Providing biomechanical plaque data for interventional device simulation systems

Abstract

A computer-implemented method of providing biomechanical plaque data (110) for an interventional device simulation system (100) includes: extracting plaque data from spectral CT data, the plaque data representing a spatial distribution of plaques (140) within a vascular region (130); converting the plaque data into biomechanical plaque data (110) representing a spatial distribution of mechanical constraints applied to an interventional device (150) of the interventional device simulation system (100) in response to contact between the interventional device (150) and the plaque (140); and outputting the biomechanical plaque data (110).

Description

Technical Field

[0001] The present disclosure relates to providing biomechanical plaque data for an interventional device simulation system. Computer-implemented methods, computer program products, and systems are disclosed. Background Art

[0002] Interventional device simulation systems are becoming increasingly important in familiarizing physicians with the use of interventional devices. Simulation systems provide a virtual experience of interventional procedures that allows physicians to gain experience with new intravascular procedures in a realistic and low-risk environment. For example, in the vascular field, simulation systems have been developed and integrated into real catheterization laboratories. Such systems are also used in planning patient-specific interventions.

[0003] An example of an interventional device simulation system in the vascular field is the Vist G5 intravascular simulator sold by Mentice AB of Gothenburg, Sweden. Another example of an interventional device simulation system in the vascular field is the Angio Mentor system of Simbionix Ltd. of Israel (which is sold by Surgical Science Sweden AB of Gothenburg, Sweden). This system provides visual feedback and tactile feedback, realistically mimicking the look and feel of actual intravascular intervention. Many different types of intravascular processes can be simulated using this system. Different types of interventional devices can be used to simulate in different types of patient anatomical structures. For example, navigation processes such as catheter insertion can be simulated. Treatment processes such as atherectomy processes can also be simulated.

[0004] An important factor in providing a realistic simulation of the vasculature is the accurate representation of vascular plaque. Vascular plaque is formed by substances such as fat, cholesterol, calcium, fibrin, and other substances that accumulate on the walls of arteries. Plaque causes hardening and narrowing of arteries, restricting blood flow and oxygen supply to vital organs and increasing the risk of blood clots (which may block blood flow to the heart or brain). Therefore, the presence of vascular plaque affects interventional device navigation, vessel deformation, and treatment decisions. However, the correlation between the simulated process and the real process and any resulting benefits are only feasible when the properties of the plaque are accurately represented in the simulation. Summary of the invention

[0005] According to one aspect of the present disclosure, a computer-implemented method for providing biomechanical plaque data to an interventional device simulation system is provided. The method comprises:

[0006] receiving spectral CT data representing a blood vessel region, the spectral CT data defining X-ray attenuation in a plurality of different energy intervals within the blood vessel region;

[0007] extracting plaque data from the spectral CT data, the plaque data representing a spatial distribution of plaque within the vascular region;

[0008] converting the plaque data into biomechanical plaque data representing a spatial distribution of mechanical constraints applied to the interventional device in response to contact between an interventional device of the interventional device simulation system and the plaque; and

[0009] The biomechanical plaque data is output.

[0010] The spectral CT data defines X-ray attenuation in multiple different energy intervals. By processing data from multiple different energy intervals, it is possible to distinguish between multiple media that have similar X-ray attenuation values ​​when measured in a single energy interval and media that are indistinguishable in conventional X-ray attenuation data. Because in the above method, plaque data is extracted from spectral CT data, the plaque data provides a more accurate representation of the spatial distribution of plaques. In addition, because the plaque data is converted into biomechanical plaque data, which represents the spatial distribution of mechanical constraints applied to the interventional device in response to contact between the interventional device and the plaque of the interventional device simulation system, a more accurate simulation of the interaction between the interventional device and the plaque is provided.

[0011] Other aspects, features and advantages of the present disclosure will become apparent from the following description of examples with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 is an example of a graphical representation of a blood vessel region 130 according to some aspects of the present disclosure.

[0013] Figure 2 is a schematic diagram illustrating an example of the spatial distribution of plaque 140 within a blood vessel region, according to some aspects of the present disclosure.

[0014] Figure 3 is a flow chart illustrating an example of a computer-implemented method of providing biomechanical plaque data to an interventional device simulation system, according to aspects of the present disclosure.

[0015] Figure 4 is a schematic diagram illustrating an example of a system 200 for providing biomechanical plaque data to an interventional device simulation system, according to some aspects of the present disclosure.

[0016] Figure 5 is a schematic diagram illustrating an example of an interventional device simulation system 100 according to some aspects of the present disclosure.

[0017] Figure 6is a flow chart illustrating an example of a computer-implemented method of simulating navigation of an interventional device within a vascular region, according to some aspects of the present disclosure.

[0018] Figure 7 is a flow chart illustrating an example of a computer-implemented method of providing guidance for a current interventional procedure involving navigation of an interventional device within a vascular region, according to aspects of the present disclosure.

[0019] Figure 8 is a flow chart illustrating an example of a computer-implemented method of identifying a type of interventional device used in a current interventional procedure according to some aspects of the present disclosure.

[0020] Fig. 9 is a flow chart illustrating an example of a computer-implemented method of providing a predictive model for controlling a robotic interventional device manipulator to navigate an interventional device within a vascular region according to some aspects of the present disclosure.

[0021] Fig.10 is a flow chart illustrating an example of a computer-implemented method of using an interventional device simulation system to simulate robotic navigation of an interventional device within a vascular region during a current interventional device navigation procedure, according to aspects of the present disclosure.

[0022] Fig.11 is a flow chart illustrating an example of a computer-implemented method of determining differences between a simulated robotic interventional device navigation procedure and a current interventional device navigation procedure according to aspects of the present disclosure. DETAILED DESCRIPTION

[0023] Examples of the present disclosure are provided with reference to the following description and accompanying drawings. In this specification, for the purpose of explanation, many specific details of certain examples are set forth. References to "examples," "implementations," or similar language in the specification mean that features, structures, or characteristics described in conjunction with the examples are included in at least one example. It should also be understood that features described with respect to one example may also be used in another example, and for the sake of brevity, all features do not have to be repeated in each example. For example, features described with respect to computer-implemented methods may be implemented in a corresponding manner in a computer program product and system.

[0024] In the following description, a computer-implemented method for providing biomechanical plaque data for an interventional device simulation system is mentioned. An example of providing biomechanical plaque data for a vascular region in the heart is mentioned. However, it should be understood that the vascular region can generally be anywhere in the body. For example, the vascular region can be in the brain, arm, leg, etc. In some examples, it is mentioned that biomechanical plaque data for an artery is provided. However, it should be noted that biomechanical plaque data can generally be provided for any blood vessel in the vascular system. For example, alternatively, biomechanical plaque data can be provided for a vein. In some instances, it is mentioned that biomechanical plaque data is provided for a mature plaque. However, it should be understood that this type of plaque is only by way of example, and the method disclosed herein can be used to provide biomechanical plaque data for plaques at different stages of maturity. For example, the plaque can be a so-called fatty streak, or a ruptured plaque. Therefore, the method disclosed herein can also be used to provide biomechanical plaque data for plaques having a different composition than a mature plaque.

[0025] Note that the computer-implemented method disclosed herein can be provided as a non-transient computer-readable storage medium, the computer-readable storage medium including computer-readable instructions stored thereon, the computer-readable instructions causing the at least one processor to perform the method when run by at least one processor. In other words, the computer-implemented method can be implemented in a computer program product. The computer program product can be provided by dedicated hardware or hardware capable of running software associated with appropriate software. When provided by a processor, the function of the method feature can be provided by a single dedicated processor, or a single shared processor, or a plurality of individual processors (some of which can be shared). The function of one or more method features in the method feature can be provided, for example, by a processor shared within a networked processing architecture (e.g., a client / server architecture, a peer-to-peer architecture, the Internet, or a cloud).

[0026] The explicit use of the term "processor" or "controller" should not be interpreted as referring exclusively to hardware capable of running software, but can implicitly include, but is not limited to, digital signal processor "DSP" hardware, read-only memory "ROM" for storing software, random access memory "RAM", non-volatile storage devices, etc. In addition, examples of the present disclosure can take the form of a computer program product accessible from a computer-usable storage medium or a computer-readable storage medium, which provides program code used by or in conjunction with a computer or any instruction execution system. For the purposes of this specification, a computer-usable storage medium or a computer-readable storage medium can be any device capable of containing, storing, transmitting, propagating or transmitting a program used by or in conjunction with an instruction execution system, device or device. The medium can be an electrical, magnetic, optical, electromagnetic, infrared or semiconductor system or device or a propagation medium. Examples of computer-readable media include semiconductor or solid-state memory, magnetic tape, removable computer disk, random access memory "RAM", read-only memory "ROM", hard disk, and optical disk. Current examples of optical disks include compact disk-read only memory (CD-ROM), compact disk-read / write (CD-R / W), Blu-ray Disc, and DVD.

[0027] As mentioned above, the correlation between the simulated and real processes, and any benefits arising from this, is only feasible if the properties of the patches are accurately represented in the simulation.

