, during this edge detection process, errors may occur due to noise or low image quality (Figure 6A). To address this issue, we introduce a modified method of casting rays from a set of origins. Since most rays from different origins should converge at a certain point on the real aortic wall, if there is a detection error from one ray, it is unlikely that rays from other origins will repeat the same error. Thus, by comparing the detected edges from all origins, we can filter out points that are unclear if they are too far from the average location. The average of the remaining points should accurately determine the aortic wall (Figure 6B). If the points detected by different rays are too scattered at a certain location, this point is discarded. Figures 6A and 6B clearly show that the multi-point approach can accurately capture the aortic wall. Finally, a cubic spline is fitted to all the detected points. In the sample slice shown in Figure 3C, the algorithm demonstrates its ability to accurately capture the aortic wall from the left ventricular outflow graft to the ascending aorta. Note that the origin of the rays is different in every slice (Figure 3C). To improve the accuracy of the edge detection, a core of five points and four satellite points derived from the borders of neighboring slices are used. The contours from all slices are stacked on top of each other to represent the aortic wall (Figure 6D). The aortic geometry is smoothed by fitting a sixth order polynomial to the points along the same longitude in the assembly process. The fitting method is not limited to polynomials. Other methods such as B-spline fitting can also be used.A triangular mesh is used to construct this geometry (Figure 6D), although other surface mesh types such as quadrilateral or pentagonal meshes can also be used.

[0024] Segmentation of coronary arteries and calcification deposits Both coronary artery and calcification deposit geometries are segmented by an intensity-based region-growing algorithm. The coordinates of the two coronary artery ostia have to be specified (human input) as starting points for the region-growing algorithm. For calcifications, as proposed, a global threshold of 850 Hounsfield Units (HU) is applied, followed by a cleaning process that discards calcification clusters smaller than 20 voxels or located outside the aorta.

[0025] Final Mesh Assembly The final mesh assembly process is based on Boolean operations in intensity space. First, the surface meshes of the aorta and valve leaflets are thickened and voxelized based on physiological values. Then, using Boolean operations, the coronary arteries and calcification deposits are added to the regions. To prevent the leaflets from fusing, slits with a width equal to the leaflet thickness are placed between them. Finally, a marching cubes algorithm converts the voxel data into an STL mesh. The assembled mesh is visually inspected and any mesh problems are corrected.

[0026] Comparison with manual segmentation results To ensure the accuracy of the method, a comparison with manual segmentation results was performed (FIGS. 7A-7D), which show a comparison between an example output of the automatic segmentation operation and an example output of the manual segmentation operation of the method of FIG.

[0027] In the aorta, the automatically generated geometry (Figure 7A) was quite similar to that from the manual segmentation process (Figure 7B). However, the manual result had more fine details due to a lower degree of smoothness. In the en face views between the models (Figures 7C and 7D), they showed similar aortic leaflet geometry and calcification distribution. However, the parametric leaflet model cannot resolve the curved coaptation line between the leaflets because only three landmark points were used for this line. If additional landmark points were added along the coaptation line or a local deformation algorithm was applied, the leaflet edge would be captured more accurately.

[0028] Although the exemplary systems and methods have been described in relation to the aortic valve and TAVR, the exemplary systems and methods are readily applicable to the mitral valve and in other structural heart pre-procedure planning.

[0029] The exemplary systems and methods may be incorporated into routine clinical practice for TAVR and other structural heart pre-procedure assessment. The exemplary systems and methods may be incorporated into software such as Materialise Mimics for users to configure patient-specific geometries with minimal user input (i.e., user-selected landmarks).

