Navigation and 3D modeling for catheter-based systems and associated methods

Abstract

The present technology relates to systems and methods for removing a thrombus from a blood vessel of a patient. In some embodiments, the present technology is directed to systems including an elongated catheter having a distal portion configured to be positioned within the blood vessel of the patient, a proximal portion configured to be external to the patient, and a lumen extending therebetween. The systems and methods discussed herein provide 3D models of the pulmonary vasculature, clots within the pulmonary vasculature, and / or of medical devices within the patient, for pre and peri-operative planning and navigation.

Description

NAVIGATION AND 3D MODELING FOR CATHETER-BASED SYSTEMS AND ASSOCIATED METHODSPRIORITY CLAIM

[0001] This patent application claims priority to U.S. provisional patent application no.63 / 745,730, titled “THROMBUS REMOVAL SYSTEMS AND ASSOCIATED METHODS,” filed January 15, 2025; U.S. provisional patent application no. 63 / 793,494, titled “NAVIGATION AND 3D MODELING FOR CATHETER-BASED SYSTEMS AND ASSOCIATED METHODS,” filed April 23, 2025; U.S. provisional patent application no.63 / 808,500, titled “NAVIGATION AND 3D MODELING FOR CATHETER-BASED SYSTEMS AND ASSOCIATED METHODS,” filed May 19, 2025; U.S. provisional patent application no. 63 / 817,844, titled “NAVIGATION AND 3D MODELING FOR CATHETER-BASED SYSTEMS AND ASSOCIATED METHODS,” filed June 4, 2025; U.S. provisional patent application no. 63 / 823,657, titled “NAVIGATION AND 3D MODELING FOR CATHETER-BASED SYSTEMS AND ASSOCIATED METHODS,” filed June 13, 2025; U.S. provisional patent application no. 63 / 857,337, titled “NAVIGATION AND 3D MODELING FOR CATHETER-BASED SYSTEMS AND ASSOCIATED METHODS,” filed August 4, 2025; U.S. provisional patent application no.63 / 903,796, titled “NAVIGATION AND 3D MODELING FOR CATHETER-BASED SYSTEMS AND ASSOCIATED METHODS,” filed October 22, 2025; and U.S. provisional patent application no. 63 / 919,432, titled “NAVIGATION AND 3D MODELING FOR CATHETER-BASED SYSTEMS AND ASSOCIATED METHODS,” filed November 17, 2025, which are all herein incorporated by reference in their entirety.INCORPORATION BY REFERENCE

[0002] All publications and patent applications mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication or patent application was specifically and individually indicated to be incorporated by reference.FIELD

[0003] The present technology generally relates to medical devices and, in particular, to systems including aspiration and fluid delivery mechanisms and associated methods for removing a thrombus from a mammalian blood vessel.- 1 - SG Docket No.: 10844-737.674BACKGROUND

[0004] Thrombotic material may lead to a blockage in fluid flow within the vasculature of a mammal. Such blockages may occur in varied regions within the body, such as within the pulmonary system, peripheral vasculature, deep vasculature, or brain. Pulmonary embolisms typically arise when a thrombus originating from another part of the body (e.g., a vein in the pelvis or leg) becomes dislodged and travels to the lungs.

[0005] Anti coagulation therapy is the current standard of care for treating pulmonary embolisms, but may not be effective in some patients. Additionally, conventional devices for removing thrombotic material may not be capable of navigating the tortuous vascular anatomy, may not be effective in removing thrombotic material, and / or may lack the ability to provide sensor data or other feedback to the clinician during the thrombectomy procedure.

[0006] Existing thrombectomy devices operate based on simple aspiration which works sufficiently for certain clots but is largely ineffective for difficult, organized clots. Many patients presenting with deep vein thrombus (DVT) are left untreated as long as the risk of limb ischemia is low. In more urgent cases, they are treated with catheter-directed thrombolysis or lytic therapy to break up a clot over the course of many hours or days.

[0007] Visualization of clots and the location of a thrombectomy device relative to the clot is also an issue during thrombectomy procedures. While positioning a thrombectomy catheter in the vicinity of a clot is fairly routine, it is particularly difficult to orient a thrombectomy catheter at the clot locally using current imaging techniques. The further away the catheter is from the clot prior to turning on aspiration, the more blood is likely to be unnecessarily aspirated in the process of capturing the clot.

[0008] Fluoroscopic imaging can be used to visualize clots and the positioning of the thrombectomy device within the anatomy, however this requires frequent injection of radiopaque dyes or contrast agent into the vasculature and pausing the procedure to obtain the fluoroscopy imaging. Additionally, the 2D imaging does not provide for accurate placement of a thrombectomy device in 3D space, particularly within the voluminous pulmonary trunk and left and right pulmonary arteries (relative to the size of a thrombectomy catheter), which can result in clots that appear to be close to the inlet of the thrombectomy catheter in the fluoroscopy images being distanced from the catheter out of the imaging plane. This deficiency with traditional 2D fluoroscopy imaging can result in larger than desirable volumes of blood being aspirated from the patient by initiating aspiration when the physician incorrectly believes that the clot is engaged with or near enough to the thrombectomy catheter to be engaged with upon initiation of aspiration. To correct for imaging out of plane with the clot and / or catheter, the physician must guess which direction the catheter is out of plane -2 - SG Docket No.: 10844-737.674from the clot, requiring additional steering and manipulation of the catheter to try to engage with the targeted clot. This can increase treatment time and also lead to procedures in which no clot is removed.

[0009] There remains the need for systems to address these and other problems with existing venous thrombectomy including, but not limited to, visualization that provides for a fast, easy-to-use, and effective device for removing a variety of clot morphologies.BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The novel features of the invention are set forth with particularity in the claims that follow. A better understanding of the features and advantages of the present invention will be obtained by reference to the following detailed description that sets forth illustrative embodiments, in which the principles of the invention are utilized, and the accompanying drawings of which:

[0011] FIGS. 1 A-1B illustrate a medical device such as a thrombectomy catheter.

[0012] FIG. 2 is a schematic view of the pulmonary vasculature.

[0013] FIGS. 3 A-3B show a system for generating a 3D anatomical model of a target anatomy.

[0014] FIGS. 4A-4B are schematic embodiments of a system for providing pre-operative assessment and / or real-time navigation of a medical device during an interventional procedure.

[0015] FIGS. 5A-7 are flowcharts for providing an assessment or recommendation or navigation before or during an interventional procedure.

[0016] FIGS. 8A-8C illustrate additional views of segmented clot and pulmonary vasculature from imaging data.

[0017] FIGS. 9A-9D show screenshots of a graphical user interface that can display medical imaging and 3D generated models of patient anatomy to a user along with a classification of the clots.

[0018] FIGS. 10A-10B illustrate embodiments of a 3D pulmonary artery model with segmented clots overlaid within the model, which can be used to map or plan out a procedure.

[0019] FIGS. 11 A-l 10 illustrate an embodiment of removing thrombus from a patient and estimating perfusion of the lungs.

[0020] FIG. 12 is a block diagram of a clot management environment in accordance with an embodiment of the present technology.

[0021] FIG. 13 is processor environment in accordance with an embodiment of the present technology.-3 - SG Docket No.: 10844-737.674

[0022] FIG. 14 is a flow diagram of a process of characterizing clots in accordance with an embodiment of the present technology.

[0023] FIG. 15 is a flow diagram of a process of performing an outcome assessment in accordance with an embodiment of the present technology.SUMMARY OF THE DISCLOSURE

[0024] A thrombus removal system is provided, comprising an elongate shaft comprising a working end, at least one fluid lumen in the elongate shaft, and two or more apertures disposed at or near the working end, the two or more apertures in fluid communication with the least one fluid lumen and configured to generate two or more fluid streams directed generally normal to a longitudinal axis of the shaft at an intersection point within the shaft to mechanically fractionate a target thrombus.

[0025] In certain examples, there is a system having: one or more processors; memory coupled to the one or more processors, in which the memory includes computer-program instructions that, when executed by the one or more processors, cause the device to perform operations including: acquiring two-dimensional (2D) and / or three-dimensional (3D) imaging data of at least a torso of a patient; segmenting the imaging data of the patient to identify 1) the pulmonary vasculature, and / or 2) one or more clots or lesions within the pulmonary vasculature; generating a 3D pulmonary vasculature model of the patient from the segmented imaging data; generating a 3D clot model of the patient from the segmented imaging data; characterizing and assessing the one or more clots or lesions within the 3D pulmonary vasculature model and the 3D clot model. Then, based on considering the characterizing and assessing, the system outputs a treatment recommendation of selected clots or lesions of the one or more clots or lesions to target for removal.

[0026] A system is provided, comprising: one or more processors; memory coupled to the one or more processors, wherein the memory includes computer-program instructions that, when executed by the one or more processors, cause the system to perform operations comprising: acquiring two-dimensional (2D) and / or three-dimensional (3D) imaging data of at least a torso of a patient; segmenting the imaging data of the patient to identify 1) the pulmonary vasculature, and / or 2) one or more clots or lesions within the pulmonary vasculature; generating a 3D pulmonary vasculature model of the patient from the segmented imaging data; generating a 3D clot model of the patient from the segmented imaging data; characterizing and assessing a chronicity of the one or more clots or lesions within the 3D pulmonary vasculature model and the 3D clot model; and considering the characterization- 4 - SG Docket No.: 10844-737.674and assessment of chronicity, outputting a treatment recommendation of selected clots or lesions of the one or more clots or lesions to target for removal.

[0027] In some aspects, characterizing and assessing the chronicity comprises evaluating pixels or voxels in the 2D and / or 3D imaging data corresponding to the one or more clots or lesions.

[0028] In some aspects, evaluating pixels or voxels comprises evaluating a Houndsfield scale value of the pixels or voxels.

[0029] In other aspects, characterizing and assessing the chronicity comprises evaluating voxels in the 2D and / or 3D imaging data of boundaries between one or more clots or lesions and a vessel wall.

[0030] In some aspects, the system is further configured to determine vessel wall integration of the one or more clots or lesions.

[0031] In some aspects, characterizing and assessing the chronicity comprises assessing the aggregation or uniformity of pixels or voxels of the one or more clots or lesions.

[0032] In one aspect, the computer-program instructions further cause the system to perform operations comprising: generating an obstruction index for the one or more clots or lesions that indicates a clot or lesions contribution to obstruction of the pulmonary vasculature; and presenting the obstruction index to the user for the one or more clots or lesions.

[0033] In some aspects, presenting the obstruction index comprises presenting a volume of the one or more clots or lesions. In other aspects, presenting the obstruction index comprises presenting a percentage of the pulmonary vasculature obstructed by the one or more clots or lesions.

[0034] In one aspect, generating the obstruction index comprises extracting a centerline of a target vessel at a location of a specific clot or lesion, determining a size of the target vessel at the location with the specific clot or lesion removed, and determining the degree of obstruction that the specific clot or lesion is responsible for within the target vessel at the location.

[0035] In other aspects, characterizing and assessing chronicity of the one or more clots comprises classifying the one or more clots into categories pertaining to the age of the clot.

[0036] In some aspects, characterizing and assessing the one or more clots comprises identifying clots that should be excluded from treatment based on their location within the 3D pulmonary vasculature model.

[0037] In some aspects, characterizing and assessing the one or more clots comprises identifying clots that should not be accessed with a thrombectomy catheter.- 5 - SG Docket No.: 10844-737.674

[0038] In additional aspects, characterizing and assessing the one or more clots comprises identifying clots that can be accessed with a thrombectomy catheter.

[0039] In some aspects, characterizing and assessing the one or more clots comprises determining a pre-operative Miller Score or a comparable index or metric based on the one or more clots to indicate an extent of obstruction to blood flow caused by the one or more clots.

[0040] In one aspect, characterizing and assessing the one or more clots comprises determining a post-operative Miller Score or a comparable index or metric based if selected clots are removed to indicate an extent of obstruction to blood flow caused by the one or more clots.

[0041] In some aspects, characterizing and assessing the one or more clots comprises determining a pre-operative Miller Score based on the one or more clots, identifying one or more selected clots to be removed, and determining a post-operative Miller Score if the selected clots are removed.

[0042] In some aspects, the 2D or 3D imaging data comprises X-ray imaging data. The imaging data can be as a function of time. In other aspects, the 2D or 3D imaging data comprises computed tomography (CT) imaging data. In other aspects, the 2D or 3D imaging data comprises magnetic resonance imaging (MRI) imaging data. In some aspects, the 2D or 3D imaging data comprises positron emission tomography (PET) imaging data. In additional aspects, the 2D or 3D imaging data comprises a fusion of one or more sources of imaging data. In some aspects, the 2D or 3D imaging data comprises ultrasound imaging data.

[0043] A computer implemented method is also provided, comprising: acquiring two-dimensional (2D) and / or three-dimensional (3D) imaging data of at least a torso of a patient; segmenting the imaging data of the patient to identify 1) the pulmonary vasculature, and / or 2) one or more clots or lesions within the pulmonary vasculature; generating a 3D pulmonary vasculature model of the patient from the segmented imaging data; generating a 3D clot model of the patient from the segmented imaging data; characterizing and assessing a chronicity of the one or more clots or lesions within the 3D pulmonary vasculature model and the 3D clot model; and considering the characterizing and assessing, outputting a treatment recommendation of selected clots or lesions of the one or more clots or lesions to target for removal.

[0044] In some aspects, characterizing and assessing the chronicity comprises evaluating pixels or voxels in the 2D and / or 3D imaging data corresponding to the one or more clots or lesions.

[0045] In some aspects, evaluating pixels or voxels comprises evaluating a Houndsfield scale value of the pixels or voxels.- 6 - SG Docket No.: 10844-737.674

[0046] In other aspects, characterizing and assessing the chronicity comprises evaluating voxels in the 2D and / or 3D imaging data of boundaries between one or more clots or lesions and a vessel wall.

[0047] In some aspects, the system is further configured to determine vessel wall integration of the one or more clots or lesions.

[0048] In some aspects, characterizing and assessing the chronicity comprises assessing the aggregation or uniformity of pixels or voxels of the one or more clots or lesions.

[0049] In one aspect, the computer-program instructions further cause the system to perform operations comprising: generating an obstruction index for the one or more clots or lesions that indicates a clot or lesions contribution to obstruction of the pulmonary vasculature; and presenting the obstruction index to the user for the one or more clots or lesions.

[0050] In some aspects, presenting the obstruction index comprises presenting a volume of the one or more clots or lesions. In other aspects, presenting the obstruction index comprises presenting a percentage of the pulmonary vasculature obstructed by the one or more clots or lesions.

[0051] In one aspect, generating the obstruction index comprises extracting a centerline of a target vessel at a location of a specific clot or lesion, determining a size of the target vessel at the location with the specific clot or lesion removed, and determining the degree of obstruction that the specific clot or lesion is responsible for within the target vessel at the location.

[0052] In other aspects, characterizing and assessing chronicity of the one or more clots comprises classifying the one or more clots into categories pertaining to the age of the clot.

[0053] In some aspects, characterizing and assessing the one or more clots comprises identifying clots that should be excluded from treatment based on their location within the 3D pulmonary vasculature model.

[0054] In some aspects, characterizing and assessing the one or more clots comprises identifying clots that should not be accessed with a thrombectomy catheter.

[0055] In additional aspects, characterizing and assessing the one or more clots comprises identifying clots that can be accessed with a thrombectomy catheter.

[0056] In some aspects, characterizing and assessing the one or more clots comprises determining a pre-operative Miller Score or a comparable index or metric based on the one or more clots to indicate an extent of obstruction to blood flow caused by the one or more clots.

[0057] In one aspect, characterizing and assessing the one or more clots comprises determining a post-operative Miller Score or a comparable index or metric based if selected - 7 - SG Docket No.: 10844-737.674clots are removed to indicate an extent of obstruction to blood flow caused by the one or more clots.

[0058] In some aspects, characterizing and assessing the one or more clots comprises determining a pre-operative Miller Score based on the one or more clots, identifying one or more selected clots to be removed, and determining a post-operative Miller Score if the selected clots are removed.

[0059] In some aspects, the 2D or 3D imaging data comprises X-ray imaging data. The imaging data can be as a function of time. In other aspects, the 2D or 3D imaging data comprises computed tomography (CT) imaging data. In other aspects, the 2D or 3D imaging data comprises magnetic resonance imaging (MRI) imaging data. In some aspects, the 2D or 3D imaging data comprises positron emission tomography (PET) imaging data. In additional aspects, the 2D or 3D imaging data comprises a fusion of one or more sources of imaging data. In some aspects, the 2D or 3D imaging data comprises ultrasound imaging data.

[0060] A thrombectomy method is provided, comprising: acquiring two-dimensional (2D) and / or three-dimensional (3D) imaging data of at least a torso of a patient; segmenting the imaging data of the patient to identify 1) the pulmonary vasculature, and / or 2) one or more clots or lesions within the pulmonary vasculature; generating a 3D pulmonary vasculature model of the patient from the segmented imaging data; generating a 3D clot model of the patient from the segmented imaging data; advancing a thrombectomy catheter into a pulmonary vasculature toward a target thrombus; delivering a bolus of contrast from the thrombectomy catheter into the pulmonary vasculature; modeling a flow of the bolus of contrast in the 3D pulmonary vasculature model; calculating a perfusion estimate of the bolus of contrast into one or more sections of the pulmonary vasculature; and displaying the perfusion estimate.

[0061] In some aspects, the method includes obtaining fluoroscopy images of the flow of the bolus of contrast, and updating modeling the flow of the bolus of contrast based on the fluoroscopy images.

