Detection and alert for device intrusion in vulnerable anatomical regions

Abstract

A system is provide for detecting intrusion of an interventional device in a vulnerable region around an aneurysm. The system includes a processor that obtains a set of medical images of the interventional device in a vasculature. The processor tracks a position of the interventional device in the vasculature based on the set of medical images and detects the aneurysm in the vasculature. The processor further accesses a risk map that defines one or more regions around the detected aneurysm. The processor determines an intrusion risk score based on the tracked position of the interventional device and the risk map. The intrusion risk score defining a risk of the interventional device contacting the detected abnormality.

Description

DETECTION AND ALERT FOR DEVICE INTRUSION IN VULNERABLE ANATOMICAL REGIONSFIELD

[0001] The following relates generally to vulnerable anatomical region mapping. More particularly, embodiments herein relate to vulnerable anatomical region mapping for interventional device treatment dynamics.BACKGROUND

[0002] In an interventional procedure, by insertion and manipulation of suitable catheters and guide wires (often called as device) at the appropriate anatomical location, the patient is treated with minimal invasion of the body. One of the clinical applications of interventional procedures is for treatment of intracranial aneurysms. Aneurysm, informally, is a bulge on the blood vessel which can be created due to the local weakening of the blood vessel. If the aneurysm ruptures (all aneurysms are potentially so), this will cause internal bleeding and the consequences for the patient are obviously harmful. Depending on the clinical decision process, during the interventional procedure, a small coil may be placed inside the aneurysm which will ensure that the aneurysm gets supported or filled up and therefore the risk of rupture is reduced greatly.

[0003] Typically, the devices are inserted not directly at the location of the aneurysm because they are not easily reachable. For example, most intracranial aneurysms occur at the major arteries at the base of the skull and are detected during diagnostic scans. Therefore, the catheters are “pushed” or threaded by making a small incision made at the patient’s wrist or groin regions. From this entry point, the main catheter is inserted and then through it, smaller wires and other small detachable coils are inserted and moved toward the target location. The device is slowly and gradually moved up toward the target anatomy by the interventionalist. The device location is tracked by fluoroscopy. Therefore, a series of low dose x-ray images are acquired live as part of the procedure to guide the interventionalist through the procedure. This is a procedure needing constant attention from the interventionalists, especially as the device nears the target location.

[0004] The following discloses certain improvements to overcome problems that exist in such treatments.SUMMARY

[0005] As discussed above, there is a need for constant attention from the interventionalists, especially as the device nears the target location. This risk is further compounded by the fact that the interventional devices may twist during the insertion process, leading to build up of stresses or tensions in the guidewire or catheter. This is also known as slack. The tension can release and cause a rapid untwisting or uncontrolled motion of the device which can lead to potential harm including rupture or dissection, especially if this happens in the vicinity of the aneurysm.

[0006] Advantageously, some embodiments herein predict a risk score of the interventional device puncturing an aneurysm by using imaging data in the live scan. More specifically, some embodiments herein describe a method to minimise the risk of aneurysm rupture during the interventional procedure. Since the risk of aneurysm rupture and complexity of procedure are closely intertwined, the method described in some embodiments also simplifies the workflow for the interventionalist. Therefore, this as a workflow improvement for the operator specifically directed toward improving patient safety.

[0007] As will be described in greater detail below, systems, apparatuses, non-transitory computer-readable storage medium, and methods discussed herein provide for interventional device intrusion support for patients, in some embodiments of the present disclosure. Such interventional device intrusion support includes obtaining a set of medical images of an interventional device in an anatomy (e.g., vasculature). The position of the interventional device is tracked in the anatomy based on the set of medical images. An abnormality (e.g., aneurysm) is detected in the vasculature. A risk map is accessed that defines one or more regions around the detected aneurysm. A risk score of the interventional device contacting (e.g., as used herein the term “contacting” or “contact” refers to coming so near or adjacent to a detected aneurysm, aneurysm neck, or aneurysm wall as to cause potential damage to the detected aneurysm) the detected aneurysm is determined based on the tracked position of the interventional device and the risk map.

[0008] In one aspect, a system includes a processor and a memory communicatively coupled to the processor. The processor is configured to obtain a set of medical images of an interventional device in a vasculature. The processor is configured to track the position of interventional device in the vasculature based on the set of medical images. The processor is further configured to detect an abnormality in the vasculature. The processor accesses a risk map that defines one or moreregions around the detected abnormality. The processor determines an intrusion risk score based on the tracked position of the interventional device and the risk map. The intrusion risk score defines a risk of the interventional device contacting the detected abnormality.

[0009] In another aspect, a method includes obtaining a set of medical images of an interventional device in a vasculature. The method further includes tracking the position of the interventional device in the vasculature based on the set of medical images. The method further includes detecting an abnormality in the vasculature. The method further includes accessing a risk map that defines one or more regions around the abnormality. The method further includes determining an intrusion risk score based on the tracked position of interventional device and the risk map. The intrusion risk score defines a risk of the interventional device contacting the detected abnormality.

[0010] In still another aspect, a non-transitory machine-readable storage having stored a computer program comprising a set of instructions, which when executed by a processor, cause the processor to: obtain a set of medical images of an interventional device in a vasculature; track position of an interventional device in the vasculature based on the set of medical images; detect an abnormality in the vasculature; accessing a risk map that defines one or more regions around the detected abnormality; and determine an intrusion risk score based on the tracked position of the interventional device and the risk map. The intrusion risk score defines a risk of the interventional device contacting the detected abnormality.

[0011] It should be appreciated that all combinations of the foregoing concepts and additional concepts discussed in greater detail below (provided such concepts are not mutually inconsistent) are contemplated as being part of the subject matter disclosed herein. In particular, all combinations of claimed subject matter appearing at the end of this disclosure are contemplated as being part of the subject matter disclosed herein. It should also be appreciated that terminology explicitly employed herein that also may appear in any disclosure incorporated by reference should be accorded a meaning most consistent with the particular concepts disclosed herein.

[0012] These and other aspects of the various embodiments will be apparent from and elucidated with reference to the embodiment s) described hereinafter.BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The various advantages of the embodiments will become apparent to one skilled in the art by reading the following specification and appended claims, and by referencing the following drawings, in which:

[0014] FIG. 1 is an illustration of a block diagram of an example interventional device intrusion support system according to an embodiment;

[0015] FIG. 2 is an illustration of a block diagram of another example interventional device intrusion support system according to an embodiment;

[0016] FIG. 3 is an illustration of an interventional treatment of an aneurysm according to an embodiment;

[0017] FIG. 4 is an illustration of a flowchart of a method for managing interventional device intrusion support according to an embodiment;

[0018] FIG. 5 is an illustration of a flowchart of a method for managing interventional device intrusion support according to an embodiment;

[0019] FIG. 6 is an illustration of a sequence diagram of a further method for managing interventional device intrusion support according to an embodiment;

[0020] FIG. 7 is an illustration of a block diagram of a computer program product according to an embodiment;

[0021] FIG. 8 is a further illustration of a system according to an embodiment;

[0022] FIG. 9 is an illustration of a hardware apparatus including a semiconductor package according to an embodiment;

[0023] FIG. 10 is a schematic diagram illustrating an example C-arm imaging system for managing interventional device intrusion support, in accordance with some aspects of the present disclosure; and

[0024] FIG. 11 is an illustration of a block diagram of another example interventional device intrusion support system according to an embodiment.DETAILED DESCRIPTION

[0025] FIG. 1 is an illustration of a block diagram of an example interventional device intrusion support system 100 according to an embodiment. In the illustrated embodiment, theinterventional device intrusion support system 100 may include a main platform 102 (e.g., also referred to as a care platform herein). In some embodiments, the main platform 102 may be embodied as a server computer or a plurality of server computers (e.g., interconnected to form a server cluster, cloud computing resource, the like, and / or combinations thereof).

