Measurement and monitoring modules for a vascular access system

Abstract

Measurement and monitoring modules for systems and methods for delivering needles or catheters within target tissue including an apparatus that includes or a method involving, structure configured for placing a needle or cannula to provide vascular access. In one approach, a vascular access device or system cooperates with one or more of a measurement module and a monitoring module.

Description

PCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCTMEASUREMENT AND MONITORING MODULESFOR A VASCULAR ACCESS SYSTEMFIELD OF THE DISCLOSURE

[0001] The present disclosure generally relates to measurement and monitoring modules for systems and methods to deliver needles or catheters within target tissue, vasculature or grafts.BACKGROUND OF THE DISCLOSURE

[0002] There is a continuing need for an effective alternative approach for needle or catheter delivery and in particular for hemodialysis. Moreover, there is a need for an access system for clinic or home use which is easy and effective to use.

[0003] Hemodialysis has become a conventional approach to treat individuals with kidney disease. A hemodialysis procedure, commonly referred to as dialysis, involves filtering wastes from a person’s blood and thus supports or replaces biological functions provided by a healthy kidney. Effective dialysis results in facilitating balancing substances residing in blood including calcium, potassium and sodium, and also can aid in controlling blood pressure.

[0004] Typically, a hemodialysis treatment lasts about four hours and is conducted three times per week, but more frequent and longer treatments can be necessary for some patients. Adding to this, the time to travel to a treatment center and waiting to be hooked up and disconnected from a dialyzer of course adds up to a very significant commitment from the patient both in time and effort.

[0005] Hemodialysis treatment is a life sustaining treatment that is performed three times per week until a patient can receive a kidney transplant. With the severe shortage of kidneys available relative to the number of end stage renal disease patients, most patients will be on three day per week hemodialysis for the rest of their life. Therefore, there is a need for a system that assesses the health of their arteriovenous fistula or graft, and assists with determining where needle insertion should be each treatment to help provide longevity of the usability of that fistula or graft. Also, there is a need to detect earlier than current technologies allow when a problem such as stenosis or thrombosis is developing.

[0006] It is due to these limitations associated with dialysis performed at treatmentPCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT centers that more efficient hemodialysis including cannulation has become a desirable alternative either in the treatment center or at home. It is thus desirable to make dialysis as well as assessment of desired locations of needle insertion sites and cannulation itself more efficient and / or require less skill. Studies have shown that home dialysis five to seven times a week has dramatically better outcomes in many ways including a longer life and better survival. When dialysis is performed at home, there is no need to travel to a dialysis center. There is also more flexibility in home dialysis as the patient is able to choose a convenient time for dialysis and has a greater sense of control from being independent and doing treatments themselves.

[0007] However, unless in-home care is engaged, there is most often no medical professional in the home setting to monitor treatments or answer immediate questions. Also, significantly, in the treatment center health care professionals are less available to assist with or perform the cannulation techniques required to insert needles within patient vessels so that the patient can be hooked up to a dialyzer. Cannulation is a skilled task. Also, self-cannulation can be a daunting task for many dialysis technicians and patients and many dialysis technicians and patients can lack the necessary dexterity or skill to repeatably, effectively and efficiently chose insertion locations and insert needles to gain access to vasculature through insertion sites. Since proper repeatable cannulation is so critical to successful hemodialysis and the avoidance of infection and other complications, unassisted home dialysis is not currently a practical alternative to large populations of patients.

[0008] Hemodialysis is a procedure done tens of thousands of times a day to help people suffering from kidney dysfunction. As a part of the system to replicate the function of (non or low performing) kidney, blood must be removed from the patient, passed through a dialysis machine to clean it of wastes, salts, and fluids, and then have the blood returned to the patient. To facilitate the path of blood through a dialysis machine, two connections or cannulations must be made to the patient. These connections are large bore needles delivered by dialysis technicians who have a wide range of skill and experience. Preferably, there would be a system or machine that would enable all technicians to safely, reliably, deliver these needles and make the connections without causing any issues or damage to the target vessels. To achieve this goal, an ultrasound-assisted vascular access system is desired.

[0009] Prior systems have focused on vascular access in emergency situations that justPCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT require identifying a needle insertion location one time and guiding insertion of the needle to that one location, usually a femoral artery or vein, or jugular artery or vein.

[0010] Arteriovenous fistulas (AVFs) are essential for hemodialysis but are prone to complications like stenosis and thrombosis, leading to dysfunction and potential failure. Early identification and intervention are crucial for maintaining AVF patency and improving patient outcomes. Current monitoring methods have limitations. Physical examination is subjective and relies heavily on operator skill. Doppler ultrasound while accurate, is time-consuming, requires trained personnel, and is not always readily available. Also, angiography is invasive and carries risks associated with contrast agents and radiation exposure.

[0011] Accordingly, there is a need for effective and efficient devices and approaches to delivering needles or catheters and for minimizing or reducing the time involved in receiving hemodialysis. There is also a need for a noninvasive approach to proactively monitor and predict the health of insertion sites and / or detecting blockages or stenoses within insertion anatomy.

[0012] The present disclosure addresses these and other needs.PCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCTSUMMARY OF THE DISCLOSURE

[0013] Briefly and in general terms, the present disclosure is directed towards methods, a device or system that eases and standardizes the safe placement of needles or catheters for dialysis. Additionally, there are many other opportunities for automatic, guided, needle or catheter delivery so that the present disclosure also provides a system that would have many other applications beyond dialysis for vascular access.

[0014] In one aspect, there is provided an integrated system for proactively monitoring and predicting the health of insertion sites and surrounding anatomy that includes one or more of a data collection module, a processing module, a machine learning and prediction module, and a user-interface module. In a particular aspect, the data collection module includes one or more of data from a dialysis machine, an ultrasound imager and a digital stethoscope auscultation device. Integrating data from a combination of an ultrasound, a digital stethoscope and the dialysis machine, combined with machine learning offers a comprehensive, objective and early predictive tool for AVF health assessment.

[0015] In another aspect, there is additionally or alternatively provided an integrated system for monitoring and detecting narrowing or stenosis in target anatomy or in AVF used for hemodialysis. In a particular aspect, the system can include one or more of an ultrasound device, a digital stethoscope, and advanced software algorithms providing comprehensive and user-friendly solutions for assessing target anatomy. Such a system provides an integrated approach that combines the benefits of multiple modalities, enabling comprehensive and objective assessments.

[0016] In one embodiment, there is provided a portable, handheld electromechanical device which assists with or enables needle or catheter insertion for vascular access in hemodialysis patients. The device may include an imaging modality in the form of an ultrasound array. In some embodiments, a display for the user presenting them with the information on location of the identified fistula vessel or graft. Furthermore, an embodiment of the device may include a bi-plane transducer arrangement in which the fistula vessel or graft can be seen in two different locations offset from each other, and as such providing information on the alignment or directionality of the vessel. Moreover, the bi-plane approachPCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT also allows for the target to be viewed at the location where the needle enters the vessel and the location where it is expected to stop, thus providing a clear view of the anatomy from a static position. In another embodiment, the device may include a multi-plane (3 or more) transducer arrangement which provides greater accuracy or fidelity regarding the alignment or directionality of the vessel.

[0017] In one embodiment, the system includes an electromechanical device including motors, sensors (a combination of position, current, force or torque, flashback, acoustic sensing with optionality for further expansion of the sensor suite) and a computer or wired or wireless connectivity to a computer. The system also can have an inertial measurement unit that allows for the system to measure velocity and acceleration in all directions during use. The computer can be responsible for acquisition of the ultrasound image stream, executing image processing algorithms and finally calculating the inverse kinematics and path planning with all inputs to generate reference trajectories for the cannulation module.

[0018] Accurate and reliable cannulation of the needle or catheter in the bloodstream is of paramount importance to avoid complications and prolong the life of the surgically created arteriovenous (AV) fistulas (AVF) and grafts (AVG) in patients for hemodialysis access. Ultrasound guided cannulation can be particularly beneficial in cases where a fistula vessel is small or deep, and real-time feedback from ultrasound can reduce needle stick errors, improve accuracy and success rates for hemodialysis access. Accordingly, ultrasound can be utilized in place of or in addition to user observed sound and feel to facilitate building a detailed image for a user, such that subjective user feedback can be proceduralized along with ultrasound imaging to accomplish desired cannulation.

[0019] To facilitate this needle or catheter insertion, ultrasound provides inputs to a control model which understands the geometry and capabilities of the delivery device and the intended target. With these two pieces of information a needle path can be determined and applied when the user activates the device for needle insertion. Additionally, the disclosed system or device can continuously monitor the anatomy to ensure that the target is correctly positioned and thus provides a platform that can be used in a dynamic and handheld embodiment.

[0020] In one particular aspect, there is provided a combination of detailed geometry and capability as well as mechanical properties such as stiffness of components of the deliveryPCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT device that are combined with the kinematic analysis to allow a needle insertion system to determine a needle path from a start position to a needle successfully placed in a target vessel.