[0062] In another aspect, the method includes obtaining fluoroscopy images of the flow of the bolus of contrast, and updating the perfusion estimate based on the fluoroscopy images.

[0063] In some aspects, the method includes removing the target thrombus with the thrombectomy catheter, delivering a second bolus of contrast into the pulmonary vasculature, modeling the flow of the second bolus, updating the perfusion estimate based on the second bolus, and displaying the updated perfusion estimate. In some aspects, the method includes repeating the steps above multiple targeted thrombus.- 8 - SG Docket No.: 10844-737.674DETAILED DESCRIPTION

[0064] This application is related to disclosure in International ApplicationNo. PCT / US2021 / 020915, filed March 8, 2021 (the ‘915 application), and International Application No. PCT / US2022 / 033028, filed June 10, 2022 (the ‘028 application), the disclosures of which are incorporated by reference herein for all purposes. The ‘915 and ‘028 applications describe general mechanisms for capturing and removing a clot. By example, multiple fluid streams are directed toward the clot to fragment the material.

[0065] The present technology is generally directed to thrombus removal systems and associated methods. A system configured in accordance with an embodiment of the present technology can include, for example, an elongated catheter having a distal portion configured to be positioned within a blood vessel of the patient, a proximal portion configured to be external to the patient, a fluid delivery mechanism configured to fragment the thrombus with pressurized fluid, an aspiration mechanism configured to aspirate the fragments of the thrombus, and one or more lumens extending at least partially from the proximal portion to the distal portion and in some embodiments including a steering mechanism located at the distal portion. In some aspects, the system may include multiple colinear shafts.

[0066] The terminology used in the description presented below is intended to be interpreted in its broadest reasonable manner, even though it is being used in conjunction with a detailed description of certain specific embodiments of the present technology.Certain terms may even be emphasized below; however, any terminology intended to be interpreted in any restricted manner will be overtly and specifically defined as such in this Detailed Description section. Additionally, the present technology can include other embodiments that are within the scope of the examples but are not described in detail with respect to the figures.

[0067] Reference throughout this specification to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present technology. Thus, the appearances of the phrases "in one embodiment" or "in an embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment.Furthermore, the particular features or characteristics may be combined in any suitable manner in one or more embodiments.

[0068] Reference throughout this specification to relative terms such as, for example, "generally," "approximately," and "about" are used herein to mean the stated value plus or minus 10%.- 9 - SG Docket No.: 10844-737.674

[0069] Although some embodiments herein are described in terms of thrombus removal, it will be appreciated that the present technology can be used and / or modified to remove other types of emboli that may occlude a blood vessel, such as fat, tissue, or a foreign substance. Additionally, although some embodiments herein are described in the context of thrombus removal from a pulmonary artery (e.g., pulmonary embolectomy), the technology may be applied to removal of thrombi and / or emboli from other portions of the vasculature (e.g., in neurovascular, coronary, or peripheral applications). Moreover, although some embodiments are discussed in terms of maceration of a thrombus with a fluid, the present technology can be adapted for use with other techniques for breaking up a thrombus into smaller fragments or particles (e.g., ultrasonic, mechanical, enzymatic, etc.).

[0070] The headings provided herein are for convenience only and do not interpret the scope or meaning of the claimed present technology.Systems for Thrombus Removal

[0071] As provided above, the present technology is generally directed to thrombus removal systems. Such systems include an elongated catheter having a distal portion positionable within a blood vessel of the patient (e.g., an artery or vein), a proximal portion positionable outside the patient's body, a fluid delivery mechanism configured to fragment the thrombus with pressurized fluid, an aspiration mechanism configured to aspirate the fragments of the thrombus, and one or more lumens extending at least partially from the proximal portion to the distal portion. In some embodiments, the systems herein are configured to engage a thrombus in a patient's blood vessel, break the thrombus into small fragments within the confines of the catheter system, and aspirate the fragments out of the patient's body. The pressurized fluid streams (e.g., jets) function to cut or macerate thrombus, before, during, and / or after at least a portion of the thrombus has entered the aspiration lumen or a funnel of the system. Fragmentation helps to prevent clogging of the aspiration lumen and allows the thrombus removal system to macerate large, firm clots that otherwise could not be aspirated. As used herein, “thrombus” and “embolism” are used somewhat interchangeably in various respects. It should be appreciated that while the description may refer to removal of “thrombus,” this should be understood to encompass removal of thrombus fragments and other emboli as provided herein.

[0072] According to embodiments of the present technology, a fluid delivery mechanism can provide a plurality of fluid streams (e.g., jets) to fluid apertures of the thrombus removal system for macerating, cutting, fragmenting, pulverizing and / or urging thrombus to be removed from a proximal portion of the thrombus removal system. The thrombus removal system can include an aspiration lumen extending at least partially from the proximal portion - 10 - SG Docket No.: 10844-737.674to the distal portion of the thrombus removal system that is adapted for fluid communication with an aspiration pump (e.g., vacuum source). In operation, the aspiration pump may generate a volume of lower pressure within the aspiration lumen near the proximal portion of the thrombus removal system, urging aspiration of thrombus from the distal portion.

[0073] FIGS. 1 A-1B illustrate a vascular access and treatment system 100 that can include a catheter 102 and optionally a medical device 108 disposed within a lumen of the catheter. The catheter can include an elongate, steerable, flexible shaft and a distal end 103 at the end of one or more lumens that runs along the shaft of the catheter. The catheter can include one or more sensors 105 disposed along, in, or within the shaft 101, including but not limited to pressure sensors, flow sensors, electrical sensors (electrodes), or any other sensor useful for measuring patient parameters or parameters of the catheter during an intravascular procedure. In the example of FIGS. 1 A-1B, the sensor 105 can comprise a pressure sensor disposed near the distal end 103.

[0074] As mentioned above, the catheter 102 can include one or more steering wires (not shown) to enable steering in multiple degrees of freedom. In some examples, the steering wires can terminate at or near the distal end of the catheter. In other examples, the steering wires can terminate at selected locations along the catheter to allow for controlled steering at those selected locations. Any number of sensors can be associated with the steering mechanism(s), including but not limited to force sensors on the steering wires, force sensors on the catheter shaft, or torque sensors on the shaft handle interfaces. Additionally, sensors can be provided that can track advancement or retraction of the catheter shaft within the patient. For example, encoders or optical sensors can be located on or near the catheter shaft external to the patient (e.g., at the access site) and can track advancement or retraction of the catheter as it is inserted or retracted into the access site. As will be discussed in more detail below, any number of the sensors or features of the catheter can be used to track, estimate, or calculate a position, location, orientation, and / or configuration of the catheter during a procedure. Additionally, the sensor(s) and features can be used to model the catheter into a 3D model or representation of the catheter, which can be presented in real-time to a user, including within a 3D model of the pulmonary vasculature.

[0075] In some embodiments, an interior lumen of the catheter can be coupled to a vacuum source or aspiration source, to provide vacuum within the interior lumen of the catheter. The vacuum source can be configured to aspirate thrombus or clots into the interior lumen to remove clots form a subject. Additionally, the distal end of the catheter may include two or more fluid ports having axes directed at each other or towards a common intersection point, and configured to produce jets or fluid streams that intersect within the - 11 - SG Docket No.: 10844-737.674distal end of the catheter at the intersection point to cut or break up thrombus. In the illustrated example, four fluid ports are shown to direct four intersecting jets towards a common intersection point within the distal end of the catheter.

[0076] The catheter 102 can optionally receive a separate medical device within a lumen, such as the interior aspiration lumen. The medical device 104 can comprise any elongate medical device insertable into the lumen of the catheter, including but not limited to balloon angioplasty catheters, dilators, guidewires, or thrombectomy devices. As shown in FIG. 1A, the medical device 104 comprises a thrombectomy device with an elongate shaft 106 and an optional expandable element or funnel 108 on a distal end of the shaft. The funnel and / or shaft can include one or more lumens 107, e.g., for aspiration of thrombus material and / or delivery of fluid or jets from the thrombectomy device. The funnel 108 can be advanced out of the distal end 103 of the catheter during a thrombectomy procedure. The medical device can have a smaller bore or outer diameter than that of the catheter 102, such that the medical device can be delivered to a target thrombus location within the interior lumen of the catheter 102. The medical device may include a similar aspiration and jetting system to that of the catheter 102, as described above.

[0077] In FIG. 1 A, a hub assembly 110 such as a Touhy Borst is shown which can provide access for the medical device 108 into the lumen of the steerable introducer catheter 102 and include an injection port for fluidic connection a fluid or contrast source 112. The injection port can direct the fluid or contrast into the lumen(s) of the introducer catheter. In the illustrated embodiment, the fluid or contrast source 112 can comprise a contrast injector configured to automatically or manually deliver a controlled volume (e.g., bolus) of a contrast agent into the patient’s vasculature via the introducer catheter 102. In some examples, injection of contrast from the injector into the hub assembly 110 provides the contrast agent into the annular space between the introducer catheter 102 and the medical device 108 (e.g., within the lumen of the introducer catheter, between the introducer catheter shaft and the shaft 106 of the medical device).

[0078] FIG. IB shows the funnel 108 of the medical device 108 axially disposed out of a distal end 103 of the introducer catheter 102. When the medical device is a thrombectomy or aspiration catheter, aspiration or vacuum generated in lumen 107 can pull thrombus material into the funnel 108 and out of the device via lumen 107. In some embodiments, jets or fluid streams can also be delivered into the funnel or aspiration lumen to interact with and / or macerate the thrombus material. In this example, contrast delivered by the fluid or contrast source 112 into the lumen of the introducer catheter can still be delivered into the patient, even when the funnel is in an expanded state. In some examples, the funnel can disperse the - 12 - SG Docket No.: 10844-737.674contrast agent as it is delivered past the funnel from the introducer catheter. Alternatively, a dilator device or other medical device can be inserted into the introducer catheter, as will be described below.

[0079] Additionally, FIG. IB shows the medical device 108 with a plurality of 3D sensors 150 disposed on or in the device. The 3D sensors can comprise, for example, spatial position sensors, or spatial mapping sensors that can provide position and orientation information of the device, including approximations or calculations about the curvature of the device. The 3D sensors on the device can work in conjunction with an external transmitter / receiver (external device 352 in FIG. 3 A) to generate 3D maps and pinpoint the exact location and orientation of the medical device in the pulmonary arteries during diagnosis and therapeutic procedures for patients with pulmonary embolisms. In some embodiments, the 3D sensors comprise magnetic sensors. While the sensors 150 are shown on the medical device in FIG. IB, they could also be on the introducer sheath shaft 101, or on both the introducer sheath and the medical device. The sensors can be positioned at selected points on either the medical device and / or the introducer sheath, including at or near a distal end of the medical device or sheath to provide information on the relative position between the two devices.

[0080] One or more of the 3D sensors can be used to define a path and curvature of the medical device, which will improve potential position estimation errors / speed up rendering. In some examples, the 3D sensors can use position at the catheter / sheath in a region of the heart such as the right atrium (RA), right ventricle (RV), root of the pulmonary artery (PA), and / or left pulmonary artery (LPA) or right pulmonary artery (RPA) to capture fiducial positions to register 3D models of the pulmonary arteries, CT or other imaging modalities with the 3D sensors and / or medical device. The sensors can also be coupled with real-time fluoroscopy captures to scale / register fluoroscopy to magnetic / CT as well.

[0081] The 3D sensors can provide a minimum-fluoroscopy / fluoroscopy-free thrombectomy procedure in real-time, especially the system utilizes pressure waveform morphology to validate position of catheter / sheath.

[0082] The fluid or contrast source 112 (e.g., contrast injector) can be configured to automatically inject or deliver selected volumes or boluses of any contrast agent into the thrombus removal system to assist with imaging of the thrombus removal device and / or a target thrombus. In some embodiments, while the volumes and timing of contrast to be delivered by the injector are selected by a user or pre-selected, the injector can be configured to automatically and / or continuously deliver contrast at the selected volumes and frequency. In the illustrated embodiment, the fluid or contrast source can comprise a cradle assembly - 13 - SG Docket No.: 10844-737.674configured to receive one or more contrast injection syringe(s). The cradle assembly can include an automatic pusher or other mechanism configured to engage with the syringe to inject a contrast agent into the lumen(s) of the introducer catheter.

[0083] The system 100 can employ control systems, algorithms, or protocols to provide consistent or controlled injection of fluid or contrast agent near the distal end of the introducer catheter. In some embodiments, the fluid or contrast source can be configured to inject a pre-determined or pre-selected bolus or volume of fluid or contrast agent into the patient at the target location within the vasculature. For example, the fluid or contrast source may be configured to deliver a bolus of contrast agent (e.g., a 5ml bolus or “shot” of contrast) at a pre-determined time interval (e.g., every 3-5 seconds).

[0084] While the embodiments herein have been described as being intended to remove thrombi from a patient’s vasculature, other applications of this technology are provided. For example, the devices described herein can be used for breaking up and removing hardened stool from the digestive tract of a patient, such as from the intestines or colon of a patient. In one embodiment, the device can be inserted into a colon or intestine of the patient (such as through the anus) and advanced to the site of hardened stool. Next, the aspiration system can be activated to engage the hardened stool with an engagement member (e.g., funnel) of the device. Finally, the jets or irrigation can be activated to break off pieces of the hardened stool and aspirate them into the system. Any of the techniques described above with respect to controlling the system or removing clots can be applied to the removal of hardened stool.

[0085] As background, FIG. 2 is a diagram of the pulmonary vasculature. Clots or pulmonary embolisms are typically found within the pulmonary vasculature at locations 1 (left and right pulmonary arteries), 2 (left and right interlobar pulmonary arteries), and 3 (left and right segmental branches including the anterior, superior, and lateral branches).Accessing clots in each of these locations with a thrombectomy catheter or device has challenges. For example, while the left and right pulmonary arteries are relatively large and easy to access, the size of the pulmonary arteries can make it difficult to position a thrombectomy catheter near or adjacent to a target clot, even with fluoroscopy imaging. Interfacing or accessing clots at locations 2 and 3 can be even more challenging, as the tortuous pathways and branching into smaller lumens can be difficult or impossible to navigate with current imaging techniques like fluoroscopy.

[0086] To overcome the limitations described above with respect to navigating the pulmonary vasculature with a medical device such as a thrombectomy catheter, the present disclosure provides systems and methods for providing real-time 3D navigation of the pulmonary vasculature, including 3D models of the pulmonary vasculature with overlays - 14 - SG Docket No.: 10844-737.674representing the medical device (e.g., thrombectomy catheter) within the pulmonary vasculature for improved navigation, steering, and positioning during medical procedures.

[0087] In one embodiment shown in FIGS. 3A-3B, a system 300 is shown configured to generate a 3D model of an anatomical structure 301 of a subject (e.g., the pulmonary arteries or pulmonary vasculature, the heart, coronary vasculature, peripheral vasculature, etc.). The system 300 can include, for example, an imaging system 314 such as a fluoroscopy imaging system, a computed tomography (CT) imaging system, a cone-beam CT imaging system (CBCT), a magnetic resonance imaging (MRI) imaging system, an ultrasound imaging system, any other high resolution medical or diagnostic imaging system, contrast-enhanced imaging (e.g., angiography), or any combination of these imaging systems. In the illustrated example, the imaging system 314 can include a C-arm 316 that supports an X-ray irradiation unit (not shown) and an X-ray detector 318. A patient table 320 and support base 322 can be positioned within the C-arm to image a patient.

[0088] The system can further include a console 324 having one or more optional displays 326. The console can include one or more input devices such as a graphical user interface (GUI), a keyboard, mousejoystick, etc. to allow a user to control the system including the imaging system and to view, manipulate, or interact with 3D models of the anatomical structure produced by the system.

[0089] The imaging system 314 can be operatively connected to the console 324.Processing of the data collected by the imaging system may be accomplished via electronics and software in the console. The console may include, for example, various processors, power supplies, memory, firmware, and software configured to receive, store, and process imaging data collected by the imaging system 314.

[0090] The 3D anatomical models can be presented on a display of the system, such as the one or more displays 326 of FIG. 3 A. The user or clinician can interact with one or more inputs or icons, such as with a GUI or with other input devices such as a keyboard or mouse. The inputs or icons allow the user to select various features or functionality of the 3D model, including revealing or hiding specific features of the vasculature, revealing or hiding specific clots, selecting or highlighting specific vessels or clots, colorizing or adjusting the contrast of specific vessels or clots, or adjusting the view angle or zoom level off the 3D model.

[0091] An external device 352 for interacting / interfacing with the 3D sensors to illustrate tool paths described in FIG. IB is also shown in FIG. 3 A. While the external device is shown near the console, it could also be situated on table 320. The external device can include a locator pad with three separate low-level magnetic field emitting coils being arranged as a triangle under the patient, configured to work with the catheter or the medical device (e.g.,- 15 - SG Docket No.: 10844-737.674thrombectomy catheter) of FIGS. 1A-1B with embedded 3D magnetic location sensors. The external device can include, for example, a data processing unit and a graphic display unit to provide visualization of the electroanatomical model being created. The field strength of the three electromagnets can be measured by a sensor element of the catheter tip and used for position determination via a triangulation algorithm which allows an exact calculation of the distance from each magnetic coil.

[0092] In various embodiments, the system uses scan, imaging and / or patient data to generate information for a clinician or user of the system. For example, the system may construct a 3D anatomical model based on medical scan or image data taken from multiple locations (e.g., multiple C-arm positions of the imaging system). In another example, the system generates image data based on composite data from multiple locations, and / or from multiple imaging modalities. For example, the system may generate the 3D model from any combination of imaging data including CT, X-ray, fluoroscopy, MRI, ultrasound, etc.

