Systems and methods for performing doppler ultrasound guided peripheral vascular procedures

Abstract

Systems and methods are disclosed for providing intraprocedural guidance and feedback during a vascular intervention via intraprocedural Doppler ultrasound performed on one more distal peripheral vessels impacted by the vascular intervention. According to various example embodiments, intraprocedural Doppler ultrasound is performed to detect the changes in flow characteristics and / or anomalous vascular flow signatures of the peripheral vasculature impacted by the intervention, such as the detection of an adverse vascular event. The detected changes in flow characteristics are intraprocedurally communicated to guide and / or assist clinical decision making, and / or employed to autonomously send control signals to interrupt or modify treatment. In other embodiments, intraprocedural guidance and / or feedback is autonomously provided during a neuromodulation procedure via Doppler ultrasound performed on one more peripheral vessels for which blood flow is modulated by the neuromodulation procedure, with the neuromodulation device optionally being controlled according to the detected changes in blood flow.

Description

SYSTEMS AND METHODS FOR PERFORMING DOPPLER ULTRASOUND GUIDED PERIPHERAL VASCULAR PROCEDURESCROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims priority to U.S. Provisional Patent Application No. 63 / 667,935, titled “SYSTEMS AND METHODS FOR PERFORMING DOPPLER ULTRASOUND GUIDED PERIPHERAL VASCULAR PROCEDURES” and filed on July 5, 2024, the entire contents of which is incorporated herein by reference.BACKGROUND

[0002] The present disclosure relates to diagnostic ultrasound. More particularly, the present disclosure relates the Doppler ultrasound for the detection of vascular pathology.

[0003] When peripheral arterial disease is diagnosed, surgical intervention can often be attempted to open the proximal vessels so as to improve blood flow to a wound, promote wound healing, and avoid amputation. Typically, the interventionalist will use fluoroscopy with contrast to image the arteries in the limbs and perform interventions, such as angioplasty (balloon), atherectomy, or stent to open a stenosis. After intervention, they use fluoroscopy with contrast again to determine how well their intervention improved distal perfusion. This technique is not ideal because fluoroscopy is not a quantitative measure of perfusion. Moreover, fluoroscopic readings are not necessarily predictive of desired outcomes, such as wound healing.

[0004] After the first intervention, the interventionalist is faced with the decision of whether to perform additional interventions. Additional interventions can be performed on the same vessel or additional vessels. While additional interventions may improve perfusion and outcomes, they carry additional risks and require additional time in the catheter lab and / or operating room. More importantly, interventionalists typically do not know which or how many interventions will lead to positive outcomes.SUMMARY

[0005] Systems and methods are disclosed for providing intraprocedural guidance and feedback during a vascular intervention via intraprocedural Dopplerultrasound performed on one more distal peripheral vessels impacted by the vascular intervention. According to various example embodiments, intraprocedural Doppler ultrasound is performed to detect the changes in flow characteristics and / or anomalous vascular flow signatures of the peripheral vasculature impacted by the intervention, such as the detection of an adverse vascular event. The detected changes in flow characteristics are intraprocedurally communicated to guide and / or assist clinical decision making, and / or employed to autonomously send control signals to interrupt or modify treatment. In other embodiments, intraprocedural guidance and / or feedback is autonomously provided during a neuromodulation procedure via Doppler ultrasound performed on one more peripheral vessels for which blood flow is modulated by the neuromodulation procedure, with the neuromodulation device optionally being controlled according to the detected changes in blood flow.

[0006] A further understanding of the functional and advantageous aspects of the disclosure can be realized by reference to the following detailed description and drawings.BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Embodiments are described with reference to the accompanying drawings. In the drawings, like reference numbers can indicate identical or functionally similar elements.

[0008] FIGS. 1A, 1 B and 1C show example autonomous processing workflows involving the use of intraprocedural Doppler ultrasound to assess flow characteristics and / or vascular flow signatures of a peripheral vessel impacted by the intervention to provide intraprocedural feedback and / or control of the intervention.

[0009] FIGS. 1 D and 1 E show example autonomous processing workflows involving the use of intraprocedural Doppler ultrasound to assess flow characteristics and vascular flow signatures of a peripheral vessel impacted by a neuromodulation intervention to provide intraprocedural feedback and / or control of the intervention.

[0010] FIG. 2A shows an example system for employing intraprocedural Doppler ultrasound to assess flow characteristics and vascular flow signatures of peripheral vasculature impacted by the intervention to facilitate intraprocedural feedback and / or control of the intervention.

[0011] FIG. 2B shows an example convolutional neural network architecture for processing spectral Doppler ultrasound data to detect a vascular event signature.

[0012] FIGS. 3A and 3B show color flow and spectral Doppler image respectively of a blood vessel with emboli within (represented by the bright white vertical “streaks” on the Doppler waveform).

[0013] FIGS. 4A, 4B and 4C show color flow and spectral Doppler images respectively with emboli. The emboli are more frequent in these figures compared to what is shown in FIGS. 3A and 3B.

[0014] FIGS. 5A and 5B show examples of a blood vessel undergoing spasm, as evidenced by the increased flow over time. The change in peak systolic velocity is clearly observable.

[0015] FIG. 6 show an example of Doppler spectrum in a blood vessel undergoing recoil.

[0016] FIGS. 7A and 7B show the spectral Doppler waveforms with (FIG. 7A) and without (FIG. 7B) stimulation.DETAILED DESCRIPTION

[0017] Various embodiments and aspects of the disclosure will be described with reference to details discussed below. The following description and drawings are illustrative of the disclosure and are not to be construed as limiting the disclosure. Numerous specific details are described to provide a thorough understanding of various embodiments of the present disclosure. However, in certain instances, well- known or conventional details are not described in order to provide a concise discussion of embodiments of the present disclosure.

[0018] As used herein, the terms “comprises” and “comprising” are to be construed as being inclusive and open ended, and not exclusive. Specifically, when used in the specification and claims, the terms “comprises” and “comprising” and variations thereof mean the specified features, steps or components are included. These terms are not to be interpreted to exclude the presence of other features, steps or components.

[0019] As used herein, the term “exemplary” means “serving as an example, instance, or illustration,” and should not be construed as preferred or advantageous over other configurations disclosed herein.

[0020] As used herein, the terms “about” and “approximately” are meant to cover variations that may exist in the upper and lower limits of the ranges of values, suchas variations in properties, parameters, and dimensions. Unless otherwise specified, the terms “about” and “approximately” mean plus or minus 25 percent or less.

[0021] It is to be understood that unless otherwise specified, any specified range or group is as a shorthand way of referring to each and every member of a range or group individually, as well as each and every possible sub-range or sub-group encompassed therein and similarly with respect to any sub-ranges or sub-groups therein. Unless otherwise specified, the present disclosure relates to and explicitly incorporates each and every specific member and combination of sub-ranges or subgroups.

[0022] As used herein, the term "on the order of", when used in conjunction with a quantity or parameter, refers to a range spanning approximately one tenth to ten times the stated quantity or parameter.

[0023] As used herein, the phrase “spectral Doppler dataset” refers to a dataset characterizing blood flow velocity over time as measured via Doppler ultrasound. Each point in time in a spectral Doppler dataset has an associated velocity distribution characterizing the distribution of velocity values within the sampled region (the Doppler window). The spectral Doppler dataset may also be referred to as a Doppler spectrogram.

[0024] As used herein, the phrase “spectral Doppler waveform” refers to a graphical representation of at least a portion of a spectral Doppler dataset.

[0025] As used herein, the phrase “real-time” refers to a refresh rate, latency or delay in an action, such as processing data, displaying a result or otherwise communicating a result, of less than 100 milliseconds. As used herein, the phrase “near-real-time” refers to a refresh rate, latency or delay in an action, such as processing data, displaying a result or otherwise communicating a result, of less than 2 seconds.

[0026] As used herein, the phrase “distal peripheral vessel” and “distal peripheral vasculature”, when employed in association with a vascular intervention, relate to one or more peripheral vessels located downstream of the location of the vascular intervention that are impacted by the vascular intervention. In such context, the vascular intervention may be performed to target a blockage or pathology to improve or restore perfusion to the distal peripheral vessel and / or distal peripheral vasculature. The intervention may be a peripheral vascular intervention or a nonperipheral (e.g. coronary) intervention.

[0027] As used herein, the term “intraprocedural” encompasses the time period during the performance of a medical procedure (intraoperative) and the time period following the procedure (postoperative) while the patient remains under procedural care, such as, for example, in a surgical suite, recovery room, or intensive care unit.

[0028] Various example implementations of the present disclosure relate to the use of the peripheral vascular Doppler ultrasound to provide actionable intraprocedural feedback and guidance during vascular interventions, such as, but not limited to, vascular procedures employing balloon angioplasty (balloon inflation and deflation), atherectomy, intravascular lithotripsy, and stenting, including peripheral and / or coronary vascular procedures. Conventionally, intraoperative decision making during peripheral vascular procedures is guided by modalities including the presence or lack of intraoperative wound blush during angiographic procedures, defined as a highly concentrated area of contrast media around the distal wound area. This methodology relies on contrast being injected into vessels to show if the contrast made it to the distal vasculature. Its major drawback is that it is subjective as the operator will look at the blush of contrast and subjectively visualize the amount of contrast. The other drawback is it is dependent on the amount of contrast used and does not at all determine any objective data on the flow to the wound but to the general area of interest. Most modalities such as ankle-brachial index, toe brachial index and TcP02 are diagnostic modalities that are bulky, timeconsuming, non-practical for intraoperative settings and are limited to clinics and diagnostic centers. Moreover, such modalities, however, often fail to provide clear, direct and sufficiently rapid feedback that is needed to facilitate intraprocedural decision making.

[0029] The present inventors have discovered that when a vascular intervention is performed, intraprocedural Doppler ultrasound performed on distal peripheral vasculature impacted by the intervention, such as one or more vessels impacted by the peripheral vascular intervention, can be employed to assess flow characteristics and anomalous vascular flow signatures of the peripheral vasculature impacted by the intervention, thereby providing useful intraprocedural feedback that can be employed to guide and / or assist clinical decision making and positively impact the outcome of the procedure. In particular, many of the example embodiments disclosed herein will demonstrate how the intraprocedural processing of spectral Doppler ultrasound data can guide and / or facilitate intraoperative decision makingwhen performing a vascular intervention directed to correcting a pathology of the vascular system.

[0030] The present disclosure is organized as follows. Methods are broadly described in which intraprocedural Doppler ultrasound is employed to assess peripheral vasculature during a wide variety of vascular procedures to facilitate intraprocedural feedback and / or guidance. Intraprocedurally acquired Doppler data characterizing flow in one or more peripheral vessels, such as spectral Doppler waveforms and / or other forms of Doppler data, are dynamically and interprocedurally processed to infer and / or detect the presence of a change in vascular flow characteristics in the one or more peripheral vessels, such as the detection of a peripheral vascular event, including, for example, the intraprocedural onset and / or persistence of a peripheral vascular event. In some example embodiments, after detection of a change in vascular flow characteristics, feedback is provided to a user to alert the user alerting the user to the detected change. In some example embodiments, after detection of a change in vascular flow characteristics, a control signal is autonomously generated and delivered to a therapeutic medical device or therapeutic medical system associated with the procedure, where the control signal is configured to control or instruct the medical device to interrupt or modulate (modify) the therapeutic intervention. These and other example embodiments will be described below in the context of the user of Doppler ultrasound for the intraprocedural detection of peripheral vascular events including the detection of emboli in peripheral vessels, the detection of peripheral vasospasm (peripheral vessel spasms), the detection of peripheral vessel recoil, and the detection of intraprocedural changes in blood flow in peripheral vessels during neuromodulation procedures such as spinal stimulation.

