Systems and methods for intravascular lithotripsy recommendations

Abstract

The present disclosure is generally directed to systems and methods including a model trained to determine one or more IVL parameters for preparing lesions within a blood vessel for further treatment. The parameters may be used to configure an IVL system that can prepare the lesions for further treatment. IVL parameters may include, for example, balloon size, balloon diameter, balloon length, balloon pressure, a number of pulses, a number of emitters for a balloon, a shockwave energy distribution for a balloon, a pulse pattern, post-treatment vessel diameter, rime for balloon to be positioned within a vessel, or catheter flexibility, treatment positions, or the like. The model may receive, as input, imaging data of a region of interest. The region of interest may include, for example, one or more lesions to be treated through IVL. In some examples, the imaging data may be intravascular imaging.

Description

ABTLLI-128SYSTEMS AND METHODS FOR INTRAVASCULAR LITHOTRIPSY RECOMMENDATIONSCROSS REFERENCE TO RELATED APPLICATION

[0001] The present application claims the benefit of priority from United States Provisional App. No.63 / 738,030 filed on December 23, 2024, which is hereby incorporated by reference in its entirety.FIELD OF THE DISCLOSURE

[0002] The present disclosure relates to determining recommended intravascular lithotripsy (IVL) parameters based on vessel image data.BACKGROUND

[0003] Intravascular lithotripsy (IVL) is an effective procedure used in the treatment of the calcification of a patient’s blood vessels. Such calcification is present in obstructive arterial diseases such as coronary artery disease (CAD) and peripheral artery disease (PAD). Here, lesions containing calcium deposits are present in the walls of the blood vessel. Such lesions can cause direct obstruction to the blood flow, typically where the calcium deposit is present on the intimal layer of the blood vessel. Treatment here centers around removing or reducing these obstructions. Calcium deposits in the medial layer of the blood vessel impinge upon the elasticity and conformity of the blood vessel, in effect hardening the wall.

[0004] While percutaneous interventions, such as stenting can be used in both cases to try and ensure the vessel remains unobstructed and suitably dilated, the ongoing presence of calcium in the lesions can lead to stent under expansion and other complications which negatively impact upon the health of the patient.

[0005] IVL techniques use pressure or sound waves to apply force to the calcified deposit to thereby cause the deposit to fracture and break up. This can result in the breaking down or reduction of calcified lesion in the intimal wall of a given vessel. For the case of calcified lesions in the medial layer of a vessel, by fracturing the lesion, the overall compliance of the vessel may be somewhat restored or, at least, improved. With current systems, it is recommended that no segment of a vessel receive more than a maximum number of pulses specified for the IVL device, which may range from about 80 pulses for coronary systems to about 300 pulses for peripheral systems.

[0006] IVL is routinely used to treat calcified lesions. This is often as a form of vessel pre-preparation in anticipation of a further procedure, such as the insertion of a stent. Hence IVL can improve the efficacy of following procedures.

[0007] IVL techniques typically generate pressure waves by selectively vaporising fluid contained within a balloon catheter placed in the subject vessel. Given the forces involved, the balloon catheter and associated internal triggering apparatus may often only be able to deliver a certain number of pressure wave pulses before the balloon ruptures and / or the triggering apparatus fails. As such, during IVL treatment for a single insertion of an IVL balloon only a limited number of pulses or doses may be delivered. ForABTLLI-128coronary systems, some are limited to 80 or 120 pulses. For some peripheral systems, some may afford up to 300 pulses before the balloon must be withdrawn and replaced. This may be thought of as the capacity of a given IVL balloon. For IVL to be effective, careful attention must be paid to the parameters of the procedure, such as the selection of the balloon size, the balloon pressure, the emitter arrangement or energy / pressure profile, the number of pulses used to ensure that the procedure is el'l'ccli ve in reducing the lesions, whilst also not causing further damage to the vessel itself, or risking a balloon failure. This currently largely relies on the experience and inluilion of the physician performing the procedure. Furthermore, current approaches typically advise a level of overtreatment to compensate for the uncertainty. For example, where lesions are longer than the el'l'ccli ve treatment area of the IVL balloon, current practice can be to simply advance the IVL balloon to a new posilion with only a small overlap over the previous position, effectively doubling the treatment area, regardless of whether the lesion is actually double the length of a single treatment area.BRIEF SUMMARY

[0008] The present disclosure is generally directed to systems and methods including a model that is trained to determine one or more IVL parameters for preparing lesions within a blood vessel for further treatment. The parameters may be used to configure an IVL system that can prepare the lesions for further treatment. IVL parameters may include, for example, balloon size, balloon diameter, balloon length, balloon pressure, a number of pulses, a number of emitters for a balloon, a shockwave energy distribution for a balloon, a pulse pattern, post-treatment vessel diameter, lime for balloon to be positioned within a vessel, catheter flexibility, treatment positions, or the like. In some examples, balloon pressure may include and / or correspond to an inflation pressure. In another example, a number of pulses may include and / or correspond to an IVL dose. The pulse pattern may, in some examples, correspond to a mode of the IVL device, such as an enhanced power delivery mode. The model may receive, as input, imaging data of a region of interest. The region of interest may include, for example, one or more lesions to be treated through IVL. In some examples, the imaging data may be intravascular imaging, such as optical coherence tomography (OCT) imaging data, intravascular ultrasound (IVUS) imaging data, near-infrared spectroscopy (NIRS) imaging data, micro-OCT imaging data, or the like. The model may be configured to synthesise the experiences of numerous physicians or surgeons performing a large number of intravascular lithotripsy procedures. In this way, the trained model is enabled to recommend one or more IVL parameters for a given patient based on image data of the patient, which may be taken as part of a treatment preparation protocol or after a treatment regimen.

[0009] In a first aspect there is provided a method of specifying an intravascular lithotripsy procedure for a treatment site of a patient vessel, which may comprise: receiving an initial set of intravascular image data of the treatment site; generating, using a trained model, at least one recommended intravascular lithotripsy configuration (or parameter) based on the set of intravascular image data; and providing for output the recommended intravascular lithotripsy configuration. The intravascular image data may be opticalABTLLI-128coherence tomography data, and / or intravascular ultrasound data. In particular, the image data may comprise a set of intravascular image frames.

[0010] The intravascular lithotripsy configuration may specify one or more intravascular lithotripsy parameters, and may optionally comprise one or more treatment positions. The intravascular lithotripsy configuration may therefore comprise a set of intravascular lithotripsy parameter values for each treatment position. Examples of intravascular lithotripsy parameters may include any one or more of: a balloon size, a balloon diameter, a balloon length, a balloon pressure, a number of pulses, a number of emitters for a balloon, a shockwave energy distribution for a balloon, a pulse pattern, post-treatment vessel diameter, time for balloon to be positioned within a vessel, catheter flexibility, or the like. In some examples, balloon pressure may include and / or correspond to an inflation pressure. In another example, a number of pulses may include and / or correspond to an IVL dose.[Oil] Providing for output, the recommended intravascular lithotripsy configuration may comprise displaying one or more parameters of the intravascular lithotripsy configuration in conjunction with the intravascular image data. Additionally, or alternatively providing for output the recommended intravascular lithotripsy configuration may comprise configuring an intravascular lithotripsy system according to at least part of the intravascular lithotripsy configuration.

[0012] The step of general! ng may comprise: determining a set of imaging characteristics (which may include angiographic characteristics and / or pathology characteristics), such as a set of vessel characteristics and / or one or more sets of lesion characteristics for lesions in the vessel at the treatment site, from the set of intravascular image data; and inputting the determined set of imaging characteristics to the trained model to generate the at least one recommended Intravascular lithotripsy system configuration. The set of vessel characteristics may comprise any one of more of: a reference vessel diameter at the treatment site, a minimum lumen diameter at the treatment site, a minimum lumen area at the treatment site, or one or more stenosis diameters at the treatment site. Similarly, the set of lesion characteristics may comprise any one of more of: length of a lesion, presence of calcium, arc length of calcium, location of a lesion, thickness of calcium, or calcification length.

[0013] Determining a set of imaging characteristics from the set of intravascular image data may comprise inputting the set of intravascular image data into a further trained model. The further trained model may be selected from a plurality of further trained models based on one or more region of interest (which may include one or more treatment sites or treatment positions) characteristics.

[0014] Alternatively, the step of generating may comprise inpulling the set of intravascular image data to the trained model, such as an image or volume segmentation model, to generate the at least one recommended IVL configuration. The trained model may be based on any of: V-net, U-net, CUMedVisionl, CUMedVision2, VGGNet, M2FCN, Coarse-to-Fine Stacked Fully Convolutional Net, Deep Active Learning Framework, ResNet and / or combinations thereof.ABTLLI-128

[0015] The method of the first aspect may further comprise the steps of: receiving a further set of intravascular image data of the treatment site following an intravascular lithotripsy procedure being carried out according to the recommended intravascular lithotripsy configuration; determining a measure of success of the intravascular lithotripsy procedure based on the further set of intravascular image data; updating the trained model based on the initial set of intravascular image data, the recommended intravascular lithotripsy configuration and the measure of success.

[0016] The method of the first aspect may further comprise displaying at least one treatment position of the intravascular lithotripsy configuration and one or more corresponding intravascular lithotripsy parameters for the treatment position. The at least one treatment position of the intravascular lithotripsy configuration may be displayed overlaid on the initial set of intravascular image data. One or more marker bands, or one or more marker band positions, may be displayed for a treatment position for aligning an IVL catheter and / or balloon.

[0017] In a second aspect there is provided a method of training a model for specifying an intravascular lithotripsy procedure, which may comprise: receiving a plurality of initial sets of intravascular imaging data for respective treatment sites of patient vessels, each set of initial intravascular imaging data corresponding to a respective intravascular lithotripsy configuration of a respective intravascular lithotripsy procedure performed at the respective treatment sites after the set of initial intravascular imaging data was generated; obtaining for each initial set of intravascular imaging data a measure of success for the respective intravascular lithotripsy procedure; training the model according to the plurality of initial sets of intravascular imaging data and the respective measures of success.

[0018] The step of training may additionally comprise annotating each initial set of intravascular imaging data with the respective intravascular lithotripsy configuration and the respective measure of success. Alternatively, the method of the second aspect may further comprise, for each initial set of intravascular image data, determining a set of characteristics, wherein the step of training comprises annotating each set of characteristics with the respective intravascular lithotripsy configuration and the respective measure of success, and training the model based on the plurality of annotated sets of characteristics.

[0019] According to a third aspect of the disclosure, there is provided a system adapted to carry out any of the above-mentioned aspects or any embodiment thereof.

[0020] To that end there is provided a system of specifying an IVL procedure for a treatment site of a patient vessel, the system may include one or more processors configured to: receive an initial set of intravascular image data, such as optical coherence tomography data, and / or intravascular ultrasound data, of the treatment site; generate, using a trained model, at least one recommended IVL configuration based on the set of intravascular image data; and provide for output the recommended intravascular lithotripsy configuration.ABTLLI-128

[0021] According to a fourth aspect of the disclosure, there is provided a computer program which, when executed by one or more processors, causes the one or more processors to carry out any of the above-mentioned aspects or any embodiment thereof. The computer program may be stored on a non-transient computer readable medium.

[0022] According to a fifth aspect of the disclosure, a system may comprise one or more processors configured to: provide, as input into an artificial intelligence (Al) model trained to determine one or more IVL parameters, image data of a region of interest may include one or more lesions. The one or more IVL parameters may include one or more treatment posilions within the region of interest. The one or more processors may also determine, by executing the Al model, the one or more IVL parameters. The one or more processors may also provide for output, via a display, an indication of the one or more IVL parameters.

[0023] In some examples, the one or more IVL parameters further may include one or more of a balloon size, a balloon diameter, a balloon length, a balloon pressure, a number of pulses, a number of emitters for a balloon, a shockwave energy distribution for a balloon, a pulse pattern, post-treatment vessel diameter, time for balloon to be positioned within the vessel, or catheter flexibility.

[0024] In some examples, the one or more treatment positions may include at least a portion of one of the one or more lesions.

[0025] In some examples, the one or more processors may be further configured to provide, as input into the model, one or more region of interest characteristics. The one or more region of interest characteristics may include one or more of a vessel tortuosity, vessel diameter values, a lesion location, vessel type, a number of administered pulses, or a calcium density.

[0026] In some examples, the one or more processors may be further configured to: determine, by executing the Al model, a set of imaging characteristics may comprise one or more of vessel characteristics of a vessel within the region of interest or lesion characteristics of a lesion within the region of interest. The vessel characteristics may be one or more of a reference vessel diameter, a minimum lumen diameter, a minimum lumen area, or a stenosis diameter. The lesion characteristics may be one or more of a length of the lesion, a presence of calcium, an arc length of calcium, a localion of the lesion, a thickness of calcium, a calcification length, or a calcium density. In some examples, the determining of the one or more IVL parameters is at least partially based on the one or more vessel characteristics or lesion characteristics.

[0027] In some examples, the one or more processors may be further configured to: determine, by executing the Al model, a confidence value for each of the determined one or more IVL parameters. The one or more processors may be further configured to: identify, based on the determined confidence values for each of the determined one or more IVL parameters, a subset of one or more IVL parameters. The subset may include IVL parameters having a confidence value above a threshold confidence value. The providing for output the indication of the one or more IVL parameters may include providing for outputABTLLI-128the subset of the one or more IVL parameters. The one or more processors may be further configured to: rank, based on the determined confidence value for each of the determined one or more IVL parameters, the determined one or more IVL parameters. The one or more IVL parameters are ranked from a highest confidence value to a lowest confidence value. The providing for output the indication of the one or more IVL parameters may include providing for output the ranked one or more IVL parameters.

[0028] In some examples, the one or more processors may be further configured to: receive, based on the indication of the one or more IVL parameters, feedback data confirming or rejecting the one or more IVL parameters; and provide the feedback data to the model as part of a feedback loop to update the model.

[0029] According to a sixth aspect of the disclosure, a method may comprise: providing, by one or more processors and as input into an Al model trained to determine one or more IVL parameters, image data of a region of interest may include one or more lesions. The one or more IVL parameters may include one or more treatment positions within the region of interest. The method may further include determining, by the one or more processors and by executing the Al model, the one or more IVL parameters. The method may further include providing for output, by the one or more processors and via a display, an indication of the one or more IVL parameters.

[0030] In some examples, the one or more IVL parameters further may include one or more of a balloon size, a balloon diameter, a balloon length, a balloon pressure, a number of pulses, a number of emitters for a balloon, a shockwave energy distribution for a balloon, a pulse pattern, post-treatment vessel diameter, time for balloon to be positioned within the vessel, or catheter flexibility.

[0031] In some examples, the one or more treatment positions may include at least a portion of one of the one or more lesions.

[0032] In some examples, the method may further include providing, as input into the model, one or more region of interest characteristics. In some examples, the one or more region of interest characlcrislics may include one or more of a vessel tortuosity, vessel diameter values, a lesion localion, vessel type, a number of administered pulses, or a calcium density.

[0033] In some examples, the method may further include determining, by executing the Al model, a set of imaging characteristics may comprise one or more of vessel characlcrislics of a vessel within the region of interest or lesion characteristics of a lesion within the region of interest.

[0034] According to a seventh aspect of the disclosure, one or more non-transitory computer-readable media for storing instructions that, when executed by one or more processors, may cause the one or more processors to perform operations comprising: providing, as input into an Al model trained to determine one or more IVL parameters, image data of a region of interest may include one or more lesions. The one or more IVL parameters may include one or more treatment posilions within the region of interest. The operations may further include determining, by executing the Al model, the one or more IVL parameters.ABTLLI-128The operations may further include providing for output, via a display, an indication of the one or more IVL parameters.BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.

[0036] Figure 1 illustrates an example method for determining one or more IVL parameters for a blood vessel of a padent displaying stenosis, according to aspects of the present disclosure.

[0037] Figure 2 is a flowchart of an example IVL configuration system, according to aspects of the present disclosure.

[0038] Figure 3 is a flowchart of another example IVL configuration system, according to aspects of the present disclosure.

[0039] Figure 4 is a flowchart of an example method of specifying IVL parameters for a region of interest of a patient vessel, which might be carried out using the example IVL configuration systems described in relation to Figures 2 or 3, according to aspects of the present disclosure.

[0040] Figure 5 is a flowchart of an example method of training a model for specifying IVL parameters, such as the models as discussed in relation to Figure 2 or 3, according to aspects of the present disclosure.

[0041] Figure 6 is a flowchart of another example method of specifying IVL parameters for a region of interest of a patient vessel, according to aspects of the present disclosure.

[0042] Figure 7 is a flowchart of another example method of specifying IVL parameters for a region of interest of a patient vessel, which might be carried out using the example IVL configuration systems described in relation to Figures 2 or 3, according to aspects of the present disclosure.

[0043] Figure 8 is an illustration of IVL emitter arrangements and corresponding energy profiles.

[0044] Figure 9 is an example output of the method illustrated in Figure 4, according to aspects of the present disclosure.

[0045] Figure 10 is another example output of the method illustrated in Figure 4, according to aspects of the present disclosure.

[0046] Figure 11 A is another example output, according to aspects of the present disclosure.

[0047] Figure 1 IB is another example of the output in Figure 11 A.

[0048] Figure 12 is another example output, according to aspects of the present disclosure.

[0049] Figure 13 is a functional diagram of an example system, according to aspects of the present disclosure.

[0050] DETAILED DESCRIPTION

[0051] In the description that follows and in the figures, certain aspects are described. However, it will be appreciated that the disclosure is not limited to the aspects that are described and that some aspects mayABTLLI-128not include all of the features that are described below. It will be evident, however, that various modifications and changes may be made herein without departing from the broader spirit and scope of the disclosure as set forth in the appended claims.

[0052] The present disclosure is generally directed to a model that is trained to determine one or more IVL parameters for preparing lesions within a blood vessel for further treatment. The model may be, for example, an artificial intelligence (Al) model, such as a machine learning (ML) model. In some examples, IVL parameters may be used to configure an IVL system. For example, when the IVL system is used to prepare the lesions, the IVL system may be configured based on the parameters determined by the model. The IVL parameters may include, for example, one or more treatment positions, a balloon size, a balloon diameter, a balloon length, a balloon pressure, a number of pulses, a number of emitters for a balloon, a shockwave energy distribution for a balloon, a pulse pattern, post-treatment vessel diameter, time for balloon to be positioned within a vessel, catheter flexibility, or the like. The model may receive, as input, imaging data of a region of interest. The imaging data may include, for example, intravascular imaging data and / or extraluminal imaging data. Intravascular imaging data may be imaging data captured during a pullback of a probe within the vessel. Intravascular imaging data may include, for example, OCT image data, IVUS image data, or the like. Extraluminal imaging data may be imaging data captured by non-invasive methods. Extraluminal imaging data may include, for example, angiography image data, CT image data, or the like.