[0093] The system may further use patient or physiological data or information in constructing the 3D anatomical model, including but not limited to patient age, sex, medical history, disease state of the patient (e.g., whether the patient is diagnosed with any known diseases relevant to the target anatomy), prior surgical or medical treatment history, physical exam results, or vital signs including but not limited to blood pressure, ECG, respiration, etc.

[0094] In some embodiments, the position and location of a medical device (e.g., a thrombectomy catheter) can be overlaid or tracked in real-time in the 3D anatomical model. For example, a graphical overlay or representation of the medical device can be displayed on the 3D anatomical model. As a physician or clinician navigates or advances the medical device within the anatomy (e.g., within the pulmonary vasculature) while referencing the 3D anatomical model on the display(s), the graphical overlay or representation of the medical device on the 3D anatomical model can track the precise location of the medical device within the patient.

[0095] Several embodiments are provided for tracking or determining the actual position and orientation of the medical device within the patient so that the system can provide the graphical overlay of the medical device on the 3D model. Alternatively, the system can generate a 3D model of the device, and incorporate the 3D model of the device into the 3D model of the pulmonary vasculature. In one example, the length of the catheter inserted into the patient past some known reference point (e.g., the femoral or brachial access points) can be monitored, such as with optical sensors or rotational encoders. The amount and / or degree of steering of the catheter can also be tracked or determined with similar sensors that monitor the steering mechanism of the catheter (e.g., pull wire displacement). The system can then - 16 - SG Docket No.: 10844-737.674model the orientation of the catheter based on the insertion length and steerage, or from information such as 3D sensors on the device, and use that information to provide the graphical overlay or 3D model of the device in the 3D model of the anatomy. In another embodiment, radiopaque markers or other trackable sensors can be placed at known positions along the length of the medical device, including along the catheter shaft and at the tip or any other deflection points of the device. The markers can be periodically imaged (e.g., with fluoroscopy) and registered to the 3D model using known landmarks in the anatomy. Other known techniques for tracking the position and orientation of a medical device such as a catheter within the body can also be implemented.

[0096] FIG. 4A is a diagram showing an example of a system 400; the system 400 may be incorporated into a portion of another system (e.g., a general treatment planning system, as described below) and may therefore also be referred to as a sub-system. Alternatively the methods and apparatuses for performing them described herein may be included as part of a different system. In any of the methods and apparatuses described herein, the system 400 may be invoked by a user control, such as a tab, button, etc., as part of treatment planning system, as part of a navigation system, or may be separately invoked.

[0097] In FIG. 4A, the system 400 may include a plurality of engines and datastores. A computer system can be implemented as an engine, as part of an engine or through multiple engines. As used herein, an engine includes one or more processors, such as a GPU, CPU, ASIC, FPGA, or a portion thereof. A portion of one or more processors can include some portion of hardware, less than all of the hardware comprising any given one or more processors, such as a subset of registers, the portion of the processor dedicated to one or more threads of a multi-threaded processor, a time slice during which the processor is wholly or partially dedicated to carrying out part of the engine’s functionality, or the like. As such, a first engine and a second engine can have one or more dedicated processors, or a first engine and a second engine can share one or more processors with one another or other engines. Depending upon implementation-specific or other considerations, an engine can be centralized, or its functionality distributed. An engine can include hardware, firmware, or software embodied in a computer-readable medium for execution by the processor. The processor transforms data into new data using implemented data structures and methods, such as is described with reference to the figures herein.

[0098] The engines described herein, or the engines through which the systems and devices described herein can be implemented, may be cloud-based engines. As used herein, a cloud-based engine is an engine that can run applications and / or functionalities using a cloudbased computing system. All or portions of the applications and / or functionalities can be - 17 - SG Docket No.: 10844-737.674distributed across multiple computing devices, and need not be restricted to only one computing device. In some embodiments, the cloud-based engines can execute functionalities and / or modules that end users access through a web browser or container application without having the functionalities and / or modules installed locally on the end-users’ computing devices.

[0099] As used herein, datastores are intended to include repositories having any applicable organization of data, including tables, comma-separated values (CSV) files, traditional databases (e.g., SQL), or other applicable known or convenient organizational formats. Datastores can be implemented, for example, as software embodied in a physical computer-readable medium on a specific-purpose machine, in firmware, in hardware, in a combination thereof, or in an applicable known or convenient device or system. Datastore-associated components, such as database interfaces, can be considered “part of’ a datastore, part of some other system component, or a combination thereof, though the physical location and other characteristics of datastore-associated components is not critical for an understanding of the techniques described herein.

[0100] Datastores can include data structures. As used herein, a data structure is associated with a particular way of storing and organizing data in a computer so that it can be used efficiently, within a given context. Data structures are generally based on the ability of a computer to fetch and store data at any place in its memory, specified by an address, a bit string that can be itself stored in memory and manipulated by a program. Thus, some data structures are based on computing the addresses of data items with arithmetic operations; while other data structures are based on storing addresses of data items within the structure itself. Many data structures use both principles, sometimes combined in non-trivial ways. The implementation of a data structure usually entails writing a set of procedures that create and manipulate instances of that structure. The datastores described herein can be cloud-based datastores. A cloud-based datastore is a datastore that is compatible with cloud-based computing systems and engines.

[0101] The system 400 may include or be part of a computer-readable medium, and may include an input engine 401 (e.g., providing and / or allowing access to the patient’s scan or imaging data, patient medical history, and / or patient characteristic(s)). The scan or imaging data 403 may include two-dimensional (2D) or three-dimensional (3D) scan or imaging data provided by a medical imaging device, including but not limited to ultrasound images, X-ray images, computed tomography (CT) images, angiogram images (pulmonary, coronary, or peripheral), real-time fluoroscopy images, magnetic resonance imaging (MRI) images, positron emission tomography (PET) images, or the like. In some embodiments, the input - 18 - SG Docket No.: 10844-737.674engine 401 may receive training images, including supervised training images. In some embodiments, the input engine 401 may receive synthetic training images generated from other patient imaging or scan data. Additionally, the input images may be run through an inference engine. As will be described herein, the training images may be used to train one or more neural networks.

[0102] The system 400 may include an anatomical segmentation engine 402 that may segment 3D models into different objects, sections, parts, or the like. Segmentation may be performed in any feasible manner. In some variations, the anatomical segmentation engine 402 may also process (e.g., convert, transform) 2D or 3D imaging or scan data into a 3D model. For example, the 2D or 3D imaging or scan data can include medical imaging data (e.g., X-ray, MRI, CT, ultrasound) of a target tissue, such as the pulmonary vasculature. In some embodiments, the 2D or 3D imaging data is processed to focus on the target tissue. For example, a chest CT may be cropped down to focus on the pulmonary vasculature. In some embodiments, the target tissue includes coronary vasculature and / or peripheral vasculature. For example, the anatomical segmentation engine 402 can receive 2D or 3D scan or imaging data of the patient’s pulmonary vasculature from the input engine 401, generate 3D models of the patient’s pulmonary vasculature and then optionally segment the 3D models into separate objects, sections, parts, or the like (e.g., into the left and right pulmonary arteries, or any of the lobes, branches or segments described above in FIG. 2). For example, the anatomical segmentation engine 402 can receive 2D or 3D scan or imaging data of the patient’s coronary or peripheral vasculature and then optionally segment the 3D models into separate objects, sections, parts, or the like representative of the imaged anatomy. In some examples the system 400 may include a memory, register or datastore storing all or some of the patient’s scan or imaging data, patient medical history, and / or patient characteristic(s). Patient characteristics may include patient demographical info such as age, gender, vital signs, and the like.

[0103] The system 400 may also include a clot segmentation engine 404. The clot segmentation engine may segment 3D models into different objects, sections, parts, surfaces, or the like. Segmentation may be performed in any feasible manner. In some variations, the clot segmentation engine 402 may also process (e.g., convert, transform) 2D or 3D imaging or scan data into a 3D model. For example, the 2D or 3D imaging or scan data can include medical imaging data (e.g., X-ray, fluoroscopy, MRI, CT, ultrasound, or fusion of any of the preceding imaging modalities) of a target tissue, such as the pulmonary vasculature. The imaging data can include not only the patient’s pulmonary vasculature, but also imaging data of clots or thrombus within the pulmonary vasculature. For example, the clot segmentation - 19 - SG Docket No.: 10844-737.674engine 402 can receive 2D or 3D scan or imaging data of the patient’s pulmonary vasculature from the input engine 401, generate 3D models of clots within the patient’s pulmonary vasculature, and then optionally segment the 3D models into separate objects, sections, parts, or the like. Alternatively, the clot segmentation engine 402 can receive the 3D model of the patient’s pulmonary vasculature from the anatomical segmentation engine 402, generate 3D models of clots within the patient’s pulmonary vasculature, and then optionally segment the 3D models into separate objects, sections, parts, or the like. The 3D model of the clots can be integrated with or combined with the 3D model of the patient’s pulmonary vasculature.

[0104] FIG. 8A, for example, shows an example of the pulmonary vasculature (PV) and clots (C) segmented from the 2D or 3D imaging data in various views (e.g., lateral, anterior, posterior, etc.). The clot segmentation engine may be configured to not only segment and generate 3D models of the clots, but also to identify the location of the clots within the anatomy. In some embodiments, the clot segmentation engine is configured to determine the volume or size of the clots compared to or relative to the vasculature. Additionally, the clot segmentation engine can determine the type or age of the clot, or alternatively, the density of the clot. The density / toughness of clot may be estimated based on assessed imaging parameters for a given imaging modality. For example, for X-ray imaging modalities signal attenuation (e.g., in Hounsfeld units) and / or contrast uptake can be assessed. For ultrasound imaging modalities, ultrasonic attenuation or ultrasonic backscatter can be assessed. For magnetic resonance imaging (MRI), signal intensity parameters can be showing navigation. In some embodiments, the engine(s) may characterize the effect of the clots or lesions on blood flow.

[0105] While herein segmentation of occlusive material may refer to ‘clot’ or ‘embolism,’ it should be appreciated that other occlusive blockages within the anatomy are contemplated. For example, the segmentation engine 404 can alternatively or additionally segment vascular lesions. The lesions can be calcified lesions. In some embodiments calcified lesions can correspond to locations within the coronary and / or peripheral vasculature of the patient.

[0106] Segmentation performed by the anatomical segmentation model and the clot segmentation model may classify pixels from imaging data, or from a 3D model into structures in another 3D structural model, such as pulmonary arteries, artery volumes, artery surfaces, or structures in an artery.

[0107] Both the anatomical segmentation engine 402 and the clot segmentation engine 404 may employ machine learning models, including neural networks. The machine learning models or neural networks for anatomical segmentation engine 402 and clot segmentation - 20 - SG Docket No.: 10844-737.674engine 404 may be trained to detect and identify the pulmonary vasculature and clots, thrombus, or other abnormalities within the pulmonary vasculature. While the present disclosure describes the anatomical segmentation engine 402 and the clot segmentation engine 404 as being separate components or engines, it should be understood that a single segmentation engine could segment both the patient anatomy and any lesions / clots within the anatomy, and provide 3D models of the anatomy and the clots. The clot’s position and / or orientation within the pulmonary vasculature, or within the 3D model of the pulmonary vasculature may be used, at least in part, to determine a patient’s treatment plan or provide a treatment recommendation or assessment. In some variations, the system 400 may store a library of training images or supervised training images 404. These images may be used to train the machine learning model or neural network for anatomical segmentation engine 402 or clot segmentation engine 404.

[0108] The system 400 may include a treatment plan engine 406. The treatment plan engine 406 may process patient imaging or scan data, 3D model or segmentation data from the anatomical segmentation engine 402 and / or the clot segmentation engine 404, patient characteristics, clinician input and the like to determine a patient’s treatment plan or provide a treatment assessment or recommendation. In some variations, a patient’s treatment plan, assessment, or recommendation may include identifying target clots or thrombi for removal. The treatment plan engine 406 may also provide an assessment or recommendation of clots or thrombi not to target (e.g., clots that are too deep within the vasculature, or would not result in significant patient improvement if removed). In some variations, the treatment plan engine 406 can generate a pre-operative assessment of the patient’s condition and / or a post-operative assessment of the patient’s condition if specific action is taken. For example, the treatment plan engine may calculate or determine pre and post-operative Miller Scores, Modified Miller Scores (MMS), or some other quantifiable measurement of the patient’s condition and / or outcome. In yet another embodiment, the treatment plan engine 406 may identify clots or thrombi to target for removal (e.g., with a thrombectomy system), and may provide or calculate an estimated Miller Score or Modified Miller Score if the targeted clots or thrombi are removed.

[0109] A Miller Score, for example, provides an assessment of flow in the pulmonary vasculature, and considers the nine major branches of the right pulmonary artery and the seven major branches of the left pulmonary artery, and an embolus in any of these branches is given a score of one point. Each lung is further considered to have an upper, middle, and lower zone, and in each of these three zones, the absence of a pulmonary artery flow confers- 21 - SG Docket No.: 10844-737.674a score of three points, severely reduced flow gives two points, mildly reduced flow gives one point, and normal flow scores 0 points. The Miller Score can therefore range from 0 to 3.

[0110] Any of these apparatuses or systems may include an output engine 410 for outputting the treatment plan from treatment planning engine 406. The system may also include a display or graphical user interface (GUI) 412 configured for displaying the 3D models and data discussed above. In some embodiments, the display or GUI 412 presents the 3D model of the anatomy, such as the pulmonary vasculature, and also presents the segmented clots and / or 3D model of the clots, including their location and orientation within the 3D model of the pulmonary vasculature. The display can further present or highlight clots to be targeted, or recommend clots for removal. In some examples, the display or GUI can present a Miller Score, Modified Miller Score, or some other quantitative assessment of patient outcomes if targeted clots are removed. Additionally, the display or GUI can provide real-time tracking or navigation of the 3D model of the anatomy. In some examples, the display or GUI can include a real-time graphical overlay or model of an interventional device such as a thrombectomy catheter, within the 3D model of the anatomy. Additionally, the display or GUI can provide or present a procedural plan for tracking the device to one or more target clots or lesions, including optionally providing real-time navigation instructions or directions for the device and any other aspects of the system including introducer sheaths, etc.[oni] In some embodiments, an image system according to the present invention is configured to comprise a training engine or user-generated feedback for anatomy and / or occlusive material segmentation. As shown in FIG. 4B, a system 450 includes many similar features to those described of system 400 in FIG. 4A: input engine 451; patient imaging data 453; anatomical segmentation engine 452; clot segmentation engine 454; library of training images 458; display or GUI 462; treatment planning engine 466; and output engine 460. FIG.4B further includes an inference engine 455, an optional training engine 465, and optional user-generated segmentation engine 470. The inference engine 455 can comprise a trained model for segmentation. The inference engine 455 can be implemented locally (e.g., “on premises”) with respect to system 450, or the inference engine 455 can be implemented in the cloud and communicate with system 450 via a network connection.

[0112] In some embodiments of the system 450, anatomical segmentation engine 452 and clot segmentation engine 454 are configured to output to, and receive inputs from, usergenerated segmentation engine 470 to provide the system 450 with user-identified improvements in segmentation. For example, a user may be able to view the 3D models generated by the segmentation engines 452 / 454, and mark-up or further identify features to - 22 - SG Docket No.: 10844-737.674segment, including additional branches or features of vasculature, and / or identifying additional clots / lesions, or marking features segmented as obstructions (e.g., clots) as not being obstructions. For example, the user may affirmatively agree with the segmentation provided by the system, or alternatively, may provide edits or further input that modifies or changes the initial segmentation. In some embodiments the user may interact with an input device such as a mousejoystick, keyboard, digital pen, or graphical user interface (GUI) to directly draw, mark, or “paint” the initial segmentation to edit labels or refine segmentation. In some embodiments, this painting may be performed on one or more 2D or 3D views of the imaging data. For example, referring to FIG. 8B, the user may mark, paint, annotate, or draw directly on the segmented model data to refine, edit, or change aspects of the segmentation. In the illustrated example, the user has marked where clot resides within the segmented pulmonary vasculature. In some embodiments, the user-input information from may be used by system 450 for further segmentation training by training engine 465. In some embodiments the training engine 465 is locally instantiated (e.g., “on premises”). In some embodiments the training engine 465 is instantiated in the cloud. In some embodiments, the inference engine 455 can be updated based on inputs from the training engine 465 and / or user-generated segmentation engine 470. In some embodiments the updates can be in realtime.

[0113] One or more of the engines of the systems 400 / 450 may be coupled to one another (e.g., through the example couplings shown in FIGS. 4A-4B) or to modules / engines not explicitly shown in FIGS. 4A-4B. The computer-readable medium may include any computer-readable medium, including without limitation a bus, a wired network, a wireless network, or some combination thereof.

[0114] FIG. 5A schematically illustrates processes and / or steps associated with generating a 3D model of the pulmonary vasculature and any clots or lesions, and providing a recommendation or assessment to a user regarding clots to target or not target for removal. In general, the assessment or recommendation of clots to target may be determined based on a treatment plan. Clots to target may be determined from patient imaging or scan data, such as 2D or 3D imaging data of the patient’s pulmonary vasculature. The imaging or scan data can be converted into a model (e.g., a 3D model) that can include the pulmonary vasculature including the left and right pulmonary arteries and the various branches of the pulmonary arteries, and any clots, thrombi, or other lesions disposed within the pulmonary vasculature. In some examples, a trained machine learning model or neural network (also referred to as a trained machine learning agent) may be executed to identify clots within the 3D model and provide a pre and or post treatment assessment or quantifiable analysis of the patient’s health - 23 - SG Docket No.: 10844-737.674or condition if one or more of the clots, thrombi, or lesions are removed (e.g., via a thrombectomy procedure).

