System and method for planning cardiac radiation therapy

Abstract

Systems (104) and methods (400) for predicting a location of a target cardiac region for cardiac ablation may include one or more processors that generate (402) a 3D model of a patient's heart based on a medical image of the patient, and estimating (404) one or more mechanical properties that drive movement of the heart of the patient using the 3D model of the heart and the electrophysiological data of the patient. The one or more processors may generate (406) a simulated motion pattern of the patient's heart over at least a portion of the cardiac cycle using the 3D model and the one or more mechanical properties, and identify (408) a region of interest (ROI) of the patient's heart to be radiated. The one or more processors may determine (410) a location of the ROI at a predefined instance of time within the cardiac cycle using a simulated motion pattern of the patient's heart.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to U.S. Application No. 17 / 855,171, filed on June 30, 2022, which is incorporated by reference in its entirety. Technical Field

[0003] The present application generally relates to systems and methods for planning cardiac radiation therapy. In particular, the present application relates to systems and methods for estimating a patient-specific motion pattern of a beating heart over a cardiac cycle, wherein the estimated motion pattern is used to estimate the position of a region of interest of the heart over a cardiac cycle. Background Art

[0004] Cardiac ablation is generally an invasive medical procedure for treating various cardiac conditions, such as atrial fibrillation (AFib), atrial flutter, atrial tachycardia, ventricular tachycardia (VT), atrioventricular nodal reentrant tachycardia (AVNRT), paroxysmal supraventricular tachycardia (PSVT), Wolff-Parkinson-White syndrome, or cardiac tumors. Standard radiocardiac procedures involve a doctor inserting a catheter into a patient's body to access the patient's heart. Heat or extreme cold is then applied to destroy abnormal areas of the heart and interrupt abnormal electrical signals traveling through the heart. The procedure may be risky for at least some types of patients, and it is usually necessary to monitor the patient in an intensive care unit afterwards. Some of the risks associated with standard radioablation include bleeding or infection at the site where the catheter is inserted, vascular damage, heart valve damage, new or worsening arrhythmias, slow heart rate, blood clots, stroke or heart attack, pulmonary vein stenosis, damage to the kidneys by contrast agents used during the procedure, and / or death in rare cases.

[0005] Cardiac radiotherapy ablation (also called cardiac radioablation) is a non-invasive form of cardiac ablation. Instead of a catheter, a radiation dose is used to target abnormal areas of the heart to destroy them. The use of radiation mitigates some of the risks associated with invasive procedures and provides relief to high-risk heart patients who are likely to have exhausted other options. However, cardiac radioablation has its own risks and brings its own challenges. Among them is the risk of damaging surrounding healthy tissue (especially in the heart area). This risk requires accurate radiation in terms of (one or more) radiation areas and (one or more) radiation doses.

[0006] Prior to a radiotherapy ablation procedure, radiotherapy treatment planning is performed. The purpose of treatment planning is to optimize the radiation angles and / or radiation doses for various angles to ensure that a high radiation dose is delivered to the target area and a low radiation dose is applied to intervening tissues. Summary of the invention

[0007] According to a first aspect of the present invention, there is provided a method of predicting the location of a target cardiac region for cardiac ablation as defined in claim 1 .

[0008] According to a second aspect of the invention, a system for predicting the location of a target cardiac region for cardiac ablation as defined in claim 12 is provided.

[0009] According to a third aspect of the invention, a computer readable medium as defined in claim 23 is provided.

[0010] Optional features are defined in the dependent claims.

[0011] Embodiments described herein relate to improved prediction of the location of a target cardiac region for cardiac ablation. A simulated motion pattern can be generated that accurately mimics or reproduces the motion of a patient's heart. The simulated motion pattern allows the location of the target cardiac region at various time instances of the cardiac cycle to be determined. The simulated motion pattern or simulated state of the heart can be used by a cardiac radiotherapy planning system to accurately plan radiation to the target cardiac region while minimizing radiation to surrounding or intervening healthy tissue.

[0012] According to one aspect, a method for predicting the location of a target cardiac region for cardiac ablation may include: one or more processors generating a three-dimensional (3D) model of the patient's heart based on medical images of the patient; and using the 3D model of the heart and the patient's electrophysiological data to estimate one or more mechanical properties that drive the movement of the patient's heart. The method may include: one or more processors using the 3D model and the one or more mechanical properties to generate a simulated motion pattern of the patient's heart along at least a portion of the cardiac cycle, and identifying a region of interest (ROI) of the patient's heart to be irradiated. The method may include using the simulated motion pattern of the patient's heart to determine the location of the ROI at a predefined time instance within the cardiac cycle.

[0013] In some implementations, the one or more mechanical properties include contraction and relaxation forces of the patient's heart. In some implementations, the medical image may include a sequence of image frames acquired at time intervals, and the method may also include: one or more processors using the sequence of image frames to determine an estimated motion pattern of the patient's heart over a cardiac cycle. Determining the estimated motion pattern may include tracking the displacement of a discrete set of points of the heart over the sequence of image frames, or using the sequence of image frames to estimate a displacement field using a machine learning model.

[0014] The method may include: using an estimated motion pattern of the patient's heart to determine one or more estimated positions of one or more points of the patient's heart at a time point in the cardiac cycle, and using a simulated motion pattern of the patient's heart to determine one or more simulated positions of the one or more points of the patient's heart at the time point in the cardiac cycle. The method may include: calculating one or more point distances between the one or more simulated positions and the one or more estimated positions, updating one or more mechanical properties of the patient's heart upon determining that the one or more point distances exceed a threshold, and updating the simulated motion pattern of the patient's heart based on the updated one or more mechanical properties. The method may include repeating the following steps until one or more point distances are below a threshold: determining one or more simulated positions, calculating one or more point distances, updating one or more mechanical properties, and updating the simulated motion pattern of the patient's heart.

[0015] Determining one or more estimated positions may include: one or more processors deforming a 3D model of the heart corresponding to time instance t0 according to an estimated motion pattern of the patient's heart to determine an estimated 3D model of the heart corresponding to time instance t1, and identifying one or more estimated positions on the estimated 3D model of the heart corresponding to time instance t1. The estimated 3D model indicates an estimated state of the heart at time instance t1. Determining one or more simulated positions may include: one or more processors deforming a 3D model of the heart corresponding to time instance t0 according to a simulated motion pattern of the patient's heart to determine a simulated 3D model of the heart corresponding to time instance t1, and identifying one or more simulated positions on the simulated 3D model of the heart corresponding to time instance t1. The simulated 3D model indicates a simulated state of the heart at time instance t1. The method may include: repeating the steps of determining one or more estimated positions and determining one or more simulated positions for a plurality of time instance pairs (t0, t1) corresponding to pairs of consecutive image frames in a sequence of image frames, calculating a plurality of point-wise distances between simulated positions and corresponding estimated positions across the plurality of time instance pairs (t0, t1), and updating one or more mechanical properties of the patient's heart when it is determined that the plurality of point-wise distances exceed a threshold.