[0053] According to some examples, the model may receive, as input, in addition to the imaging data, region of interest characteristics. The region of interest characlcrislics may include, for example, vessel tortuosity, vessel diameter, lesion location, vessel type, number of pulses, calcium density, or the like. The model may be trained to provide, as output, the determined IVL parameters.

[0054] According to some examples, the same model, or a separate model, may be trained to determine one or more imaging characteristics (which may include pathology characteristics and / or angiographic characteristics) of the blood vessel in the imaging data. Imaging characlcrislics may be, for example, characteristics of the blood vessel that are determined based on image data of the blood vessel. The imaging characteristics may include, for example, a reference vessel diameter, a minimum lumen diameter, a minimum lumen area, a stenosis diameter, or the like. The imaging characteristics may additionally or alternatively include lesion characteristics, such as a length of the lesion, a presence of calcium, an arc length of calcium, a location of the lesion, a thickness of calcium, calcification length, a calcium density, or the like. Where, the portion of the vessel corresponding to (or imaged in) the imaging data comprises a plurality of lesions, the set of characteristics may comprise respective sets of lesion characlcrislics for one or more of the plurality of lesions.

[0055] The determined imaging characteristics, along with optionally the imaging data itself and / or additional region of interest characteristics, as discussed above, may be provided as input into the model trained to determine IVL parameters configurations. According to some examples, the model may beABTLLI-128trained to determine, identify, and / or select one or more IVL parameters. In some examples, such selection may include discarding one or more determined IVL parameters based on predefined criteria of the region of interest characteristics. In other examples, such selection may include determining a confidence value for each of the IVL parameters. The IVL parameters may then be ordered or ranked based on corresponding confidence values. According to some examples, IVL parameters with a confidence value below a predefined value may be discarded. Alternatively, the IVL parameter having the highest confidence value may be selected. The model may be trained to provide, as output, the selected IVL parameters.

[0056] The output of the model may be provided for output, or display, via a graphical user interface (GUI) or the like. The output provided for display may include a representation of the output of the model, such as representations of the IVL parameters determined by the model.

[0057] According to some examples, the output provided for display may include an indication of each treatment position. The treatment position may be an annotation on a representation of imaging data. The annotation may be at a position on the representation corresponding to the predicted, recommended, and / or determined treatment position. In some examples, the indication may also be a color coded indication, with a first color corresponding to a completed treatment, and a second color corresponding to a non-completed, or suggested, treatment. A completed treatment may correspond to a treatment position that has received the determined number of pulses and / or the limit of pulses, and a non-completed or suggested treatment may correspond to a treatment position that has not yet received that determined number of pulses and / or the limit of pulses. Such indication may additionally, or alternatively, include a textual indicator, such as Zone 1, Zone 2, etc. In some examples, the textual indicator may provide an indication as to whether a treatment has been completed or not. When a treatment position is provided for output, the output may also include an indication of guide points corresponding to where marker bands, located at each end of a balloon, should be positioned during an IVL procedure. Such indication of the guide points may be a shape and / or color indicator. For example, the color indicator may be a shape positioned at opposing ends on the shape indicator of the treatment position, and which may, in some examples, be a different color than that shape indicator.

[0058] The output provided for display may also include indications of the determined IVL size, such as a balloon size, and / or an IVL dose, such as a number of pulses. Such indications of the IVL size and / or dose may be a textual indicator. In some examples, the output may also include an indication of the region of interest on one or more representations of imaging data. The representations may include, for example, an extraluminal image of the region of interest, a two-dimensional representation of the vessel, an intraluminal image, a three-dimensional representation, or the like. The extraluminal image may include, for example, an angiography image. The two-dimensional representation of the vessel may be based on the diameter values of the vessel. According to some examples, the two-dimensional representation is symmetric about a longest axis of the representation. The intraluminal image may be, for example, a cross-sectional image of the vessel captured during a pullback of a probe within the vessel. The region of interestABTLLI-128may be indicated on any one of the representations via a color indicator along a length of the representation of the vessel. In some examples, the output may also include an indication of each lesion in the region of interest. Such indication may be a shape indicator along a length of a representation of the vessel. In some examples, the output may additionally or alternatively include one or more of a balloon pressure, a number of emitters for a balloon, a shockwave energy distribution for a balloon, a pulse pattern, post-treatment vessel diameter, time for balloon to be positioned within a vessel, catheter flexibility, or the like. The outputs of the balloon pressure, number of emitters for a balloon, shockwave energy distribution for a balloon, pulse pattern, post-treatment vessel diameter, time for balloon to be positioned within a vessel, catheter flexibility, or the like may be the IVL parameters determined by the model.

[0059] The model, or models, may be trained based on training data, which may include labelled data. For example, the training data may include image data and optionally region of interest characteristics. The label may be, for example, a ground truth label based on actual region of interest characteristics and / or actual IVL parameters, such as those of previously treated patients. In some examples, the ground truth labels may be received, or determined, based on feedback data. For example, after the model is executed, the model may provide, as output, region of interest characteristics, IVL parameters, a confidence value for each IVL parameter, or the like. Based on and / or in conjunction with such determinations provided for output, the system may receive feedback data confirming or rejecting the same. For example, the system may be configured to receive inputs, such as user inputs, corresponding to the confirmation and / or rejection of a given parameter, characteristics, or the like. A label may be associated with the input data to the model, e.g., the imaging data and / or the region of interest cliaraclerislics, and provided as additional training data for the model as part of a feedback loop. In some examples, the training data may be updated and / or confirmed based on the feedback data. The updated training data may be provided as additional training data to the model. The model may be updated using the additional training data.

[0060] The IVL configuration system may quickly and efficiently provide a suitably efficacious IVL treatment procedure based on image data and, optionally, region of interest characteristics. Particularly, IVL devices deliver acoustic shockwaves via a balloon in a blood vessel to fracture lesions, without cutting or tearing the vessel. However, IVL treatments are limited in the number of pulses the IVL device can administer, both to avoid balloon rupture and also to avoid damaging the vessel walls. The model mitigates the limiting feature of IVL procedures, e.g., the limited number of pulses, by determining the most effective and efficient IVL parameters, including treatment position, to treat or perform a PCI using a single IVL device. As a result, the model maximizes the usage of the allotted number of pulses without exceeding the burden on the device itself or on the vessel walls of the patient. Moreover, by determining the parameters for an IVL procedure, the model can mitigate the need to introduce more than one IVL device to the palienl , thereby mitigating procedural risk to the patient. Following the IVL treatment of the one or more lesions, the planned further interventions, such as sl nting, may then be carried out.ABTLLI-128

[0061] Figure 1 illustrates an example method for determining one or more IVL parameters for a blood vessel 100 of a patient displaying stenosis. The IVL parameters may be subsequently used as part of an IVL procedure performed on the blood vessel 100. In this example, the IVL procedure may be performed as part of vessel preparation in anticipation of stenting or other percutaneous interventions.

[0062] As shown in Figure 1 , a region of interest 102 of the vessel 100 has one or more lesions 101. While three lesions 101 are shown, the region of interest 102 may include any number of lesions 101 such that the three lesions 101 shown is just one example and is not intended to be limiting. Each lesion 101 may be the result of either intimal calcification where calcium is deposited in the intimal layers of the arterial wall leading to vessel occlusion, or medial calcification where calcium deposits occur in the medial layers of the arterial wall substantially reducing the vessel wall’s cl asl icily and compliance. The treatment of such lesions 101 using IVL reduces the effect of the calcified deposits in each lesion 101 by breaking up or fracturing said calcified deposits in order to improve the compliance and elasticity of the vessel wall, potentially in preparation for further treatment, such as stenting.

[0063] For each lesion 101 to be treated, intravascular and / or extraluminal image data 202a is obtained corresponding to the lesions 101, which may include a portion of the vessel on either side of the lesion. The image data 202a may include images obtained using an imaging system 130. Imaging system 130 may be any device that can image a blood vessel. In examples in which the image data is extraluminal, the imaging system 130 may include, for example, a computer tomography (CT) scanner, a ma netic resonance imaging (MRI) scanner, an ultrasound probe, fluoroscopy, a combination thereof, or the like. In examples in which the image data is intravascular, the imaging system 130 may include, for example, an optical coherence tomography (OCT) probe, an intravascular ultrasound (IVUS) catheter, a micro-OCT probe, a near infrared spectroscopy (NIRS) sensor, an optical frequency domain imaging (OFDI), a combination thereof, or the like.

[0064] In some examples, the imaging system 130 may include a pressure wire, a flow meter, etc. The imaging system 130 may include a device tip, one or more radiopaque markers, an optical fiber, a torque wire, or the like. Additionally, the device tip may include one or more data collecting subsystems such as an optical beam director, an acoustic beam director, a pressure detector sensor, other transducers or detectors, and combi nations of the foregoing. In examples where the imaging system 130 includes a probe 132, such as an ophcal beam director, the optical fiber may be in optical communication with the imaging system 130 and / or with the beam director. The torque wire may define a bore in which an optical fiber is disposed. According to some examples, the imaging system 130 may include the sheath such as a polymer sheath, which forms part of a catheter. The optical fiber, which in the context of an intravascular system is a portion of the sample arm of an interferometer, may be optically coupled to IVL configuration system 200 and / or a patient interface unit (PIU).ABTLLI-128

[0065] A guide wire, not shown, may be used to introduce a portion of the imaging system 130 into the blood vessel. In examples where the imaging system 130 includes an intravascular data collection device, the device may be introduced and pulled back along a length of a blood vessel while collecting data.

[0066] In some examples, the image data 202a may be a set of intravascular and / or extraluminal images of the region of interest 102, which may include each lesion 101 and the vessel on either side of said lesion 101. In some examples, the image data 202a may be received as input into a model trained to determine imaging characteristics, such as vessel characteristics and / or lesion cliaraclcrislics. The vessel characteristics may include one or more of a reference vessel diameter, a minimum lumen diameter, a minimum lumen area, a stenosis diameter, or the like. The lesion characteristics may include a length of a lesion, a presence of calcium, an arc length of calcium, a location of the lesion, a thickness of calcium, a calcification length, a calcium density, or the like.

[0067] The image data 202a may be provided as input to an IVL configuration system 200. The IVL configuration system 200 may, in some examples, receive, as input, one or more region of interest characteristics associated with the region of interest 102. Such region of interest characteristics may include lesion location and / or vessel tortuosity of the vessel in the vicinity of said lesion. The IVL configuration system 200 may include one or more models trained to determine one or more IVL parameters for treating said lesions 101. The IVL parameters may include one or more treatment positions or locations within the vessel. Each treatment posilion may be a posilion or location where the IVL device should be placed in the vessel and subsequently triggered. As such, each treatment posilion may be a position at which a particular number or dose of IVL shockwaves or pulses should be applied, delivered, or administered. It should be understood that a given IVL procedure may comprise multiple separate positions at which IVL pulses are delivered. This may allow for the treatment of multiple lesions in a region of interest, and / or the treatment of a given lesion which extends across multiple treatment positions.

[0068] For each treatment position specified by or present in the determined IVL parameters (or configuration), the IVL parameters may further include a determination and / or recommendation defining how the IVL treatment of said lesion 101 at the treatment position should be carried out. For example, the IVL parameters may, in addition to or as an alternative of the treatment positions, comprise any of a balloon size, a balloon diameter, a balloon length, a balloon pressure, a number of pulses, a number of emitters for a balloon, a shockwave energy distribution for a balloon, a pulse pattern, post-treatment vessel diameter, time for balloon to be positioned within a vessel, catheter flexibility, or the like. In some examples, balloon pressure may include and / or correspond to an inflation pressure. In another example, a number of pulses may include and / or correspond to an IVL dose. The model may provide for output the determined IVL parameters to be used with an appropriately configured IVL system 140. Examples of such IVL systems are set out in WO 2024 / 138212 Al along with WO 2024 / 102896 A2. According to some examples, the outputs of the model may be provided as output, via a display, to a user, such as a physician.ABTLLI-128

[0069] It will be appreciated that in effect the IVL configuration system 200 may execute one or more trained machine learning models based on, for example, the image data received as input to determine an IVL configuration (or protocol) for treating one or more lesions in the region of interest (including one or more treatment sites or treatment positions). The IVL configuration (or protocol) may specify (or comprise) one or more treatment positions (or locations) within the vessel. As such, one or more treatment positions may be an IVL parameter determined by the IVL configuration system 200. As a single IVL procedure may comprise multiple separate positions at which IVL pulses are delivered, an IVL configuration may comprise multiple such positions.

[0070] For each treatment position specified by (or present in) the IVL configuration, the IVL configuration may comprise one or more IVL parameters defining how the IVL treatment of said lesion at the treatment position should be carried out. Using such a configuration a surgeon may then carry out the IVL procedure in accordance with the treatment position(s) and the calculated IVL parameters, using an appropriately configured IVL system 140. Whilst in the discussion that follows the term IVL parameter(s) is used throughout, it will be appreciated that this discussion applies equally to IVL configurations comprising such parameters as set out in the preceding paragraphs.

[0071] The effectiveness of the determined IVL parameter (or an IVL configuration) may be ascertained by further imaging. In particular, following the IVL procedure a further set of image data 202a may be acquired for each lesion 101 in the region of interest 102. A change in one or more vessel and / or lesion characteristics between the initial image data, which may be pre-IVL image data, and the further image data, which may be post-IVL image data, may be used to characterize the success of the IVL procedure. Additionally, or alternatively, such further imaging may be carried out following the further treatment procedure, such as stenting, and the overall success of the IVL procedure and the further procedure may be calculated. As the IVL configuration system 200 uses one or more models trained to determine the recommended IVL parameters, pre-IVL imaging data, such as image data 202a, for a lesion 101, and / or the IVL parameters used to treat the lesion and labelled by the measure of success may be used to further train or update the one or more models to improve the efficacy of the IVL parameters generated by the IVL configuration system 200. In other words, the IVL configuration system 200 may use the outcomes of the procedures of the IVL parameters (or configurations) it generates to continuously train the one or more models. In this way, the IVL configuration system 200 can quickly and efficiently refine the recommended IVL parameters, thereby enabling efficacious IVL parameters to be recommended for a wide variety of different lesions 101 and vessel types rather than relying on the individual know how of an individual surgeon.

[0072] Example Systems

[0073] Figure 2 depicts a block diagram of an example IVL configuration system 200, which can be implemented on one or more computing devices. The IVL configuration system 200 may be configured to determine IVL parameters. The IVL configuration system 200 may include a IVL parameter subsystemABTLLI-128206 and optionally a selection subsystem 208, both of which may be implemented in a system configured to execute one or more algorithms or instructions. According to some examples, each subsystem may include respective processors, memory, data, instructions, etc. configured to implement the task and / or processes of the given subsystem. While shown as separate subsystems, the subsystems 206, 208 may be a single subsystem configured to perform the tasks and / or processes described with respect to the separate subsystems 206, 208. Accordingly, the example described above and herein illustrating the subsystems being separate subsystems within the IVL configuration system 200 is just one example and is not intended to be limiting.

[0074] The IVL configuration system 200 may receive the input data 202 as input through an application programming interface (API) coupling the IVL configuration system 200 to one or more computing devices, through a storage system, such as remote storage connected to the one or more computing devices over a network, and / or as input through a user interface on one or more computing devices coupled to the IVL configuration system 200. In some examples, the one or more computing devices are computer 1020 and / or the storage system is storage system 1040, as described in connection with Figure 5.

[0075] The input data 102 may be image data 202a and, optionally, region of interest characteristics 202b. The input data 102 may have been previously obtained and stored in a storage system, such as storage system 1040 described in connection with Figure 5, upon authorization from the patient and in compliance with any applicable privacy regulations.

[0076] As discussed in connection with Figure 1, image data 202a may include data corresponding to a lesion 101 or a suspected lesion 101 within a region of interest 102 of a vessel 100. Additionally, or alternatively, image data 202a may include a plurality of lesions 101 in the region of interest 102 of the vessel 100. For example, image data 202a may correspond to a continuous length of a vessel 100 having a plurality of lesions 101. The image data 202a may be collected using intravascular and / or extraluminal imaging systems, such as the imaging system 130 detailed in connection with Figure 1. In some examples, the imaging data 202a is OCT imaging data, and specifically, a sequence of OCT images generated by way of an OCT pullback through the vessel 100. However, as detailed in connection with Figure 1, other intravascular and / or extraluminal imaging systems may be used, such as IVUS, micro-OCT, NIRS, OFDI, CT, MRI, angiography, a combination thereof, or the like.

[0077] In some examples, the IVL configuration system 200 may receive image data 202a directly from the intravascular imaging system. For example, an imaging system may stream image data 202a to the IVL configuration system 200. Additionally, or alternatively, the IVL configuration system 200 may be configured to retrieve, receive, or otherwise obtain, the image data 202a from a storage device. Such image data 202a may have been generated from a previously completed imaging process.

[0078] The IVL configuration system 200 may receive, as input, the region of interest characteristics 202b. Additionally, or alternatively, as discussed in connection with Figure 3, the IVL configuration system 200 may receive region of interest characteristics 202b as metadata alongside, associated with, or otherwiseABTLLI-128comprised in, the image data 202a. A region of interest may be a portion of the vessel 100 where the IVL procedure and, in some cases, further percutaneous intervention, will be performed. In some examples, the region of interest may be determined automatically. In some examples, the region of interest may be determined based on one or more user inputs received identifying a proximal and distal reference of the region of interest.

[0079] In examples that include region of interest characteristics 202b as input data 202, the region of interest characlcrislics 202b may correspond to a vessel tortuosity, a vessel diameter, a lesion localion, a vessel type, a number of previously administered pulses in a previous IVL procedure or during an IVL procedure, a calcium density, a type of calcium, or the like. The type of calcium may be, for example, inliinal, medial, adventitial, nodules, eccentric, concentric, etc.

[0080] Vessel tortuosity, for example, may refer to a degree to which a vessel deviates from a straight path. Vessel tortuosity may be classified as mild, moderate, or severe based on imaging data, such as angiographic imaging data or coronary CTA imaging data. A mild classification may be less than three bends that are 45° or more. A moderate class! 11 cal ion may be three or more bends that range between 45° to 90°. A severe classification may be two or more consecutive 180° turns or may be extreme angulation more than 90°. Additionally, or alternatively, vessel tortuosity can be quantified using a Tortuosity Index, which is a ratio of actual vessel length to straight-line distance or sum of bend angles.