[0021] In another aspect, the vascular access system eases and standardizes the safe placement of needles for dialysis. There is a broad range of user variability, anatomy variation, and daily repetition. All of which points to a need to simplify and automate the workflow both for the technician doing the placement and the patient who is undergoing dialysis. In addition, the use of ultrasound can lead to a more standardized assessment of the patient’s anatomy to ensure the vessel / graft health is maintained over time, as well as facilitate observing even small changes in the health of a fistula such as seeing a stenosis develop early or detecting the presence of recirculation or aid in identifying signs of a hematoma or clotting.

[0022] These and other features of the disclosure will become apparent to those persons skilled in the art upon reading the details of the systems and methods as more fully described below.PCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCTBRIEF DESCRIPTION OF THE DRAWINGS

[0023] Fig. 1 is a schematic representation, depicting a hemodialysis procedure and equipment.

[0024] Fig. 2 is a top view, depicting a graft in a first vascular access approach.

[0025] Fig. 3 is a top view, depicting an AV fistula in a second vascular access approach.

[0026] Fig. 4 is a top view, depicting cannulation of the second vascular access approach.

[0027] Fig. 5 is a top view, depicting the flow of blood resulting from cannulation.

[0028] Fig. 6A is a flow chart, depicting a cooperating relationship between software and hardware components of the vascular access device.

[0029] Fig. 6B is a partial cross-sectional view, depicting one embodiment of a vascular access device.

[0030] Fig. 6C is a flow chart, depicting a vessel detection and alignment flow chart.

[0031] Fig. 7 is a flow chart, depicting a workflow for use of the vascular access device.

[0032] Fig. 8 is a schematic, depicting computer communication of the vascular access device.

[0033] Figs. 9A-M are various views, depicting another approach to a vascular access device.

[0034] Figs. 10A-E are schematic and perspective views, depicting approaches to ultrasound transducer or array or probe assemblies.

[0035] Fig. 11 is a flow chart, depicting an integrated health monitoring system.

[0036] Fig. 12 is a flow chart, depicting a stenosis detection system.PCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCTDETAILED DESCRIPTION OF THE DISCLOSURE

[0037] Before the present systems and methods are described, it is to be understood that this disclosure is not limited to particular embodiments described, as such may, of course, vary. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to be limiting, since the scope of the present disclosure will be limited only by the appended claims.

[0038] Where a range of values is provided, it is understood that each intervening value, to the tenth of the unit of the lower limit unless the context clearly dictates otherwise, between the upper and lower limits of that range is also specifically disclosed. Each smaller range between any stated value or intervening value in a stated range and any other stated or intervening value in that stated range is encompassed within the disclosure. The upper and lower limits of these smaller ranges may independently be included or excluded in the range, and each range where either, neither or both limits are included in the smaller ranges is also encompassed within the disclosure, subject to any specifically excluded limit in the stated range. Where the stated range includes one or both of the limits, ranges excluding either or both of those included limits are also included in the disclosure.

[0039] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs. Although any methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present disclosure, the preferred methods and materials are now described.

[0040] It must be noted that as used herein and in the appended claims, the singular forms "a", "an", and "the" include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to "the system" includes reference to one or more systems and equivalents thereof known to those skilled in the art, and so forth.

[0041] In one embodiment, there is provided an integrated monitoring system configured to proactively monitor and predict the health of target anatomy or arteriovenous fistulas in hemodialysis patients. The system includes one or more of a data collection module, a processing module, a machine learning and prediction module, and a user-interface module. In a particular aspect, the data collection module includes one or more of: dataPCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT received from a dialysis machine, an ultrasound imager and data received from a digital stethoscope auscultation device. Integrating data from ultrasound, digital stethoscopes and the dialysis machine, combined with machine learning offers a comprehensive, objective and early predictive tool for AVF health assessment.

[0042] In an additional or alternative embodiment, there is provided an integrated system for monitoring and detecting narrowing or stenosis in target anatomy or in arteriovenous fistulae (AVF) used for hemodialysis earlier or with greater detail than is detectable using the conventional look, listen and feel technique. In a particular aspect, the system can include one or more of: an ultrasound device, data received from a digital stethoscope, and advanced software algorithms providing comprehensive and user-friendly solutions for assessing target anatomy. Such a system provides an integrated approach that combines the benefits of multiple modalities, enabling comprehensive and objective assessments.

[0043] It is to be noted that various different procedures can benefit from guided cannulation. In certain procedures, the depth of a target, size of the target and surrounding anatomy may differ. Additionally, the type of device being inserted may change as well and may not be a needle and instead can be another type of device (such as biopsy, targeted energy device etc.). Accordingly, the system can be modular or a two-part system. The first part can be the ultrasound assembly or module, and the second part can be the cannulation or insertion module. In one aspect, the main assembly can contain the ultrasound functionality and a connector along with electronics that can communicate with the second part. The second part can be the insertion or cannulation module that contains the degrees of freedom required for needle or tool insertion along with any additional electronics required for actuation and sensing. The system would thus have a universal interface and the common part to any procedure would be the ultrasound module, thus resulting in an approach that is more universally applicable to alternative applications by allowing for modifications to the kinematics of the device insertion mechanism.

[0044] With reference to Fig. 1, a hemodialysis system 50 and associated equipment is shown. During hemodialysis, the patient’s blood is routed through a dialyzer 52 which filters the blood. The patient is prepared by cleaning the skin sites from which blood flow out and then back into the patient’s blood vessels. As described more below, steps are taken to preparePCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT the patient for dialysis. Once the patient is so prepared, at the start of hemodialysis, a pair of needles 54, 56 are inserted into the patient’s arm one at a time. A numbing compound can be used to facilitate needle insertion and to minimize pain. Each needle is attached to a soft tube or cannula 58 that connects to the dialysis system 50. The dialysis system 50 pumps blood through a dialyzer 52 and returns the blood to the patient’s body. During this process, blood pressure and flow rates are monitored and controlled by the dialysis machine thereby controlling the flow of blood through the filter and speed of which blood flows from and to the patient.

[0045] As blood enters the filter, it is forced through a large number of thin, hollow fibers. At the same time, a dialysis solution passes in the opposite direction about the fibers. Waste products are thereby removed from blood and carried by the dialysis solution. Filtered blood is then returned to the patient’s vasculature. In this way, extra salt, potassium, calcium and fluid is removed from blood.

[0046] An important step before starting hemodialysis treatment is having surgery to create a vascular access site. Vascular access is a phrase used to describe the place on a patient’s body where blood flows from and returns to the patient’s vasculature such as during hemodialysis. A hemodialysis vascular access site may be a catheter, an arteriovenous (AV) graft 60 (Fig. 2) or an arteriovenous (AV) fistula 70 (Fig. 3). Notably, the catheter approach is generally employed for temporary access and not suited as a permanent solution for home or treatment center dialysis because of risk of serious or deadly infection.

[0047] To create an AV graft 60 (Fig. 2), during an outpatient procedure, a surgeon cuts the skin to gain access to target vessels. The surgeon then uses a synthetic tube graft 60 to connect an artery which carries blood away from the heart to a vein which carries blood to the heart. The surgical site is then closed, leaving the graft available for dialysis. Dialysis needles are then repeatedly used to access the tube during a hemodialysis procedure as described above. Generally, the AV graft approach is employed for patients that have veins that prevent them from having an AV fistula since the AV graft approach is more often associated with infection and repeated blood clots that can block the flow of blood and make it hard or impossible to have dialysis.

[0048] The generally accepted best type of long-term vascular access is an AV fistula 70 (Fig. 3). This approach is characterized as providing the highest blood flow forPCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT dialysis, is less likely to become infected or clot, and as lasting longer than other approaches to vascular access. Here, the surgeon connects an artery directly to a vein, usually within the patient’s arm, to create the AV fistula. When the vein is so connected to the artery, it grows or remodels over weeks so that its walls are wider and thicker, making it easier to repeatedly place the needles for dialysis. The AV fistula itself also has a large diameter that allows blood to quickly flow out and back into the patient’s body, the goal being creating a system where there is high blood flow so that the largest amount of blood can pass through the dialyzer. Fig. 4 shows the cannulation of two needles 54, 56 at an AV fistula site 70, and Fig. 5 depicts the flow of blood to and from the AV fistula 70 during dialysis. It is to be noted that in using the presently disclosed vascular access systems inventions, any number of treatment protocols can be specified by the treating healthcare professional and embedded into software such that needle cannulation sites may be alternated or rotated in a rope ladder technique or repeatedly used as in the buttonhole technique.

[0049] Once the patient is provided with a vascular access site, the challenge becomes placing the needles within the site. In one aspect, the system can manage, but is not limited to, rope ladder or buttonhole site strategies for cannulation as necessary or dictated by the system or health care professional. Accordingly, various approaches to needle delivery methods and apparatus are presented. The disclosed approaches are configured to provide a repeatable, effective and accurate approach to cannulation. The disclosed approaches are intended for use in an in-center setting or a home setting for hemodialysis.