[0026] In some embodiments, the main platform 102 is a care platform designed for one or more aspects of patient care. In such an example, the main platform 102 includes one or more of the following care platforms Performance Management Data Platform, Medical Asset Track & Trace System, Virtualized Imaging Solution, Centralized Care Management System, Interoperability Solution, Electronic Medical Record (EMR) platform, radiology information system (RIS), picture archiving and communication system (PACS), etc. In a patient care setting, there may be many care platforms connected to various clinical and operational data sources (e g., Health Level Seven (HL7), Fast Healthcare Interoperability Resources (FHIR), machine logs, Real-Time Location System (RTLS), sensors, etc.).

[0027] In some embodiments, a first application 104 though an Nth application 106, which may include an interventional device intrusion support application 108, may be associated with the main platform 102. As will be described in greater detail below, the operations of the interventional device intrusion support application 108 provide for receiving two-dimensional (2D) fluoroscopy sequences (or the like) from an imaging system (e.g., an interventional X-ray imaging system or the like) during treatment (e.g., endovascular coiling, flow diverter deployment, stent deployment, balloon angioplasty, atherectomy procedures, or the like).

[0028] The interventional device intrusion support application 108 then performs the following actions: to receive a set of medical images of an anatomy from the medical imager 119. An interventional device 116 is tracked in the anatomy based on the set of medical images. An abnormality (e.g., aneurysm) is detected in the anatomy (e.g., vasculature). A risk map is determined to define one or more risk regions around the detected aneurysm. A risk score of the interventional device contacting the detected aneurysm is determined based on the tracking of the interventional device and the risk map.

[0029] Additionally, or alternatively, the interventional device intrusion support system 100 may include a patient monitor 110, a sensor 112 (e.g., one or more sensors 112 that may be associated with a care facility, sensors 112 that may be associated with a patient 114, the like, and / or combinations thereof), an interventional device 116, a medical management device 118,medical imager 119, a database 120, a user interface 122 (e.g., one or more user interfaces 122 may be associated with a user 124), an information system 126, the like, and / or combinations thereof. For example, the main platform 102, the patient monitor 110, the sensor 112, the interventional device 116, the medical management device 118, medical imager 119, the database 120, the user interface 122, and / or the information system 126, may be in communication with one another via Internet based communicating, cloud based communication, wired communication, wireless communication, the like, and / or combinations thereof

[0030] In an example, the patient monitor 110 may be utilized to access patient data from and / or enter patient data to the database 120. For example, the patient monitor 110 may communicate measured patient data (e.g., via one of more of the sensors 112). In such an example, the patient monitor 110 may be configured to monitor a patient for vital signs and the like, and the patient monitor 110 may communicate such measured patient data to the database 120 or to a user interface 122. For example, the sensors 112 may determine dynamic patient condition data including patient vitals (e.g., blood pressure, pulse, temperature, respiration, and / or the like) and / or patient test results (e.g., creatine level, urine flow, potassium level, oxygen saturation, blood glucose level, carbon dioxide level, and / or the like).

[0031] In some embodiments, the patient monitor 110 may include a bedside-type monitor, a transport-type monitor, a central station-type monitor, the like, and / or combinations thereof.

[0032] In some embodiments, the sensors 112 may be minimally invasive-style sensors (e.g., by puncturing the skin, sensing through the skin, the like, and / or combinations thereof). Sensors 112 may be wired or wireless.

[0033] In some embodiments, the interventional device 116 may be utilized to access patient vasculature or other anatomical structures accessible through patient vasculature (e.g., using catheters, guidewires, sheaths, and so on) and / or to deliver treatment to the patient (e.g., using stents, coils, flow diverters, and so on). For example, the interventional device 116 may be used to navigate to and deliver treatment to an intracranial aneurysm. In such an example, the interventional device 116 may be an access sheath used to gain femoral access at the patient’s groin or to gain radial access at the patient’s wrist. In another such example, the interventional device 116 may be a guidewire and a catheter used to navigate to the intracranial aneurysm. In another such example, the interventional device 116 may be a flow diverter or a set of coils used to deliver treatment to the intracranial aneurysm.

[0034] In some embodiments, the interventional device 116 may enable the administration of one or more patient procedures (e.g., endovascular therapy).

[0035] As used herein, the term “endovascular therapy” refers to endovascular coiling, flow diverter deployment, stent deployment, balloon angioplasty, atherectomy procedures, the like, and / or combinations thereof.

[0036] In a further example, the medical management device 118 may be utilized to access patient data from and / or enter patient data to the database 120. For example, the medical management device 118 may communicate measured patient data. In such an example, the medical management device 118 may be configured to monitor medication delivery to a patient and may communicate such measured patient data to the database 120. In some embodiments, the medical management device 118 may supply and / or monitor the administration of one or more patient medications (e.g., blood pressure medications, diuretic medications, anesthesia medications, anti -coagulant medications, vitamin supplements, and / or the like) and / or other fluids (e.g., contrast, saline, and / or the like).

[0037] In some examples, the medical imager 119 may include one or more medical imaging devices, medical imaging systems, the like, and / or combinations thereof. For example, the medical imager 119 may include an interventional imaging device including an X-ray device (such as a fixed C-arm imaging system and / or a mobile C-arm imaging device), an ultrasound device, and / or the like. In another example, the medical imager 119 may include a non-interventional imaging device such as a magnetic resonance imaging (MRI) device, a computed tomography (CT) device, and / or the like. In another example, the medical imager 119 may include a radiology information system (RIS), a picture archiving and communication system (PACS), the like, and / or combinations thereof, where images acquired from interventional or non-interventional imaging systems may be stored. In such an embodiment, the medical imager 119 is associated with standardized format medical imaging information to acquire, transmit, store, retrieve, print, process, and display medical imaging information (e.g., Digital Imaging and Communications in Medicine (DICOM) data).

[0038] More specifically, FIG. 10 is a schematic diagram illustrating an example interventional imaging system 1000, in accordance with some aspects of the present disclosure. As illustrated, the system 1000 includes one or more processors 1010 that are configured to perform one or more aspects of the below-described methods. The system 1000 may consist of anX-ray imaging system 1020, which may be configured to provide a single X-ray image, or a temporal sequence of X-ray images also known as fluoroscopy images. The X-ray imaging system 1020 may also be configured to provide a temporal sequence of contrast-enhanced images, from which digital subtraction angiography (DSA) images are generated by subtracting a mask image (and upon which motion compensation may be performed). In FIG. 10, X-ray imaging system 1020 is illustrated as a C-arm X-ray system with the top of the X-ray imaging system 1020 being the X-ray detector and the bottom of the X-ray imaging system 1020 being the X-ray source. The system 1000 may also include a display 1030 for displaying the acquired X-ray and / or DSA image. The system 1000 may also include a patient bed 1040. The system 1000 may also include a user interface device such as a keyboard, and / or a pointing device such as a mouse, and / or a joystick to control any of the components of the system 1000. These items may be in communication with each other via wired or wireless communication.

[0039] FIG. 11 is an illustration of a block diagram of another example interventional device intrusion support system 1100 according to an embodiment. As illustrated, the interventional device intrusion support system 1100 includes an X-ray imaging system 1120 (e.g., such as the example interventional imaging system 1000 of FIG. 10), a workstation 1140, an Artificial Intelligence (Al) controller 1170 and an Al training system 1180. The X-ray imaging system 1120 captures medical images that include imagery of an interventional medical device 1101.