[0081] Lesion localion, for example, may refer to the location a lesion within the vessel, such as whether the lesion is located in an intimal, a superficial, a medial, or a deep portion of the vessel. An inliinal lesion may be the innermost layer of a vessel, a superficial lesion may be near the surface of the vessel, a medial lesion may be a middle layer of the vessel, and a deep lesion may be farther from the surface of the vessel. The location of the lesion may be determined from angiography system or coronary CTA. Alternatively, or additionally, as detailed in connection with Figure 3, the vessel diameter may be determined based on the image data 202a, such as OCT or IVUS imaging data.

[0082] Vessel type, for example, may refer to a type and / or location of the vessel. For example, the vessel type may be a coronary artery, which supplies blood to the heart, or peripheral artery, which supply blood to limbs and other extremities. Coronary arteries may include proximal left anterior descending (LAD) coronary artery, distal LAD, left circumflex artery (LCx), right circumflex artery (RCX), right coronary artery (RCA), and / or septal perforator arteries. Peripheral arteries may include femoral artery, popliteal artery, anterior libial artery, posterior tibial artery, and / or dorsalis pedis artery.

[0083] Vessel diameter, for example, may refer to one or more vessel diameters in the region of interest. The vessel diameters may be determined based on the imaging data, such as angiographic imaging data. For example, the IVL configuration system 200 and / or another system in communication with the IVL configuration system 200 may be configured to determine vessel diameters along the region of interest. In some examples, the vessel diameters may be used to generate one or more representations of the vessel. For example, the vessel diameters may be used to generate a two-dimensional representation of the vessel.ABTLLI-128One axis of the two-dimensional representation may be the location along the vessel and another axis may be the diameter values. The two-dimensional representation may be, for example, a graphical representation. In another example, the two-dimensional representation may be symmetrical about a longest axis of the representation. The vessel diameters input into IVL configuration system 200 may correspond to a predetermined vessel diameter, or a predetermined range of vessel diameters, associated with the location of the region of interest. For example, the system 200 may be configured to receive inputs, such as user inputs, corresponding to the vessel diameter. In examples where the vessel diameters are provided as user inputs, the vessel diameters may be region of interest characteristics 202b. Allcrnali vely, or additionally, as detailed in connection with Figure 3, the vessel diameter may be determined based on the image data 202a, such as OCT or IVUS imaging data. If the vessel diameters are determined based on the image data 202a, the vessel diameter may be imaging characteristics. The vessel diameters may be used, in part, to determine IVL parameters, such as balloon sizing, inflation pressure, and treatment positions. In some examples, the vessel diameter values may be used to generate two-dimensional representations of the vessel, such as a location versus diameter graph or representation, which can be used for visualization and planning.

[0084] According to some examples, the vessel diameters may include predicted post-procedure vessel diameters and / or actual post-procedure vessel diameters. Predicted post-procedure, or predicted post-IVL, vessel diameter may correspond to the vessel diameter that should be obtained once the IVL procedure is executed. Actual post-procedure, or actual post-IVL, vessel diameter may correspond to the vessel diameter that is determined after the IVL procedure is completed. The actual post-procedure vessel diameters may be determined based on post-procedural image data captured after the IVL procedure is completed.

[0085] The number of previously administered pulses may, for example, correspond to a number of IVL pulses administered in a previously performed IVL procedure in the same region of interest. According to some examples, the number of previously administered pulses may be used to ensure compliance with a treatment limit, such as the 80 to 300 pulse limit discussed above. In some examples, a pulse limit may be subject to a time period, so data may only include the number of pulses administered within such lime period.

[0086] Calcium density, for example, may be obtained through imaging data from a CT scan, such as a coronary computed tomography angiography (CCTA) and / or photon-counting computed tomography (PCCT), and may be measured in Hounsfield Units (HU). PCCT in parlicular may prevent blooming arlilacls and thus may better distinguish calcil'i calion over conventional techniques. Specifically, PCCT may provide spectral data and high-resolution imaging that allows for virtual non-iodine reconstructions, enabling calcium quantification during routine contrast-enhanced CCTA, without requiring a separate noncontrast scan. A higher-density calcium may correlate with more effective fracture formation during IVL,ABTLLI-128leading to improved stent expansion. As such, IVL efficacy may increase with both calcium arc and calcium density.

[0087] According to some examples, based on at least the input data 202, IVL configuration system 200 may determine IVL parameters. For example, the IVL configuration system 200 may be trained to determine IVL parameters. The IVL configuration system 200 may receive training data 204 as input through an API, from a storage system, or the like. The IVL configuration system 200 may be trained based on, at least, training data 204. In some examples, when the training data 204 includes data from feedback loop 212, the IVL configuration system 200 may be updated based on the training data 204. According to some examples, the IVL configuration system 200 may include one or more models, such as Al or machine learning models, that are trained to determine IVL parameters. The training data 204 may be used to train the models.

[0088] Similar to the input data 202, the training data 204 may include image data, such as image data 202a, and / or region of interest characteristics, such as region of interest characteristics 202b. The training data 204 may, in some examples, include IVL parameters. In some examples, a label corresponding to a confirmation or rejection of the IVL parameters may be associated with the input data 202, e.g., the image data and / or region of interest characteristics. For example, the label may be the ground truth label corresponding to the IVL parameters of a completed IVL procedure for the imaged region of interest. The IVL parameters may include, for example, treatment positions, a balloon size, a balloon diameter, a balloon length, a balloon pressure, a number of pulses, a number of emitters for a balloon, a shockwave energy distribution for a balloon, a pulse pattern, post-treatment vessel diameter, time for balloon to be positioned within a vessel, catheter flexibility, or the like.

[0089] The training data 204 may be split into a training set, a validation set, and / or a testing set, and may be in any form suitable for training a model, according to one of a variety of different learning techniques. For example, the training data may include multiple training examples that can be received as input by a model.

[0090] The input data 202 and / or training data 204 may be received as input into the IVL configuration system 200. In some examples, the input data 202 and / or training data 204 may be received as input into one or more of a given subsystem within the IVL configuration system 200, e.g. the IVL parameter subsystem 206 and optionally the selection subsystem 208.IVL Parameter Subsystem

[0091] The IVL parameter subsystem 206 may include and / or be configured to execute a model trained to determine one or more IVL parameters as output data 212, or, alternatively, as input data for the selection subsystem 208. While illustrated as a separate subsystem and, therefore, model within the IVL configuration system 200, the subsystem 206 and the corresponding model may be a given layer within the overall model trained to determine IVL parameters. For example, the model to determine IVL parameters may be a first layer and the model to select IVL parameters from the determined IVL parameters may be aABTLLI-128second layer. Each layer may be part of a larger model, which includes both layers. The models may be executed sequentially and / or in parallel.

[0092] In some examples, input data 202 may be received, as input, into a model trained to determine IVL parameters. As detailed above, input data 202 may include image data 202a and, optionally, region of interest characteristics 202b.

[0093] In examples in which input data 202 includes region of interest characteristics 202a corresponding to vessel tortuosity, highly tortuous vessels can make it challenging to advance and position the IVL catheter accurately. The bends and curves in the vessel can also affect the transmission of shockwaves. Intermediate-to-severe tortuosity, or moderate-severe grades, is correlated with increased procedural risk, including challenges in wiring, balloon and / or stent delivery, and higher PCI adverse outcomes. According to some examples, highly tortuous vessels may require smaller balloons and flexible catheters to navigate bends and maintain effective energy delivery. The model may determine, as one or more of the IVL parameters, catheter flexibility, a balloon size, balloon pressure, pulse pattern, etc., based on the vessel.

[0094] Catheter flexibility may include enhanced flexibility, e.g., improved shaft torque response and softer distal segment. The balloon size may include a smaller balloon diameter than the reference vessel size. For example, the smaller balloon diameter may be a balloon ranging from 0.25 mm to 0.5 mm smaller than the measured reference diameter. The balloon size may also include a shorter balloon length, such as a length ranging from 12 mm to 20 mm, to reduce friction and improve deliverability. Balloon pressure may be adjusted downward, such as a balloon pressure of about 3 atm to about 4 atm. A lower balloon pressure may be required because a balloon may only be inflated enough to contact with the calcification before administering pulses. The pulse pattern may be a lower-energy or enhanced power delivery mode to minimize stress on vessel curves. Additionally, the number of pulses and the energy profile might be adjusted to ensure effective treatment despite the challenging vessel geometry.

[0095] In examples in which input data 202 includes region of interest characteristics 202a corresponding to vessel diameter, the diameter of the vessel in a region of interest can heavily influence the appropriate balloon size and / or inflation pressure. Undersized balloons smaller than a vessel diameter may not deliver sufficient energy, while oversized balloons larger than a vessel diameter can cause vessel damage. Thus, the model may determine balloon size based on the vessel diameter, as part of the IVL parameters.

[0096] In examples in which input data 202 includes region of interest characteristics 202a corresponding to vessel type, different vessels, such as coronary arteries or peripheral arteries, have varying anatomical and physiological characlcrislics that influence IVL treatment. Since IVL stops blood flow, certain vessel types may require blood to continue flowing. As such, the determined IVL parameter may indicate that a balloon cannot be in a certain vessel type for a specific amount of time. For example, coronary arteries might require more precise control and lower energy settings compared to larger peripheral arteries. To provide the user with more precise control, one or more IVL parameters may be determined. In some examples, the IVL parameters may further determined based on vessel type, e.g., coronary or peripheral.ABTLLI-128

[0097] The IVL parameters may include, for example, balloon size, balloon pressure, number of pulses, and / or a pulse pattern. The balloon size may include a smaller balloon diameter, such as a balloon size ranging from 2.5 mm to 4 mm, and / or a shorter balloon length, such as a balloon length ranging from 12 mm to 20 mm. The balloon pressure may include a lower inflation pressure, such as a balloon pressure of about 4 atm. The number of pulses may include fewer pulses, such as a range from 80 to 120 pulses, to minimize vessel trauma. For concentric calcification, the pulse pattern may include a uniform profile. For focal lesions, the pulse pattern may include an enhanced power delivery mode.

[0098] According to some examples, for peripheral arteries, the balloon size may include a larger balloon diameter, such as a balloon diameter ranging from 4 mm to 7 mm and / or a longer balloon length, such as a balloon length ranging of 40 mm to 60 mm or more than 60 mm. In some examples, for peripheral arteries, the number of pulses may include a higher number of pulses, such as up to 300 pulses, and a balloon pressure of a standard balloon pressure, such as a balloon pressure ranging from 4 atm to 6 atm. According to some examples, treatment of peripheral arteries may include broader energy profiles and longer treatment sites.

[0099] For distal or highly tortuous segments, the balloon size may be a smaller diameter, such as a range from 0.25 to 0.5 mm below the reference vessel diameter, and / or a shorter balloon length, such as a balloon length ranging from 12 mm to 20 mm. Distal or highly tortious segment may also include a lower balloon pressure, such as a balloon pressure ranging from about 3 atm to about 4 atm, to reduce risk of vessel injury. For proximal or large-caliber vessels, the balloon size may include standard or slightly larger balloon sizes, and balloon pressure may be a higher pressure of about 4 atm to about 6 atm.

[0100] As described above and herein, the model may determine time for a balloon to be positioned within a vessel, balloon size, balloon pressure, pulse pattern, and / or the number of pulses, based on the vessel type, as part of the IVL parameters. Additionally, the model may use the vessel type to select a specialized model for a region of interest out of a plurality of models to determine characteristics in image data, as described in connection with Figure 3.

[0101] In examples in which the region of interest characteristics 202a includes a number of pulses from previous IVL procedures, the model may provide, as output, a recommended number of pulses, the previously administered number of pulses, a limit of pulses, or the initially determined number of pulses for the same region of interest. In some examples, the model may determine the recommended number of pulses based on the previously administered number of pulses, the limit of pulses, and / or the initially determined number of pulses. For example, if a difference between the limit of pulses and the previously administered number of pulses is larger than the initially determined number of pulses, then the model will adjust the recommended number of pulses so as not to exceed the limit. In such examples, the model may reduce the initially determined number of pulses to a number equal to or less than the difference between the limit of pulses and the previously administered number of pulses. If a difference between the limit of pulses and the previously administered number of pulses is equal to or smaller than the initially determinedABTLLI-128number of pulses, the model may not adjust the recommended number of pulses since the limit was not determined to be exceeded. Thus, the model may determine a recommended number to pulses based on a number of previously provided pulses in the same region, as part of the IVL parameters.

[0102] Additionally, or allcrnali vely, as an IVL procedure is occurring, the model may receive, as input, the number of pulses during the IVL procedure. For example, if there is an 80 pulse limit for the region of interest and the IVL system administers 5 pulses, the model may receive, as input, that 5 pulses were administered. The IVL configuration system may determine the number of remaining pulses based on the input number of pulses, e.g., 75 remaining pulses. In some examples, the IVL configuration system may be configured to determine one or more IVL parameters to utilize the remaining number of pulses cITccli vely within the region of interest and in view of the limit of pulses in that region, and / or the number of pulses previously administered in that region. The IVL configuration system may provide for output the remaining number of pulses, as described in connection with Figure 10. In some examples, the IVL configuration system may be configured to determine IVL parameters based on the determined remaining number of pulses.

[0103] In examples in which the region of interest characteristics 202a includes calcium density, higher-density calcium may require different treatment than lower-density calcium. Higher-density calcium may be, for example, identified as over 800 HU on CT images or may be thick arcs or calcium modi I'icalion on OCT images. Calcium modification on OCT imaging may include fracturing, cracking, crushing, or the like, of calcium. According to some examples, higher-density calcium may require a higher number of pulses, a pulse pattern of an enhanced power delivery mode for concentrated energy, and / or an energy profile including a peaked energy profile for focal nodules. As compared to higher-density calcium, lower-density calcium may require fewer pulses and / or uniform energy profiles. In some examples, calcium density may be present as intermittent calci I'icalion, such as spiral calcium, which alternates on one side of the vessel. As such, I real i ng spiral calcification can cause damage to healthy tissue, and a number of pulses available may thus be dependent on risk of perforation to healthy tissue. Additionally, as described in connection with Figure 10, when remaining pulse capacity is limited, the model may prioritize zones with a highest calcium burden to maximize therapeutic benefit. Thus, the model may determine a number of pulses, a pulse pattern, and / or an energy profile based on the vessel diameter, as part of the IVL parameters.

[0104] In some examples, the model may be executed and may provide, as output, the determined one or more IVL parameters. The determined IVL parameters may be a single IVL parameter or a plurality of IVL parameters. The output of the model may be provided for output, such as via a display screen, e.g., user output interface 1120 described in connection with Figure 5. In other examples, the determined IVL parameters may be provided as input into the selection subsystem 208 before an output is provided for display via a display screen.Selection SubsystemABTLLI-128

[0105] The selection subsystem 208 may include and / or be configured to execute a model trained to select one or more of the determined IVL parameters as output data 212. The While illustrated as a separate subsystem and, therefore, model within the IVL configuration system 200, the subsystem 208 and the corresponding model may be a given layer within the model trained to determine and select one or more IVL parameters.

[0106] In some examples, the plurality of IVL parameters, such as those provided as output from IVL parameter subsystem 206, may be received as input into a model trained to select one or more of the determined IVL parameters. In some examples, such selection may include filtering or otherwise discarding one or more of the determined plurality of IVL parameters. The IVL parameters determined from the IVL parameter subsystem 206 may be filtered or discarded, for example, based on the region of interest characteristics 202b and one or more predefined criteria. Predefined criteria for filtering or discarding IVL parameters may include anatomical, procedural, and / or device safety constraints. According to some examples, the filtering may be based on balloon diameter, balloon length, inflation pressure, pulse count, energy profile, vessel tortuosity or access, etc. For example, a determined balloon diameter may be filtered or discarded if not within ±0.5 mm of the reference vessel diameter. Additionally or alternatively, a determined balloon length may be filtered or discarded if there is excessive overlap in treatment positions, such as 50% or more overlap. Additionally or alternatively, a determined balloon pressure may be filtered or discarded if it exceeds device specifications, which typically only enables a balloon pressure of 6 atm or less. A determined balloon pressure may also be filtered or discarded if based on vessel type, since coronary arteries require lower pressure and peripheral arteries require higher pressure. Additionally or alternatively, a determined number of pulses may be filtered or discarded if it exceeds balloon and / or IVL device capacity, which typically enables a number of pulses ranges of 80 to 120 pulses for coronary arteries or up to 300 pulses for peripheral arteries. Additionally or alternatively, a determined energy profile may be filtered or discarded if it does not match calcification pattern. An energy profile may not match a calcification pattern if concentric calcium is not determined to have a uniform energy profile and / or if eccentric calcium is not determined to have a peaked energy profile. Regarding anatomical constraints, determined IVL configurations may be filtered or discarded that are incompatible with vessel tortuosity or access, such as long balloons in highly tortuous segments. Determined IVL configurations may also be filtered or discarded due to safety thresholds to prevent oversizing, ensure catheter flexibility for complex anatomy, and / or avoid overlapping treatment zones beyond predefined limits. The model may then provide for output the remaining IVL parameters that were not filtered or otherwise discarded. For example, the remaining IVL parameters may be provided as output data 210 via a display screen, GUI, or the like.

[0107] According to some examples, the filtering, discarding, and / or selection of IVL parameters may include determining a confidence value for each of the determined plurality of configurations. The confidence value may, in some examples, correspond to an indication of expected success should the IVLABTLLI-128parameter be used in an IVL procedure. For example, the predicted or expected lesion modification and / or post-PCI parameters after the IVL procedure using the IVL parameter(s) may be used to determine the confidence value in the lesion modification and / or post PCI parameters may, in some examples, indicate effective lesion modification and / or facilitates optimal stent expansion without causing vessel injury. The lesion modification and / or post-PCI parameters may include calcium fracture formation, post-IVL lumen gain, stent expansion percentage, and / or an absence of major complications. Calcium fracture formation may be at least one fracture per lesion or multiple fractures for severe calcification. Post-IVL lumen gain may be a minimum lumen area increase of at least 1 mm2. Stent expansion percentage may be 80% or more of nominal diameter post-PCI. The major complications that may not indicate a successful IVL procedure may include dissection and / or perforation. Such success metrics may be determined based on imaging data, such as OCT imaging data, IVUS imaging data, angiographic imaging data, or a combination thereof. For example, post-PCI imaging data may determine calcium fracture formation to generate a confidence value of whether the IVL treatment provided sufficient fracturing or whether a subsequent treatment is required for additional fracturing. Additionally, stent expansion percentage may be determined based on imaging data. These models may calculate the probability of achieving predefined success thresholds based on input, such as vessel diameter, calcification arc length, calcium thickness, lesion length, and prior treatment data. Confidence values may be expressed as a normalized score, such as 0-1, or percentage likelihood, and may be used to rank or filter IVL parameters. Parameters with confidence values below a predetermined threshold, such as 0.7, may be discarded, while those with higher confidence values may be prioritized for recommendation.

