System and method for monitoring vascular function

Abstract

A method of monitoring a blood vessel, the method comprising using a system for monitoring blood vessel function, imaging a patient's body to obtain a geometry of a blood vessel; imaging a patient's body to obtain a shape of a location of the patient's body using image analysis; and analyzing vibrations of the patient's body at a location of the patient's body including the blood vessel to observe, listen to, and feel blood vessel function. A method for monitoring blood vessel function, the method comprising illuminating one or more blood vessels passing through a patient's skin; taking at least one image of the blood vessels; analyzing the at least one image; and calculating a parameter related to blood vessel function based on image analysis. Related devices and methods are also described.

Description

[0001] Related applications

[0002] This application is a partial successor to U.S. Patent Application No. 16 / 656,585, filed October 18, 2019. The contents of the foregoing application are incorporated herein by reference as if fully set forth herein.

[0003] Technical Field and Background Technology

[0004] This invention generally relates to the field of monitoring patient blood vessels. Some aspects more specifically relate to the early diagnosis of vascular dysfunction, and even more specifically to the early detection of vascular access failure in patients undergoing hemodialysis. Some aspects more specifically relate to the measurement of fistulas.

[0005] Vascular access (VA, fistula, or graft) makes life-saving hemodialysis treatment possible, but it can also lead to access-related problems.

[0006] The term “VA” or “vascular access”, used in all its grammatical forms throughout this specification and claims, refers to all types of vascular access constructions, biological and synthetic, including, as some non-limiting examples, arteriovenous (AV) fistulas, synthetic grafts, and venous catheters.

[0007] One type of long-term access is the AV fistula. A surgeon connects an artery to a vein, usually in the arm or leg, to create an AV fistula. When the surgeon connects the artery to the vein, the vein becomes wider and thicker, making it easier to place the needle used for dialysis. AV fistulas also have a large diameter, allowing blood to flow out and back into the body quickly. The goal of an AV fistula is to allow high blood flow so that a large amount of blood can pass through a dialyzer.

[0008] In the United States, more than 25% of hemodialysis (HD) patients are hospitalized due to access-related factors.

[0009] In 2013, more than 30,000 patients required dialysis-related arterial and venous thrombectomy.

[0010] The cost of the annual incidence rate is estimated to be close to $1 billion.

[0011] VA function and patency are crucial for optimal management of hemodialysis patients. Low VA flow and loss of patency limit the implementation of hemodialysis, prolong treatment duration, and may lead to inadequate dialysis, resulting in increased morbidity and mortality. In long-term VA, especially in grafts, thrombosis is a major cause of VA patency loss and increased healthcare costs.

[0012] The basic concept of VA monitoring and examination is that the vast majority of VAs will develop progressive stenosis at different time intervals, and that if it is detected and corrected (corrective surgery, such as percutaneous transluminal angioplasty - PTA), underdialysis can be minimized or avoided (dialysis dose protection) and the rate of thrombosis can be reduced. Many monitoring and examination methods are available: continuous VA flow, continuous dynamic or static pressure, recirculation measurements, and physical examination.

[0013] Monitoring is the examination and assessment of VA to diagnose VA dysfunction through physical examination, usually within the hemodialysis unit, in order to detect the presence of dysfunction and correctable lesions before VA loss.

[0014] Physical examination can serve as a monitoring tool to rule out low flow associated with impending fistula formation and graft failure. Typically, a VA examination consists of three components: examination, palpation, and auscultation.

[0015] A simple examination can reveal swelling, ischemic fingers, aneurysms, and abundant collateral veins. A strong pulse and weak fremitus in the central part of the venous anastomosis indicate stenosis of the draining vein. Stenosis is palpable, and the intensity and characteristics of the murmur can suggest its location. Localized enhancement of the murmur at the graft or venous anastomosis compared to adjacent segments suggests stenosis. The physical examination may also include an elevation test, which involves elevating the limb with the venous artery (VA) and examining for collapse of the normal pathway. This test is considered normal when the organ is elevated above the level of the patient's heart and the fistula collapses.

[0016] Other background art includes:

[0017] Besarab et al., “Access Monitoring Is Worthwhile and Valuable,” *Blood Purif*, 2006; 24:77–89, February 2006.

[0018] Ehsan Rajabi-Jaghargh and Rupak K Banerjee, “Combined functional and anatomical diagnostic endpoints for assessing arteriovenous fistula dysfunction,” World J Nephrol, 6 Feb 2015; 4(1): 6-18 ISSN 2220-6124;

[0019] Luc Turmel-Rodrigues, “Salvage of immature forearm fistulas for hemodialysis by interventional radiology”, Nephrol Dial Transplant, December 2001; 16(12):2365-71;

[0020] Jürg Schmidli et al., Editor's Choice e Vascular Access: 2018 European Society for Vascular Surgery (ESVS) Clinical Practice Guidelines, Eur J Vasc EndovascSurg (2018) 55, 757e818.

[0021] All publicly available information in the foregoing references, as well as in this specification and all publicly available information in references mentioned therein, are incorporated herein by reference. Summary of the Invention

[0022] This invention relates generally to the automatic monitoring of a patient's blood vessels, more specifically to the early diagnosis of vascular dysfunction, and even more specifically to the early detection of vascular access failure in patients undergoing hemodialysis.

[0023] According to one aspect of some embodiments of the present invention, a system for monitoring vascular function is provided, comprising an illumination source, a detector, a display, and a processor, configured to identify changes in pulse wave velocity relative to a baseline measurement, identify changes in at least one parameter indicating the development of one or more collateral vessels relative to a baseline measurement, identify changes in vessel diameter relative to a baseline measurement, identify changes in vascular spectral analysis, correlate the identified changes, and determine the probability of vascular function failure based on the correlated identified changes.

[0024] According to some embodiments of the present invention, at least one parameter indicating the development of one or more collateral vessels includes one or more of shape, density, and distance from the vessel.

[0025] According to some embodiments of the present invention, the processor is configured to calculate pulse velocity, at least one parameter indicating the development of one or more collateral vessels, vessel diameter, and the rate of change of spectral analysis of the vessel.

[0026] According to some embodiments of the invention, the blood vessels are in the patient's arm, and the processor is further configured to recognize changes in blood vessel collapse when the patient's arm or leg is raised.

[0027] According to some embodiments of the invention, the processor is also configured to calculate the rate of change of vascular collapse when the patient's arm or leg is raised.

[0028] According to some embodiments of the invention, the processor is also configured to identify changes in the composition of blood flowing within the blood vessels.

[0029] According to one aspect of some embodiments of the present invention, a system for monitoring vascular function is provided, comprising an illumination source, a detector, a display, and a processor, the processor being configured to identify changes in at least one parameter indicating the development of one or more collateral vessels relative to a baseline, and to determine the probability of vascular function failure based on the identified changes.

[0030] According to some embodiments of the invention, the processor is further configured to identify changes in one or more of the pulse wave velocity, vessel diameter, and spectral analysis of the vessel relative to a baseline measurement, associate the identified changes in one or more of the pulse wave velocity, vessel diameter, and spectral analysis of the vessel with changes in at least one parameter indicating the development of one or more collateral vessels relative to a baseline measurement, and determine the probability of vascular dysfunction based on the associated changes.

[0031] According to some embodiments of the invention, the blood vessel is located in the patient's arm or leg, and the processor is further configured to identify changes in vascular collapse when the patient's arm or leg is raised, associate the identified changes in vascular collapse when the patient's arm or leg is raised with changes in at least one parameter identified as indicating the development of one or more collateral vessels relative to a baseline, and determine the probability of vascular dysfunction based on the associated changes.

[0032] According to one aspect of some embodiments of the present invention, a method for monitoring vascular function is provided, comprising identifying changes in pulse wave velocity relative to a baseline measurement, identifying changes in parameters indicating collateral vessel development relative to a baseline measurement, identifying changes in parameters indicating collateral vessel development relative to a baseline measurement, identifying changes in vessel diameter relative to a baseline measurement, correlating the identified changes, and determining the probability of vascular function failure based on the correlating identified changes.

[0033] According to some embodiments of the present invention, the method further includes the steps of calculating pulse wave velocity, at least one parameter indicating the development of one or more collateral vessels relative to a baseline, and the rate of change of vessel diameter.

[0034] According to one aspect of some embodiments of the present invention, a method for monitoring vascular function is provided, comprising identifying changes in at least one parameter indicating the development of one or more collateral vessels relative to a baseline, and determining the probability of vascular function failure based on the identified changes.

[0035] According to some embodiments of the invention, the method further includes identifying changes in one or more of pulse wave velocity, vessel diameter, and vascular spectral analysis relative to a baseline measurement; associating the identified changes in one or more of the pulse wave velocity, vessel diameter, and vascular spectral analysis with changes in at least one parameter identified as indicative of the development of one or more collateral vessels relative to a baseline measurement; and determining the probability of vascular dysfunction based on the associated changes.