[0040] In some embodiments, the interventional medical device 1101 may enable the administration of one or more patient procedures (e g., endovascular therapy). For example, such an interventional medical device 1101 is used to perform endovascular coiling, flow diverter deployment, stent deployment, balloon angioplasty therapy, atherectomy therapy, the like, and / or combinations thereof. Machine learning, artificial intelligence and / or image processing techniques can be used to segment and track the interventional medical device 1101 present in the image during live fluoroscopic imaging.

[0041] Examples of the X-ray imaging system 1120 include a fixed C-arm X-ray system such as Azurion from Koninklijke Philips N.V. or a mobile X-ray system such as Veradius from Koninklijke Philips N.V. The X-ray imaging system 1120 is a medical apparatus that generates a plurality of images during an interventional medical procedure and may include an X-ray tube adapted to generate X-rays and an X-ray detector configured to acquire X-ray images.

[0042] The workstation 1140 includes a controller 1150, an interface 1153, a monitor 1155 and a touch panel 1156. The controller 1150 includes a memory 1151 that stores instructions and a processor 1152 that executes the instructions.

[0043] The Al controller 1170 includes a memory 1171 that stores instructions and a processor 1172 that executes the instructions. The Al controller 1170 may dynamically implement trained machine learning models based on images received by the workstation 1140 from the X-ray imaging system 1120. In an embodiment, the Al controller 1170 is integrated with the workstation 1140. In another embodiment, functionality of the Al controller 1170 as described herein may be performed by the controller 1150. The Al controller 1170 may include a network computer configured to receive a stream of a plurality of images from a system computer implemented by the workstation 1140 and apply the trained machine learning model to the plurality of images. The significance of the images may be determined according to a standardized metric that may be applied across different clinical sites and users. The metric used to determine the significance may vary based on the context in which the images are being analyzed. For example, the metric may vary based on subject matter being sought as triggers in the images, and the subject matter may include medical instruments, anatomical structures, motion(s), or presence of people. The significance metric may be determined by experts in the field(s) in which the machine learning model described herein is applied. Accordingly, the significance may be appropriately termed a contextual significance that reflects the context of the subject matter in the images deemed significant. The significance may be appropriately termed a predetermined significance that reflects significance previously set for the images that will be deemed significant. The significance may be appropriately termed a standardized significance that reflects significance set as a standard for a particular type of subject matter. The significance may be appropriately described in other terms consistent with the descriptions herein.

[0044] The Al training system 1180 includes an Al training controller 1181. The Al training controller 1181 may include a memory (not shown) that stores instructions and a processor (not shown) that executes the instructions. The Al training system 1180 may train machine learning models as described herein and provide the trained machine learning model to the Al controller 1170 and / or the workstation 1140 in FIG. 11. The Al training system 1180 may include a controller configured to train the machine learning model. The machine learning model may be aconvolutional neural network (CNN) model, a temporal convolutional network (TCN) model, a recurrent neural network (RNN) model, a transformer model, a Hidden Markov model (HMM), and so forth.

[0045] When executed by the processor 1152, the instructions from the memory 1151 cause the controller 1150 to perform some or all aspects of methods described herein. The controller 1150 may be a data processing controller that is configured to receive consecutive fluoroscopy images from the X-ray imaging system 1120. The fluoroscopy images may be images that are generated and stored by the X-ray imaging system 1120 during a clinical procedure. The fluoroscopy images may be further enhanced by other information from the X-ray imaging system 1120 such as a C-arm position, radiation dose, identification of procedure phase, identification of the interventional medical device 1101 in the image, image generation settings, type of scanning protocol type, as well as patient characteristics including information from the patient's electronic medical record (EMR).

[0046] The interface 1153 interfaces the workstation 1140 to the X-ray imaging system 1120. The interface 1153 may include a receiver that receives a stream of a plurality of images from the X- ray imaging system 1120. For example, the interface 1153 may be a port or an adapter that accepts a cable line from the X-ray imaging system 1120.

[0047] The monitor 1155 displays images generated by the X-ray imaging system 1120. The monitor 1155 may also display instructions or alerts for a clinician using the interventional device intrusion support system 1100.

[0048] The touch panel 1156 accepts touch instructions from a clinician, such as instructions input via a mouse or keyboard. The touch panel 1156 may also accept touch input such as via a keypad, touchpad, touchscreen, or the like.

[0049] The workstation 1140 may receive time-series interventional X-ray images such as fluoroscopy X-ray images from the X-ray imaging system 1120. The workstation 1140 may also receive digital subtraction angiography (DSA) images from the X-ray imaging system 1120. The workstation 1140 may also receive radiation exposure information, and other system information such as C-arm position, radiation settings and table position. As will be discussed in greater detail below, the workstation 1140 applies trained machine learning models to the X-ray images to detect aneurysms in the anatomy, track an interventional medical device 1101 in the anatomy, determine a risk map to define one or more risk regions around the detected aneurysm, and determine a riskscore of the interventional device contacting the detected aneurysm based on the tracking of the interventional device and the risk map.

[0050] Referring back to FIG. 1 additionally, or alternatively, in a still further example, the user interface 122 may be utilized to access patient data from and / or enter patient data to the database 120. In some embodiments, user interface 122 may be implemented via one or more formfactor devices (e.g., a monitor, a smart phone, a tablet, a laptop, a workstation, and / or the like), an interface associated with the main platform, and / or an interface associated with the patient monitor 110. Additionally, or alternatively, a care provider (e.g., user 124) may access patient data and / or enter patient data through an analog device, a non-networked patient monitor, a nonnetworked interventional device, a non-networked medical management device, the like, and / or combinations thereof.

[0051] In the illustrated embodiment, the database 120 may include one or more types of patient data. For example, the database 120 may include patient data including medical images, patient health information (PHI) (including patient age, height, weight, and so on), laboratory result data, microbiology data, medication data, vital sign data, care order data, admission discharge and transfer data, and / or the like. As used herein, the term "database" refers to a collection of data and information organized in such a way as to allow the data and information to be stored, retrieved, updated, and / or manipulated. The term "database" as used herein may also refer to databases that may reside locally or that may be accessed from a remote location (e.g., via remote network servers).

[0052] As used herein, the term "patient data" refers to clinical data, imaging, or information related to an individual patient. Patient data may include measured patient data from a medical imaging device, an analog medical device, a sensor, a patient monitor, an interventional device, a medical management device, the like, and / or combinations thereof.

[0053] In the illustrated embodiment, the information system 126 may have or have access to one or more types of patient data that are the same or in addition to the patient data of the database 120. For example, the information system 126 may be a Hospital Information System (HIS). Such a Hospital Information System (HIS) has patient data including Health Level Seven (HL7) data, Fast Healthcare Interoperability Resources (FHIR) data, the like, and / or combinations thereof. Additionally, or alternatively, in some embodiments, the information system 126 has or has access to one or more of the following information sources: an Electronic Medical Record (EMR), aradiology information system (RIS), a picture archiving and communication system (PACS), medical imaging devices, a Real-Time Location System (RTLS), sensors, machine logs, a Performance Management Data Platform, a Medical Asset Track & Trace System, a Virtualized Imaging Solution, a Centralized Care Management System, an Interoperability Solution, etc.

[0054] Additionally, or alternatively, in some embodiments, the database 120 and / or the information system 126 may include or be associated with a simulated database. In such an example, such a simulated database may store simulated or estimated patient data. For example, interventional device intrusion support system 100 may utilize some measured patient data from the patient monitor 110, the sensor 112, the medical management device 118, the medical imager 119, the user interface 122, the information system 126, and / or the database 120 to generate some other simulated or estimated patient data and store in the simulated database. The interventional device intrusion support system 100 may utilize digital twin technology to perform the estimation, for example. In such an example, such simulated or estimated patient data may be marked to indicate its simulated or estimated nature (rather than measured patient data). Additionally, or alternatively, a weight factor may be applied to the simulated or estimated patient data so that the simulated or estimated patient data may have a lower weight than corresponding measured patient data.