[0016] In some implementations, updating one or more mechanical properties of the patient's heart may include: generating multiple second simulated motion patterns of the heart corresponding to various changes in one or more mechanical properties and various changes in electrical properties of the heart, determining one or more corresponding simulated positions of one or more points of the patient's heart at time points in the cardiac cycle for each second simulated motion pattern, calculating one or more corresponding point-wise distances between the one or more corresponding simulated positions and one or more estimated positions for each second simulated motion pattern, selecting a second simulated motion pattern from the multiple second simulated motion patterns based on the one or more corresponding point-wise distances, and updating the one or more mechanical properties according to the changes in one or more mechanical properties corresponding to the selected second simulated motion pattern. In some implementations, the ROI includes one or more segments of a standardized N-segment model, where N is an integer. In some implementations, the method may also include: modeling fixed conditions to be applied to the patient during the cardiac ablation procedure as boundary conditions incorporated into the simulated motion patterns of the patient's heart.

[0017] According to another aspect, a system for predicting the location of a target cardiac region for cardiac ablation may include one or more processors and a memory for storing computer code instructions. The computer code instructions, when executed, may cause the one or more processors to: generate a three-dimensional (3D) model of the patient's heart based on medical images of the patient, estimate one or more mechanical properties that drive motion of the patient's heart using the 3D model of the heart and electrophysiological data of the patient, generate a simulated motion pattern of the patient's heart over the cardiac cycle using the 3D model and the one or more mechanical properties, identify a region of interest (ROI) of the patient's heart to be irradiated, and determine the location of the ROI at a predefined time instance within the cardiac cycle using the simulated motion pattern of the patient's heart.

[0018] In some implementations, the one or more mechanical properties include contraction and relaxation forces of the patient's heart. In some implementations, the medical image may include a sequence of image frames acquired at time intervals, and the one or more processors may also use the sequence of image frames to determine an estimated motion pattern of the patient's heart over a cardiac cycle. In determining the estimated motion pattern, the one or more processors may track the displacement of a discrete set of points of the heart over the sequence of image frames, or use the sequence of image frames to estimate a displacement field using a machine learning model.

[0019] One or more processors may use an estimated motion pattern of the patient's heart to determine one or more estimated positions of one or more points of the patient's heart at a point in time of the cardiac cycle, and use a simulated motion pattern of the patient's heart to determine one or more simulated positions of the one or more points of the patient's heart at the point in time of the cardiac cycle. The one or more processors may calculate one or more point distances between the one or more simulated positions and the one or more estimated positions, update one or more mechanical properties of the patient's heart upon determining that the one or more point distances exceed a threshold, and update the simulated motion pattern of the patient's heart based on the updated one or more mechanical properties. The one or more processors may repeat the following steps until the one or more point distances are below a threshold: determine one or more simulated positions, calculate one or more point distances, update one or more mechanical properties, and update the simulated motion pattern of the patient's heart.

[0020] When determining one or more estimated positions, the one or more processors may deform the 3D model of the heart corresponding to time instance t0 according to the estimated motion pattern of the patient's heart to determine the estimated 3D model of the heart corresponding to time instance t1, and identify one or more estimated positions on the estimated 3D model of the heart corresponding to time instance t1. The estimated 3D model indicates an estimated state of the heart at time instance t1. When determining one or more simulated positions, the one or more processors may deform the 3D model of the heart corresponding to time instance t0 according to the simulated motion pattern of the patient's heart to determine the simulated 3D model of the heart corresponding to time instance t1, and identify one or more simulated positions on the simulated 3D model of the heart corresponding to time instance t1. The simulated 3D model indicates a simulated state of the heart at time instance t1. The one or more processors may repeat the steps of determining one or more estimated positions and determining one or more simulated positions for multiple time instance pairs (t0, t1) corresponding to pairs of consecutive image frames in the sequence of image frames, calculate multiple point distances between simulated positions and corresponding estimated positions across multiple time instance pairs (t0, t1), and update one or more mechanical properties of the patient's heart when it is determined that the multiple point distances exceed a threshold.

[0021] In some implementations, when updating one or more mechanical properties of the patient's heart, one or more processors can generate multiple second simulated motion patterns of the heart corresponding to various changes in one or more mechanical properties and various changes in the electrical properties of the heart, determine one or more corresponding simulated positions of one or more points of the patient's heart at time points in the cardiac cycle for each second simulated motion pattern, calculate one or more corresponding point-wise distances between the one or more corresponding simulated positions and one or more estimated positions for each second simulated motion pattern, select a second simulated motion pattern from the multiple second simulated motion patterns based on the one or more corresponding point-wise distances, and update the one or more mechanical properties according to the changes in one or more mechanical properties corresponding to the selected second simulated motion pattern. In some implementations, the ROI includes one or more segments of a standardized N-segment model, where N is an integer. In some implementations, the one or more processors can also model fixed conditions to be applied to the patient during the cardiac ablation procedure as boundary conditions incorporated into the simulated motion pattern of the patient's heart.

[0022] According to yet another aspect, a computer-readable medium may include computer code instructions stored thereon. The computer code instructions, when executed, may cause one or more processors to: generate a three-dimensional (3D) model of a patient's heart based on medical images of the patient, estimate one or more mechanical properties that drive motion of the patient's heart using the 3D model of the heart and electrophysiological data of the patient, generate a simulated motion pattern of the patient's heart over a cardiac cycle using the 3D model and the one or more mechanical properties, identify a region of interest (ROI) of the patient's heart to be irradiated, and determine a location of the ROI at a predefined time instance within the cardiac cycle using the simulated motion pattern of the patient's heart. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 A block diagram illustrating a computer environment for implementing the methods and processes described herein according to one embodiment is shown.

[0024] Figure 2 is a block diagram depicting one implementation of a system architecture according to one embodiment.

[0025] Figure 3 is a block diagram of a system for predicting the location of a target cardiac region, according to one embodiment.

[0026] Figure 4 is a flow chart illustrating a method of predicting the location of a target cardiac region according to one embodiment.

[0027] Figure 5 A diagram illustrating estimated and simulated states of a heart according to one embodiment is shown.

[0028] Fig. 6A and Figure 6B Shown is a graph illustrating the distance between an estimated state and a simulated state of a heart according to one embodiment.

[0029] For illustrative purposes, some or all of the drawings are schematic representations. The foregoing information and the following detailed description include illustrative examples of various aspects and implementations, and provide an overview or framework for understanding the nature and characteristics of the claimed aspects and implementations. The accompanying drawings provide illustration and further understanding of the various aspects and implementations, and are incorporated into and constitute a part of this specification. DETAILED DESCRIPTION

[0030] The following is a more detailed description of various concepts and implementations thereof related to methods, apparatus, and systems for predicting the location of a cardiac region for planning cardiac radiotherapy ablation. The various concepts introduced above and discussed in more detail below can be implemented in any of a variety of ways, as the concepts described are not limited to any particular implementation. Examples of specific implementations and applications are provided primarily for illustrative purposes.

[0031] Compared to catheter-based ablation, cardiac radiotherapy ablation is expected to be a safer and more effective ablation strategy, mainly due to its non-invasive nature and the potential ability to target the full thickness of the myocardium. However, in order to make this prospect a reality, various challenges still need to be overcome. These challenges are mainly due to the strict requirements for radiation accuracy. The heart consists of multiple risk structures (including coronary arteries and valves). These structures must be shielded from radiation (or subjected to relatively low radiation) to avoid permanent and severe damage to the heart. At the same time, the treatment aims to apply a sufficiently high radiation dose to abnormal areas of the heart in order to destroy or at least change the characteristics (e.g., electrophysiological properties) of these areas. This compromise makes radiation therapy in general an obscure and complex task, and even more so for the heart.