[0115] Some clots or lesions may be more difficult to detect or locate, particularly when the clots are of a particular age, location within the anatomy, or size. A machine learning model or neural network, trained with models that include these types of clots or lesions, may more accurately locate and / or identify clots or lesions, clot types, or provide assessments or recommendations on clots to target or remove.

[0116] Patient imaging or scan data 502 is converted to a 3D model 506 at block 508, which can include extracting image data comprising special information associated specifically with pulmonary vasculature features. As described above, user input can further confirm or modify the segmentation, including the size and location of any clots within the vasculature. The 3D model can include a point cloud data representation of the patient’s pulmonary vasculature and the size and location of any clots, thrombi, or lesions, which can be presented to the user. For example, referring to FIG. 8C, the system can generate size / volume estimations or calculations of a selected clot and additionally provide 3D positional data of the clot (e.g., the center of a selected clot). While specific x,y,z coordinates are shown in FIG. 8C (e.g., in the imaging data coordinate space), other ways of conveying position information can be provided. For example, the system may instead convey the coordinates as an anatomical location (e.g., LUL apical posterior seg branch, or any other location within the target anatomy such as those described in FIG. 2). Segmented clots, thrombi, lesions, or other data from the 3D model can be provided to a machine learning model or neural network at block 508 for determining additional information or parameters about the clots or lesions, including the clot type, or providing an assessment or recommendation of what clots to target, the expected patient outcome or patient improvement if selected clots are removed, and / or navigation guidelines or instructions for navigating to the targeted clot(s).

[0117] FIG. 5B is a flowchart 500 describing one method of generating a 3D model of a target anatomy. The method can comprise generating a 3D model from 2D images using one or more anatomical landmarks in the images and knowledge of the C-arm position when each 2D image is obtained. Referring to step 501 of flowchart 500, the method can include obtaining a plurality of 2D CT images of a target anatomy, such the pulmonary vasculature, at a plurality of C-arm angles (e.g., a new C-arm position for each 2D image).

[0118] Next, at step 503 of flowchart 500, the method can include identifying one or more anatomical landmarks in each of the plurality of 2D CT images. The landmark may comprise, for example, specific organs, bones, nerves, or vascular locations. For example, if - 24 - SG Docket No.: 10844-737.674the anatomical location comprises the pulmonary vasculature, the anatomical landmarks may include features of the heart or surrounding vasculature, the lungs, branches, or surrounding vasculature, or, for example, the pulmonary trunk, the branching of the left and right main pulmonary arteries, or any of the segmental branches.

[0119] At step 505 of flowchart 500, the method can include generating a 3D model with a correlation function using the plurality of C-arm angles from step 501 and the one or more anatomical landmarks from step 503. The anatomical landmarks in combination with the known pose / orientation / angle of the c-arm when the image was obtained allows for each 2 image pixel to be assigned a location in 3D space. Then all of the 2D images can be correlated into this 3D model.

[0120] FIG. 6 is a flowchart showing an example method 600 for training a machine learning model or neural network to provide an assessment or recommendation on clots to target or treat. Some examples may perform the operations described herein with additional operations, fewer operations, operations in a different order, operations in parallel, and some operations differently. Patients with pulmonary embolisms may have numerous clots, emboli, thrombi, or lesions within the pulmonary vasculature. Some clots or thrombi may be a higher priority for removal, or lead to improved patient outcomes, relative to other clots or thrombi which may not lead to patient improvement, or may be too difficult to reach or remove. Thus, various factors on clot location, size, and type may be used to determine the location of the clots or lesions to be targeted. Any of the machine learning models or neural networks described herein may be used to identify or recommend one or more clots or lesions for removal or treatment. The method 600 is described below with respect to the system 400 of FIG. 4A, however, the method 600 may be performed by any other suitable system or device.

[0121] In block 610, the system 400 receives supervised training data. Supervised training data can include images or scans of the pulmonary vasculature that have been manually labeled by skilled personnel. The labeled images include any and all clots or lesions or other characteristics that the machine learning model or neural network will be trained to recognize. In some variations, the supervised training data can be labeled to identify clot locations, clot sizes or volumes, and / or clot types or densities.

[0122] In some variations, the supervised training data is “balanced.” That is, the supervised training data should include cases with regular anatomy and unusual anatomy in approximately equal proportions. In some cases, some of the training data can be manipulated or synthesized to transform regular pulmonary anatomies into unusual pulmonary anatomies. Alternatively, clots or lesions can be added to pulmonary anatomies without clots or lesions.- 25 - SG Docket No.: 10844-737.674In some examples, the supervised training data can include up to 10,000 different cases, up to 50,000 different cases, or up to 100,000 different cases.

[0123] In block 620, the system 400 trains the machine learning model or neural network with the supervised training data received in block 610. As described above, the system 400 can train the machine learning model or neural network (machine learning agent) to respond with two channels. A first channel can identify all clots or lesions within the pulmonary vasculature and a second channel can identify clots or lesions to be targeted, and optionally provide an assessment or score of an expected patient outcome if the targeted clots or lesions are removed. The neural network can be trained to respond with the first and second channels to identify the classes described herein within the point cloud files.

[0124] The training of a neural network to identify or locate clots or lesions to be treated may be associated with various aspects of a computing environment that is used to determine pulmonary embolism treatment. By way of example and not limitation, the training of a neural network may be associated with treatment planning or a treatment planning system. For example, a treatment planning system may include one or more modules configured to receive or obtain supervised training data and determine or train a neural network to identify or locate clots or lesions to be treated based on the supervised training data.

[0125] FIG. 7 is a flowchart showing an example method 700 for providing pre-operative planning for pulmonary embolism. The method 700 is described below with respect to the system 400 of FIG. 4A, however, the method 700 may be performed by any other suitable system or device.

[0126] The method 700 begins in block 702 as the system 400 converts 2D or 3D imaging or scan data into one or more 3D models. In some variations, the 2D or 3D scan data can be imaging data provided by a medical imaging device, including but not limited to ultrasound images, X-ray images, computed tomography (CT) images, magnetic resonance imaging (MRI) images, positron emission tomography (PET) images, or the like. The system 400 can transform the imaging or scan data into a 3D model that includes a 3D model of the patient’s pulmonary vasculature and any clots or lesions within the anatomy. In some examples, the 3D model of the pulmonary vasculature is separate from 3D models of each clot or lesion identified from the imaging or scan data. In other embodiments, the anatomy and clots are located within the same 3D model.

[0127] Next, in block 708, the system 400 segments the 3D model(s). In some examples, the system 400 can segment the 3D model into individual vessels, arteries, branches, clots, and / or lesions. In some cases, the system 400 can execute a machine learning model or neural- 26 - SG Docket No.: 10844-737.674network, as described above, to segment the 3D model into various portions of the anatomy and the clots / lesions.

[0128] Next, in block 706 the system 400 selects or identifies one or more clots or lesions in the 3D model. The selected clot(s) can be any or all clots in the 3D model. In block 708, the system 400 determines if the selected clot(s) can be accessed via the thrombectomy removal device (e.g., the thrombectomy catheter). Numerous factors can be used to determine whether or not clots can be accessed. For example, clots within lumens or branches of the pulmonary vasculature smaller than a minimum diameter may be flagged as inaccessible. Alternatively, any clots more distal (e.g., further within the anatomy) to a specific portion of the pulmonary vasculature may be deemed inaccessible (e.g., clots located beyond the left or right pulmonary artery, clots within the upper, middle, or lower lobes, etc.). Additionally, tortuosity of the pulmonary vasculature (or the 3D model of the pulmonary vasculature) may be used to determine if a specified clot can be accessed with the medical device(s). In another example, the size and / or location of the clots, such as the size / volume / location parameters estimated by the system, can be used to determine whether or not a clot can be accessed.

[0129] If the selected clot cannot be accessed, then in block 710 the system 400 can provide a recommendation or assessment that the clot should not be treated. If the selected clot can be accessed, then in block 712 the system 400 can mark or flag the selected clot as being accessible with the thrombectomy system. In some examples, the clots that are recommended and not recommended for treatment can be identified or flagged on the display or GUI, such as with color coding (e.g., green for can be accessed, red for cannot be accessed) or can be accompanied with indication or text to the user alongside the 3D model that the clot can or cannot be accessed.

[0130] The method 700 proceeds to block 718 where the system 400 can provide or generate a quantitative assessment of the accessible clots. This can include, for example, generating a treatment plan or identifying or recommending specific clots to target for removal. In some embodiments, the system can generate a Miller Score, a Modified Miller Score, or some other quantitative assessment of the accessible clots. The assessment can include a recommendation of clots to remove, and can further include information for the user or physician on the expected outcomes or patient recovery / improvement if the targeted clots are removed. The assessment can be based on, but not limited to, clot location, clot size or volume, and clot type or density. Additionally, the system can provide a Miller Score if the targeted clots are removed. The assessment or recommendation may be output at block 716, such as on a display or GUI of the system.- 27 - SG Docket No.: 10844-737.674

[0131] In any of the embodiments described herein, including any embodiment that shows the catheter within a 3D model or within medical imaging of the pulmonary arteries (e.g., CT, MRI, fluoro, etc.), the location, position, and orientation of the catheter can be determined or located within the model or imaging using the system described above, particularly the catheter location system including the 3D sensors described in FIG. IB and the external device shown in FIG. 3 A.

[0132] FIGS. 9A-9D illustrate a graphical user interface that can display medical imaging and 3D generated models of patient anatomy to a user. The medical imaging and / or 3D generated models may be applicable, for example, to help guide or navigate a user during a thrombectomy procedure, and may include 3D models of relevant patient anatomy including the pulmonary vasculature, clots within the pulmonary vasculature, and optionally a 3D representation of a medical device such as a thrombectomy catheter within the models.

[0133] FIG. 9 A shows a user interface 901 that includes screen 903 and 905 that allows the user to view a multitude of imaging and or 3D model views of the anatomy and clots before, during, or after a procedure. View controls 902 can be used to select the appropriate view for each screen 903, 905. For example, a user may interact with option menus within view controls to select a desired imaging modality (e.g., CT, X-Ray, MRI, ultrasound, etc.) and specific features of the 3D model for view in one or both of the screens 903 and 905. Generally, CT or MRI imaging data is obtained prior to a procedure, and fluoroscopy or X-ray imaging is obtained during a procedure. As shown, the user can select / deselect which aspects of the 3D model are to be displayed. In this example, the user has selected both the artery / vasculature view, as well as segmented clots. The model is shown overlaid in both screens 903 and 905 over the chosen imaging. In this example, CT imaging is selectable in the screen 903, and X-ray imaging is selectable in screen 905. However, the user in this example has deselected or hidden the CT view in screen 903, so only the artery and clot models are shown. In screen 905, an outline of the artery model is shown overlaid upon the X-ray imaging. The user can adjust many features of these views, including the transparency of the imaging, the transparency of the model, and the size, angle view, and rotation of the imaging and / or the model, as shown.

[0134] At the bottom of the user interface, the system can include screens 907, 909, 911, and 913 which can provide additional information pertaining to the patient, the 3D model, or the treatment. In some examples, this information can pertain to a specific classification of the segmented clot(s) in the 3D model. As shown in FIG. 9 A, screen 903 identifies individual clots 1, 2, 3, 4, which are also shown within the model in screen 903, and screen 909 shows the clots’ estimated or calculated contribution to obstruction (e.g., obstruction - 28 - SG Docket No.: 10844-737.674index contribution). Screen 911 provides estimates / calculations of the clot volume and screen 913 provides an assessment of the severity of the clot (e.g., High, Medium, Low, etc.). This severity assessment may, for example, indicate to a user which clots to target, or alternatively, may indicate which clots should be removed to improve patient health.

[0135] While the type of classification shown in screen 909 of FIG. 9A is an obstruction based classification of the clots and their effect on the patient, it should be understood that other ways of classifying the clots can be determined by the system and presented in screen 909, or any of the other screens on user interface 901. For example, clots can be classified by volume, chronicity, or obstruction (e.g., ratio of clot to lumen), and effect on flow, as discussed below.

[0136] FIG. 9B is an example of a screen 915 that shows the pulmonary vasculature and clot models in combination with obstruction based classification. This screen 915 can be shown, for example, in the UI 901 of FIG. 9 A. In this example, the system can employ obstruction-based classification to classify the clots and their contribution to obstruction of local flow within the anatomy. As shown in this view, the pulmonary vasculature and clot models are displayed to the user. In this example, the clots are labeled 1, 2, 3, 4, and 5. On the side of screen 915, the system can identify individual clots and their estimated or calculated contribution to obstruction (e.g., obstruction index contribution). The system can also provide estimates / calculations of the clot volume for each clot of interest. The labeling / number or color coding of clots in both the model and the classification chart or index allows the user to quickly and efficiently identify the clots within the 3D pulmonary vasculature model and identify the volume and or obstruction index of a given clot. Other graphical techniques can be used for identifying the individual clots besides numbering or color. For example, shading or outlining techniques may be used to identify each clot and its volume / obstruction index.

[0137] In some embodiments, the system can determine the obstruction index by extracting the centerline of the target vessel (e.g., the pulmonary artery) at the location of a specific emboli or clot. The system can determine or calculate the size of the lumen at this location with the clot or emboli removed (e.g., a clean artery score), and then add the clot or emboli back into the model and calculate the obstruction index (e.g., the amount of obstruction the clot is responsible for within the lumen or vessel). The algorithm can employ a union function between the artery and the emboli to determine the clean artery. For a given cross-section in the images or the model, pixels or voxels can be tagged as either pulmonary artery (PA) or pulmonary embolism (PE). This initial tagging results in some portions of the PA being marked as PE, so the initial data set does not provide a true “clean” PA. Therefore,- 29 - SG Docket No.: 10844-737.674these two datasets can be unioned to mark all those pixels as “clean” PA. Generally in this specification, the term “pixel” or “voxel” refers to any representation of a the location of unit of 2D or 3D space, which may additionally capture additional characteristics of that location.

[0138] FIG. 9C is another example of a screen 917 that shows the pulmonary vasculature and clot models in combination with clot density or chronicity based classification. In this example, the system can employ clot density -based classification to classify the chronicity of the clots (e.g., how old the clot is). The clot density-based classification uses the model and / or other imaging or patient data to determine how dense the clot is, which correlates to how old the clot is. As shown in this screen, the pulmonary vasculature and clot models are displayed to the user, labeled in this example as clots 1-5. On the side of the screen, the system can identify individual clots and their estimated or chronicity or density. The system may present additional information relating to the density, or attempt to classify the chronicity or density into categories (e.g., fresh clot, mixed fresh / chronic clot, chronic clot) as shown.

[0139] In some embodiments, the system can use a Houndsfield scale (HU) to characterize the clot chronicity. A sample Houndsfield scale is shown, and can optionally be displayed to the user. For example, acute clots may have a HU between 60-90, mixed clot can have a HU between 90-150, and chronic clot can have a HU above 150. In some embodiments, the system can only assess the chronicity of clots that are of a certain size (e.g., larger than X cc’s in size). This could be, for example, clots larger than lee, 2cc, 3cc, etc. The system can also provide estimates / calculations of the clot volume and or clot chronicity, either with a categorization as discussed above or an objective Hounsfield scale measurement or estimate. The clots within the model can be individually numbered as shown, or colored to correspond to the classification on the right. The numbering or color coding allows the user to quickly and efficiently identify the clot within the pulmonary vasculature model and identify the chronicity / density of a given clot. Other graphical techniques can be used for identifying the individual clots besides color. For example, shading or outlining techniques may be used to identify each clot and its chronicity.

[0140] Referring to FIG. 9D, the system may evaluate CT imaging data to determine the chronicity of clot for each clot within the 3D model. The system can evaluate voxels in the CT imaging data corresponding to a clot, including a distribution of HU values for each voxel within a specific clot to determine the chronicity of the clot. The intensity of pixels in the CT imaging data corresponds to the density of that pixel. The system can evaluate the distribution of HU / density values and then classify the clot based on the evaluation. In some examples, the evaluation can include assessing the aggregation or uniformity of the pixels for - 30 - SG Docket No.: 10844-737.674a given clot. This evaluation can identify the age of the clot, including mixed clots that have portions of old clot surrounded by or attached to portions of new clot. The system can also evaluate pixels in CT imaging to identify how voxels or pixels of the clot material blends into or forms into pixels of the vessel wall. In some embodiments, the system incorporates or utilizes trained machine learning algorithms or modules to review and evaluate the CT imaging data to classify the chronicity of the clots based on the above parameters.