[0055] In some embodiments, the interventional device intrusion support system 100 may be utilized as an element of an Integrated Clinical Environment (ICE). As used herein, the “Integrated Clinical Environment (ICE)” refers to a platform to create a medical Internet of Things (loT) associated with the care of a patient. In such an embodiment, the interventional device intrusion support system 100 may support many real-time clinical decision support algorithms. Additionally, or alternatively, the interventional device intrusion support system 100 may support closed loop control algorithms of medical devices in the ICE.

[0056] For example, in some embodiments discussed herein, 2D fluoroscopy sequences are received from an imaging system (e.g., an interventional X-ray imaging system or the like) during endovascular treatment (e.g., endovascular coiling, flow diverter deployment, stent deployment, balloon angioplasty, atherectomy procedures, or the like).

[0057] An interventional X-ray imaging system includes an X-ray tube adapted to generate X- rays and an X-ray detector configured to acquire X-ray images. Examples of such systems are a fixed monoplane and biplane C-arm X-ray system, a mobile C-arm X-ray system, the like, and / orcombinations thereof. The X-ray imaging system can acquire both regular fluoroscopy (X-ray) images as well as contrast enhanced fluoroscopy images like digital subtraction angiography (DSA) images.

[0058] Machine learning, artificial intelligence and / or image processing techniques can be used to segment and track the interventional device 116 present in the image during live fluoroscopic imaging.

[0059] As will be described in greater detail below, the interventional device intrusion support system 100 and / or the interventional device intrusion support application 108 may include several operable components.

[0060] FIG. 2 is an illustration of a block diagram of another example interventional device intrusion support system 200 according to an embodiment. As illustrated, the interventional device intrusion support system 200 includes a control unit 201 with an interventional device intrusion support logic 208, an interventional device 216, a medical imager 219, a robotic control manipulator 221, the like, and / or combinations thereof. For example, the control unit 201, the interventional device 216, the medical imager 219, and / or the robotic control manipulator 221 may be in communication with one another via Internet based communication, cloud-based communication, wired communication, wireless communication, the like, and / or combinations thereof.

[0061] In the illustrated example, the control unit 201 may include a processor 202 and a memory 204 communicatively coupled to the processor 202. The memory 204 may include interventional device intrusion support logic 208 as a set of instructions. In some embodiments the interventional device intrusion support logic 208 may be implemented as software (e g., as illustrated by interventional device intrusion support application 108 in FIG. 1). In an embodiment, the interventional device intrusion support logic 208, when executed by the processor 202, implements one or more aspects of the method 500 (FIG. 5), the method 600 (FIG. 6), as will be discussed in greater detail below.

[0062] In some embodiments, the processor 202 may include a general purpose controller, a special purpose controller, a storage controller, a storage manager, a memory controller, a microcontroller, a general purpose processor, a special purpose processor, a central processor unit (CPU), a graphics processing unit (GPU), a tensor processing unit (TPU), the like, and / or combinations thereof. Further, embodiments may include distributed processing,component / object distributed processing, parallel processing, the like, and / or combinations thereof. For example, virtual computer system processing may implement one or more of the methods or functionalities as described herein, and the processor 202 described herein may be used to support such virtual processing.

[0063] In some examples, the memory 204 is an example of a computer-readable storage medium. For example, memory 204 may be any memory which is accessible to the processor 202, including, but not limited to RAM memory, registers, and register files, the like, and / or combinations thereof. References to “computer memory” or “memory” should be interpreted as possibly being multiple memories. The memory may for instance be multiple memories within the same computer system. The memory may also be multiple memories distributed amongst multiple computer systems or computing devices.

[0064] In some embodiments, the interventional device 116 may enable the administration of one or more patient procedures (e.g., endovascular therapy). For example, such an interventional device 116 is used to perform endovascular coiling, flow diverter deployment, stent deployment, balloon angioplasty therapy, atherectomy therapy, the like, and / or combinations thereof.

[0065] In some examples, the medical imager 219 includes one or more medical imaging devices, medical imaging systems, the like, and / or combinations thereof. For example, the medical imager 219 includes X-ray imaging system. Such an X-ray imaging system typically will include an X-ray tube adapted to generate X-rays and an X-ray detector configured to acquire X-ray images. For example, such an X-ray imaging system is implementable by a fixed monoplane X- ray system, a biplane C-arm X-ray system, a mobile C-arm X-ray system, the like, and / or combinations thereof.

[0066] In some embodiments, the robotic control manipulator 221 is configured to perform one or more operations in response to a risk score. For example, such operations may be utilized to inhibit movement of the interventional device 216, relieve rotational torque of the interventional device 216, relieve linear tension of the interventional device 216, the like, and / or combinations thereof.

[0067] In operation, the memory 204 stores logic (e.g., interventional device intrusion support logic 208) that includes a set of instructions executable by the processor 202, which when executed by the processor 202, cause the processor 202 to receive a set of medical images of an anatomy from the medical imager 219. An interventional device 216 is tracked in the anatomy based onthe set of medical images. An aneurysm is detected in the anatomy. A risk map is determined to define one or more risk regions around the detected aneurysm. A risk score of the interventional device contacting the detected aneurysm is determined based on the tracking of the interventional device and the risk map.

[0068] FIG. 3 is an illustration of an interventional treatment of an aneurysm according to an embodiment. As illustrated, techniques described herein are applicable to intracranial aneurysms 304. It will be understood, however, that the techniques described herein are applicable to any interventional procedure where the motion of the intervention device can cause a significant harm to the patient.

[0069] In the illustrated example, treatment of aneurysm 304 in the brain 302 (or intracranial aneurysm) is performed using interventional X-ray imaging systems. For example, the intracranial aneurysm 304 is to be "filled" in as part of the procedure with coils 300. The coils 300 are delivered via an interventional device, which includes a microcatheter 306 in the present example. This procedure is known as endovascular coiling. The microcatheter 306 is typically introduced into the aneurysm 304 via the parent vessel, which is the vessel from which the aneurysm develops.

[0070] FIG. 4 is an illustration of a flowchart of a method 400 for managing interventional device intrusion support according to an embodiment. As illustrated, method 400 includes detection of the aneurysm at operation 408. For example, the aneurysm may be detected from fluoroscopic image data 402.

[0071] In the event that no aneurysm is detected a risk score may be set to zero at operation 410. In the event that an aneurysm is detected, a risk map is determined to define one or more risk regions around the detected aneurysm. For instance, a distance-based risk map may be defined around the detected aneurysm to define a high, medium, and / or low risk region around the aneurysm at operation 412.

[0072] In some embodiments, the interventional device may be detected and tracked at operation 404. For example, the interventional device tip and length may be tracked from fluoroscopic image data 402. For example, the device location at a given time is determined at operation 406.

[0073] In some examples, a risk score (e.g., a probability of the device entering the high-risk region) may be predicted at operation 414 and assigned at operation 416 based on the motion model derived from the device tip and / or device length detection and tracking (e.g., determined atoperation 406) in combination with the distance-based risk map (e.g., determined at operation 412).

[0074] As illustrated, in method 400 Rtdenotes the “risk” or a “score” of the device rupturing the aneurysm at time t, and P is the set of learnable parameters for the risk prediction model. Examples of such models include Hidden Markov models (HMMs), temporal convolutional models (TCNs), recurrent neural networks (RNNs), transformer networks, simple statistical models, and / or combinations thereof.