[0108] Once executed, the model may, in some examples, provide, as output, each of the IVL parameters and the corresponding confidence values. In some examples, the IVL parameters may be ranked or otherwise ordered based on their respective confidence values. In some examples, the model may provide for output the ranked or otherwise ordered list of IVL parameters. For example, the IVL parameters may be ranked and / or ordered from a highest confidence value to a lowest confidence value.

[0109] In some examples, IVL parameters having a confidence value below a predetermined threshold may be filtered or otherwise discarded. The model may then provide for output the remaining IVL parameters, e.g., the IVL parameters that were not filtered or otherwise discarded. In other examples, IVL parameters having a highest confidence value may be selected with all other determined IVL parameters discarded. The model may provide for output the IVL parameters having the highest confidence value.

[0110] The IVL parameters identified by the selection subsystem 208 may correspond to output data 212. In some examples, the output data 212 may be provided for output on a display screen, such as user output interface 1120 described in connection with Figure 5.

[0111] In examples in which image data 202a includes a portion of a vessel, or region of interest 102, having more than one lesion 101, the determined and / or selected IVL parameters may correspond to one or more IVL procedures, such as an IVL procedure for each of the lesions 101. For example, the IVLABTLLI-128parameters may include and / or correspond to a treatment position or location for a given IVL procedure in addition to other IVL parameters, such as treatment positions, balloon size, balloon diameter, balloon length, balloon pressure, a number of pulses, a number of emitters for a balloon, a shockwave energy distribution for a balloon, a pulse pattern, post-treatment vessel diameter, time for balloon to be positioned within a vessel, catheter flexibility, treatment positions, or the like.

[0112] The selection subsystem 208, as part of the IVL configuration system 200, may provide for quick and efficient idcnli I'icalion of IVL parameters for use with the IVL system 140. Besides the speed at which the IVL parameters can be identified using the IVL configuration system 200, the efficacy of the IVL parameters identified by the IVL configuration system may maximize the usage of the allotted number of pulses without exceeding the burden on the device itself or on the vessel walls of the patient and / or mitigate the need to introduce more than one IVL device to the patient, thereby mitigating procedural risk to the patient.

[0113] As discussed above and herein, some or all of the determined IVL parameters, e.g., output data 210, may be output by the IVL configuration system 200 to the IVL system 140. In this way, the IVL configuration system 200 may directly configure the IVL system 140 for the recommended IVL procedure. Such direct configuration may comprise selling a number of shockwaves to be triggered in a given burst for the IVL apparatus, setting a target inflation pressure, and so on. The configuration of the IVL system 140 may comprise selling one or more limits in line with the IVL parameters. For example, the IVL configuration system 200 may set any of: a maximum number of shockwaves, a maximum inflation pressure, and so on.

[0114] The IVL configuration system 200 may be arranged to receive directly or indirectly the actual IVL parameter values used with the IVL system 140 in the subsequent IVL procedure. In this way, the IVL configuration system 200 may store or record any change of parameter values with respect to the recommended IVL parameters. The IVL configuration system 200 may be arranged to label the image data 202a with the actual IVL parameter values used with the IVL system 140 in the subsequent IVL procedure and store the labeled imaging data sets for further training of the one or more models 232. For example, the labeled image data 202a, labelled with the actual IVL parameter values used with the IVL system 140, may be provided as feedback data within feedback loop 212.

[0115] It will be particularly noted that the IVL configuration system 200 may be arranged to track the number of shock waves or pulses delivered at each position when carrying out a given IVL procedure. Such tracking may be used for further training of the models. However, such tracking may also be useful for further IVL procedures at the same treatment site. For example, this may allow further IVL procedures to be carried out on the same lesions whilst not exceeding a total or cumulative dose or number of shock waves at a given position. The model may also enable modi I'icalion of the protocol during a procedure if, after the initial protocol has been followed, there is still additional capacity of the IVL balloon to deliver further pulses.ABTLLI-128

[0116] Figure 3 depicts a block diagram of another example IVL configuration system 300, which can be implemented on one or more computing devices. The IVL configuration system 300 may be configured to determine IVL parameters. The IVL configuration system 300 may be substantially similar to the IVL conl'iguralion system 200. In that regard, both the IVL conl'iguralion systems 200, 300 include an IVL parameter subsystem 206 and, optionally, a selection subsystem 208, as discussed above with respect to Figure 2. For example, IVL parameter subsystem 206 may be configured to execute a model trained to determine one or more IVL parameters as output data 312 or as input into selection subsystem 208, as described in connection with Figure 2. Selection subsystem 208 may be configured to execute a model trained to select one or more of the determined IVL parameters as output data 312, as described in connection with Figure 2. IVL configuration system 300 may differ from IVL configuration system 200 in that IVL conl'iguralion system 300 includes a characterization subsystem 314.Characterization Subsystem

[0117] The characterization subsystem 314 may include and / or be configured to execute a model trained to generate, determine, or otherwise calculate one or more imaging characteristics from image data. The determined imaging characteristics may be provided as input into the IVL parameter subsystem 206. While the subsystem 314 is illustrated as a separate subsystem and, therefore, model within the IVL conl'iguralion system 300, the subsystem 314 and the corresponding model may be a given layer within the overall model trained to determine IVL parameters.

[0118] In some examples, the input data 202, e.g., image data 202a, may be provided as input into a model trained to determine imaging characteristics in image data. In some examples, the imaging characteristics may be pathology characteristics and / or imaging characteristics associated with the blood vessel 100, region of interest 102, and / or lesions 101. Pathology characteristics may include structural and compositional features of the vessel and lesion. The IVL parameters determined by the IVL configuration system 200 may be based on, at least in part, the pathology characteristics. For example, pathology characteristics may include calcification severity and distribution, calcium density, plaque composition, lesion complexity, and / or pathological vessel changes. Calcification severity and distribution may be determined based on arc length, thickness, and / or length of calcium deposits. Calcium density may be measured in Hounsfield Units on CT imaging or inferred from OCT reflectivity. Plaque composition may include lipid-rich, fibrotic, or mixed morphology adjacent to calcium. Lesion complexity may include concentric versus eccentric calcification, bifurcation involvement, and / or proximity to side branches. Pathological vessel changes may include positive or negative remodeling, chronic total occlusion characteristics, and / or presence of thrombus. The imaging characteristics may include vessel and / or lesion characteristics for the region of interest corresponding to or imaged in the image data 202a. The vessel characteristics may include one or more of a reference vessel diameter, a minimum lumen diameter, a minimum lumen area, a stenosis diameter at the region of interest, or the like. The lesion characteristics may include one of more of a length of a lesion, a presence of calcium, an arc length of calcium, a locationABTLLI-128of a lesion, a thickness of calcium, calcification length, calcium density, or the like. In examples in which image data 202a corresponds to a plurality of lesions 101, the characteristics may comprise respective sets of lesion characteristics for each lesion 101. The pathology characlerislics and imaging characteristics may be used by the model to determine IVL parameters, including balloon size, energy profile, pulse count, and treatment positions.

[0119] In some examples, determining the lesion characlerislics may include the characterization subsystem 314 idenli lying lesions 101 depicted or present in the image data 202a. Identifying the lesions 101 may include, for example, the execution of the model trained to identify imaging characteristics. Additionally, or alternatively one or more lesions 101 may be indicated or annotated in the intravascular image data set 202a. Such indication may be carried out by a separate processing system prior to the image data 202a being provided to the IVL parameter subsystem 206, or, as described in connection with Figure 2, may be performed by manual inspection of the intravascular image data set 202a to be provided as user input.

[0120] In some examples, determining the imaging characteristics may include one or more heuristic algorithms. Such heuristics algorithms may include one or more image processing algorithms to identify various characlerislics based on areas of contrast and other features of the image data 202a. Additionally, or alternatively the characterization subsystem 314 may be arranged to use one or more trained models described in U.S. Pat. App. No. 11,819,309 (the ’309 patent), which is incorporated herein by reference in its entirety.

[0121] In some examples, one model may be trained to generate or determine both the vessel characteristics and the lesion characteristics. In other examples, distinct trained models may be used to generate or determine each of the vessel characlerislics and the lesion characteristics. In some examples, at least some the model(s) may be specialized or specifically trained for a specific lesion localion and / or vessel type.

[0122] In such cases, the characterization subsystem 314 and / or its corresponding model may select a specialized model trained to determine IVL parameters for a lesion location or type corresponding to the lesion location or type in the region of interest based on the determined imaging characteristics. For example, if the characterization subsystem 314 and / or its corresponding model determined that a region of interest in the image data corresponds to a coronary artery, the characterization subsystem 314 and / or its corresponding model further may select a model trained to determine IVL parameters for coronary arteries. Similarly, if the characterization subsystem 314 and / or its corresponding model determines that a region of interest 102 in the image data corresponds to a peripheral artery, the characterization subsystem 314 and / or its corresponding model may select a model trained to determine IVL parameters for peripheral arteries. The selected model(s) trained to determine IVL parameters may be model(s) corresponding to IVL parameter subsystem 206. In some examples, an indication of the identified model may be providedABTLLI-128to the IVL parameter subsystem 206 such that the IVL parameter subsystem 206 can execute the identified model to determine the IVL parameters.

[0123] In some examples, the determined characlcrislics may be provided as input into a model trained to determine IVL parameters. For example, the determined characteristics may be provided as input to the IVL parameter subsystem 206. The IVL parameter subsystem 206 and / or the IVL configuration system 300 may execute the model trained to determine IVL parameters. The IVL parameter subsystem 206 and / or IVL configuration system 300 may provide, as output data 312, the determined IVL parameters.

[0124] In some examples, the same or different model may be trained to select one or more IVL parameters from the determined IVL parameters. The model may be, for example, the model within and / or executed by selection subsystem 208, as described above with respect to Figure 2. The model may then provide, as output data 310, the selected IVL parameters from those determined from the model of the IVL parameter subsystem 206. The determined and / or selected IVL parameters may be output on a display screen, such as user output interface 1120 described in connection with Figure 5.

[0125] It will be appreciated that the image data 202a may comprise a plurality of images or frame) of the vessel. As such, the trained model(s) may provide an end-to-end pipeline in processing the images of the region of interest to directly generate or determine IVL parameters for an IVL procedure. As such, the trained model may be or comprise a deep neural network model, such as a convolutional neural network model.Example Method

[0126] Figure 4 illustrates a flowchart of an example method 400 of determining IVL parameters for a region of interest of a patient vessel. The IVL parameters may determined using the example IVL configuration systems 200, 300 described in connection with Figures 2 and 3. The following operations are not required to be performed in the precise order described below. Rather, various operations can be handled in a different order or simultaneously, and operations may be added or omitted.

[0127] In block 410, an initial set of image data 202a of the region of interest is received. As set out above, the image data 202a may be received directly from an intravascular imaging apparatus, such as OCT apparatus, or via an intermediate device, such as a storage device.

[0128] In block 420, at least one IVL parameter is generated based on the image data 202a. In this way an IVL parameter which may, in some examples, specify an IVL procedure suitable for the lesion 101 or lesions present in the region of interest imaged in the image data 202a may be obtained. In some examples, the IVL parameters may include parameters for configuring an IVL system for an IVL procedure. A model may be trained to generate the at least one IVL parameter. As discussed in connection with Figure 3, the model may receive the imaging data 202a as input data 202. As such, the model may provide an end-to-end processing of the image data.ABTLLI-128

[0129] Additionally, or alternatively, block 420 may also comprise block 422. Imaging characteristics are obtained or determined from the image data 202a, such as by the characterization subsystem 314. The system may use one or more heuristic processing or image processing techniques to determine at least part of the characteristics. Additionally, or alternatively the model may be trained to generate the imaging characteristics from image data 202a. The imaging characteristics may comprise a set of vessel characteristics and one or more sets of lesion characteristics for lesions in the vessel at the region of interest, as defined in connection with Figures 2 and 3.

[0130] In such examples, block 424 may also include a model trained to determine IVL parameters receiving the determined set of characteristics as input, as detailed in connection with Figure 3.

[0131] In block 430, the model may provide for output the IVL parameters. For example, the model may provide the IVL parameters as output alongside or with some or all of the image data 202a. Equally, the image data 202a may be displayed with annotations representing or showing some or all of the characteristics. Additionally, or alternatively, block 430 may comprise or consist of directly configuring a connected IVL system, in line or in accordance with the IVL parameter.

[0132] Figure 5 schematically illustrates a method 700 of training a model to determine IVL parameters for an IVL procedure. A system of one or more processors, such as the system 1050 described in connection with Figure 13, can perform the method 700. Some operations in the processes described herein can be omitted or performed multiple times, for example iteratively or in parallel. In some examples, other operations are added to the processes and / or performed in different orders.

[0133] In block 710, the system receives training data, such as imaging data 202a as described in connection with Figures 2 and 3. When the training data is imaging data, the imaging data may be captured prior to an IVL procedure and thus may represent an initial state of the respective vessel 100. The imaging data may further include annotations of labels corresponding to one or more IVL parameters of an IVL procedure performed on the region of interest in the images.

[0134] According to some examples, imaging data 202a for respective treatment sites of patient vessels are received. Each set of initial intravascular imaging data 202a corresponds to a respective IVL parameter of a respective IVL procedure performed at the respective treatment sites the initial intravascular imaging data 202a having been captured prior to the IVL procedure represents the initial state of the respective vessel 100.

[0135] Vessel data and / or patient data that can be included as features in training data examples can include one or more of the following: age, presence of hypertension, dyslipidemia, or diabetes, whether the patient is a current or former smoker, BMI, whether the patient had a prior myocardial infarction, renal insufficiency, and angina scoring from 0 to IV (4). Additional examples include characteristics of target vessels, including left anterior descending, circumflex, right coronary artery, or left main. Other examples can include the amount of severe calcification identified in the target vessel, as well as lesion location,ABTLLI-128including whether the lesion is proximal, mid, distal, or oslial. Other lesion characteristics include lesion length and calcification length. Other examples include patient historical data such as other conditions the patient may have, patient allergy data, patient medication data, or patient product-specific adverse event data including but not limited to hypersensitivity reactions, excess bleeding, and / or product performance failures. Patient historical data is stored securely and in compliance with any applicable privacy regulations. The patient historical data may be useful if a patient is receiving a subsequent procedure at the same facility or if the patient is receiving a procedure at a different facility from the facility in which the patient historical data was collected. The training data may be from cadaveric and / or living subjects. The training data examples can include a reference vessel diameter, a minimum lumen diameter, and / or a diameter stenosis. The training data may be readily available and may include, for example, pre-operative image data, historical patient, and / or indications of post-operative success or post-operative failures. In some examples, the training data may be publicly available, for example, with anonymized individual patient information for privacy.

[0136] If a percutaneous coronary intervention (PCI) was previously performed, other example features can include a reference vessel diameter, a lumen diameter, a diameter stenosis, a stent length, and / or acute gain. Post-PCI final complications can also be included as features, including severe dissection, slow flow or no reflow, abrupt closures, and / or perforations in the vessel

[0137] Other examples of features that may be provided as part of the training data, include the following, for example in reference to calcified coronary lesions identified using OCT: lesion length, minimal lumen area, mean lumen area, area stenosis, calcium length, maximum continuous calcium arc, mean calcium arc, minimum calcium thickness, and / or calcium volume index. Other examples include characteristics related to a pre-IVL at MLA sites, area stenosis percentage, lumen area, calcium angle, and / or max calcium thickness. Other examples include characteristics related to a post-stent at an MLA site, such as area stenosis, stent area, stent expansion, acute area gain, and calcium fracture. Other examples include characteristics related to pre-IVL at a maximum calcium site, such as area stenosis, lumen area, calcium angle, and / or maximum calcium thickness. Other examples include characteristics related to pre-IVL at the final minimal stent area (MSA) site, such as area stenosis, stent area, and / or stent expansion, acute area gain, and / or calcium fracture. Other examples include characteristics related to post-stent at the final MSA site, including area stenosis, stent area, stent expansion, acute area gain, and / or calcium fracture.

[0138] Other examples of features that may be provided as part of the training data include OCT characteristics for calcium fracture, such as the presence of any fracture, one fracture, two fractures, or more than three fractures, fracture length, fracture depth, maximum calcium arc at a calcium fracture, minimum calcium angle at the calcium fracture, calcium thickness at the calcium fracture, calcium fractures per lesion, calcium fracture per mm, lumen gain at fracture site, lumen area at fracture site, and / or stent expansion percentage. Other examples of features that may be provided as part of the training data include characteristics for OCT symmetry, eccentricity, and malapposition characteristics.ABTLLI-128

[0139] Other examples of features that may be provided include whether the patient has stable ischemic heart disease, acute coronary syndrome, non-ST-elevation myocardial infarction (NSTEMI), unstable angina, or stabilized recent ST-elevation myocardial infarction (STEMI).

[0140] The training data may be received from a data store or database. For example, the system may receive training data including examples of OCT image data annotated with labels corresponding to an IVL procedure performed on the vessel characterized in the OCT image data. The training data can include OCT image data as described herein, which may be stored in a database, along with labels annotating the configurations of IVL procedure that was performed, for example on a padent from whom the OCT image data was taken. The annotations may be, in some examples, an indication of one or more of a vessel tortuosity, vessel diameter, lesion location, vessel type, number of pulses, calcium density, number of cracks, a reference vessel diameter, a minimum lumen diameter, a minimum lumen area, a stenosis diameter, a length of the lesion, a presence of calcium, an arc length of calcium, a location of the lesion, a thickness of calcium, calcification length, proximal frames, distal frames, or the like. In some examples, the annotations may be related and / or associated with the determined IVL parameter. For example, the system may receive an input corresponding to as a balloon size, a balloon diameter, a balloon length, a balloon pressure or an inflation pressure, a number of pulses or an IVL dose, a number of emitters for a balloon, a shockwave energy distribution for a balloon, a pulse pattern, post-treatment vessel diameter, time for balloon to be positioned within a vessel, catheter flexibility, or the like. The database can be managed and updated by the system, which in turn can provide for additional training examples for further training or fine-tuning the model.

[0141] Other examples of features that may be provided include data corresponding to a number of pulses to complete cracking hardened calcium in a vessel using an IVL procedure. In some examples, such training data may be obtained from an imaging system so as to detect cracks and crack propagation in a region of interest. For example, the data may be CT and / or OCT image data, or may be micro-CT image data obtained from a cadaver.