[0036] According to some embodiments of the present invention, the blood vessels are in the patient's arm or leg, and the method further includes identifying changes in vascular collapse when the patient's arm or leg is raised, associating the identified changes in vascular collapse when the patient's arm or leg is raised with changes in at least one parameter identified as indicating one or more collateral vessels; and determining the probability of vascular dysfunction based on the associated changes.

[0037] According to some embodiments of the present invention, the method further includes the step of identifying changes in the composition of blood flowing within the blood vessels.

[0038] According to some embodiments of the invention, the blood vessels are located in the patient's arm or leg, and the measurements are taken with the patient's arm or leg approximately parallel to the ground.

[0039] According to some embodiments of the invention, the blood vessels are located in the patient's arm or leg, and the measurements are taken at a position where the patient's arm or leg is substantially perpendicular to the ground.

[0040] According to some embodiments of the invention, the blood vessels are located in the patient's arm or leg, and the measurements are taken when the patient's arm or leg is positioned below the patient's heart.

[0041] According to some embodiments of the invention, the blood vessels are located in the patient's arm or leg, and the measurements are taken when the patient's arm or leg is positioned above the patient's heart.

[0042] According to one aspect of some embodiments of the present invention, a method for monitoring vascular function is provided, the method comprising illuminating one or more blood vessels through the skin of a patient; capturing at least one image of the blood vessels; analyzing the at least one image; and calculating parameters related to vascular function based on the image analysis.

[0043] According to some embodiments of the present invention, the method further includes automatically detecting the location of the vascular access (VA) in the at least one image.

[0044] According to some embodiments of the present invention, the method further includes automatically detecting the location of the fistula in the at least one image.

[0045] According to some embodiments of the invention, the method further includes a physician marking the location of the vascular access (VA) in the at least one image.

[0046] According to some embodiments of the present invention, the acquisition of at least one image of the blood vessel is performed by a device configured to provide images including at least one artery and at least one vein under the patient's skin.

[0047] According to some embodiments of the present invention, irradiating one or more blood vessels through the patient's skin includes transmission irradiation of the patient's organs.

[0048] According to some embodiments of the present invention, calculating parameters related to vascular function includes calculating at least one parameter selected from the group consisting of pulse wave velocity, parameters indicating the development of one or more collateral vessels, collateral vessel count, vessel diameter, spectral analysis of the vessel, size of the arteriovenous fistula, and size of the synthetic graft VA.

[0049] According to some embodiments of the present invention, the method further includes estimating the probability of vascular function failure based on parameters.

[0050] According to some embodiments of the invention, it further includes calculating the rate of change of one or more parameters based on performing several measurements on one or more parameters at different times, some of which are based on patient-associated historical data retrieved from a database.

[0051] According to some embodiments of the present invention, the method further includes estimating the probability of vascular function failure based on the rate of change of one or more parameters.

[0052] According to some embodiments of the invention, the method further includes estimating the maturity of the VA based on the rate of change of one or more parameters.

[0053] According to some embodiments of the present invention, the method further includes automatically detecting collateral veins by counting the number of veins in a specific image region in different images taken at different times.

[0054] According to some embodiments of the present invention, the automatic detection of the location of a vascular pathway (VA) in at least one image includes: detecting the location of the vascular pathway (VA) by detecting the junction of a vein and an artery.

[0055] According to some embodiments of the invention, the detection of the junction of veins and arteries is performed using a device capable of providing images of the arteries and veins beneath the patient's skin.

[0056] According to some embodiments of the present invention, the automatic detection of the location of a vascular pathway (VA) in at least one image includes performing spectral analysis of at least one image.

[0057] According to some embodiments of the present invention, the method further includes: measuring pulse parameters by detecting pulse positions in two images taken at different times and comparing the pulse positions in the two images.

[0058] According to some embodiments of the present invention, the method further includes: measuring the pulse velocity by detecting the position of a pulse in two images and dividing the distance along the center line of the blood vessel in the two images by the time difference between the two images.

[0059] According to one aspect of some embodiments of the present invention, a method is provided to monitor vascular function in dialysis patients as an alternative to physical examinations performed by medical personnel, the method comprising: generating at least one image of a patient's organ, rather than manually manipulating the patient's organ; analyzing the at least one image; and classifying the patient's condition as either suitable for dialysis or at risk of stenosis.

[0060] According to some embodiments of the present invention, it further includes: irradiating one or more blood vessels through the patient's skin, and wherein analyzing the at least one image includes calculating parameters related to vascular function based on image analysis.

[0061] According to one aspect of some embodiments of the present invention, a system for monitoring vascular function is provided, the system comprising: an illuminator configured to illuminate a patient's blood vessels through the patient's skin; a camera configured to capture an image of at least one blood vessel through the patient's skin; an image analyzer configured to process the at least one image; a calculator configured to calculate parameters related to vascular function based on the image analysis; a classifier configured to classify the patient's condition as either suitable for dialysis or at risk for dialysis; and a display configured to provide a caregiver with a report of the patient's condition and at least one of the parameters related to vascular function.

[0062] According to one aspect of some embodiments of the present invention, a system and method for measuring parameters related to fistulas are provided.

[0063] In some embodiments, a system is provided that includes an optical device for acquiring one or more images of the same patient's fistula during monitoring.

[0064] In some embodiments, one or more measurements and / or features may be extracted from the image—and their changes over time may be monitored. In some embodiments, the features are the time-line derivatives of parameters measured or estimated in the image; as a non-limiting example, the number and size of collateral veins may change over a period of time, such as days / weeks / months.

[0065] In some embodiments, the features of interest are derived from a graphical representation of the identified blood vessels, and / or how the representation changes over time, by varying the number of nodes, the average number of bifurcations in the graphical nodes, or their distribution, through non-limiting examples.

[0066] In some embodiments, machine learning-derived methods are used to identify patterns in the aforementioned variations, which may lead to significant clinical endpoints (e.g., fistula stenosis) before the appearance of clinical signs or symptoms that a human nurse can recognize.

[0067] In some embodiments, a system is provided that measures parameters related to fistulas using optical means.

[0068] In some embodiments, structured light is projected onto a patient's body or limbs to image the body. In some embodiments, the structured light may include horizontal and / or vertical stripes of equal or different widths and / or various light patterns other than stripes.

[0069] In some embodiments, structured light imaging is used to provide information about the extent of the fistula by means of some non-limiting examples: the length of the fistula along the long axis of the body; the width of the fistula along the short axis of the body; the shape of the fistula appearing in the image; the shape and / or segments of the fistula circumference; the eccentricity index and / or aspect ratio of the fistula or its segments; and the smoothness and / or roughness index of the fistula profile.

[0070] In some embodiments, structured light patterns are projected onto the patient’s body or limbs and imaged to provide information about the three-dimensional shape of the fistula or body organ (such as an arm or leg).

[0071] In some embodiments, the system can identify changes in the shape of fistulas and / or body organs. In some embodiments, a projector is used to project one or more light patterns (e.g., structured light). In some embodiments, a method measures and / or estimates how patterns on a patient's organ deform to measure the shape of the organ and changes in shape over time.

[0072] In some embodiments, structured light patterns are projected onto the patient’s body or limbs and imaged to provide information about the three-dimensional shape of the fistula, such as one or more of the following: the volume of the entire fistula or fistula segment (e.g., the needle insertion point), the characteristics and variations of curvature, the shape / volume variations of the lower arm / organ portion near the fistula, and three-dimensional surface features such as smoothness and / or roughness.

[0073] In some embodiments, laser speckle interferometry (LSI) is used. In some embodiments, LSI is used to record and observe vibrations on the fistula surface associated with internal blood flow and turbulence. Changes in blood flow and turbulence are often associated with the potential development of stenosis events and clinical conditions.

[0074] In some embodiments, speckle imaging is used to provide information about dynamic effects in the fistula, such as heart pulse and blood flow turbulence, and optionally to generate a spectrogram of fistula vibrations.

[0075] In some embodiments, images of the body are taken at time intervals, and differences between the images may optionally be used to determine differences in fistula shape.

[0076] In some embodiments, these images are taken at intervals of days, weeks, months, or years, and the differences between the images may optionally be used to measure and / or monitor changes in the size or shape of the fistula.

[0077] In some embodiments, these images are taken at intervals of several seconds or minutes, for example with limbs (e.g., a hand held horizontally, then a hand held vertically), and the differences between the images are optionally used to measure and / or monitor whether at least some blood in the fistula can drain from the fistula, the rate of blood drainage, and / or the extent of drainage from the fistula or a specific portion of the fistula, such as the collapse of the needle insertion point.