[0075] One of the aspects of method 400 is that the model uses the information from the current observation of the aneurysm location along with temporal device tracking and the past states of the risk. For instance, interventional devices are tracked temporally (e.g., as devices are navigated up the aorta, through the aortic arch, and into the carotid artery) to estimate torque or tension buildup such that when the devices are near the aneurysm, the aggregate information can be used to estimate risk of sudden movement or sudden intrusion of the device into the aneurysm. Similarly, as the device is twisted or pulled back to release tension, temporal states of risk can be updated to compute the new reduced risk. These are the primary inputs to and outputs of the algorithm.

[0076] In another embodiment, the method 400 can include additional information from hardware including a robot manipulating the interventional device. For example, the extra information can be hardware information like translation data, rotation information from the robotic controller, the like, and / or combinations thereof.

[0077] In another example, a combination of two-dimensional and three-dimensional (2D / 3D) information can be used to predict the risk score. Specifically, fluoroscopy images are 2D projection images of the three-dimensional (3D) anatomy. Therefore, the risk map including a distance map or any such measure will necessarily contain errors (e.g., due to foreshortening) in the depth direction, which is not captured by 2D projection images. However, in the case that preoperative 3D image data and / or intraoperative 3D image data is available (e g., computed tomography angiography (CTA) or 3D rotational angiography (3DRA) images), registering or combining the 2D fluoroscopy data and the 3D image data can provide additional information to the algorithm.

[0078] In another embodiment, as a preventive measure to minimise harm, the system can provide the risk score back to a robotic control manipulator in a closed loop control fashion. In such a case, this algorithm can act as a soft inhibitor for the robotic controller. For instance, if therisk is too high, the robotic controller may slow down the translation of the interventional device near the aneurysm. Additionally, or alternatively, the robot could also be trained to take specific action when the device tip is close to the aneurysm and the risk of intrusion is high. That is, since the robot has information about how the device was twisted to navigate the device up to the aneurysm, it could apply opposing twist to the device to release the tension before the device is pushed closer to the aneurysm.

[0079] Additional details regarding such procedures are described below in FIGS. 5-6.

[0080] FIG. 5 shows an example method 500 for interventional device intrusion support according to an embodiment. The method 500 may generally be implemented in the interventional device intrusion support system 100 (FIG. 1) and / or the interventional device intrusion support system 200 (FIG. 2), already discussed.

[0081] In an embodiment, the method 500 (as well as method 600 (FIG. 6) may be implemented in logic instructions (e.g., software), configurable logic (e.g., firmware), fixed- functionality hardware logic (e.g., hardware), etc., or any combination thereof.

[0082] In some examples, the methods described herein (e.g., method 500 and / or method 600) may be performed at least in part by cloud processing.

[0083] It will be appreciated that some or all of the operations described herein (e.g., method 500 and / or method 600) that have been described using a “pull” architecture (e g., polling for new information followed by a corresponding response) may instead be implemented using a “push” architecture (e.g., sending such information when there is new information to report), and vice versa.

[0084] Illustrated processing block 502 provides for receiving or obtaining a set of medical images of an anatomy (e.g., vasculature). For example, the set of medical images include one or more two-dimensional fluoroscopy images. Multiple fluoroscopy images may be acquired over time or may be acquired from various arms of a C-arm imaging system (e.g., a biplane system has two arms, a frontal arm and a lateral arm, that can acquire images from two views at the same time). The set of medical images may include an interventional device in the anatomy (e g., vasculature).

[0085] In some embodiments, the set of medical images include one or more three-dimensional estimate images. In some examples, the one or more two dimensional and three dimensional images are combined to generate overlays. Overlays are based on two-dimensional fluoroscopyimages taken during a procedure used in conjunction with a three dimensional image taken for the procedure. The two-dimensional and three-dimensional images must be aligned or registered so that three-dimensional information can appropriately enhance the two-dimensional image via the overlay. Such overlays are used to track the interventional devices (imaged in two-dimensional X- ray imaging) in the three-dimensional anatomy.

[0086] Illustrated processing block 504 provides for tracking the interventional device in the anatomy. For example, the position of the interventional device may be tracked in the anatomy based on the set of medical images.

[0087] In some examples, machine learning algorithms, artificial intelligence (Al) algorithms, image processing algorithms, the like, or some combination thereof can be used to segment and track the interventional device present in the image. For example, the interventional device tip and length may be tracked from fluoroscopic image data. For instance, interventional devices are tracked temporally (e.g., as devices are navigated up the aorta, through the aortic arch, and into the carotid artery) to estimate location with respect to the aneurysm. In such embodiments, machine learning algorithms, artificial intelligence (Al) algorithms, image processing algorithms, the like, or some combination thereof can be used to train models that can segment and track the interventional device present in the image using training data consisting of annotated interventional devices present in fluoroscopic images containing various anatomies including the aorta, aortic arch, carotid artery, aneurysm, and so forth. Annotations may be made manually by an expert. Additionally, or alternatively, annotations may be made automatically by collecting fluoroscopic image data with the interventional device on known anatomical background which can then be subtracted from the fluoroscopic image data. Additionally, or alternatively, location may be determined from the data from the interventional device itself, which may include positional location data.

[0088] In some examples, interventional devices are also tracked temporally to estimate torque or tension build-up such that when the devices are near the aneurysm, the aggregate information can be used to estimate risk of sudden movement or sudden intrusion of the device into the aneurysm. In some embodiments, estimations of torque and / or tension build-up may be based on data from the interventional device itself. For example, the data from the interventional device may itself include device type, positional location data, rotational torque data, linear tension data, the like, and / or combinations thereof. Additionally, or alternatively, this data may be estimatedfrom fluoroscopic image data and / or other devices that may be used to manipulate (e.g., push, pull, rotate) the interventional device such as an endovascular robot.

[0089] Illustrated processing block 506 provides for detecting an aneurysm in the anatomy (or other abnormality in the vasculature). For example, such aneurysms may be detected in the anatomy based on the set of medical images. For example, the aneurysm may be detected from fluoroscopic image data. Additionally, or alternatively, the aneurysm may be detected in 3D imaging data.

[0090] In some examples, detection of the aneurysm may be performed using machine learning algorithms, artificial intelligence (Al) algorithms, image processing algorithms, the like, or some combination thereof. In such embodiments, algorithms may be used to train models that can detect aneurysms using training data consisting of annotated aneurysms of varying sizes and in various locations collected from a large set of patients. Additionally, or alternatively, the aneurysm may be identified manually by a user using a user interface such as by pointing on a touch screen, or clicking using a mouse, or drawing a bounding box using a mouse, and so forth.

[0091] Illustrated processing block 508 provides for determining a risk map. For example, a risk map may be determined or accesses that defines one or more regions around the detected aneurysm.

[0092] For example, in the event that an aneurysm is detected, a risk map may be used to define one or more regions around the detected aneurysm, which each have a level of risk associated with an interventional device inserted into the anatomy contacting the aneurysm. For instance, a distance-based risk map may be defined around the detected aneurysm to define a high, medium, and / or low risk region around the aneurysm. The term “around” the aneurysm generally refers to regions near the detected aneurysm, including within the aneurysm and the parent vessel.

[0093] In some embodiments, the risk map is a distance-based risk map. For example, the distance-based risk map may be calculated within the aneurysm and the parent vessel by computing the shortest distance to the aneurysm wall from any point within the aneurysm and the parent vessel. As this distance increases, the risk decreases. In another example, the risk map may include other factors such as regions of the aneurysm wall that are more vulnerable to rupture. For instance, the aneurysm wall near the dome of the aneurysm is typically thinner and, therefore, weaker than the aneurysm wall near the base or neck of the aneurysm. The distance-based risk map may, therefore, be weighted by distance from the aneurysm neck. That is, the same distanceaway from the aneurysm wall near the neck may be associated with lower risk than that near the aneurysm dome. In another example, the distance-based risk map may also be weighted by the size of the aneurysm. For example, the same distance away from the aneurysm wall in a smaller aneurysm may be associated with lower risk than that in a larger aneurysm.