[0030] Consideration To address the issue of time-consuming segmentation process when the valve cusps are barely visible in medical images, parametric heart valve models have been used for segmentation. Ionasec, R. et al. used parametric aortic and valve cusp models to reconstruct both aortic and mitral valves from 4D cardiac CT and TEE. A learning-based algorithm was applied to the 4D images to identify and track landmarks throughout the cardiac cycle. Pouch AM et al. implemented a medial representation method to model the mitral valve from 3D echo images. They succeeded in capturing the thickness of the mitral valve and its deformation during the cardiac cycle. This method has been applied to reconstruct the anatomy of the aortic valve from 3D echo data (Pouch 2015). However, the authors did not mention whether the method is also applicable to CT data, since the aortic valve cusps have much higher contrast in echo images. Lalys, F. et al. used a centerline detection approach to reconstruct the aorta, and then used a registration algorithm to detect landmarks associated with the aortic valve. These landmarks were then used to predict TAVR outcomes. Hosny, A. et al. used a similar approach to that described herein to generate parametric leaflet models for 3D printing. After printing the manually segmented aorta and calcifications together, the models were constructed for the purpose of valve sizing and pre-TAVR evaluation. However, because 3D printing generally has lower mesh quality requirements, these meshes could not be directly used in computational simulations. Unlike the studies mentioned above, the exemplary system and method can generate computational meshes directly from 3D CT data with only minor manual input.

[0031] It should be appreciated that the logical operations described above may be implemented as (1) a sequence of computer-implemented operations or program modules executed on a computing system, and / or (2) as interconnected machine logic circuits or circuit modules within a computing system. The implementation is a matter of choice dependent on the performance and other requirements of the computing system. Accordingly, the logical operations described herein are referred to variously as state operations, operations, or modules. These operations, operations, and / or modules may be implemented in software, firmware, special purpose digital logic, hardware, and any combination thereof. It should also be appreciated that more or fewer operations may be performed than illustrated in the figures and described herein. These operations may also be performed in different orders than described herein.

[0032] 8 illustrates an example computer architecture of a computer system 200 capable of executing software components that can use the output of the example methods described herein. The computer architecture illustrated in FIG. 8 illustrates an example computer system configuration in which computer 200 may be utilized to execute any aspect of the components and / or modules presented herein that are described as executing on an analysis system or any components in communication therewith.

[0033] In one aspect, computing device 200 may comprise two or more computers in communication with each other that cooperate to perform a task. For example, but not limited to, an application may be divided in a manner that allows for simultaneous and / or parallel processing of instructions of the application. Alternatively, data processed by an application may be divided in a manner that allows for simultaneous and / or parallel processing of different portions of the data set by two or more computers. In one aspect, virtualization software may be employed by computing device 200 to provide the functionality of multiple servers that are not directly bound to the number of computers in computing device 200. For example, the virtualization software may provide 20 virtual servers on four physical computers. In one aspect, the functionality disclosed above may be provided by running an application and / or multiple applications in a cloud computing environment. Cloud computing may include providing computing services over a network connection using dynamically scalable computing resources. Cloud computing may be supported at least in part by virtualization software. Cloud computing environments may be established by enterprises and / or rented as needed from third party providers. Some cloud computing environments may include cloud computing resources that are owned and operated by the enterprise, as well as cloud computing resources rented and / or leased from third-party providers.

[0034] In its most basic configuration, computing device 200 typically includes at least one processing unit 220 and system memory 230. Depending on the exact configuration and type of computing device, system memory 230 may be volatile (such as random-access memory (RAM)), non-volatile (such as read-only memory (ROM), flash memory, etc.), or some combination of the two.

[0035] This most basic configuration is illustrated in FIG. 8 by dashed line 210. Processing unit 220 may be a standard programmable processor that performs arithmetic and logical operations necessary for the operation of computing device 200. Although one processing unit 220 is shown, there may be multiple processors. As used herein, processing unit and processor refer to physical hardware devices that execute coded instructions to perform functions on inputs and create outputs, including, for example, but not limited to, microprocessors (MCUs), microcontrollers, graphical processing units (GPUs), and application specific circuits (ASICs). Thus, although instructions may be described as being executed by one processor, the instructions may be executed simultaneously, sequentially, or alternatively by one or more processors. Computing device 200 may also include a bus or other communication mechanism for communicating information between various components of computing device 200.