[0031] In contrast to commercial Doppler ultrasound systems that enable an interventionalist to observe the presence of vascular flow characteristics by viewing a display, such as a spectral Doppler waveform or a power M-mode Doppler display, or to hear such events via audio output, the example embodiments of the present disclosure provide systems and methods that involve the autonomous processing of Doppler ultrasound data for the continuous (temporally longitudinal) detection and monitoring of changes in vascular flow characteristics during therapeutic interventions.

[0032] Referring now to FIG. 1 A, an example method is shown in which feedback and / or guidance is provided during a vascular procedure based on the detection of a change in vascular flow characteristics of a peripheral vessel via the intraprocedural processing of Doppler ultrasound data characterizing flow within a distal peripheral vessel, such as the identification of a flow signature associated with the presence of an adverse vascular event. In step 100, Doppler ultrasound data is acquired from the distal peripheral vessel during the vascular procedure. This data is dynamically and intraprocedurally processed, such as in real-time or near-real-time, as shown at step 110, to detect the presence of changes in vascular flow characteristics (such as the presence of a vascular event) associated with the vascular intervention.

[0033] In step 120, the occurrence of a detected change in vascular flow characteristics, such as the detection of a flow signature associated with the presence of an adverse vascular event, is communicated to an interventionalist (e.g. user, clinician, nurse). Communication may occur according to a wide variety of methods and modalities, such as, but not limited to, a visual display on a user interface, an audible alarm, and a combination thereof.

[0034] In one example implementation, the workflow illustrated in FIG. 1A may be implemented after first having affixed the Doppler ultrasound device to an anatomical region suitable for detecting a target peripheral vessel, such as the foot, and this step may be performed prior to initiating the therapy. Prior to the initiation of therapy, a baseline ultrasound spectral Doppler dataset may be detected (e.g. for at least 5 seconds). The spectral Doppler dataset is autonomously analyzed, for example, to determine a waveform envelope, and optionally to determine Doppler indices such as, but not limited to, maximum velocity, mean velocity, pulsatility index, and resistive index. An “energy” waveform may also be computed, where, for each time interval (e.g. every 50 ms some other suitable time interval), a sum of the pixel values (or squares of the pixel values, or another sum in which each term in the sum is based on a respective pixel value along the frequency (or velocity) axis) is computed. This initial, preoperative energy waveform is labeled as a baseline energy waveform. Once therapy begins, the Doppler waveform is continuously measured at the same location and the aforementioned measurements are repeated during the procedure. Depending on the use case, metrics characterizing the change in flow pattern are extracted at intermittent or periodic intervals and intraprocedurally displayed to the interventionist.

[0035] Referring now to FIG. 1 B, an alternative example method is shown in which steps 100 and 110 proceed according to the method illustrated in FIG. 1A, and where in addition or an alternative to step 120 from FIG. 1 A, step 130 is autonomously performed such that the detection of a change in vascular flow characteristics, such as the detection of a flow signature associated with a vascular event, autonomously triggers the generation and delivery (e.g. communication, transmission) of a control signal or instruction to a therapeutic medical device or therapeutic medical system that is employed to deliver therapy during the procedure. The control signal may be configured, for example, to control or instruct the medical device, or a system controlling the medial device, to interrupt or modulate (modify) the therapeutic intervention, as illustrated in step 130 of FIG. 1 B.

[0036] Referring now to FIG. 1 C, an alternative example method is shown in which steps 100 and 110 proceed according to the method illustrated in FIG. 1A, and where in addition or an alternative to step 120 from FIG. 1 A and / or step 130 from FIG. 1 B, step 140 is autonomously performed such that the detection of the change in vascular flow characteristics (e.g. the detection of a flow signature associated with an adverse vascular event) autonomously triggers the generation and delivery (e.g. communication, transmission) of a control signal to a device or system that is configured to delivery an additional therapeutic intervention that mitigates the impact of the detected change in vascular flow characteristics (e.g. mitigates the impact, severity and / or risk of a detected adverse vascular event). Examples of such therapeutic interventions include vascular robotic, surgical and endovascular interventions including and not limited to vascular interventions performed in the aortic, iliac, femoral, tibial and pedal vessels, and also includes spinal stimulation devices intended for peripheral and / or chronic pain, and extracorporeal membrane oxygenation (ECMO) devices and / or ventricular assist devices (VAD) utilized for the treatment of shock.

[0037] The device or system that is controlled to mitigate the change in vascular flow characteristics (and / or adverse vascular event) may be the same device or system that is employed for delivery of the primary therapy associated with the vascular intervention (provided that such as device or system has the capability to perform the additional intervention), or alternatively, a separate device or system may be controlled to deliver the additional therapeutic intervention.

[0038] In the example autonomous workflows illustrated in FIGS. 1 A to 1 C, the change in vascular flow characteristics that is detected may be an adverse vascular event that is associated within a worsening or a potentially worsening of vascular health of the subject, and / or the potential for co-morbidities associated with or predicated by the occurrence of the detected vascular event. However, in some other example implementations, a detected change in flow characteristics may not necessarily be associated with a worsening of vascular pathology and can instead be associated with an improvement of vascular perfusion as a consequence of the therapeutic intervention.

[0039] In some example embodiments, after having detected an adverse vascular event, additional contextual information associated with the occurrence of the adverse vascular event, such as, but not limited to, a severity of the adverse vascular event, is determined, at least in part, by employing a change in one or more Doppler indices, where the change is evaluated before and after the onset of the detected adverse vascular event. One or more of the Doppler indices may be determined based on the processing of spectral Doppler data. Various examples of the present example embodiment are described below in association with adverse vascular events including emboli, vasospasm and vessel recoil, where changes in one or more Doppler indices are employed to provide additional multimodal contextual information that facilitate an assessment of the severity and / or risk associated with the adverse vascular event. The estimated or inferred severity or risk may be employed when generating feedback associated with the adverse vascular event. In some example implementations, the change in a given Doppler index is determined based measurements of the Doppler index before the onset of the adverse vascular event and during the vascular event. In some example implementations, the change in a given Doppler index is determined based measurements of the Doppler index before the onset of the adverse vascular event and after the conclusion of the vascular event (after the vascular event has ended).

[0040] Non-limiting examples of Doppler indices that may be employed (e.g. based on their pre- and post-event changes) when autonomously and intraprocedurally evaluating the severity and / or risk associated with a detected vascular event include peak systolic velocity, end diastolic velocity, mean velocity, pulsatility index, acceleration time (which can be used to evaluate distal perfusion and detect improvements or impairments following intervention, e.g. pedalacceleration time in cases in which the interrogated peripheral vessel is a pedal vessel), and resistive index (a measure of downstream resistance, which can optionally provide an indirect marker of vasospasm, recoil, or tissue-level ischemia). In some example implementations, at least one Doppler index is an acceleration time and / or a resistive index. In some example implementations, the peripheral vessel is a pedal vessel and at least one Doppler index is a pedal acceleration time, and can additionally include the resistive index.

[0041] It will be understood that in some example implementations, the Doppler indices include one or more standard Doppler indices known the art, while in other example implementations, the Doppler indices additionally alternatively include one or more Doppler parameter values or measures that are not standard or not well- known Doppler indices.

[0042] In some example implementations, the severity or risk associated with the adverse vascular event may be determined, in part, based on one or more non- Doppler-index metrics that characterize an adverse vascular event are which are obtained by processing of one or more spectral Doppler waveforms that correspond to the adverse vascular event. Such metrics can be combined, for example, with the changes in the Doppler indices, as noted above, to provide an estimation of severity and / or risk associated with the adverse vascular event. Non-limiting examples of such metrics in the case of an embolic adverse vascular event include metrics characterizing a quantity and / or type of emboli, an emboli duty cycle, and / or a duration of a detected embolic shower. Non-limiting examples of such metrics in the case of an adverse vascular event involving vasospasm include metrics characterizing an intensity of the vasospasm and / or a duration of the vasospasm. In some example embodiments, the intraprocedural feedback is generated by processing, with a machine learning algorithm (examples of which are described below), at least one metric associated with the adverse vascular event, and at least one additional metric associated with a change in at least one Doppler index before and after an onset of the adverse vascular event.

[0043] In some example implementations, after identifying the onset or occurrence of an adverse vascular event via the processing of Doppler data such as a spectral Doppler dataset, additional monitoring of the Doppler data may be performed to detect a completion of the vascular event. This may be achieved, for example, via the continuous or repeated acquisition and monitoring (processing) ofDoppler data until a flow signature associated with the adverse vascular event is no longer detected. Upon or after the determination that an adverse vascular event has ended, feedback may be provided indicating the completion of the adverse vascular event.

[0044] While many of the example embodiments disclosed herein may be implemented such that the acquisition of Doppler data is performed during the portion of a clinical procedure involving the performing of the vascular intervention, it will be understood that Doppler data acquisition may be additionally or alternatively performed during a post-operative phase of medical procedure, after completion of a vascular intervention, for example, while the patient remains under procedural care, such as, for example, in a surgical suite, recovery room, or intensive care unit. Indeed, while some adverse vascular events may occur during the interventional (operative) phase of a medical procedure, some vascular events, such as, for example, vessel recoil and / or vasospasm, may occur immediately or shortly after intraoperative phase of a procedure.

[0045] Furthermore, while some of the example autonomous workflows described herein involve the detection of a change in vascular flow characteristics via the intraprocedural processing of Doppler ultrasound data characterizing flow within a distal peripheral vessel, where the distal peripheral is downstream, within the vasculature, of a location of a vascular intervention, it will be understood that some example implementations of the present disclosure involve the use of Doppler ultrasound to detect changes in flow characteristics (and / or detect an adverse vascular event) in one or more peripheral vessels in response to a non-vascular therapeutic intervention. For example, as will be described in detail below, the present inventors have found that the application of neurostimulation, such as spinal neurostimulation, can result in changes in the flow characteristics in peripheral vasculature, and the detection of this change in peripheral blood flow can be employed as a feedback measure to monitor and / or control the neurostimulation procedure.

[0046] For example, referring now to FIG. 1 D, an alternative example method is shown in which in which feedback and / or guidance is provided during a neuromodulation (e.g. neurostimulation) procedure based on the intraprocedural Doppler ultrasound characterization of blood flow within one or more peripheral vessels. As shown at step 150, Doppler ultrasound data is acquired from aperipheral vessel during the neuromodulation procedure, where the neuromodulation therapy results in a change in peripheral flow characteristics. This data is dynamically and intraprocedurally processed, such as in real-time or near-real-time, as shown at step 160, to detect changes in vascular flow characteristics via the intraprocedural processing of Doppler ultrasound data characterizing flow within the peripheral vessel. In step 170, the occurrence of a changes in vascular flow characteristics of the interrogated peripheral vessel is communicated to an interventionalist (e.g. user, clinician, nurse), thereby providing the interventionalist with feedback suitable for monitoring and / or controlling the neuromodulation procedure.