[0028] Figure 1 is an example of a graphical representation of a blood vessel region 130 according to some aspects of the present disclosure. Figure 1 The blood vessel region 130 represented in FIG. 1 is a portion of the heart, and the blood vessel region includes coronary arteries. Figure 1 The lumen 160 of the coronary artery is labeled. Figure 2 is a schematic diagram illustrating an example of the spatial distribution of plaque 140 within a blood vessel region, according to some aspects of the present disclosure. Figure 2 The spatial distribution of the patch 140 shown in FIG. Figure 1 Plaque in a section of a coronary artery, such as the connection Figure 1 and Figure 2 Indicated by the line. Figure 2 The plaque 140 shown is a so-called mature plaque and includes a fibrous cap. The fibrous cap is formed by intimal smooth muscle cells and connective tissue. The fibrous cap separates the lumen 160 of the coronary artery from the thrombotic core of the plaque and is therefore the ultimate barrier to thrombosis. Figure 2 The nuclei of the plaques shown include foam cells, cholesterol, and other lipids.

[0029] Over time, the composition and distribution of plaque in blood vessels can change. For example, plaques may begin as so-called fatty streaks and then evolve into Figure 1 Mature plaque is shown. Mature plaque may subsequently evolve further by rupture, leading to thrombosis. During its evolution, vascular calcification may also occur in the blood vessel 130. Vascular calcification is defined as mineral deposition in the form of calcium phosphate complexes in the vasculature. Vascular calcification may occur in the intima layer and the media layer.

[0030] Therefore, the spatial distribution of a plaque and its composition depends on the stage of its evolution. This in turn affects its biomechanical properties, i.e., the mechanical properties that govern its response to applied forces. Therefore, in order to accurately represent the interaction between an interventional device and the plaque, it is important to accurately represent the spatial distribution of the biomechanical properties of the plaque.

[0031] Figure 3 is a flow chart illustrating an example of a computer-implemented method of providing biomechanical plaque data to an interventional device simulation system, according to aspects of the present disclosure. Figure 4 is a schematic diagram illustrating an example of a system 200 for providing biomechanical plaque data to an interventional device simulation system according to some aspects of the present disclosure. The system 200 includes one or more processors 210. Note that Figure 3 The operations described by the methods shown can also be performed by Figure 4 Similarly, the operations described with respect to the one or more processors 210 of the system 200 may also be performed in reference to Figure 3 The method described in the reference Figure 3 , a computer-implemented method for providing biomechanical plaque data 110 to an interventional device simulation system 100 includes:

[0032] Receive S110 spectral CT data 120 representing a blood vessel region 130, the spectral CT data defining a plurality of different energy intervals DE within the blood vessel region. 1..m X-ray attenuation in

[0033] extracting S120 plaque data from the spectral CT data 120 , the plaque data representing a spatial distribution of plaque 140 within the blood vessel region 130 ;

[0034] converting S130 the plaque data into biomechanical plaque data 110 representing a spatial distribution of mechanical constraints applied to the interventional device 150 in response to contact between the interventional device 150 and the plaque 140 of the interventional device simulation system 100; and

[0035] The biomechanical plaque data 110 is output S140 .

[0036] The spectral CT data defines X-ray attenuation in multiple different energy intervals. By processing data from multiple different energy intervals, it is possible to distinguish between multiple media that have similar X-ray attenuation values ​​when measured in a single energy interval and media that are indistinguishable in conventional X-ray attenuation data. Because in the above method, plaque data is extracted from spectral CT data, the plaque data provides a more accurate representation of the spatial distribution of plaques. In addition, because the plaque data is converted into biomechanical plaque data, which represents the spatial distribution of mechanical constraints applied to the interventional device in response to contact between the interventional device and the plaque of the interventional device simulation system, a more accurate simulation of the interaction between the interventional device and the plaque is provided.

[0037] refer to Figure 3 In the method shown, in operation S110, spectral CT data 120 is received. The spectral CT data represents a blood vessel region 130, and defines a plurality of different energy intervals DE within the blood vessel region. 1..m X-ray attenuation in .

[0038] In general, the spectral CT data 120 received in operation S110 may be raw data, i.e., data that has not yet been reconstructed into a volume or 3D image, or it may be image data, i.e., data that has been reconstructed into a volume image. In general, there may be two or more energy intervals; i.e., m is an integer, and m≥2.

[0039] The spectral CT data 120 received in operation S110 may be generated by a spectral CT imaging system, or, as described in more detail below, it may alternatively be generated by rotating or stepping an X-ray source and an X-ray detector of a spectral X-ray projection imaging system around a vascular region. More generally, the spectral CT data 120 received in operation S110 may be generated by a spectral X-ray imaging system.

[0040] The spectral CT imaging system generates spectral CT data by rotating or stepping an X-ray source-detector arrangement around an object and collecting X-ray attenuation data of the object from multiple rotation angles about the object. The spectral CT data can then be reconstructed into a 3D image of the object. Examples of spectral CT imaging systems that can be used to generate the spectral CT data 120 include cone-beam spectral CT imaging systems, photon counting spectral CT imaging systems, dark-field spectral CT imaging systems, and phase-contrast spectral CT imaging systems. An example of a spectral CT imaging system that can be used to generate the spectral CT data 120 received in operation S110 is the spectral CT7500 sold by Philips Healthcare of Best, the Netherlands.

[0041] As described above, the spectral CT data 120 received in operation S110 can alternatively be generated by rotating or stepping the X-ray source and X-ray detector of the spectral X-ray projection imaging system around the vascular region. The spectral X-ray projection imaging system generally includes a support arm (e.g., a so-called "C-arm") that supports the X-ray source and the X-ray detector. Alternatively, the spectral X-ray projection imaging system may include a support arm (e.g., an O-arm) having a different shape than the support arm of this example. Compared with the spectral CT imaging system, the spectral X-ray projection imaging system generates X-ray attenuation data of the object when the X-ray source and the X-ray detector are in a static position with respect to the object. Compared with the volume data generated by the spectral CT imaging system, the X-ray attenuation data can be referred to as projection data. The X-ray attenuation data generated by the spectral X-ray projection imaging system is generally used to generate a 2D image of the object. However, the spectral X-ray projection imaging system can generate spectral CT data (i.e., volume data) by rotating or stepping its X-ray source and X-ray detector around the object and collecting the projection data of the object from multiple rotation angles about the object. Then, the projection data obtained from the multiple rotation angles can be reconstructed into a volume image using image reconstruction techniques, in a manner similar to reconstructing a volume image using X-ray attenuation data acquired from a spectral CT imaging system. Therefore, the spectral CT data 120 received in operation S110 can be generated by a spectral CT imaging system, or alternatively, it can be generated by a spectral X-ray projection imaging system.

[0042] In multiple different energy ranges DE 1..m The ability to generate X-ray attenuation data in a single energy interval distinguishes spectral X-ray imaging systems from conventional X-ray imaging systems. By processing data from multiple different energy intervals, multiple media with similar X-ray attenuation values ​​when measured in a single energy interval and media that are indistinguishable in conventional X-ray attenuation data can be distinguished.

[0043] In general, the spectral CT data 120 may be generated by various different configurations of a spectral X-ray imaging system including an X-ray source and an X-ray detector. The X-ray source of the spectral X-ray imaging system may include multiple monochromatic sources, or one or more polychromatic sources, and the X-ray detector of the spectral X-ray imaging system may include: a common detector for detecting multiple different X-ray energy intervals; or multiple detectors, wherein each detector detects a different X-ray energy interval DE 1..mor a multi-layer detector in which X-rays having energies within different X-ray energy intervals are detected by corresponding layers; or a photon counting detector that bins detected X-ray photons into one of a plurality of energy intervals based on their individual energies. In a photon counting detector, the associated energy interval of each received X-ray photon can be determined by detecting the pulse height induced by electron-hole pairs generated in response to absorption of the X-ray photon in the direct conversion material.

[0044] The various configurations of the aforementioned X-ray sources and detectors can be used to detect different X-ray energy ranges DE 1..m In general, discrimination between different X-ray energy intervals can be provided at the source by temporally switching the X-ray tube potential of a single X-ray source (i.e., "fast kVp switching"), or by temporally switching or filtering the X-ray emission from multiple X-ray sources. Temporal switching can be performed within a rotation of the X-ray source-detector arrangement so that multiple different X-ray energy intervals DE 1..m The X-ray attenuation data is collected during rotation; or alternatively, the X-ray attenuation data of one energy interval can be collected during a specified number of revolutions of the gantry, and then the X-ray attenuation data of another energy interval can be similarly collected. In this configuration, a common X-ray detector can be used to detect X-rays in multiple different energy intervals, and the X-ray attenuation data of each energy interval is generated in a time-series manner. Alternatively, the distinction between different X-ray energy intervals can be provided at the detector by using a multi-layer detector or a photon counting detector. Such a detector is capable of detecting X-rays from multiple X-ray energy intervals DE nearly simultaneously. 1..m Therefore, multi-layer detectors or photon counting detectors can be used in conjunction with a polychromatic source to generate X-rays at different energy ranges. 1..m X-ray attenuation data at .