[0036] Computing device 200 may have additional features / functionality. For example, computing device 200 may include additional storage, such as removable storage 240 and non-removable storage 250, including, but not limited to, magnetic or optical disks or tape. Computing device 200 may also include network connections 280 that enable the device to communicate with other devices, such as via communication paths described herein. Network connection 280 may take the form of a modem, a modem bank, an Ethernet card, a universal serial bus (USB) interface card, a serial interface, a token ring card, a fiber distributed data interface (FDDI) card, a wireless local area network (WLAN) card, a wireless transceiver card, such as a code division multiple access (CDMA), global system for mobile communications (GSM), long-term evolution (LTE), worldwide interoperability for microwave access (WiMAX) and / or other air interface protocol wireless transceiver card, and other well known network devices. Computing device 200 may have input devices 270, such as a keyboard, keypad, switches, dials, a mouse, a track ball, a touch screen, a voice recognition device, a card reader, a paper tape reader, or other well-known input devices.Also included may be output devices 260 such as a printer, video monitor, liquid crystal display (LCD), touch screen display, display, speakers, etc. Additional devices may be connected to the bus to facilitate communication of data between the components of computing device 200. All of these devices are well known in the art and need not be described at length here.

[0037] The processing unit 220 may be configured to execute program code encoded in a tangible computer-readable medium. A tangible computer-readable medium refers to any medium capable of providing data that causes the computing device 200 (i.e., a machine) to operate in a specific manner. A variety of computer-readable media may be utilized to provide instructions to the processing unit 220 for execution. Exemplary tangible computer-readable media may include, but are not limited to, volatile, non-volatile, removable, and non-removable media implemented in any method or technology for storage of information, such as computer-readable instructions, data structures, program modules, or other data. System memory 230, removable storage 240, and non-removable storage 250 are all examples of tangible computer storage media. Example tangible computer-readable recording media include, but are not limited to, integrated circuits (e.g., field programmable gate arrays, or application specific ICs), hard disks, optical disks, magneto-optical disks, floppy disks, magnetic tape, holographic storage media, solid state devices, RAM, ROM, electrically erasable program read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices.

[0038] In view of the above, it should be appreciated that many types of physical transformations may take place in computer architecture 200 to store and execute the software components presented herein. It should also be appreciated that computer architecture 200 may include other types of computing devices, including handheld computers, embedded computer systems, personal digital assistants, and other types of computing devices known to those of skill in the art. It is also contemplated that computer architecture 200 may not include all of the components shown in FIG. 8, may include other components not explicitly shown in FIG. 8, or may utilize a different architecture than that shown in FIG. 8.

[0039] In the illustrated embodiment, processing unit 220 may execute program code stored in system memory 230. For example, a bus may carry data to system memory 230, from which processing unit 220 receives and executes instructions. Data received by system memory 230 may optionally be stored on removable storage 240 or non-removable storage 250 before or after execution by processing unit 220.

[0040] It is to be understood that the various techniques described herein may be implemented in connection with hardware or software or, where appropriate, a combination thereof. Thus, the methods and apparatus of the presently disclosed subject matter, or certain aspects or portions thereof, may take the form of program code (i.e., instructions) embodied in a tangible medium, such as a floppy diskette, CD-ROM, hard drive, or any other machine-readable storage medium, which when loaded into and executed by a machine, such as a computing device, causes the machine to become an apparatus for practicing the presently disclosed subject matter. In the case of program code execution on a programmable computer, the computing device generally includes a processor, a processor-readable storage medium (including volatile and non-volatile memory and / or storage elements), at least one input device, and at least one output device. One or more programs may perform or utilize the processes described in connection with the presently disclosed subject matter, for example, through application programming interfaces (APIs), reusable controls, or the like. Such programs may be implemented in a high-level procedural or object-oriented programming language to communicate with a computer system. However, the programs may be implemented in assembly or machine language, if desired. In any case, the language may be a compiled or interpreted language, and combined with hardware implementations.

[0041] Further, the various components may be in communication via wireless and / or wired or other desired available communication means, systems, and hardware. Further, the various components and modules may be replaced by other modules or components providing similar functionality.