[0141] The system may evaluate CT imaging data to determine the chronicity of an embolus for each clot within the 3D model. The chronicity of the embolus may refer to how long a clot has been at a particular location within the pulmonary artery. The system can evaluate voxels of the clot or clot boundaries in the CT imaging data corresponding to one or more results of an embolus or settled materials of a thrombus within a blood vessel. The one or more results of the embolus or a settled materials of the thrombus within blood vessel may include vascular stenosis, retraction with total obstruction, retraction with partial obstruction, recanalization, or residual fibrous cords (e.g., web or bands). The system can evaluate voxels in the CT imaging data corresponding to the one or more of a clot, an embolus, or a settled materials of the thrombus, including the size, shape, density, vessel wall integration, and other parameters to determine chronicity of the embolus. The system can use image analysis of the voxels of the clot, embolus, thrombus, and the like to determine its respective size, shape, density, and vessel wall integration. For example, image analysis performed on one or more voxels in the CT imaging data may indicate a chronicity of an embolus or settled materials of a thrombus is old if the clot integration with the vessel wall is smooth or the chronicity of the embolus or the settled materials of the thrombus is new if the clot integration with the vessel wall includes hard angular edges. In some embodiments, the system incorporates or utilizes trained machine learning algorithms or modules to review and evaluate the CT imaging data to classify the chronicity of the embolus or settled materials of the thrombus. The chronicity of the embolus or settled materials of the thrombus may include one or more levels of chronicity, such as new, recent, old, and the like. There may be several different levels of chronicity used by the system to classify the chronicity of the embolus or settled materials of the thrombus. The system may also generate an estimation and assign a risk score to obstruction using the classification of the chronicity of the embolus. The system may calculate an obstruction index of a clot using the classification of the chronicity of the embolus corresponding to the clot. The system may assess the clot environment by analyzing the vascular stenosis and recanalization using the chronicity of the embolus and / or image analysis of the clot corresponding to the embolus. The assessment of the clot environment may be presented with the 3D model to show the stenosis and recanalization. The assessment - 31 - SG Docket No.: 10844-737.674of the clot environment may also be presented to identify a degree of obstruction caused by one or more of the stenosis, recanalization, and other parameters of the chronicity of the embolus.

[0142] The system may generate an assessment based on clot position within a blood vessel. The system may evaluate CT imaging data to generate the assessment on clot position within the blood vessel. The assessment on clot position within the blood vessel may also include an angle at which the clot is positioned within the blood vessel. For example the angle at which the clot is positioned may be the angles which the clot forms with respect to the vessel wall.

[0143] The system may identify potential pulmonary hypertension or pulmonary infarction of a patient. The system may evaluate CT imaging data to identify the potential pulmonary hypertension or pulmonary infarction of the patient. The system may use the CT imaging data to compare a size and shape of the pulmonary arteries to an average shape of the arteries within the CT imaging data and / or average shape of the arteries from an accumulation of patient data to identify potential pulmonary hypertension or pulmonary infarction of a patient. In some embodiments, the assessment may include one or more of identifying an increase in diameter of pulmonary arteries or the tortuosity of the pulmonary arteries. In some embodiments, the system incorporates or utilizes trained machine learning algorithms or modules to review and evaluate the CT imaging data to identify potential pulmonary hypertension or pulmonary infarction of the patient.

[0144] The system may identify calcification or assess a degree of calcification. The system may evaluate CT imaging data to identify calcification or assess a degree of calcification of the clot. The system can evaluate voxels in the CT imaging data corresponding to a clot, including a distribution of HU values for each voxel within a specific clot to determine the calcification or assess a degree of calcification of the clot. The system may generate a scale of calcification of a clot that ranges between 100 HU to 1100 HU. The scale of calcification of the clot may include a plurality of different levels, such as partial, severe, and the like. In some embodiments, the system incorporates or utilizes trained machine learning algorithms or modules to review and evaluate the CT imaging or other imaging data to identify calcification or assess a degree of calcification of the clot.

[0145] The system may identify chronic thromboembolism within a patient. The system may evaluate CT imaging data to identify chronic thromboembolism within a patient. The system can evaluate voxels in the CT imaging data to identify air trapping that can occur within the pulmonary arteries. The system can evaluate voxels in the CT imaging data to identify air trapping in areas of lower attenuation of the CT imaging data. The system may - 32 - SG Docket No.: 10844-737.674evaluate the CT imaging data to identify one or more of a mosaic perfusion pattern and diminished size of vessels in a lung lobe in comparison to one or more other lung lobes to identify chronic thromboembolism within the patient. In some embodiments, the system incorporates or utilizes trained machine learning algorithms or modules to review and evaluate the CT imaging data to identify chronic thromboembolism within a patient.

[0146] The system may distinguish soft tissue from a clot within a 3D model. The system may evaluate CT imaging data to identify soft tissue to distinguish the soft tissue from the clot. In some embodiments, the system incorporates or utilizes trained machine learning algorithms or modules to review and evaluate the CT imaging data to distinguish soft tissue from a clot within a 3D model.

[0147] After a procedure, pre and post CT images can be analyzed to evaluate the effectiveness of the procedure. In one embodiment, this process can include:

[0148] 1. The segmentation model outputs labels for the pulmonary artery and clots for both pre and post CT scans.

[0149] 2 For each CT scan, the artery and clot labels are merged into a single label, which can be referred to as the "clean artery label.

[0150] 3. The registration machine learning model takes the clean artery labels from both CT scans and outputs a 3D deformation field, that registers the post- and pre-clean artery labels.

[0151] 4. The system can then apply this deformation field to the clot labels from the post-CT scan, effectively positioning the post-CT clots into the pre-CT artery.

[0152] FIGS. 10A-10B illustrate embodiments of a 3D pulmonary artery model PV with segmented clots C overlaid within the model, which can be used to map or plan out a procedure. In the model presented in FIG. 10A, a user can see that the clots C are located predominantly on the right and left sides of the pulmonary arteries. In the model presented in FIG. 10B, however, a bilateral saddle clot is shown with a string-like saddle that connects clots on both sides of the pulmonary arteries. The physician / user can manipulate, rotate, zoom, and review the model to determine or plan out how to treat the clots. This can include, for example, determining the order and approach of treating the clot sections including the saddle and clot portions on both sides of the pulmonary arteries.

[0153] FIGS. 11 A-l 10 describe systems and methods for estimating perfusion during a thrombectomy procedure to evaluate the efficacy of the thrombectomy procedure in realtime. This embodiment can include any of the systems and methods described above, particularly the catheter 102 and medical device 108 of FIGS. 1A-1B. The system and methods can include using fluoroscopy imaging in combination with flow modeling and flow - 33 - SG Docket No.: 10844-737.674estimation to estimate perfusion of the lungs resulting from a bolus or puff of contrast and identify changes in perfusion after one or more clots are removed to assist the user or physician in identifying the efficacy of a given thrombectomy procedure.

[0154] In FIG. 11 A, a catheter 1102 fitted with a dilator 1150 can be advanced over a guidewire 1152 towards a target thrombus. With the catheter in this position, a puff or bolus of contrast can be delivered from the catheter into the pulmonary artery, towards the target thrombus. The contrast can be delivered directly from the catheter 1082. In some examples, the catheter is delivered through one or more ports or openings between the catheter and the dilator.

[0155] FIG. 1 IB shows three different views of the catheter of FIG. 11 A. View 1101 is a fluoroscopy view of the lungs and catheter, obtained with fluoroscopy imaging to evaluate contrast within the pulmonary artery and lungs. View 1103 is a segmented perfusion estimation that is modeled using the 3D model of the pulmonary arteries, the segmented models of the clots, and the fluoroscopy imaging from view 1103. View 1105 is an outline view of the 3D model of the pulmonary artery and downstream branches of the pulmonary vasculature, optionally including views or models of the thrombectomy system which may include the guidewire, catheter, etc.

[0156] FIG. 11C shows the same three views as in FIG. 1 IB, after a specified passage of time. The passage of time between delivery of contrast and the views of FIG. 11C allows for the contrast to perfuse the lungs, as shown in view 1101. The perfusion modeling or estimation in view 1103 now includes one or more perfusion estimates. The perfusion estimates can be directed to specific locations, sections, or partitions of the lungs. In the illustrated example, perfusion estimates are provided for three sections of the lungs (e.g., upper, mid, lower). In the example of FIG. 11C, the upper section of both lungs have perfusion estimates of 10%, and the mid and lower sections see 20% perfusion. The perfusion estimates can be, for example, an estimate of the percentage or volume of each lung section that is perfused. As is known, clots or emboli within the pulmonary arteries can affect perfusion of the lungs, so more clots, or more obstructive clots in the pulmonary arteries will reduce or prevent perfusion of the lungs, or perfusion of segments or sections of the lungs. As will be described below, a sequence of injecting contrast, evaluating / estimating perfusion, removing one or more clots, and repeating the contrast injection and perfusion estimation steps can be used to evaluate a thrombectomy procedure in real-time.

[0157] In FIG. 1 ID, after the initial perfusion estimate, the dilator and guidewire can be removed from catheter 1102, and the catheter can be advanced into or towards the clot.Aspiration can be activated to aspirate or remove some or all of the clot at that location. In - 34 - SG Docket No.: 10844-737.674some examples, the catheter can further employ the use of intersecting jets to assist in the aspiration by cutting or fragmenting the thrombus, as discussed above. View 1107 shows a cross-sectional view of the catheter 1102 with four intersecting jets within the catheter to cut or fragment the thrombus.

[0158] FIG. 1 IE shows the pulmonary arteries after the first clot or portion of clot has been removed. At this point in the procedure, the user may repeat the contrast delivery and perfusion estimate procedure described above to assess the effectiveness of the initial clot removal. In FIG. 1 IF, a second puff or bolus of contrast is delivered, after the clot removal of FIGS. 1 ID-1 IE. In FIG. 11G, the perfusion of the contrast is modeled / estimate, as shown in view 1103, and the perfusion estimate(s) for one or more sections of the lungs is provided or displayed to the user. In this example, it can be seen that perfusion has increased after removing some of the clot in the pulmonary arteries, in this instance perfusion increasing to 20% in the upper sections and 30% in the mid and lower sections.

[0159] At this point in the procedure, the physician or user may determine that clot within the main branch of the pulmonary artery has been removed, but since perfusion is still limited, unaffected, or restricted in both lungs, clots may still exist in both the left and right branches of the pulmonary artery. The process described above can be repeated to further remove these clots and re-assess perfusion to determine treatment efficacy. In FIG. 11H, the catheter 1102 is advanced into the left pulmonary artery towards remaining clot within that branch. In FIGS. 1 II- 11 J, contrast is delivered and perfusion is calculated / estimated as previously described. In FIG. 1 IK, clot is removed from the left branch. In FIGS. 1 IL-1 IM, the process is repeated for the right branch. In FIG. 1 IM, the new perfusion estimate, prior to removing the clot in the right branch, but after removing the clot in the left branch, shows the new perfusion estimate for the left lung to be 90% / 100% / 100% in the upper / mid / lower sections of the lung, which is a large improvement over the prior estimate in the left lung. The perfusion estimates for the right lung remain the same, since the clot in the right branch has not been removed yet.

[0160] In FIG. 1 IN, since the remaining clot in the right branch is in a narrower or deeper section of the right branch, the physician may opt to introduce the medical device 1108 to access that clot. As discussed above, the medical device can be a smaller bore aspiration catheter that can be delivered to a target location via the lumen of the catheter 1102. View 1109 shows a cross-section of medical device 1108, which can also include aspiration and jetting as discussed above. The medical device 1108 can be used to remove clots in these deeper / harder to reach locations within the anatomy.- 35 - SG Docket No.: 10844-737.674

[0161] Referring to FIG. 110, with the clots removed from the main branch, the left branch, and the right branch of the pulmonary artery, another bolus of contrast can be delivered and the perfusion can be calculated / estimated as shown. In this example, perfusion in the right lung has increased to 90% / 100% / 90%, while perfusion in the left lung is at 90% / 90% / 100%. These perfusion estimates can be used by the physician to determine if more clot needs to be removed, or if the procedure can be terminated.

[0162] Flow within the vasculature / arteries can be computed or estimated with a variety of approaches. In one embodiment, flow is estimated as follows:

[0163] Data Input & Preprocessing: The code begins by loading three files in the system: the 3D volume of the pulmonary artery (NIFTI), its geometric centerline (OBJ), and the associated metadata (XML) which contains information like voxel size for converting pixel coordinates to physical dimensions.

[0164] Graph Construction & Simplification: The centerline's vertices and edges from the OBJ file are used to construct a mathematical graph. To correct for potential artifacts from the centerline extraction process, nearby vertices are merged (VertexContract), resulting in a cleaner, more representative graph of the artery's branching structure.

[0165] Geometric Property Extraction: The NIFTI image is converted into a 3D mesh. A coordinate transformation is applied using the metadata to ensure the 3D mesh and the centerline graph are perfectly aligned in the same physical space.

[0166] Link Geometric & Graph Data: This is a step where the two data sources are combined. The code iterates through each edge of the simplified graph and "probes" the 3D artery mesh to determine length and radius of vessel segment. Length is the simple Euclidean distance between the edge's start and end nodes. Radius is estimated by measuring the distance from centerline segment to the nearest surface of the 3D mesh.

[0167] Flow Simulation (effectiveK function): This is the computational core where fluid dynamics principles are applied to the graph.

[0168] It first calculates the hydraulic conductance for each vessel, a value representing how easily fluid can flow through it. This is derived from the Hagen-Poiseuille equation, where conductance is proportional to the radius to the fourth power (rA4) and inversely proportional to the length (L).

[0169] A weighted graph Laplacian matrix is constructed. This matrix represents the entire connected system, with each entry encoding information about the conductivity between nodes. Boundary conditions are set by assigning a high pressure (e.g., P=l) at the graph's main inlet and a low pressure (P=0) at all the outlets.- 36 - SG Docket No.: 10844-737.674

[0170] A linear system of equations (Lp=b) is solved to find the unknown pressure (p) at every internal junction in the network.

[0171] With the pressure at every node known, the pressure gradient (\Delta P) across each vessel is calculated. Finally, the volumetric flow rate (Q) is computed for each vessel using the fluid dynamics equivalent of Ohm's Law: Flow = Conductance \times Pressure\Gradient.

[0172] Additional approaches for measuring flow are provided below, and can include:

[0173] 1) Measuring contrast clearance rates at one or more points along the pulmonary vasculature. If contrast introduction is at the base of a divide, then the system can integrate the clearance rates at points along the branches to estimate the flow at the base injection point. This approach could use dynamic management of contrast injection timing given cardiac cycles and related flow periodicity. While measuring contrast washout curves where the curves are well sampled, timing the injection puff of contrast to sync with the cardiac cycle and the limited fluoro sampling rate would make this process more efficient and effective. Having clear understanding of the dimensions of the vasculature involved is important to make this approach accurate.

[0174] 2) Measuring time from pulmonary artery injection to aortic clearance. This technique would measure initiation, peak, and clearance times for contrast flow first at the point of injection (root of the pulmonary artery). The same view will provide a view of the aorta. A large bolus injection is probably necessary given it has to fully dilute and then show up again in the aortic flow. The time length of the injection vs. the time length of the bolus passing through the aorta gives an indication of total flow rate through the lungs. In some aspects, the system can automatically pick the injection point and the position of the aorta from the fluoro.

[0175] 3) Measuring pulmonary opacification at the outer boundaries of the lungs with a bolus contrast injection at the pulmonary artery root (perfusion imaging). In this embodiment, the system can inject a bolus of contrast (enough to at least do a detailed venogram) with the lungs in AP view and automatically segment the captured image to several regions of interest that are anatomically relevant. The system could use standard die-dilution / perfusion measurement approaches to understand relative flow into each section and a cumulative wash-out time that is indicative of cardiac output / pulmonary vascular resistance. The system could then take two successive frames of fluoro during a timed contrast injection and calculate volumetric flow based on displacement of contrast front. This approach potentially allows the system to measure flow with small puffs of contrast. The contrast could be injected by the fluid source / fluid pump that is otherwise used for jetting, provided - 37 - SG Docket No.: 10844-737.674that there is a separate switched source of contrast of known concentration. Alternatively, the fluid source could include contrast to be delivered by the jets.

[0176] The algorithm would operate as follows: To use a snapshot of contrast concentration at two consecutive points in time (ideally separated by the minimum possible increment) to estimate volumetric flow rate. Using the snapshots and a 3D model of the blood vessel, create al. Each point on this centerline corresponds to a 2D cross-section of the vessel; average the concentration of contrast (given by its color intensity) over this crosssection for both snapshots and call this function C(x,tn), as it is a function of one’s position along the centerline and time. We now have an initial condition for C(x,ti) as well as (from the second snapshot) a final C(x,tf). The ID advection-diffusion equation (ADE) dictates the evolution of C(x,t). The system is missing a velocity (scalar in this case, as it is approximating it to be one-dimensional along the centerline) and a diffusion coefficient (which is not simply molecular diffusion, but an effective diffusion determined both by molecular diffusion and potentially turbulent advection). The system can use the solution to the ADE with these unknown parameters to determine the final C(x,t). This can then minimize |C_final(x,t) - C_initial(x,t)| by varying the velocity v(x,t) and the diffusion coefficient D. The velocity and coefficient that satisfy this optimization problem are used to approximate the true fields. Next, the algorithm can multiply v(x,t) by the area at each point along the centerline A(x) to obtain an approximate expression for volumetric flow rate.

[0177] It should be appreciated that since the catheter / thrombectomy systems described herein can include pressure sensors, the system can continuously monitor pressure within the pulmonary artery during a procedure. This pressure information in combination with the flow modeling / estimate described above can be used by the system to also estimate cardiac output.