[0050] In one or more embodiments, the system includes an electromechanical device comprising of motors, sensors (a combination of position, current, force, flashback, acoustic sensing with optionality for further expansion of the sensor suite) and a computer or wired or wireless connectivity to a computer. In one or each of the disclosed embodiments, the computer can be in the device or can be separate from the device such as on a cart or the like. In one particular aspect, at least a part of host computing occurs is a cloud-based server. An embodiment of the device can include a mode where the underlying algorithms are continually improving via a feedback loop created from the collected data being used as training input. The computer can be responsible for acquisition of the ultrasound image stream, executing image processing algorithms and finally calculating the inverse kinematics and path planning with all inputs to generate reference trajectories for the cannulation module. Further, thePCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT computer may comprise a non-transitory computer readable medium comprising instructions to collect and process signals received from data collection devices or functionality or include or communicate with a non-transitory computer readable medium having stored thereon instructions executable by a computing device of the system or external to the system to cause the computing devices to perform functions associated with and directed by the firmware or software. The computer may further comprise instructions to transmit collected and / or processed signals to a memory for storage, or to a module for transmission to another computing device. While the insertion is occurring, the system can also continue to monitor the device relative to the vessel position, shape, and location and ensure the anatomy and device stay positioned relative to each other for delivery of the tip of the needle into the center of the vessel so as to not scrape along the internal surface of the vessel or pass all the way through the vessel with the tip of the needle being positioned outside the vessel.. If it does change the kinematic path accommodates such changes as necessary or stops delivery of the needle. Thus, real-time control can be provided in that an insertion path is initially calculated and then a safety loop that runs in the background will continuously monitor the ultrasound stream, to check for any movement leading up to a change in the initial target position. The real-time control loop will be looking at the position of the target anatomy and calculating any relative motion compared to the already calculated kinematic path and target. If the system detects that there is any change in location of the vessel and device relative to each other that would impact the pre-calculated trajectory it can then intervene by either stopping the system, even retracting the needle if insertion has begun, or updating and resending the trajectory to the system. In one particular aspect, there is provided a combination of detailed geometry and capability of the current device combined with the kinematic analysis to allow a needle insertions system to determine a needle path from a start position to a needle successfully placed in a target vessel.

[0051] The disclosed needle insertion systems are configured and function to position a needle or needles and, in some embodiments, advance the needle within the targeted AV graft, fistula, vessel or vessels. The system further includes functionality for positioning a needle into the correct position and trajectory and, in some embodiments, to advance the needle at pre-determined angle(s) and depth(s) within a patient’s body and within target vasculature. The system is effective for providing vascular access and assists with cannulationPCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT throughout the body including specifically for radio-cephalic, brachio-cephalic and brachio- basilic fistulas and use may range from the forearm to the upper arm or other locations on the body. The system can be easily operated in a home setting or in a treatment center and functions to cannulate the patient successfully while improving outcomes and reducing complications. Further, the system can provide access to various anatomical structures that are identifiable via ultrasound such as targeting a biopsy or to gain access to the lymphatic system or the like.

[0052] A standard of care is to initially insert needles for fistulas at a 25 degrees angle to the skin, and grafts at 45 degrees. In one or more embodiments, a final insertion angle can be 15 degrees. Therefore, a pivot location can be selected such that the portion of a target circle that is utilized to pitch from 45 to 15 degrees is centered at approximately 3mm deep into the patient because standard for mature fistula access is less than 6mm depth. This depth could also be adjusted to an amount other than the average of the possible vessel depths.

[0053] One piece of information employed in needle insertion is the identification of a pivot location that is the offset between the pitch axis and the needle axis. This is a detail that is a function of the mechanical design of the device arm and needle holder.

[0054] Also provided can be a laptop or other computer (not shown) for example a tablet. For this embodiment, the current programming and control is from a USB cable attached laptop. In addition to being able to work with the ultrasound images coming to the computer from the ultrasound array, the laptop allows for flexibility in programming and refining the movements of the system 100. In an alternative embodiment, the system 100 could still be attached to a laptop with a fully developed user-interface to allow the laptop / cart to accomplish computational or control work. In yet another embodiment, all the system work can be connected with the electronics within the handheld device requiring only a power connection (or even battery power) to operate. In one aspect, there can be a full integration of processing (computer) function into the device itself.

[0055] In terms of automated or robotic systems, the number of controllable motions or actions is a source of complexity and cost. In many cases the additional precision or control needs to be countered by cost in terms of physical parts, manufacturing complexity, or software development / processing burden. The disclosed embodiment distills the design toPCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT minimal degrees of freedom so that the system 100 can perform as desired in all conditions of use. Notably, by making the system user controlled or positioned and / or to be easily maneuvered, additional degrees of freedom of motion are inherently provided.

[0056] The degrees of freedom present in this embodiment are pitch and insertion.Pitch rotates the needle to a desired insertion angle. Additionally, as the needle advances pitch can be used to alter the needle path angle forward along the target vessel. The combination of these two degrees of freedom allows the system to advance the needle tip into the target fistula or graft and then adjust to allow the needle tip to pass down the fistula or graft, safely, without scraping along the interior of the vessel or passing into and then out of the wall of the vessel, for complete insertion.

[0057] In the disclosed embodiment, a range of motion is 2.5 degrees to 47.5 degrees with the lower angles for final insertion, such as approximately 15 degrees. As shown in Fig. 8, a brushed DC motor 128 is used to drive a spur or pitch gear 132 and control the angle of pitch. In alternative approaches, brushless DC motors or stepper motors or other types of actuators (piezo motors etc.) can be employed for drive functionality. The motors can further include drive bevel gear trains, belt, capstan plus cable structure, a worm gear or similar drive train to get motion from the motor to the pitch axis. Moreover, a linear actuator can be used to push or pull the pivoting arm 116 offset from a pivot axis.

[0058] The angle of the needle 130 for insertion needs to be sensed for the control circuit and such sensing can be provided by a motor encoder (not shown), with a homing step. This can include a drive to a current threshold, torque threshold, or up to a limit switch.Further, an absolute encoder could be used in the device body or arm to measure without a required homing step.

[0059] With reference to Figs. 6A-C, a needle insertion system 100 involves cooperation between software 170 and hardware 176 modules. On the software side, code is provided to one or more of accomplish vessel localization, vessel tracking and device-vessel alignment. Vessel localization in turn is based upon sub-programs handling ultrasound device drivers and Inertial Measurement Unit (IMU) driver information. Through communication with a graphical user interface, device-vessel alignment information is communicated to a main display on the hardware side. With reference to Fig. 6B and as explained in further detail below (See Figs. 9A-M), a needle insertion system can be embodied in a vascular accessPCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT system 400 that includes a body housing 410 containing the mechanical components to alter the pitch of a pivoting arm 416, an ultrasound transducer or probe or array 420, a needle or catheter holder assembly 460, and internal control electronics for both motor control and activation and the transducer 420.

[0060] Software also addresses and makes calculations or assessments concerning, among other things, device kinematics and trajectory planning leading to feedback position control and control of motor drivers. As stated, the computer is responsible for acquisition of ultrasound image stream, executing image processing algorithms and calculating inverse kinematics and path planning with all inputs to generate reference trajectories for the cannulation hardware modules 176. In one or more aspects, these reference trajectories are transmitted as inputs to the cannulation modules 176, which may consist of onboard processors, that implement feedback position control for the motors using a sensor suite for positioning along with measurements that inform the controller about system constraints.

[0061] As the vascular access system 100 is positioned on the patient’s upper extremity, and the user initiates imaging, in one approach the position of an identified vessel, and the lumen in particular, is computed by the computer application using a combination of computer vision techniques (See Fig. 6C). This position is used as a target input for the cannulation module 176. The cannulation module algorithms can include a kinematic model of the vascular access system 100. This kinematic model performs the mathematical transformations of representing the needle tip position in several different coordinate frames. In one form of the vascular access system 100, consisting of a pitch and a slide degree of freedom (2 DOF), these frames would consist of the imaging frame, the vascular access device frame and the intermediary slide and needle frames respectively (See Fig. 6B).

[0062] Transformation matrices which are constructed depending on the physical parameters and arrangement of the DOF’s, facilitate the algorithm in determining needle tip position from a vascular access system 100 perspective in an imaging frame through a forward kinematic model, thereby aiding in position feedback control. The position feedback control may consist of one of or a combination of linear position control methods such as a proportion integration derivative (PID) or a linear quadratic gaussian (LQG) regulator or a model predictive controller (MPC), In one or all formulations, the position controller can utilizePCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT position feedback from motor encoders, reference inputs received from the host computer and an embedded processor for implementation of motor drivers and control algorithm. The feedback position controller, in addition, may have motor output constraints imposed via measurements from a motor current sensor, or a strain-gauge based force sensor, to limit the force applied during execution of the calculated cannulation trajectory. In some instances, the position controller may have constraints, to limit motor output, be informed by an infrared or other sensor that detects the presence of blood flow in the attached cannula.