[0042] The computer architecture 200 includes the necessary software and / or hardware components and modules to enable the functionality of the modeling, simulations and methods disclosed in this disclosure. In some embodiments, the computer architecture 200 may include artificial intelligence (AI) modules or algorithms and / or machine learning (ML) modules or algorithms (e.g., stored in the system memory 230, the removable storage 240, the non-removable storage 250, and / or a cloud database). The AI ​​and / or ML modules / algorithms may enhance the predictive power of the models, simulations, and / or methods disclosed in this disclosure. For example, by using deep learning, AI and / or ML model training, including patient information and any relevant input data to the computational model, the predictive power of the computational model may be significantly improved. The AI ​​and / or MI modules / algorithms also help improve the sensitivity and specificity of predictions as the database grows. In some aspects, the computer architecture 200 may include virtual reality (VR), augmented reality (AR), and / or mixed reality displays, headsets, glasses, or any other suitable display devices as part of the output devices 260 and / or input devices 270. In some aspects, the display devices may be interactive to allow a user to select from options including with or without AR, with or without VR, or fusion with real-time clinical imaging to help the clinician interact and make decisions.

[0043] Although illustrative aspects of the disclosure are described in detail in certain instances herein, it should be understood that other aspects are contemplated. Thus, the disclosure is not intended to be limited in scope to the details of construction and the arrangement of components set forth in the following detailed description or illustrated in the drawings. The disclosure is capable of other aspects and of being practiced or carried out in various ways.

[0044] As used in this specification and the appended claims, the singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise. Ranges may be expressed herein as from "about" or "approximately" one particular value and / or to "about" or "approximately" another particular value. When such a range is expressed, another exemplary embodiment includes from the one particular value and / or to the other particular value.

[0045] "Comprising," "containing," or "including" means that the named compounds, elements, particles, or method steps are present in a composition, article, or method, but do not exclude the presence of other such compounds, materials, particles, or method steps, even if those other compounds, materials, particles, or method steps have the same function as the one named.

[0046] In describing the exemplary embodiments, technical terms are used for clarity. Each term is intended to have the broadest meaning as understood by those skilled in the art, and is intended to include all technical equivalents that operate in a similar manner to achieve a similar purpose. It should also be understood that the reference to one or more steps of a method does not preclude the presence of additional or intervening method steps between those steps that are explicitly identified. The steps of the method may be performed in a different order than described herein without departing from the scope of the present disclosure. Similarly, it should also be understood that the reference to one or more components in a device or system does not preclude the presence of additional or intervening components between those components that are explicitly identified. As used herein, a "subject" may be any applicable human, animal, or other organism, living or dead, or other biological or molecular structure or chemical environment, and may relate to a particular component of the subject, such as a particular tissue or fluid of the subject (e.g., human tissue in a particular area of ​​the body of a living subject), which may be a particular location of the subject, referred to herein as an "area of ​​interest" or "region of interest."

[0047] As described herein, it is to be appreciated that the subject may be a human or any animal. It is to be appreciated that the animal may be of any of a wide variety of applicable types, including, but not limited to, mammals, veterinary animals, livestock animals or pet-type animals, etc. By way of illustration, the animal may be a laboratory animal (e.g., rats, dogs, pigs, monkeys) specifically selected to have certain characteristics similar to humans, etc. It is to be appreciated, for example, that the subject may be any applicable human patient.

[0048] The term "about" as used herein means approximately, in the region of, roughly, or around. When the term "about" is used in conjunction with a numerical range, it modifies the range by extending the boundaries above and below the numerical values ​​set forth. In general, the term "about" is used herein to modify a numerical value above and below the stated value with a variance of 10%. In one embodiment, the term "about" means plus or minus 10% of the numerical value of the number with which it is used. Thus, about 50% means within a range of 45% to 55%. Numerical ranges described herein by endpoints include all numbers and decimals subsumed within that range (e.g., 1 to 5 includes 1, 1.5, 2, 2.75, 3, 3.90, 4, 4.24, and 5).