[0045] Other combinations of the aforementioned X-ray sources and detectors may also be used to provide spectral CT data 120. For example, in yet another configuration, by mounting the X-ray source-detector pair to a gantry at a rotationally offset position about a rotation axis, sequential switching of different X-ray sources for emitting X-rays in different energy intervals may not be required. In this configuration, each source-detector pair operates independently, and by virtue of the rotational offset of the source-detector pair, different energy intervals DE 1..m Improved DE for different energy intervals 1..m The separation between the spectral CT data can be achieved by applying an energy selective filter to the (one or more) X-ray detectors to reduce the impact of X-ray scatter.

[0046] In general, the spectral CT data 120 received in operation S110 may be received via any form of data communication, including wired, optical, and wireless communications. As some examples, when wired or optical communications are used, the communication may be performed via signals transmitted over an electrical or optical cable, and when wireless communications are used, the communication may be performed, for example, via RF or optical signals. The spectral CT data 120 received in operation S110 may be received from a variety of sources. For example, the spectral CT data 120 may be received from a spectral X-ray imaging system (e.g., one of the spectral X-ray imaging systems described above). Alternatively, the spectral CT data 120 may be received from another source (e.g., a computer-readable storage medium, the Internet, or the cloud).

[0047] Back to Figure 3 In the illustrated method, in operation S120, plaque data is extracted from the spectral CT data 120. The plaque data represents the spatial distribution of the plaque 140 within the blood vessel region 130.

[0048] Various material decomposition techniques may be used to extract the patch data in operation S120. These techniques include using various material decomposition algorithms, and generating so-called "Z eff Image", the Z eff The image represents the effective atomic number of the medium, and from it the various materials can be identified.

[0049] An example of a material decomposition technique that can be used to extract plaque data in operation S120 is disclosed in the document “Empirical, projection-based basis-component decomposition method” by Brendel, B. et al. (Medical Imaging 2009, Physics of Medical Imaging, edited by Ehsan Samei and Jiang Hsieh, SPIE Conference Proceedings, Vol. 7258, 72583Y). Another example of a material decomposition algorithm that can be used is disclosed in the document “K-edge imaging in X-ray computed tomography using multi-bin photon counting detectors” by Roessl, E. and Proksa, R. (Biomedical Physics, August 7, 2007, Vol. 52, No. 15, pp. 4679-4696). Another example of a material decomposition algorithm that can be used is disclosed in the published PCT patent application WO / 2007 / 034359A2. These and other material decomposition algorithms may be used to extract plaque data for different types of plaque (eg, soft plaque, calcified plaque, etc.) from the spectral CT data 120 .

[0050] As described above, in another method, in operation S120, a so-called "Z" indicating the effective atomic number of the medium may be used. eff The method for generating Z is disclosed in the document “A simple formulation forderiving effective atomic numbers via electron density calibration from dual-energy CT data in the human body” by Saito, M. et al. (Medical Physics, Vol. 44, No. 6, June 2017, pp. 2293-2303). eff An example of the technique. eff After the image is formed, the data of the desired material can be extracted from the image by selecting the data within the range of effective atomic numbers corresponding to the material. eff Plaque data of the calcified plaque or the soft plaque is extracted from the spectral CT data by selecting data in the image within a relevant range of effective atomic numbers corresponding to the calcified plaque or the soft plaque, respectively.

[0051] Back to Figure 3 The method shown, in operation S130, converts the plaque data into biomechanical plaque data 110. The biomechanical plaque data represents the spatial distribution of mechanical constraints applied to the interventional device 150 in response to contact between the interventional device 150 and the plaque 140 of the interventional device simulation system 100.

[0052] Figure 5 is a schematic diagram illustrating an example of an interventional device simulation system 100 according to some aspects of the present disclosure. Figure 5 The illustrated interventional device simulation system 100 includes a user interface device 170 and one or more processors 180. The user interface device 170 generates user input data for guiding the interventional device 150 within the vascular region 130 in response to received user input. The one or more processors 180 determine a virtual position of the interventional device 150 within the vascular region based on the user input data generated by the user interface device 170. The vascular region is defined by a 3D mathematical model, so the position of the interventional device within the vascular region is a "virtual" position. The position is then displayed on a display device (e.g., Figure 5 The virtual position of the interventional device 150 within the vascular region is displayed on the monitor 190 shown in FIG. By facilitating the user to navigate the interventional device in the vascular system, the user can use Figure 5 The illustrated interventional device simulation system 100 is used to gain experience in various interventional procedures.

[0053] During navigation of an interventional device in a vascular region, the interventional device often comes into contact with the plaque. The contact is "virtual" in that the position of the interventional device in the vascular region is virtual, so the contact with the plaque is also virtual. As mentioned above, the spatial distribution of the plaque and its composition depend on the stage of its evolution. This in turn affects its biomechanical properties, i.e., the mechanical properties that control its response to applied external forces. Therefore, in order to accurately represent the interaction between the interventional device and the plaque, it is important to accurately represent the spatial distribution of the biomechanical properties of the plaque.

[0054] In operation S130, the plaque data is converted into biomechanical plaque data 110. The biomechanical plaque data may represent a spatial distribution of one or more mechanical constraints applied to the interventional device 150 in response to contact between the interventional device 150 and the plaque 140 of the interventional device simulation system 100.

[0055] An example of a mechanical constraint that can be applied to the interventional device 150 in operation S130 is a spatial restriction. In this example, in response to contact between the interventional device 150 and the plaque 140, a spatial restriction is applied to the position of the interventional device 150. The spatial restriction can define that the contact portion of the interventional device cannot extend beyond the surface of the plaque. Therefore, the spatial restriction affects the freedom of movement of the interventional device. The spatial restriction can also vary depending on the amount of force applied to the plaque by the interventional device. This can be used to model the deformation of the plaque surface in response to the force, thereby defining new spatial restrictions based on the deformed plaque surface. The deformation of the plaque surface can also be modeled to represent that the plaque ruptures in response to excessive force, thereby defining further spatial restrictions based on the expected shape of the ruptured plaque. If the plaque data extracted from the spectral CT data 120 in operation S110 distinguishes different types of plaques, the amount of deformation can be calculated for the relevant type of plaque.

[0056] In general, the amount of deformation of the plaque surface can be modeled using the mechanical properties of the plaque. In this regard, the amount of deformation of the plaque surface can be modeled using the mechanical properties of the plaque (e.g., its elasticity, plasticity, ductility, malleability, hardness, toughness, brittleness, toughness, fatigue, fatigue resistance, impact resistance, strength, strain energy, resilience, standard rebound energy, modulus of resilience, creep, rupture, and modulus of toughness). One way to express the mechanical properties of the plaque is via its stress-strain curve. The stress-strain curve of a material determines the amount of deformation of the material when an external force is applied. Therefore, the stress-strain curve of the plaque can be used to model the amount of deformation of the plaque surface in response to the force applied to the plaque by the interventional device at the contact point. The modeled plaque surface deformation then presents new spatial constraints applied to the location of the interventional device 150 in response to contact between the interventional device 150 and the plaque 140. If the rupture point on the stress-strain curve is exceeded due to the application of excessive force, it is expected that the plaque may rupture. If the rupture point on the stress-strain curve is exceeded, this effect can also be modeled using the stress-strain curve by imposing further spatial constraints defined based on the expected shape of the rupture patch. In the simulation, this event will be an important performance indicator that is recorded and may cause visual notification.

[0057] Another example of a mechanical constraint that can be applied to the interventional device 150 in operation S130 is friction. In this example, in response to contact between the interventional device 150 and the plaque 140, an amount of friction is applied to the interventional device 150. Friction reduces the mobility of the interventional device contacting the plaque. In turn, friction may cause the interventional device to bend, which in turn affects the movement of the interventional device in the vascular region. For example, the amount of friction can be calculated using a friction coefficient for the plaque.

[0058] Another example of a mechanical constraint that can be applied to the interventional device 150 in operation S130 is the amount of force feedback applied to a force feedback user interface device of the simulation system in response to contact between the interventional device 150 and the plaque 140. Some user interface devices of the simulation system include so-called force feedback, in which the user feels an opposite force in response to contact between the interventional device 150 and the medium in the vascular system. With this user interface, the user can feel an opposite force in response to, for example, an attempt to force a guidewire against an obstacle in the vascular system. Force feedback improves the authenticity of the user training experience. With this user interface, the amount of force feedback applied is determined by simulating the mechanical properties of the interventional device. In this example, the mechanical constraint applied (i.e., the amount of force feedback) is determined by further considering the above-mentioned spatial constraints. In other words, the amount of force feedback applied to the force feedback user interface device is determined using the spatial constraints provided by the plaque data, in which plaque deformation can be modeled and plaque rupture can also be modeled. Therefore, for the user of the user interface device, the so-called soft plaque may feel softer than a hard plaque. Additionally, users of force feedback user interface devices may also experience plaque disruption - changes in the amount of force feedback.