[0178] A. Overview of Imaging System

[0179] In one or more embodiments, referring to FIG. 12, a clot management environment 1200 may be used to provide imaging related to a thrombectomy procedure, a balloon pulmonary angioplasty (BP A) procedure, and other procedures managing clot in a patient 1202. The clot management environment 1200 may perform one or more functions to related to a thrombectomy procedure, a balloon pulmonary angioplasty (BP A) procedure, and other procedures managing clot in a patient 1202. As an example, the clot management environment 1200 may be used to remove clot from a patient 1202 through a catheter system 1204 coupled to a console system 1208 as described herein. The clot management environment 1200 may include a patient 1202, a catheter system 1204, a console system 1208, and a fluoroscopy system 1210. During a thrombectomy procedure, a physician may - 38 - SG Docket No.: 10844-737.674introduce a catheter system 1204 into a patient 1202 through, for instance, a femoral artery. A console system 1208 may be coupled to a catheter system 1204 to receive one or more control signals for one or more components of the console system 1208. The console system 1208 may receive data corresponding to one or more of pressure measurements, flow measurements, and the like. The console system 1208 may be coupled to a catheter system 1204 in such a way to generate a vacuum pressure in one or more components of the catheter system 1204. As an example, the console system 1208 may generate a vacuum pressure for the jetting catheter 1220 to aspirate one or more clots in the patient 1202. The console system 1208 may be coupled to the fluoroscopy system 1210 to receive one or more of data and / or images generated by the fluoroscopy system 1210 and process the data and / or images to present images to a physician. The fluoroscopy system 1210, such as a C-arm, may be placed directly over the patient 1202 to capture X-ray images of the target area (e.g., lungs of the patient). The X-ray images may be focused on the position of the catheter system 1204 within the patient 1202. While this disclosure describes a clot management environment 1200 with a number of components arranged in a particular way, this disclosure contemplates a clot management environment 1200 with any number of components arranged in any suitable way. For instance, the clot management environment 1200 may comprise a wireless network connection between one or more components of the clot management environment 1200.

[0180] In one or more embodiments, the catheter system 1204 may include one or more of one or more control knobs 1212, reciprocating mechanism 1214, an aspiration control 1216, a contrast control 1218, jetting catheter 1220, outer sheath 1222, one or more sensors 1224, and a jetting control 1226. While this disclosure describes the catheter system 1204 as comprising one or more components in a particular arrangement, this disclosure contemplates a catheter system with any number of components in any suitable arrangement. As an example, the reciprocating mechanism 1214 may be a subcomponent of the outer sheath 1222. As another example, the catheter system 1214 may include additional catheters. The control knobs 1212 may be embodied as one or more knobs to steer one or more catheters of the catheter system 1204. For instance, the one or more control knobs 1212 may steer a delivery catheter (not shown) or a jetting catheter 1220. The control knobs 1212 may be located on one or more of the catheters of the catheter system 1204. As an example, the control knobs 1212 may be located on jetting catheter 1220. The reciprocating mechanism 1214 may enable a reciprocation process where an outer sheath 1222 encompassing a jetting catheter 1220 may be translated along the jetting catheter 1220 by a specific degree of motion or otherwise lock the outer sheath 1222 to a particular location of the jetting catheter 1220. As an example, the reciprocating mechanism 1214 may enable the outer sheath 1222 to move - 39 - SG Docket No.: 10844-737.674coaxially to the jetting catheter 1220 by up to 3 inches in one or more directions. The reciprocating mechanism 1214 may be coupled to one or more components of the catheter system 1204. As an example, the reciprocating mechanism 1214 may be coupled to the outer sheath 1222.

[0181] In one or more embodiments, the aspiration control 1216 may be used to initiate an aspiration through one or more catheters of the catheter system 1204. As an example, aspiration control 1216 may send a signal to console system 1208 to initiate aspiration through jetting catheter 1220 by activating a pump (e.g., pump 1250) to create a vacuum pressure. The aspiration control 1216 may be embodied as any suitable user-actuatable input mechanism, including a physical or virtual button or other control element, configured to receive user input and generate a corresponding control signal for the system. The aspiration control 1216 may be integrated into a housing, bezel, surface, or user interface panel of one or more components of the catheter system 1204, or may be remotely located, detachable, or wirelessly coupled to one or more components of catheter system 1204.

[0182] In one or more embodiments, the contrast control 1218 may be used to initiate injection of contrast into patient 1202 through one or more catheters of the catheter system 1204. As an example, contrast control 1218 may send a signal to console system 1208 to initiate injection of contrast through jetting catheter 1220 by activating a pump (e.g., pump 1250) to push a saline and contrast mixture into the patient 1202. The contrast control 1218 may initiate a mixing process that mixes a saline solution with contrast to be injected into the patient 1202. The contrast control 1218 may be embodied as any suitable user-actuatable input mechanism, including a physical or virtual button or other control element, configured to receive user input and generate a corresponding control signal for the system. The contrast control 1218 may be integrated into a housing, bezel, surface, or user interface panel of one or more components of the catheter system 1204, or may be remotely located, detachable, or wirelessly coupled to one or more components of catheter system 1204.

[0183] In one or more embodiments, the jetting catheter 1220 may be a catheter of the catheter system 1204 used to aspirate one or more clots in the patient 1202. The jetting catheter 1220 may comprise one or more high-velocity jets located at a distal tip of the jetting catheter 1220 to macerate one or more clots through expelling a saline solution towards the one or more clots. The one or more high-velocity jets may be directed to a central focal point to intersect the one or more streams of the respective one or more high-velocity jets. As an example, for a jetting catheter 1220 that has a circular distal end, it may have four high-velocity jets may be positioned at the 12 o’clock, 3 o’clock, 6 o’clock, and 9 o’clock position of a clockface at the distal end of the jetting catheter 1220. The jetting catheter 1220 may be - 40 - SG Docket No.: 10844-737.674coupled to a pump 1250 of the console system 1208 to receive a vacuum pressure within an inner lumen of the jetting catheter 1220 that is generated by the pump 1250 and / or a pressure generated by the pump 1250 to push a solution through one or more jets of the jetting catheter 1220.

[0184] In one or more embodiments, the outer sheath 1222 may be larger in diameter than the jetting catheter 1220 to envelop the jetting catheter 1220. The outer sheath 1222 may provide a mechanism to readjust clot engaged with the distal tip of the jetting catheter 1220 through the reciprocating mechanism 1214 by extending out past the distal tip of the jetting catheter 1220 and contacting engaged clot to reposition the engaged clot to another position to be aspirated by the jetting catheter 1220.

[0185] In one or more embodiments, the one or more sensors 1224 may measure one or more of a pressure at a point within a catheter or a flow at a point within a catheter of the catheter system 1204. As an example, the one or more sensors 1224 may measure a pressure at a distal tip of a jetting catheter 1220 of the catheter system 1220. The pressure measurement may indicate engagement of the catheter with a surface (e.g., clot or blood vessel). The one or more sensors 1224 may be embodied as one or more of a fiber optic pressure sensor, a flow sensor, or another sensor to measure pressure or flow within the catheter system 1220. The data from the one or more sensors 1224 may be sent to the console system 1208 to process and determine to perform one or more functions. As an example, once pressure measurements from one or more sensors 1224 indicate clot engagement, the console system may enable actuation of the one or more high-velocity jets of the jetting catheter 1220, where jetting of the clot with the jetting catheter 1220 may previously be disabled until clot engagement has been determined.

[0186] In one or more embodiments, the jetting control 1226 may be used to initiate jetting of one or more clots through one or more jetting ports of one or more catheters of the catheter system 1204. As an examplejetting control 1226 may send a signal to console system 1208 to initiate jetting through jetting catheter 1220 by activating a pump (e.g., pump 1250) to create a pressure to expel one or more of a saline solution, contrast solution, and the like through one or more jetting ports of the jetting catheter 1220. The jetting control 1226 may be embodied as any suitable user-actuatable input mechanism, including a physical or virtual button or other control element, configured to receive user input and generate a corresponding control signal for the system. The jetting control 1226 may be integrated into a housing, bezel, surface, or user interface panel of one or more components of the catheter system 1204, or may be remotely located, detachable, or wirelessly coupled to one or more components of catheter system 1204.- 41 - SG Docket No.: 10844-737.674

[0187] In one or more embodiments, the console system 1208 may perform one or more functions for a clot management environment 1200 for a thrombectomy procedure, a BP A procedure, and / or other procedures to manage clots within a patient 1202. The console system 1208 may manage a pump 1250 to control a vacuum pressure within a catheter of the catheter system 1204 to perform aspiration of one or more clots. The console system 1208 may manage a pump 1250 to control jetting of a jetting catheter 1220 of the catheter system 1204 to expel a high-velocity saline solution towards one or more clots. The console system 1208 may manage a pump 1250 to control contrast injection into the patient 1202 through a reservoir 1256. The console system 1208 may perform processing of data (e.g., one or more images from fluoroscopy system 1210) to generate and update a 3D model of pulmonary vasculature of a patient 1202. The console system 1208 may present a current view of the patient 1202 anatomy received from the fluoroscopy system 1210. The console system 1208 may use one or more algorithms, machine-learning models, or a combination of the two to perform the one or more functions for the clot management environment 1200. In one or more embodiments, console system 1208 may include one or more of a processor 1240, a memory 1242, a storage 1244, a communication unit 1246, a display 1248, a pump 1250, a pump controller 1252, a contrast controller 1254, a reservoir 1256, and a user interface 1258. While this disclosure describes a console system 1208 as performing one or more functions for a clot management environment 1200, this disclosure contemplates a console system 1208 performing any suitable function for a clot management environment 1200.

[0188] In one or more embodiments, the processor 1240 may execute instructions stored in memory 1242 to perform one or more functions of the console system 1208. The processor 1240 may be embodied as one or more processors 1240. The processor 1240 may be embodied as one or more of, but is not limited to general -purpose CPUs, microcontrollers, graphics processing units (GPUs), digital signal processors (DSPs), neural processing units (NPUs), field-programmable gate arrays (FPGAs), or application-specific integrated circuits (ASICs). The console system 1208 may store data collected from one or more sources, one or more models of patient anatomy, one or more generated outputs, and one or more models of devices (e.g., catheters of the catheter system 1204) in storage 1244. The communication unit 1246 may receive and transmit data to one or more other components of the clot management environment 1200, such as the fluoroscopy system 1210. The processor 1240 may process data received by the fluoroscopy system 1210 and / or other imaging systems to generate one or more outputs as described herein.

[0189] In one or more embodiment, the user interface 1258 may receive user input from one or more operators. As an example, the user interface 1258 may receive a user input to - 42 - SG Docket No.: 10844-737.674view a 3D model of a pulmonary vasculature of a patient 1202. The user interface 1258 may be embodied as one or more of a graphical, tactile, auditory, or multimodal interfaces. As an example, the user interface 1258 may be coupled to the display 1248 to be a touch-screen display. The user input may generate one or more outputs to be displayed on a display 1248 of the console system 1208. The display 1248 may display one or more outputs of the clot management environment 1200. The user interface 1258 may comprise one or more user-actuatable input mechanisms corresponding to one or more functions that may be performed by the clot management environment 1200. The console system 1208 may display one or more images from the fluoroscopy system 1210 and one or more generated outputs simultaneously. As an example, the console system 1208 may present one or more fluoroscopy images from the fluoroscopy system 1210 and a 3D model overlaid onto one or more fluoroscopy images on the display 1248. The display 1248 may display one or more views of patient anatomy of the patient 1202.

[0190] In one or more embodiments, the pump 1250 may be used to generate a vacuum pressure to aspirate one or more clots through the catheter system 1204 and generate pressure to expel one or more of a saline solution, a contrast solution, and the like from high-velocity jet ports. While this disclosure describes console system 1208 as comprising one pump 1250, this disclosure contemplates a console system 1208 as comprising any number of pumps. As an example, the console system 1208 may have one pump to generate a vacuum pressure to aspirate one or more clots and another pump to generate pressure to expel one or more of a saline solution, a contrast solution, and the like from high-velocity jet ports. The pump 1250 may be coupled to the reservoir 1256 to receive one or more of aspirated clot, saline solution, and blood. The pump 1250 may be coupled to a reservoir 1256 that contains one or more of a saline solution, a contrast solution, and the like to be expelled from high-velocity jet ports.

[0191] In one or more embodiments, the pump controller 1252 may be used to control actuation of a pump 1250 to either generate a vacuum pressure to aspirate one or more clots through a catheter or generate pressure to expel one or more of a saline solution, a contrast solution, and the like from high-velocity jet ports. The pump controller 1252 may be embodied as one or more of a one or more of, but is not limited to, general-purpose CPUs, microcontrollers, GPUs, DSPs, NPUs, FPGAs, or ASICs. The pump controller 1252 can receive a signal from one or more components of the catheter system 1204 to actuate the pump 1250 to either generate a vacuum pressure to aspirate one or more clots through a catheter or generate pressure to expel a solution from high-velocity jet ports. As an example, the aspiration control 1216 may send a signal to aspirate one or more clots to pump controller- 43 - SG Docket No.: 10844-737.6741252, which actuates the pump 1250 to generate a vacuum pressure to aspirate one or more clots.

[0192] In one or more embodiments, the contrast controller 1254 may be used to control injection of a contrast mixture into a patient 1202. The contrast controller 1254 may be used to generate a contrast mixture to be injected into a patient 1202. The contrast mixture may contain a radiopaque contrast agent mixed with a saline solution. One or more algorithms and / or one or more machine-learning models may be used to determine a contrast mixture to inject into patient 1202 based at least on one or more parameters. As an example, an obstruction index for a blood vessel where a catheter injecting contrast is located, previous amount of contrast injected, size of blood vessel, a determined flow, and other possible parameters to determine amount of contrast to be injected into a patient 1202. The contrast controller 1254 may use reservoir 1256 to generate a contrast mixture to inject into a patient 1202. The console system 1208 may have access to one or more containers of contrast to mix with a container of a saline solution. The contrast controller 1254 may receive a signal from contrast control 1218 to inject contrast into patient 1202. The contrast controller 1254 may have a premixed contrast solution and / or generate a contrast mixture to be injected into the patient 1202. The contrast controller 1254 may send a signal to the pump 1250 to inject a contrast mixture through one or more catheters of the catheter system 1204. As an example, the contrast controller 1254 may send a signal to pump 1250 to generate pressure to push the contrast mixture into a lumen of a jetting catheter 1220 (e.g., jetting ports or aspiration lumen of jetting catheter). The contrast controller 1254 may send a signal to pump 1254 to follow a contrast mixture with a saline solution.

[0193] In one or more embodiments, the reservoir 1256 may be used to contain one or more solutions, liquids, and the like for the clot management environment 1200. As an example, the reservoir 1256 may be used to contain one or more of aspirated clot, blood, saline solution, contrast mixture, and the like. While this disclosure describes console system 1208 as having one reservoir 1256, this disclosure contemplates the console system 1208 as having any number of reservoirs needed for one or more of a thrombectomy procedure, BPA procedure, or any other clot management procedures.

[0194] In one or more embodiments, the fluoroscopy system 1210 may be used to perform an imaging process of patient 1202 anatomy using X-ray imaging. The fluoroscopy system 1210 may be embodied as a C-arm system or another kind of fluoroscopy system 1210 that may be used for one or more of a thrombectomy procedure, BPA procedure, or any other clot management procedures. In one or more embodiments, the fluoroscopy system 1210 may include one or more of an imaging module 1228, processor 1230, storage 1232,- 44 - SG Docket No.: 10844-737.674memory 1234, one or more sensors 1236, and an I / O interface 1238. The imaging module 1228 may include one or more components to perform X-ray imaging for the fluoroscopy system 1210. While not shown, the fluoroscopy system 1210 may include one or more components that interface the patient 1202 to inject contrast into the patient 1202 to perform the X-ray imaging. As an example, the fluoroscopy system 1210 may have an activatable control (e.g., button, knob, foot pedal, and the like) to control the injection of contrast into the patient 1202. In another example, the fluoroscopy system 1210 may have an activation module (not shown) to receive one or more control signals from a source (e.g., catheter system) that has an activatable user-input mechanism to control the injection of contrast into the patient 1202. The processor 1230 may execute instructions stored in memory 1234 to perform one or more functions of the fluoroscopy system 1210. The processor 1230 may be embodied as one or more processors 1230. The processor 1230 may be embodied as one or more of, but is not limited to, general-purpose CPUs, microcontrollers, graphics processing units (GPUs), digital signal processors (DSPs), neural processing units (NPUs), field-programmable gate arrays (FPGAs), or application-specific integrated circuits (ASICs). The fluoroscopy system 1210 may store data collected from one or more sources, such as the imaging module 1228 in storage 1232. The fluoroscopy system 1210 may send data to the console system 1208 to process. As an example, the fluoroscopy system 1210 may send imaging data collected by the imaging module 1228 to be presented on a display 1248 of the console system 1208. The console system 1208 may additionally process the imaging data received from the fluoroscopy system 1210 prior to presenting on the display 1248. While this disclosure describes a fluoroscopy system 1210 as performing one or more functions for presenting patient anatomy, this disclosure contemplates a fluoroscopy system 1210 as performing any suitable function for presenting patient anatomy.

[0195] In one or more embodiments, while this disclosure describes using a fluoroscopy system 1210 to capture one or more images of the catheter system 1204 during a procedure, this disclosure contemplates using other systems to capture one or more images of the catheter system 1204 during a procedure. As an example, an intravascular ultrasound system may be used to capture one or more ultrasound images that may be processed and used similarly to X-ray images captured by the fluoroscopy system 1210.

[0196] In one or more embodiments, while this disclose describes capturing one or more images (e.g., X-ray images or ultrasound images) of a catheter system 1204 interacting with patient 1202 anatomy, this disclosure contemplates capturing one or more images of another system, such as a catheter system that comprises pulmonary angiography catheter, balloon- 45 - SG Docket No.: 10844-737.674dilation catheters, balloon inflation device, and / or other devices for a BPA procedure in addition or in place of the catheter system 1204.