[0032] During radiotherapy planning, various parameters of the external radiation beam are determined to achieve the desired radiation dose distribution. These parameters include, for example, the angle and intensity of the radiation beam. The external beam parameters depend on the size, shape and position of the target area at the time of radiation, the geometry and nature of the surrounding or intervening tissues, and the prescribed dose. However, the heart is an organ that moves continuously, and radiation is applied while the heart is moving. In addition, the movement of various regions of the heart (including the target area) is much more complex than the movement of other organs (for example, the brain or spine, among other organs), for which radiation therapy has been proven to be an effective option. In addition, radiation therapy has the potential to cause toxicity to healthy heart tissue. The target area representing the abnormal area to be radiated is also referred to as a region of interest (ROI) in this article.

[0033] Technical problems associated with cardiac radiotherapy ablation involve the difficulty of accurately targeting a region of interest, and the challenge of determining optimal radiation levels to maximize ablation success and minimize toxicity. In the present disclosure, targeting of a region of interest is addressed by describing systems, methods, and apparatus for accurately predicting the position of a target region over at least a portion of the cardiac cycle. The determined position of the ROI can be used during radiotherapy planning to ensure accurate targeting of the ROI. In the prior art, cardiac motion is not accurately estimated. As a result, an overestimated target volume is often used to account for the uncertainty in the position of the target volume (or ROI) at any time instance in the cardiac cycle.

[0034] Figure 1 An example computer environment 100 for planning cardiac radiation therapy according to an example embodiment is illustrated. Briefly, the computer environment 100 may include a cardiac radiation therapy planning system 102, a region of interest (ROI) localization system 104, a database 106, and / or a communication network 108. The cardiac radiation therapy planning system 102 may include an imaging device 110, an electrophysiology system 114, and / or one or more computing devices 112. The ROI localization system 104 may include one or more computing devices, such as computing devices 116a and 116b, which are referred to herein individually or collectively as (one or more) computing devices 116. The (one or more) computing devices 116 are configured to predict the location of the ROI. The cardiac radiation therapy planning system 102, the ROI localization system 104, and the database 106 may be communicatively coupled to each other via the communication network 108.

[0035] The communication network 108 may include a local area network (LAN), a wireless local area network (WLAN), a metropolitan area network (MAN), a wide area network (WAN), the Internet, a cellular network, other types of networks, or combinations thereof. The network 108 may include both wired and wireless communications according to one or more standards and / or via one or more transport media. Communications through the network 108 may be performed according to various communication protocols such as the Transmission Control Protocol and Internet Protocol (TCP / IP), the User Datagram Protocol (UDP), and the IEEE communication protocol.

[0036] The cardiac radiation therapy planning system 102 may include one or more imaging devices 110, an electrophysiology system 114, and one or more computing devices 112. The imaging device 110 may include, among other things, a computed tomography (CT) scanner, a magnetic resonance (MR) scanner, a positron emission tomography (PET) scanner, or a combination thereof. The imaging device 110 may acquire medical images of the patient's heart along at least a portion of the cardiac cycle. In some implementations, the imaging device 110 may acquire a sequence of images depicting heart motion or deformation along the cardiac cycle or a portion of the cardiac cycle. The acquired medical images may include CT images, computed tomography angiography (CTA) images, MR images, PET images, other types of medical images, or a combination thereof.

[0037] The electrophysiology system 114 is configured to perform an electrophysiology study of the patient, for example, to evaluate the electrical system of the heart and diagnose abnormal heartbeats or arrhythmias. The electrophysiology system 114 may include one or more catheters, a plurality of wire electrodes, and a computing device connected to the wire electrodes to record electrical signals. The doctor inserts the catheter into a vein in the patient's groin, and then inserts the wire electrodes into the patient's heart via the catheter and the vein. The natural electrical impulses of the heart may travel through the wire electrodes to be recorded by the computing device of the electrophysiology system 112. In some implementations, the electrophysiology system 114 may include a plurality of electrocardiogram (ECG) electrodes to be placed on the surface of the patient's body (e.g., chest) via a multi-electrode vest. The natural electrical impulses of the heart may travel through the wire electrodes to be recorded by the computing device of the electrophysiology system 114. The computing device of the electrophysiology system 114 may send electrical signals through the electrodes to stimulate the heart tissue in an attempt to cause an abnormal heart rhythm.

[0038] The computing device 112 may receive medical images from the imaging device 110 and electrophysiology study data from the electrophysiology system 114. The computing device 112 may also obtain other medical data of the patient, such as electrocardiogram (ECG) data, blood pressure data, and / or other patient data. The computer device 112 may be configured to run or perform radiation simulations as part of radiation therapy planning. A radiation therapy planner may use the computer device 112 to simulate one or more radiation therapy parameter sets to determine which parameter set results in a desired radiation dose distribution. The computer device 112 may also send the acquired patient data, such as medical images and electrophysiology study data, to the database 106.

[0039] The ROI positioning system 104 (or the corresponding (one or more) computing devices 116) can be configured to use the acquired medical images and electrophysiological study data of the patient to estimate the true motion pattern of the heart or ROI, which will reflect the dynamic position of the ROI across at least a portion of the cardiac cycle. When estimating the motion pattern of the heart or ROI, the ROI positioning system 104 estimates and uses the mechanical properties of the heart that drive the movement of the heart during the cardiac cycle or a portion of the cardiac cycle, such as contraction force / pressure and relaxation force / pressure. The ROI positioning system 104 (or the corresponding (one or more) computing devices 116) can generate a simulation model that mimics the motion pattern of the heart and ROI along at least a portion of the cardiac cycle. The following is about Figures 3 to 6B The functional properties of the ROI localization system 104 (or corresponding computing device(s) 116 ) are discussed in further detail.

[0040] In some implementations, (one or more) computing devices 116 may be configured to execute computer instructions to perform any of the methods described herein or operations thereof. (One or more) computing devices 116 may generate and display an electronic platform to display information indicating or associated with a motion pattern of the heart and / or ROI. The electronic platform may include a graphical user interface (GUI) for receiving input data and / or displaying a predicted motion pattern and / or position of the heart or ROI. An example of an electronic platform generated and hosted by (one or more) computing devices 106 may be a web-based application or a website configured to be displayed on different electronic devices such as a mobile device, a tablet computer, a personal computer, etc.

[0041] although Figure 1 A network-based implementation is shown, but it should be noted that the methods described herein can be implemented by a single computing device that receives medical images and electrophysiological data of a patient and predicts the motion pattern and / or dynamic position of an ROI according to the methods described herein. The computer environment 100 is not necessarily limited to the components described herein and can include additional or alternative components (not shown for brevity) that should be considered within the scope of the embodiments described herein. For example, the computer environment 100 can include additional or alternative databases, such as additional or alternative databases within the cardiac radiation therapy planning system 102 or within the ROI positioning system 104. The number of computing devices within the cardiac radiation therapy planning system 102 or within the ROI positioning system 104 can also vary according to various implementations.