[0059] In general, the operation S130 of converting the plaque data into the biomechanical plaque data 110 may be performed using a functional relationship between the plaque data and the mechanical constraints. For example, such a functional relationship may be provided by a graph or a lookup table.

[0060] For example, the plaque data extracted from the spectral CT data 120 can represent the spatial distribution of two categories of plaque (hard plaque and soft plaque). For soft plaque, a functional relationship in the form of a graph representing the relationship between stress and strain can be provided; and for hard plaque, another graph can be provided. In response to contact between an interventional device and a plaque, a stress-strain graph for the relevant soft / hard plaque is queried based on the amount of force applied to the plaque by the interventional device to model the amount of deformation induced in the plaque. The deformation is then used to modify the spatial constraints applied to the location of the interventional device by modifying the plaque data so that the shape of the plaque represented by the plaque data is deformed according to the deformation. If, based on the stress-strain graph, the amount of force applied to the plaque by the interventional device is sufficient to rupture the plaque, this effect can also be represented in the modified plaque data by deforming the plaque so that the modified plaque data represents the estimated shape of the ruptured plaque.

[0061] As another example, a lookup table can be used to store values ​​of the coefficient of friction for each plaque category (soft plaque and hard plaque). In response to contact between the interventional device and the plaque, the lookup table for the relevant soft / hard plaque is queried to determine the coefficient of friction to be used. The coefficient of friction is then used to calculate the amount of friction to be applied to the interventional device as a constraint in response to contact between the interventional device 150 and the plaque 140.

[0062] Back to Figure 3 In the method shown, in operation S140, the biomechanical plaque data 110 is output. The biomechanical plaque data 110 can be output in various ways. In one example, the biomechanical plaque data 110 is output to a computer-readable storage medium. In another example, a graphical representation of the biomechanical plaque data 110 is output to a display device. For example, the graphical representation of the biomechanical plaque data 110 can be output to a monitor, or to another type of display device (e.g., a virtual reality head-mounted device, or an augmented reality head-mounted device). The graphical representation of the biomechanical plaque data 110 can be output to a display device as a 3D image or a 2D image. A 2D image can be generated by projecting a 3D image. The graphical representation of the biomechanical plaque data 110 may include color coding, or shading or another type of formatting that depicts the plaque, and if the biomechanical plaque data 110 is extracted from spectral CT data, the biomechanical plaque data 110 of different types of plaques (e.g., soft / hard plaques) can also be depicted. In another example, the biomechanical plaque data 110 is output to the Internet or the cloud. In another example, the biomechanical plaque data 110 is output to a printer.

[0063] In one example, a graphical representation of the vessel region 130 is output, and the graphical representation of the vessel region 130 includes a graphical representation of plaque data and / or a graphical representation of the output biomechanical plaque data 110. In this example, the graphical representation of the vessel region 130 represents (one or more) vessels in the vessel region. The graphical representation of the vessel region 130 may be generated based on lumen data extracted from the spectral CT data 120 received in operation S110. For example, a material decomposition algorithm or a Z eff The image extracts lumen data from the spectral CT data, and an image of (one or more) blood vessels in the vascular region can be generated based on the lumen data. The image can then be superimposed with a graphical representation of plaque data and / or a graphical representation of the output biomechanical plaque data 110. In a related example, the graphical representation can include a virtual angiographic projection of the vascular region 130. Such an image can be used to facilitate a physician's clinical diagnosis of the vascular region.

[0064] In another example, lumen data of a blood vessel region is output. Figure 5In the simulations performed by the illustrated interventional device simulation system 100, lumen data may be used in conjunction with biomechanical plaque data. For example, lumen data may be used to generate a 3D mathematical model representing a vascular region. Simulations may be performed using lumen data in conjunction with biomechanical plaque data, wherein both lumen data and biomechanical plaque data provide spatial constraints applied to the position of an interventional device as it is navigated through (one or more) lumens in a vascular region. In this example, reference is made to Figure 3 The methods described include:

[0065] extracting luminal data from the spectral CT data 120, the luminal data representing a three-dimensional shape of one or more lumens 160 within the vessel region 130; and

[0066] Output lumen data.

[0067] Luminal data may be extracted from the spectral CT data in a manner similar to that for plaque data (i.e., by using material decomposition techniques). Luminal data may be output in a manner similar to that for biomechanical plaque data 110. In one example, the lumen data and plaque data are output as a graphical representation. For example, a 3D image may be generated from the lumen data and the 3D image may be overlaid with a graphical representation of the biomechanical plaque data 110. By providing lumen data from the spectral CT data and the biomechanical plaque data 110, personalized simulation data of a vascular region may be provided, wherein the two data sets are inherently registered to one another.

[0068] In another example, reference Figure 3 The described method includes calculating a spatial distribution of plaque rupture risk from plaque data and outputting a graphical representation of the plaque rupture risk.

[0069] Because the plaque data is extracted from the spectral CT data, the plaque data extracted from the spectral CT data provides a more accurate description of the plaque than if the plaque data was extracted from conventional CT data. For example, as described above, the use of spectral CT data facilitates the determination of the type of plaque and its composition. Figure 1 As shown, high risk of rupture is associated with factors such as the presence of a fibrous cap on the plaque. Thus, in this example, the risk of plaque rupture can be calculated by detecting these features in the plaque data. The graphical representation of the risk of plaque rupture can be output in various ways. For example, the graphical representation of the risk of plaque rupture can be output as an overlay image on the biomechanical plaque data, where color coding, shading, or other types of formatting are used in the overlay to depict the risk of plaque rupture. Figure 5In the simulation performed by the illustrated interventional device simulation system 100, in addition to biomechanical plaque data, plaque rupture risk may also be used. For example, plaque rupture risk may be used to simulate the plaque region that ruptures due to contact between the interventional device and the plaque.

[0070] In another example, a computer-implemented method of providing biomechanical plaque data 110 to an interventional device simulation system 100 is performed by the system. Figure 4 The system 200 for providing biomechanical plaque data 110 to the interventional device simulation system 100 includes one or more processors 210 configured to:

[0071] Receive S110 spectral CT data 120 representing a blood vessel region 130, the spectral CT data defining a plurality of different energy intervals DE within the blood vessel region. 1..m X-ray attenuation in

[0072] extracting S120 plaque data from the spectral CT data 120 , the plaque data representing a spatial distribution of plaque 140 within the blood vessel region 130 ;

[0073] converting S130 the plaque data into biomechanical plaque data 110 representing a spatial distribution of mechanical constraints applied to the interventional device 150 in response to contact between the interventional device 150 and the plaque 140 of the interventional device simulation system 100; and

[0074] The biomechanical plaque data 110 is output S140 .

[0075] The system 200 may also include a spectral CT imaging system 220 for providing the spectral CT data 120. The system 200 may also include a display (e.g., Figure 4 A monitor 230 is shown. A display may be used to display graphical representations of spectral CT data, plaque data, biomechanical plaque data 110, etc. The system 200 may also include a patient bed 240.

[0076] In another example, an interventional device simulation system 100 is provided for simulating navigation of an interventional device 150 within a vascular region 130. The simulation system includes:

[0077] User interface device 170; and

[0078] one or more processors 180;

[0079] wherein the user interface device 170 is configured to generate user input data for guiding the intervention device 150 within the blood vessel region 130 in response to the received user input; and

[0080] The one or more processors 180 are configured to:

[0081] receiving lumen data and biomechanical plaque data 110;

[0082] determining a virtual position of the interventional device 150 within the one or more lumens 160 represented by the lumen data based on the user input data generated by the user interface device 170; and

[0083] Therein, determining the virtual position of the interventional device 150 includes applying a mechanical constraint represented by the biomechanical plaque data 110 to the interventional device 150 in response to contact between the interventional device 150 and the plaque 140 .

[0084] Figure 5 An example of the system 100 is illustrated. The user interface 170 may be provided by various types of devices. For example, the user interface 170 may include one or more input devices, such as: a joystick, a mouse, a touchpad, a keyboard, a knob, a switch, a slider control, a gesture control device, etc. Such a user interface generates user input data in the form of electrical, RF or optical signals using various sensors coupled to the user input device. Sensors may be provided by various types of sensors, including one or more switches, potentiometers, strain gauges, optical sensors, (depth) cameras, etc. In one example described below, a real interventional device forms part of the user interface device. In this example, various sensors (e.g., the sensors described above) may be used to generate user input data in response to user manipulation of the real interface device. In another example, the user interface is provided by a real controller of the interventional device. An example of such a controller is a catheter "handle" used in an actual procedure. In Figure 5 The example is illustrated within the dashed outline in . The catheter handle includes controls in the form of switches, knobs, dials, etc., which are used to advance, retract, and rotate the catheter (sometimes via one or more motors or actuators). Sensors can be used to monitor the state of the controls to provide user input data. Using a real controller (e.g., a catheter handle) provides the user with a more realistic interventional procedure experience. In another example, the user interface device 170 applies force feedback to the user of the user interface device so that the user can, among other things, feel an opposing force in response to contact between the interventional device 150 and the medium in the vascular system.