[0063] To generate the reference position trajectory, in one or more aspects, a path planning algorithm utilizes an inverse kinematic model of the vascular access system to translate a desired needle tip position that is expressed in the imaging frame into corresponding desired position configuration of the vascular access system component joints that achieve the desired needle tip position, so translation occurs from image to system. In various aspects, a path planning algorithm also takes into account the depth of the desired target position to determine intermediate points, and construct a trajectory that passes through these points. The algorithm is flexible as to a selection between piecewise linear trajectory or creating cubic splines for smoother transitions between calculated trajectory points. In one or more embodiments, other factors that influence the trajectory point calculations include the approach and the insertion angle that are determined based on desired insertion depth and the guideline put forth by the National Kidney Foundation (NKF).

[0064] On the hardware side, the main display communicates with a main processor which in turn receives information from an ultrasound array. The IMU drivers on the software side communicate with an IMU on the hardware side, and the IMU communicates with a supplemental microprocessor which receives information from force and flashback sensors as well as position sensors and motor driver information transmitted from the software side. The position sensors are additionally in communication with the system motors which are driven to effect positional changes in a needle hub. In another embodiment, so as to not confuse the user with ultrasound imagery or require advanced training regarding interpreting ultrasound imagery, the display shows minimal information to the user such as a red light for do not activate the needle insertion and a green light for activate, or even no information to the user but rather provides audible or haptic sensation to direct the user.PCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT

[0065] Employing the needle insertion system can involve a workflow including one or more of the following steps (Fig. 7). A patient is assessed to determine whether a needle insertion system can be used for cannulation (steps 1-4) by moving the system 100 along the length of, or substantial portion thereof, the fistula or graft to record ultrasound image data. One or more of AVF localization, ultrasound imaging and vessel segmentation and classification are conducted on the anatomy of the patient. Fistula health, such as changes in vessel wall thickness, lumen diameter, shifts in location of the vessel, flow rate, presence of thrombosis, fibrosis, stenosis, etc. is assessed and the system logs data for trending and prediction of patient access issues. The system can identify a patient by scanned fistula anatomy and pulls up the patient record. Alternatively or additionally, the user can enter a patient identifier into the computer. Records can be stored remotely or on the cloud, from which assessment of fistula and guidance for next treatments are stored. The record directs the user to a next location on the patient using reference from previous cannulation history. In addition, measurements of the fistula can be made with both ultrasonic imaging as well as flow assessment based on doppler scanning.

[0066] When it is determined that intervention using the needle insertion system can be employed, a device setup (step 5) is conducted. If so, the user cleans the area for cannulation. If not, the patient can be assessed again using the information learned during a previous assessment.

[0067] Once it is determined that an individual is ready for cannulation, the needle insertion device and cart, if any, associated with the device, which can be plugged into an outlet for charging overnight, is used. In one embodiment, a battery can be provided that is configured to last a working day. The user attaches the needle to the system with a needle guard in place, ensuring to keep area clean, and removes the needle guard provided with each needle (step 5). Just prior to use, an operator cleans the needle insertion module for infection control, (step 6).

[0068] Once the patient is registered, or alternatively without registering the patient, the user places the needle insertion device on a patient’s arm near where they want to cannulate (step 7). Using the ultrasound array, the user guides the needle insertion device to be centered on and aligned to the target vessel (fistula or graft; step 7b). Ultrasound interface material (gelPCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT or other) is employed for visualization. Once aligned to the user’s satisfaction and confirmed by the device, the user activates needle insertion by pressing the activation button (step 8). Notably, the system can use its sensors to choose not to insert the needle even if commanded by the user , if the system for example detects unacceptable movement of the device or the patient’s arm relative to one another. The system uses the ultrasound alignment information to make this decision, or uses the IMU position velocity data, or other sensors. The needle insertion system is nevertheless configured to insert and advance the needle tip through the skin and into the target vessel. During the motion, the system can adjust needle angle during advancement to ensure the needle tip is both in the vessel but not contacting vessel walls. Once complete and before moving the cannulation module, the user releases the needle (step 9) from the device. The needle will then be taped and secured to the patient. User repeats the steps for placement of a second needle.

[0069] Turning now to Fig. 8, there is provided a schematic depicting the relationships and connections between a computer (laptop, tablet, etc.) 200, main board (located in the device) 210, and motors and sensors. In alternative embodiments, the various connections and processing can be moved between the various platforms or isolated to a single platform. The main board 210 is in communication with both an arm board 230 that communicates with a force sensor 232 and a flashback sensor 234, and an ultrasound board 240 that communicates with a transducer 242. The force sensor 232 is associated with and provides information concerning needle penetration into tissue, and the flashback sensor 234 is associated with and provides information concerning blood flashback within the clear tubing attached to a needle. Two-way communication is further provided between the arm board 230 and an insertion actuator 252 and an insertion position sensor 254, and between the main board 200 and a pitch actuator 262 and a pitch position sensor 264. The insertion actuator 252 operates to actuate insertion or movement of a needle during a use procedure and the position sensor 254 is associated with and provides information concerning a position of a needle tip within space or relative to target tissue. Further, the pitch actuator 262 operates to actuate pitch control, and the position sensor 264 is associated with and provides information concerning catheter or needle tip positioning relative to target tissues or generally within space. The look / listen / feel sensor 265 can also be employed to provide further feedback to the user concerning thePCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT position of the needle or catheter of the system during use or the environment in which they are used or have been advanced.

[0070] With all the inputs from the physical system, a kinematic model is created to know the location of the needle tip with known inputs of pitch angle and linear travel. In order to track the location so the needle insertion system knows where the needle tip is, encoders (not shown) are used on the motors so that locations can be returned to the control system. In addition, the properties of the motors controlling the motion will also be used in the control system, though not specifically required for the kinematic model. There are many options for encoders and motors in a wide range of embodiments.

[0071] In one or more embodiments, the needle insertion device includes a drivetrain including pitch and insertion controls. To control pitch, an encoder is provided and has a resolution of .01 degree / tick and is embodied in an absolute encoder having a 12bit ADC (360 degree) on motor gearbox output shaft and an 8.75: 1 geartrain between a gearbox output shaft and a pivot shaft. The drivetrain associated with pitch has a rated torque / speed of 243 N-mm at a joint and a 40.5 degree / sec speed at rated torque values. A max torque / speed can be 810 N-mm stall torque at a joint and 46.6 degree / sec motor at a no load speed. To control insertion, an encoder is provided and has a resolution of .029 mm / tick and embody an incremental encoder having a ,2mm nut travel per motor gearbox revolution and 7 encoder ticks per motor revolution. The insertion drivetrain can further have a rated torque / speed of 4.43 N rated force at nut and 41.7 mm / sec at rated force. A max torque / speed can include a 19 N stall force at nut and a 53.3 mm / sec motor no load speed.

[0072] In the full control loop, with the understanding of the geometry and kinematics of the system, combined with the knowledge of the locations / positions of these degrees of freedom, a next step is the assessment of where the target is located. In the current embodiment, the location is assessed or sensed by way of an ultrasound image. From the ultrasound image, the vessel / fistula will be identified in terms of depth and location. Once the user has followed prompts to align and hold the vascular access device in place, the user then triggers the system. Consequently, the targeting control loop of the vascular access system creates the desired needle path to accomplish vessel placement.PCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT

[0073] With reference to Figs. 9A-M, another embodiment of a vascular access system 400 is presented. This system 400 can embody one or more or all of the functionality and features described above. The vascular access system includes a body housing 410 containing the mechanical components to alter the pitch of a pivoting arm 416, an ultrasound transducer or probe or array 420, a needle holder assembly 460 that retains a needle 430, and internal control electronics for both motor control and activation and the transducer 420. In addition, connections to both power and a computer run though the body housing 410. Figures 9A-F show various side and perspective views of the vascular access system with the needle holder assembly 416 spaced from the housing 410.

[0074] Figures 9G-I highlight certain of the internals of the body 410 and the ultrasound array or transducer 420, with shell components of the body housing 410 removed. It is to be noted that a custom integrated transducer 420 is incorporated into the design. By integrating an ultrasound transducer in the device, there is a distinct rigid body connection between the needle holder and the imaging modality, which allows for deterministic positioning of the needle in relation to the anatomy of interest in the scanned volume. Moreover, housed within the body 410 is a circuit board 422 associated with the transducer 420, and a controller board423 that controls the functioning of the system and moving components. An activation button424 (See Fig. 9F) is configured to be in electronic communication with the controller 423.

[0075] Mechanically and electronically connected to the body 410 is the pivoting arm 416 (See also Fig. 9K-L). Through a geared transmission, the angle of the arm 416 can be changed via an electric motor 428 mounted within the body. Aligning with the traditional guidance for needles entering the AV Fistula for hemodialysis, the angle of the arm 416 can range from 0 (parallel to the bottom of the body) to 46 or more degrees, with typical range for use in the 15 to 45 degrees range. In addition to the pitch control, the pivoting arm 416 has a needle mounting 462 assembly and a needle drive system 464 (best seen in Fig. 9J). The needle mounting assembly 462 is configured to receive the needle holder assembly 460.