[0047] Referring now to FIG. 1 E, an alternative example method involving neuromodulation is shown in which steps 150 and 160 proceed according to the method illustrated in FIG. 1 D, and where in addition or an alternative to step 170 from FIG. 1 D, step 180 is autonomously performed such that the detection of a change in the flow characteristics of the peripheral vessel, such as a change in velocity and / or a change in flow rate, autonomously triggers the generation and delivery (e.g. communication, transmission) of a control signal to the neurostimulation device or neurostimulation system that is employed to deliver neurostimulation during the procedure. The control signal may be configured, for example, to control or instruct the medical device to interrupt or modulate (modify) the neurostimulation intervention, as illustrated in step 180 of FIG. 1 E.

[0048] In the example embodiments illustrated in FIGS. 1 A to 1 E, the intraprocedurally acquired Doppler ultrasound data characterizing blood flow in the one or more peripheral vessels may be acquired, and dynamically and interprocedurally processed to infer and / or detect the presence of a change in vascular flow characteristics of a peripheral vessel, such as the occurrence of a vascular event, according to a wide variety of acquisition, pre-processing and postprocessing algorithms. For example, in some example implementations that are described in detail below, the intraprocedurally acquired Doppler ultrasound data may be acquired and pre-processed to generate a spectral Doppler dataset that is subsequently processed to detect the change in vascular flow characteristics.

[0049] Accordingly, in some example embodiments, a spectral Doppler dataset is intraprocedurally processed to autonomously detect a change in vascular flow characteristics. Spectral Doppler waveform is currently a standard data output formaton commercial Doppler ultrasound scanners and can be accessed by any end-user via DICOM or screen capture. Many of the example embodiments of the present disclose may be implemented by adapting a standard Doppler ultrasound scanner to facilitate the intraprocedural detection of peripheral vascular events by processing the spectral Doppler dataset generated by such scanners according to the methods disclosed herein. Accordingly, many of the example embodiments here may be implemented in a hardware-agnostic manner that is adaptable to any existing Doppler ultrasound system capable of generating a spectral Doppler ultrasound dataset with sufficiently low latency to facilitate continuous intraprocedural processing and feedback according to the systems and methods disclosed herein (e.g. as a retrofit or enhancement).

[0050] In other example implementations, the systems and methods of the present disclosure may be implemented by processing other Doppler ultrasound dataset formats, such as raw Doppler time-series signals, optionally processed according to a pre- and post- framework in which Doppler ultrasound signals in the time domain are compared at time points before and after a vascular event.

[0051] It will be understood that the detection of a vascular event (a peripheral vascular event) may be based on a wide variety of threshold(s), criteria and / or comparative measures. In some example implementations, a change in vascular flow characteristics, and / or a specific vascular event, may be determined to have occurred when one or more conditions are met that are indicative of the presence of a vascular signature in a time-dependent Doppler dataset, such as a time-dependent spectral Doppler ultrasound dataset or a raw time-domain time-series Doppler ultrasound dataset.

[0052] For example, a vascular signature associated with a specific vascular event may be characterized by one or more thresholds associated with one or more features of a Doppler ultrasound dataset, such as one or more parameters representing a sudden change in velocity distribution, peak velocity, or signal intensity. For example, a vascular event may be determined to have occurred when a sudden change in the velocity distribution within a given time window of a spectral Doppler ultrasound waveform satisfies pre-determined criteria, such as one or more aggregate measures. In another example, a vascular event may be determined to have occurred when one or more parameters representing a sudden change slope associated with the time-dependence of a waveform feature of a spectral Dopplerultrasound waveform, such as a peak systolic velocity (PSV) or an end diastolic velocity (EDV).

[0053] In some example implementations, a measure such as a severity, intensity, frequency, or other measure associated with change in vascular flow characteristics, and / or a specific type of vascular event, may be dynamically and intraprocedural determined, in addition or in alternative to the mere detection of a change in vascular flow characteristics or the occurrence of a vascular event. In such cases, such measures may be communicated to the interventionalist and / or may be employed to autonomously control the delivery of a vascular or non-vascular therapy.

[0054] In some example implementations, the detection of a vascular event may represent a discrete vascular event that occurs over a short time interval during a vascular procedure, such as the presence of an emboli. In other example implementations, the detection of a vascular event may represent a blood flow signature that extends over a time interval. For example, the detection of a vascular event may be associated with the onset of a vascular condition, such as, for example, a vessel spasm (vasospasm) or vessel recoil. The vascular condition may persist over time, either for a time duration during the vascular intervention, or both during and after the vascular intervention. Examples of the detection of such persistent vascular events are described in detail below.

[0055] While many of the example embodiments disclosed herein refer to the assessment or characterization of blood flow within a peripheral vessel, it will be understood that Doppler ultrasound data may be collected and processed from two or more vessels.

[0056] In embodiments disclosed herein that involve the use of intraprocedural acquisition of Doppler ultrasound data associated with a distal peripheral vessel impacted by a vascular intervention, it will be understood that the vascular procedure may involve any vascular intervention that is capable of impacting blood flow within the distal peripheral vessel, such as but not limited to, coronary vascular interventions including cardiac stenting, balloon angioplasty, intravascular lithotripsy, valvular interventions / replacements, thrombectomy, and peripheral vascular interventions including stenting, balloon angioplasty, drug-coated balloon angioplasty, intravascular lithotripsy, deep venous arterialization, and otherinterventions such as aortic endovascular aneurysm repairs, aortic stenting, ECMO, left ventricular assist devices, hemodialysis / ultrafiltration.

[0057] It will be understood that the workflows illustrated in FIGS. 1 A to 1 E may be implemented in a fully autonomous manner, provided that the ultrasound Doppler probe is appropriately positioned to detect Doppler ultrasound signals from the peripheral vessel of interest. Referring now to FIG. 2A, an example system is shown that is suitable for performing the methods disclosed herein. The example system includes ultrasound device 330 that includes ultrasound transducer array 340, a transmit beamformer 300 with pulser-receiver circuitry 320, a receive beamformer 310 and control and processing hardware 200 (e.g. a controller, computer, or other computing system). The ultrasound device may be capable of being secured to the subject during the intervention, for example, via an adhesive support and / or one or more straps, thereby enabling the continuous or intermitted measurement of Doppler ultrasound data during the procedure.

[0058] The control and processing hardware 200 may also include, or may be capable of connecting with, an interventional therapeutic system 350, such as an intravascular therapeutic device (e.g. an intravascular therapeutic catheter, or subsystem associated therewith, controllable to perform an intravascular intervention such as, but not limited to, balloon angioplasty (balloon inflation and deflation), atherectomy, intravascular lithotripsy, and stenting. As also shown in the figure, the control and processing hardware 200 may also include, or may be capable of connecting with, an additional therapeutic device or system 360 capable of delivering an additional therapy for mitigating a detected adverse therapeutic event.

[0059] Control and processing hardware 200 is employed to control transmit beamformer 300 and receive beamformer 310, and for processing the beamformed receive signals. As shown in FIG. 2A, in one embodiment, control and processing hardware 200 may include a processor 210, a memory 220, a system bus 205, one or more input / output devices 230, and a plurality of optional additional devices such as communications interface 260, display 240, external storage 250, and data acquisition interface 230.

[0060] The present example methods involving the control of the ultrasound transducer array 340 for performing hemodynamic measurements (e.g. the detection of Doppler ultrasound data) can be implemented via processor 210 and / or memory 220. As shown in FIG. 2A, the control of the ultrasound transducer array 340 may beimplemented by control and processing hardware 200, via executable instructions represented as calculation module 290. The control and processing hardware 200 may include and execute scan conversion software (e.g. real-time scan conversion software).

[0061] In some example implementations, the transducer array 340 is controlled to obtain one or more Doppler ultrasound datasets, as indicated by Doppler processing module 280. In some example embodiments, the transducer array may be controlled to scan the ultrasound beam and generate an ultrasound image, for example, as controlled via image processing module 285. Each Doppler dataset may correspond to a different location, such as a location within an image acquired by the ultrasound transducer array 340. The ultrasound image may be a Doppler ultrasound image.

[0062] In some example implementations, the control and processing hardware 200 may be employed to intraprocedurally process a given Doppler ultrasound dataset to detect intraprocedural changes in blood flow characteristics (including detection of an adverse vascular event), as schematically shown by peripheral blood flow assessment module 290. The blood flow assessment module may also perform calculations to determine one or more peripheral hemodynamic measures, including, but not limited to, acceleration time (i.e. pedal acceleration time in the case of ultrasound interrogation of pedal peripheral vessels), flow volume, peak velocity, beats per minute, and velocity slope from start to end of systole.

[0063] In some example implementations, the ultrasound beam transmitted by the ultrasound transducer array 340 may be scanned in order to identify one or more regions associated with a sufficiently high Doppler signal, such as a location corresponding to an arterial vessel of interest. For example, the transmit beamformer 300 can be controlled to scan the ultrasound beam across a 1 D or 2D angular range (2D if the transducer array is a 2D transducer array) and the Doppler signals may be processed to determine an angle that corresponds to maximal signal, as schematically shown by scanning module 295. In some example implementations, a preferred angle may be determined by processing the collected Doppler signals according to a machine learning algorithm, such as a neural network, that was trained with Doppler signals having a desired shape and / or signal-to-noise ratio.

[0064] The functionalities described herein can be partially implemented via hardware logic in processor 210 and partially using the instructions stored in memory 220. Some embodiments may be implemented using processor 210 withoutadditional instructions stored in memory 220. Some embodiments are implemented using the instructions stored in memory 220 for execution by one or more general purpose microprocessors. In some example embodiments, customized processors, such as application specific integrated circuits (ASIC) or field programmable gate array (FPGA), may be employed. Thus, the disclosure is not limited to a specific configuration of hardware and / or software.

[0065] It is to be understood that the example system shown in FIG. 2A is not intended to be limited to the components that may be employed in a given implementation. For example, the system may include one or more additional processors. Furthermore, one or more components of control and processing hardware 200 may be provided as an external component that is interfaced to a processing device. For example, as shown in the figure, any one or more of transmit beamformer 300 and receive beamformer 310 may be included as a component of control and processing hardware 200 (as shown within the dashed line), or may be provided as one or more external devices.

[0066] While some embodiments can be implemented in fully functioning computers and computer systems, various embodiments are capable of being distributed as a computing product in a variety of forms and are capable of being applied regardless of the particular type of machine or computer readable media used to actually effect the distribution.

[0067] At least some aspects disclosed herein can be embodied, at least in part, in software. That is, the techniques may be carried out in a computer system or other data processing system in response to its processor, such as a microprocessor, executing sequences of instructions contained in a memory, such as ROM, volatile RAM, non-volatile memory, cache or a remote storage device.

[0068] A computer readable storage medium can be used to store software and data which when executed by a data processing system causes the system to perform various methods. The executable software and data may be stored in various places including for example ROM, volatile RAM, nonvolatile memory and / or cache. Portions of this software and / or data may be stored in any one of these storage devices. As used herein, the phrases “computer readable material” and “computer readable storage medium” refers to all computer-readable media, except for a transitory propagating signal perse.