[0085] Continue to refer Figure 5 , one or more processors 180 receive the above-mentioned lumen data and biomechanical plaque data 110. For example, the one or more processors can receive the lumen data from a computer readable storage medium or from the Internet or the cloud. In one example, Figure 5As shown, the system 100 is provided in the form of a training console for delivering a training experience to a user. In this example, the one or more processors 180 may form part of the training console. In another example, the one or more processors 180 are the same as the one or more processors 210 for providing biomechanical plaque data to an interventional device simulation system, as described in reference to Figure 4 Thus, in this example, the one or more processors 180 may also perform the following operations: receive S110 the spectral CT data 120, extract S120 plaque data from the spectral CT data 120, convert S130 the plaque data into biomechanical plaque data 110, and output S140 the biomechanical plaque data 110, as shown in reference Figure 3 described.

[0086] Continue to refer Figure 5 , the one or more processors 180 determine a virtual position of the interventional device 150 within the one or more lumens 160 represented by the lumen data based on the user input data generated by the user interface device 170 .

[0087] In this operation, the lumen data extracted from the spectral CT data 120 is used to generate a 3D mathematical model of one or more lumens 160, and the virtual position of the interventional device 150 is determined in the 3D mathematical model. Because both the lumen data and the biomechanical plaque data are extracted from the spectral CT data 120, a personalized model of the vascular region is provided, wherein the two data sets are inherently registered with each other. For example, a 3D mathematical model can be generated from the spectral CT data using a technique similar to that disclosed in document US2012 / 197619A1 for generating a model from conventional CT data.

[0088] The position of the interventional device 150 within the one or more lumens 160 determined in this operation is a "virtual" position because the position is determined in the 3D mathematical model. The interventional device 150 is also virtual and provided by the 3D mathematical model, which models the shape of the interventional device when navigating the interventional device in the 3D mathematical model of the one or more lumens. The 3D mathematical model of the interventional device can be based on the mechanical properties of the real interventional device. For example, the stiffness of the real interventional device can be modeled to provide a more realistic interventional procedure experience for the user. The virtual position of the interventional device 150 can be determined by replicating the user's manipulation of the interventional device in the 3D mathematical model of the interventional device using user input data, and using the 3D mathematical model of the one or more lumens to constrain the position and shape of the interventional device. Therefore, the user interface device 170 is used to navigate the virtual interventional device in one or more lumens based on the user input data generated by the user interface device 170. Note that although the interventional device is virtual in the sense provided by the 3D mathematical model, as described above, the user interface device can include a "real" interventional device. For example, a real catheter may be provided for manipulation by a user, wherein the various sensors described above are used to monitor movement of the catheter, thereby generating user input data that is used to replicate the movement of the real catheter in a 3D mathematical model of the catheter.

[0089] In general, interventional device 150 can be any type of interventional device suitable for navigating in a vascular system. For example, depending on the type of interventional procedure being simulated, the interventional device can be a guidewire, a catheter, an intravascular ultrasound "IVUS" imaging device, an optical coherence tomography "OCT" imaging device, a blood flow sensing device, a blood pressure monitoring device, a balloon catheter, a stent, an atherectomy device, etc.

[0090] Continue to refer Figure 5 , when there is contact between the interventional device 150 and the plaque 140, the one or more processors 180 determine the virtual position of the interventional device 150 within one or more lumens represented by the lumen data by applying the mechanical constraints represented by the biomechanical plaque data 110 to the interventional device 150. In this operation, the above-mentioned mechanical constraints are applied to the interventional device. In other words, in response to the contact between the interventional device 150 and the plaque 140, spatial constraints may be applied to the position of the interventional device 150. Alternatively or additionally, in response to the contact between the interventional device 150 and the plaque 140, an amount of friction may be applied to the interventional device 150. Alternatively or additionally, in response to the contact between the interventional device 150 and the plaque 140, an amount of force feedback may be applied to a force feedback user interface device of the simulation system.

[0091] Continue to refer Figure 5 , which can then be displayed on a display (e.g. Figure 5 A graphical representation of the vascular region is displayed on a monitor 190 (shown in FIG. 1 ), the graphical representation also including a virtual position of the interventional device 150 within one or more lumens 160. Alternatively, the graphical representation may be displayed on other types of display devices (e.g., a virtual reality headset or an augmented reality headset). By facilitating the user to navigate the interventional device in the vascular system, the user may use Figure 5 The illustrated interventional device simulation system 100 is used to gain experience in various interventional procedures.

[0092] In one example, reference Figure 5 The described system applies force feedback to a user of a force feedback user interface device. As described above, the use of force feedback facilitates a more realistic training experience. In this example, the user interface device 170 includes a force feedback interface device, and the force feedback interface device is configured to receive a control signal for applying force feedback to a user of the force feedback user interface device, and the mechanical constraint includes an amount of force feedback applied to the force feedback user interface device in response to contact between the intervention device 150 and the plaque 140.

[0093] In this example, the one or more processors 180 are configured to:

[0094] generating a control signal for applying force feedback to a user of the user input device 170 based on the virtual position of the interventional device 150 within the one or more lumens 160; and

[0095] Therein, control signals generated by the one or more processors 180 in response to contact between the interventional device 150 and the plaque 140 in the one or more lumens 160 are generated based on the biomechanical plaque data 110 .

[0096] Thus, in this example, virtual contact between the interventional device and the plaque causes force feedback to be applied to the force feedback interface device. This can be used to provide a realistic experience in the event that the interventional device contacts the plaque or further deforms the plaque or even ruptures the plaque.

[0097] In another example, reference Figure 5 The described system outputs a graphical representation of the lumen data and the biomechanical plaque data 110. In this example, the one or more processors 180 are configured to output a graphical representation of the lumen data and the biomechanical plaque data 110. For example, the graphical representation may be provided as an overlay image.

[0098] In another example, reference Figure 5The described system is used to modify the biomechanical plaque data 110. In this example, the one or more processors 180 are also configured to receive inputs defining modifications to the biomechanical plaque data 110, and determine control signals generated in response to contact between the interventional device 150 and the plaque 140 in the one or more lumens 160 based on the modified biomechanical plaque data 110.

[0099] In this example, modifications to the biomechanical plaque data may be provided via the user interface device 170. For example, modifications to the biomechanical plaque data may define modifications to its mechanical properties or shape. The input may be used to modify the reference biomechanical plaque data to provide, for example, a more complex or more direct simulation experience to the user. For example, the input may be received via a menu, where the user selects a level of complexity for the simulation process.

[0100] In another example, reference Figure 5 The described system is used to modify lumen data. In this example, one or more processors 180 are configured to:

[0101] receiving input defining a modification to lumen data;

[0102] modifying the lumen data based on the received input; and

[0103] Therein, the modified lumen data is used to determine a virtual position of the interventional device 150 within one or more lumens 160 represented by the lumen data.

[0104] In this example, the modification of the lumen data may be provided via the user interface device 170. The modification of the lumen data may define a modification of the shape of the lumen. For example, the lumen area or lumen curvature may be changed to increase the complexity of the simulation process.

[0105] In a related example, the modification to the lumen data represents a virtual implant device inserted in one or more lumens 160. In this example, the operation of determining the virtual position of the interventional device 150 includes applying mechanical constraints to the interventional device 150 based on the mechanical properties of the virtual implant device in response to contact between the interventional device 150 and the virtual implant device. In this example, for example, the implant device can be a stent. For example, the mechanical properties of the stent can be its hardness or its coefficient of friction. The hardness of the stent affects the deformation of the lumen in response to its contact with the interventional device. The coefficient of friction affects how the interventional device moves during contact with the stent. Therefore, this example can be used to provide a more realistic simulation of the navigation of an interventional device through a stent.

[0106] In another example, a reference is provided. Figure 5The described interventional device simulation system 100 is used to simulate a computer-implemented method for navigating an interventional device 150 within a vascular region 130. The method includes:

[0107] receiving S210 lumen data and biomechanical plaque data 110;

[0108] receiving S220 user input data generated by the user interface device 170;

[0109] determining S230 a virtual position of the interventional device 150 within the one or more lumens 160 represented by the lumen data based on the user input data generated by the user interface device 170; and

[0110] Therein, determining the virtual position of the interventional device 150 includes applying a mechanical constraint represented by the biomechanical plaque data 110 to the interventional device 150 in response to contact between the interventional device 150 and the plaque 140 .

[0111] exist Figure 6 The method is illustrated in Figure 6 is a flow chart illustrating an example of a computer-implemented method for simulating navigation of an interventional device 150 within a vascular region 130 according to some aspects of the present disclosure. Figure 5 The operations described in the system can also be performed in Figure 6 Execute the method shown.