[0078] In some embodiments, these images are captured at fractions of a second interval, as video clips or movies, and the differences between image frames may optionally be used to measure and / or monitor dynamic parameters associated with the fistula, such as heart pulse, blood flow turbulence, and optionally generate a spectrogram of fistula vibration.

[0079] In some embodiments, changes in dynamic parameters associated with the fistula are analyzed between imaging sessions to monitor changes in the fistula and the patient's condition.

[0080] In some embodiments, performing the above operations in conjunction with near-infrared imaging can collect data related to examinations that nurses and / or physicians need to perform, as well as data that has been clinically proven to have predictive value in identifying stenosis events.

[0081] According to one aspect of some embodiments of the present invention, a system and method are provided for implementing and recording more than one technology or mode, such as one or more structured lights; laser speckle interferometers; image analysis and near-infrared imaging modes, using an imaging device.

[0082] In some embodiments, the system includes a processor and an imaging device, the imaging device including a Digital Light Processing (DLP) projector and a near-infrared camera.

[0083] According to one aspect of some embodiments of the present disclosure, a method for monitoring blood vessels is provided, the method comprising observing, listening to and sensing blood vessel function by imaging a patient’s body to obtain blood vessel geometry using a system for monitoring blood vessel function; imaging the patient’s body using image analysis to obtain the shape of the patient’s body position; and analyzing vibrations of the patient’s body at a position on the patient’s body including blood vessels.

[0084] According to some embodiments of this disclosure, obtaining the shape of a patient's body position includes illuminating the position using structured light.

[0085] According to some embodiments of this disclosure, vibration analysis includes illumination using a laser speckle interferometer.

[0086] According to some embodiments of this disclosure, illumination using a laser speckle interferometer is performed at a location based on the shape of the patient's body position obtained using image analysis.

[0087] According to some embodiments of this disclosure, illumination using a laser speckle interferometer is automatically performed at the location by controlling a digital light processing (DLP) projector.

[0088] According to some embodiments of this disclosure, illumination using a laser speckle interferometer is performed by a physician who guides the laser speckle interferometer illumination to a location on the patient's body.

[0089] According to some embodiments of this disclosure, obtaining the shape of the patient's body position includes calculating the three-dimensional (3D) shape of the patient's fistula.

[0090] According to some embodiments of this disclosure, the drainage rate of a patient's fistula is calculated based on changes in the three-dimensional shape of the fistula.

[0091] According to some embodiments of this disclosure, the elevation test is performed during the test.

[0092] According to some embodiments of this disclosure, one or more parameters related to vascular function are calculated based on image analysis.

[0093] According to some embodiments of this disclosure, an estimate of the probability of vascular function failure is calculated based on one or more parameters.

[0094] According to some embodiments of this disclosure, an estimate of when a blood vessel may fail is calculated based on one or more parameters.

[0095] According to some embodiments of this disclosure, an estimate of the probability of vascular function failure is calculated based on the rate of change of one or more parameters.

[0096] According to some embodiments of this disclosure, an estimate of the maturity of a VA is calculated based on the rate of change of one or more parameters.

[0097] According to some embodiments of this disclosure, calculating one or more parameters includes calculating parameters that indicate the development of one or more collateral vessels.

[0098] According to some embodiments of this disclosure, calculating one or more parameters includes calculating the count of collateral vessels.

[0099] According to some embodiments of this disclosure, collateral veins are automatically detected by counting the number of veins in a specific image region in different images taken at different times.

[0100] According to some embodiments of this disclosure, the location of a vascular access (VA) is automatically detected by detecting the junction of a vein and an artery.

[0101] According to some embodiments of this disclosure, image analysis includes automatically detecting the location of a vascular access (VA) in at least one image.

[0102] According to some embodiments of this disclosure, image analysis includes automatically detecting the location of the fistula in at least one image.

[0103] According to some embodiments of this disclosure, observing the vascular geometry of a patient includes taking images including at least one artery and at least one vein under the patient's skin.

[0104] According to some embodiments of this disclosure, near-infrared wavelengths are used to illuminate the patient's body.

[0105] According to some embodiments of this disclosure, the lighting includes the use of a digital light processing (DLP) projector.

[0106] According to some embodiments of this disclosure, analyzing vibrations in a patient's body includes analyzing the light intensity at a specific location in an image of the patient's body position, including blood vessels.

[0107] According to some embodiments of this disclosure, the spectrum is generated by producing a light intensity vector at a specific location and by converting the intensity vector into a frequency vector to generate a vibration spectrum.

[0108] According to some embodiments of this disclosure, vibration analysis includes analyzing the vibration spectrum within a frequency range. According to some embodiments of this disclosure, the vibration spectrum is within a range corresponding to human audible frequencies. According to some embodiments of this disclosure, the vibration spectrum is within a range corresponding to frequencies below human audible frequencies. According to some embodiments of this disclosure, body vibrations are analyzed in a frequency range less than 1,000 Hz.

[0109] According to some embodiments of this disclosure, vibrations of a patient's body are analyzed by analyzing images captured at a frame rate greater than 150 frames per second (FPS). According to some embodiments of this disclosure, the analysis of vibrations of a patient's body is accomplished by analyzing images captured at a frame rate greater than 500 frames per second (FPS).

[0110] According to some embodiments of this disclosure, vibrations of a patient's body are analyzed by analyzing selected pixels in captured images.

[0111] According to some embodiments of this disclosure, pulse wave parameters are measured by detecting the pulse wave position in two images taken at different times and comparing the pulse wave positions in the two images.

[0112] According to some embodiments of this disclosure, pulse velocity is measured by detecting the pulse position in two images and dividing the distance along the center line of the blood vessel in the two images by the time difference between the two images.

[0113] According to one aspect of some embodiments of the present disclosure, a method is provided that replaces a physical examination performed by a medical professional for monitoring vascular function, the method comprising: generating at least one image of a patient's organ; analyzing the at least one image; and generating parameter values ​​related to vascular function.

[0114] According to some embodiments of this disclosure, it further includes classifying the patient's condition as either suitable for dialysis or at risk of stenosis.

[0115] According to some embodiments of this disclosure, illuminating one or more blood vessels through a patient's skin and analyzing at least one image includes: calculating parameters related to vascular function based on image analysis.

[0116] According to some embodiments of this disclosure, the method described is used to replace doctors in performing observation, listening, and sensory examinations.

[0117] According to some embodiments of this disclosure, the method is performed by a device that does not contact the patient's fistula. According to some embodiments of this disclosure, the method is performed by a device that does not contact the patient's body.

[0118] According to one aspect of some embodiments of the present disclosure, a system for monitoring vascular function is provided, comprising: an illuminator configured to provide laser spots and structured light for a laser speckle interferometer (LSI); a camera configured to image the location illuminated by the illuminator; and a processor for processing images captured by the camera to extract data about shape from the camera images obtained using the structured light, and data about vibration from the camera images obtained using the laser speckle interferometer.

[0119] According to some embodiments of this disclosure, a classifier is further included, configured to classify a patient’s condition as either suitable for dialysis or at risk of stenosis.

[0120] According to some embodiments of this disclosure, the illuminator includes a near-infrared wavelength light source.

[0121] According to some embodiments of this disclosure, the illuminator includes a digital light processing (DLP) projector.

[0122] According to some embodiments of this disclosure, the camera includes a camera capable of capturing images at a frame rate greater than 150 frames per second (FPS).

[0123] According to some embodiments of this disclosure, the camera includes a camera capable of capturing images at a frame rate of more than 500 FPS and a resolution lower than the camera's maximum resolution.

[0124] According to some embodiments of this disclosure, a stent for positioning a patient's limb is further included, wherein the classifier is configured to classify the patient's condition as either suitable for dialysis or at risk of stenosis.

[0125] According to one aspect of some embodiments of the present disclosure, a method for calculating collateral vessel counts is provided, the method comprising: imaging a patient’s body to obtain vascular geometry; and calculating the count of collateral vessels.

[0126] According to some embodiments of this disclosure, calculating the number of collateral vessels includes automatically detecting collateral veins by counting the number of veins in a specific image region in different images taken at different times.

[0127] According to some embodiments of this disclosure, the location of the vascular access (VA) is also automatically detected by detecting the junction of veins and arteries.

[0128] Unless otherwise defined, all technical and / or scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. While similar or equivalent methods and materials to those described herein may be used in the implementation or testing of embodiments of the invention, exemplary methods and / or materials are also described below. In case of conflict, the patent specification, including its definitions, shall prevail. Furthermore, the materials, methods, and examples described are for illustrative purposes only and are not intended to be limiting.