[0069] While many of the example embodiments disclosed herein employ the intraprocedurally processing of Doppler ultrasound data acquired from one or more peripheral vessels to detect a vascular event and / or a change of peripheral blood flow characteristics based on thresholds or criteria associated with time-dependent features or parameters, it will be understood that the preceding example processing methods are intended to be non-limiting and illustrative of a broad range of example methods.

[0070] In some example alternative implementations, Doppler ultrasound data may be intraprocedurally processed using a machine learning algorithm to detect the presence of a vascular event. For example, the adverse vascular event can be detected via a machine learning algorithm configured to receive, as an input, a spectral Doppler waveform segment, and to generate output comprising a classification of the type of adverse vascular event among a plurality of types of adverse vascular events. In such an example embodiment, the machine learning algorithm can be trained according to training spectral Doppler waveform data associating waveform signatures with labeled adverse vascular events of different types.

[0071] In one example implementation, an ultrasound Doppler spectrogram is analyzed via multiple streams to extract different Doppler indices. For example, a first processing stream detects the presence or absence of emboli, a second processing stream calculates the resistance index after extracting the peak velocity profile, and a third processing stream analyzes the spectrogram to compute the acceleration time (e.g. the pedal acceleration time in the case of interrogation of a pedal vessel). Optionally, the peak velocity profile could be first be extracted and the resistance index and acceleration time could be measured. After these measures have been computed, they can be fed into a machine learning classifier to perform multi-modal classification, where the inputs are the different Doppler indices. In one example implementation, the output is a binary outcome variable indicating whether or not the patient should progress to intervention. Non-limited example types of the machine learning classifier include random forest, logistic regression, a fully connected neural network, and XGBoost. In some example applications, additional metrics are used as input. As an example, venous arterialization monitoring can be implemented to use additional features such as peak systolic velocity, end diastolic velocity, resistive index, and flow volume.

[0072] In another example implementation, reference clinically adjudicated Doppler ultrasound data, such as spectral Doppler data, may be labeled with expert user labels associating the reference data with known vascular events such as, for example, the presence of emboli, vasospasm and / or recoil. The reference datasets may also be labeled to indicate a change in therapy (e.g. an increase or decrease) was employed to mitigate the vascular event. If such a labeled data set is available, a machine learning algorithm can be trained to process intraprocedurally acquired unlabeled Doppler data and output one or more measures such as, for example, a presence / absence of emboli, a number of peripheral emboli, a duty cycle associated with peripheral emboli, a presence / absence of recoil and / or vasospasm, a severity level of associated with recoil and / or vasospasm, a determination of whether or not the interrupt the intervention, a determination of a suitable modification of the therapy, and a determination of a suitable additional therapeutic intervention (and optionally a degree of such an intervention), for mitigating the detected vascular event.

[0073] In one non-limiting example, a spectral Doppler waveform, represented by a 2D image with velocity along the vertical axis and time along the horizontal axis, can serve as the input image to a convolutional neural network (CNN), as illustrated in FIG. 2B. Examples of CNNs include VGG-16, DenseNet, AlexNet etc. Typically, CNNs require a pre-determined input image size. Hence, the spectrogram will be resampled using image interpolation to the expected size (e.g. 224 by 224 for VGG- 16). Also, many CNNs are designed to work with 3-channel RGB images. With a grayscale image like the spectrogram, the image is replicated to the three channels.

[0074] FIG. 2B shows an example of a standard CNN architecture. The input on the left may be the Doppler spectrogram (replicated to the 3 channels). The input spectrogram will be of a fixed time length, for example, such as 5 seconds. The CNN may be employed to perform classification of the Doppler spectrogram to determine the presence or absence of one or more vascular events.

[0075] In another example embodiment, the softmax and output layers of the CNN may be replaced with a regression layer that will provide a continuous output. In one example implementation, the regression output layer may contain multiple outputs, each being a scalar value. For example, in the case of a CNN trained to perform emboli detection, one scalar output may provide a determination of the number of detected emboli, and a second scalar output may provide a determination of a dutycycle associated with the presence of emboli over the time duration of the input Doppler dataset.

[0076] In such an example, the training data may consist of an expert clinician or sonographer manually reviewing the spectrogram of Doppler signals obtained in the presence of emboli. Each time an emboli occurs, the dataset is annotated with the time of occurrence and its duration by placing cursors on a screen where the spectrogram is displayed. The total number of emboli will be recorded as will the total time duration. The duty cycle will be calculated as the ratio of total emboli time over the dataset time interval (e.g. 5 seconds). Both the number of emboli recorded and the duty cycle will be the ground truth for the CNN above.

[0077] In this example case, the CNN is trained to predict the scalar outputs described above. During inference, the input is a spectral Doppler dataset and the output will be the numerical values. A data driven approach with machine learning will allow the system to work in a variety of conditions including low SNR, artifacts etc. Also, to increase robustness of the system, augmentation can be used during training to increase the number of input spectrum samples. In this case, the spectral doppler waveforms are injected with noise maintaining the same corresponding ground truth output.

[0078] The preceding example embodiments, and variations thereof contemplated by the present disclosure, may be beneficial in addressing problems associated with need for skilled ultrasound sonographers during vascular ultrasound procedures. Indeed, the embodiments provided herein enable the autonomous processing of intraprocedurally acquired Doppler ultrasound data acquired from one or more peripheral vessels to autonomously provide feedback, guidance and / or therapeutic device control during a therapeutic procedure, thus significantly reducing the required skill level and experience of the user.

[0079] While some of the systems and methods described and illustrated herein refer to applications involving pedal wounds and pedal ultrasound vascular diagnostic assessment during a therapeutic procedure, it will be understood that the embodiments described herein can be employed or adapted to be employed such that intraprocedural Doppler ultrasound is acquired from peripheral vasculature in other anatomical regions, such as, but not limited to, the hand.Intraoperative Doppler Ultrasound Detection of Emboli in of Peripheral Vessels During Peripheral Vascular Procedures

[0080] In some example implementations, the aforementioned systems and methods are adapted to facilitate the intraprocedural detection of emboli in one or more peripheral vessels (for example, peripheral pedal vessels) during a therapeutic intervention. Prior to present disclosure, clinicians were not able to determine whether or not embolization occurred within the pedal vasculature during a peripheral vascular intervention (such as balloon angioplasty (inflation and deflation), atherectomy, intravascular lithotripsy, and stenting), and the rate of emboli occurrence, volume of emboli, was also unknown. Pedal embolization was hypothesized to impact pedal vessel flow and wound healing, however there is no current automated real-time methodology to quantify the emboli. Emboli, which arise from a non-natural process due to a therapeutic intervention, as opposed to as natural response of the body, can be, for example, atherosclerotic plaque or thrombus that is impacted or moved with devices which releases downstream and potentially creates occlusions in the pedal arch or microcirculation, often leading to poor outcomes.

[0081] Accordingly, in some example implementations, the preceding embodiments are adapted to facilitate the detection of embolic events occurring in the peripheral vasculature due to interventional therapies (e.g. targeted in the vessels proximal to the Doppler ultrasound sensor), where the example embodiments disclosed above are adapted to facilitate, for example, a characterization of emboli presence or burden via flow disturbance analysis, including detection of transient high-frequency components or specific waveform patterns.

[0082] In some example implementations, systems and methods are provided that intraprocedurally determine (e.g. in real-time or near-real-time), the number of embolic hits, the size of the particle hits, and differentiation between artifacts (not true hits) and true embolic hits. In some example implementations, the Doppler ultrasound data is processed to categorize a detected emboli according to various embolic morphologies, for example, using thresholds associated with measures obtained by processing the spectral Doppler data within a specific time window, and / or based on the use of machine learning (as described) above to detect embolic signatures associated with different sizes of emboli (e.g. large embolic events andsmaller / micro embolic events). Intraprocedural peripheral embolic detection and optional differentiation could facilitate a realization of clinical worsening (or potential thereof) to wound perfusion via larger emboli compared to micro embolic hits. For example, such a system could detect the presence of a “shower” or gross amount of embolic hits, which could prompt a clinician to halt the therapy or lead the clinician to provide medical therapy such as TPA (blood thinning agent) to dissolve the thrombus particles. In some example implementations, the present example methods and systems can provide continuous autonomous ultrasound monitoring of statistical and / or aggregate measures relating to peripheral emboli, such as, but not limited to, total number of embolic hits that have occurred throughout the procedure, thereby providing useful intraprocedural insights for a clinician.

[0083] In one non-limiting example implementation involving the use of Doppler ultrasound to achieve intraprocedural detection, during a vascular intervention, of emboli in peripheral vasculature, the following steps may be performed. Firstly, a baseline doppler image is acquired and PSV and EDV are recorded. A continuous Doppler dataset (waveform) is activated “live” when therapy is performed (e.g. balloon angioplasty, atherectomy, vascular laser therapy, or stenting). The spectral Doppler waveform is collected (optionally such that the data occupies approximately % of the spectrum), optionally with a sweep speed reflecting 3 to 4 cardiac cycles. The waveform gain may be configured to detect embolic signatures. Embolic signatures are characterized by a vertical, bright echogenic line above the base line, as determined based on autonomously processing the spectral Doppler data within a plurality of time windows and applying suitable thresholding for emboli detection.The number of embolic hits are autonomously counted during (e.g. throughout) therapy and communicated to the clinician. Post therapy PSV and EDV are recorded and compared to the baseline values.

[0084] In one non-limiting example implementation involving the use of Doppler ultrasound to achieve intraprocedural detection, during a vascular intervention, of emboli in peripheral vasculature, the following steps are performed. The spectral Doppler waveform is computed by taking a FFT of the ultrasound Doppler signal, for example, as described in Ubeyli, E. et al., Applications of FFT and ARMA Spectral Analysis to Arterial Doppler Signals, Mat. and Comp. App. 8, 311 (2003). Typically, the FFT length is 128 and the FFT calculation is repeated on overlapping windows of length 25 ms with 50% overlap. The magnitude of the FFT is then computed andsquared to produce the spectral power. This produces the 2D matrix where the vertical axis is the frequency bins, and the horizontal dimension is the time step of 12.5 ms. For each step, the values in the vertical dimension (i.e. frequency) are summed, thereby generating the “energy” waveform characterized by a sum (scalar) value at each time interval. In the preceding calculation, the underlying data that is evaluated is the Fourier spectrum of the Doppler shift as a function of time (equivalent to a spectrogram in signal processing). Thus, at each time interval, what is plotted is the spectral energy density - each pixel represents the energy in a frequency bin: brighter the pixel, higher the energy in that frequency bin. The summation thus provides an energy measure within each time interval. These scalar values can be assembled as a time series. Two such time-series can be created: one for the pre-therapy phase, and another for therapy and post-therapy phase. Typically, the sum time-series would each be a few seconds long to contain multiple cardiac cycles.

[0085] As an optional initial step, a determination may be made as to whether or not an artifact may be present in each time window. For example, this may be performed by: (i) summing along the positive side of the waveform to obtain a first number, (ii) summing along the negative side of the waveform to obtain a second number, (iii) subtracting the negative from the positive value to obtain a difference, and (iv) identifying an artifact if the difference falls in within a range known to be associated with artifacts.