[0112] In another example, reference Figure 5 The described system is used to generate a database of process simulation data. In this regard, a computer-implemented method of using the interventional device simulation system 100 to generate a database of process simulation data is provided. The method includes:

[0113] Record process simulation data. For each simulation process, the process simulation data includes:

[0114] user input data generated by the user interface device 170;

[0115] Position data representing a virtual position of the intervention device 150 at each of a plurality of time points;

[0116] Biomechanical plaque data 110 used in the simulation process;

[0117] Device data representing the type of interventional device 150 used in the simulated procedure; and

[0118] Result data from the simulation process; and

[0119] Store process simulation data into the database.

[0120] In this example, the process simulation data may be recorded using a computer-readable storage medium and may also be stored to the computer-readable storage medium. The user input data represents input provided by a user and may be recorded, for example, by storing signals generated by a user interface device. For example, these signals may represent Figure 5 Movement of the joystick shown. Position data can be recorded by storing a virtual position of the interventional device in a 3D mathematical model of one or more lumens. For example, the device data can indicate whether the interventional device 150 is a catheter or a guidewire or another type of interventional device. The result data can represent various factors, such as one or more of the following: the duration of the simulated procedure, a measure of success of the procedure, and the location of one or more confounding factors encountered during the procedure in one or more lumens 160.

[0121] Use reference Figure 5 The database of procedure simulation data generated by the described system can be used for various purposes. In one example, the database of procedure simulation data is used to provide guidance for the current interventional procedure. In another example, the database of procedure simulation data is used to identify the type of interventional device 150 used in the current interventional procedure. In another example, the database of procedure simulation data is used to train a predictive model to control a robotic interventional device manipulator.

[0122] Figure 7 is a flow chart illustrating an example of a computer-implemented method for providing guidance for a current interventional procedure involving navigation of an interventional device within a vascular region, according to some aspects of the present disclosure. Figure 7 , a method for providing guidance for a current interventional procedure involving navigation of an interventional device 150 within a vascular region 130 includes:

[0123] A database for receiving S310 process simulation data;

[0124] receiving S320 spectral CT data 120 representing a blood vessel region 130 of a current interventional procedure;

[0125] extracting S330 plaque data of the current interventional procedure from the spectral CT data 120 , the plaque data representing the spatial distribution of plaque 140 within the blood vessel region 130 of the current interventional procedure;

[0126] identifying S340 one or more similar simulated procedures from a database based on similarities between plaque data of the current interventional procedure and plaque data of the simulated procedure;

[0127] predicting S350 an outcome metric of the current interventional procedure based on the identified outcome data of one or more similar simulated procedures; and

[0128] The predicted outcome metrics are output S360 to provide guidance for the current intervention process.

[0129] In this example, the operations of identifying S340, predicting S350, and outputting S360 may be repeated during the current interventional procedure to provide an updated predicted outcome metric corresponding to the current time in the current interventional procedure.

[0130] In this example, for the current intervention procedure, the operation of receiving S320 the spectral CT data 120 and the operation of extracting S330 the plaque data are similar to those described above with reference to Figure 3 The method described above is performed in the same manner. The similarity between the plaque data of the current interventional procedure and the plaque data of the simulated procedure can be calculated based on factors such as the type of plaque identified in the procedure and the distribution of plaque identified in the procedure. Metrics such as the Dice coefficient, local root mean square error, and mutual information can be used to calculate the similarity between images. The similarity between the types or categories of plaques can be calculated by applying a weighted distance metric to the (multidimensional) category space. The result metric is predicted based on the result data, which, as described above, can represent one or more of the following: the duration of the simulated procedure, the success metric of the procedure, and the location of one or more confounding factors encountered during the procedure in one or more lumens 160. For example, predicting the result metric can include selecting the result metric of the most similar simulated procedure from a database, or calculating the result metric as the average of multiple similar simulated procedures from a database. By providing guidance for the current interventional procedure based on this example, doctors can be helped to perform more successful procedures.

[0131] As described above, in another example, the database of procedure simulation data is used to identify the type of interventional device 150 used in the current interventional procedure. Figure 8 To describe this example, Figure 8 is a flow chart illustrating an example of a computer-implemented method for identifying the type of interventional device used in a current interventional procedure according to some aspects of the present disclosure. The computer-implemented method for identifying the type of interventional device 150 used in a current interventional procedure includes:

[0132] A database for receiving S410 process simulation data;

[0133] receiving S420 spectral CT data 120 representing a blood vessel region 130 of a current interventional procedure;

[0134] extracting S430 plaque data of the current interventional procedure from the spectral CT data 120 , the plaque data representing the spatial distribution of plaque 140 within the blood vessel region 130 of the current interventional procedure;

[0135] identifying S440 one or more similar simulated procedures from a database based on similarities between plaque data of the current interventional procedure and plaque data of the simulated procedure;

[0136] identifying S450 a type of interventional device 150 used in a current interventional procedure based on the identified device data of one or more similar simulated procedures; and

[0137] An indication of the type of interventional device 150 identified is output S460 .

[0138] In this example, for the current intervention procedure, the operation of receiving S420 the spectral CT data 120 and the operation of extracting S430 the plaque data are similar to those described above with reference to Figure 3 The same method as described above can be used. Figure 6 The similarity between the plaque data of the current interventional procedure and the plaque data of the simulated procedure is calculated. The device data represents the type of the interventional device 150. For example, the device data may indicate that the interventional device 150 is a catheter or a guidewire or another type of interventional device. In this example, the operation of outputting the indication of the type of the interventional device 150 identified by S460 may include, for example, outputting the device data of the most similar simulated procedure from the database, or determining the most common device data of multiple similar simulated procedures from the database and outputting the most common device data.

[0139] As described above, in another example, a database of procedure simulation data is used to train a predictive model to control a robotic interventional device manipulator. Fig. 9 To describe this example, Fig. 9 is a flow chart illustrating an example of a computer-implemented method for controlling a robotic interventional device manipulator to navigate an interventional device within a vascular region according to some aspects of the present disclosure. Fig. 9 , a computer-implemented method for providing a predictive model for controlling a robotic interventional device manipulator to navigate an interventional device 150 within a vascular region 130 includes:

[0140] A database receiving S510 process simulation data; and

[0141] The prediction model is trained S520 to control a robotic interventional device manipulator to navigate the interventional device 150 within the vascular region 130 in order to replicate position data representing a virtual position of the interventional device 150 .

[0142] In this example, the robotic interventional device manipulator can be a handle of a catheter as described above. Alternatively, the robotic interventional device manipulator can be another type of manipulator, such as the manipulator of the Corindus CorPath GRX system sold by Siemens Healthcare of Erlangen, Germany, or the R-One robotic-assisted platform sold by Robocath of Rouen, France. The latter two devices employ manipulators that can be used to navigate elongated interventional devices in the vascular system.

[0143] In this example, the database of process simulation data may include processes for various different spatial distributions of patches and various compositions of patches. The predictive model in this example may be provided by a machine learning model or a neural network. Examples of suitable machine learning models include random forests and shallow predictors (e.g., support vector machines). Examples of suitable neural networks include convolutional neural networks "CNNs", recurrent neural networks "RNNs", or transformers.

[0144] In this example, the operation of training the prediction model may include controlling a user interface device 170 of the interventional device simulation system 100 via the robotic device manipulator, and wherein the control is based on an output of the prediction model. Thus, in this example, the robotic interventional device manipulator controls the user interface device 170 of the system 100 to provide a virtual position of the interventional device 150. The prediction model is trained using the difference between the virtual position of the interventional device and the position of the interventional device in the procedure simulation data.

[0145] In one example, a predictive model may be trained to control a robotic interventional device manipulator to navigate an interventional device 150 within a vascular region 130 using procedure simulation data by:

[0146] receiving process simulation data;

[0147] Inputting process simulation data into the predictive model; and

[0148] For each of a plurality of simulation runs, and for each of a plurality of time points:

[0149] inputting position data representing a virtual position of the interventional device 150 at a point in time from a simulated procedure;

[0150] using the prediction model to predict a position of the interventional device 150 at a point in time;

[0151] adjusting parameters of the prediction model based on a difference between the predicted position of the interventional device 150 at the point in time and the virtual position of the interventional device 150 at the point in time; and

[0152] Prediction and adjustment are repeated until a stopping criterion is met.

[0153] In another example, a neural network can be trained in an unsupervised manner. Reinforcement learning on a simulated flow allows the AI ​​to be trained based on one or more success measurements without relying on labeled real-world data. The biomechanical plaque data used during simulation training defines the specific clinical scenario for which the AI ​​is trained. CT-based plaque data from clinical cases is used to inform whether this trained AI is appropriate for a specific clinical case.

[0154] In general, training a neural network involves inputting a training data set into the neural network and iteratively adjusting the parameters of the neural network until the trained neural network provides accurate outputs. Training is typically performed using a graphics processing unit "GPU" or a dedicated neural processor (e.g., a neural processing unit "NPU" or a tensor processing unit "TPU"). Training typically employs a centralized approach, where a cloud-based or mainframe-based neural processor is used to train the neural network. After the neural network has been trained using the training data set, the trained neural network can be deployed to a device for analyzing new input data during inference. The processing requirements during inference are significantly less than the processing requirements during training, allowing the neural network to be deployed to a variety of systems (e.g., laptops, tablets, mobile phones, etc.). Inference can be performed, for example, by a central processing unit "CPU", GPU, NPU, TPU on a server or in the cloud.