[0076] In one embodiment, the ultrasound array 420 of the system 400 is rigidly attached or mounted at the bottom of the housing 410. The integrated ultrasound array 420 allows for more refined packaging and better integration of the ultrasound output with the needle insertion system.PCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT

[0077] As noted, a laptop or other computer (not shown) can be provided to work with the system. For this embodiment, the current programming and control is from a USB cable attached laptop. Here also, in addition to being able to work with the ultrasound images coming to the computer from the ultrasound array, the laptop allows for flexibility in programming and refining the movements of the system 400. In an alternative embodiment, the system 400 could still be attached to a laptop with a fully developed user-interface to allow the laptop / cart to accomplish computational or control work. In yet another embodiment, all the system work can be connected with the electronics within the handheld device requiring only a power connection (or even battery power) to operate. In one aspect, there can be a full integration of processing (computer) function into the device itself. As with the previously disclosed embodiment, this system 400 distills the design to minimal degrees of freedom so that the system 400 can perform as desired in all conditions of use.

[0078] Again, the degrees of freedom present in this embodiment are pitch and translation (e.g., insertion and retraction). Pitch rotates the needle to a desired insertion angle. Additionally, as the needle advances, pitch is used to alter the needle tip path angle along the target vessel. The combination of these two degrees of freedom allows the system to advance the needle tip into the target fistula or graft and then adjust to allow the needle tip to advance inside of the fistula or graft, safely without the needle tip scraping the interior vessel wall, for complete insertion. Or to retract the needle tip back to a safe location if the system detects unsafe patient or device movement.

[0079] In the disclosed embodiment, a range of motion is 10 degrees to 46 degrees with the lower angles for final insertion, such as approximately 15 degrees. Still referencing Figs. 9J-L, the motor 428 is used to drive a spur or pitch gear 432 and control the angle of pitch. A proximal end of the pivoting arm 416 is equipped with a connector 464 that engages the pitch gear 432 so that as the pitch gear 432 is caused to rotate, the arm 416 pivots relative to the body housing 410 and relative to target tissue. Figure 9K depicts the connector 436 with the rest of the pivoting arm removed and Fig. 9L shows the connector also removed.

[0080] The angle of the needle 430 for insertion needs to be sensed for the control circuit. Various approaches can be taken, such as the same being provided by a motor encoder, with a homing step, or an absolute encoder could be used in the device body or arm to measure without a required homing step.PCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT

[0081] Insertion of a needle 430 within target tissue involves the linear advancement of the needle 430 to the target location. This can be along the angle created / controlled by the pitch. Insertion consists of linear motion to insert the needle 430 into a patient along a path to target vessel or location. The tip of needle 430 starts outside of patient and is advanced through skin layer, into vessel / target (not shown). In one embodiment, up to 35mm of travel is provided. Length of travel is selected to ensure the tip of needle 430 can be held away from the patient to minimize the likelihood of inadvertent needle sticks but close enough to make the insertion fast once the user decides to insert. Here also, the needle can be protected (i.e. sheathed or covered) until it is needed. In an alternative embodiment (not shown), the needle protective structure can be removed by the system itself (rather than the user) as part of a delivery process.

[0082] As best seen in Fig. 9J, the pivoting arm assembly 416 includes a motor 462 that is connected to a leadscrew 454 to drive and translate a needle holding assembly 460 toward and away from a target. The needle holding assembly 460 is removable for cleaning and is designed to present smooth, wipeable surfaces. In this way, the assembly is easier to clean and simplifies and eases user workflow. In one alternative embodiment, the needle holding assembly 460 is equipped with magnetic encoder technology that involves a single magnet configured within the assembly that communicates or cooperates with one or more sensors integrated into the system arm to thereby provide information regarding needle holder presence and / or absolute positioning. With additional reference to Fig. 9F, the needle assembly 460 includes a leadscrew receiving structure 463 that receives and engages with the leadscrew 454. As shown, the motor 462 driving the lead screw 454 is located in the arm 416, and gears 464 are employed to transmit rotating energy from the motor 462 to the leadscrew 454. The leadscrew 454 and leadscrew receiving structure 463 thus cooperate to position the needle assembly 460 as desired relative to patient anatomy. Through the connection of the leadscrew 454 and leadscrew receiving structure 463 of the needle holder assembly 460, rotating the leadscrew 454 results in translating the needle holder assembly 416 as desired relative to target tissue. As shown in Fig. 9M, the pivoting arm assembly 460 additionally includes electronics 470 that cooperate and communicate with the controller 423 to control operation of the arm 460.PCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT

[0083] Turning now to Figs. 10A-E, there are presented details concerning ultrasound transducer or probe or array arrangements that can be embodied in one or more of the vascular access systems disclosed herein. With the use of an automated needle or catheter delivery system, the system uses a sensing system in place to understand the target anatomy and determine the path for the needle prior to placement. Conventionally, when placing cannulas for dialysis, the user relies on visual, tactile, and auditory feedback (i.e., “look, listen and feel”) to make the assessment of where and how to insert the needle. While this feedback is still available to the user in order to make determinations for automatic placement of the needle, additional sensing can be advantageous. Due to the complexity of many patients’ fistula or target anatomy, ultrasound provides a significant increase in the information available and allows for the automatic system to determine needle path and monitor its placement. Additionally, with ultrasound, more than just imaging data may be collected and tracked. Thus, an ultrasound array can provide the needed information to better automate needle delivery. It has been noted that while a single linear ultrasound array can be employed to image internal anatomy, it lacks an ability to facilitate linking what is imaged with targeting or sensing information for automated needle delivery. That is, in certain instances, there can exist a challenge with the introduction of slightly off parallel angles where usable information available begins to drop off. In particular, when a single linear array is placed longitudinally with respect to a target vessel, the information available begins to drop off as the array presents a view that is not actually down a center of the target vessel and as a result, can lead to incorrect assessments by the automated system. Further, when placing a single linear array transverse to a target vessel, only a single cross-section of the vessel is provided, and while some assumptions can be made regarding how the target vessel or fistula changes along its length, it can be difficult to fully understand the path a needle or catheter must follow during placement.

[0084] Accordingly, to provide a more useful and complete imaging of target anatomy for automated needle or catheter insertion, a T-array, I-array, or a two parallel ultrasound transducer array arrangement can be employed. A long linear array ultrasound is made up of many transducer elements. This allows for system electronics controlling the ultrasound to control adjacent elements as a phased array and increases the imaging capability from thePCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT ultrasound probe. In one or more embodiments, an array of one hundred twenty-eight elements can be employed, and these elements can be split into two equal arrays of sixty -four elements. Such a configuration is beneficial to a vessel identification and targeting application as the arrays are separate but electrically connected and controlled.

[0085] In one embodiment (Figs. 10A), the ultrasound transducer, probe or array 600 includes a first array 602 that is oriented perpendicularly to a second array 604. The first array 602 is transverse to the vessel or fistula 610 and will clearly observe the cross section of the vessel 610, thereby allowing for assessment of size and location (depth). The second array 604 will provide longitudinal information for a portion of the length of the vessel or fistula 610. Combined together, the arrays 602, 602 provide more complete data for feedback so the user can position the system in alignment with the vessel or fistula 610 for needle or catheter 612 delivery based on signals from the system to the user to move the system a desired direction and / or distance.

[0086] In another embodiment (Fig. 10B), two arrays 620, 622 are arranged parallel to one another and arranged transverse to the vessel or fistula 610. The resultant imaging presents two cross-sections of the vessel or fistula 610 at a known distance apart. This distance between arrays can be based on the length of the needle or catheter 612 to be inserted and the specific application. Additionally, distances between transducers can be informed by anatomic assumptions and the ability to infer the anatomy between the two images. In one particular aspect, a distance of 14mm is used, but can range from l-3cm or more depending on anatomy and application. The parallel transducer arrangement simplifies the process of aligning the transducers to the vessel since entire cross-section of the target vessel will be observable in each transverse view. With the cross-sectional image of the vessel present, the centerline of the vessel is measured relative to the known device centerline. Additionally, with two parallel transducers at a known distance from each other the center points of each vessel cross-section is utilized to determine directionality (angled left, right, up, down) of the vessel via interpolation, thus providing a means to ensure the device can be translationally aligned to the center of the vessel as well as rotationally aligned to any directional change of the vasculature that is present. Notably, there is an interaction between the first cross-section and the point of intersection with the needle / catheter 612 and the vessel or fistula 610 to facilitate proper device insertion and advancement. One such arrangement is depicted in Figs. 10C-D, whichPCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT show the array attached to a flex circuit 630 that communicates with a controller (not shown).

[0087] In alternative or additional approaches, an ultrasound transducer, probe or array can embody one or more of a tilting primary array, an “I” shaped array, or an assembly configured to control the distance between the two transverse arrays. Array tilting allows for a wider view of the needle entry to the vessel. Additionally, array tilt allows for doppler assessment which increases the information from the ultrasound, especially in the case of wanting to track the quality and condition of the fistula over time (See Fig. 6E for example). Additionally, one or more embodiments include an array element perpendicular to two parallel arrays to create an I- shape. The resulting imaging can be the two transverse views of the target anatomy at a fixed spacing as well as a short segment of longitudinal imaging. With alignment aided with the transverse images, a centered longitudinal image provides feedback as the needle progresses down the vessel.