[0094] In some examples, operations to determine the distance-based risk map are further based on a risk model. For example, such a risk model is based on one or more machine learnable parameters. In such an example, determining the distance-based risk map based on such a risk model may be performed using machine learning algorithms, artificial intelligence (Al) algorithms, the like, or some combination thereof. For example, the risk model may be learnt offline using training datasets consisting of aneurysms in various locations and in varying sizes collected from a large set of patients, as explained below. In other embodiments, the risk map may be determined by other techniques, such as a database may include various historical images of different vasculatures having different aneurysms and each image annotated with a risk map specific to the vasculature and aneurysm in the respective historical image. The risk map in each historical image may also be specific to a type of interventional device entering the vasculature. In some embodiments, the risk map in each historical image may be annotated with risk scores indicating the risk of the interventional device at different positions in the images. The risk map in each historical image may be defined based on position of the interventional device with the vasculature, conditions related to the anatomy at or near the aneurysm (e.g., narrowness of vessel, tortuosity of vessel, angle of vessel, etc), characteristics the aneurysm (e.g., size, shape, strength, location in vasculature), and / or characteristics of the interventional device (e.g., size, type, flexibility).

[0095] Illustrated processing block 510 provides for determining a risk score (e.g., intrusion risk score) of the interventional device contacting the detected aneurysm. For example, a risk score of the interventional device contacting the detected aneurysm may be determined based on the position of the tracking interventional device in the anatomy and the risk map.

[0096] For example, determining the risk score may be performed using machine learning algorithms, artificial intelligence (Al) algorithms, the like, or some combination thereof. As discussed above with respect to FIG. 4, Rtdenotes the “risk” or a “score” of the device rupturing the aneurysm at time t, which is dependent on Dt, device location at time t, and P, a set of learnable parameters for the risk prediction model. The set of learnable parameters, P, may, for instance,encode the risk associated with various distances from the aneurysm wall. Additionally, or alternatively, Rtmay be dependent on device locations tracked temporally through a time window from t — ( / + 1) to t, where N is the size of the time window. Additionally, or alternatively, risk calculated at a prior time step,or tracked temporally through a prior time window may also be used to calculate risk Rtat time t. Models that can accommodate temporal data such as Hidden Markov models (HMMs), temporal convolutional models (TCNs), recurrent neural networks (RNNs), transformer networks, simple statistical models, and / or combinations thereof may be used to determine the risk score. For example, the risk score may be learnt offline using training datasets. Training datasets may consist of simulation data, phantom data, preclinical data, clinical data from prior procedures, the like, and / or combinations thereof. During training, an initial risk map may be determined at the start of training in processing block 508. For instance, the initial risk map may assign a risk of 1 at the aneurysm wall (i.e., when the distance from the aneurysm wall is 0) and a risk of 0 everywhere else (i.e., for all distances > 0 from the aneurysm wall). Over several training iterations, the risk map may be updated based on the likelihood of the interventional device contacting the aneurysm wall from various positions of the interventional device. As the risk map updates, the risk score of the interventional device contacting the detected aneurysm in processing block 510 will also change over training iterations. In such an embodiment, many different instances of navigating the interventional device into an aneurysm may be captured in a phantom set up to generate training data. During training, for each interventional device position, the number of times the interventional device contacted the aneurysm wall in the next M time steps may be recorded and compared to the total number of times the interventional device was in or near its current position in the training dataset. The training data may also capture other information regarding the different instances of navigating the interventional device, such as characteristics of the interventional device (e.g., type, size, flexibility), conditions related to the anatomy at or near the aneurysm (e.g., narrowness of vessel, tortuosity of vessel, angle of vessel, etc), and / or characteristics of the aneurysm (e.g., size, shape, strength, location in vasculature), etc. With each training iteration, as more interventional device positions are analysed across different aneurysms in the training dataset, the general risk map around aneurysms is learned by the model by updating its learnable parameters, P. Training may end when a required number of training iterations are completed or when the change in the risk score of the interventional device contacting the detected aneurysm across several trainingiterations is smaller than some threshold. Therefore, during training, the algorithm may iterate several times between blocks 508 and 510. Alternatively, or additionally, during training, the risk score can be based on an expert’s hand annotated categorical levels of risks associated with various distances from the aneurysm wall in aneurysms in various locations and in varying sizes collected from a large set of patients, for example. In some embodiments, the risk score provides a prediction of the probability of the outcome of the interventional device contacting the aneurysm.

[0097] In some embodiments, the number of times the interventional device contacted the aneurysm wall may be weighted by the location on the aneurysm wall. For instance, the interventional device contacting the aneurysm wall near the aneurysm neck may be weighted lower than the interventional device contacting the aneurysm wall near the aneurysm dome in calculating the risk of aneurysm rupture. Other information about where the aneurysm wall is weaker can also be incorporated into the risk calculation in a similar fashion.

[0098] In some embodiments, the training may be performed using a reinforcement learning technique, where minimizing the risk of the interventional device contacting the detected aneurysm is rewarded. In the case that the interventional device is being navigated using an endovascular robot, the robot control may be updated to navigate the interventional device in such a way that the reinforcement learning model’s reward function is maximized and, therefore, the risk of the interventional device contacting the detected aneurysm is minimized.

[0099] In some embodiments, the risk score of the interventional device contacting the detected aneurysm may be used to inform other tasks. For instance, if tension in the interventional device must be released, then a region may be determined where tension may be released safely without the risk of the device tip jumping forward and contacting the aneurysm. The region may be determined such that the risk of the interventional device contacting the detected aneurysm is below a predetermined threshold (e g., risk < 0.25)

[0100] In some embodiments, such a risk score describes a probability of the device entering the high-risk region and may be defined based on a motion model derived from the device tip / length detection and tracking. For example, the risk score is determined further based on a motion model. For example, the motion model considers device type, positional location, rotational torque, linear tension of the interventional device, the like, and / or combinations thereof. For example, the device type is associated with an associated stiffness level of the interventional device (e.g., stiffer devices will have higher risk than softer devices).

[0101] In some examples, such a motion model is further based on data from the interventional device itself. For example, the data from the interventional device itself includes device type, positional location data, rotational torque data, linear tension data, the like, and / or combinations thereof.

[0102] Additional and / or alternative operations for method 500 are described in greater detail below in the description of FIG. 6.

[0103] FIG. 6 is a sequence diagram of an example of another method 600 for interventional device intrusion support according to an embodiment. The method 600 may generally be implemented in the interventional device intrusion support system 100 (FIG. 1) and / or the interventional device intrusion support system 200 (FIG. 2), already discussed.

[0104] As illustrated, various processing blocks are illustrated as being performed by interventional device intrusion support logic 608, an interventional device 616, a medical imager 619, a display 620, and / or a robotic control manipulator 621 in conjunction with one another (e g., as discussed above in FIG. 2).

[0105] Illustrated processing blocks 622 provide for training the interventional device intrusion support model.

[0106] Illustrated processing blocks 624 provide for utilizing the interventional device intrusion support logic once the interventional device intrusion support model is trained.

[0107] Training operations are illustrated in processing blocks 630-640. In some examples, the inputs to the algorithm may be solely imaging data. For example, the risk model or score may be learnt offline using training datasets. During training, the risk score can be based on an expert’s hand annotated categorical levels of risks, for example.

[0108] During training, illustrated processing block 630 provides for receiving a set of medical images. For example, the set of medical images include two-dimensional and / or three-dimensional images for training.