[0042] refer to Figure 2, shows a block diagram of one implementation of a system architecture of a computing system 200 that can be used to implement the methods described herein, according to an example embodiment. The computing system 200 may include a computing device 202. The computing device 202 may represent Figure 1 The computing device 202 may include, but is not limited to, a computed tomography (CT) scanner, a medical linear accelerator device, a desktop computer, a laptop computer, a hardware computer server, a workstation, a personal digital assistant, a mobile computing device, a smart phone, a tablet computer, or other types of computing devices. The computing device 202 may include one or more processors 204 for executing computer code instructions, a memory 206, and a bus 208 that communicatively couples the processor 204 and the memory 206.

[0043] The one or more processors 204 may include a microprocessor, a general purpose processor, a multi-core processor, a digital signal processor (DSP) or a field programmable gate array (FPGA), an application specific integrated circuit (ASIC) or other type of processor. The one or more processors 204 may be communicatively coupled to the bus 208 for processing information. The memory 206 may include a main memory device 210, such as a random access memory (RAM), other dynamic storage device, coupled to the bus 208, for storing information and instructions to be executed by the processor 204. The main memory device 210 may be used to store temporary variables or other intermediate information during the execution of instructions (e.g., related to the methods described herein (such as method 400)) by the processor 204. The computing device 202 may include a read-only memory (ROM) 212 or other static storage device coupled to the bus 208 for storing static information and instructions for the processor 204. For example, the ROM 212 may store, for example, a medical image of a patient received as input. The ROM 212 may store computer code instructions related to the methods described herein or representing an implementation of the methods. A storage device 214, such as a solid-state device, magnetic disk, or optical disk, may be coupled to bus 208 for storing (or providing as input) information and / or instructions.

[0044] The computing device 202 may be communicatively coupled to or may include an input device 216 and / or an output device 218. The computing device 202 may be coupled to the output device 218 via the bus 208. The output device 218 may include a display device for displaying information to a user, such as a liquid crystal display (LCD), a thin film transistor LCD (TFT), an organic light emitting diode (OLED) display, an LED display, an electronic paper display, a plasma display panel (PDP), or other display, etc. The output device 218 may include a communication interface for transmitting information to other external devices. An input device 216, such as a keyboard including alphanumeric and other keys, may be coupled to the bus 208 for transmitting information and command selections to the processor 204. In another implementation, the input device 216 may be integrated into a display device, such as in a touch screen display. The input device 216 may include a cursor control, such as a mouse, a trackball, or cursor direction keys, which is used to transmit direction information and command selections to the processor 204 and to control cursor movement on the display device.

[0045] According to various implementations, the methods or corresponding operations described herein may be implemented as an arrangement of computer code instructions executed by (one or more) processors 204 of the computing system 200. The arrangement of computer code instructions may be read into the main memory device 210 from another computer-readable medium such as ROM 212 or storage device 214. The execution of the arrangement of computer code instructions stored in the main memory device 210 may cause the computing system 200 to perform the methods or operations described herein. In some implementations, one or more processors 204 in a multi-processor arrangement may be used to execute computer code instructions representing implementations of the methods or processes described herein. In some other implementations, hard-wired circuits may be used instead of software instructions or in combination with software instructions to implement illustrative implementations of the methods or operations described herein. In general, implementations are not limited to any specific combination of hardware circuits and software. The functional operations described in this specification may be implemented in other types of digital electronic circuits, in computer software, firmware, hardware, or a combination thereof.

[0046] Reference now Figure 3, a block diagram of a ROI positioning system 104 for predicting the location of a target cardiac region according to one embodiment is shown. According to one embodiment, the ROI positioning system 104 is a system for accurately predicting the location of a target cardiac region. In short, the ROI positioning system 104 may include a geometric model generator 302, a motion pattern estimator 304, a mechanical property estimator 306, a motion pattern simulator 308, and a motion pattern comparator 310. The ROI positioning system 104 may include or may be connected to a database 106. The database 106 may include image data 312, electrophysiology data 314, and / or other patient data 316. The image data 312 may include a CT image, a CTA image, and / or other medical images of the patient. The electrophysiology data 314 may include electrophysiology study data of the patient, which depicts the electrical activity of the patient's heart along at least one cardiac cycle or a portion thereof. Other patient data 316 may include blood pressure data, ECG data, medical history data, demographic data, or a combination thereof.

[0047] Each of components 302, 304, 306, 308, and / or 310 may be implemented as a software component, a hardware component, a firmware component, or a combination of software, firmware, and / or hardware. For example, any of these components may be implemented as computer code instructions that are executed by one or more processors (e.g., processor 204) to perform corresponding functional steps or processes. Any of components 302, 304, 306, 308, and / or 310 may be implemented as digital circuits. Figure 4 The functional steps or processes associated with each of these components are described in further detail.

[0048] Figure 4 A flow chart illustrating one embodiment of a method 400 for predicting the location of a target cardiac region for cardiac radiotherapy ablation according to an example embodiment is shown. Briefly, the method 400 may include generating a geometric model of a patient's heart based on medical images of the patient (step 402), and estimating one or more mechanical properties of the patient's heart (step 404). The method 400 may include generating a simulated motion pattern of the patient's heart along at least a portion of a cardiac cycle (step 406), identifying a region of interest (ROI) of the patient's heart to be irradiated (step 408), and determining the location of the ROI at a predefined time instance within the cardiac cycle (step 410). The method 400 may be implemented by computing device(s) 116 or processor(s) thereof (e.g., processor 204).

[0049] Return to reference Figures 1 to 4, method 400 may include computing device(s) 116 or processor(s) 204 generating a geometric model of a patient's heart based on medical images of the patient (step 402). Prior to a cardiac ablation procedure, imaging device 110 may acquire image data 312 of the patient. Image data 312 may include cardiac gated CTA images, CT images, MR images, other types of medical images, or combinations thereof. Similarly, electrophysiology system 114 may record electrophysiology study data 314 of the patient depicting the electrical activity of the patient's heart. Computing device 112 may store image data 312 and electrophysiology data 314 in database 106.

[0050] A doctor or other medical professional may mark or outline the ROI on one or more of the acquired medical images of the patient, for example, based on the image data 312 and the electrophysiology data 314. For example, the doctor may manually identify or mark the boundaries of the ROI to be irradiated (e.g., on a display of the computing device 112). In some implementations, the computing device 112 may process the marked (one or more) images, for example, using an image segmentation algorithm, an object identification algorithm, other image processing algorithms, or a combination thereof, to refine the boundaries of the ROI. The computing device 112 may use the refined (or originally marked) ROI boundaries to identify the ROI in other unmarked images of the patient. In some implementations, the doctor may mark the ROI on all acquired medical images.

[0051] In some implementations, the ROI may include or may be one or more segments of a standardized N-segment model, where N is an integer. For example, the standardized N-segment model may include a standardized AHA 17-segment heart model. The standardized N-segment model represents a standard geometric description of the heart. The computing device 112 may identify the segments based on a parametric model of the heart chamber, for example, as described in "Four-Chamber Heart Modeling and Automatic Segmentation for 3-D Cardiac CT Volumes Using Marginal Space Learning and Steerable Features" (IEEE Transactions on Medical Imaging 27, no. 11 (November 2008) 1668-81). In some implementations, other standardized geometric models of the heart may be used. An operator of the computing device 112 may identify the ROI as one or more segments of the standardized N-segment model, for example, via a user interface displayed by the computing device 112.