[0142] In block 720, for each initial set of imaging data 202a, a measure of success for the respective IVL procedure is obtained. Suitable measures of success are discussed shortly below. However, it will be appreciated that these may be determined from further intravascular imaging, such as post-operative imaging, of the respective patient. Similarly, it will be understood that such measures of success may have been calculated or determined as part of the treatment of the respective patient. As such, the measures of success may be received alongside or as part of the initial sets of intravascular imaging data 202a.

[0143] In block 730, the model is trained according to the plurality of initial sets of imaging data 202a and the respective measures of success. In examples in which the model receives imaging data directly as input, block 730 may comprise annotating each initial set of imaging data with the respective IVL parameter and the respective measure of success. Alternatively, in examples in which the model is trained to determine vessel and or lesion characteristics, such as in the example discussed above in relation to Figure 3, blockABTLLI-128730 may comprise, for each initial set of image data, determining a set of imaging characteristics. Block 730 may further comprise annotating each set of characteristics with the respective IVL parameter and the respective measure of success, and training the model based on the plurality of annotated sets of characteristics. It will be appreciated that the sets of imaging characteristics may be obtained by a characterization subsystem 320, such as that described above in relation to Figure 3.

[0144] It will be appreciated that the IVL procedure carried out at the treatment site, based on the determined IVL parameters, may often be done to prepare the treatment site for a further i nlervenlion. Such further intervention may include the insertion of a stent. As such, when training the model(s) discussed herein, the measure of success used for the recommended IVL parameters may be measured post further intervention, such as post stenting. One such measure of success is stent expansion, which may be measured as a percentage of full expansion. In this case, the measure of success may be proportional to the stent expansion. Here, the model(s) would be trained to determine IVL parameters with the aim of maximizing stent expansion. Alternatively, the measure of success could be a binary measure, with stent expansion below a pre-determined threshold, referred to as under expanded stents, causing the IVL parameter and corresponding imaging data or imaging characteristics in the training data to be labelled as a fail. Stent expansion at or above the pre-determined threshold would cause the IVL parameter and corresponding intravascular imaging data or imaging characteristics to be labelled as a pass.

[0145] Similarly, the minimum lumen area post further intervention may be used as a measure of success. The difference between the minimum lumen area pre- IVL treatment and the minimum lumen area post stenting may be used. Again, the measure of success may be directly proportional to the minimum lumen area, or difference thereof, or may be a binary measure of whether the minimum lumen area, or difference thereof, is above or below a pre-determined success threshold.

[0146] By using such post further intervention measures of success training data is readily available from existing data on historical IVL procedures as post further intervention imaging from which the above measures can be readily derived is routinely carried out.

[0147] It will be understood that the measure of success may also be made after the IVL treatment. This may be useful in cases where no further intervention is planned. It also has the advantage that any confounding effects of the choices made during the further i nlervenlion are avoided. The minimum lumen area measure of success can in such cases be the difference between the minimum lumen area pre-IVL treatment and the minimum lumen area post-IVL treatment. This may be derived or calculated by calculating the minimum lumen area from post-IVL treatment intravascular images of the treatment site, using the techniques described above as applied to the pre-IVL procedure intravascular image data set.

[0148] The measure of success may be or comprise measure of calcium fracturing. Intravascular imaging may be used to determine a degree of fracturing of a given lesion post IVL treatment. More fracturing may indicate a greater success of the IVL treatment.ABTLLI-128

[0149] In some examples, the trained model can be deployed in the system, so that when a new OCT pullback is conducted, the system can process the new data through the model for providing a rccoinmcndalion on the most appropriate preparation method or technique, or a plurality of options from which to choose. In some examples, the OCT system can include a display device onto which any lesion type information can be displayed for operator review and / or any vessel modi I'icalion or treatment regimen that is recommended or provided as an option.

[0150] Figure 6 schematically illustrates a variant method 800 of specifying an IVL procedure for a treatment site of a patient vessel. The variant method 800 is a variant of the method 400 described above in relation to Figure 4. The steps of the variant method 800 are the same as the steps of the method 400 described above, except where noted below. In the variant method 800 the model is updated based on the outcome of the IVL procedures carried out subsequent to the recommended IVL parameter being provided.

[0151] In block 840, the model receives a further set of image data, such as intravascular image data, of the region of interest. The further image data may be received directly from an imaging apparatus, such as an OCT apparatus, or via an intermediate device, such as a storage device. The further set of image data is generated or captured following an IVL procedure at the region of interest using an IVL parameter. The IVL parameter may be the recommended IVL parameter provided in block 430. However, it will be appreciated that the IVL parameter may differ from the recommended IVL parameter. For example, the system may enable a user to override and / or change one or more IVL parameters specified in the recommended IVL parameters.

[0152] In block 850, a measure of success of the IVL procedure is determined based on the further set of image data. As discussed above, a number of different measures of success may be used. The measure of success may be determined based on a comparison of or difference between the initial set of intravascular image data received in block 420 and the further set of intravascular image data. The comparison may comprise determining one or more imaging characteristics from the initial set of intravascular image data received in block 420 and the further set of image data.

[0153] In block 860, the trained model used to generate the at least one IVL parameter in block 420 is updated based on the initial set of image data, the IVL parameter used for the IVL procedure and the measure of success. The update may be carried out in the same manner as the initial training of the model as set out above in relation to Figure 5. It will be appreciated that the measure of success used in the method 800 may differ from the measure of success used in the method 700 of initial training. For example, the measure of success for the method 700 of initial training may be a post further procedure measure of success, such as stent expansion. This may allow the model to be trained using historical data from existing stenting procedures where post IVL but pre stenting imaging is not carried out. The measure of success used in the updating and / or ongoing learning process of the method 800 may instead use measures of success determined from post IVL intravascular imaging, which may be performed as part of the treatment protocol in future cases, such as the calcium fracturing measures or the minimum lumen diameter measureABTLLI-128discussed above. In this way the system of the disclosure can be introduced trained on existing procedure and outcome data, allowing a baseline of IVL parameter recommendations consistent with current good practice to be generated. The system may then refine the model in use using the more direct measures of success of the IVL procedure alone, allowing IVL parameter recommendations of improved quality to be provided.

[0154] Figure 7 illustrates a flowchart for an example method 1500 of determining IVL parameters. The following operations are not required to be performed in the precise order described below. Rather, various operations can be handled in a different order or simultaneously, and operations may be added or omitted.

[0155] In block 1502, input data, such as input data 202 including imaging data 202a and optionally region of interest characteristics 202b as described in connection with Figure 2, may be received as input into a model trained to determine one or more IVL parameters, The IVL parameters may be a treatment site, a balloon size, a balloon diameter, a balloon length, a balloon pressure, a number of pulses, a number of emitters for a balloon, a shockwave energy distribution for a balloon, a pulse pattern, post-treatment vessel diameter, time for balloon to be positioned within a vessel, catheter flexibility, or the like.

[0156] In block 1504, the model may be executed to determine IVL parameters. Such determination may include an IVL parameter subsystem, such as IVL parameter subsystem 206 detailed in connection with Figures 2 and 3, which may include and / or be configured to execute a model trained to determine one or more IVL parameters. In some examples, such determination may first include a characlerislics subsystem, such as characteristic subsystem 314 detailed in connection with Figure 3, which may include and / or be configured to execute a model trained to determine imaging characlerislics for use in the IVL parameter subsystem. Such determination may also lastly include a selection subsystem, such as selection subsystem 208 detailed in connection with Figures 2 and 3, which may include and / or be configured to execute a model trained to select one or more of the determined IVL parameters.

[0157] In block 1506, the model may provide for output an indication of the IVL parameters. The output may be on a display screen, such as user output interface 1120 described in connection with Figure 5.

[0158] As a single IVL procedure may be used to treat multiple lesions and / or lesions with complex structures Across a variety of different vessel sizes and locations, there are extensive options for how the procedure is carried out. Each option, or parameter, may have a direct impact on the efficacy of the treatment as a whole. In particular, there are numerous parameters that together specify the particular IVL procedure to be carried out. The IVL parameters determined by IVL configuration systems 200, 300, discussed above, may comprise or be specified by any combination of such IVL parameters.

[0159] In some examples, IVL parameters may include balloon size for the IVL apparatus. As will be understood, the IVL procedure involves placing an IVL shockwave emitter device disposed in a balloon on a guidewire in the vessel to be treated. The shock or pressure waves may be generated by the emitters vaporizing a portion of fluid in the balloon, with the remaining fluid in the balloon acting as a medium for transmission of the pressure wave to the walls of the vessel and therefore the calcareous lesions itself. AsABTLLI-128such appropriate sizing of the balloon is important to ensure an effective treatment. A balloon that is too small will not allow for the pressure waves to be effectively transmitted into the walls of the vessel. Similarly, a balloon that is too large risks damaging the vessel once the balloon is inflated, as the inflation pressure will be constrained by the vessel wall and not by the balloon itself.

[0160] In some examples, balloon size may include the inflated balloon diameter, the balloon length, and / or shock wave emitter arrangement. IVL balloons and the corresponding emitters may be offered in different lengths. As would be understood, the length of the lesion itself strongly influences the length of balloon.

[0161] In some examples, IVL parameters may include balloon pressure of the fluid in the balloon during treatment affects how effectively the pressure waves are transmitted to the walls of the vessel. An inflation value may be around 4 atmospheres or around 405 kPa. However, under or overinflalion of the balloon may be employed to adjust the energy transfer.

[0162] In some examples, IVL parameters may include a number of emitters for a balloon and / or a shockwave energy distribution for a balloon, as shown in Figure 8. An IVL device may comprise a plurality of shockwave or pressure wave emitters which are triggered simultaneously. The arrangement and strengths of the emitters determine the overall energy distribution along the device. As will be appreciated pressure waves emitted from different emitters in the IVL device may constructively and / or destructively interfere. This can give rise to complex energy profiles depending on the IVL emitter arrangement used. Figure 8 illustrates two different IVL emitter arrangements 610; 620. In the first emitter arrangement 610 five emitters 601 provide an energy or pressure wave profile 615 having five distinct peaks of intensity. An example of such an arrangement is the M5+ IVL device manufactured by Shockwave Medical Inc., Santa Clara, CA, USA. Such a peaked energy profile may be suited for segments, lesion, and / or treatment positions with focal / high-density calcification. The middle of the projected pressure profile may be placed nearest the peak calcil'icalion. This ensures that the maximum energy is delivered to the most calcified area, enhancing the effectiveness of the treatment.

[0163] In the second emitter arrangement 620, six emitters 601 provide an almost flat energy or pressure wave profile 625. An example of such an arrangement is the L6 IVL device manufactured by Shockwave Medical Inc., Santa Clara, CA, USA. Such a flat energy profile is often more suitable for segments with uniform calcification. Here the energy is evenly delivered across the calcified area.

[0164] As will be appreciated by selecting different emitter arrangements, the pressure wave energy delivered to particular parts of the vessel will change, thus altering the treatment. As such an IVL parameter may be energy profile of the device, and / or the emitter arrangement. An emitter arrangement may be selected from a number of pre-defined available emitter arrangements rather than being wholly adjustable on a given device.

[0165] In some examples, the IVL configuration system may be in communication with an IVL system that includes different power delivery modes. For example, one such IVL system may include a standardABTLLI-128power delivery mode but also include an enhanced power delivery mode. When operating in the enhanced power delivery mode, the IVL device may be configured to more effectively and / or efficiently crack the calcium in the vessel. An example of technology suitable for delivering such modes is disclosed in published U.S. Pat. App. Nos. 2025 / 0261957, 2024 / 0156476, 2024 / 0156477, and 2025 / 0228580, the entire contents of which are incorporated by reference. An IVL system with an enhanced power delivery mode may be capable of increasing or altering voltage, timing, duty cycle, intensity, duration, power, frequency and / or shape of the signal delivered to a specific electrode pair or plurality of pairs for creating spark(s). Similarly, an IVL system with a laser-based shockwave generation system may also have multiple operating modes, including multiple power delivery modes. In one aspect, the IVL configuration system may determine IVL parameters based on the operational modes of the IVL system. The IVL configuration system may output as an IVL parameter a power delivery mode, e.g., enhanced power delivery model, for a region of interest.

[0166] In some examples, the IVL parameters may include a number of pulses or shocks delivered in a region of interest. The number of pulses may change the efficacy of the treatment as a whole, with larger lesions with more calcium requiring more shocks in order to provide effective fracturing.

[0167] In some examples, the IVL parameters may include a pulse pattern. The IVL configuration system 200 may recommend the pulse pattern based on lesion morphology, calcification density, vessel type, etc. The pulse pattern may a continuous single-pulse mode or an enhanced power delivery mode. The pulse pattern may be determined based on lesion morphology, calcification density, and / or vessel type.

[0168] The continuous single-pulse mode may be a continuous pulse pattern, in which single pulses are delivered sequentially at regular intervals. This continuous single-pulse mode may provide precise control for uniform or concentric calcification, such as in coronary arteries.

[0169] The enhanced power delivery mode may include delivering clusters of pulses, such as three to five pulses per burst, followed by short pauses, such as one to two seconds, to allow balloon pressure stabilization and reduce mechanical stress. The enhanced power delivery mode may reduce the overall number of pulses provided to a region of interest and may increase the likelihood of safe, efficient calcium cracking without undue tissue damage. Thus, the enhanced power delivery mode may improve energy efficiency for dense or focal calcification by concentrating IVL acoustic energy waves while minimizing catheter fatigue.

[0170] As discussed above and herein, an IVL parameter for a given IVL procedure may comprise a plurality of IVL treatment positions. Each treatment position is a position or location where the IVL device should be placed in the vessel and subsequently triggered. As such different each treatment position may include additional IVL parameters for the given IVL procedure. Moreover, the treatment positions may be chosen to provide overlap between adjacent treatment positions. This may enable the treatment of lesions that are longer than the IVL device.ABTLLI-128

[0171] It will also be understood that the above IVL parameters can be interdependent or non-orthogonal. In particular, each of the above IVL parameters effects the amount of energy delivered to a given position on the vessel wall. For a parlicular amount of energy, there may be multiple different combinations of parameter values that will achieve the same energy delivery.

[0172] In some examples, the system may determine one or more IVL parameters based on the type of calcium. For example, the type of calcium may be calcium nodules, which are focal, protruding calcific deposits often associated with severe eccentric calcification. Calcium nodules may be identified from intravascular imaging data, such as OCT, based on features including high backscatter with signal attenuation, protrusion into the lumen, and localized thick calcium caps. For regions containing calcium nodules, the IVL parameters may include an emitter alignment, a pulse pattern, a balloon size, a balloon pressure, treatment positions, and / or a number of pulses.

[0173] Emitter alignment for a calcium nodule may be a position of the IVL balloon that one or more emitters are aligned with the nodule to maximize localized energy delivery. The alignment may be guided by marker-band positioning and confirmed via imaging overlays. In one mode, the determined emitter alignment may include aligning the emitter directly with a calcium nodule to concentrate IVL acoustic wave energy at the focal point, thereby maximizing fracture formation in dense calcified regions. In another mode, the determined emitter alignment may include dynamic balloon movement across the lesion while triggering pulses in a sequence designed to leverage constructive interference between emitters, creating overlapping IVL acoustic wave energy zones for broader energy distribution. These modes account for eccentric versus concentric calcification patterns and may be determined based on imaging characteristics, such as calcium arc length, thickness, and / or lesion geometry.

[0174] The pulse pattern for a calcium nodule may include an enhanced energy mode, such as a crack mode, which may be a specialized pulse pattern configured to intensify energy delivery at the nodule site. The pulse pattern may include short, high-frequency bursts of pulses, timing offsets between emitters to create constructive interference at the nodule location, and / or dynamic modulation of energy distribution to concentrate acoustic pressure on the protruding calcium. The short, high-frequency bursts of pulses may be clusters of three to five pulses with inter-burst cooling intervals.

[0175] The balloon size for a calcium nodule may include slight undersizing of the balloon relative to a reference vessel diameter. Such undersizing may improve coupling and reduce risk of vessel injury.

[0176] The balloon pressure for a calcium nodule may be adjusted within safe limits, e.g., ±0.5 atm from nominal, to optimize energy transfer without compromising balloon integrity.

[0177] The treatment posilions for a calcium nodule may include a threshold amount of overlap zones to ensure complete coverage of the nodule while avoiding unnecessary energy delivery to adjacent healthy tissue.ABTLUI-128

[0178] The number of pulses for a calcium nodule at the nodule site may be capped, or limited, based on device capacity and safety thresholds.

[0179] To determine whether a treatment of a calcium nodule is successful, the model may receive as input post-treatment imaging to determine fracture presence, fracture length and / or depth, lumen gain at the nodule site, and / or predicted improvement in minimum stent area (MSA) if slenling is planned. These metrics may be used to update the model as part of a feedback loop.

[0180] In some examples, a graphical user interface (GUI) may provide visual indicators for calcium nodules and display the enhanced energy mode activation status, recommended pulse pattern, and / or emitter alignment guide points for marker bands.

[0181] As discussed above, the IVL configuration system 200 determines, or recommends, IVL parameters based on imaging data of the region of interest. This can be achieved as the model can accurately determine vessel and / or lesion characteristics based on the imaging data.

[0182] For example, intravascular imaging data can allow for accurate determination of vessel characteristics. In parlicular, the internal geometry of a vessel may be readily inferred or determined from intravascular images, such as OCT or IVUS images. As the intravascular images may represent one or more sequences of cross-sectional images of a vessel across a given pullback, reference vessel diameter, minimum lumen diameters, minimum lumen areas, stenosis diameters, calcium severity, and so may be calculated.

[0183] Similarly, lesion characteristics can be readily determined from intravascular images, such as OCT or IVUS images. Such intravascular images, or image data, may allow the various structures or tissues that make up the vessel wall and lesion to be discerned. In this way the calcium that makes up the lesion may be distinguished from the materials of the vessel itself, such as externa, media, intima etc. Similarly other deposits, such as lipids, may often be distinguished or identified. Again, as the intravascular images may represent a sequence of cross-sectional images of a vessel across a given pullback, idem i licat ion of these regions allow lesion characteristics such as the length of a lesion, presence of calcium, arc length of calcium, location of calcium, thickness of calcium, and so on, to be determined.

[0184] As such, the intravascular imaging data provides suitable information for the models to discriminate between lesions and vessels required different IVU treatments, as discussed shortly below. In parlicular, by training the models on intravascular image data and corresponding IVU parameters, labelled according to the outcome of the IVU procedure in each case the model may learn or identify a mapping between the intravascular data space and the IVU parameter space that maximizes the measure of success chosen for the outcome of the IVU treatment.