[0109] Illustrated processing block 632 provides for tracking the interventional device. For example, an interventional device may be tracked in the anatomy based on the set of medical images.

[0110] Illustrated processing block 634 provides for detecting an aneurysm. For example, detection of the aneurysm may be performed using machine learning algorithms, Al algorithms, image processing algorithms, the like, or some combination thereof.

[0111] Illustrated processing block 636 provides for determining a risk map.

[0112] Illustrated processing block 638 provides for providing data to model for the training routine.

[0113] Illustrated processing block 640 provides for training the interventional device intrusion support model. The motion model and / or risk model may be trained prior to the training of the interventional device intrusion support model, or they may be trained concurrently.

[0114] Utilization operations are illustrated in processing blocks 650-664.

[0115] During utilization, illustrated processing block 650 provides for receiving a set of medical images. For example, the set of medical images include two-dimensional fluoroscopy images.

[0116] In some embodiments, the set of medical images include one or more three-dimensional estimate images. In some examples, the one or more three-dimensional estimate image overlays are generated based on two dimensional fluoroscopy images taken during a procedure used in conjunction with a three dimensional image taken for the procedure.

[0117] Illustrated processing block 651 provides for providing positional data, rotational torque data, linear tension data, the like, and / or combinations thereof. For example, such data may be provided from an interventional device 616 being utilized.

[0118] Illustrated processing block 652 provides for tracking the interventional device. For example, an interventional device may be tracked in the anatomy based on the set of medical images.

[0119] Illustrated processing block 654 provides for detecting an aneurysm. For example, detection of the aneurysm may be performed using machine learning algorithms, Al algorithms, image processing algorithms, the like, or some combination thereof.

[0120] Illustrated processing block 656 provides for determining a risk map. For example, a risk map may be determined to define one or more risk regions around the detected aneurysm.

[0121] In some embodiments, the risk map is a distance-based risk map. In some examples, operations to determine the distance-based risk map are further based on a risk model. For example, such a risk model is based on one or more machine learnable parameters.

[0122] Illustrated processing block 662 provides for determining a risk score. For example, the determined risk score may be transferred to the display 620 and / or robotic control manipulator621. For example, a risk score of the interventional device contacting the detected aneurysm may be determined based on the tracking of the interventional device and the risk map.

[0123] In some embodiments, such a risk score describes a probability of the device entering the high-risk region and may be defined based on a motion model derived from the device tip / length detection and tracking. For example, the risk score is determined further based on a motion model. For example, the motion model considers device type, positional location, rotational torque, linear tension of the interventional device, the like, and / or combinations thereof. For example, the device type is associated with an associated stiffness level of the interventional device (e.g., stiffer devices will have higher risk than softer devices).

[0124] In some examples, such a motion model is further based on data from the interventional device itself. For example, the data from the interventional device itself includes device type, positional location data, rotational torque data, linear tension data, the like, and / or combinations thereof.

[0125] Illustrated processing block 664 provides for generating guidance. For example, the generated guidance may be transferred to the display 620 and / or robotic control manipulator 621 in addition to or in the alternative to the risk score.

[0126] Illustrated processing block 666 provides for presenting the risk score and / or risk map to a user via the display 620. For example, this presentation may be continually updated as the intervention proceeds.

[0127] Illustrated processing block 668 provides for providing guidance to a user via the display 620. For example, such guidance may instruct a user regarding risks or techniques to reduce risks for the current intervention.

[0128] Illustrated processing block 670 provides for controlling the interventional device. For example, the robotic control manipulator may be utilized to take over full control of the interventional device, provide haptic feedback, soft inhibit movement (e.g., provide resistance to slow down the intervention), relieve tension (e.g., rotational torque and / or linear tension) in the interventional device, the like, and / or combinations thereof.

[0129] In some embodiments, the robotic control manipulator is to perform one or more operations in response to the risk score. For example, the robotic control manipulator is to inhibit movement of the interventional device, relieve rotational torque of the interventional device, or relieve linear tension of the interventional device.

[0130] Additionally, or alternatively, the procedures described herein may provide a framework for clinical deployment of decision support algorithms. These procedures can work together with many clinical decision support (CDS) algorithms (such as acute kidney injury (AKI), acute respiratory distress syndrome (ARDS), acute decompensated heart failure (ADHF), etc ). Clinical decision support (CDS) refers to computer-based support of clinical staff responsible for making decisions for the care of patients. Computer-based support for clinical decision-making staff may take many forms, from patient-specific visual / numeric health status indicators to patientspecific health status predictions and patient-specific health care recommendations. Further, the procedures described herein may be deployed on analytics platforms (such as Inference Engine, Critical Care Information System, Interoperability Solution, etc.) in conjunction with CDS algorithms.

[0131] FIG. 7 illustrates a block diagram of an example computer program product 700. In some examples, as shown in FIG. 7, computer program product 700 includes a machine-readable storage 702 that may also include logic 704. In some embodiments, the machine-readable storage 702 may be implemented as a non-transitory machine-readable storage. In some embodiments the logic 704 may be implemented as machine-readable instructions, such as software, for example. In an embodiment, the logic 704, when executed, implements one or more aspects of the method 500 (FIG. 5), the method 600 (FIG. 6), and / or realize the system 100 (FIG. 1 and / or FIG. 2), already discussed.

[0132] FIG. 8 shows an illustrative example of a system 800. In the illustrated example, the system 800 may include a processor 802 and a memory 804 communicatively coupled to the processor 802. The memory 804 may include logic 806 as a set of instructions. In some embodiments the logic 806 may be implemented as software. In an embodiment, the logic 806, when executed by the processor 802, implements one or more aspects of the method 500 (FIG. 5), the method 600 (FIG. 6), and / or realize the system 100 (FIG. 1 and / or FIG. 2), already discussed.

[0133] In some embodiments, the processor 802 may include a general purpose controller, a special purpose controller, a storage controller, a storage manager, a memory controller, a microcontroller, a general purpose processor, a special purpose processor, a central processor unit (CPU), the like, and / or combinations thereof.

[0134] Further, embodiments may include distributed processing, component / object distributed processing, parallel processing, the like, and / or combinations thereof. For example,virtual computer system processing may implement one or more of the methods or functionalities as described herein, and the processor 802 described herein may be used to support such virtual processing.

[0135] In some examples, the memory 804 is an example of a computer-readable storage medium. For example, memory 804 may be any memory which is accessible to the processor 802, including, but not limited to RAM memory, registers, and register files, the like, and / or combinations thereof. References to “computer memory” or “memory” should be interpreted as possibly being multiple memories. The memory may for instance be multiple memories within the same computer system. The memory may also be multiple memories distributed amongst multiple computer systems or computing devices.

[0136] FIG. 9 shows an illustrative semiconductor apparatus 900 (e.g., chip and / or package). The illustrated semiconductor apparatus 900 includes one or more substrates 902 (e.g., silicon, sapphire, or gallium arsenide) and logic 904 (e.g., configurable logic and / or fixed-functionality hardware logic) coupled to the substrate(s) 902. In an embodiment, the logic 904 implements one or more aspects of the method 500 (FIG. 5), the method 600 (FIG. 6), and / or realize the system 100 (FIG. 1 and / or FIG. 2), already discussed.

[0137] In some embodiments, logic 904 may include transistor array and / or other integrated circuit / IC components. For example, configurable logic and / or fixed-functionality hardware logic embodiments of the logic 904 may include configurable logic such as, for example, programmable logic arrays (PLAs), field programmable gate arrays (FPGAs), complex programmable logic devices (CPLDs), or fixed-functionality logic hardware using circuit technology such as, for example, application specific integrated circuit (ASIC), complementary metal oxide semiconductor (CMOS) or transistor-transistor logic (TTL) technology, the like, and / or combinations thereof.