[0052] The medical images acquired by the imaging device 110 may be two-dimensional images. The processor 204 or the geometric model generator 302 may use the medical image data 312 to generate a patient-specific geometric model of the patient's heart. The patient-specific geometric model may include a three-dimensional (3D) model that depicts anatomical characteristics of the patient's heart, such as shape, size, and / or various anatomical regions. For example, the processor 204 or the geometric model generator 302 may use the medical image of the patient to generate a 3D mesh of the patient's heart. The processor 204 or the geometric model generator 302 may add one or more layers on top of the 3D mesh to reflect different regions or segments of the patient's heart.

[0053] In some implementations, the geometric model generator 302 can be integrated within the cardiac treatment planning system 102. For example, the imaging device 110 or the computing device 112 can generate a 3D model of the patient's heart and store the generated 3D model in the database 106. The processor 204 or the ROI positioning system 104 can obtain or access the generated 3D model from the database 106. In some implementations, the processor 204 or the geometric model generator 302 can register the electrophysiology data with the generated 3D model of the patient's heart. In some implementations, the processor 204 or the simulation model generator 302 can combine the electrophysiology data with the generated 3D model(s) of the patient's heart, as described in U.S. Pat. No. 9,463,072, the contents of which are incorporated herein by reference.

[0054] The processor 204 or the motion pattern estimator 304 may use the image data 312 to determine an estimated motion pattern of the patient's heart along a cardiac cycle or a portion of a cardiac cycle. For example, the processor 204 or the geometric model generator 302 may generate a plurality of 3D models depicting various deformation states of the patient's heart along a cardiac cycle or a portion of a cardiac cycle. Given a sequence of acquired medical images of the patient's heart, the processor 204 or the geometric model generator 302 may generate a corresponding 3D model for each acquired medical image, the 3D model representing the deformation state (or simply the state of the heart) of the heart at a time instance (a time instance of the cardiac cycle) at which the medical image was acquired. The plurality of 3D models of the heart corresponding to different time instances of the cardiac cycle may be viewed as a four-dimensional (4D) model of the patient's heart, wherein the fourth dimension represents time. The processor 204 or the motion pattern estimator 304 may track the displacement of a discrete set of points of the heart on the sequence of generated 3D models (or 4D models) of the patient's heart. The processor 204 or the motion pattern estimator 304 may select a set of points in one of the generated 3D models of the patient's heart and employ motion tracking techniques to track the selected set of points on a sequence of 3D models. The processor 204 or the motion pattern estimator 304 may select the set of points using mesh segmentation. The processor 204 or the motion pattern estimator 304 may select a set of points representing landmarks of the heart for motion tracking. In some implementations, the processor 204 or the motion pattern estimator 304 may select a set of points in the acquired images (or image frames) and track the selected points on a sequence of acquired image frames (e.g., two-dimensional (2D) images). The processor 204 or the motion pattern estimator 304 may map the set of points to a corresponding set of points in the sequence of 3D models of the patient's heart.

[0055] In some implementations, the processor 204 or the motion pattern estimator 304 can estimate the motion pattern of the patient's heart by estimating a displacement field based on image voxel values ​​of a sequence of 3D models or a sequence of acquired image frames. The processor 204 or the motion pattern estimator 304 can use deep learning techniques to estimate the displacement field, for example, as described in "Learning a Probabilistic Model for Diffeomorphic Registration" (IEEE Transactions on Medical Imaging 38, no. 9 (September 2019): 2165-76). The estimated motion pattern depicts the movement of the patient's heart along the cardiac cycle or a portion of the cardiac cycle based on the acquired medical images of the patient.

[0056] Method 400 may include estimating one or more mechanical properties of the patient's heart (step 404), and generating a simulated motion pattern of the patient's heart along at least a portion of the cardiac cycle (step 406). The processor 204 or mechanical property estimator 306 may use the electrophysiological data 314 and the generated geometric (or 3D) model of the patient's heart to estimate the mechanical properties of the patient's heart at various time instances of the cardiac cycle (or a portion of the cardiac cycle). The processor 204 or mechanical property estimator 306 may also use other patient data 316 to estimate the mechanical properties of the patient's heart. The processor 204 or mechanical property estimator 306 may analyze the electrophysiological data 314 (and possibly other patient data 316) to extract measurements of electrical and mechanical activity in the patient's heart. The measurements of electrical activity may include: total activation time of the left or right ventricle, measurements extracted from ECG data (such as QRS duration), electrical axis, QT interval duration, point activation time, or voltage values ​​measured by a catheter device as part of electroanatomical mapping. The measurements of mechanical activity may include one or more of blood pressure acquired invasively or non-invasively (e.g., via an arm cuff), mechanical strain derived from strain imaging, displacement values ​​derived from imaging. The processor 204 or mechanical property estimator 306 may simulate corresponding measurements of electrical and mechanical activity and compare them to the measurements extracted from the data. The processor 204 or mechanical property estimator 306 may modify the electrical and mechanical properties of the patient's heart to minimize the difference between the simulated measurements of electrical and mechanical activity and the measurements based on the data.

[0057] In a second step, the processor 204 or the mechanical property estimator 306 uses the electrical properties modified in the first step to modify the mechanical properties of the patient's heart to minimize the difference between the simulated measurement results of mechanical activity and the data-based measurement results. Additional methods of modifying the electrical and mechanical properties of the patient's heart to minimize the difference between the simulated measurement results of electrical and mechanical activity and the data-based measurement results are disclosed in U.S. Patents 9,129,053, 9,245,091, and 10,733,910. The mechanical properties may include the contractile force or pressure that drives the movement of the patient's heart during the contraction phase and the relaxation (or elastic) force or pressure that drives the movement of the patient's heart during the relaxation phase. The contractile force / pressure or relaxation force / pressure at a given time instance within the cardiac cycle is related to or depends on the electrical activity of the heart at the same time instance. In particular, the electrical activity of the heart triggers or generates the contractile force / pressure or relaxation force / pressure, which in turn drives the movement of the heart. When the electrical activity of the heart changes over time, the force / pressure that drives the movement of the heart also changes over time.

[0058] Given the estimated force / pressure at a given time instance, the processor 204 or the motion pattern simulator 308 can be configured to determine the motion produced by the estimated force / pressure at different points or regions of the heart. In particular, given the force / pressure distribution on the patient's heart at the time instance, the processor 204 or the motion pattern simulator 308 can calculate the displacement of various regions or segments of the 3D model of the patient's heart (e.g., the 3D model corresponding to the time instance) based on the local force or pressure. The calculated displacement can represent the movement of various regions or segments of the 3D model over a time interval, such as the time interval between two consecutive 3D models or the time interval between two consecutive image frames.

[0059] Typically, given a generated 3D model of the patient's heart at time instance t0, the processor 204 or the motion pattern simulator 308 can calculate the displacements of various regions or segments of the 3D model between time instance t0 and time instance t1 (t1>t0) based on the estimated mechanical properties of the patient's heart at time instance t0 or within the time interval [t0 to t1]. In some implementations, the processor 204 or the mechanical property estimator 306 can estimate the mechanical properties at discrete time points and assume that the mechanical properties are constant within the interval between consecutive time points. The processor 204 or the motion pattern simulator 308 can deform the 3D model corresponding to time instance t0 according to the calculated displacement to determine a simulated state (or simulated deformed state) of the patient's heart at time instance t1.