[0086] If the time window is not found to represent an artifact, the energy measure within the time interval is further processed to determine if it is indicative of the presence of an emboli. This may optionally be performed via comparison with a pre-therapy waveform. In an example of such a comparative method, the start and end of each cardiac cycle are computed from the waveform. This is performed by extracting the waveform envelope (using the tracing algorithms in the literature). From the envelope, the end-diastole of each cycle is computed. This process is repeated on the baseline energy waveform as noted above. For the baseline energy waveform, the average energy profile is calculated across 5 cycles. The energy profile for each cycle in the treatment / post-treatment time frame is then calculated. In an optional step, a correlation coefficient between the energy envelope of the posttreatment and the baseline trace is computed. If the correlation coefficient is above apre-selected threshold, such as 0.9, this indicates that the two waveforms can now be compared for further evaluation.

[0087] Emboli detection is then performed on a per-time-window basis. For example, in the present example comparative method, the difference between the baseline average waveform (per cycle) and each cycle of waveform envelope of the post-intervention period may be computed for each time window. If the difference exceeds a threshold value, the time window is identified as corresponding to the presence of an embolic event.

[0088] The number of emboli events per cardiac cycle can be displayed as an output trace. When the number of emboli events exceed a threshold set by the clinician, an alarm may be generated that informs the interventionalist to stop the therapy. Other parameters, such as, but not limited to, the emboli duty cycle (percentage of the time windows associated with emboli) may also be calculated. Such parameters may optionally be employed to determine when to interrupt the procedure. For example, when the emboli duty cycle exceeds 20%, an alert is generated cautioning the clinician to interrupt the treatment, or a control signal id generated and deliverer to the treatment device to automatically interrupt the treatment.

[0089] Some of the aforementioned methods have involved the processing of the spectral Doppler waveform. Alternatively, if the raw time domain Spectral doppler gate signal is available, other time-domain techniques such as cross-correlation for time delay estimation at two different positions along the vessel can also be used to determine if it is an artifact or not. Unlike the case of transcranial Doppler, in which spectral gates are typically placed at two different depth positions, the gates in the present example implementation could be placed laterally, and with the ultrasound transmit beam switched between the two lateral locations along the vessel for Doppler acquisition. Accordingly, in other example implementations that do not involve the processing of spectral Doppler data, the raw time-domain Doppler gate signal (pre-FFT) can be processed using time-domain techniques such as, but not limited to, cross-correlation at two different positions along the vessel, which would include employing two Doppler windows at different locations along the vessel.

[0090] Referring now to FIG. 3A and 3B, color flow and spectral Doppler image respectively of a blood vessel with emboli within (represented by the bright white vertical “streaks” on the Doppler waveform. FIGS. 4A, 4B and 4C show color flowand spectral Doppler image respectively with emboli. The flow profile is dominated by emboli in FIGS. 4A-4C, as compared to FIGS. 3A and 3B that show a combination of blood flow and emboli.

[0091] In some example implementations, after having detected the onset and or presence of emboli, additional contextual information associated with the occurrence of the presence of the emboli, such as, but not limited to, a severity of the adverse vascular event, is determined, at least in part, by employing a change in one or more Doppler indices, where the change is evaluated before and after the onset of the detected adverse vascular event, and this additional contextual Doppler-index-based information may be employed, according to a multi-modal processing workflow, to estimate or infer a severity and / or risk associated the presence of the emboli, and can optionally be employed to infer whether distal perfusion is compromised.

[0092] For example, in some example implementations, changes in acceleration time (e.g. pedal acceleration time) and resistive index, before and after the onset of the presence of emboli, are employed to generate a measure associated with a severity and / or risk associated the presence of the emboli, where a prolongation in acceleration time (e.g. pedal acceleration time) and / or a change in resistive index are employed as clinically significant indications (e.g. weighted as being associated with increase severity when generating a composite severity measure) as they indicate a possible compromise in downstream perfusion, and can inform, guide or suggest immediate clinical intervention. This assessment can be employed to therapeutic interventions such as thrombolysis or aspiration. Conversely, in some example implementations, emboli detected without corresponding changes in acceleration time (e.g. pedal acceleration time) or resistive index, or with detected with changes below prescribed thresholds are treated as potentially subclinical and processed (e.g. weighted) to lower the severity and / or risk assessment. In such cases, such lower-risk classification can enable a clinician to deprioritize or monitor the emboli without further therapeutic intervention.Intraoperative Doppler Ultrasound Detection of Vasospasm and Vessel Recoil of in Peripheral Vessels During Peripheral Vascular Procedures

[0093] Vessel spasm, also known as vasospasm, is a sudden, involuntary contraction of the smooth muscle in the walls of a blood vessel, causing it to narrow or constrict. This phenomenon can occur during or after vascular interventions, suchas angioplasty, stenting, or catheterization procedures. Vessel spasm can significantly affect the success of the intervention and the patient's outcome by reducing blood flow through the affected vessel.

[0094] Vessel recoil refers to the natural tendency of a blood vessel to return to its original diameter after being expanded or dilated during a vascular intervention, such as angioplasty. Vessel recoil essentially involves the clamping down of the artery. For example, with radial or tibial access of 2-3mm arteries, once a sheath is removed, vessel recoil can occur with the media tunica of the arterial (muscular wall) reacts and “clamps shut”. When this occurs, intra-arterial medication or a balloon is delivered to open the artery. This phenomenon is an important consideration in vascular interventions because it can affect the long-term success of a vascular procedure. During an angioplasty, a balloon catheter is used to dilate a narrowed or blocked blood vessel, typically due to atherosclerosis. However, once the balloon is deflated and removed, the vessel may partially collapse or "recoil" back toward its pre-dilated state. This can lead to a reduction in the lumen diameter, which can compromise blood flow and potentially lead to restenosis (re-narrowing of the vessel). Vessel recoil can be a devastating result if complete shutdown of an artery occurs.

[0095] If vasospasm or recoil are detected during a procedure, appropriate therapy can be delivered prior to the conclusion of the procedure and the transfer of the patient to recovery. Once in recovery, there are minimal tools to reopen the artery that shut down, other than medicine. In the example case of one artery supplying the foot with the presence of significant recoil, having access to balloons, stents, can avoid severe ischemia which can lead to possible future amputation.

[0096] While vessel recoil and vasospasm (vessel spasm) are known to be common during vascular interventions, there is currently an absence of automated solutions that perform autonomous intraprocedural detection of vasospasm and / or recoil. Accordingly, there is currently an absence of available solutions for the automatically detection of the occurrence of vessel spasm / recoil, the severity of vessel recoil, the duration of vessel recoil and the resolution of vessel spasms / recoil. Moreover, prior to the present disclosure, the present inventors were not aware of the potential utility of intraprocedural monitoring of peripherical vasculature, such as within the pedal system window, to determine the onset or occurrence of vasospasm and recoil.

[0097] Furthermore, it is noted that while transcranial Doppler (TCD) has been employed to perform vasospasm detection, the conventional TCD workflow for vasospasm detection involves the use of separate measurements in different arteries at different points in time to detect vasospasm, with detection happening after the vasospasm has already occurred due to an increased velocity in a monitored artery. For example, in the conventional approach to TCD-based vasospasm detection, the blood flow in the middle cerebral artery is compared to that in the distal internal carotid artery, and a ratio is obtained (the Lindegaard ratio), which is assessed to infer the presence of vasospasm. In stark contrast to this conventional method, the methods of the present disclosure, when adapted for vasospasm detection, involve the use of Doppler ultrasound monitoring of blood flow characteristics in peripheral vasculature, during a procedure involving a vascular intervention, to intraprocedurally detect the onset of a vessel spasm, thereby enabling the use of intraprocedural therapy to treat the vasospasm soon after it occurs, and to potentially improve the overall outcome of the procedure.

[0098] The present inventors have found that vasospasm and recoil can be intraprocedurally detected, via Doppler ultrasound monitoring of a peripheral vessel distal to an intervascular intervention, by detecting a spectral Doppler waveform signature indicative of the onset of vasospasm or recoil, as described in further detail below. Moreover, the present example methods facilitate ongoing assessment of vasospasm and recoil duration to guide therapeutic windows, and facilitate confirmation of resolution, helping determine treatment efficacy. Accordingly, embodiments of the present disclosure can be adapted to facilitate the intraprocedural detection of vasospasm and recoil, and may be further adapted, for example, to quantify the percentage of change in flow rates associated with vasospasm. This percentage of change in flow, measured as a difference prior to and after the onset of vasospasm, can be employed by an interventionalist to determine whether or not to provide medical therapy such as nitrogycerin to relax the affected vessels.

[0099] As can be seen in FIGS. 5A and 5B, the onset of a vasospasm is characterized, in a spectral Doppler waveform, by the presence of an increasing peak velocity in successive cardiac cycles, generating a “crescendo” vascular flow signature. This flow signature can be detected according to a wide variety of processing algorithms, with non-limiting algorithm examples described below.

[0100] According to one example workflow, spectral Doppler data is acquired during a vascular intervention based on the interrogation of a distal vessel, and the spectral Doppler dataset is processed over a time span that includes several consecutive cardiac cycles. The envelope can be computed with established methods in the literature, such as, for example, the method described in Kathpalia, A., et al., Adaptive Spectral Envelope Estimation for Doppler Ultrasound, IEEE Trans Ultrason Ferroelectr Freq Control 63, 1825-1838 (2016).

[0101] According to the present example vasospasm detection method, the peak detection algorithm is applied on the envelope to compute the peak systolic velocity (PSV). Knowing the peak for cardiac cycle, the end diastolic velocity is computed. This is repeated for each cardiac cycle. From this history of PSV values, a linear fit can be computed on the PSV values for a set of consecutive cardiac cycles (e.g. 3, 4, 5 or 6 cardiac cycles). The slope of the fit to the PSV values can be evaluated and compared to a threshold value, such that if the slope exceeds the threshold, a determination is made that the onset of vasospasm has occurred, while normal flow is inferred if the slope of the consecutive PSV values is below the threshold. In another example implementation, the beat-to-beat PSV change is computed, and if the difference in PSV values exceeds a threshold value for N successive cardiac cycles, a determination is made that a vasospasm has occurred. In other example vasospasm detection implementations, a machine learning algorithm can be employed to detect the presence of the crescendo blood flow signature in the spectral Doppler data, which can be implemented, for example, as per the example machine learning methods described above.

[0102] Considering now the adaptation of the methods of the present disclosure for the intraprocedural detection of recoil, it is noted that recoil is typically associated with a cessation of flow after a few cardiac cycles. Accordingly, a recoil signature in a spectral Doppler waveform is typically associated with a decrease in flow velocity to near-zero PSV, with the absence of a detectable diastolic portion of the waveform when compared to a baseline pre-intervention spectral Doppler waveform. After the occurrence of vessel recoil, the spectral Doppler waveform reduces to a staccato signature, characterized by a series of “blips” on the spectrogram with high distal resistance, as illustrated, for example, in FIG. 6. As can be seen in the figure, vessel recoil has resulted in muted spectral Doppler waveform that is absent of detectable diastolic flow.

[0103] Recoil can therefore be intraprocedurally identified by monitoring the spectral Doppler data acquired from a distal peripheral vessel impacted by a vascular intervention, such that when a beat-to-beat changes in the PSV are significantly reduced or the flow / waveform ceases to exist, at least within the diastolic portion, the presence of recoil is detected. Such a spectral Doppler flow signature can be detected, for example, according to a wide variety of processing methods, including methods that monitor for a reduction in PSV, beyond a threshold reduction, over a set of cardiac cycles, or, for example pattern detection methods such as machine learning detection of the aforementioned staccato recoil flow signature.