[0155] Therefore, the above-mentioned process of training a neural network includes adjusting its parameters. Parameters (or more specifically weights and biases) control the operation of activation functions in a neural network. In supervised learning, the training process automatically adjusts weights and biases so that when input data is provided, the neural network accurately provides the corresponding expected output data. To this end, the value or error of the loss function is calculated based on the difference between the predicted output data and the expected output data. The value of the loss function can be calculated using functions such as negative log-likelihood loss, mean absolute error (or L1 norm), mean square error, root mean square error (or L2 norm), Huber loss or (binary) cross entropy loss. During training, the value of the loss function is usually minimized, and training is terminated when the value of the loss function meets the stopping criteria. Sometimes, training is terminated when the value of the loss function meets one or more criteria in multiple criteria.

[0156] Various methods are known to solve the loss minimization problem, such as gradient descent, quasi-Newton methods, etc. Various algorithms have been developed to implement these methods and their variants, including but not limited to stochastic gradient descent "SGD", batch gradient descent, mini-batch gradient descent, Gauss-Newton, Levenberg Marquardt, Momentum, Adam, Nadam, Adagrad, Adadelta, RMSProp, and Adamax "optimizers". These algorithms use the chain rule to calculate the derivatives of the loss function with respect to the model parameters. This process is called backpropagation because the derivatives are calculated starting from the last layer or output layer and moving toward the first layer or input layer. These derivatives inform the algorithm how the model parameters must be adjusted to minimize the error function. That is, the model parameters are adjusted starting from the output layer and working backwards through the network until the input layer is reached. In the first training iteration, the initial weights and biases are usually randomized. The neural network then predicts the output data, which is also randomized. Backpropagation is then used to adjust the weights and biases. The training process is performed iteratively by adjusting the weights and biases in each iteration. When the error or difference between the predicted output data and the expected output data is within an acceptable range for the training data or some validation data, the training is terminated. The neural network can then be deployed and the trained neural network uses the trained values ​​of its parameters to make predictions on new input data. If the training process is successful, the trained neural network will accurately predict the expected output data based on the new input data.

[0157] In another example, the process simulation data is grouped based on the type of plaque, and separate prediction models are provided for different types of plaque. Fig. 9 The methods described include:

[0158] grouping the flow simulation data based on the type of plaque represented by the biomechanical plaque data 110; and

[0159] The method includes providing a separate prediction model for controlling a manipulator of a robotic interventional device for each type of plaque among a plurality of different types of plaques, each prediction model being trained using procedure simulation data for the corresponding type of plaque.

[0160] Therefore, different prediction models can be provided for different types of plaques.

[0161] In another example, a computer-implemented method of controlling a robotic interventional device manipulator to navigate an interventional device 150 within a vascular region 130 during a current interventional device navigation procedure is provided. The method includes:

[0162] receiving a plurality of prediction models;

[0163] receiving spectral CT data 120 representing a blood vessel region 130 of a current interventional device navigation procedure;

[0164] Extracting plaque data of a current interventional device navigation procedure from the spectral CT data 120 , the plaque data representing the spatial distribution of plaques 140 within a blood vessel region 130 of the current interventional device navigation procedure;

[0165] selecting a prediction model from the prediction models based on similarity between plaque data of a current interventional device navigation process and plaque data used to train the prediction model; and

[0166] The selected prediction model is used to control a robotic interventional device manipulator to navigate an interventional device 150 within the vascular region 130 during a current interventional device navigation procedure.

[0167] In this example, the received plurality of prediction models includes providing a separate prediction model for controlling the robotic interventional device manipulator for each type of plaque in a plurality of different types of plaques. The operation of selecting the prediction model is based on the similarity between the plaque data of the current interventional device navigation process and the plaque data used to train the prediction model. Figure 6 The similarity between the plaque data of the current intervention procedure and the plaque data used to train the prediction model is calculated. This example provides a prediction model for controlling a robotic intervention device manipulator to navigate an intervention device 150, wherein the prediction model is suitable for the plaque in the current intervention procedure.

[0168] In another example, a computer-implemented method of controlling a robotic interventional device manipulator to navigate an interventional device 150 within a vascular region 130 during a current interventional device navigation procedure is performed by the system. Therefore, a system for controlling a robotic interventional device manipulator to navigate an interventional device 150 within a vascular region 130 during a current interventional device navigation procedure is provided. The system includes one or more processors configured to:

[0169] receiving a plurality of prediction models;

[0170] receiving spectral CT data 120 representing a blood vessel region 130 of a current interventional device navigation procedure;

[0171] Extracting plaque data of a current interventional device navigation procedure from the spectral CT data 120 , the plaque data representing the spatial distribution of plaques 140 within a blood vessel region 130 of the current interventional device navigation procedure;

[0172] selecting a prediction model from the prediction models based on similarity between plaque data of a current interventional device navigation process and plaque data used to train the prediction model; and

[0173] The selected prediction model is used to control a robotic interventional device manipulator to navigate an interventional device 150 within the vascular region 130 during a current interventional device navigation procedure.

[0174] In another example, a computer-implemented method of using the interventional device simulation system 100 to simulate robotic navigation of an interventional device 150 within a vascular region 130 during a current interventional device navigation procedure is provided. Fig.10 To describe this example, Fig.10 is a flow chart illustrating an example of a computer-implemented method for simulating robotic navigation of an interventional device within a vascular region during a current interventional device navigation procedure using an interventional device simulation system according to some aspects of the present disclosure. In this example, the method includes:

[0175] Receive S610 multiple prediction models;

[0176] receiving S620 spectral CT data 120 of a blood vessel region 130 representing a current interventional device navigation process;

[0177] Extracting S630 plaque data of the current interventional device navigation process from the spectral CT data 120 , the plaque data representing the spatial distribution of plaque 140 within the blood vessel region 130 of the current interventional device navigation process;

[0178] selecting 640 a prediction model from the prediction models based on similarity between the plaque data of the current interventional device navigation procedure and the plaque data used to train the prediction model; and

[0179] Using the selected prediction model, the user interface device 170 of the interventional device simulation system 100 is robotically controlled S650 to navigate the interventional device 150 within the blood vessel region 130 during the current interventional device navigation procedure.

[0180] In this example, the plurality of prediction models received in operation S610 include providing a separate prediction model for controlling the robotic interventional device manipulator for each type of plaque in a plurality of different types of plaques. By using the selected prediction model to robotically control the user interface device 170 of the interventional device simulation system 100 S650, a more optimized simulation is provided. This is because the selected model is an appropriate model for the type of plaque in the current interventional device navigation procedure. The simulation can be used to demonstrate to the doctor a method of performing an interventional procedure.

[0181] In another example, a difference between a simulated robotic interventional device navigation process and a current interventional device navigation process is determined. This difference can be used in the navigation process to determine when the current interventional device navigation deviates from the expected process. Fig.11 To describe this example, Fig.11 is a flow chart illustrating an example of a computer-implemented method for determining a difference between a simulated robotic interventional device navigation procedure and a current interventional device navigation procedure according to some aspects of the present disclosure. In this example, the method includes:

[0182] Simulating S710 robotic navigation of the interventional device 150 within the blood vessel region 130 during the current interventional device navigation procedure;

[0183] controlling S720 the robotic interventional device manipulator to navigate the interventional device 150 within the blood vessel region 130 during a current interventional device navigation procedure; and

[0184] A difference between the simulated robotic navigation of the interventional device 150 and the navigation of the interventional device 150 performed by controlling the robotic interventional device manipulator to navigate the interventional device 150 within the blood vessel region 130 during the current interventional device navigation procedure is determined S730 .

[0185] In this example, according to Fig.10 The method shown in FIG. 1 is used to perform the operation of simulating S710 robot navigation. The difference can be output in various ways. For example, the difference can be output to a display device, or the difference can be output in an auditory manner. In one example, if the size of the difference exceeds a threshold, a user is warned via a message.

[0186] In another example, a computer-implemented method for determining a difference between a simulated robotic interventional device navigation process and a current interventional device navigation process is performed by a system. Therefore, a system for determining a difference between a simulated robotic interventional device navigation process and a current interventional device navigation process is provided. The system includes one or more processors, the one or more processors being configured to:

[0187] Simulating S710 robotic navigation of the interventional device 150 within the blood vessel region 130 during the current interventional device navigation procedure;

[0188] controlling S720 the robotic interventional device manipulator to navigate the interventional device 150 within the blood vessel region 130 during a current interventional device navigation procedure; and

[0189] A difference between the simulated robotic navigation of the interventional device 150 and the navigation of the interventional device 150 performed by controlling the robotic interventional device manipulator to navigate the interventional device 150 within the blood vessel region 130 during the current interventional device navigation procedure is determined S730 .