[0129] As those skilled in the art will understand, some embodiments of the present invention can be embodied as systems, methods, or calculator program products. Therefore, some embodiments of the present invention may take the form of entirely hardware embodiments, entirely software embodiments (including firmware, resident software, microcode, etc.), or embodiments combining software and hardware aspects, which are generally referred to herein as “circuits,” “modules,” or “systems.” Furthermore, some embodiments of the present invention may take the form of calculator program products containing calculator-readable program code on one or more calculator-readable media. Implementation of the methods and / or systems of some embodiments of the present invention may involve manually, automatically, or in combination thereof, performing and / or completing selected tasks. Furthermore, actual instruments and devices according to some embodiments of the methods and / or systems of the present invention can implement multiple selected tasks through hardware, software, or firmware and / or combinations thereof (e.g., using an operating system).

[0130] For example, according to some embodiments of the invention, the hardware for performing a selected task can be implemented as a chip or circuit. As software, the selected task according to some embodiments of the invention can be implemented as a plurality of software instructions executed by a calculator using any suitable operating system. In exemplary embodiments of the invention, one or more tasks according to some exemplary embodiments of the methods and / or systems described herein are performed by a data processor, such as a computing platform for executing a plurality of instructions. Optionally, the data processor includes volatile and / or non-volatile memory for storing instructions and / or data, such as a disk and / or removable media. Optionally, a network connection is also provided. A display and / or a user input device such as a keyboard or mouse are also provided.

[0131] Any combination of one or more calculator-readable media can be used in some embodiments of the present invention. A calculator-readable medium can be a calculator-readable signal medium or a calculator-readable storage medium. A calculator-readable storage medium can be, for example, but not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or apparatuses, or any suitable combination thereof. Examples of specific calculator-readable storage media (a non-exhaustive list) will include the following: electrical connections having one or more wires, portable calculator disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable optical disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In the context of this document, a calculator-readable storage medium can be any tangible medium that can contain or store programs used or connected to an instruction execution system, apparatus, or device.

[0132] Calculator-readable signal media may include propagated data signals, such as calculator-readable program code in baseband or as part of a carrier wave. Such propagated signals may take any of a variety of forms, including but not limited to electromagnetic, optical, or any suitable combination thereof. Calculator-readable signal media may be any calculator-readable medium that is not a calculator-readable storage medium but can communicate, propagate, or transmit programs for use by or associated with an instruction execution system, apparatus, or device.

[0133] The program code embodied on the calculator's readable medium and / or the data used therefrom can be transmitted using any suitable medium, including but not limited to wireless, wired, cable, radio frequency, or any suitable combination thereof.

[0134] The calculator program code used to perform operations according to some embodiments of the present invention can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, Smalltalk, C++, etc., and conventional programming languages ​​such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's calculator, partially on the user's calculator as a standalone software package, partially on the user's calculator, partially on a remote calculator, or entirely on a remote calculator or server. In the latter case, the remote calculator can be connected to the user's calculator via any type of network including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external calculator (e.g., via the Internet provided by an Internet service provider).

[0135] Some embodiments of the present invention will now be described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and calculator program products according to embodiments of the present invention. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by calculator program instructions. These calculator program instructions can be provided to the processor of a general-purpose calculator, a special-purpose calculator, or other programmable data processing device to generate a machine that, when executed via the processor of the calculator or other programmable data processing device, creates the functions / actions specified in the apparatus flowchart illustrations and / or block diagrams for implementation.

[0136] These calculator program instructions can also be stored in a calculator-readable medium that can instruct a calculator, other programmable data processing device or other device to operate in a particular manner, such that the instructions stored in the calculator-readable medium produce an article of art including instructions that implement the functions / actions specified in the flowcharts and / or block diagrams.

[0137] Calculator program instructions may also be loaded onto a calculator, other programmable data processing device or other device to cause a series of operational steps to be performed on the calculator, other programmable device or other device, thereby producing a process implemented by the calculator, such that the instructions executed on the calculator or other programmable device provide a process for implementing the functions / actions specified in the flowchart and / or block diagram.

[0138] Some of the methods described in this article are typically designed for use with calculators only and may be infeasible or impractical for human experts to perform purely manually. Human experts who wish to perform similar tasks manually might use entirely different approaches, such as leveraging expert knowledge and / or the pattern recognition capabilities of the human brain, which would be more efficient than manually executing the steps of the methods described in this article. Attached Figure Description

[0139] Some embodiments of the invention are described herein by way of example only with reference to the accompanying drawings and images. Detailed reference is now made to the accompanying drawings, whereby the details shown are exemplary and for purposes of illustrative discussion of embodiments of the invention. In this regard, it will become apparent to those skilled in the art from the description taken in conjunction with the drawings how embodiments of the invention can be practiced.

[0140] In the attached diagram:

[0141] Figure 1 This is a graph showing that the probability of vascular pathway thrombosis within 3 months depends on flow velocity and flow velocity changes, as reported by Besarab et al., “Pathway monitoring is worthwhile and valuable,” Blood Purification, February 2006.

[0142] Figure 2 This is a simplified illustration of a system for measuring blood vessels according to an exemplary embodiment of the present invention;

[0143] Figure 3 This is a simplified block diagram of a system for measuring blood vessels according to an exemplary embodiment of the present invention;

[0144] Figures 4A-4E This is a simplified flowchart of the algorithm according to an exemplary embodiment of the present invention;

[0145] Figure 5A and 5B It is a simplified diagram of a pulse wave traveling along a vein;

[0146] Figure 6 This is a simplified flowchart of a classifier method according to an exemplary embodiment of the present invention;

[0147] Figure 7 This is a simplified block diagram of a system for measuring blood vessels according to an exemplary embodiment of the present invention;

[0148] Figure 8A and 8B These are images of optical components in a system constructed according to exemplary embodiments of the present invention;

[0149] Figure 9 This is a simplified flowchart of the segmentation method according to an exemplary embodiment of the present invention;

[0150] Figure 10 This is a simplified flowchart of the registration method according to an exemplary embodiment of the present invention;

[0151] Figure 11This is a simplified flowchart illustration of a method according to an exemplary embodiment of the present invention;

[0152] Figure 12 This is a simplified flowchart illustration of a method according to an exemplary embodiment of the present invention;

[0153] Figure 13 This is a simplified flowchart illustration of a method according to an exemplary embodiment of the present invention;

[0154] Figure 14 This is a simplified flowchart of a classifier method according to an exemplary embodiment of the present invention;

[0155] Figure 15A -C shows three different images of the same patient's arm according to an exemplary embodiment of the present invention;

[0156] Figure 16A It is a procedure for medical personnel to examine patients with vascular stenosis or thrombosis;

[0157] Figure 16B This is a simplified flowchart illustrating a method for examining a patient according to an exemplary embodiment of the present invention;

[0158] Figure 17 This is a simplified block diagram of a method for examining a patient according to an exemplary embodiment of the present invention;

[0159] Figure 18A and 18B These are images of the patient's fistula taken at two different times;

[0160] Figure 19A and 19B This is a simplified diagram of a system for monitoring vascular access (VA) and / or fistulas according to two exemplary embodiments of the present invention;

[0161] Figure 20 Images are of a system for monitoring vascular access (VA) and / or fistulas according to exemplary embodiments of the present invention;

[0162] Figure 21A -C shows three different images of the same patient's arm.

[0163] Figure 22A It is a graph that displays the measured vibration power spectrum by analyzing images produced by laser speckle imaging;

[0164] Figure 22B This is a simplified flowchart of a method for converting data from an image stream to a spectrum according to an exemplary embodiment of the present invention;

[0165] Figure 23This is a simplified flowchart illustrating a method for monitoring vascular function according to an exemplary embodiment of the present invention; and

[0166] Figure 24 This is a simplified flowchart of a method for replacing a physical examination performed by medical personnel to monitor vascular function, according to an exemplary embodiment of the present invention. Detailed Implementation

[0167] This invention generally relates to the field of monitoring blood vessels in patients. Some aspects more specifically relate to the early diagnosis of vascular dysfunction, and even more specifically to the early detection of vascular access failure in patients undergoing hemodialysis. Some aspects more specifically relate to the measurement of fistulas.

[0168] introduce

[0169] Monitoring through physical examinations is cost-effective and an effective method for detecting vitamin A (VA) abnormalities. Unfortunately, the availability and access to information among nephrology and hemodialysis staff are often limited. Therefore, routine physical examinations for VA are not typically performed in hemodialysis units.

[0170] Furthermore, due to the complexity of the graft membrane (VA), monitoring strategies have failed to consistently detect stenosis in different scenarios. Although low VA flow is associated with an increased risk of thrombosis, this association is not accurate enough in predicting thrombosis. In contrast, VA flow and dynamic or static pressure monitoring have been found to be inaccurate in predicting graft thrombosis rather than preventing many unnecessary invasive procedures. Additionally, PTA causes mechanical trauma, accompanied by intimal hyperplasia (NIH), increased risk of stenosis, and impaired VA survival.

[0171] During and between dialysis sessions, and more importantly, the patient's flow and pressure fluctuate. This makes each measurement a potentially inaccurate predictor of stenosis, as well as thrombosis, and potentially overlooked, as well as developing lesions.