[0088] Further, structure can be provided to accomplish spacing between two arrays with a lead screw or servo, manually or automatically in response to system data collection. The motion would maintain the parallel nature but could move the array to be suited to the application, such as a different needle or catheter length, or making an adjustment to review the presented anatomy.

[0089] In another particular aspect, the system is configured to prompt the user to move the device in a set pattern over the target location. In doing so, imaging data can be collected to create a 3D model of the target vessel along with landmarks so that the system would know where it is prior to needle placement and compute the path for needle entry. Further, ultrasound in the system is employed to scan along the length of the fistula or graft and provide identification of the patient, similar to a fingerprint, so that even without technician entry the system would know the patient and be able to bring up information about their treatment and treatment history.

[0090] In one embodiment, patient scanning is performed as needed such as on a periodic basis so that the position of the targeted vasculature or graft is confirmed. In one approach, additional scanning is conducted on a schedule such as up to every six or more months. The pre-scanning process can involve medical personnel that develop a strategy and plan to map an approach to advancing one or more needles into targeted vasculature through an acceptablePCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT path. The strategy or plan can include a small range of acceptable paths as well as a most desired path to vascular access. This plan can include mapping paths to numerous predetermined cannulation sights which can then be accessed on a rotating basis via a rope ladder access technique. The acceptable cannulation sights and paths can all be determined after the pre-scanning process and prior to the patient undergoing dialysis. Accordingly, in one embodiment, the cannulation site and path to cannulation does not need to be calculated in real time - but can be determined before the patient puts their arm under the system.

[0091] In one embodiment, a vascular access system includes structure and functionality to target vasculature and to position one or more needles or catheters within the target vasculature without the need for skilled personnel. In one aspect, the needle or needles or catheters are advanced at a pre-determined angle and depth within a patient’s body and within target vasculature. Significantly, the system is effective for providing vascular access and assists with cannulation throughout the body including specifically for radio-cephalic, brachiocephalic and brachio-basilic fistulas and may range from the forearm to the upper arm or other locations on the body.

[0092] During the pre-scan, mapping and planning process, one or more of an MRI scan, a CT scan, ultrasound scan, infrared view, or 3-D photography is employed to collect information on a patient’s vascular morphology and anatomy.

[0093] In various embodiments, the system tracks insertion sites, sizes, location and geometry, insertion dates, flow rates, treatment frequency and treatment length, as well as allows for patient input so that complications or infections are tracked and monitored. In certain alternative embodiments, the system suggests insertion sites based upon a combination of historical data about the patient’s previous cannula insertions with the device and the patient’s anatomic imaging data, as well as in view of patient input, to provide options to the patient or healthcare provider concerning needle placement. In certain alternative embodiments, the system uses data from other patients with similar anatomy and vascular morphology that had successful cannula insertion at a given site, in order to further inform the system’s recommendation. Further, in alternative embodiments, the system includes a remote interface or computer that allows the patient, health care provider, or other connected health device (e.g. by Bluetooth) to enter patient health information including heart rate, blood pressure, and blood flow as well as patient diet, medication regiment, and exercise.PCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT

[0094] In alternative embodiments, the system controller manages or provides the assessment of a fistula or graft prior to cannulation. The system can be used to identify a fistula of a particular patient using ultrasound along a length of a target area or a portion thereof, and using data stored in the system, for a present dialysis session. Should an obstruction be detected, the system will prevent a cannulation procedure and can alert the patient or health care provider so that further assessment and / or intervention can be conducted. Various sensors and actuation mechanisms are provided to automate the assessment process, or portions thereof. In certain approaches, the system embodies one or more sensors that recognize thrill or vibrations, that operate like a stethoscope to track for bruit (i.e. sounds of heartbeats or blood flow), or listens to flow to look for obstructions or restrictions in the vessel. In one particular approach, the system employs a laser or other pointer to show a technician where the insertion site should be and then the technician uses standard cannulation techniques to insert a needle or catheter. In yet other approaches, the system is configured to monitor the fistula diameter or cross-section area throughout each heartbeat cycle for changes, as such changes over time can be used to detect access issues such as a stenosis, for example.

[0095] With reference to Fig. 11, there is an integrated monitoring system 700 that cooperates with or includes a vascular access system. In one or more embodiments, the integrated monitoring system 700 is configured to proactively monitor and predict the health of target anatomy or arteriovenous fistulas in hemodialysis patients. The system includes one or more of a data collection module 702, a processing module 704, a machine learning and prediction module 706, and a user-interface module 708. In a particular aspect, the data collection module 702 is a noninvasive measurement module that includes one or more of data received from a dialysis machine 710, an ultrasound imager 712 and data received from a digital stethoscope auscultation device 714. Integrating data from ultrasound 712, digital stethoscopes 714 and the dialysis machine 710, combined with machine learning offers a comprehensive, objective and early predictive tool for AVF health assessment.

[0096] In one aspect, the data collection module 702 gathers data from one or more of a portable or other ultrasound imager 712, the digital stethoscope 714, and in real-time from the dialysis machine 710. The processing module 704 processes acquired data and extracts relevant features for analysis. The machine learning and prediction module 706 analyzes extracted features and predicts future AVF or anatomy health. The user-interface module 708PCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT provides users or clinicians with clear visualizations, reports, risk scores, and / or actionable recommendations for proactive anatomy or AVF management. By synergistically integrating these modules and employing advanced machine learning algorithms, the system 700 offers an approach for comprehensive and objective assessments to predict a risk of future complications like stenosis or thrombosis.

[0097] In one or more aspects, the system 700 combines morphological data from ultrasound imaging, functional data from auscultation sounds, and real-time dialysis parameters for a holistic understanding of AVF or anatomy health. The system 700 leverages machine learning to analyze combined data and to identify subtle changes indicative of potential AVF or anatomy dysfunction, enabling proactive interventions. The system 700 reduces reliance on subjective interpretation of physical examination, leading to more reliable and consistent assessments and enables rapid automated screening of AVFs or other anatomy to thereby optimize workflow for healthcare professionals or others in busy dialysis settings. Moreover, in one or more embodiments, the system 700 can employ portable and user-friendly home-based monitoring and can tailor monitoring strategies and interventions based on individual patient risk priorities, to optimize resource allocation, and improve patient outcomes. Early detection and prevention of complications can therefore reduce hospitalizations and healthcare costs.

[0098] In one particular embodiment, the system 700 includes a portable ultrasound system such as described above (See Figs. 9A-10E, for example). In one or more aspects, the portable ultrasound system 712 includes a frequency (ranging from 2MHz to 30MHz or more) linear array transducer that captures high-resolution 2D B-mode and Doppler ultrasound images of the AVF and surrounding vasculature or anatomy. A 3D image reconstruction module 720 is provided which processes 2D ultrasound images to generate a 3D fistulogram of the AVF, and offers spatial information and visualization of vessel geometry. There can also be provided an integrated display that provides real-time visualization of ultrasound images, Doppler waveforms and 3D fistulograms. A data storage and transfer module is also provided and configured to store and transmit ultrasound data to the processing unit or module 704.

[0099] The digital stethoscope 714 is a high-sensitivity microphone that captures AVF or other blood flow sounds with high fidelity and minimal noise. Sound amplification and filtering circuitry is further provided to improve sound quality and removes background noise.PCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT The data storage and transfer module cooperates with the digital stethoscope 714 and stores and transmits auscultation sound data to the processing unit 704.

[0100] There is a data interface with the dialysis machine 710. A communication module enables secure and reliable data transfer between the dialysis machine 710 and the processing unit 704. In one or more embodiments, there can be provided data extraction software that extracts relevant data from the dialysis machine including one or more of access flow information including a volume of blood flowing through the AVF or other anatomy per unit time; dynamic venous pressure within a venous segment of the AVF during dialysis; dynamic arterial pressure within an arterial segment of the AVF or other anatomy during dialysis; and dialysis dose (Kt / V) measuring dialysis adequacy.

[0101] In one or more embodiments, the processing unit 704 is a microprocessor or embedded system that executes 3D reconstruction, feature extraction and / or machine learning algorithms. The processing unit 704 includes a memory that stores one or more of software programs, trained models and patient data, and there can be provided a user interface that offers a user-friendly interface for data visualization, result interpretation and system interaction.

[0102] Further, the system 700 includes various software algorithms including one or more of a 3D fistulogram reconstruction functionality, a sound processing and feature extraction functionality, and / or machine learning and prediction algorithms.

[0103] The fistulogram reconstruction functionality can involve one or more of an image segmentation algorithms that identify and delineate the boundaries of the AVF or other anatomy and surrounding vessels in 2D ultrasound images, and / or 3D reconstruction algorithms that generate a 3D model of the AVF or other anatomy by stitching together segmented 2D images, offering spatial information and visualization of vessel geometry.