[0138] All definitions, as defined and used herein, should be understood to control over dictionary definitions, definitions in documents incorporated by reference, and / or ordinary meanings of the defined terms.

[0139] The herein described subject matter sometimes illustrates different components contained within, or connected with, different other components. It is to be understood that such depicted architectures are merely exemplary, and that in fact many other architectures can be implemented which achieve the same functionality. In a conceptual sense, any arrangement ofcomponents to achieve the same functionality is effectively "associated" such that the desired functionality is achieved. Hence, any two components herein combined to achieve a particular functionality can be seen as "associated with" each other such that the desired functionality is achieved, irrespective of architectures or intermedial components. The term “coupled” may be used herein to refer to any type of relationship, direct or indirect, between the components in question, and may apply to electrical, mechanical, fluid, optical, electromagnetic, electromechanical, or other connections. Likewise, any two components so associated can also be viewed as being "operably connected", or "operably coupled", to each other to achieve the desired functionality, and any two components capable of being so associated can also be viewed as being "operably couplable", to each other to achieve the desired functionality. Specific examples of operably couplable include but are not limited to physically mateable and / or physically interacting components.

[0140] In the claims, as well as in the specification above, the terms “first”, “second”, etc. may be used herein only to facilitate discussion, and carry no particular temporal or chronological significance unless otherwise indicated.

[0141] In the claims, as well as in the specification above, all transitional phrases such as “comprising,” “including,” “carrying,” “having,” “containing,” “involving,” “holding,” “composed of,” and the like are to be understood to be open-ended, i.e., to mean including but not limited to. Only the transitional phrases “consisting of’ and “consisting essentially of’ shall be closed or semi-closed transitional phrases, respectively.

[0142] The indefinite articles “a” and “an,” as used herein in the specification and in the claims, unless clearly indicated to the contrary, should be understood to mean “at least one.”

[0143] As used herein, the term “or” or “and / or” is inclusive and not exclusive, unless expressly indicated otherwise or indicated otherwise by context. Therefore, herein, “A or B” means “A, B, or both,” unless expressly indicated otherwise or indicated otherwise by context. Moreover, “and” is both joint and several, unless expressly indicated otherwise or indicated otherwise by context. Therefore, herein, “A and B” means “A and B, jointly or severally,” unless expressly indicated otherwise or indicated otherwise by context.

[0144] As used herein in the specification and in the claims, the phrase “at least one,” in reference to a list of one or more elements, should be understood to mean at least one element selected from any one or more of the elements in the list of elements, but not necessarily includingat least one of each and every element specifically listed within the list of elements and not excluding any combinations of elements in the list of elements. This definition also allows that elements may optionally be present other than the elements specifically identified within the list of elements to which the phrase “at least one” refers, whether related or unrelated to those elements specifically identified.

[0145] As used in this application and in the claims, a list of items joined by the term “one or more of’ may mean any combination of the listed terms. For example, the phrases “one or more of A, B or C” may mean A; B; C; A and B; A and C; B and C; or A, B and C.

[0146] As is described above in greater detail, one or more processor, other unit, the like, and / or combinations thereof may fulfill the functions of several items recited in the claims.

[0147] As is described above in greater detail, a computer program may be stored / distributed on a suitable computer readable medium, such as an optical storage medium or a solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunication systems.

[0148] It should also be understood that, unless clearly indicated to the contrary, in any methods discussed herein that include more than one step or act, the order of the steps or acts of the method is not necessarily limited to the order in which the steps or acts of the method are recited. Further, such methods may include additional or alternative steps or acts.

[0149] As used in the claims, the mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measured cannot be used to advantage.

[0150] It is also noted that the claims may include reference signs / numerals in accordance with PCT Rule 6.2(b). However, the present claims should not be considered to be limited to the exemplary embodiments corresponding to the reference signs / numerals.

[0151] Those skilled in the art will appreciate from the foregoing description that the broad techniques of the embodiments of the present invention can be implemented in a variety of forms. Therefore, while the embodiments of this invention have been described in connection with particular examples thereof, the true scope of the embodiments of the invention should not be so limited since other modifications will become apparent to the skilled practitioner upon a study of the drawings, specification, and following claims.

Claims

CLAIMS:

1. A system for detecting an interventional device in a vasculature, the system comprising: a processor communicatively coupled to memory, the processor configured to: obtain a set of medical images of the interventional device in the vasculature; track a position of the interventional device in the vasculature based on the set of medical images; detect an abnormality in the vasculature; access a risk map that defines one or more regions around the detected abnormality; and determine an intrusion risk score based on the tracked position of the interventional device and the risk map, the intrusion risk score defining a risk of the interventional device contacting the detected abnormality.

2. The system of claim 1, wherein the set of medical images comprises two-dimensional (2D) fluoroscopy images.

3. The system of claim 1, wherein the set of medical images comprises one or more three- dimensional (3D) images generated based on 2D fluoroscopy images captured during a medical procedure, the generated 3D images used in conjunction with a 3D image captured to track the interventional device in the vasculature.

4. The system of claim 1, wherein the processor is further configured to apply a motion model configured to determine the intrusion risk score based on at least one of type, location, rotational torque, and linear tension of the interventional device.

5. The system of claim 4, wherein the motion model is generated based on data captured by a sensor associated with the interventional device, wherein the data comprises at least one of device type data, location data, rotational torque data, and linear tension data.

6. The system of claim 1, wherein the risk map is a distance-based risk map.

7. The system of claim 6, wherein the processor is configured to apply a risk model to determine the intrusion risk score, the risk model comprising a trained machine-learning model.

8. The system of claim 7, wherein the risk model is configured to predict the risk scor.

9. The system of claim 1, further comprising: a display communicatively coupled to the processor, wherein the display is configured to present at least one of the risk map and the risk score to a user.

10. The system of claim 1, wherein the processor is further configured to perform, in response to the intrusion risk score, one or more operations comprising at least one of inhibit movement of the interventional device, relieve rotational torque of the interventional device, and relieve linear tension of the interventional device.

11. A non-transitory machine-readable storage medium having stored a computer program comprising a set of instructions, which when executed by a processor, causes the processor to: obtain a set of medical images of the interventional device in the vasculature ; track a position of the interventional device in the vasculature based on the set of medical images; detect an abnormality in the vasculature; access a risk map that defines one or more regions around the detected abnormality; and determine an intrusion risk score based on the tracked position of the interventional device and the risk map, the intrusion risk score defining a risk of the interventional device contacting the detected abnormality.

12. The non-transitory machine-readable storage of claim 11, wherein the set of medical images comprises one or more 3D images generated based on 2D fluoroscopy images captured during a medical procedure, the generated 3D images used in conjunction with a 3D image captured to track the interventional device in the vasculature.

13. The non-transitory machine-readable storage of claim 11, wherein the set of instructions, when executed by the processor, further cause the processor to apply a motion model configured to determine the intrusion risk score based on at least one of device type, positional location data, rotational torque data, or linear tension data, wherein the device type is associated with a stiffness level.

14. The non-transitory machine-readable storage of claim 11, the set of instructions, when executed by the processor, further cause the processor to: perform, in response to the intrusion risk score, one or more operations that at least one of inhibit movement of the interventional device, relieve rotational torque of the interventional device, and relieve linear tension of the interventional device.

15. A method for detecting an interventional device in a vasculature, the method comprising: obtaining a set of medical images of the interventional device in the vasculature; tracking a position of the interventional device in the vasculature based on the set of medical images; detecting an abnormality in the vasculature; accessing a risk map that defines one or more regions around the detected abnormality; and determining an intrusion risk score based on the tracked position of the interventional device and the risk map, the intrusion risk score defining a risk of the interventional device contacting the detected abnormality.

Images