[0172] If the outflow is the sole source of stenosis, a hyperbolic relationship between flow and pathway pressure is expected to appear within a given pathway. Unfortunately, this is not the case. Lesions do occur in the inflow and within the pathway itself, with, on average, nearly two lesion / pathway locations in a typical pathway at referral. Lesions alter the relationship between flow and pressure. Due to these confounding factors, such as anatomical factors and the location of the stenosis, there is little correlation (if any) between single measurements of flow and pressure.

[0173] For reference Figure 1This study showed that the probability of vascular access thrombosis within 3 months depends on flow velocity and its variation, as reported by Besarab et al., “Access monitoring is worthwhile and valuable,” *Blood Purification*, February 2006.

[0174] Figure 1 The charts include a Y-axis 101 showing the probability of vascular access thrombosis during a 3-month period, various lines 103 showing flow rates in milliliters per minute, and an X-axis 102 showing monthly flow rate changes in milliliters per minute.

[0175] Figure 1 It was shown that the probability of vascular access thrombosis within 3 months depends not only on the absolute flow rate at any given time, but also on the rate of change in flow rate (if the flow rate changes) (Besarab et al., “Access monitoring is worthwhile and valuable”, Blood Purification, February 2006).

[0176] The pathway with an initial flow rate of 600 ml / min and a monthly flow rate reduction of 20 ml / min had a lower probability of thrombosis (22%) compared to the pathway with an initial flow rate of 1200 ml / min and a monthly flow rate reduction of 100 ml / min, even though the absolute flow rate of the former (540 ml / min) was lower than that of the latter (900 ml / min) at the start of the observation period.

[0177] Therefore, there is a need for a monitoring solution that can detect developing stenosis early and predict thrombosis, overcoming at least some of the shortcomings of existing monitoring practices:

[0178] As the guidelines state, dialysis centers have poor compliance with routine VA body examinations;

[0179] Inherent inaccuracies associated with a single physical examination or VA pressure / flow measurement;

[0180] For example, the inherent inaccuracy of a single parameter such as flow rate or pressure;

[0181] The results of regular measurements may be affected by unrelated hemodynamic events; and

[0182] Measurements taken by different human caregivers may introduce inconsistencies.

[0183] Overview

[0184] One aspect of some embodiments of the present invention relates to replacing or adding to physical examinations performed by medical personnel / nurses.

[0185] When a nurse or doctor examines a patient's blood vessels, they typically use a three-step procedure: observation, listening, and feeling.

[0186] One aspect of some embodiments involves performing observation, listening, and sensing through instrumental measurements and computerized analysis.

[0187] In some embodiments, the system described herein performs observation, listening, and sensing based on illumination and imaging of a patient's limb, and analysis of data collected from the imaging. In some embodiments, the system teaches how to predict fistula conditions and may be able to prevent failure early.

[0188] In some embodiments, the system described herein performs observation, listening, and sensing based on illumination and imaging of a patient's limb, and analysis of data collected from the imaging. In some embodiments, the system teaches how to predict fistula conditions and may be able to prevent failure early.

[0189] In some embodiments, blood flow is measured non-invasively based on image processing of human vascular images. Physiological parameters known to affect vascular access (VA) are measured, and these measurements may optionally be used to determine whether a patient should be scheduled for corrective surgery or continue dialysis.

[0190] As described herein, one aspect of some embodiments relates to sensing through instrumentation and computerized analysis.

[0191] In some embodiments, the listening, as described herein, is performed by an instrument, optionally the same instrument.

[0192] In some embodiments, the observations described herein are performed by an instrument, optionally the same instrument.

[0193] One aspect of some embodiments of the present invention relates to the automatic detection and / or monitoring of AV fistulas in vascular images.

[0194] In some embodiments, images of blood vessels are analyzed, and the location where an artery connects to a vein is optionally determined as the location of an AV fistula.

[0195] In some embodiments, images of blood vessels are analyzed, and the location where an artery appears to connect to a vein is optionally identified as the location of an AV fistula.

[0196] In some embodiments, images of blood vessels are analyzed, and optionally, AV fistulas are measured to estimate geometric characteristics.

[0197] One aspect of some embodiments of the present invention relates to the automatic, non-invasive measurement of blood flow-related parameters.

[0198] In some embodiments, non-invasive measurements include imaging blood vessels through the skin using reflected and / or transmitted light.

[0199] In some embodiments, the probability of vascular access failure may be estimated. In some embodiments, the estimation is based on one or more measured parameters.

[0200] In some embodiments, the probability of occlusion formation may be estimated. In some embodiments, the estimation is based on one or more measured parameters.

[0201] In some embodiments, the probability of thrombosis may be estimated. In some embodiments, the estimation is based on one or more measured parameters.

[0202] In some embodiments, the severity of stenosis may be estimated. In some embodiments, the estimation is based on one or more measured parameters.

[0203] In some embodiments, the narrowing formation rate may be estimated. In some embodiments, the estimation is based on one or more measured parameters.

[0204] In some embodiments, the maturity level of VA is optionally estimated. In some embodiments, the estimation is based on one or more measured parameters.

[0205] In some implementations, the rate of VA maturation is optionally estimated. In some embodiments, the estimation is based on one or more measured parameters.

[0206] One aspect of some embodiments of the present invention relates to providing a visual report to caregivers.

[0207] The following are listed properties, one or more of which relate to some embodiments of the present invention:

[0208] 1. One or more patient-related parameters, including images, can be easily measured and / or integrated into routine dialysis appointments in a cost-effective and / or non-invasive (optionally, non-contact) manner.

[0209] 2. The input to the algorithm described herein may optionally include one or more patient-related parameters to estimate the probability of vascular access failure, wherein each parameter may be used on a single measurement basis or as multiple measurements over time.

[0210] 3. Some patient-related parameters are obtained using objective measurements, which may not require a high level of expertise from the user (e.g., the patient and / or healthcare professional).

[0211] 4. Some patient-related parameters may optionally be taken from the patient's specific medical records, including elements such as demographic data (e.g., age, sex, weight, and height), laboratory test results, imaging test results (e.g., X-ray, magnetic resonance imaging "MRI"), and physical examination results. Those skilled in the art will appreciate that parameters can be extracted in various ways, such as – directly inputting examination results into a keyboard connected to the system described herein, computer processes retrieving electronic medical records using a specific patient identification number (ID), speech-to-text conversion, speech recognition algorithms applied to verbal analysis by staff, and optical character recognition (OCR) for printed / written documents.

[0212] Measurement of VA maturity: Compared to veins and arteries, VAs have a unique tissue structure. This structure changes during VA maturation and stenosis.

[0213] Structural changes affect the mechanical and optical properties of the VA, therefore, in some embodiments, the changes can be monitored by one or more of the following:

[0214] • Imaging: By way of non-limiting example, by measuring changes in the contrast or intensity of reflected and / or transmitted light;

[0215] • Non-imaging: The intensity of reflected or transmitted light;

[0216] • Measurement of scattering and absorption coefficients (e.g., two-distance steadystate photon migration measurement).

[0217] For example, in some embodiments, the system is configured to detect veins, and monitoring VA during the maturation period may alter the detection results. In examples of optical sensing, the VA's response to light (one or more of transmission, reflection, absorption, and scattering) may change during the maturation period. Monitoring maturation may improve the success rate of VA maturation by suggesting timely, proactive corrections. Measurements of vascular layers, or changes in the ratio between vascular layers or the absolute value of layers during maturation or stenosis.

[0218] By using one or more parameters derived from non-invasive measurements, the accuracy of estimating maturity (maturity level, stage, rate, completion) or the probability of vascular access failure, occlusion, and thrombosis can be improved. The parameters used can be directly measured or are the result of preprocessing applied to the measurement. This preprocessing can involve the application of various algorithms, as well as the combination of several parameters and the utilization of multiple measurements over time.

[0219] In some embodiments of the present invention, examples of metrics or phenomena that may be optionally extracted and used include:

[0220] 1. Pulse velocity - In some embodiments, the reflection or absorption of light radiation from at least two points in an image frame is detected. In some embodiments, changes in impedance are measured by electrodes placed between and / or along the two points, or along a blood vessel or tissue region. Optionally, these two points include portions known to be prone to stenosis. More generally, the pulse waveform (e.g., pulse amplitude, full-width half-maximum, FWHM) is measured using at least one point.

[0221] In some embodiments, pulse wave amplitude is optionally measured. An optional method for measuring pulse wave amplitude includes a first measurement measuring a region along a vein identified as widening due to the pulse wave. A second measurement also measures the same region in different images when the pulse wave is not at that location. The difference between the first and second measurements is optionally correlated with the pulse wave amplitude. In some embodiments, pulse wave amplitude is considered a characteristic corresponding to the mechanical properties of the vein and / or the maturity of the AV through which the pulse wave travels.