[0104] In some example implementations, after having detected the onset and or presence of vasospasm or vessel recoil, additional contextual information associated with the occurrence of the adverse vascular event, such as, but not limited to, a severity of the adverse vascular event, is determined, at least in part, by employing a change in one or more Doppler indices, where the change is evaluated before and after the onset of the detected adverse vascular event, and this additional contextual Doppler-index-based information may be employed, according to a multi-modal processing workflow, to estimate or infer a severity and / or risk associated the presence of the vascular event, and can optionally be employed to infer whether or not an intervention is warranted. For example, in the example case of a detected vasospasm event, an elevated resistive index with a corresponding prolongation of acceleration time (e.g. pedal acceleration time) may suggest a more severe vasospasm that requires or could benefit from therapy, such as use of a vasodilator. In the example case of a detected vessel recoil event, acute or significant (e.g. satisfying pre-established criteria or threshold(s)) changes in resistive index or acceleration (e.g. pedal acceleration time) after angioplasty can suggest elastic recoil, and can inform, depending on the determined severity, a need or benefit involving use of therapeutic intervention such as ballooning.

[0105] In some example implementations, the severity of vessel recoil may be measured based on the pre-procedure baseline PSV and waveform. For example, if the intraprocedural PSV reduced below a selected threshold, such as 5 cm / sec or less, a determination of severe recoil may be made, while other thresholds may be established and employed to measure or infer other levels of recoil severity.Guidance of Spinal Stimulation and Neuromodulation Procedures via Doppler Ultrasound Monitoring of Peripheral Vessels

[0106] Spinal stimulators provide electrical current focused on the spine to treat neuropathic pain. The present inventors have made the discovery that such spinal stimulation treatment has a direct impact on the peripheral vascular flow. The present inventors were able to determine, based on clinical studies, that the activation of spinal stimulation leads to an increase in peripheral vascular flow, as experimentally confirmed via intraprocedural Doppler ultrasound assessment of peripheral vascular perfusion in the pedal vasculature. In particular, the present inventors found that the application of spinal stimulation was found to modify Doppler-derived hemodynamic parameters such as pedal acceleration time, peak and end diastolic flow velocity, resistive index, and pedal vascular flow volume. It was experimentally found that adjustment of the electrical dosage leads to immediate real-time pedal vascular flow changes.

[0107] Accordingly, in some example embodiments, feedback and / or guidance is provided during a neurostimulation procedure, such as a spinal stimulation procedure, based on the detection of a change in vascular flow characteristics of a peripheral vessel via the intraprocedural processing of Doppler ultrasound data characterizing flow within a peripheral vessel, as illustrated, for example, in FIGS. 1 D and 1 E. For example, detected vascular flow changes can be employed to direct, in an autonomous manner, the dosage or intensity of neurostimulation therapy, such as spinal stimulation, according to the detected vascular flow characteristics. For example, a pre-determined relationship between neurostimulation dosage and / or intensity and one or more peripheral vascular flow measures detectable via peripheral Doppler ultrasound may be employed to modify neuromodulation therapy. For example, spine stimulation devices have adjustable settings to control pain, and these settings can be changed according to peripheral vascular flow characteristics detected via Doppler ultrasound.

[0108] In one non-limiting example workflow, feedback and / or guidance is provided during a neurostimulation procedure, such as a spinal stimulation procedure, based on the detection of a change in vascular flow characteristics of a peripheral vessel via the intraprocedural processing of Doppler ultrasound data characterizing flow within a peripheral vessel, according to the following example method. With the ultrasound transducer in position to interrogate a peripheral vessel,such as peripheral pedal vessel, a baseline arterial spectral Doppler waveform is measured prior to the application of neurostimulation (e.g. spinal stimulation). One or more Doppler hemodynamic indices may be measured prior to the application of neurostimulation, such as, for example, acceleration time (e.g. pedal acceleration time), PSV and EDV. As seen in the figure on the right, the waveform will be characterized by a low PSV, low resistance index, and a large PAT value. These values may be stored and a historical record may be created with values recorded for each cardiac cycle within an initial waveform.

[0109] As noted above, the present inventors have found that the application of spinal stimulation results in increased peripheral blood flow due to vasodilation. This can result in marked differences in the Doppler flow characteristics and flow signatures, and the aforementioned methods of Doppler data processing for vascular flow characterization can be employed to detect changes induced by the applied stimulation. In some example implementations, indices including PSV, EDV, Rl, PI and PAT (pedal acceleration time), as shown in the FIGS. 7A and 7B, can be measured and employed to detect and / or quantify changes in flow characteristics induced by the application of the stimulation (FIG. 7B shows the spectral Doppler waveforms before stimulation, while FIG. 7A shows the spectral Doppler waveform after stimulation). It is noted that the pedal acceleration time is significantly reduced in FIG. 7A compared to FIG. 7B due to a stronger flow in the former resulting from the stimulation. In such implementations, the difference between the hemodynamic (flow characterizing) values for each of the indices can be computed and displayed as a rolling waveform. Such a multi-channel display can be provided to the interventionist throughout the procedure.

[0110] In addition, when the change in one or more aspects of the detected flow characteristics (e.g. a parameter or index, such as, for example, acceleration Time (e.g. pedal acceleration time), peak systolic velocity (PSV), and resistive index (Rl)) exceeds a predetermined threshold, this can be associated with a clinical concern, for example, due to sudden increased vascularization. In such an instance, it may be appropriate or beneficial to reduce the therapy energy, and this reduction may be recommended, and / or autonomously controlled, according to the ratio between the Doppler indices before and during the stimulation. As the stimulation energy is reduced, the quantitative information from the doppler indices is used to modify the therapy.

[0111] While the present example embodiment illustrates the intraprocedural processing of Doppler ultrasound data characterizing flow within a peripheral vessel to monitor and / or modify spinal stimulation, it will be understood that the present example embodiment may be adapted for use with a wide variety of neuromodulation devices, including, but not limited to, electrical spinal stimulation, spinal drug delivery, deep brain stimulation, vagus nerve stimulation, peripheral nerve stimulation, transcutaneous electrical nerve stimulation, transcranial magnetic stimulation and other neuromodulation devices, provided that the Doppler ultrasound measurements are performed on a peripheral vessel that exhibits blood flow characteristics that are responsive to the neuromodulation therapy. Moreover, it will be understood that while the present example embodiments employ peripheral vessel Doppler ultrasound as a diagnostic modality to monitor and / or control neuromodulation, the neuromodulation modality may take on many forms, including, but not limited to electrical, acoustic (ultrasound), magnetic and optical (e.g. infrared) neurostimulation modalities. The skilled artisan can determine, through routine experimentation, which neuromodulation modalities cause detectable changes in blood flow characteristics in peripheral vessels, and which peripheral vessels are suitable for Doppler ultrasound detection for a given neuromodulation modality.

[0112] In some example implementations involving the use of Doppler ultrasound to detect neuromodulation-induced changes in the flow characteristics of a peripheral vessel, a user interface may be provided that facilitates control of the neuromodulation and also displays the detected blood flow measures, and optionally includes one or more autonomous feedback modules that control / lim it the applied neuromodulation based on the detected blood flow measures.

[0113] In some example implementations, a threshold associated with autonomous interruption of neuromodulation based on neuromodulation-induced changes in the flow characteristics of a peripheral vessel detected via Doppler ultrasound, and / or a pre-determined relationship between applied stimulus and one or more flow indices or parameters generated by intraprocedurally processing Doppler ultrasound data, may be customized to a specific patient. For example, different patients will have different severity of pathology, different pain thresholds, different physiological responses to a given neuromodulation invention, and / or other relevant differences, which may result in a benefit to the patient of such customization.

[0114] It will be understood that a medical procedure pertaining to neuromodulation (e.g. neurostimulation) may take on many forms without departing from the intended scope of the present disclosure. For example, some neuromodulation procedures involve the use of an extracorporeal neuromodulation device that is contacted with the patient, or proximally positioned relative to the patient, on a temporary basis during the delivery of neuromodulation therapy (for example, but not limited to, transcranial magnetic stimulation and transcutaneous electrical nerve stimulation, external vagus nerve stimulation and photobiomodulation). In other example cases, neuromodulation may be performed using an implantable device, such as, for example, spinal cord stimulators and deep brain stimulators), and in such cases, a neuromodulation procedure may include any procedure involving the implantation or subsequent use or modification of the implanted device. For example, a Doppler-ultrasound-assisted neuromodulation medical procedure involving an implanted neuromodulation device may include any one of more of (i) a medical procedure involving the initial implantation of the device, (ii) any therapeutic use of the device, (iii) procedures involving adjustments, calibrations or other modifications, and (iv) procedures involving removal of the device, provided that Doppler ultrasound is employed to monitor peripheral blood flow.Intraoperative Doppler Ultrasound Detection of Peripheral Vessel Flow During Peripheral Vascular Balloon Procedures

[0115] In another example embodiment, the preceding example methods and workflows may be adapted for use during a peripheral vascular intervention such as proximal ballooning of an artery. It can be important to achieve correct vessel sizing for balloon angioplasty. If a balloon is undersized, it may not provide a clinical benefit, while if a balloon is oversized, it can cause dissection. Accordingly, in some example implementations, a distal peripheral vessel downstream of a location of where balloon angioplasty is to be performed, is intraprocedurally assessed via Doppler ultrasound to detect changes in blood flow characteristics during balloon angioplasty therapy. For example, when the balloon is delivered to a given artery (e.g. an anterior tibial balloon), a relevant downstream peripheral vessel (e.g. the dorsal is pedis or arcuate artery) is monitored according to the methods described above. If autonomous processing of the Doppler data results in a determination thatflow has ceased with inflation of the balloon, a warning is communicated and / or a control signal is autonomously generated and delivery to the balloon catheter to cease further inflation. If flow is detected as not having changed or only diminishing (as per a suitable pre-established threshold), then an alert is communicated that the balloon sizing may not be appropriate.Enumerated Embodiments

[0116] Embodiment 1 . A method of monitoring and detecting an adverse vascular event via Doppler ultrasound interrogation of peripheral vasculature, the method comprising: intraprocedurally acquiring Doppler ultrasound data from a peripheral vessel during a medical procedure involving a vascular intervention, wherein blood flow within the peripheral vessel is downstream from a location of associated with the vascular intervention; intraprocedurally and autonomously processing the Doppler ultrasound data to identify a flow signature associated with a presence of an adverse vascular event; detecting an occurrence of the adverse vascular event; and generating intraprocedural feedback associated with the adverse vascular event.

[0117] Embodiment 2. The method according to embodiment 1 further comprising, after detecting the occurrence of the adverse vascular event, determining a severity of the adverse vascular event, wherein the severity of the adverse vascular event is determined, at least in part, by employing a change in one or more Doppler indices before and after the onset of the adverse vascular event; wherein the intraprocedural feedback is generated based on the severity of the adverse vascular event.

[0118] Embodiment s. The method according to embodiment 2 wherein one or more Doppler indices are generated based on processing spectral Doppler data.