[0060] In some implementations, the mechanical property estimator 306 and the motion pattern simulator 308 may operate together in an iterative manner to estimate the mechanical properties of the heart and simulate the motion pattern. For example, the processor 204 or the mechanical property estimator 306 may use the generated 3D model and electrophysiological data 314 to estimate the mechanical properties of the patient's heart, and the processor 204 or the motion pattern simulator 308 may use the 3D model and electrophysiological data 314 to generate the motion pattern of the patient's heart. The processor 204 may also use other patient data 316 to estimate the mechanical properties of the patient's heart and / or generate the motion pattern of the patient's heart. The processor 204 or the motion pattern comparator 310 may compare the estimated motion pattern generated by the motion pattern estimator 304 and the simulated motion pattern generated by the motion pattern simulator 308. If the estimated motion pattern and the simulated motion pattern are not similar enough (e.g., based on one or more thresholds), the processor 204 or the mechanical property estimator 306 may adjust the estimated mechanical properties, and the processor 204 or the motion pattern simulator 308 may determine a new simulated motion pattern based on the adjusted mechanical properties. The motion pattern comparator 310 may compare the estimated motion pattern to the new simulated motion pattern. The processor 204 may repeat these steps (eg, adjusting mechanical properties, updating the simulated motion pattern, and comparing the estimated motion pattern to the simulated motion pattern) until the two motion patterns are sufficiently close.

[0061] When comparing the estimated motion pattern and the simulated motion pattern, the processor 204 or the comparator 310 may select or identify one or more estimated states (also referred to herein as ground truth positions or ground truth states) of the patient's heart based on the generated 3D model and the estimated motion pattern. Each estimated state represents a deformation and / or position state of the patient's heart determined by applying the estimated motion pattern to the generated 3D model of the patient's heart. For example, the estimated state S i,e It can be represented by applying the estimated motion pattern to the generated 3D model of the patient's heart until the time instance t of the cardiac cycle i The deformation and / or position state of the patient's heart is determined, where i is an integer. The processor 204 or the comparator 310 can select or identify one or more simulated states by applying the simulated motion pattern to the generated 3D model of the patient's heart. Each simulated state corresponds to one of the estimated states. For example, the simulated state S i,s and the estimated state S i,e represents the patient's heart at time instance t i Each of the estimated state and the simulated state represents a deformation of the generated 3D model. The processor 204 or the motion mode comparator 310 may compare the simulated state S i,s With the estimated state S i,eThe processor 204 or the motion mode comparator 310 can convert the simulated state S i,s With the estimated state S i,e The point-to-point distance, point-to-curve (also called point-to-surface) distance, or curve-to-curve (also called surface-to-surface) distance between the two is compared with a threshold. If the distance (or distances) is less than the threshold, the simulation state S i,s and the estimated state S i,e If the estimated motion pattern is determined to be sufficiently similar, the steps of adjusting the mechanical property, updating the simulated motion pattern, and comparing the estimated motion pattern to the simulated motion pattern are repeated in another iteration. In the iterative method, the processor 204 can set an initial estimate of the mechanical property based on random guesswork, literature data, and / or patient information (e.g., demographic and / or medical information).

[0062] Reference now Figure 5 , a diagram illustrating estimated and simulated states of a patient's heart is shown according to an example embodiment. 3D shape 502 represents an example of a 3D model (or geometric model) of the patient's heart generated at step 402. 3D shape 504 represents the state of the patient's heart generated by (e.g., until time instance t of the cardiac cycle) i An example of an estimated state of a patient's heart generated by applying an estimated motion pattern of t to 3D model 502. 3D shape 506 represents a state of a patient's heart generated by applying an estimated motion pattern of t to 3D model 502. i 5. An example of a simulated state of a patient's heart generated by applying a simulated motion pattern of 3D model 502. After multiple iterations of adjusting mechanical properties and updating the simulated state, expected simulated state 506 converges toward estimated state 504.

[0063] Reference now Fig. 6A and Figure 6B , shows a diagram illustrating the distance between an example estimated state and an example simulated state of a heart according to one embodiment. Fig. 6A , curve 602 is a 2D representation of the estimated state of the heart, and curve 604 is a 2D representation of the simulated state of the heart. The arrow between the two curves 602 and 604 represents the distance between the two curves. These distances can be point distances (e.g., distances between pairs of points, where each pair includes a point of the simulated state 604 and a point of the estimated state 602), point-to-surface distances, or surface-to-surface distances. The motion pattern comparator 310 can compare the distance between the estimated state 602 and the simulated state 604 with a threshold, and if the distance (or its maximum value) is below the threshold, it is determined that the two states are similar, otherwise continue to adjust the estimated mechanical properties and update the simulated motion pattern until the two states are sufficiently similar. Figure 6BIllustrated is another example of a 2D representation of an estimated state 606 and a 2D representation of a simulated state 608. The arrow between the curves 606 and 608 represents the distance between the two states of the patient's heart.

[0064] When comparing the distance between the estimated state and the simulated state, the processor 204 or the motion pattern comparator 310 can evaluate the distance between the estimated state and the corresponding simulated state at any given time along the cardiac cycle and compare each of the evaluated distances with a threshold value. In some implementations, the processor 204 or the motion pattern comparator 310 can average the distances over multiple time points in the cardiac cycle and compare the average distance with the threshold value.

[0065] In some implementations, the processor 204 may perform a comparison of the estimated heart state and the simulated heart state at separate time intervals corresponding to the time range between consecutive image frames. For two time instances t1 and t2 corresponding to two consecutive image frames F1 and F2, the processor 204 or the motion pattern estimator 304 may deform the 3D model of the patient's heart corresponding to the time instance t1 according to the estimated motion pattern to determine the estimated state of the heart at the time instance t2. Then, the processor 204 deforms the 3D model of the patient's heart corresponding to the time instance t1 according to the simulated motion pattern to determine the simulated state of the heart at the time instance t2. The processor 204 may iteratively adjust the mechanical properties of the heart and update the simulated motion pattern until the estimated state and the simulated state at the time instance t2 are sufficiently similar (e.g., the distance is below a threshold). In other words, the mechanical properties of the heart tissue are adjusted to minimize the distance between the estimated state and the simulated state at the time instance t2. The processor 204 may perform a comparison of the estimated heart state and the simulated state at the time instance t2 for two time instances t1 and t2 corresponding to the two consecutive image frames F1 and F2. i ,F i+1 ) for various time instances (t i ,t i+1 ) This process is repeated. This method results in multiple estimates of the mechanical properties of the cardiac tissue over the cardiac cycle (e.g., for each time instance (t i ,t i+1 ) of the estimated mechanical properties. The processor 204 may define the global optimum as an average or median value.

[0066] In some implementations, the processor 204 may employ a plurality of random variations of mechanical and electrical properties of cardiac tissue that are initially estimated based on available measurements. The processor 204 may generate a plurality of simulated motion patterns based on the plurality of random variations of mechanical and electrical properties of cardiac tissue. The processor 204 may compare each of the plurality of simulated motion patterns to the estimated motion pattern based on the distance between the simulated states associated with the different simulated motion patterns and the estimated state. The processor 204 may select the simulated motion pattern that minimizes the distance between the corresponding simulated state and the estimated state.