[0222] In some embodiments, pulse wave analysis (PWA) may optionally be performed to estimate changes associated with vascular stiffness, which are associated with additional risk factors such as cardiovascular disease or atherosclerosis, which in turn may affect the activity of the VA over time. In some embodiments, pulse quality may optionally be scored, and the analysis may optionally include changes over time and changes between different segments.

[0223] 2. The presence and development of collateral veins and their characteristics, such as density, size, distance from the VA, orientation, and filling, are assessed using image processing and / or other detection methods, such as measuring contrast—by absorbing visible or near-infrared wavelengths; or emitting at far-infrared wavelengths. Other measurement options include measuring changes in the absorption of visible and near-infrared light and the emission of far-infrared light. Another optional method for measuring collateral development is to optionally measure temperature changes around the VA. In some embodiments, the detection of the presence and development of collateral veins may optionally use reference images or measurements obtained from previous examinations. In some embodiments, trend analysis of the rate of collateral vein development may optionally use recurring examinations. Examinations may optionally be performed daily, per dialysis session, weekly, bi-weekly, or monthly.

[0224] In some embodiments, collateral veins are detected by comparing a new image with a previous image and counting the number of veins—optionally, an increase in the number of veins indicates that a new vein is a collateral vein.

[0225] In some embodiments, the presence and / or development of collateral veins are detected by extracting features from an image or measurement.

[0226] Basic principle: The detection of collateral vessels may indicate the presence of a flow-limiting lesion (significant hemodynamics). Collateral vessels may develop and dilate, relieving increased pressure within the pathway in cases of outflow tract narrowing.

[0227] 3. Minimum diameter of blood vessel obtained through image processing (narrow location).

[0228] 4. Narrowing points can be detected by estimating mechanical reflection waves or local pressure / flow changes, for example, by measuring changes in electrical impedance.

[0229] 5. Maximum diameter of the blood vessel obtained through image processing (appearance and size of the aneurysm).

[0230] 6. Detect vascular collapse when the arm or leg is raised.

[0231] 7. Use near-infrared (NIR) (700-1000nm) reflectance and / or transferred spectroscopy to measure the amount of oxygenated and deoxygenated hemoglobin (Hb).

[0232] 8. Spectroscopic analysis of oxygenated and deoxygenated hemoglobin (Hb).

[0233] 9. VA (noise) audible sound.

[0234] 10. Palpation of the VA (tremor).

[0235] 11. Analyze the change in electrical impedance at VA using signal processing methods known in the art.

[0236] 12. In some embodiments where the same parameter is measured multiple times over time, the measurements may be synchronized according to the detected respiratory cycle and categorized against the detection algorithm relative to their time with respect to the respiratory cycle. This synchronization and categorization has potential benefits, for example, in estimating changes in the oxygen mixture over time, but may also improve the accuracy of other measurements, such as pulse velocity.

[0237] The output of the system described herein may be an audible alarm, a visual alarm, an image, an image sequence, or a video providing rapid and accurate guidance for invasive procedures to healthcare professionals (e.g., recommending the optimal invasive location). The system may recommend treatment (PTA, non-invasive, thrombectomy) for the patient. Optionally, the recommendation is based on information collected by the system.

[0238] According to one aspect of some embodiments of the invention, during testing, the system's output may optionally be analyzed and / or the system may optionally be used to guide the patient through the test to ensure proper performance of the test. As a non-limiting example, in a limb elevation test—verifying that the limb elevation / position is correct. In some embodiments, an alert may also be issued to a nurse / technician if the patient does not perform the test correctly or requires assistance.

[0239] In some embodiments, the system output can optionally be used to support remote physical examinations performed by patients, while the system provides feedback and / or alerts to remote support personnel, such as nurses or technicians who require additional guidance, regarding the correct execution of the examination.

[0240] In some embodiments, the system output may optionally be provided to different data consumers in different ways. For example, a general interpretation of the likelihood of clinically significant stenosis formation may optionally be provided to dialysis nurses, and alerts with annotated images may optionally be provided to interventional radiologists, and / or reports highlighting parameters such as the location, severity, and speed of stenosis formation may optionally be provided to interventional radiologists.

[0241] According to one aspect of some embodiments of the present invention, a system and method for measuring parameters related to fistulas are provided.

[0242] In some embodiments, a system is provided that includes optical devices for acquiring one or more images of fistulas in the same patient during monitoring.

[0243] In some embodiments, one or more measurements and / or features may be optionally extracted from the image—and optionally, their changes over time may be monitored. In some embodiments, features are timeline derivatives of parameters measured or estimated in the image; as a non-limiting example, the number, branching, and size of collateral veins may change over a period of time, such as days / weeks / months.

[0244] In some embodiments, machine learning-derived methods are used to identify patterns in the aforementioned variations that may potentially lead to significant clinical endpoints (e.g., fistula stenosis) before the appearance of clinical signs or symptoms that can be recognized by a human nurse.

[0245] In some embodiments, a system is provided that measures parameters related to fistulas using optical means.

[0246] In some embodiments, structured light is projected onto the patient's body or limbs, and the body is imaged. In some embodiments, the structured light may include horizontal and / or vertical stripes of the same or different widths and / or various light patterns beyond the stripes.

[0247] In some embodiments, structured light imaging is used to provide information about the extent of the fistula, such as the length of the fistula along the long axis of the body, the width of the fistula along the short axis of the body, the shape of the fistula appearing in the image, the segmentation of the fistula circumference, the eccentricity and / or aspect ratio of each segment, and the smoothness and / or roughness of the fistula profile.

[0248] In some embodiments, structured light patterns are projected onto the patient’s body or limbs, and the body is imaged to provide information about the three-dimensional shape of the fistula or organ.

[0249] In some embodiments, the system identifies shape changes in fistulas and / or organs near fistulas. In some embodiments, a projector is used to project one or more light patterns (e.g., structured light). In some embodiments, a method measures and / or estimates how patterns on a patient's organ deform to measure the shape of the organ and changes in shape over time.

[0250] In some embodiments, a structured light pattern is projected onto the patient’s body or limb, and the body is imaged to provide information about the three-dimensional shape of the fistula in a manner that is not limited to certain examples, such as the volume of the entire fistula or a segment of the fistula (e.g., the point of needle insertion); the characteristics and / or variations of curvature; the shape and / or volume variations of the arm / organ portion below the fistula; and three-dimensional surface features such as smoothness and / or roughness.

[0251] In some embodiments, a laser speckle interferometer is used. In some embodiments, the laser speckle interferometer is used to record and observe vibrations on the fistula surface associated with internal blood flow and turbulence. Changes in blood flow and turbulence are often associated with the potential development of stenosis events and clinical conditions.

[0252] In some embodiments, speckle imaging is used to provide information about dynamic effects in the fistula, such as heart rate and blood flow turbulence, and optionally to generate a spectrogram of fistula vibrations.

[0253] In some embodiments, images of the body are taken at intervals, and the differences between the images may optionally be used to determine differences in the shape of the fistula.

[0254] In some embodiments, the images are taken at intervals of days, weeks, months, or years, and the differences between the images may optionally be used to measure and / or monitor changes in the size or shape of the fistula.

[0255] In some embodiments, images are taken at intervals of several seconds or minutes, for example with a limb (e.g., a hand held horizontally, then a hand held vertically), and the differences between the images may optionally be used to measure and / or monitor one or more of the following: whether at least a portion of the blood within the fistula is draining from the fistula; the rate of blood drainage; the extent of blood drainage from the fistula and / or specific portions of the fistula; and the collapse of one or more needle insertion points.

[0256] In some embodiments, the images are spaced a fraction of a second apart, as a video clip or movie, and the differences between the image frames may optionally be used to measure and / or monitor dynamic parameters associated with the fistula, such as heart pulse, blood flow turbulence, and optionally generate a spectrogram of fistula vibration.

[0257] In some embodiments, the spectrum may be generated by selecting one or more pixels in an image frame, showing large or even maximum changes in intensity over time. In some embodiments, the number of pixels selected may be in the range of 1 to 100 pixels. In some embodiments, the pixel intensity values ​​of this or these pixels are used to calculate a function of light intensity changing over time. In some embodiments, the spectrum of light intensity may be generated by transforming the time domain to the frequency domain, for example, by a Fast Fourier Transform (FFT).

[0258] In some embodiments, changes in dynamic parameters related to the fistula between imaging sessions are analyzed to monitor changes in the fistula and the patient's condition.

[0259] In some embodiments, performing the above operations in conjunction with near-infrared imaging can collect data related to examinations that nurses and / or physicians need to perform, as well as data that has been clinically proven to have predictive value in identifying stenosis events.