[0185] The IVU parameters may be thought of as unknown functions of vessel and lesion characteristics. It will be understood that physically the total energy delivered by IVU to a given lesion at a given point on the vessel wall may be assumed to be in part determined by or a function of the mass of the lesion at thatABTLLI-128point. As noted above, the various IVL parameters each effect the energy delivered at a given point on the vessel wall. For example, the more shocks that are delivered, the more energy is delivered to all points at the treatment site. Similarly, the energy profile of the IVL arrangement directly effects the energy delivered to different points along the vessel wall. Over or under sizing of the IVL balloon also effects the energy delivered depending on the thickness of the calcium in a given lesion.

[0186] For example, a lesion with a particularly thick section may require an undersized balloon compared to the reference diameter of the vessel. This in turn will reduce the energy transmitted to the other less thick parts of the lesion. This may be parlicularly relevant where there are multiple treatment posilions in a given IVL parameter as a single balloon size is needed for each posilion.

[0187] In general, the lesion characteristics, such as the presence, thickness, and distribution of calcium within the vessel wall, can vary and strongly affect the IVL parameters. Concentric and eccentric calcifications require different approaches. Concentric calci Ileal ions may be uniform around the vessel, and eccentric calcifications may be localized. For concentric calcifications, a uniform energy profile would usually be used, while eccentric calcifications would usually require targeted energy delivery. As such, the optimum and / or suitably effective number of pulses and / or emitter arrangement maybe dependent on the calcification’s arc length and thickness.

[0188] In some examples, a separate treatment may be determined to be completed before IVL. Such separate procedures, for example, may be applicable for eccentric calcification. As a result, the IVL parameter may be the determination of the separate treatment, such as those detailed in PCT / US2025 / 036276, which is incorporated herein by reference in its entirety. For example, when calcium nodules are detected in a region of interest, if predicted fracture probability at the nodule site remains below a predefined confidence threshold after initial application of the enhanced energy mode, the system may recommend adjunctive therapies, such as specialty balloons and / or atherectomy, as a separate procedure.

[0189] Further, longer lesions may require multiple treatment positions to ensure complete coverage and cITccli ve calcium modification. Here, the IVL parameter might include overlapping treatment posilions with specific parameters for each position to ensure the entire lesion is adequately treated.

[0190] As set out above, the diameter of the vessel at the treatment position heavily influences, and in some cases completely determines, the appropriate balloon size and inflation pressure. Undersized balloons may not deliver sufficient energy, while oversized balloons can cause vessel damage. As such, accurate determination of the vessel diameter from the intravascular imaging data, such as OCT or IVUS, helps in selecting the correct balloon size and inflation pressure to optimize energy delivery and minimize risks.

[0191] In terms of the lesion characteristics, the calcification extent and / or distribution, such as arc length, may influence the number of pulses to be delivered. A vessel segment with 360-degree calcification is more likely to benefit from IVL pulses. The comprehensive calcification around the vessel wall can be effectively fractured by the pressure waves generated by IVL, improving vessel compliance and facilitating subsequent procedures like stenting. In contrast, a treatment segment with only a few degrees ofABTLLI-128calcification might not require the extensive application of IVL pulses. The localized nature of the calcification can be targeted more precisely, potentially reducing the number of pulses needed.

[0192] As discussed above in relation to Figure 8, the energy profile and / or emitter placement of the IVL parameter may be heavily influenced by the calcification extent and / or distribution, such as arc length. Tthe choice of emitter arrangement may be guided by the calcification pattern. Energy profiles or pressure profiles of similar shape to the calcification distribution of the lesion are preferable, with the peak or peaks of the energy profile being placed nearest the peak or peaks of the calcification distribution. This ensures that the maximum energy is delivered to the most calcified area, enhancing the effectiveness of the treatment. In this way the energy profile and / or emitter arrangement may be correlated with the actual treatment position or positions.

[0193] As such, the various IVL parameters and possible treatment positions described above may be dependent on the vessel and lesion characteristics, as these characteristics determined the energy that needs to be delivered at each point on the vessel walls, which in turn determines the IVL parameters and treatment positions. The vessel and lesion characteristics may also provide various constraints on the IVL parameters with respect to balloon size, balloon pressure and so on. As such, the determination of the IVL parameters and treatment positions may be thought of as a constrained optimization problem, albeit one where the specific functional relationship between the IVL parameters and the vessel / lesion characteristics are unknown.

[0194] It will be appreciated that the sets of vessel characteristics discussed herein may comprise any combination of the vessel characteristics discussed above. Similarly, the sets of lesion characteristics discussed herein may comprise any combination of the lesion characteristics discussed above.

[0195] It will also be understood that there are other region of interest characteristics which may be readily determined without recourse to intravascular imaging. These include: lesion location - i.e. the physical location of the vessel in the body of the patient, and vessel tortuosity. These may be used to determine a balloon size and / or range of balloon sizes based on the reference vessel diameters for a given lesion location. Reference vessel diameters may be readily determined based on lesion location and age.

[0196] As discussed above in relation to Figure 3, the characterization subsystem 310 may include one or more models trained to determine the set of imaging characteristics from the image data 202a. It will be understood that such determination is in effect an image segmentation problem which can be addressed using image segmentation techniques. Various neural network architectures may be used for image segmentation such as V-net, U-net, CUMedVisionl, CUMedVision2, VGGNet, M2FCN, Coarse-to-Fine Stacked Fully Convolutional Net, Deep Active Learning Framework, ResNet and / or combinations thereof. Such architectures can be trained on image data annotated with imaging characteristics, often generated through manual inspection of the images. Specific examples of such image segmentation to generate imaging characteristics can described in ’309 patent.ABTLLI-128

[0197] Also as set out in Figure 2, a model may receive as input the set of characteristics, may be trained to determine one or more IVL parameters, and may provide as output the one or more IVL parameters. It will be appreciated that various Al and / or ML architectures may be used for the trained model. In effect, in view of the discussion above describing the relationship between the imaging characteristics and the IVL parameters, any architecture suitable for classification may be used. For example, decision trees, such as random forest architectures, clustering models and so on.

[0198] Particular benefit though may be observed by the use of an image or volume segmentation architecture. In particular, the V-Net model may be used for the model trained to determine IVL parameters. Here, the training data for this model may include the imaging data along with the imaging characteristics of each lesion as addilional inputs. The training data is labelled according to the IVL parameter used and the achieved stent expansion. As such, the model will receive as input both the region of interest characteristics 202b and the image data 202a.

[0199] An alternative approach is a single trained model is arranged to receive as input the image data 202a and trained to generate as output the one or more IVL parameters 235. In effect, the model is during training left to infer the relevant imaging characteristics rather than having these explicitly set. Any suitable image segmentation architecture as described above may be used for the single trained model. However, the V-Net architecture is parlicularly suitable for this application.Example GUIs

[0200] Figure 9 illustrates an example display 1300, which may be implemented by output interface 1120 of system 100 of Figure 1. The display 1300 may be an output generated in block 430 of method 400, as described in connection with Figure 4. As discussed below, display 1300 includes image data and the determined IVL parameter. As such, the model may output one or more indications of the determined IVL parameter in relation to the particular lesion morphology and vessel structure before an IVL procedure.

[0201] The display 1300 may include a first portion 1302, a second portion 1304, and a third portion 1306, each including a representation of a region of interest. As shown, first portion 1302 may be positioned adjacent to second portion 1304. First and second portions 1302, 1304 may be positioned above third portion 1306 including the indications of treatment positions 1318, 1302, 1322, which may be positioned above the indications of the IVL parameters 1324, 1326. However, in other examples, each porlion may be positioned anywhere and in any order on the display. The first portion 1302 may include an angiographic representation of the region of interest, the second portion 1304 may include an image frame of the region of interest, and the third portion 1306 may include a two-dimensional rcprcscnlalion of the region of interest. The image frame in the second porlion 1304 may be a cross-sectional image of the blood vessel. The cross-sectional image may correspond to a location 1319 along the two-dimensional representation shown in the third portion 1306. The location may be determined based on a user input received by the system, where the user input corresponds to a selection of a location along the two-dimensional representation.ABTLLI-128

[0202] The first portion 1302, second portion 1304, and / or third portion 1306 may provide information associated with the vessel. For example, the first portion 1302, an angiographic image that includes, at least, the region of interest, may include annotations. The annotations may provide an indication of the presence of calcium, lipid, etc. within the region of interest. Each deposit, e.g., calcium, lipid, etc., may be represented with a different color, pattern, or combination thereof. As shown in Figure 9, the orange indication 1312 corresponds to the presence of calcium at and / or along that portion of the vessel. Similarly, the corresponding presence of calcium may be illustrated with annolalions in the two-dimensional representation, cross-sectional image, numerical and / or textual indications, or the like. As shown in Figure 9, the image frame in the second portion 1304 may include a calcium arc, or curve, that extends at least partially circumferentially around the image. In examples where the portions 1302-1306 include an indication of lipid, the image frame may include an arc, curve, or the like illustrating the presence and / or extent of lipid. The two-dimensional representation in the third portion 1306 may include an indication of the calcium, lipid, etc. by color coding, hashing, hatching, etc. of the two-dimensional representation. As shown, the orange portions of the two-dimensional representation provide an indication of the presence of calcium. Between the second portion 1304 and third portion 1304 may include annolalions of an indication of the calcium arc 1306, 1308, which may be represented numerically, e.g., 184 degrees, or visually, e.g., as a coaxial arc or ring around a perimeter of an intravascular image frame, which may be color-coded. Additional annotations between the second portion 1304 and third portion 1304 may include an indication of a maximum thickness 1310 of calcium, which may be represented numerically, e.g. 0.54mm, or the like. In examples where an indication of lipid is provided for display, another color and / or form of indication may be provided on or along the two-dimensional representation.

[0203] According to some examples, the annotations of the different portions may include an indication of the lesion 1314. As shown in Figure 9, the indication of the lesion 1314 may be highlighting and / or a different in color within the lesion, but in other examples, may be an outline around a lesion and / or a difference in pattern within the region of interest. Though Figure 9 depicts only one lesion, in other examples, the display 1300 may include an indication for each of a plurality of lesions.

[0204] The two-dimensional representation in third portion 1306 may, in some examples, be a symmetrical representation of the vessel . For example, the representation may be symmetrical about a longest axis the represent al ion. In some examples, the two-dimensional representation may include a first axis corresponding to the localion along the vessel and a second axis corresponding to other data associated with the vessel. The other data associated with the vessel may include, for example, diameter values, EEE values, FFR values, VFR values, etc. In some examples, the two-dimensional representations may be a graphical representation. As shown in display 1300, the two-dimensional representation is generated based on diameter values of the vessel and is symmetrical about the longest axis of the representation. As shown, the two-dimensional representation may extend horizontally across the display 1300. However, in other examples, the two-dimensional representation may extend vertically on the display 1300. According toABTLLI-128some examples, the two-dimensional representation may be referred to as a longitudinal view of the vessel, even in examples where the two-dimensional representations extends vertically on the display 1300.

[0205] Below the third portion 1306 may include annotations or values associated with the determined IVL parameters, such as one or more treatment positions and / or one or more IVL parameters, such as IVL size and / or IVL dose, corresponding to each treatment position.

[0206] In examples where the IVL parameters include treatment positions, the treatment positions may be provided as annotations on one or more representations of the vessel. For example, the representations of the vessel in the first portion 1302 and / or the third portion 1306 may include annotations indicating the determined treatment position. As shown in Figure 9, the annotations may be a rectangle indicator 1318, 1320, 1322, each surrounding one of three treatment positions of the two-dimensional representation in third portion 1306. While the indicators are shown as rectangular indicators, the annotations may be an indication annotating the position of the respective proximal and distal ends of the treatment positions. In that regard, the annotations may be a line, flag, or other indicator marking the boundaries of the treatment positions.

[0207] As shown in Figure 9, two of the treatment position indicators 1318, 1320 overlap. As discussed above, an overlap in treatment positions may ensure proper treatment of a lesion that is axially longer than the IVL shock or pressure wave and / or emitter region of the IVL apparatus. In some examples, the annotations may include an indication of guide points 1318a-b, 1320a-b, 1322a-b on opposing ends of each treatment position indicator 1318, 1320, 1322. As shown in Figure 9, the annotation of the guide points 1318a-b, 1320a-b, 1322a-b may be a shape and / or color indicator, such as a rectangle having a different color than a color of the treatment position indicator 1318, 1320, 1322. The guide points 1318a-b, 1320a-b, 1322a-b may indicate where the corresponding marker bands, which may be placed at each end of an IVL balloon, should be located to ensure that an IVL device is in the specified treatment position. While different configurations and / or representations of the treatment positions and guide points are discussed above and herein, the examples provided are merely some examples of how the treatment positions and / or guide points may be provided for output via a GUI and, therefore, are not intended to be limiting.

[0208] In some examples, the output provided via the GUI may include an indication of the IVL parameters determined by the IVL configuration system 200, 300. The determined IVL parameters, such as the IVL size and / or IVL dose for a given treatment position, may be provided for output via the GUI. As shown, the GUI may include an indication of the IVL size 1324 and / or IVL dose 1326. In some examples, the IVL parameters, such as IVL size and / or IVL dose, may be provided for output via the GUI after receiving an input corresponding to the selection of a treatment position. By providing the determined IVL size and / or dose for a given treatment position in response to receiving a user input, the GUI provides a clear and efficient output of the IVL parameters determined for the given treatment position. In some examples, the output provided via display 1300 may include IVL parameters, such as IVL size and / or IVL dose, for all treatment positions, which may be located alongside their respective treatment positions.ABTLLI-128

[0209] In other examples, any of the annotations on display 1300, may be a textual indicator, a color coded indicator, a shape indicator, or the like. For example, the color coded indicator may be highlighting, a difference in color, and / or a difference in pattern. The difference in pattern may be, for example, different hashings, stipples, or the like. The shape indicator may be a shape or directional. The shape may be, for example, a circle, square, rectangle, triangle, polygon, etc. The directional may be, for example, an arrow.

[0210] Figure 10 illustrates a further example display 1400. The display 1400 shown in Figure 10 is similar to display 1300 shown in Figure 9. In that regard, the display 1400 includes a first portion 1402, corresponding to first portion 1302, a second portion 1404, corresponding to second portion 1304, and a third portion 1406, corresponding to third portion 1306. Display 1400 differs from display 1300 in that display 1400 may be generated during and / or after the IVL procedure.

[0211] The IVL configuration system 200, 300 may receive or determine a number of shock waves or pulses as they are delivered at each position along a vessel segment during an IVL procedure. For example, the IVL configuration system 200, 300 may be in communication with an IVL system such that, as the IVL procedure is being performed, the IVL configuration system 200, 300 may receive information from the IVL system. The information may include, for example, the number of pulses administered, the number of pulses remaining, or the like. Such information may be reflected in a similar way to or alongside the IVL parameters.

[0212] As shown in Figure 10, the two-dimensional representation in portion 1406 may include an indication and / or annotation denoting the treatment positions 1418, 1420. As shown the treatment positions are indicated via a rectangular indicator, as described in connection with Figure 9. As compared to the display 1300 in Figure 9, the display 1400 in Figure 10 may additionally include an indication that a predetermined number of pulses has been delivered to a given treatment position. The predetermined number of pulses may be, for example, determined based on an input corresponding to a user input indicating the number of pulses to be delivered. In some examples, the predetermined number of pulses may be the number of pulses determined by the IVL configuration system 200, 300 For example, the annotation regarding the number of pulses may be a color-coded indicator. Specifically, the treatment position indicator 1418, 1420 may change color, such as from green to red, when the predetermined number of pulses has been delivered, when the threshold number of pulses has been delivered, or the like. Such color change may indicate that no further pulses should be delivered to that treatment position.

[0213] Positions where the predetermined number of pulses have not yet been delivered may be indicated as such. For example, the treatment posilion may start as a first color. As shown via display 1400, the first color may be green. The treatment position may remain a color, e.g., green, if the predetermined number of pulses has not yet been delivered to that treatment position.

[0214] According to some examples, in addilion to or as an alternative of the color annotation associated with the predetermined number of pulses, the display 1400 may include a text indicator associated with the predetermined number of pulses. For example, one of the treatment positions 1418 that corresponds to aABTLLI-128completed treatment may be indicated by a first color, such as red, different from the non-completed treatments. A completed treatment may correspond to a treatment site that has received the determined number of pulses and / or the limit of pulses. A non-completed or suggested treatment may correspond to a treatment site that has not yet received that determined number of pulses and / or the limit of pulses. The non-completed treatment may be indicated by a second color, different from the first color, such as green, and / or may be indicated by a textual indication. In some examples, the textual indication may be provided via display 1400 relative to the two-dimensional representation below the third portion 1406. The textual indication may be a label of the specified zone, such as Zone 1 or “1,” as shown below the third portion 1406, corresponding to the treatment site, with the textual indication located next to “Completed Treatment.” Similarly, one of the treatment positions 1420 that corresponds to a non-completed or suggested treatment may be indicated by a color, such as green, different from the completed treatments and / or may be indicated by a textual indication below the longitudinal view. The textual indication may be a label of the specified zone, such as Zone 2 or “2,” as shown below the third portion 1406, corresponding to the treatment site, with the textual indication located next to “Suggested Treatment.” The “Suggested Treatment” may be positioned above the “Completed Treatment.” Thus, the model may quickly and efficiently indicate which treatment position are yet to be adopted during the procedure. This may be particularly useful in the case of unforeseen circumstances, such as when advancing an IVL balloon to a certain position becomes difficult or impossible during a given procedure. In such a scenario, the IVL balloon itself may not be nearing capacity and therefore may still be capable of delivering further pulses. Here, the model may quickly and efficiently indicate that the IVL balloon should be move to alternative and accessible treatment positions so that the remaining pulses can be used in a therapeutically beneficial way. As such, unforeseen treatment position access issues may be obviated, the use of available catheter pulses may be optimized and the catheter position for therapy can be correlated to actual vessel lesions as opposed to simply advancing the catheter.

[0215] Figures 11A-B illustrate a further example display 1500, 1500’. The display 1500, 1500’ (individually and / or collectively referred to as display 1500) shown in Figures 11 A-B is similar to display 1300 shown in Figure 9. In that regard, the display 1500 includes a first portion 1502 substantially corresponding to first portion 1302, a second portion 1504 substantially corresponding to second portion 1304, and a third portion 1506 substantially corresponding to third portion 1306. As compared to the display 1300 in Figure 9, the display 1500 further includes a fourth portion 1508 adjacent to the second portion 1504 and the third portion 1506. However, in other examples, each portion may be positioned anywhere and in any order on the display.

[0216] Like first portion 1302 in Figure 9, the first portion 1502 in Figures 11 A-B may include colored indications 1512 corresponding to the presence of calcium at and / or along that portion of the vessel.