[0067] In some implementations, the processor 204 or the motion pattern simulator 308 may employ a machine learning model. The machine learning model may use a database of synthetically generated cardiac motions generated using a 3D model of the patient's heart, electrophysiological data 314, and other data 316. The database of synthetically generated cardiac motions may be generated using patient-specific data (e.g., the patient's image data 312 and physiological data 314) or using data from multiple patients. For example, a database of synthetically generated cardiac motions may be generated using a 3D model of the heart or a combination of multiple models from multiple patients and a set of values ​​representing physical (e.g., electrical and mechanical) properties of cardiac tissue randomly sampled from data from multiple patients. The machine learning model or data-driven motion model may be configured to generate a database of synthetically generated cardiac motions for each point in the heart at a given time t i+1 The position at is expressed as a function of the same point at the previous time instance t i The location at t, the physical properties of the cardiac tissue associated with that point, and the same point at time t i+1 In the synthetically generated database of cardiac motion, multiple examples of cardiac motion can be generated based on the simulated state of the heart plus random perturbations that simulate the effects of noise. The machine learning model is configured to minimize the distance between the state of the heart generated by the machine learning model and the corresponding simulated state at multiple time instances. Return to reference Figure 3 Once trained, the machine learning model can be used by the motion pattern simulator 308 instead of simulation.

[0068] Return to reference Figure 4, the method 400 may include identifying a region of interest (ROI) of the heart of the patient to be irradiated (step 408), and determining the location of the ROI at a predefined time instance within the cardiac cycle (step 410). The processor 204 may identify the ROI in a plurality of simulation states determined by applying the final simulated motion pattern to the 3D model generated at step 402. The processor 204 may map the ROI identified or marked in the acquired image or image frame into the simulation state. The processor 204 may use an image mask to indicate the location of the ROI in each of the simulation states.

[0069] The processor 204 may provide the simulated motion pattern and / or the plurality of simulated states to the cardiac radiation therapy planning system 102 or a corresponding computing device 112. In some implementations, identification of the ROI within the simulated state may be performed by the computing device 112. The computing device 112 may use the simulated motion pattern or the corresponding simulated state to determine the location of the ROI at one or more time instances of the cardiac cycle. The one or more time instances of the cardiac cycle may represent time instances during which the radiation beam will be applied to the ROI. Given the location of the ROI, the cardiac radiation therapy planning system 102 or the computing device 112 may accurately target or irradiate the ROI while avoiding surrounding or intervening tissue as much as possible.

[0070] The final simulated motion pattern (output by the ROI positioning system 104) is optimized to mimic or reproduce the real or actual motion of the patient's heart. The final simulated motion pattern is used by the cardiac radiotherapy planning system 102 or the computing device 112 to estimate the position of the ROI for radiation therapy under the conditions encountered during the radiation therapy treatment. In most clinical workflows, the patient is immobilized during radiation therapy delivery by applying rigid or vacuum-based fixation, abdominal compression, and / or a shoulder mask, with or without anesthesia. In some implementations, the computing device(s) 116 or the computing device 112 can model the fixation conditions as boundary conditions incorporated into the simulated motion pattern. The boundary conditions can be used as constraints on the displacement of the heart inside the torso. The boundary conditions can be expressed as, for example, a specified pressure on the epicardial surface or an opposite displacement along a predefined direction.

[0071] It should be noted that the examples discussed in this specification are provided for illustrative purposes and should not be construed as limiting. For example, other techniques may be used to estimate or adjust the mechanical properties of the heart. Likewise, other types of distances may be used to compare simulated states and corresponding estimated states.

[0072] Each method described in this disclosure (such as Figure 4The method 400) can be performed by computer code instructions stored on a computer readable medium. When executed by one or more processors of a computing device, the computer code instructions can cause the computing device to perform the method.

[0073] While the present disclosure has been particularly shown and described with reference to particular embodiments, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention described in the present disclosure.

[0074] Although the present disclosure contains many specific embodiment details, these should not be interpreted as limitations on the scope of any invention or what may be claimed, but rather as descriptions of features specific to specific embodiments of specific inventions. Certain features described in the context of separate embodiments in this specification may also be implemented in combination in a single embodiment. Conversely, the various features described in the context of a single embodiment may also be implemented in multiple embodiments individually or in any suitable sub-combination. In addition, although features may be described above as working in certain combinations and even initially claimed as such, one or more features from the claimed combination may be deleted from the combination in some cases, and the claimed combination may involve a sub-combination or a variation of a sub-combination.

[0075] Similarly, although operations are depicted in a specific order in the accompanying drawings, this should not be understood as requiring that such operations be performed in the specific order shown or in a sequential order, or that all of the operations shown be performed to achieve the desired result. In some cases, multitasking and parallel processing can be advantageous. In addition, the separation of the various system components in the above-described embodiments should not be understood as requiring such separation in all embodiments, and it should be understood that the described program components and systems can generally be integrated into a single software product or packaged into multiple software products.

[0076] References to "or" may be construed as inclusive such that any item described using "or" may refer to any of a single, more than one, and all of the described items.

[0077] Thus, specific embodiments of the subject matter have been described. Other embodiments are within the scope of the appended claims. In some cases, the actions recited in the claims can be performed in a different order and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing can be advantageous.

Claims

1. A method for predicting the location of a target cardiac region for cardiac ablation, the method comprising: generating, by one or more processors, a three-dimensional 3D model of a heart of a patient based on medical images of the patient; estimating, by the one or more processors, one or more mechanical properties that drive motion of the heart of the patient using the 3D model of the heart and electrophysiological data of the patient; generating, by the one or more processors, a simulated motion pattern of the heart of the patient over a cardiac cycle using the 3D model and the one or more mechanical properties; identifying, by the one or more processors, a region of interest (ROI) of the heart of the patient to be irradiated; as well as The simulated motion pattern of the heart of the patient is used to determine, by the one or more processors, a location of the ROI at a predefined time instance within the cardiac cycle. 2 . The method of claim 1 , wherein the one or more mechanical properties include contractility and relaxation force of the heart of the patient.

3. The method according to claim 1 or claim 2, wherein the medical image comprises a sequence of image frames acquired at time intervals, and the method further comprises: The sequence of image frames is used by the one or more processors to determine an estimated motion pattern of the heart of the patient over the cardiac cycle.

4. The method of claim 3, wherein determining the estimated motion mode comprises: Tracking the displacement of a discrete set of points of the heart over the sequence of image frames; or The sequence of image frames is used to estimate a displacement field using a machine learning model.

5. The method according to claim 3 or claim 4, comprising: using the estimated motion pattern of the heart of the patient to determine one or more estimated positions of one or more points of the heart of the patient at points in time of the cardiac cycle; using the simulated motion pattern of the heart of the patient to determine one or more simulated positions of the one or more points of the heart of the patient at the time points of the cardiac cycle; calculating one or more point-wise distances between the one or more simulated locations and the one or more estimated locations; updating the one or more mechanical properties of the heart of the patient upon determining that the one or more point-wise distances exceed a threshold; as well as The simulated motion pattern of the heart of the patient is updated based on the updated one or more mechanical properties.

6. The method according to claim 5, further comprising: The following steps are repeated until the one or more point-wise distances are below the threshold: determining the one or more simulated positions, calculating the one or more point-wise distances, updating the one or more mechanical properties, and updating the simulated motion pattern of the heart of the patient.