[0260] According to one aspect of some embodiments of the present invention, a system and method are provided for implementing and recording more than one technique or modality (e.g., one or more structured light; laser speckle interferometer; image analysis and near-infrared imaging modalities) and using an imaging device.

[0261] In some embodiments, the system includes a processor and an imaging device, the imaging device including a digital light processing projector and a near-infrared camera.

[0262] According to one aspect of some embodiments of the present invention, a system and method are provided for analyzing the vibrations of light reflected from a patient's body.

[0263] In some embodiments, the heartbeat is monitored.

[0264] In some embodiments, the analysis of vibration patterns caused by flow through or near the fistula can selectively detect complete or partial blockage of the inflow or outflow path.

[0265] In some embodiments, vibration patterns caused by flow through or near a fistula are analyzed while local pressure is applied to the inflow or outflow path, selectively detecting complete or partial blockage of the inflow or outflow path.

[0266] In some embodiments, the analysis of vibrations may optionally detect the initiation of flow through a fistula associated with normal cardiac activity (the diastolic or systolic phase of the heart cycle).

[0267] In some embodiments, vibration analysis can optionally detect the initiation of flow through a fistula associated with a sudden release (partial or complete collapse or expansion of the fistula).

[0268] In some embodiments, the analysis of vibrations may optionally detect the time period during which blood flows into the fistula, followed by the sudden opening of the obstruction that allows blood to flow out of the fistula. This opening may occur during the high-pressure period of cardiac contraction. In some cases, the sudden opening is referred to as hammering. In some embodiments, hammering is detected by measuring the amplitude of the vibrations, optionally relative to amplitudes at other times, such as other times during a heartbeat.

[0269] In some embodiments, vibrations associated with flow initiation are analyzed. For any or all types of initiation, the parameter value or variation of the parameter value or the variation of the characteristic parameter value, or the variance of the parameter value, may be measured. The parameter may be one or more of the following: intensity, energy, steepness (derivative of the value), relaxation time, temporal width, duty cycle, spectral content, spectral width, or any combination thereof.

[0270] In some embodiments, the analysis of vibrations associated with the onset of flow may optionally measure parameter values ​​related to the time delay or phase delay between the onset associated with sudden release and the onset associated with normal cardiac activity.

[0271] In some embodiments, the analysis of vibrations associated with the initiation of flow may optionally measure parameter values ​​related to regularity or self-similarity of a series of starting points from the same source.

[0272] Before explaining at least one embodiment of the invention in detail, it should be understood that the invention is not necessarily limited in its application to the construction and arrangement of the components and / or methods set forth in the following description and / or in the drawings and / or examples. The invention can have other embodiments or can be practiced or performed in various ways.

[0273] For reference Figure 2 This is a simplified illustration of a system for measuring blood vessels according to an exemplary embodiment of the present invention.

[0274] Figure 2 The top-level configuration of an exemplary system 200 for measuring blood vessels is shown.

[0275] In some embodiments, the system 200 may include at least one illumination source 202 and at least one detector 204, such as a camera.

[0276] In some embodiments, the system 200 may further include a control unit 206 that may optionally activate the illumination source 202 and the camera 204, and an optional processor 208 that may optionally receive and analyze images generated by the camera 202.

[0277] In some embodiments, the generated images and / or data generated after analyzing the images can be displayed on an optional display 210 coupled to the processor 208 via a wireless or wired connection.

[0278] In some embodiments, the processor 208 and the display 210 may be implemented in a single device, such as a laptop, tablet, or smartphone. In some embodiments, a scanning system may be applied that optionally moves the detection unit (automatically or manually) and optionally scans organs at more than one point. Figure 2 A system 200 applied to arm 212 is described.

[0279] The system and method can be implemented with other components without limitation.

[0280] Now for reference Figure 3 This is a simplified block diagram of a system for measuring blood vessels according to an exemplary embodiment of the present invention.

[0281] Figure 3 The top-level block diagram of the example system is described.

[0282] In some embodiments, the system may include at least two main units: a detection unit 302 and a software unit 306.

[0283] The system may include additional units, such as a workstation 304, optional cloud infrastructure 308, etc.

[0284] In some embodiments, the software unit 306 includes at least two sub-units: an embedded unit 330 and an algorithm unit 334. The software unit 306 may include additional blocks, such as a graphical user interface (GUI) unit 332, etc.

[0285] Detection unit

[0286] In some embodiments, the detection unit 302 may optionally use:

[0287] 1. Visual / optical inspection to acquire images containing information for further analysis.

[0288] 2. Speckle Imaging—When an object is illuminated by a laser, the backscattered light forms an interference pattern consisting of dark and bright areas. This pattern is called a speckle pattern. If the illuminated object is stationary, the speckle pattern is static. When the object moves, such as red blood cells in tissue, the speckle pattern changes over time. Speckle images contain information related to changes in blood vessels, which can optionally be analyzed and extracted through image processing.

[0289] 3. Dark Field / Side Lighting —

[0290] a. The specular reflection did not reach the camera.

[0291] b. The camera only captured diffused light.

[0292] c. Reduce surface reflection

[0293] d. Contrast data varies with the angle between the light source and the detector.

[0294] 4. Transmitted illumination—illuminating the back of a sample. The sample is placed between the illumination source and the sensor device. Transmitted illumination potentially improves image contrast and / or potentially increases the depth at which blood vessels can be imaged.

[0295] 5. Photoacoustic imaging potentially enhances the contrast between different media due to the differences in their optical properties. By averaging the refractive index gradient within the tissue components, photoacoustic imaging potentially reduces scattering within the tissue, potentially leading to a greater depth of light penetration.

[0296] In some embodiments, the detection unit 302 may optionally include one or more of the following components:

[0297] 1. One or more detectors / sensors / cameras 310 (e.g., CCD or CMOS, gallium indium arsenide sensor, microbolometer) sensitive to one or more visible, near-infrared, or short-wave infrared (SWIR) light sources. In some embodiments, the sensor's frame rate can vary from a single frame to a high frame rate. The sensor frame rate may optionally be in the range of, for example, 5, 10, 16, 24, 30, 50, 60, 100, 165, 200, or even up to 300 frames per second (fps).

[0298] 2. One or more lenses 312 (zoom or fixed focal length) and / or filters.

[0299] 3. One or more illuminators 314 or emitters (e.g., coherent or incoherent, narrow-spectrum or broadband, UV, visible, SWIR, far-IR, NIR illumination sources—e.g., NIR-led or green (532nm) lasers). The emitters may be coaxial with respect to detectors 310 and VA or at different angles.

[0300] The operation mode can be still image or video.

[0301] 4. One or more polarizing filters (elliptical and / or linear).

[0302] 5. One or more optical bandpass filters.

[0303] 6. The detection unit may optionally include a scanning system or a moving strip scanner.

[0304] In some embodiments, the detection unit 302 may optionally use an audio / sound detection sensor 316 in place of or as a supplement to visual / optical detection, and the detection unit 302 may optionally include one or more audio sensors.

[0305] In some embodiments, the detection unit 302 may include a vital signs sensor.

[0306] Software unit

[0307] In some embodiments, the software unit 306 may include one or more of the following components:

[0308] 1. GUI - Graphical user interface / application 332, used for one or more of the following: operating test procedures, displaying images and / or results and / or inserting or importing patient clinical information.

[0309] 2. Embedded unit 330 - used to control detection unit 302.

[0310] 3. Algorithm Unit 334 — The algorithm unit may optionally include an algorithm or a software module, used for:

[0311] Image processing;

[0312] Machine learning (ML);

[0313] In some embodiments, the input to the machine learning algorithm may optionally be the images and / or data captured by the detection unit 302.

[0314] In some embodiments, the input may also include the patient’s clinical information and / or vital signs.

[0315] In some embodiments, the workstation 304 may optionally include a computer, a screen, a keyboard, one or more rotary controls, a mechanical interface for the imaging unit, and a power supply or a power interface. In some embodiments, the workstation 304 may also include an "organ fixation surface".

[0316] In some embodiments, the workstation 304 may optionally include one or more of the following:

[0317] A control unit 320 is used to control the operation of the detection unit 302 and / or one or more components of the detection unit 302;

[0318] One computer 320;

[0319] One monitor has 324;

[0320] An optional organ fixation surface or device 326 for optionally positioning the organ at a specific location relative to the illumination 314 and / or the detector 310; and

[0321] A stent 328 is used to place components of the system at a specific location relative to the patient's organ.

[0322] In some embodiments, the cloud infrastructure 308 may optionally include one or more of the following cloud services:

[0323] One storage (database) server 340;

[0324] One Web application server 336;

[0325] A computational service for machine learning, such as improving algorithms based on new data; and / or for analytics to provide a user with functionalities and metrics; and / or insights to provide metrics relevant to the current or predicted future clinical condition of the VA.