[0104] The sound processing and feature extraction functionality involves one or more of noise reduction that filters out background noise and artifacts from audio recordings and segmentation functionality that identifies and isolates individual cardiac cycles for detailed analysis. Feature extraction functions to extract relevant features from both ultrasound and auscultation data including one or more of ultrasound features, auscultation features and dialysis parameters. Ultrasound features include one or more of vessel diameter, blood flow velocity, vessel wall characteristics, and anatomical variations from 3D fistulograms. DialysisPCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT parameters include one or more of access flow, dynamic venous pressure, dynamic arterial pressure, dialysis dose and the trends thereof over time.

[0105] As stated, the system 700 can include machine learning and prediction algorithms. These can include one or more of unsupervised learning that functions to identify patterns and clusters in the combined data to detect anomalies or deviations from normal AVF or other anatomy function. Alternatively or additionally, there is supervised learning that train models to predict the risk of future AVF or other anatomy complications (e.g. stenosis, thrombosis) based on extracted feature and labeled training data. This involves one or more of regression algorithms to estimate stenosis severity or predict future function decline or survival analysis models used to analyze time-to-event data for prediction time until AVF or other anatomy failure or intervention. Further, there is additionally or alternatively provided a time-series analysis that involves analysis of temporal trends in ultrasound, auscultation and dialysis data to predict future changes in function.

[0106] Moreover, as noted, the system 700 can be configured to provide useful integration and workflow. A portable or other ultrasound 712 and digital stethoscope 714 can be used to collect data from the patient during routine dialysis sessions, while simultaneously extracting data from the dialysis machine 710. The system 700 reconstructs a 3D fistulogram and extracts features from both ultrasound 712 and stethoscope 714 data to integrate with dialysis paraments set for the patient. The extracted features and dialysis data are analyzed by machine learning 706 algorithms to predict the risk of future complications. The 3D fistulogram, relevant features, dialysis parameters and predicted risk scores are then presented to clinicians through the user interface. Thereafter, based upon the predicted risk and individual patient factors, the system 700 can recommend one or more of personalized monitoring plans, potential interventions, and / or follow-up strategies that aid clinicians in proactively managing patient or AVF health and prevent complications.

[0107] In various alternative or additional embodiments, the user interface has one or more of a user-friendly platform for data input, ultrasound and audio capture and model prediction display. The system can also additionally or alternatively generate reports summarizing AVF or other assessments, including ultrasound images, audio recordings, and the predicted risk of failure and / or reports can be integrated into electronic health records or shared with other healthcare providers for coordinated patient management.PCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT

[0108] In one or more embodiments, a computer-implemented system and method for ultrasound-guided arteriovenous fistula cannulation provides real-time guidance and monitoring for improved cannulation success rates and reduced complications.

[0109] In one embodiment, the system receives real-time ultrasound image data from an ultrasound probe operating at a predetermined frequency range of 7-15 MHz. This frequency range has been optimized for superficial vascular imaging while maintaining adequate penetration depth for typical fistula locations.

[0110] The system implements a sophisticated image pre-processing pipeline to enhance image quality and optimize vessel visualization. Initially, the system applies a median filter with a 5x5 kernel size, which effectively reduces speckle noise characteristic of ultrasound imaging while preserving critical vessel boundary information. Following noise reduction, the system performs adaptive histogram equalization with precisely tuned parameters - a clip limit of 0.03 and tile size of 8x8 pixels - to enhance local contrast while preventing over-amplification of noise. A final pre-processing step involves normalizing pixel intensities to a standardized range of 0-1, ensuring consistent input scaling for subsequent analysis steps.

[0111] In a preferred embodiment, the system employs a deep convolutional neural network (DCNN) for image analysis, which has been trained on a comprehensive labeled dataset comprising at least 1000 ultrasound images of arteriovenous fistulas. The DCNN implements an encoder-decoder architecture with skip connections, enabling both efficient feature extraction and precise localization capabilities. This neural network generates a binary segmentation mask that precisely identifies the arteriovenous fistula location while simultaneously calculating critical parameters including vessel depth and diameter measurements.

[0112] The system can feature automatic optimization of ultrasound scanning parameters based on real-time analysis of detected vessel characteristics. This includes dynamic modulation of gain settings based on detected vessel depth for optimal visualization and to ensure consistent image quality, and precise focusing where the focus position is automatically maintained at the detected vessel depth ± 0.5 cm.

[0113] In an embodiment, real-time visual guidance is provided through an integrated display system that overlays the segmentation mask on the live ultrasound feed. ThePCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT segmented mask may also be provided in a standalone manner without the overlay on ultrasound stream. The system can continuously display numerical indicators of vessel depth and calculates the optimal needle insertion angle. To enhance procedural safety, the system can generate immediate visual alerts when the needle trajectory deviates more than allowed thresholds from the calculated optimal path.

[0114] In various approaches, the system maintains continuous vessel tracking and updates all guidance parameters at a minimum frame rate of 15 frames per second, ensuring smooth, real-time feedback during the cannulation procedure. For quality assurance and procedure optimization, the system records successful cannulation parameters including vessel depth, insertion angle, ultrasound settings, and patient-specific annotations.

[0115] In another embodiment, the system incorporates a comprehensive safety monitoring module that actively monitors for potential complications during the procedure. This module can continuously analyze vessel characteristics to detect sudden changes in diameter or shape that might indicate infiltration. Upon detecting potential complications, the system immediately generates both visual and audible alerts to notify the operator. All such events are automatically recorded and stored for subsequent quality assurance review and system optimization.

[0116] In certain approaches, all recorded data and parameters can be stored in a secure HIPAA complied database that enables subsequent analysis for procedure optimization and quality improvement initiatives. This data repository facilitates ongoing system refinement and the development of patient-specific cannulation protocols.

[0117] In various embodiments, a system and method for predictive arteriovenous fistula monitoring provides comprehensive analysis and early detection of potential complications through multi-modal data integration and advanced machine learning techniques.

[0118] In one embodiment, the system comprises a multi-modal data acquisition subsystem that simultaneously collects data from multiple sources. The data acquisition subsystem includes an ultrasound probe operating at frequencies between 7-15 MHz with a frame rate of 15-30 frames per second, enabling high-resolution imaging of the fistula and surrounding vessels. A digital stethoscope, operating at a sampling rate of 44.1 kHz with 24- bit resolution, captures acoustic signatures associated with blood flow patterns. The subsystemPCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT further includes a real-time interface with dialysis machine data streams, enabling continuous monitoring of critical parameters during treatment sessions. All data collection and transmission components are designed to maintain HIPAA compliance through encrypted data storage and secure transmission protocols.

[0119] In an embodiment, system implements a sophisticated 3D fistulogram reconstruction module that processes the acquired ultrasound data. This module employs a deep learning-based segmentation network built on a U-Net architecture with a ResNet-50 backbone, achieving vessel lumen segmentation accuracy exceeding 95% while maintaining real-time performance of at 20 frames per second or more. The segmentation output feeds into a spatial reconstruction algorithm that utilizes electromagnetic tracking data with submillimeter accuracy (±0.5mm). The algorithm employs iterative closest point (ICP) methods for precise registration between consecutive frames, ultimately generating a high- resolution 3D mesh model with voxel resolution of 0.1mm3.

[0120] In another embodiment, the system includes an acoustic signal processing module that analyzes data from the digital stethoscope. This module implements a multi-stage processing pipeline, beginning with adaptive noise to isolate relevant acoustic signals. Wavelet-based denoising with soft thresholding further enhances signal quality. The module segments individual cardiac cycles using a modified Pan -Tompkins algorithm for QRS detection, followed by dynamic time warping for precise cycle alignment. The processor extracts a comprehensive set of features including spectral characteristics (FFT coefficients, spectral centroid, bandwidth), temporal features (RMS energy, zero-crossing rate), and statistical measures (kurtosis, skewness).

[0121] The unified feature extraction and analysis system can process data from all input sources to generate quantitative measurements and derived metrics. For ultrasound data, the system measures vessel parameters with high precision, including diameter (±0.1mm), flow velocity (±5% accuracy), and depth (±0.1mm), These measurements enable calculation of critical derived metrics such as flow rate and resistance index. Acoustic data analysis quantifies bruit characteristics including intensity (measured in dB), frequency components (Hz), and duration (ms), while also identifying abnormal patterns such as high-frequency turbulence and amplitude variations. The system continuously monitors dialysis machine parameters, tracking access flow (Qa) at 15-minute intervals, pressure measurements, andPCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT Kt / V calculations for each session, maintaining trending metrics across weekly sessions for comprehensive analysis.

[0122] In a preferred embodiment, the system implements a hybrid machine learning prediction system that combines multiple analytical approaches. An anomaly detection module utilizes an Isolation Forest algorithm with a contamination factor of 0.1, supplemented by a One-class SVM with RBF kernel and an autoencoder network for robust detection of unusual patterns. The predictive modeling pipeline incorporates a Gradient Boosting Classifier for complication risk assessment, an LSTM network for temporal pattern analysis, and a Cox Proportional Hazards model for survival analysis. This hybrid approach enables real-time risk assessment with updates every 5 minutes, generating alerts when risk scores exceed predetermined thresholds and providing confidence intervals for all predictions.

[0123] The system can present information through a decision support user interface that renders the 3D fistulogram with color-coded risk areas, enabling intuitive visualization of potential problem areas. The interface provides comprehensive trend analysis across multiple sessions and generates detailed reports including quantitative measurements, risk assessments, and recommended interventions.