[0197] In one or more embodiments, while this disclosure describes the fluoroscopy system 1210 interfacing a console system 1208, this disclosure contemplates the console system 1208 wirelessly communicating with another device, such as a server-side system, which may handle one or more processing functions of the console system 1208. As an example, the console system 1208 may use a server-side system to handle processing for the processor 1240 and / or one or more other processing functions. For instance, the console system 1208 may send data to be processed by a machine-learning model executed on a server-side system.

[0198] In one or more embodiments, referring to FIG. 13, a processor environment 1300 may be executed on a processor 1240 of the console system 1208 to perform one or more functions of a clot management environment 1200 related to a thrombectomy procedure, a BPA procedure, and other procedures managing clot in a patient 1202. While this disclosure describes a console system 1208 comprising a processor 1240 to execute a processor environment 1300, this disclosure contemplates a console system comprising a plurality of processors to execute a processor environment 1300. While this disclosure describes a processor 1240 located on the console system 1208 executing the processor environment 1300, this disclosure contemplates a processor 1240 located separate from a console system 1208 (e.g., a server-side system) executing the processor environment 1300. While this disclosure describes a processor environment being executed on a processor 1240, this disclosure contemplates executing the processor environment on one or more components of the clot management environment 1200 and / or a combination of one or more components of the clot management environment 1200. In one or more embodiments, the processor 1240 of the processor environment 1300 may execute one or more software modules, including an image management module 1302, 3D model module 1304, a pump control module 1306, contrast injection module 1308, clot detection module 1310, real-time update module 1312, clot characterization module 1314, outcome assessment module 1316, procedure planning module 1318, and navigation assistance module 1320, wherein such modules may be executed individually, sequentially, concurrently, or in any combination thereof. While this disclosure describes the processor 1240 executing one or more software modules in a particular manner, this disclosure contemplates a processor executing one or more software modules in any suitable manner. While this disclosure describes a processor environment 1300 as comprising a number of components in a particular arrangement, this disclosure contemplates a processor environment 1300 as comprising any number of components in any - 46 - SG Docket No.: 10844-737.674suitable arrangement. As an example, the processor environment 1300 may comprise two separate processors that are executing one or more of the described software modules.

[0199] In one or more embodiments, the image management module 1302 may be configured to manage image data received by the console system 1208, including fluoroscopic images, computed tomography (CT) scans, ultrasound images, and other medical imaging data. In one or more embodiments, the image management module 1302 may receive CT scans and fluoroscopic images for analysis. The image management module 1302 may receive the CT scans and / or fluoroscopic images from a fluoroscopy system 1210 and / or previously stored data corresponding to a patient 1202. The image management module 1302 may preprocess such image data prior to providing the image data or analysis results to one or more other modules, including but not limited to a 3D model module 1304, a clot detection module 1310, a real-time update module 1312, a clot characterization module 1314, and an outcome assessment module 1316. In one or more embodiments, the image management module 1302 may perform segmentation of the CT scans and fluoroscopic images to generate segmented image data, and may employ one or more of one or more algorithms, artificial intelligence or one or more machine-learning models to process, analyze, or otherwise derive information from the image data. While this disclosure describes the image management module 1302 managing image data in a particular manner, this disclosure contemplates the image management module 1302 managing the image data in any suitable manner.

[0200] In one or more embodiments, the 3D model module 1304 may be configured to generate, manage, and update 3D models representing pulmonary vasculature, identified clots, and one or more medical devices used during a procedure by the clot management environment 1200. In one or more embodiments, the 3D model module 1304 may receive computed tomography (CT) scans and / or preprocessed image data from the image management module 1302 and applies one or more algorithms, including artificial intelligence or machine-learning techniques, to generate a 3D anatomical model that includes the pulmonary vasculature and associated clot structures. The 3D model module 1304 may further receive data relating to device geometry, dimensions, or schematics, such as catheter size or configuration (e.g., one or more catheters or components of the catheter system 1204), and generate corresponding device models for integration with the anatomical model. The generated 3D models may be transmitted to one or more other modules, including a pump control module 1306, a contrast injection module 1308, a clot detection module 1310, a realtime update module 1312, a clot characterization module 1314, an outcome assessment module 1316, and a navigation assistance module 1318. The 3D model module 1304 may - 47 - SG Docket No.: 10844-737.674also provide instructions to a display (e.g., display 1248) to present the 3D model via an interactive user interface (e.g., user interface 1258), which may receive user input from a physician to annotate, modify, or correct aspects of the model, including identified clot locations or anatomical features. While this disclosure describes the 3D model module 1304 managing one or more 3D models representing pulmonary vasculature, identified clots, and one or more medical devices in a particular manner, this disclosure contemplates the 3D model module 1304 managing one or more 3D models representing pulmonary vasculature, identified clots, and one or more medical devices in any suitable manner.

[0201] In one or more embodiments, the pump control module 1306 may be configured to control aspiration of clot material and operation of a jetting catheter (e.g., jetting catheter 1220) during a procedure. In one or more embodiments, the pump control module 1306 may receive user input from a physician to initiate, modify, or terminate aspiration and / or jetting operations and may apply logic to inhibit jetting until contact with a clot has been detected, such as based on an indication received from a clot detection module 1310. The pump control module 1306 may further determine aspiration and jetting parameters based on procedural data, including one or more of a model of a catheter, dimensions of blood vessels, or a 3D anatomical model, and may control one or more valves, pumps, or flow-regulating elements, including adjusting a valve opening or flow rate, to achieve a desired level of aspiration or jetting. The pump control module 1306 may generate and transmit pump control instructions to one or more pumps (e.g., pump 1250 of the console system 1208) to control aspiration and jetting operations and may additionally provide instructions to a display (e.g., display 1248) to present a user interface indicative of operational status or control parameters, such as current aspiration or jetting intensity levels. While this disclosure describes a pump control module 1306 controlling a pump in a particular manner, this disclosure contemplates a pump control module 1306 controlling a pump in any suitable manner.

[0202] In one or more embodiments, the contrast injection module 1308 may be configured to control delivery of a contrast agent into one or more lumens, including a jetting sheath (e.g., outer sheath 1222), a jetting catheter (e.g., jetting catheter 1220), an aspiration lumen (e.g., aspiration lumen of jetting catheter 1220), or another delivery lumen of the clot management system 1200 (e.g., one or more catheters of catheter system 1204). In one or more embodiments, the contrast injection module 1308 may receive user input from a physician to initiate contrast injection for fluoroscopic imaging. As an example, a physician may actuate a button through contrast control 1218 to send a signal to the contrast injection module 1308 to initiate contrast injection. The contrast injection module 1308 may determine a volume, rate, or concentration of contrast to be delivered based on at least procedural data,- 48 - SG Docket No.: 10844-737.674including dimensions of blood vessels derived from a 3D anatomical model received from the 3D model module 1304. The contrast injection module 1308 may instruct the console system 1208 to mix the contrast agent with a diluent, such as saline, including adjusting a mixing ratio based on image quality metrics, such as fluoroscopic brightness or contrast levels associated with visualizing clots or other anatomical structures. In one or more embodiments, the contrast injection module 1308 may apply one or more safety thresholds to limit contrast usage, with an optional override capability, and may further interface with a real-time update module 1312 to generate enhanced imaging output that permits reduced contrast delivery while maintaining image clarity. The contrast injection module 1308 may generate and transmit control instructions to one or more contrast injector components (e.g., contrast controller 1254) and may coordinate operation with a pump control module 1306 to control delivery of contrast through one or more of a jetting catheterjetting sheath, aspiration lumen, or other delivery pathways. The contrast injection module 1308 may send an indication to a pump control module 1306 when a premixed contrast solution is available or being injected. While this disclosure describes a contrast injection module 1308 controlling delivery of a contrast agent in a particular manner, this disclosure contemplates a contrast injection module 1308 controlling delivery of a contrast agent in any suitable manner.

[0203] In one or more embodiments, the clot detection module 1310 may be configured to detect engagement of clot material with one or more components of the clot management environment 1200 (e.g., one or more catheters of the catheter system 1204) based on one or more inputs, including sensor data and image data received from the image management module 1302. In one or more embodiments, the clot detection module 1310 may utilize one or more sensors, such as pressure sensors or impedance sensors associated with a catheter (e.g., jetting catheter 1220) or pump (e.g., 1250), to detect clot engagement or aspiration of clot material. The clot detection module 1310 may apply one or more algorithms to evaluate catheter position within the pulmonary vasculature, including based on insertion depth or a 3D anatomical model received from the 3D model module 1304, and to adjust or qualify detection logic to reduce false indications, such as misinterpreting vessel wall contact as clot engagement. The clot detection module 1310 may use one or more algorithms to adjust aspiration based on how far a catheter (e.g., a jetting catheter 1220) is inserted into the pulmonary vasculature to avoid inadvertently vacuuming blood vessels and misinterpreting the blood vessel contact as engagement with clot material. The clot detection module 1310 may use a 3D model to infer whether a clot is near and clot engagement is possible. Upon detecting clot engagement or a likelihood thereof, the clot detection module 1310 may generate notifications or alerts for presentation via a display (e.g., display 1248) or user - 49 - SG Docket No.: 10844-737.674interface of clot engagement. The clot engagement module 1310 may further transmit control signals or permissions to a pump control module 1306 to enable or initiate aspiration operations. While this disclosure describes the clot detection module 1310 as detect engagement of clot material in a particular manner, this disclosure contemplates engagement of clot material in any suitable manner.

[0204] In one or more embodiments, the real-time update module 1312 may be configured to update and present a 3D model of pulmonary vasculature and associated clots in real time during a procedure. In one or more embodiments, the real-time update module 1312 may receive a 3D model from the 3D model module 1304 and provide the model for display to a physician via a display (e.g., 1248) of a console system 1208. The real-time update module 1312 may receive fluoroscopic images and / or preprocessed image data from the image management module 1302 to update the 3D model based on current procedural conditions. The real-time update module 1312 may generate simulated flow representations, including simulated volumetric flow or contrast “puff’ models, and compare such simulations with corresponding fluoroscopic image data to refine or validate the 3D model. The real-time update module 1312 may further update a pose or orientation of the 3D model in real time based on sensor data indicative of movement of an imaging system, such as a C-arm, and / or based on image analysis of real-time fluoroscopic images. The real-time update module 1312 may additionally receive user input from a physician to modify the 3D model during the procedure, transmit updated model information to the 3D model module 1304, and provide real-time updates to a navigation assistance module 1320 to support procedural guidance. The physician may manipulate a 3D model in real-time (e.g., adjusting a pose or zoom of a view of the 3D model), and the real-time update module 1312 may track the adjustments of one or more parameters of the 3D model (e.g., pose or zoom) and send the adjustments to a navigation assistance module 1320 to generate instructions on how to manipulate the fluoroscopy system 1210 (e.g.,. a C-arm) to get the same view. While this disclosure describes a real-time update module 1312 updating and presenting a 3D model in a particular manner, this disclosure contemplates a real-time update module 1312 updating and presenting a 3D model in any suitable manner.

[0205] In one or more embodiments, the clot characterization module 1314 may be configured to analyze image data, including fluoroscopic images received from the image management module 1302 and / or 3D anatomical models received from the 3D model module 1304, to characterize one or more identified clots within the pulmonary vasculature. In one or more embodiments, clot characterization may include determining one or more attributes of a clot, such as volume, spatial extent relative to a vessel lumen, and chronicity, including - 50 - SG Docket No.: 10844-737.674density, uniformity, aggregation patterns, or morphological features indicative of clot age or integration with a vessel wall. The clot characterization module 1314 may provide clotspecific information for presentation via a user interface, including interactive displays (e.g., display 1348 coupled with user interface 1358) that allow a physician to select a clot within a 3D model to view associated characterization data. The clot characterization module 1314 may further classify or grade clots and generate treatment-related recommendations, including whether a clot is suitable for thrombectomy, balloon pulmonary angioplasty, staged treatment, or no treatment, such as based on clot features, anatomical context, or historical or previous procedural data (e.g., data corresponding to similar procedures done in the past). As an example, the clot characterization module 1314 may characterize one or more clots in such a way to recommend treatment in stages to reduce perfusion risk based on historical data. In one or more embodiments, the clot characterization module 1314 may identify particular clot configurations, such as saddle clots, and may provide corresponding procedural guidance. The clot characterization module 1314 may transmit clot characterization information to the 3D model module 1304 to update clot representations within the 3D model and to an outcome assessment module 1316 to support tracking of treatment outcomes for identified clots. The clot characterization module 1314 may send clot characterization information to procedure planning module 1318 to plan a procedure to address one or more clots. While this disclosure describes a clot characterization module 1314 analyzing data to characterize one or more clots within a pulmonary vasculature, this disclosure contemplates a clot characterization module 1314 analyzing data to characterize one or more clots within a pulmonary vasculature in any suitable manner.

[0206] In one or more embodiments, the outcome assessment module 1316 may be configured to evaluate procedural outcomes by comparing pre-procedure and post-procedure image data received from the image management module 1302 and / or 3D anatomical models received from the 3D model module 1304. In one or more embodiments, the outcome assessment module 1316 may determine an extent of clot removal by analyzing differences between pre-procedure and post-procedure images. The outcome assessment module 1316 may estimate perfusion within the pulmonary vasculature based on remaining clot burden as represented in the 3D model and corresponding fluoroscopic image data. The outcome assessment module 1316 may generate one or more quantitative metrics, including an obstruction index, a Miller index, or a refined modified Miller index, and may estimate an impact of clot removal on patient outcomes. The outcome assessment module 1316 may present outcome-related information via a user interface for physician reference and may allow a physician to preselect one or more clots as procedural targets while tracking partial or - 51 - SG Docket No.: 10844-737.674complete removal of such clots. In one or more embodiments, the outcome assessment module 1316 may generate treatment-related recommendations, including staged treatment planning to reduce reperfusion injury risk, and may provide procedural planning support for one or more procedures. As an example, the outcome assessment module 1316 may recommend a type a treatment (e.g., thrombectomy or BP A) and / or to treat in stages to reduce risk of reperfusion injuries. The outcome assessment module 1316 may provide procedural data for training or updating one or more machine-learning models and may transmit outcome-related data or procedural guidance to one or more of a procedure planning module 1318 or a navigation assistance module 1320. As an example, the outcome assessment module 1316 may send data of selected clots to target to the navigation assistance module 1320. While this disclosure describes an outcome assessment module 1316 evaluating procedural outcomes in a particular manner, this disclosure contemplates an outcome assessment module 1316 evaluating procedural outcomes in any suitable manner.

[0207] In one or more embodiments, the procedure planning module 1318 may be configured to generate procedural plans to assist a physician in navigating pulmonary vasculature during a procedure. In one or more embodiments, the procedure planning module 1318 may generate a navigation plan based on one or more inputs, including a 3D anatomical model received from the 3D model module 1304, models of medical devices (e.g., one or more catheters of catheter system 1204), clot characterization information received from a clot characterization module 1314, and perfusion estimates or outcome metrics received from an outcome assessment module 1316. The procedure planning module 1318 may receive user input identifying one or more clots selected for treatment. As an example, a physician may select one or more identified clots within a 3D model presented on a display 1248 coupled to a user interface 1258 (e.g., a touch-screen display). The procedure planning module 1318 may generate a step-by-step or sequential plan for advancing and manipulating a catheter (e.g., a jetting catheter 1220) through the pulmonary vasculature, including guidance relating to device orientation, advancement, knob actuation, applied forces, and the like to control the catheter through a clot management procedure. The generated procedural plan may be transmitted to a navigation assistance module 1320 for presentation of corresponding instructions or guidance to the physician during the procedure. While this disclosure describes the procedure planning module 1318 generating procedural plans to assist a physician in a particular manner, this disclosure contemplates the procedure planning module 1318 generating procedural plans in any suitable manner.

[0208] In one or more embodiments, the navigation assistance module 1320 may be configured to provide procedural guidance to assist a physician in navigating pulmonary - 52 - SG Docket No.: 10844-737.674vasculature during a procedure. In one or more embodiments, the navigation assistance module 1320 may receive a procedural plan from a procedure planning module 1318 and generate navigation instructions for operating a catheter to approach and aspirate clot material and / or manage clot (e.g., through a BPA procedure). The navigation assistance module 1320 may generate a sequence of instructions in advance and / or provide real-time guidance during the procedure, including instructions for adjusting an imaging system (e.g., fluoroscopy system 1210), such as a C-arm, to obtain desired views. The navigation assistance module 1320 may further monitor procedural progress and generate alerts when detected movements or device positions deviate from planned or estimated positions beyond a threshold. The navigation assistance module 1320 may transmit navigation instructions to a display (e.g., display 1248) or console system 1208 for presentation to the physician in one or more formats, including visual and audible outputs. While this disclosure describes the navigation assistance module providing procedural guidance in a particular manner, this disclosure contemplates providing procedural guidance in any suitable manner.

[0209] In one or more embodiments, while this disclosure describes one or more components of processor environment 1300 as performing one or more functions, this disclosure contemplates one or more components of the processor environment 1300 performing one or more other functions either not described or described as being performed by one or more other components. As an example, the procedure planning module 1318 may perform one or more functions of the navigation assistance module 1320. In one or more embodiments, while this disclosure describes a processor environment 1300 as comprising a number of software modules, this disclosure contemplates a processor environment 1300 as comprising any number of software modules. As an example, the procedure planning module 1318 may be combined with the navigation assistance module 1320 to perform both functions described herein.

[0210] In one or more embodiments, referring to FIG. 14, an example process 1400 of characterizing clots is shown. Process 1400 may be performed by one or more components of the clot management environment $$00 or any other suitable system. As an example, process 1400 may be performed by a system (e.g., console system $$08). For instance, a processor $$40 and / or a processor environment ##00 executed on the processor $$40 may perform one or more steps of process 1400.