[0190] The above examples should be understood as illustrations of the present disclosure, rather than limitations. Additional examples are also envisioned. For example, examples describing methods implemented by computers may also be provided in a corresponding manner by a computer program product or a computer-readable storage medium or system. It should be understood that the features described with respect to any one example may be used alone, or in combination with other described features, and may be used in combination with one or more features of another example, or in combination with a combination of other examples. In addition, equivalents and modifications not described above may also be adopted without departing from the scope of the present invention as defined by the appended claims. In the claims, the word "including" does not exclude other elements or operations, and the word "one" or "an" does not exclude multiple. The fact that certain features are recorded in mutually different dependent claims does not indicate that a combination of these features cannot be used to advantage. Any figure mark in the claims should not be interpreted as limiting its scope.

Claims

1. A computer-implemented method for providing biomechanical plaque data (110) to an interventional device simulation system (100), the method comprising: Receiving (S110) spectral CT data (120) representing a vascular region (130), the spectral CT data defining X-ray attenuation in the vascular region at a plurality of different energy intervals (DE 1..m ) extracting (S120) plaque data from the spectral CT data (120), the plaque data representing the spatial distribution of plaques (140) within the vascular region (130); converting (S130) the plaque data into biomechanical plaque data (110), the biomechanical plaque data representing the spatial distribution of mechanical constraints applied to the interventional device (150) in response to contact between the interventional device (150) of the interventional device simulation system (100) and the plaque (140); and outputting (S140) the biomechanical plaque data (110).

2. The computer-implemented method according to claim 1, wherein the mechanical constraints include one or more of the following: spatial restrictions on the position of the interventional device (150) applied in response to the contact between the interventional device (150) and the plaque (140), and / or the amount of friction applied to the interventional device (150) in response to the contact between the interventional device (150) and the plaque (140), and / or the amount of force feedback applied to the force feedback user interface device of the simulation system in response to the contact between the interventional device (150) and the plaque (140).

3. The computer-implemented method according to claim 1 or claim 2, further comprising: outputting a graphical representation of the vascular region (130); and wherein the graphical representation of the vascular region (130) includes a graphical representation of the plaque data and / or a graphical representation of the output biomechanical plaque data (110).

4. The computer-implemented method according to any one of claims 1-3, wherein the method further comprises: extracting lumen data from the spectral CT data (120), the lumen data representing the three-dimensional shape of one or more lumens (160) within the vascular region (130); and outputting the lumen data.

5. The computer-implemented method according to any one of claims 1-4, wherein the method further comprises: calculating the spatial distribution of plaque rupture risk based on the plaque data; and outputting a graphical representation of the plaque rupture risk.

6. An interventional device simulation system (100) for simulating the navigation of an interventional device (150) within a vascular region (130), the simulation system comprising: a user interface device (170); and one or more processors (180); wherein the user interface device (170) is configured to generate user input data for guiding the interventional device (150) within the vascular region (130) in response to received user input; and wherein the one or more processors (180) are configured to: receive the lumen data and the biomechanical plaque data (110) output by the method according to claim 4 Determine a virtual position of the interventional device (150) within the one or more lumens (160) represented by the lumen data based on the user input data generated by the user interface device (170); and wherein determining the virtual position of the interventional device (150) includes applying the mechanical constraints represented by the biomechanical plaque data (110) to the interventional device (150) in response to contact between the interventional device (150) and the plaque (140).

7. The interventional device simulation system (100) according to claim 6, wherein, the user interface device (170) includes a force feedback interface device, and wherein the force feedback interface device is further configured to receive a control signal for applying force feedback to a user of the force feedback user interface device, and wherein the mechanical constraint includes the amount of force feedback applied to the force feedback user interface device in response to the contact between the interventional device (150) and the plaque (140); and wherein the one or more processors (180) are further configured to: generate the control signal for applying the force feedback to the user of the user input device (170) based on the virtual position of the interventional device (150) within the one or more lumens (160); and wherein the control signal generated by the one or more processors (180) in response to contact between the interventional device (150) and a plaque (140) within the one or more lumens (160) is generated based on the biomechanical plaque data (110).

8. A computer-implemented method of simulating navigation of an interventional device (150) within a vascular region (130) using the interventional device simulation system (100) according to claim 6, the method comprising: receiving (S210) the lumen data and the biomechanical plaque data (110) output by the method according to claim 4; receiving (S220) user input data generated by the user interface device (170); determining (S230) a virtual position of the interventional device (150) within one or more lumens (160) represented by the lumen data based on the user input data generated by the user interface device (170); and wherein determining the virtual position of the interventional device (150) includes applying the mechanical constraints represented by the biomechanical plaque data (110) to the interventional device (150) in response to contact between the interventional device (150) and the plaque (140).

9. A computer-implemented method of generating a database of process simulation data using the interventional device simulation system (100) according to claim 6, the method comprising: recording process simulation data, for each simulation process, the process simulation data including: the user input data generated by the user interface device (170); Position data representing the virtual position of the intervention device (150) at each of a plurality of time points; The biomechanical plaque data (110) used in the simulation process; Device data representing the type of the intervention device (150) used in the simulation process; and The result data of the simulation process; and Store the process simulation data in a database.

10. A computer-implemented method for guiding a current intervention process, the current intervention process involving navigating an intervention device (150) within a vascular region (130), the method comprises: Receiving (S310) the database of the process simulation data generated by the method according to claim 9; Receiving (S320) spectral CT data (120) representing the vascular region (130) of the current intervention process; Extracting (S330) plaque data of the current intervention process from the spectral CT data (120), the plaque data representing the spatial distribution of plaques (140) within the vascular region (130) of the current intervention process; Identifying (S340) one or more similar simulation processes from the database based on the similarity between the plaque data of the current intervention process and the plaque data of the simulation process; Predicting (S350) the result metric of the current intervention process based on the result data of the identified one or more similar simulation processes; and Outputting (S360) the predicted result metric to provide the guidance for the current intervention process.

11. A computer-implemented method for providing a prediction model for controlling a robotic intervention device manipulator to navigate an intervention device (150) within a vascular region (130), the method comprises: Receiving (S510) the database of the process simulation data generated by the method according to claim 9; and Training (S520) the prediction model to control the robotic intervention device manipulator to navigate the intervention device (150) within the vascular region (130) so as to replicate the position data representing the virtual position of the intervention device (150).

12. The computer-implemented method according to claim 11, wherein, The method further comprises: Grouping the process simulation data based on the type of plaques represented by the biomechanical plaque data (110); and wherein the method includes providing a separate prediction model for controlling the robotic intervention device manipulator for each of a plurality of different types of plaques, and each prediction model is trained using the process simulation data of the corresponding type of plaques.

13. A computer-implemented method for controlling a robotic intervention device manipulator to navigate an intervention device (150) within a vascular region (130) during a current intervention device navigation process, the method comprises: Receiving a plurality of prediction models provided by the method according to claim 12; Receiving spectral CT data (120) representing the vascular region (130) of the current intervention device navigation process; Extract plaque data of the current interventional device navigation process from the spectral CT data (120), where the plaque data represents the spatial distribution of plaques (140) within the vascular region (130) of the current interventional device navigation process; Select a prediction model from the prediction models based on the similarity between the plaque data of the current interventional device navigation process and the plaque data used for training the prediction model; and Use the selected prediction model to control the robotic interventional device manipulator to navigate the interventional device (150) within the vascular region (130) during the current interventional device navigation process.

14. A computer-implemented method for simulating robotic navigation of an interventional device (150) within a vascular region (130) during a current interventional device navigation process using the interventional device simulation system (100) according to claim 4, the method comprises: Receiving (S610) a plurality of prediction models provided by the method according to claim 12; Receiving (S620) spectral CT data (120) representing the vascular region (130) of the current interventional device navigation process; Extracting (S630) the plaque data of the current interventional device navigation process from the spectral CT data (120), where the plaque data represents the spatial distribution of plaques (140) within the vascular region (130) of the current interventional device navigation process; Selecting (640) a prediction model from the prediction models based on the similarity between the plaque data of the current interventional device navigation process and the plaque data used for training the prediction model; and Using the selected prediction model to robotically control (S650) the user interface device (170) of the interventional device simulation system (100) to navigate the interventional device (150) within the vascular region (130) during the current interventional device navigation process.

15. A computer-implemented method for determining the difference between a simulated robotic interventional device navigation process and a current interventional device navigation process, the method comprises: Simulating (S710) robotic navigation of an interventional device (150) within a vascular region (130) during a current interventional device navigation process according to the method of claim 14; Controlling (S720) a robotic interventional device manipulator to navigate the interventional device (150) within the vascular region (130) during the current interventional device navigation process according to the method of claim 13; and Determining (S730) the difference between the simulated robotic navigation of the interventional device (150) and the navigation of the interventional device (150) performed by controlling the robotic interventional device manipulator to navigate the interventional device (150) within the vascular region (130) during the current interventional device navigation process.

Images