[0217] Like second portion 1304 in Figure 9, the second portion 1504 in Figures 11 A-B may include an arc 1510, or curve, that extends at least partially circumferentially around the image. The arc, or curve,ABTLLI-128may correspond to an indication of the presence and / or amount of calcium. Second portion 1504 may include a numerical representation 1520 of the arc 1510, e.g., 180 degrees. The arc 1510 and / or numerical representation 1520 may be color coded. In some examples, the color of the arc 1510 and / or numerical representation 1520 may correspond to the color of the colored indications 1512. Second portion 1504 may include an indication of the presence, location, and / or amount of EEL 1514, 1517and lipid 1516. Second portion 1504 may include an indication of the presence, location, and / or measurement of a lumen area 1518, 1519.

[0218] Like third portion 1306 in Figure 9, the third portion 1506 in Figures 11A-B may include an indication of the presence, amount, and / or location of calcium, lipid, etc. by color coding, hashing, hatching, etc. of the two-dimensional representation. As shown, the orange portions of the two-dimensional representation provide an indication of the presence of calcium, and the blue portions provide an indication of the presence of lipids. Third portion 1506 may also include an indication of the location and / or value of a minimum lumen area 1524.

[0219] In examples where the IVL parameters include treatment positions, the treatment positions may be provided as annotations on one or more representations of the vessel. For example, the representations of the vessel in the first portion 1502 and / or the third portion 1506 may include annotations indicating the determined treatment position. As shown in Figures 11A-B, the annotations of the treatment positions may include an indication of each treatment position 1522a-c. Also as shown in Figures 11 A-B, the indication of the region of interest 1522a-c may be highlighting and / or a textual indicator, e.g. Rl, R2, or R3, but in other examples, may be an outline around each treatment position and / or a difference in pattern within the treatment position. As shown in Figures 11 A-B, the annotations may be a highlighting and / or a textual indicator 1513a-c on the angiographic representation in the first portion 1502 and / or a highlighting and / or a textual indicator 1522a-c on the two-dimensional representation in the third portion 1506, each surrounding one of three treatment positions. Though Figures 11 A-B depict three treatment positions, in other examples, the display 1500 may include an indication for any number of treatment positions.

[0220] The display 1500 may, in some examples, provide an indication of the IVL size, dose, or other parameters determined by the IVL configuration system 200. For example, the IVL parameters determined by the IVL configuration system 200 may, in addition to the treatment positions, include IVL size and / or IVL dose for each treatment position. As shown in Figures 11 A-B, the fourth portion 1508 may include a subsection 1526a-c for each treatment position, e.g., R1-R3. As shown, the subsections 1526a-c are stacked vertically. However, this is just one example and is not intended to be limiting. For example, the subsections 1526a-c may be selectable from a drop down list such that only one subsection is displayed at a given time. In another example, the subsection may be provided for display side by side, e.g., aligned horizontally. Accordingly, the configuration of the subsections 1526a-c within the fourth portion 1508, as shown in Figures 11 A and 1 IB, is just one example.ABTLLI-128

[0221] Each subsection 1526a-c may include information associated with the IVL parameters determined by the IVL configuration system 200. For example, the subsection may include a textual indication of the treatment position 1528. The textual indication of the treatment position 1528 may correspond to the treatment regions identified in the first portion 1502 and the third portion 1506, e.g., Region 1 (“Rl”), Region 2 (“R2”), Region 3 (“R3”), etc. The subsections 1526a-c may include a textual indication of IVL device size 1530, e.g., 2.5mm x 20mm, and / or of IVL dose 1532, e.g., 10 , 20, 40, etc. As detailed above, the IVL dose may correspond to the number of pulses remaining in the treatment and / or the number of pulses to be administered in the treatment position.

[0222] In some examples, the IVL size and / or IVL dose for each treatment position may be provided for output via a GUI after receiving user input corresponding to the selection of the IVL information section. As shown in Figures 11A-B, the fourth portion 1508 may include several information sections. Each information section may provide information associated with the vessel, IVL parameters, etc. For example, the information sections may include a section for sizing and IVL parameters. The sizing information section may provide, for example, information related to measurements of the vessel, calcium measurements within the vessel, lipid measurements within the vessel, and lumen measurements of the vessel. The IVL information section may provide, for example, information related to the IVL parameters determined by the IVL configuration system.

[0223] The system may receive user input, for example, corresponding to the selection of a symbol, such as an arrow, associated with a given information section. Respective information may be provided for output upon receipt of the selected information section. For example, in response to receiving a selection associated with the sizing information section, an information subsection may be provided for further selection. The information subsections may include, for example, measurements, calcium, lipid, and lumen. In some examples, in response to receiving a selection associated with the IVL information section, IVL parameters, such as the IVL size and / or IVL dose, for one or more treatment positions may be provided for output. In response to receiving a selection associated with the measurement information section or the lumen information section, measurements, such as minimum lumen diameter, minimum lumen area, EEL, etc., for a region of interest may be provided for output. In response to receiving a selection associated with the calcium information section, information about calcium, such as calcium arc, calcium thickness, calcium location, calcium length, calcium type, etc., for a region of interest may be provided for output. In response to receiving a selection associated with the lipid information section, information about lipid, such as lipid arc, lipid thickness, lipid location, lipid length, etc., for a region of interest may be provided for output.

[0224] According to some examples, display 1500 may include an indication of a given treatment position to be treated next. The indication may include color coding, highlighting, annotating, outlining, etc. Such indication may be automatically provided for output, such as in sequential order for all treatments that are not completed. In other examples, such indication may be provided for output after receiving a user input.ABTLLI-128As shown in Figure 11 A, subsection 1526a in the fourth portion 1508 may include an indication of an IVL size and / or IVL dose for a given treatment position, e.g., region 1 (Rl). The indication 1526a may be a color indication, such as an orange box including the IVL dose and / or IVL size for that treatment position, but in other examples, may be an outline around and / or a difference in pattern including the IVL dose and / or IVL size for that treatment position. In some examples, the two-dimensional representation in third portion 1506 may include an indication 1522a for a given treatment position. Indication 1522a may be a different font size of the textual indicator, such as a larger text size, but in other examples, may be a different color, etc.

[0225] According to some examples, one or more portions may include an indication of a completed treatment. As shown in Figure 11 B , the fourth portion 1508 may include a textual indication of a completed treatment, e.g. “Complete.” The fourth portion 1508 and third portion 1506 may include an indication of a recommended or selected next treatment corresponding to one of the treatment positions, as described in connection with Figure 11 A. Specifically, as shown in Figure 1 IB, subsection 1526b in fourth portion 1508 may include an indication of an IVL size and / or IVL dose for one of the treatment positions, and the longitudinal representation in third portion 1506 may also include an indication for one of the treatment positions 1522b. Such indications may be the same as described in connection with Figure 11 A. In some examples, such indication may be automatically provided for output upon completion of the treatment for the preceding treatment position.

[0226] In other examples, the display 1500 may include any combination of the portions and / or outputs as described in connection with Figures 9 or 10.

[0227] Figure 12 illustrates a further example display 1600. The display 1600 shown in Figure 12 is similar to display 1500’ shown in Figure 1 IB. In that regard, the display 1600 includes a first portion 1602 corresponding to first portion 1502’, a second portion 1604 substantially corresponding to second portion 1504’, a third portion 1606 corresponding to third portion 1506’, and a fourth portion 1608 substantially corresponding to fourth portion 1508’ . As compared to the display 1500’ in Figure 1 IB, the fourth portion 1608 in Figure 12 is adjacent to the third portion 1606 and below the second portion 1604. Additionally as compared to the display 1500’ in Figure 1 IB, the display 1600 in Figure 12 further includes a fifth portion 1636 adjacent to the first portion 1602 and above the third portion 1606. However, in other examples, each portion may be positioned anywhere and in any order on the display.

[0228] According to some examples, the first portion 1602 may be co-registered image data. For example, the first portion may be an angiographic image co-registered with vessel data from a pullback of an intravascular imaging probe. The fifth portion 1636 may be a live view. The live view may provide the user with an indication of a current position of an IVL device within the vessel or region of interest. For example, angiographic images, such as low dose angiographic image or fluoroscopic images, taken during the IVL procedure may be co-registered with a high dose angiographyic image that is used as a vessel map.ABTLLI-128Details for how the live view of the device is generated may be found in PCT / US2024 / 039350, the cnlirc contents of which are incorporated by reference.

[0229] According to some examples, the third portion 1606 may, be a representation of the vessel based on pre-PCI image data and / or post-PCI image data. For example, image data may be captured via a pullback of an intravascular imaging device before and / or after the IVL procedure or other PCI procedure, such as stenting. For pre-procedural image data, predicted metrics associated with the IVL procedure or other PCI may be provided. For example, a predicted percent expansion for a stent in the region of interest may be provided for output relative to one or more of the representations. As shown in Figure 12, the fifth portion 1636 may include an indication of the predicted percent expansion of the stent 1638.

[0230] In some examples, the display 1600 may include information associated with post procedural and / or post-PCI image data. For example, the information may be associated and / or relative to an implanted stent. The information may be provided relative to one of the reprcsenlalions. For example, the display 1600 may include an indication of apposition, proximal and distal boundaries of the stent, or the like provided relative to the two-dimensional representation in the third portion 1608 and / or relative to the image frames in the first and / or fifth portions 1602, 1636. For example, if the image data is post procedural image data, the percent expansion of the stent 1638 may be the actual percent expansion, not the predicted percent expansion.

[0231] In other examples, the display 1600 may include any combination of the portions and / or outputs as described in connection with Figures 9, 10, 11 A, or 1 IB.Example System

[0232] Figure 13 illustrates an example system 1000, for use in collecting vessel data, such as intravascular and extravascular data, and for use in implementing an IVL configuration system, such as IVL configuration system 200 described in connection with Figure 2 or IVL configuration system 300 described in connection with Figure 3, or as otherwise set out in any of the examples above. The system 1000 comprises a computer 1020 and an imaging system, such as imaging system 130 as detailed in connection with Figure 1. The computer 1020 comprises: a storage system 1040, a memory 1060, one or more processors 1080, an interface 1100, a user output interface 1120, a user input interface 1140 and a network interface 1160, which are all linked together over one or more communication buses 1180.

[0233] The storage system 1040 may be any form of non-volatile data storage device such as one or more of a hard disk drive, a magnetic disc, an optical disc, a ROM, etc. The storage system 1040 may store an operating system for the processor 108 to execute in order for the computer 1020 to function. The storage system 1040 may also store one or more computer programs, software, instructions, or code. Although Figure 13 functionally illustrates storage system 1040 as integrated into the computer 1020, in other examples, the storage system 1040 may be a workstation or server. In some examples, a subsystem, a server or workstation handles the functions of storage system 1040.ABTLLI-128

[0234] The memory 1060 may store information that is accessible by the processors, including instructions 1062 that may be executed by the processors 1080, and data 1064. The memory 1060 may be a type of memory operative to store information accessible by the processors 1080, including a non-transitory computer-readable medium, or other medium that stores data that may be read with the aid of an electronic device, such as a hard-drive, memory card, read-only memory (ROM), random access memory (RAM), optical disks, as well as other write-capable and read-only memories. The subject matter disclosed herein may include different combinations of the foregoing, whereby different portions of the instructions 1062 and data 1064 are stored on different types of media.

[0235] Memory 1060 may be retrieved, stored or modified by processors 113 in accordance with the inst ructions 1062. For instance, although the present disclosure is not limited by a particular data structure, the data 1064 may be stored in computer registers, in a relational database as a table having a plurality of different fields and records, XML documents, or flat files. The data 1064 may also be formatted in a computer-readable format such as, but not limited to, binary values, ASCII or Unicode. By further way of example only, the data 1064 may be stored as bitmaps comprised of pixels that are stored in compressed or uncompressed, or various image formats such as JPEG, vector-based formats such as SVG or computer instructions for drawing graphics. Moreover, the data 1064 may comprise information sufficient to identify the relevant information, such as numbers, descriptive text, proprietary codes, pointers, references to data stored in other memories, including other network locations, or information that is used by a function to calculate the relevant data.

[0236] The instructions 1062 can be any set of instructions to be executed directly, such as machine code, or indirectly, such as scripts, by the processor 1080. In that regard, the terms “instructions,” “application,” “steps,” and “programs” can be used interchangeably herein. The instructions can be stored in object code format for direct processing by the processor, or in any other computing device language including scripts or collections of independent source code subsystems that are interpreted on demand or compiled in advance. Functions, methods and routines of the instructions are explained in more detail below.

[0237] The processor or processors 1080 may be any data processing unit suitable for executing one or more computer programs, such as those stored on the storage system 1040 and / or in the memory 1060, some of which may be computer programs according to embodiments of the disclosure or computer programs that, when executed by the processor 1080, cause the processor 1080 to carry out a method according to an embodiment of the disclosure and configure the system 1000 to be a system according to an embodiment of the disclosure. The processor 1080 may comprise a single data processing unit or multiple data processing units operating in parallel or in cooperation with each other. The processor 1080, in carrying out data processing operations for embodiments of the disclosure, may store data to and / or read data from the storage system 1040 and / or the memory 1060. Alternatively, the one or more processors may be a dedicated device such as an application specific integrated circuit (ASIC) or other hardware-based processor. Although Figure 13 functionally illustrates the processor, memory, and other elements ofABTLLI-128computer 1080 as being within the same block, it will be understood by those of ordinary skill in the art that the processor, computing device, or memory may actually include multiple processors, computing devices, or memories that may or may not be stored within the same physical housing. Similarly, the memory may be a hard drive or other storage media located in a housing different from that of computer 1020. Accordingly, references to a processor or computing device will be understood to include references to a collection of processors or computing devices or memories that may or may not operate in parallel.

[0238] In examples, the data obtained by an imaging system may be processed by one or more subsystems 1090, such as IVL parameter subsystem 206, to determine IVL parameters, as discussed in connection with Figures 2 and 3. The subsystems 1090 may additionally include a selection subsystem, such as selection subsystem 208, to select a subset of the determined IVL parameters, as discussed in connection with Figures 2 and 3. The subsystems may also include a characterization subsystem, such as characterization subsystem 314, to determine imaging characteristics from image data, as discussed in connection with Figure 3.

[0239] Further, the subsystems 1090 may include a dynamic visualization subsystem, an EEL detection subsystem, a lumen detection subsystem, a lesion detection subsystem, a co-regislralion subsystem, and / or a registration and / or linking subsystem. The dynamic visualization subsystem may automatically correlate high dose and low dose extraluminal images such that a vessel map may be overlaid on the live low dose extraluminal images, a treatment zone identified on the high dose extraluminal images and / or vessel map may be overlaid on the live low dose extraluminal images, and / or the imaging system may be tracked on the live low dose extraluminal images with reference to the vessel map. The dynamic visualization subsystem may correlate the high and low dose XA based on motion features detected in both the high and low dose extraluminal images. In some examples, the dynamic visualization subsystem may use Al, such as one or more Al or machine learning (ML) models, to identify the working vessel, detect markers on the device, and / or track the motion of the device.

[0240] The computer 1020 may be adapted to co-register vessel data obtained during a pullback of the imaging system with intravascular image and / or an extraluminal image. For example, the computer 1020 may be configured to receive and store extraluminal image data, such as image data generated by an imaging system and obtained by a frame grabber, which may be a device configured to capture frames from analog video signal or digital video stream. The computer 1020 may be configured to receive and store intravascular image data, such as image data generated by an imaging system and obtained by the frame grabber. In some examples, computer 1020 may access the co-registration subsystem to co-register the vessel data with the luminal image. The luminal image may be an extraluminal image, such as an angiograph, x-ray, or the like. The co-registration subsystem may co-register intravascular data, such as an intravascular image, plaque burden, EEL measurement, lumen diameter measurements, pressure readings, VFR, FFR, resting full-cycle ratio (RFR), flow rates, etc. with the extraluminal image. In some examples,ABTLLI-128the co-registration subsystem may co-register intravascular data with an intraluminal image, such as an intraluminal image captured by an OCT probe, IVUS probe, micro-OCT probe, or the like.

[0241] In one example, the co-registration subsystem may co-register intraluminal data captured during a pullback with one or more extraluminal images. For example, the extraluminal image frames may be pre-processed. Various matrices such as convolution matrices, Hessians, and others can be applied on a per pixel basis to change the intensity, remove, or otherwise modify a given angiography image frame. As discussed herein, the preprocessing stage may enhance, modify, and / or remove features of the extraluminal images to increase the accuracy, processing speed, success rate, and other properties of subsequent processing stages. A vessel centerline may be determined and / or calculated. In some examples, the vessel centerline may be superimposed or otherwise displayed relative to the pre-processed extraluminal image. According to some examples, the vessel centerline may represent a trajectory of the imaging system, such as an intravascular device, through the blood vessel during a pullback. In some examples, the centerline may be referred to as a trace. Additionally or allcrnali vely, marker bands or radiopaque markers may be detected in the extraluminal image frames. According to some examples, the extraluminal image frames and the data received by an imaging system may be co-registered based on the determined location of the marker bands.

[0242] According to some examples, the subsystems may additionally or al I er natively include a video processing software subsystem, a preprocessing software subsystem, an image file size reduction software subsystem, a catheter removal software subsystem, a shadow removal software subsystem, a vessel enhancement software subsystem, a blob enhancement software subsystem, a Laplacian of Gaussian filter or transform software subsystem, a guide wire detection software subsystem, an anatomic feature detection software subsystem, stationary marker detection software subsystem, a background subtraction subsystem, a Frangi vesselness software subsystem, an image intensity sampling subsystem, a moving marker software detection subsystem, iterative centerline testing software subsystem, a background subtraction software subsystem, a morphological close operation software subsystem, a feature tracking software subsystem, a catheter detection software subsystem, a bottom hat filter software subsystem, a path detection software subsystem, a Dijkstra software subsystem, a Viterbi software subsystem, fast marching method based software subsystems, a vessel centerline generation software subsystem, a vessel centerline tracking subsystem software subsystem, a Hessian software subsystem, an intensity sampling software subsystem, a superposition of image intensity software subsystem and / or other suitable software subsystems as described herein. According to some examples, the subsystems may include software such as preprocessing software, transforms, matrices, and / or other software-based components that are used to process image data or respond to patient triggers to facilitate co-registration of different types of image data by other software-based components or to otherwise perform annotation of image data to generate ground truths and other software, subsystems, and / or functions suitable for implementing various features of the disclosure. The subsystems can include lumen detection using a scan line based or image-based approach,ABTLLI-128stent detection using a scan line based or image-based approach, indicator generation, apposition bar generation for stent planning, guide wire shadow indicator to prevent confusion with dissention, side branches and missing data, and / or others.

[0243] In some examples, the subsystems 1090 may be configured to process the vessel data obtained by an imaging system using artificial intelligence (Al) algorithms, machine learning techniques, or the like.