7. The method of claim 5 or claim 6, wherein determining the one or more estimated positions comprises: deforming the 3D model of the heart corresponding to time instance t0 to determine an estimated 3D model of the heart corresponding to time instance t1 according to the estimated motion pattern of the heart of the patient, the estimated 3D model indicating an estimated state of the heart at the time instance t1; as well as identifying the one or more estimated locations on the estimated 3D model of the heart corresponding to the time instance t1, and wherein determining the one or more simulated locations comprises: deforming the 3D model of the heart corresponding to the time instance t0 according to the simulated motion pattern of the heart of the patient to determine a simulated 3D model of the heart corresponding to the time instance t1, the simulated 3D model indicating a simulated state of the heart at the time instance t1; as well as The one or more simulated locations are identified on the simulated 3D model of the heart corresponding to the time instance t1.

8. The method according to claim 7, comprising: repeating the steps of determining the one or more estimated positions and determining the one or more simulated positions for a plurality of pairs of time instances (t0, t1) corresponding to pairs of consecutive image frames in the sequence of image frames; calculating a plurality of point-wise distances between simulated positions and corresponding estimated positions across the plurality of time instance pairs (t0, t1); as well as Upon determining that the plurality of point-by-point distances exceed the threshold, the one or more mechanical properties of the heart of the patient are updated.

9. The method of any one of claims 5 to 8, wherein updating the one or more mechanical properties of the heart of the patient comprises: generating a plurality of second simulated motion patterns of the heart corresponding to various changes in the one or more mechanical properties and various changes in electrical properties of the heart; for each second simulated motion pattern, determining one or more corresponding simulated positions of the one or more points of the heart of the patient at the time point of the cardiac cycle; for each second simulated motion pattern, calculating one or more corresponding point-wise distances between the one or more corresponding simulated positions and the one or more estimated positions; selecting a second simulated motion pattern from among the plurality of second simulated motion patterns based on the one or more corresponding point-by-point distances; and The one or more mechanical properties are updated based on changes in the one or more mechanical properties corresponding to the selected second simulated motion pattern.

10. A method according to any preceding claim, wherein the ROI comprises one or more segments of a normalised N-segment model, where N is an integer.

11. The method according to any preceding claim, further comprising: Fixed conditions to be applied to the patient during a cardiac ablation procedure are modeled as boundary conditions incorporated into the simulated motion pattern of the heart of the patient.

12. A system for predicting the location of a target cardiac region for cardiac ablation, the system comprising: one or more processors; as well as a memory for storing computer code instructions which, when executed, cause the one or more processors to: generating a three-dimensional 3D model of the patient's heart based on the medical image of the patient; estimating one or more mechanical properties that drive motion of the heart of the patient using the 3D model of the heart and electrophysiological data of the patient; generating a simulated motion pattern of the heart of the patient over a cardiac cycle using the 3D model and the one or more mechanical properties; identifying a region of interest ROI of the heart of the patient to be irradiated; as well as The simulated motion pattern of the heart of the patient is used to determine the location of the ROI at predefined time instances within the cardiac cycle.

13. The system of claim 12, wherein the one or more mechanical properties include contraction force and relaxation force of the heart of the patient.

14. The system of claim 12 or claim 13, wherein the medical image comprises a sequence of image frames acquired at time intervals, and the one or more processors are further configured to: The sequence of image frames is used to determine an estimated motion pattern of the heart of the patient over the cardiac cycle.

15. The system of claim 14, wherein when determining the estimated motion pattern, the one or more processors are configured to: Tracking the displacement of a discrete set of points of the heart across the sequence of image frames; or The sequence of image frames is used to estimate a displacement field using a machine learning model.

16. A system according to claim 14 or claim 15, wherein the one or more processors are configured to: using the estimated motion pattern of the heart of the patient to determine one or more estimated positions of one or more points of the heart of the patient at points in time of the cardiac cycle; using the simulated motion pattern of the heart of the patient to determine one or more simulated positions of the one or more points of the heart of the patient at the time points of the cardiac cycle; calculating one or more point-wise distances between the one or more simulated locations and the one or more estimated locations; updating the one or more mechanical properties of the heart of the patient upon determining that the one or more point-wise distances exceed a threshold; as well as The simulated motion pattern of the heart of the patient is updated based on the updated one or more mechanical properties.

17. The system of claim 16, wherein the one or more processors are configured to: The following steps are repeated until the one or more point-wise distances are below the threshold: determining the one or more simulated positions, calculating the one or more point-wise distances, updating the one or more mechanical properties, and updating the simulated motion pattern of the heart of the patient.

18. The system of claim 17, wherein when determining the one or more estimated locations, the one or more processors are configured to: deforming the 3D model of the heart corresponding to time instance t0 to determine an estimated 3D model of the heart corresponding to time instance t1 according to the estimated motion pattern of the heart of the patient, the estimated 3D model indicating an estimated state of the heart at the time instance t1; as well as identifying the one or more estimated locations on the estimated 3D model of the heart corresponding to the time instance t1, and wherein determining the one or more simulated locations comprises: deforming the 3D model of the heart corresponding to the time instance t0 according to the simulated motion pattern of the heart of the patient to determine a simulated 3D model of the heart corresponding to the time instance t1, the simulated 3D model indicating a simulated state of the heart at the time instance t1; as well as The one or more simulated locations are identified on the simulated 3D model of the heart corresponding to the time instance t1.

19. The system of claim 18, wherein the one or more processors are further configured to: repeating the steps of determining the one or more estimated positions and determining the one or more simulated positions for a plurality of pairs of time instances (t0, t1) corresponding to pairs of consecutive image frames in the sequence of image frames; calculating a plurality of point-wise distances between simulated positions and corresponding estimated positions across the plurality of time instance pairs (t0, t1); and Upon determining that the plurality of point-by-point distances exceed the threshold, the one or more mechanical properties of the heart of the patient are updated.

20. The system of any one of claims 17 to 19, wherein when updating the one or more mechanical properties of the heart of the patient, the one or more processors are configured to: generating a plurality of second simulated motion patterns of the heart corresponding to various changes in the one or more mechanical properties and various changes in electrical properties of the heart; for each second simulated motion pattern, determining one or more corresponding simulated positions of the one or more points of the heart of the patient at the time point of the cardiac cycle; for each second simulated motion pattern, calculating one or more corresponding point-wise distances between the one or more corresponding simulated positions and the one or more estimated positions; selecting a second simulated motion pattern from among the plurality of second simulated motion patterns based on the one or more corresponding point-by-point distances; and The one or more mechanical properties are updated based on changes in the one or more mechanical properties corresponding to the selected second simulated motion pattern.

21. The system of any one of claims 12 to 20, wherein the ROI comprises one or more segments of a standardized N-segment model, where N is an integer.

22. A system according to any one of claims 12 to claim 21, wherein the one or more processors are further configured to: model fixed conditions to be applied to the patient during a cardiac ablation procedure as boundary conditions incorporated into the simulated motion pattern of the heart of the patient.

23. A computer readable medium comprising computer code instructions stored thereon, which when executed cause one or more processors to: generating a three-dimensional 3D model of the patient's heart based on the medical image of the patient; estimating one or more mechanical properties that drive motion of the heart of the patient using the 3D model of the heart and electrophysiological data of the patient; generating a simulated motion pattern of the heart of the patient over a cardiac cycle using the 3D model and the one or more mechanical properties; identifying a region of interest ROI of the heart of the patient to be irradiated; as well as The simulated motion pattern of the heart of the patient is used to determine the location of the ROI at predefined time instances within the cardiac cycle.

Images