[0211] In one or more embodiments, process 1400 may start with step 1402, where a system may access one or more images. As an example, a console system 1408 may receive one or more fluoroscopic images from a fluoroscopy system 1410. The console system 1408 may store the received images in storage 1414. After accessing the one or more images, the - 53 - SG Docket No.: 10844-737.674system may segment the one or more images to identify one or more clots within a pulmonary vasculature in step 1404. The system may use a processor (e.g., processor 1240) to execute one or more algorithms and / or one or more machine-learning models to segment the one or more images to identify one or more clots within a pulmonary vasculature. The system may then determine one or more characteristics of the one or more clots in step 1406. The system may use one or more algorithms and / or one or more machine-learning models to determine one or more characteristics as described herein. The system may then associate the determined one or more characteristics to the respective one or more identified clots in step 1408. The system may generate a table to associate one or more characteristics to the respective one or more identified clots. As an example, the system may generate an ID for each of the identified clots within a pulmonary vasculature and assign one or more characteristics to the respective one or more identified clots. The system may determine whether one or more determined characteristics need to be modified in step 1410. As an example, the system may perform a check on whether an input is received to modify one or more characteristics. In step 1412, the system may determine whether one or more characteristics need to be modified. If the system determines that one or more characteristics of one or more clots need to be modified, the process 1400 continues to step 1414, where the system modifies one or more determined characteristics. As an example, if a clot was incorrectly characterized as chronic and a physician inputs a correction that the clot should be labelled as acute, the system may update the determined characteristic of the clot. After modifying the one or more determined characteristics, the process 1400 may return to step 1410, where the system determines if one or more determined characteristics need to be modified. If the system determines no other determined characteristics need to be modified, then the process 1400 may continue to step 1416, where the system may finalize the clot characterization of the one or more identified clots. The system may update clot characterization information stored in a 3D model. The system may store the clot characterization information to be later accessed for a clot management procedure. While this disclosure describes a process of characterizing clot in a particular manner, this disclosure contemplates characterizing clot in any suitable manner.

[0212] In one or more embodiments, referring to FIG. 15, an example process 1500 of performing an outcome assessment is shown. Process 1500 may be performed by one or more components of the clot management environment 1200 or any other suitable system. As an example, process 1500 may be performed by a system (e.g., console system 1208). For instance, a processor 1240 and / or a processor environment 1300 executed on the processor 1240 may perform one or more steps of process 1500.- 54 - SG Docket No.: 10844-737.674

[0213] In one or more embodiments, process 1500 may start with step 1502, where a system may access one or more of a 3D model of a pulmonary vasculature of a patient (e.g., patient 1202), clot characterization information for one or more clots identified in the 3D model, and one or more images (e.g., fluoroscopic images from fluoroscopy system 1210). After accessing the data, the system may determine an obstruction index for one or more clots using one or more of the 3D model, clot characterization information, and one or more images in step 1504 as described herein. After determining the obstruction index for one or more clots, the system may calculate a perfusion estimate for one or more lung lobe groupings using one or more of the 3D model, clot characterization information, and the one or more images in step 1506 as described herein. The system may use one or more algorithms and / or one or more machine-learning models to generate a perfusion estimation using the obstruction index of the one or more clots. The perfusion estimate may provide a physician a context of the amount of perfusion that is currently possible for a patient prior to any clot management procedure. The system may identify one or more recommended clots for removal in step 1508. The system may use one or more of historical data on previous procedures and one or more algorithms and / or one or more machine-learning models to identify one or more recommended clots for removal as described herein. The system may then estimate the change in the obstruction and perfusion in removing the one or more identified clots in step 1510. The system may determine if any additional input is received to change the identified clots for removal in step 1512. The system may check to determine whether one or more user inputs is received through a user interface (e.g., user interface $$58) indicative of a request to change one or more identified clots for removal. The system determines whether one or more inputs have been received to alter the identified clots for removal in step 1514. If one or more inputs are received to change the identified clots for removal, the process 1500 continues to step 1516, where the system modifies the identified clots for removal based on the received input. After modifying the identified clots for removal, the system may return to step 1512, where the system may determine if additional input is received to change the identified clots for removal. If the system determines that no input has been received to change the identified clots for removal, the process 1500 may continue to step 1518, where the system presents an outcome assessment on a console system. The system may present an outcome assessment on a display (e.g., display 1248 of a console system 1208), where the outcome assessment may detail a predicted outcome from removal of one or more identified clots. While this disclosure describes a process of performing an outcome assessment in a particular manner, this disclosure contemplates a process of performing an outcome assessment in any suitable manner.- 55 - SG Docket No.: 10844-737.674

[0214] As one of skill in the art will appreciate from the disclosure herein, various components of the thrombus removal systems described above can be omitted without deviating from the scope of the present technology. As discussed previously, for example, the present technology can be used and / or modified to remove other types of emboli that may occlude a blood vessel, such as fat, tissue, or a foreign substance. Further, although some embodiments herein are described in the context of thrombus removal from a pulmonary artery, the disclosed technology may be applied to removal of thrombi and / or emboli from other portions of the vasculature (e.g., in neurovascular, coronary, or peripheral applications). Likewise, additional components not explicitly described above may be added to the thrombus removal systems without deviating from the scope of the present technology.Accordingly, the systems described herein are not limited to those configurations expressly identified, but rather encompasses variations and alterations of the described systems.Conclusion

[0215] The above detailed description of embodiments of the technology are not intended to be exhaustive or to limit the technology to the precise forms disclosed above. Although specific embodiments of, and examples for, the technology are described above for illustrative purposes, various equivalent modifications are possible within the scope of the technology as those skilled in the relevant art will recognize. For example, although steps are presented in a given order, alternative embodiments may perform steps in a different order. The various embodiments described herein may also be combined to provide further embodiments.

[0216] From the foregoing, it will be appreciated that specific embodiments of the technology have been described herein for purposes of illustration, but well-known structures and functions have not been shown or described in detail to avoid unnecessarily obscuring the description of the embodiments of the technology. Where the context permits, singular or plural terms may also include the plural or singular term, respectively.

[0217] Unless the context clearly requires otherwise, throughout the description and the examples, the words "comprise," "comprising," and the like are to be construed in an inclusive sense, as opposed to an exclusive or exhaustive sense; that is to say, in the sense of "including, but not limited to." As used herein, the terms "connected," "coupled," or any variant thereof, means any connection or coupling, either direct or indirect, between two or more elements; the coupling of connection between the elements can be physical, logical, or a combination thereof. Additionally, the words "herein," "above," "below," and words of similar import, when used in this application, shall refer to this application as a whole and not to any particular portions of this application. Where the context permits, words in the above - 56 - SG Docket No.: 10844-737.674Detailed Description using the singular or plural number may also include the plural or singular number respectively. As used herein, the phrase "and / or" as in "A and / or B" refers to A alone, B alone, and A and B. Additionally, the term "comprising" is used throughout to mean including at least the recited feature(s) such that any greater number of the same feature and / or additional types of other features are not precluded. It will also be appreciated that specific embodiments have been described herein for purposes of illustration, but that various modifications may be made without deviating from the technology. Further, while advantages associated with some embodiments of the technology have been described in the context of those embodiments, other embodiments may also exhibit such advantages, and not all embodiments need necessarily exhibit such advantages to fall within the scope of the technology. Accordingly, the disclosure and associated technology can encompass other embodiments not expressly shown or described herein.- 57 - SG Docket No.: 10844-737.674

Claims

1. CLAIMS:What is claimed is:

1. A system comprising:one or more processors;memory coupled to the one or more processors, wherein the memory includes computer-program instructions that, when executed by the one or more processors, cause the system to perform operations comprising:acquiring two-dimensional (2D) and / or three-dimensional (3D) imaging data of at least a torso of a patient;segmenting the imaging data of the patient to identify 1) the pulmonary vasculature, and / or 2) one or more clots or lesions within the pulmonary vasculature;generating a 3D pulmonary vasculature model of the patient from the segmented imaging data;generating a 3D clot model of the patient from the segmented imaging data; characterizing and assessing a chronicity of the one or more clots or lesions within the 3D pulmonary vasculature model and the 3D clot model; andconsidering the characterization and assessment of chronicity, outputting a treatment recommendation of selected clots or lesions of the one or more clots or lesions to target for removal.

2. The system of claim 1, wherein characterizing and assessing the chronicity comprises evaluating pixels or voxels in the 2D and / or 3D imaging data corresponding to the one or more clots or lesions.

3. The system of claim 2, wherein evaluating pixels or voxels comprises evaluating a Houndsfield scale value of the pixels or voxels.

4. The system of claim 1, wherein characterizing and assessing the chronicity comprises evaluating voxels in the 2D and / or 3D imaging data of boundaries between one or more clots or lesions and a vessel wall.

5. The system of claim 4, further comprising determining vessel wall integration of the one or more clots or lesions.- 58 - SG Docket No.: 10844-737.6746. The system of claim 1, wherein characterizing and assessing the chronicity comprises assessing the aggregation or uniformity of pixels or voxels of the one or more clots or lesions.

7. The system of claim 1, wherein the computer-program instructions further cause the system to perform operations comprising:generating an obstruction index for the one or more clots or lesions that indicates a clot or lesions contribution to obstruction of the pulmonary vasculature; and presenting the obstruction index to the user for the one or more clots or lesions.

8. The system of claim 7, wherein presenting the obstruction index comprises presenting a volume of the one or more clots or lesions.

9. The system of claim 7, wherein presenting the obstruction index comprises presenting a percentage of the pulmonary vasculature obstructed by the one or more clots or lesions.

10. The system of claim 7, wherein generating the obstruction index comprises extracting a centerline of a target vessel at a location of a specific clot or lesion, determining a size of the target vessel at the location with the specific clot or lesion removed, and determining the degree of obstruction that the specific clot or lesion is responsible for within the target vessel at the location.

11. The system of claim 1, wherein characterizing and assessing chronicity of the one or more clots comprises classifying the one or more clots into categories pertaining to the age of the clot.

12. The system of claim 1, wherein characterizing and assessing the one or more clots comprises identifying clots that should be excluded from treatment based on their location within the 3D pulmonary vasculature model.

13. The system of claim 1, wherein characterizing and assessing the one or more clots comprises identifying clots that should not be accessed with a thrombectomy catheter.- 59 - SG Docket No.: 10844-737.67414. The system of claim 1, wherein characterizing and assessing the one or more clots comprises identifying clots that can be accessed with a thrombectomy catheter.

15. The system of claim 1, wherein characterizing and assessing the one or more clots comprises determining a pre-operative Miller Score or an equivalent index or metric based on the one or more clots to indicate an extent of obstruction to blood flow caused by the one or more clots.

16. The system of claim 1, wherein characterizing and assessing the one or more clots comprises determining a post-operative Miller Score or an equivalent index or metric based if selected clots are removed to indicate an extent of obstruction to blood flow caused by the one or more clots.

17. The system of claim 1, wherein characterizing and assessing the one or more clots comprises determining a pre-operative Miller Score based on the one or more clots, identifying one or more selected clots to be removed, and determining a postoperative Miller Score if the selected clots are removed.

18. The system of claim 1, wherein the 2D or 3D imaging data comprises X-ray imaging data.

19. The system of claim 1, wherein the 2D or 3D imaging data comprises computed tomography (CT) imaging data.

20. The system of claim 1, wherein the 2D or 3D imaging data comprises magnetic resonance imaging (MRI) imaging data.

21. The system of claim 1, wherein the 2D or 3D imaging data comprises positron emission tomography (PET) imaging data.- 60 - SG Docket No.: 10844-737.67422. The system of claim 1, wherein the 2D or 3D imaging data comprises a fusion of one or more sources of imaging data.

23. The system of claim 1, wherein the 2D or 3D imaging data comprises ultrasound imaging data.

24. A computer implemented method, comprising:acquiring two-dimensional (2D) and / or three-dimensional (3D) imaging data of at least a torso of a patient;segmenting the imaging data of the patient to identify 1) the pulmonary vasculature, and / or 2) one or more clots or lesions within the pulmonary vasculature;generating a 3D pulmonary vasculature model of the patient from the segmented imaging data;generating a 3D clot model of the patient from the segmented imaging data; characterizing and assessing a chronicity of the one or more clots or lesions within the 3D pulmonary vasculature model and the 3D clot model; andconsidering the characterizing and assessing, outputting a treatment recommendation of selected clots or lesions of the one or more clots or lesions to target for removal.

25. The method of claim 24, wherein characterizing and assessing the chronicity comprises evaluating voxels in the 2D and / or 3D imaging data corresponding to the one or more clots or lesions.

26. The method of claim 25, wherein evaluating voxels comprises evaluating a Houndsfield scale value of the voxels.

27. The method of claim 24, wherein characterizing and assessing the chronicity comprises evaluating voxels in the 2D and / or 3D imaging data of boundaries between one or more clots or lesions and a vessel wall.

28. The method of claim 27, further comprising determining vessel wall integration of the one or more clots or lesions.- 61 - SG Docket No.: 10844-737.67429. The method of claim 24, wherein characterizing and assessing the chronicity comprises assessing the aggregation or uniformity of pixels of the one or more clots or lesions.

30. The method of claim 24, further comprising:generating an obstruction index for the one or more clots or lesions that indicates a clot or lesions contribution to obstruction of the pulmonary vasculature; and presenting the obstruction index to the user for the one or more clots or lesions.

31. The method of claim 30, wherein presenting the obstruction index comprises presenting a volume of the one or more clots or lesions.

32. The method of claim 30, wherein presenting the obstruction index comprises presenting a percentage of the pulmonary vasculature obstructed by the one or more clots or lesions.

33. The method of claim 30, wherein generating the obstruction index comprises extracting a centerline of a target vessel at a location of a specific clot or lesion, determining a size of the target vessel at the location with the specific clot or lesion removed, and determining the degree of obstruction that the specific clot or lesion is responsible for within the target vessel at the location.

34. The method of claim 24, wherein characterizing and assessing chronicity of the one or more clots comprises classifying the one or more clots into categories pertaining to the age of the clot.

35. The method of claim 24, wherein characterizing and assessing the one or more clots comprises identifying clots that should be excluded from treatment based on their location within the 3D pulmonary vasculature model.

36. The method of claim 24, wherein characterizing and assessing the one or more clots comprises identifying clots that should not be accessed with a thrombectomy catheter.- 62 - SG Docket No.: 10844-737.67437. The method of claim 24, wherein characterizing and assessing the one or more clots comprises identifying clots that can be accessed with a thrombectomy catheter.

38. The method of claim 24, wherein characterizing and assessing the one or more clots comprises determining a pre-operative Miller Score or an equivalent index or metric based on the one or more clots to indicate an extent of obstruction to blood flow caused by the one or more clots.

39. The method of claim 24, wherein characterizing and assessing the one or more clots comprises determining a post-operative Miller Score or an equivalent index or metric based if selected clots are removed to indicate an extent of obstruction to blood flow caused by the one or more clots.

40. The method of claim 24, wherein characterizing and assessing the one or more clots comprises determining a pre-operative Miller Score based on the one or more clots, identifying one or more selected clots to be removed, and determining a postoperative Miller Score if the selected clots are removed.

41. The method of claim 24, wherein the 2D or 3D imaging data comprises X-ray imaging data.

42. The method of claim 24, wherein the 2D or 3D imaging data comprises computed tomography (CT) imaging data.

43. The method of claim 24, wherein the 2D or 3D imaging data comprises magnetic resonance imaging (MRI) imaging data.

44. The method of claim 24, wherein the 2D or 3D imaging data comprises positron emission tomography (PET) imaging data.- 63 - SG Docket No.: 10844-737.67445. The method of claim 24, wherein the 2D or 3D imaging data comprises a fusion of one or more sources of imaging data.

46. The method of claim 24, wherein the 2D or 3D imaging data comprises ultrasound imaging data.

47. A thrombectomy method, comprising:acquiring two-dimensional (2D) and / or three-dimensional (3D) imaging data of at least a torso of a patient;segmenting the imaging data of the patient to identify 1) the pulmonary vasculature, and / or 2) one or more clots or lesions within the pulmonary vasculature;generating a 3D pulmonary vasculature model of the patient from the segmented imaging data;generating a 3D clot model of the patient from the segmented imaging data; advancing a thrombectomy catheter into a pulmonary vasculature toward a target thrombus;delivering a bolus of contrast from the thrombectomy catheter into the pulmonary vasculature;modeling a flow of the bolus of contrast in the 3D pulmonary vasculature model; calculating a perfusion estimate of the bolus of contrast into one or more sections of the pulmonary vasculature; anddisplaying the perfusion estimate.

48. The method of claim 47, further comprising obtaining fluoroscopy images of the flow of the bolus of contrast, and updating modeling the flow of the bolus of contrast based on the fluoroscopy images.

49. The method of claim 47, further comprising obtaining fluoroscopy images of the flow of the bolus of contrast, and updating the perfusion estimate based on the fluoroscopy images.

50. The method of claim 47, further comprising removing the target thrombus with the thrombectomy catheter, delivering a second bolus of contrast into the pulmonary vasculature, modeling the flow of the second bolus, updating the perfusion estimate based on the second bolus, and displaying the updated perfusion estimate.- 64 - SG Docket No.: 10844-737.6741. The method of claim 50, further comprising repeating the steps of claim 50 formultiple targeted thrombus.- 65 - SG Docket No.: 10844-737.674

Images