[0119] Embodiment 4. The method according to embodiment 2 wherein the flow signature is detected based on processing spectral Doppler data.

[0120] Embodiment 5. The method according to embodiment 2 wherein the change in at least one Doppler index is evaluated based on a first value of the Doppler index obtained before the adverse vascular event and a second value of the Doppler index obtained during the adverse vascular event.

[0121] Embodiment 6. The method according to embodiment 2 wherein the change in at least one Doppler index is evaluated based on a first value of the Doppler index obtained before the adverse vascular event and a second value of the Doppler index obtained after the completion of the adverse vascular event.

[0122] Embodiment 7. The method according to any one of embodiments 2 to 6 further comprising processing a spectral Doppler waveform associated with the adverse vascular event to determine a quantitative metric characterizing the adverse vascular event, and further employing the quantitative metric to infer the severity of the adverse vascular event.

[0123] Embodiment 8. The method according to any one of embodiments 1 to 7 wherein the adverse vascular event is detected via a machine learning algorithm configured to receive, as an input, a spectral Doppler waveform segment, and to generate output comprising a classification of the type of adverse vascular event among a plurality of types of adverse vascular events, the machine learning algorithm having been trained according to training spectral Doppler waveform data associating waveform signatures with labeled adverse vascular events of different types.

[0124] Embodiment 9. The method according to any one of embodiments 2 to 7 wherein the intraprocedural feedback is generated by processing, with a machine learning algorithm, at least one metric associated with the adverse vascular event, and at least one additional metric associated with a change in at least one Doppler index before and after an onset of the adverse vascular event.

[0125] Embodiment 10. The method according to any one of embodiments 2 to 9 wherein the one or more Doppler indices comprise one or more of a peak systolic velocity and an end diastolic velocity.

[0126] Embodiment 11 . The method according to any one of embodiments 2 to 9 wherein the one or more Doppler indices are selected from the group consisting of an acceleration time and a resistive index.

[0127] Embodiment 12. The method according to any one of embodiments 2 to 9 wherein the one or more Doppler indices comprise an acceleration time and a resistive index.

[0128] Embodiment 13. The method according to any one of embodiments 2 to 9 wherein the peripheral vessel is a pedal vessel and the one or more Doppler indices comprise a pedal acceleration time.

[0129] Embodiment 14. The method according to embodiment 13 wherein the one or more Doppler indices further comprise a resistive index.

[0130] Embodiment 15. The method according to any one of embodiments 1 to 14 wherein the flow signature is indicative of an onset of the adverse vascular event.

[0131] Embodiment 16. The method according to any one of embodiments 1 to 9 further comprising: detecting a completion of the adverse vascular event; and providing additional feedback indicating the completion of the adverse vascular event.

[0132] Embodiment 17. The method according to any one of embodiments 2 to 6 wherein the adverse vascular event is associated with detection of one or more emboli.

[0133] Embodiment 18. The method according to embodiment 17 wherein an embolic event is identified via detection, within a given time window of a spectral Doppler waveform, of a bright echogenic vertical line satisfying emboli detection criteria.

[0134] Embodiment 19. The method according to embodiment 17 wherein an embolic event is identified via processing a spectral Doppler waveform to generate an energy waveform and comparing the energy waveform with a previously measured pre-intervention energy waveform known to be absent of emboli.

[0135] Embodiment 20. The method according to embodiment 17 wherein emboli detection is performed via a machine learning algorithm configured to receive, as an input, a spectral Doppler waveform segment, and to generate output comprising at least one scalar measure characterizing detected emboli.

[0136] Embodiment 21 . The method according to embodiment 20 wherein the machine learning algorithm is configured such that the output comprises a determination of a number of emboli within a spectral Doppler waveform segment.

[0137] Embodiment 22. The method according to embodiment 21 wherein the intraprocedural feedback comprises a total number of detected emboli.

[0138] Embodiment 23. The method according to embodiment 20 wherein the machine learning algorithm is configured such that the output comprises a determination of a duty cycle of emboli within a spectral Doppler waveform segment.

[0139] Embodiment 24. The method according to embodiment 20 wherein the machine learning algorithm is configured such that the output comprises a differentiation between true emboli and artifacts.

[0140] Embodiment 25. The method according to embodiment 20 wherein the machine learning algorithm is configured such that the output comprises a classification of emboli according to emboli size.

[0141] Embodiment 26. The method according to any one of embodiments 17 to 25 wherein the one or more Doppler indices are selected from the group consisting of an acceleration time and a resistive index.

[0142] Embodiment 27. The method according to any one of embodiments 17 to 25 wherein the one or more Doppler indices comprise an acceleration time and a resistive index.

[0143] Embodiment 28. The method according to any one of embodiments 17 to 25 wherein the peripheral vessel is a pedal vessel and the one or more Doppler indices comprise a pedal acceleration time.

[0144] Embodiment 29. The method according to embodiment 28 wherein the one or more Doppler indices further comprise a resistive index.

[0145] Embodiment 30. The method according to any one of embodiments 2 to 6 wherein the adverse vascular event is associated with detection of vasospasm.

[0146] Embodiment 31 . The method according to embodiment 30 wherein the vasospasm is detected based on identification of a time-dependent spectral Doppler waveform having a flow signature characterized by a presence of an increasing peak velocity in successive cardiac cycles.

[0147] Embodiment 32. The method according to embodiment 31 wherein the flow signature associated with vasospasm is identified using a machine learning algorithm.

[0148] Embodiment 33. The method according to any one of embodiments 30 to 32 wherein the peripheral vessel is a pedal vessel and the one or more Doppler indices comprise a pedal acceleration time.

[0149] Embodiment 34. The method according to any one of embodiments 30 to 32 wherein the one or more Doppler indices are selected from the group consisting of an acceleration time and a resistive index.

[0150] Embodiment 35. The method according to embodiment 34 wherein the severity of the vasospasm is determined based on an increase in the acceleration time and an increase in the resistive index.

[0151] Embodiment 36. The method according to any one of embodiments 2 to 6 wherein the adverse vascular event is associated with detection of vessel recoil.

[0152] Embodiment 37. The method according to embodiment 36 wherein the recoil is detected based on identification of a time-dependent spectral Doppler waveform having a flow signature characterized by a decrease of peak velocity in successive cardiac cycles to a near-zero peak systolic velocity with an absence of a detectable diastolic portion of the waveform when compared a baseline pre-intervention spectral Doppler waveform.

[0153] Embodiment 38. The method according to embodiment 37 wherein the flow signature associated with vessel recoil is identified using a machine learning algorithm.

[0154] Embodiment 39. The method according to any one of embodiments 36 to 38 wherein the peripheral vessel is a pedal vessel and the one or more Doppler indices comprise a pedal acceleration time.

[0155] Embodiment 40. The method according to any one of embodiments 36 to 38 wherein the one or more Doppler indices are selected from the group consisting of an acceleration time and a resistive index.

[0156] Embodiment 41 . The method according to embodiment 40 wherein the severity of the vessel recoil is determined based on changes in the acceleration time and the resistive index.

[0157] Embodiment 42. The method according to any one of embodiments 1 to 41 wherein the intraprocedural feedback comprises an alert.

[0158] Embodiment 43. The method according to any one of embodiments 1 to 42 wherein a vascular intervention device is employed to perform the vascular intervention, the method further comprising sending control signals to the vascular intervention device to interrupt or modify the vascular intervention to mitigate the adverse vascular event.

[0159] Embodiment 44. The method according to any one of embodiments 1 to 42 wherein a vascular intervention device is employed to perform the vascular intervention, the method further comprising sending control signals to an additionaltherapeutic device to provide therapy suitable for mitigating the adverse vascular event.

[0160] Embodiment 45. The method according to any one of embodiments 1 to 44 wherein the Doppler ultrasound data is acquired while performing the vascular intervention.

[0161] Embodiment 46. The method according to any one of embodiments 1 to 45 wherein the Doppler ultrasound data is acquired after having performed the vascular intervention.

[0162] Embodiment 47. An ultrasound system comprising: a Doppler ultrasound detection subsystem configured to detect Doppler ultrasound data from a peripheral vessel during a medical procedure involving a vascular intervention; and control and processing circuitry operatively coupled to said Doppler ultrasound subsystem, said control and processing circuitry comprising a processor and associated memory, said memory comprising instructions executable by said processor for performing operations comprising: acquiring Doppler ultrasound data from the peripheral vessel; autonomously processing the Doppler ultrasound data to identify a flow signature associated with a presence of an adverse vascular event; detecting an occurrence of the adverse vascular event; and generating feedback associated with the adverse vascular event.

[0163] Embodiment 48. The system according to embodiment 47 wherein said control and processing circuitry is further configured to perform operations comprising: after detecting the occurrence of the adverse vascular event, determining a severity of the adverse vascular event, wherein the severity of the adverse vascular event is determined, at least in part, by employing a change in one or more Doppler indices before and after the onset of the adverse vascular event; wherein the feedback is generated based on the severity of the adverse vascular event.

[0164] Embodiment 49. A method comprising: intraprocedurally acquiring Doppler ultrasound data from a peripheral vessel during a neuromodulation procedure, wherein blood flow within the peripheral vessel is modulated according to the neuromodulation procedure;intraprocedurally and autonomously processing the Doppler ultrasound data to detect a change in vascular flow characteristics in the peripheral vessel resulting from the neuromodulation procedure; and providing intraprocedural feedback indicating the occurrence of the change in vascular flow characteristics.

[0165] Embodiment 50. The method according to embodiment 49 wherein a neuromodulation device is employed to perform the neuromodulation procedure, the method further comprising sending control signals to the neuromodulation device to interrupt or modify neuromodulation therapy based on the change in vascular flow characteristics.

[0166] Embodiment 51 . The method according to embodiment 49 wherein the Doppler ultrasound data is processed to identify a flow signature associated with a presence of an adverse vascular event, the method further comprising detecting an occurrence of the adverse vascular event.

[0167] Embodiment 52. The method according to embodiment 51 wherein a neuromodulation device is employed to perform the neuromodulation procedure, the method further comprising sending control signals to neuromodulation device to interrupt or modify the neuromodulation procedure to mitigate the adverse vascular event.

[0168] Embodiment 53. A neuromodulation system comprising: a neuromodulation device; a Doppler ultrasound detection subsystem configured to detect Doppler ultrasound data from a peripheral vessel during a neuromodulation procedure performed using the neuromodulation device; and control and processing circuitry operatively coupled to said Doppler ultrasound subsystem, said control and processing circuitry comprising a processor and associated memory, said memory comprising instructions executable by said processor for performing operations comprising: intraprocedurally and autonomously processing the Doppler ultrasound data to detect a change in vascular flow characteristics in the peripheral vessel resulting from the neuromodulation procedure; and providing intraprocedural feedback indicating the occurrence of the change in vascular flow characteristics.

[0169] Embodiment 54. The neuromodulation system according to embodiment 53 wherein the neuromodulation device is a neurostimulation device.

[0170] Embodiment 55. A method comprising: acquiring Doppler ultrasound data from a peripheral vessel during a medical procedure involving a vascular intervention, wherein blood flow within the peripheral vessel is downstream from a location of associated with the vascular intervention and is thus impacted by the vascular intervention; intraprocedurally and autonomously processing the Doppler ultrasound data to detect a change in vascular flow characteristics in the peripheral vessel; and providing intraprocedural feedback indicating the occurrence of the change in vascular flow characteristics.