[0049] Similarly, numerical ranges recited herein by endpoints include the subranges subsumed within that range (e.g., 1 to 5 includes 1 to 1.5, 1.5 to 2, 2 to 2.75, 2.75 to 3, 3 to 3.90, 3.90 to 4, 4 to 4.24, 4.24 to 5, 2 to 5, 3 to 5, 1 to 4, and 2 to 4). It is also to be understood that all numbers and decimals thereof are intended to be modified by the term "about."

[0050] The exemplary system and method can significantly reduce the manual effort required to build a patient-specific model from hours to only a few minutes. It will be apparent to those skilled in the art that various modifications and variations can be made in the present disclosure without departing from the scope or spirit of the invention. Other aspects of the present disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the methods disclosed herein. It is intended that the specification and examples be considered as exemplary only, with the true scope and spirit of the invention being indicated by the following claims.

Claims

1. A method for reconstructing a patient-specific heart valve model from medical image data of the patient, comprising: constructing a parametric leaflet model representing a geometry of the heart valve based on at least one landmark point from the medical image data associated with the heart valve; constructing a ray-cast anatomical structure based on an automatic aortic root reconstruction algorithm to extract the shape of the patient's aorta; Constructing additional anatomical structures; and combining the parametric leaflet model, the ray cast anatomical structure, and the additional anatomical structure to obtain the patient-specific heart valve model; A method comprising:

2. The method of claim 1 , wherein the at least one landmark point is any identifiable anatomical feature.

3. The method of claim 1 , wherein the at least one landmark point is obtained based on a CT scan.

4. The method of claim 1 , wherein the at least one landmark point is based on an MRI scan.

5. The method of claim 1 , wherein the at least one landmark point is based on an ultrasound scan.

6. The method of claim 1 , wherein the at least one landmark point is patient-specific.

7. The method of claim 1 , wherein the automatic aortic root reconstruction algorithm comprises an intensity-based, slice-wise, multi-point ray casting edge detection algorithm.

8. The method of claim 7 , wherein the automatic aortic root reconstruction algorithm compares detected edges from all origins to capture the edges of the aortic wall.

9. The method of claim 1 , wherein the additional anatomical structures include the geometry of coronary arteries and calcification deposits.

10. The method of claim 9 , wherein the geometry of the coronary arteries and the calcification deposits are segmented by an intensity-based region growing algorithm.

11. The method of claim 1 , wherein the parametric leaflet model, the ray casted anatomical structures, and the additional anatomical structures are combined to obtain a geometric mesh file suitable for computational modeling.

12. Heart valve model, including multi-point ray casting of the aorta.

13. 13. The heart valve model of claim 12, further comprising a combination of the multi-point ray casting and a parametric leaflet model representing a geometry of the heart valve.

14. 13. The heart valve model of claim 12, further comprising geometries of calcification deposits and coronary arteries.

15. 15. The heart valve model of claim 14, wherein the geometry of the calcification deposits and the coronary arteries are segmented by an intensity-based region growing algorithm.

16. A method for reconstructing a patient-specific anatomical structure model from medical image data, comprising: constructing a first anatomical structure, the first anatomical structure including a parametric valve leaflet model representing the first anatomical structure based on at least one landmark point from the medical image data associated with the first anatomical structure; constructing a second anatomical structure from the medical image data, the second anatomical structure including a ray casted second anatomical structure constructed based on an automatic anatomical structure reconstruction algorithm for extracting a geometry of the second anatomical structure; constructing a third anatomical structure; and combining the first, second, and third anatomical structures to generate the patient-specific anatomical structure model; A method comprising:

17. The method of claim 16 , wherein the at least one landmark point is obtained using a CT scan, an MRI scan, or an ultrasound scan.

18. The method of claim 16 , wherein the at least one landmark point is patient-specific.

19. The method of claim 16 , wherein the parametric leaflet model is based on at least seven or more landmark points.

20. The method of claim 16 , wherein the third anatomical structure comprises a coronary artery or a calcified deposit.

Images