[0124] In various alternative or additional embodiments, the system can support remote monitoring of AVFs or other anatomy by transmitting data and reports securely to healthcare providers or a central monitory system. This can enable timely intervention and personalized surveillance strategies based on the predicted risk of AVF or other failure.

[0125] In various alternative or additional embodiments, the system can incorporate robust data privacy and security measures to protect patient information and comply with relevant regulations (e.g., HIPP A, GDPR). Encryption and secure data transmission protocols can additionally or alternatively be implemented for remote monitoring and data sharing.

[0126] In various alternative or additional embodiments, the system can include a machine learning model that can be periodically retrained and updated as more data becomes available to thus improve accuracy over time. The system may also incorporate feedback mechanisms to collect user input and refine the model or user experience. Moreover, the system can additionally or alternatively include a module configured to simulate blood flow within the AVF or other anatomy based on 3D fistulogram and Doppler data.PCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT

[0127] With reference to Fig. 12, a treatment approach can additionally or alternatively include an anatomy or stenosis monitoring system 800. The system 800 includes an integrated system for monitoring and detecting stenosis in AVFs or other anatomy used for hemodialysis. The system 800 can combine one or more of an ultrasound device 802, a digital stethoscope 804 and advanced software algorithms to provide comprehensive and user-friendly solutions for AVF or other assessments. The system 800 captures both anatomical and acoustic data, allowing for early detection of stenosis and timely intervention, improving AVF patency and reducing complications. The system 800 is designed for various settings including one or more of dialysis centers, hospitals and home-care environments.

[0128] The system 800 combines anatomical imaging (e.g., portable ultrasound) and acoustic information (e.g., digital stethoscope) for a more comprehensive AVF or anatomy assessment compared to single-modality approaches. The system 800 can additionally or alternatively use sensors and advanced algorithms to provide objective and quantitative measurements to reduce subjectivity and inter-observer variability. Early detection of stenosis or other obstructions or narrowing is provided through objective sound analysis and high- resolution imaging, enabling timely intervention and potentially preventing failures. The system 800 can additionally or alternatively be portable for use in various settings to increase access to regular monitoring and the system can additionally or alternatively include a user- friendly interface for ease of operation by both healthcare professionals and patients to thereby promote self-monitoring and patient engagement. Further, in one or more embodiments, the system 800 can additionally or alternatively construct a 3D model 806 of the fistula vessel or other anatomy by stitching together 2D ultrasound scans to provide a detailed anatomical visualization for improved diagnosis and intervention planning.

[0129] The system 800 includes a portable ultrasound device 802 such as disclosed above (See Figs. 9A-10E). In one or more embodiments, the ultrasound device 802 include sone or more of a linear transducer (2-10 MHz), integrated software for image acquisition, processing and display, and / or wireless connectivity capability for data transfer. In one aspect, the ultrasound device 802 is compact and lightweight for portability.

[0130] The digital stethoscope 804 can include one or more of a high-sensitivity microphone configured to capture subtle AVF or other anatomy sounds, noise-cancellationPCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT technology for clear audio recording, an integrated amplifier for enhanced clarity, and / or wireless connectivity for data transfer and integration with ultrasound data.

[0131] In one or more embodiments, the system 800 can include a sound analysis algorithm 808 that employs deep learning models (e.g. convolutional neural networks, recurrent neural networks) to analyze AVF or other sound characteristics to identify patterns indicative of stenosis or other characteristics. In one or more aspects, the sound analysis algorithm has one or more extract features from sound including frequency spectrum, temporal variations and harmonic components. Further, the algorithm is trained on one or more of a large dataset of labeled AVF or other sounds, validated against gold-standard methods like DUS, and / or output quantitative stenosis or other features probability scores and identifies specific sound abnormalities.

[0132] The system 800 can additionally or alternatively include an image processing algorithm. The image processing algorithm 810 can one or more of perform real-time image process for noise reduction and edge enhancement for improving image quality, include algorithms for automated measurement of vessel diameter, blood flow velocity or other relevant parameters, and / or feature automated detection of anatomical abnormalities, including stenosis, aneurysms and thrombi.

[0133] Moreover, the system 800 can additionally or alternatively include a fistulogram construction algorithm. The fistulogram construction algorithm 812 can one or more utilize image registration and stitching techniques to create a 3D model of the AVF vessel or other feature from multiple 2D ultrasound scans, include algorithms for accurate alignment and interpolation of images to ensure smooth and realistic 3D reconstruction, and / or provide a detailed anatomical visualization for improved diagnosis, intervention, planning and patient education.

[0134] Furthermore, the system 800 can be characterized by one or more of an integration and / or a user interface 814. In one or more aspects, the user interface can be used for entering data into the system 800, selecting options, setting modes, or otherwise controlling functionality of the device or system 800, The system 800 can include one or more of software algorithms that are integrated into a unified platform, providing a seamless user experience and a user-friendly interface for image display, sound playback, data analysis and reporting. The interface can additionally or alternatively allow for easy navigation and visualization of bothPCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT anatomical and acoustic data to facilitate efficient and informed deci si on -making. The system 800 can also alternatively or additionally provide alerts or notifications based on pre-defined thresholds for stenosis or other probability to prompt further investigation or intervention.

[0135] While the present disclosure has been described with reference to the specific embodiments thereof, it should be understood by those skilled in the art that various changes may be made and equivalents may be substituted without departing from the true spirit and scope of the disclosure. In addition, many modifications may be made to adapt a particular situation, material, composition of matter, process, process step or steps, to the objective, spirit and scope of the present disclosure. All such modifications are intended to be within the scope of the present disclosure.

Claims

PCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCTCLAIMS:That which is claimed is:

1. A system for assisting with positioning a needle or catheter into an anatomical site of a patient, comprising: an integrated monitoring system configured to monitor the health of target anatomy or arteriovenous fistulas in hemodialysis patients.

2. The system of claim 1, further comprising a data collection module.

3. The system of claim 1, further comprising a processing module.

4. The system of claim 1, further comprising a machine learning and prediction module.

5. The system of claim 1, further comprising a user-interface module.

6. The system of claim 1, further comprising a dialysis machine.

7. The system of claim 1, further comprising an ultrasound imager.

8. The system of claim 1, further comprising a digital stethoscope auscultation device.

9. The system of claim 1, wherein the system is configured to integrate data from ultrasound, a digital stethoscope and a dialysis machine, and to employ machine learning to provide a comprehensive, objective and early predictive tool for AVF health assessment.

10. The system of claim 1, wherein the system is configured to be employed with a handheld electromechanical device.PCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT11. The system of claim 1, further comprising an integrated system for monitoring and detecting narrowing or stenosis, thrombosis, or a presence of other vessel abnormalities in target anatomy or in arteriovenous fistulae used for hemodialysis.

12. A system for positioning a needle into an anatomical site of a patient, comprising: an integrated system for monitoring and detecting narrowing or stenosis in target anatomy, grafts, endovascular fistulae, or in arteriovenous fistulae used for hemodialysis.

13. The system of claim 12, further comprising an ultrasound device.

14. The system of claim 12, further comprising a digital stethoscope.

15. The system of claim 12, further comprising advanced software algorithms providing comprehensive and user-friendly solutions for assessing target anatomy.

16. The system of claim 12, wherein the system is configured to provide an integrated approach that combines the benefits of multiple modalities, enabling comprehensive and objective assessments.

17. The system of claim 12, further comprising an integrated monitoring system configured to proactively monitor and predict the health of target anatomy or arteriovenous fistulas in hemodialysis patients.

18. The system of claim 1 or 12, wherein the system simultaneously processes and correlates ultrasound, acoustic and dialysis machine data streams in real time.

19. The system of claim 19, wherein the system implements synchronized data processing of ultrasound data at 15-30 frames per second, acoustic data at 44.1 kHz, and dialysis parameters every 15 minutes.PCT / US25 / 13146 27 January 2025 (27.01.2025)Atty. Docket: CANN-003PCT20. The system of claim 1 or 12, further comprising a 3D reconstruction that is accomplished by a deep learning-based segmentation network built on a U-Net architecture with a ResNet-50 backbone specifically optimized for vasculature structures.

21. The system of claim 1 or 12, further comprising an electromagnetic tracking that is integrated for spatial registration with accuracy metrics of plus or minus 0.5mm.

22. The system of claim 1 or 12, further comprising a 3D reconstruction that involves a realtime performance guarantee of 20 frames per second minimum while maintaining accuracy of over 95%.

23. The system of claim 1 or 12, further comprising a hybrid machine learning approach involving one or more of a multi-level analysis combining an Isolation Forest algorithm for anomaly detection, an LSTM for temporal patterns and a Cox Proportional Hazards model for survival analysis.

24. The system of claim 23, wherein the system provides both immediate and long-term risk assessment.

25. The system of claim 1 or 12, further comprising an acoustic data analysis that quantifies bruit characteristics including intensity, wherein there is real-time correlation with ultrasound and dialysis data.

Images