[0244] The interface 1100 may be any unit for providing an interface to a device 1220 external to, or removable from, the computer 1020. The device 1220 may be a data storage device, for example, one or more of an optical disc, a magnetic disc, a solid-state-storage device, etc. The device 122 may have processing capabi lilies - for example, the device may be a smart card. The interface 1100 may therefore access data from, or provide data to, or interface with, the device 1220 in accordance with one or more commands that it receives from the processor 1080.

[0245] The user input interface 1140 is arranged to receive input from a user, or operator, of the system 1000. The user may provide this input via one or more input devices of the system 1000, such as a mouse 1260 or other pointing device, a keyboard 1240, that are connected to or in coin muni cation with the user input interface 1140. However, it will be appreciated that the user may provide input to the computer 1020 via one or more additional or alternative input devices, such as a touch screen and / or verbal commands. The computer 1020 may store the input received from the input devices via the user input interface 1140 in the memory 106 for the processor 108 to subsequently access and process, or may pass it straight to the processor 108, so that the processor 108 can respond to the user input accordingly.

[0246] In some examples, the information input by the user may be annotations, as detailed in connection with Figure 5. According to some examples, the annotations may be automatically determined by the system. For example, the system may, based on image data and / or region of interest characteristics, determine one or more of a vessel tortuosity, vessel diameter, lesion location, vessel type, number of pulses, calcium density, number of cracks, a reference vessel diameter, a minimum lumen diameter, a minimum lumen area, a stenosis diameter, a length of the lesion, a presence of calcium, an arc length of calcium, a location of the lesion, a thickness of calcium, calcification length, proximal frames, distal frames, or the like. The system may automatically provide such determinations for output as one or more annotations on at least one of the vessel representations.

[0247] The user output interface 1120, or a display, is arranged to provide a graphical / visual and / or audio output to a user, or operator, of the system 1000. As such, the processor 108 may be arranged to instruct the user output interface 1120 to form an image / video signal representing a desired output, and to provide this signal to a monitor 1200, screen, or display unit of the system 1000 that is connected to the user output interface 1120. Additionally, or alternatively, the processor 1080 may be arranged to instruct the user output interface 1120 to form an audio signal representing a desired audio output, and to provide this signal to one or more speakers 1210 of the system 1000 that is connected to the user output interface 1120.ABTLLI-128

[0248] The output interface 1120 may be integrated with the computer 1020, or it may be a standalone unit electronically coupled to the computer 1020. The output interface 1120 may output intravascular data relaling to one or more features detected in the blood vessel and / or obtained during a pullback. For example, the output may include, without limitation, cross-scclional scan data, longitudinal scans, three-dimensional representations generated based on intraluminal and / or extraluminal images, diameter graphs, image masks, lumen border, plaque sizes, plaque circumference, visual indicia of plaque localion, visual indicia of risk posed to stent expansion, flow rate, suggested treatment zones, or the like. The output interface 1120 may identify features with text, arrows, color coding, highlighting, contour lines, or other suitable human or machine-readable indicia.

[0249] In some examples, inputs received by the system corresponding to input information and / or annotations may be provided for output on output interface 1120. Annotations received with respect to one vessel representation may be provided for display on the vessel representation in which the annotations were received. In some examples, the annotations may be provided for display on a plurality of vessel representations, such as the extraluminal representations, graphical two-dimensional representations, longitudinal representations, three-dimensional represenlalions, or the like, the annotations may be provided for output on one, some, or all the representations. According to some examples, when the system receives updated annotations on at least one of the represenlalions, the annotations on the other representations may be updated to correspond to the updated annotations.

[0250] The output interface 1120 alone or in combination with computer 1020 may allow for toggling between one or more viewing modes in response to user inputs. For example, a user may be able to toggle between different intravascular data, images, etc. recorded during each of the pullbacks. In some examples, the user may be able to toggle between different representations, such as a graphical two-dimensional representation, longitudinal representation, a cross-sectional representation, a three-dimensional representation, intravascular images, color images, black and white images, live images, or the like. The graphical two-dimensional representation may include, for example, a graphical representation of the vessel where a first axis corresponds to a location along the vessel of interest and a second axis corresponds to another value, such as diameter, pressure-based measurements or metrics such as FFR or VFR, or any other value derived from the vessel data. In some examples, the graphical two-dimensional representation and / or the longitudinal representation of the vessel of interest may be symmetric about the longest axis of the representation.

[0251] The represenlalions of the vessel may provide various viewing angles and section views. In some examples, EEL positions, diameters thereof, or other EEL based parameters may be output for display relali ve to angular measurements, detected calcium arcs, plaque burden, or the like.

[0252] In some examples, the output may include possible treatment positions. For example, the output may include an indication corresponding to one or more possible treatment positions. The treatment positions may be determined based on the image data, region of interest characteristics, or the like. TheABTLLI-128indications may be provided on any of the vessel representations, e.g., the three-dimensional representation, the longitudinal representation, the graphical representation, image data such as the external images, etc.

[0253] According to some examples, the output interface 1120 and / or computer 1020 may be configured to receive one or more inputs corresponding to a selection on one or more reprcsenlalions. For example, an input may be received corresponding to a selection of an image frame on the longitudinal rcpresenlalion. In response, the other reprcsenlalions provided for output may be updated to display a corresponding indication or image frame. For example, the extraluminal image may be updated to have an indication along the vessel corresponding to the localion of the image frame selected in the longitudinal representation, a circumferential indication may be provided on a three-dimensional representation corresponding to the location of the image frame selected in the longitudinal rcpresenlalion, the cross-sectional image frame may be updated to correspond to the image frame selected in the longitudinal representation, etc. In some examples, the vessel data associated with the selected location may be updated and provided for display.

[0254] In some examples, the output interface 1120 and / or computer 1020 may present one or more menus as output to the physician, and the physician may provide input in response by sei eel i ng an item from the one or more menus. For example, the menu may allow a user to show or hide various features. As another example, there may be a menu for selecting blood vessel features to display.

[0255] Finally, the network interface 1160 provides functionality for the computer 1020 to download data from and / or upload data to one or more data coin muni cal ion networks.

[0256] It will be appreciated that the architecture of the system 1000 illustrated in Figure 13 and described above is merely exemplary and that other computer systems 1000 with different architectures may be used in embodiments of the disclosure. As examples, the system 1000 may have fewer components than shown in Figure 13 or additional and / or alternative components than shown in Figure 13. Further, the computer system 1000 could comprise one or more of: a personal computer; a server computer; a tablet; a laptop; other mobile devices or consumer electronics devices; a cloud computing resource, etc.

[0257] It will be appreciated that the methods described have been shown as individual steps carried out in a specific order. However, the skilled person will appreciate that these steps may be combined or carried out in a different order whilst still achieving the desired result.

[0258] It will be appreciated that embodiments of the disclosure may be implemented using a variety of different information processing systems. In particular, although the figures and the discussion thereof provide an exemplary computing system and methods, these are presented merely to provide a useful reference in discussing various aspects of the disclosure. Embodiments of the disclosure may be carried out on any suitable data processing device, such as a personal computer, laptop, personal digital assistant, mobile telephone, set top box, television, server computer, etc. Of course, the description of the systems and methods has been simplified for purposes of discussion, and they are just one of many different typesABTLLI-128of system and method that may be used for embodiments of the disclosure. It will be appreciated that the boundaries between logic blocks are merely illustrative and that alternative embodiments may merge logic blocks or elements, or may impose an alternate decomposition of functionality upon various logic blocks or elements.

[0259] It will be appreciated that the above-mentioned functionality may be implemented as one or more corresponding subsystems as hardware and / or software. For example, the above-mentioned functionality may be implemented as one or more software components for execution by a processor of the system. Alternatively, the above-mentioned functionality may be implemented as hardware, such as on one or more field-programmable-gate-arrays (FPGAs), and / or one or more applicati on-specific-integrated-circuits (ASICs), and / or one or more digital-signal-processors (DSPs), and / or other hardware arrangements. Method steps implemented in flowcharts contained herein, or as described above, may each be implemented by corresponding respective subsystems; multiple method steps implemented in flowcharts contained herein, or as described above, may be implemented together by a single subsystem.

[0260] It will be appreciated that, insofar as embodiments of the disclosure are implemented by a computer program, then a storage system and a transmission medium carrying the computer program form aspects of the disclosure. The computer program may have one or more program instructions, or program code, which, when executed by a computer carries out an embodiment of the disclosure. The term “program” as used herein, may be a sequence of instructions designed for execution on a computer system, and may include a subroutine, a function, a procedure, a subsystem, an object method, an object implementation, an executable application, an applet, a servlet, source code, object code, a shared library, a dynamic linked library, and / or other sequences of instructions designed for execution on a computer system. The storage system may be a magnetic disc, such as a hard drive or a floppy disc), an optical disc, such as a CD-ROM, a DVD-ROM or a BluRay disc, or a memory, such as a ROM, a RAM, EEPROM, EPROM, Flash memory or a portable / removable memory device, etc. The transmission medium may be a communications signal, a data broadcast, a communications link between two or more computers, etc.

Claims

ABTLLI-128CLAIMS1. A system comprising one or more processors, the one or more processors configured to: provide, as input into an artificial intelligence (Al) model trained to determine one or more intravascular lithotripsy (IVL) parameters, image data of a region of interest including one or more lesions, wherein the one or more IVL parameters includes one or more treatment positions within the region of interest;determine, by executing the Al model, the one or more IVL parameters; andprovide for output, via a display, an indication of the one or more IVL parameters.

2. The system of claim 1 , wherein the one or more IVL parameters further includes one or more of a balloon size, a balloon diameter, a balloon length, a balloon pressure, a number of pulses, a number of emitters for a balloon, a shockwave energy distribution for a balloon, a pulse pattern, posttreatment vessel diameter, time for balloon to be positioned within the vessel, or catheter flexibility.

3. The system of claim 1 or 2, wherein the one or more treatment positions includes at least a portion of one of the one or more lesions.

4. The system of any preceding claim, wherein the one or more processors are further configured to provide, as input into the model, one or more region of interest characteristics.

5. The system of claim 4, wherein the one or more region of interest characteristics include one or more of a vessel tortuosity, vessel diameter values, a lesion location, vessel type, a number of administered pulses, or a calcium density.

6. The system of any preceding claim, wherein the one or more processors are further configured to:determine, by executing the Al model, a set of imaging characteristics comprising one or more of vessel characteristics of a vessel within the region of interest or lesion characteristics of a lesion within the region of interest.

7. The system of claim 6, wherein the vessel characteristics are one or more of a reference vessel diameter, a minimum lumen diameter, a minimum lumen area, or a stenosis diameter.

8. The system of claim 6 or 7, wherein the lesion characteristics are one or more of a length of the lesion, a presence of calcium, an arc length of calcium, a location of the lesion, a thickness of calcium, a calcification length, or a calcium density.ABTLLI-1289. The system of any one of claims 6-8, wherein the determining of the one or more IVL parameters is at least partially based on the one or more vessel characteristics or lesion characteristics.

10. The system of any preceding claim, wherein the one or more processors are further configured to:determine, by executing the Al model, a confidence value for each of the determined one or more IVL parameters.

11. The system of claim 10, wherein the one or more processors are further configured to: identify, based on the determined confidence values for each of the determined one or more IVL parameters, a subset of one or more IVL parameters, wherein the subset includes IVL parameters having a confidence value above a threshold confidence value;wherein providing for output the indication of the one or more IVL parameters includes providing for output the subset of the one or more IVL parameters.

12. The system of claim 10 or 11, wherein the one or more processors are further configured to:rank, based on the determined confidence value for each of the determined one or more IVL parameters, the determined one or more IVL parameters, wherein the one or more IVL parameters are ranked from a highest confidence value to a lowest confidence value;wherein providing for output the indication of the one or more IVL parameters includes providing for output the ranked one or more IVL parameters.

13. The system of any preceding claim, wherein the one or more processors are further configured to:receive, based on the indication of the one or more IVL parameters, feedback data confirming or rejecting the one or more IVL parameters; andprovide the feedback data to the model as part of a feedback loop to update the model.

14. A method comprising:providing, by one or more processors and as input into an artificial intelligence (Al) model trained to determine one or more intravascular lithotripsy (IVL) parameters, image data of a region of interest including one or more lesions, wherein the one or more IVL parameters includes one or more treatment positions within the region of interest;ABTLLI-128determining, by the one or more processors and by executing the Al model, the one or more IVL parameters; andproviding for output, by the one or more processors and via a display, an indication of the one or more IVL parameters.

15. The method of claim 14, wherein the one or more IVL parameters further includes one or more of a balloon size, a balloon diameter, a balloon length, a balloon pressure, a number of pulses, a number of emitters for a balloon, a shockwave energy distribution for a balloon, a pulse pattern, posttreatment vessel diameter, time for balloon to be positioned within the vessel, or catheter flexibility.

16. The method of claim 14 or 15, wherein the one or more treatment positions includes at least a portion of one of the one or more lesions.

17. The method of any one of claims 14-16, further comprising:providing, as input into the model, one or more region of interest characteristics.

18. The method of claim 17, wherein the one or more region of interest characteristics includes one or more of a vessel tortuosity, vessel diameter values, a lesion localion, vessel type, a number of administered pulses, or a calcium density.

19. The method of any one of claims 14-18, further comprising:determining, by executing the Al model, a set of imaging characteristics comprising one or more of vessel characteristics of a vessel within the region of interest or lesion characteristics of a lesion within the region of interest.

20. One or more non-transitory computer-readable media for storing instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:providing, as input into an artificial intelligence (Al) model trained to determine one or more intravascular lithotripsy (IVL) parameters, image data of a region of interest including one or more lesions, wherein the one or more IVL parameters includes one or more treatment positions within the region of interest;determining, by executing the Al model, the one or more IVL parameters; andproviding for output, via a display, an indication of the one or more IVL parameters.

21. A method of specifying an intravascular lithotripsy procedure for a treatment site of a patient vessel, the method comprising:ABTLLI-128receiving an initial set of intravascular image data of the treatment site;generating, using a trained machine learning model, at least one recommended intravascular lithotripsy configuration based on the initial set of intravascular image data; andproviding for output the recommended intravascular lithotripsy configuration.

22. The method of claim 21 wherein the intravascular image data is optical coherence tomography data, and / or intravascular ultrasound data.

23. The method of claim 22 wherein the intravascular image data comprises a set of intravascular image frames.

24. The method of any preceding claim wherein the intravascular lithotripsy configuration specifies one or more intravascular lithotripsy parameters.

25. The method of claim 24 wherein the one or more intravascular lithotripsy configuration comprises one or more treatment positions.

26. The method of claim 25 wherein the intravascular lithotripsy parameter comprises a set of intravascular lithotripsy parameter values for each treatment position.

27. The method of any of claims 24 to 26 wherein the intravascular lithotripsy parameters comprise any one or more of:balloon size;inflation pressure;number of pulses;emitter arrangement; andenergy profile.

28. The method of any one of claims 21 to 27 wherein providing for output the recommended intravascular lithotripsy configuration comprises displaying one or more parameters and / or treatment positions of the intravascular lithotripsy configuration in conjunction with the intravascular image data.

29. The method of any one of claims 21 to 28 wherein providing for output the recommended intravascular lithotripsy configuration comprises configuring an intravascular lithotripsy system according to at least part of the intravascular lithotripsy configuration.ABTLLI-12830. The method of any one of claims 21 to 29 wherein the step of generaling comprises: determining a set of imaging characteristics from the set of intravascular image data; and inputting the determined set of imaging characlcrislics to the trained machine learning model to generate the at least one intravascular lithotripsy configuration.

31. The method of claim 30 wherein the set of imaging characteristics comprises a set of vessel characteristics and / or one or more sets of lesion characlcrislics for lesions in the vessel at the treatment site.

32. The method of claim 31 wherein the set of vessel characteristics comprise any one of more of:a reference vessel diameter at the treatment site,a minimum lumen diameter at the treatment site,a minimum lumen area at the treatment site, orone or more stenosis diameters at the treatment site.

33. The method of claim 31 or 32 wherein the set of lesion characteristics comprise any one of more of:length of a lesion,presence of calcium,arc length of calcium,location of a lesion,thickness of calcium, orcalcification length.

34. The method of any one of claims 30 to 32 wherein determining a set of imaging characteristics from the set of intravascular image data comprises inputting the set of intravascular image data into a further trained machine learning model.

35. The method of claim 34 wherein the further trained machine learning model is selected from a plurality of further trained machine learning models based on one or more treatment site characteristics.

36. The method of any one of claims 31 to 39 wherein the step of generaling comprises inputting the set of intravascular image data to the trained machine learning model to generate the at least one recommended intravascular lithotripsy configuration.ABTLLI-12837. The method of any one of claims 21 to 36 further comprising:receiving a further set of intravascular image data of the treatment site following an intravascular lithotripsy procedure being carried out according to the recommended intravascular lithotripsy configuration;determining a measure of success of the intravascular lithotripsy procedure based on the further set of intravascular image data; andupdating the trained machine learning model based on the initial set of intravascular image data, the recommended intravascular lithotripsy configuration and the measure of success.

38. The method of any one of claims 21 to 38 further comprising displaying at least one treatment position of the intravascular lithotripsy configuration and one or more corresponding intravascular lithotripsy parameters for the treatment position.

39. The method of claim 38 wherein the at least one treatment position of the intravascular lithotripsy configuration is displayed overlaid on the initial set of intravascular image data.

40. A method of training a machine learning model for specifying an intravascular lithotripsy procedure, the method comprising:receiving a plurality of initial sets of intravascular imaging data for respective treatment sites of patient vessels, each set of initial intravascular imaging data corresponding to a respective intravascular lithotripsy configuration of a respective intravascular lithotripsy procedure performed at the respective treatment sites after the set of inilial intravascular imaging data was generated;obtaining for each initial set of intravascular imaging data a measure of success for the respective intravascular lithotripsy procedure;training the machine learning model according to the plurality of inilial sets of intravascular imaging data and the respective measures of success.

41. The method of claim 40 wherein the step of training comprises annotating each initial set of intravascular imaging data with the respective intravascular lithotripsy parameter and the respective measure of success.

42. The method of claim 40 further comprising, for each initial set of intravascular image data, determining a set of characteristics, wherein the step of training comprises annotating each set of characteristics with the respective intravascular lithotripsy configuration and the respective measure of success, and training the machine learning model based on the plurality of annotated sets of characteristics.ABTLLI-12843. An apparatus arranged to carry out a method according to any one of claims 21 to 42.

44. A computer program which, when executed by a processor, causes the processor to carry out a method according to any one of claims 21 to 42.

45. A computer-readable medium storing a computer program according to claim 44.

Images