[0171] The specific embodiments described above have been shown by way of example, and it should be understood that these embodiments may be susceptible to various modifications and alternative forms. It should be further understood that the claims are not intended to be limited to the particular forms disclosed, but rather to cover all modifications, equivalents, and alternatives falling within the spirit and scope of this disclosure.

Claims

CLAIMS1 . A method of monitoring and detecting an adverse vascular event via Doppler ultrasound interrogation of peripheral vasculature, the method comprising: intraprocedurally acquiring Doppler ultrasound data from a peripheral vessel during a medical procedure involving a vascular intervention, wherein blood flow within the peripheral vessel is downstream from a location of associated with the vascular intervention; intraprocedurally and autonomously processing the Doppler ultrasound data to identify a flow signature associated with a presence of an adverse vascular event; detecting an occurrence of the adverse vascular event; and generating intraprocedural feedback associated with the adverse vascular event.

2. The method according to claim 1 further comprising, after detecting the occurrence of the adverse vascular event, determining a severity of the adverse vascular event, wherein the severity of the adverse vascular event is determined, at least in part, by employing a change in one or more Doppler indices before and after the onset of the adverse vascular event; wherein the intraprocedural feedback is generated based on the severity of the adverse vascular event.

3. The method according to claim 2 wherein one or more Doppler indices are generated based on processing spectral Doppler data.

4. The method according to claim 2 wherein the flow signature is detected based on processing spectral Doppler data.

5. The method according to claim 2 wherein the change in at least one Doppler index is evaluated based on a first value of the Doppler index obtained before the adverse vascular event and a second value of the Doppler index obtained during the adverse vascular event.

6. The method according to claim 2 wherein the change in at least one Doppler index is evaluated based on a first value of the Doppler index obtained before theadverse vascular event and a second value of the Doppler index obtained after the completion of the adverse vascular event.

7. The method according to any one of claims 2 to 6 further comprising processing a spectral Doppler waveform associated with the adverse vascular event to determine a quantitative metric characterizing the adverse vascular event, and further employing the quantitative metric to infer the severity of the adverse vascular event.

8. The method according to any one of claims 1 to 7 wherein the adverse vascular event is detected via a machine learning algorithm configured to receive, as an input, a spectral Doppler waveform segment, and to generate output comprising a classification of the type of adverse vascular event among a plurality of types of adverse vascular events, the machine learning algorithm having been trained according to training spectral Doppler waveform data associating waveform signatures with labeled adverse vascular events of different types.

9. The method according to any one of claims 2 to 7 wherein the intraprocedural feedback is generated by processing, with a machine learning algorithm, at least one metric associated with the adverse vascular event, and at least one additional metric associated with a change in at least one Doppler index before and after an onset of the adverse vascular event.

10. The method according to any one of claims 2 to 9 wherein the one or more Doppler indices comprise one or more of a peak systolic velocity and an end diastolic velocity.11 . The method according to any one of claims 2 to 9 wherein the one or more Doppler indices are selected from the group consisting of an acceleration time and a resistive index.

12. The method according to any one of claims 2 to 9 wherein the one or more Doppler indices comprise an acceleration time and a resistive index.

13. The method according to any one of claims 2 to 9 wherein the peripheral vessel is a pedal vessel and the one or more Doppler indices comprise a pedal acceleration time.

14. The method according to claim 13 wherein the one or more Doppler indices further comprise a resistive index.

15. The method according to any one of claims 1 to 14 wherein the flow signature is indicative of an onset of the adverse vascular event.

16. The method according to any one of claims 1 to 9 further comprising: detecting a completion of the adverse vascular event; and providing additional feedback indicating the completion of the adverse vascular event.

17. The method according to any one of claims 2 to 6 wherein the adverse vascular event is associated with detection of one or more emboli.

18. The method according to claim 17 wherein an embolic event is identified via detection, within a given time window of a spectral Doppler waveform, of a bright echogenic vertical line satisfying emboli detection criteria.

19. The method according to claim 17 wherein an embolic event is identified via processing a spectral Doppler waveform to generate an energy waveform and comparing the energy waveform with a previously measured pre-intervention energy waveform known to be absent of emboli.

20. The method according to claim 17 wherein emboli detection is performed via a machine learning algorithm configured to receive, as an input, a spectral Doppler waveform segment, and to generate output comprising at least one scalar measure characterizing detected emboli.21 . The method according to claim 20 wherein the machine learning algorithm is configured such that the output comprises a determination of a number of emboli within a spectral Doppler waveform segment.

22. The method according to claim 21 wherein the intraprocedural feedback comprises a total number of detected emboli.

23. The method according to claim 20 wherein the machine learning algorithm is configured such that the output comprises a determination of a duty cycle of emboli within a spectral Doppler waveform segment.

24. The method according to claim 20 wherein the machine learning algorithm is configured such that the output comprises a differentiation between true emboli and artifacts.

25. The method according to claim 20 wherein the machine learning algorithm is configured such that the output comprises a classification of emboli according to emboli size.

26. The method according to any one of claims 17 to 25 wherein the one or more Doppler indices are selected from the group consisting of an acceleration time and a resistive index.

27. The method according to any one of claims 17 to 25 wherein the one or more Doppler indices comprise an acceleration time and a resistive index.

28. The method according to any one of claims 17 to 25 wherein the peripheral vessel is a pedal vessel and the one or more Doppler indices comprise a pedal acceleration time.

29. The method according to claim 28 wherein the one or more Doppler indices further comprise a resistive index.

30. The method according to any one of claims 2 to 6 wherein the adverse vascular event is associated with detection of vasospasm.31 . The method according to claim 30 wherein the vasospasm is detected based on identification of a time-dependent spectral Doppler waveform having a flow signature characterized by a presence of an increasing peak velocity in successive cardiac cycles.

32. The method according to claim 31 wherein the flow signature associated with vasospasm is identified using a machine learning algorithm.

33. The method according to any one of claims 30 to 32 wherein the peripheral vessel is a pedal vessel and the one or more Doppler indices comprise a pedal acceleration time.

34. The method according to any one of claims 30 to 32 wherein the one or more Doppler indices are selected from the group consisting of an acceleration time and a resistive index.

35. The method according to claim 34 wherein the severity of the vasospasm is determined based on an increase in the acceleration time and an increase in the resistive index.

36. The method according to any one of claims 2 to 6 wherein the adverse vascular event is associated with detection of vessel recoil.

37. The method according to claim 36 wherein the recoil is detected based on identification of a time-dependent spectral Doppler waveform having a flow signature characterized by a decrease of peak velocity in successive cardiac cycles to a nearzero peak systolic velocity with an absence of a detectable diastolic portion of the waveform when compared a baseline pre-intervention spectral Doppler waveform.

38. The method according to claim 37 wherein the flow signature associated with vessel recoil is identified using a machine learning algorithm.

39. The method according to any one of claims 36 to 38 wherein the peripheral vessel is a pedal vessel and the one or more Doppler indices comprise a pedal acceleration time.

40. The method according to any one of claims 36 to 38 wherein the one or more Doppler indices are selected from the group consisting of an acceleration time and a resistive index.41 . The method according to claim 40 wherein the severity of the vessel recoil is determined based on changes in the acceleration time and the resistive index.

42. The method according to any one of claims 1 to 41 wherein the intraprocedural feedback comprises an alert.

43. The method according to any one of claims 1 to 42 wherein a vascular intervention device is employed to perform the vascular intervention, the method further comprising sending control signals to the vascular intervention device to interrupt or modify the vascular intervention to mitigate the adverse vascular event.

44. The method according to any one of claims 1 to 42 wherein a vascular intervention device is employed to perform the vascular intervention, the method further comprising sending control signals to an additional therapeutic device to provide therapy suitable for mitigating the adverse vascular event.

45. The method according to any one of claims 1 to 44 wherein the Doppler ultrasound data is acquired while performing the vascular intervention.

46. The method according to any one of claims 1 to 45 wherein the Doppler ultrasound data is acquired after having performed the vascular intervention.

47. An ultrasound system comprising:a Doppler ultrasound detection subsystem configured to detect Doppler ultrasound data from a peripheral vessel during a medical procedure involving a vascular intervention; and control and processing circuitry operatively coupled to said Doppler ultrasound subsystem, said control and processing circuitry comprising a processor and associated memory, said memory comprising instructions executable by said processor for performing operations comprising: acquiring Doppler ultrasound data from the peripheral vessel; autonomously processing the Doppler ultrasound data to identify a flow signature associated with a presence of an adverse vascular event; detecting an occurrence of the adverse vascular event; and generating feedback associated with the adverse vascular event.

48. The system according to claim 47 wherein said control and processing circuitry is further configured to perform operations comprising: after detecting the occurrence of the adverse vascular event, determining a severity of the adverse vascular event, wherein the severity of the adverse vascular event is determined, at least in part, by employing a change in one or more Doppler indices before and after the onset of the adverse vascular event; wherein the feedback is generated based on the severity of the adverse vascular event.

49. A method comprising: intraprocedurally acquiring Doppler ultrasound data from a peripheral vessel during a neuromodulation procedure, wherein blood flow within the peripheral vessel is modulated according to the neuromodulation procedure; intraprocedurally and autonomously processing the Doppler ultrasound data to detect a change in vascular flow characteristics in the peripheral vessel resulting from the neuromodulation procedure; and providing intraprocedural feedback indicating the occurrence of the change in vascular flow characteristics.

50. The method according to claim 49 wherein a neuromodulation device is employed to perform the neuromodulation procedure, the method further comprisingsending control signals to the neuromodulation device to interrupt or modify neuromodulation therapy based on the change in vascular flow characteristics.51 . The method according to claim 49 wherein the Doppler ultrasound data is processed to identify a flow signature associated with a presence of an adverse vascular event, the method further comprising detecting an occurrence of the adverse vascular event.

52. The method according to claim 51 wherein a neuromodulation device is employed to perform the neuromodulation procedure, the method further comprising sending control signals to neuromodulation device to interrupt or modify the neuromodulation procedure to mitigate the adverse vascular event.

53. A neuromodulation system comprising: a neuromodulation device; a Doppler ultrasound detection subsystem configured to detect Doppler ultrasound data from a peripheral vessel during a neuromodulation procedure performed using the neuromodulation device; and control and processing circuitry operatively coupled to said Doppler ultrasound subsystem, said control and processing circuitry comprising a processor and associated memory, said memory comprising instructions executable by said processor for performing operations comprising: intraprocedurally and autonomously processing the Doppler ultrasound data to detect a change in vascular flow characteristics in the peripheral vessel resulting from the neuromodulation procedure; and providing intraprocedural feedback indicating the occurrence of the change in vascular flow characteristics.

54. The neuromodulation system according to claim 53 wherein the neuromodulation device is a neurostimulation device.

55. A method comprising: acquiring Doppler ultrasound data from a peripheral vessel during a medical procedure involving a vascular intervention, wherein blood flow within the peripheralvessel is downstream from a location of associated with the vascular intervention and is thus impacted by the vascular intervention; intraprocedurally and autonomously processing the Doppler ultrasound data to detect a change in vascular flow characteristics in the peripheral vessel; and providing intraprocedural feedback indicating the occurrence of the change in vascular flow characteristics.

Images