[0326] A machine learning algorithm—which may be supervised or unsupervised—learns from a database of images and / or patient parameters generated by embodiments of the present invention, and / or metadata, such as the patient’s disease, vital signs, parameters from the dialysis machine, and / or other data available in the electronic medical records, optionally including the patient’s previous interventional treatments, other risk factors, comorbidities, etc.

[0327] These steps include one or more of the following:

[0328] 1. Extract features from images;

[0329] 2. Calculation of feature trends;

[0330] 3. Run machine learning on the feature vector and / or feature vector trends.

[0331] In some embodiments, the result of machine learning is a statistical classifier model that distinguishes between less than or greater than 50% AV fluency.

[0332] In some embodiments, Analytics and Insights runs on metadata and patient records and calculates statistics on AV failures (metadata and health records) based on patient data.

[0333] In some embodiments, the clinic’s performance may be optionally analyzed, such as the number of narrow events per year.

[0334] Now for reference Figures 4A-4E This is a simplified flowchart illustration of the algorithm according to an exemplary embodiment of the present invention.

[0335] Figures 4A-4E A flowchart describing exemplary algorithms is shown, which can be implemented in the system's software unit 306 or cloud unit 308 by way of non-limiting example.

[0336] Figure 4A A program flow diagram is shown. Figure 4A An example illustrates the procedure for vascular access.

[0337] First, the body's organ (e.g., an arm) is placed within a fixed sleeve (402) below the detection unit. In some embodiments, the organ is an arm or leg, and all measurements are performed when the organ is substantially perpendicular to the ground (pointing upwards or downwards). In some embodiments, some measurements are performed when the organ is substantially parallel to the ground, while others are performed when the organ is perpendicular to the ground (pointing upwards or downwards). In some embodiments, some measurements are performed when the organ is below the patient's heart, while others are performed when the organ is above the patient's heart.

[0338] Next, a region of interest (ROI) is detected (404). In some embodiments, the ROI is the vascular access body and / or the area surrounding it. This detection can be performed automatically by the system or manually by a doctor / user.

[0339] The next step is to perform one or more measurements (406), such as an image of the region of interest.

[0340] The image is processed by a processing algorithm (408), such as an image processing algorithm, and then optionally saved to a database (410).

[0341] The next step is to extract features (414) from the current inspection measurement, such as an image, and from previous inspection measurements (412), such as a picture.

[0342] The features are sent to a statistical model that can classify the vascular access body between “early detection failure” (418) and “stable” state (420) (416).

[0343] Figure 4B An exemplary algorithm flow for extracting features of the "pulse velocity" phenomenon is shown.

[0344] The first step is preprocessing (422), for example, to detect the image scale, for example in mm.

[0345] The second step is to subtract the first image (424) from the second image. The result includes two bright spots.

[0346] The next step is to detect bright spot centers (426) and calculate the distance between bright spot centers along a path of the blood vessel (428).

[0347] The next step is to divide the calculated path by the time interval (430) between the two images to produce the pulse velocity result.

[0348] Figure 4C An exemplary algorithm flow for extracting features of collateral vein phenomena is shown.

[0349] The first step is preprocessing (434), for example, to detect the image scale, for example in mm.

[0350] The second step is to detect the vascular pathways and / or branches (436).

[0351] The next step is to calculate the length of each branch and its distance from the fistula along the venous route (438).

[0352] The methods described above are also described in more detail below, under the heading "Algorithm for Detecting Vascular Pathways (VA)," and in several paragraphs.

[0353] Further steps include calculating parameters (442) describing the collateral vein phenomenon, including one or more parameters, such as:

[0354] 1. Number of branches.

[0355] 2. The length of the main branch.

[0356] 3. The centroid of the branch.

[0357] 4. Distance from the centroid of the fistula.

[0358] 5. Ellipse blocks.

[0359] 6. Branch diameter, optionally detecting the blood filling the vascular lumen.

[0360] Figure 4D An exemplary algorithm flow for extracting features from aneurysms and stenosis is shown.

[0361] The first step is preprocessing (446), such as detecting the image scale, for example in mm.

[0362] The next step is to detect venous and / or arterial routes (448).

[0363] The next step is to divide the venous and / or arterial routes (450).

[0364] The next step is to find and calculate the narrowest and widest widths along the vein and / or artery route (452).

[0365] Figure 4E An exemplary algorithm flow for extracting features from an arm elevation examination is shown. It should be understood that the algorithm flow is applicable to organs of other subjects and is not limited to the arm.

[0366] In some embodiments, two images are acquired after the arm is raised to track changes in outflow, which translates to changes in fistula volume over a short period. Under normal outflow: the fistula contents are emptied “rapidly” (within seconds), and shape / area differences between the two images are detected and / or measured. Under obstructed outflow: the fistula contents are not emptied quickly enough, and smaller changes (if any) in the shape / area of ​​the fistula are detected / measured. Tracking these changes over time allows for the tracking of changes in fistula patency.

[0367] In some embodiments, the arm-raising test may be applied when the arm is raised (up or down, perpendicular to the ground, and / or above heart level).

[0368] In some embodiments, the first step is preprocessing (456), for example, to detect the image scale, for example in mm.

[0369] The next step is to detect vascular pathways (fistulas) in the images (458).

[0370] The next step is to segment the vascular pathway (fistula) in the image (460), optionally segmenting the fistula from other parts of the image.

[0371] The next step is to calculate the area of ​​the vascular pathway in the image (462).

[0372] In some embodiments, the arm-raising test first takes a first image when the arm is parallel to the ground, and then takes a second image when the arm is perpendicular to the ground (up or down) and above heart level.

[0373] After obtaining the first image and the second image, the next step is preprocessing, for example, to detect the image ratio in the two images, for example, in mm.

[0374] The next step is to detect vascular pathways in the two images.

[0375] The next step is to calculate the area of ​​the vascular pathways in the two images.

[0376] The next step is to subtract the area of ​​the first vascular pathway from the area of ​​the second vascular pathway.

[0377] In some embodiments, the arm-raising test first moves the arm (or any other main organ) from a first position (where the arm is roughly parallel to the ground) to a second position (where the arm is roughly perpendicular to the ground, pointing upwards or downwards).

[0378] The next step is to take two images of the raised arms.

[0379] The next step is preprocessing, such as detecting the image ratio between the two images, for example, in mm.

[0380] The next step is to detect vascular pathways in the two images.

[0381] The next step is to calculate the area of ​​the vascular pathways in the two images.

[0382] The next step is to subtract the area of ​​the first vascular pathway from the area of ​​the second vascular pathway.

[0383] The next step is to divide the calculated difference by the time interval between the two images.

[0384] For reference Figure 5A and 5BIt is a simplified diagram of the pulse wave traveling along the vein.

[0385] Figure 5A The first image is displayed, while Figure 5B The second image is shown; it was taken within a short period of time.

[0386] Figure 5A and 5B The image shows an arm 502, a vein 504, an artery 506, and wherein the vein 504 is connected to a fistula 508 of the artery 506.

[0387] Figure 5A A first position 510 is shown, in which at time t0, the vein is amplified by the pressure of the pulse wave.

[0388] Figure 5A A second position 512 is shown, in which at time t1, the vein is amplified by the pressure of the pulse wave.

[0389] The second position 512 is further along the vein 504 relative to the first position 510.

[0390] Pulse velocity can optionally be measured by dividing the distance between the first position 510 and the second position 512 by the time difference between the first and second images being captured.

[0391] In some embodiments, the time difference is a fraction of a second. As a non-limiting example, when the image frame is approximately 15-30 cm wide, the pressure wave traveling along the blood vessel can optionally be imaged at a frame rate higher than 120 fps, for example, at 165 fps.

[0392] In some embodiments, to ensure that the two locations have image frames with a field of view of 160 mm, and for a pulse velocity of approximately 20 m / s, the time difference is <6 ms. Such a time difference applies to all pulse velocities less than 20 m / s.

[0393] One or more of the following algorithms can be implemented in the system described in this paper:

[0394] Predictor / Classifier Methods

[0395] For reference Figure 6 This is a simplified flowchart illustration of a classifier method according to an exemplary embodiment of the present invention.

[0396] Figure 6 The input of one or more feature descriptors is shown, such as: a lateral branch vein descriptor 602, a pulse velocity descriptor 604, an arm elevation descriptor 606, and an aneurysm and / or stenosis descriptor 608.

[0397] In some embodiments, inputs 602, 604, 606, and 608 are fed into a threshold calculation unit 612.

[0398] In some embodiments, a trend calculation unit 614 may optionally accept input from a local or remote database a historical and / or trend descriptor 610.

[0399] In some embodiments, the trend calculation unit 614 generates a trend data output.

[0400] In some embodiments, the threshold calculation unit 612 generates a threshold data output.

[0401] In some embodiments, one or more of the above-described outputs are fed into a statistics unit 616.

[0402] In some embodiments, the output of the statistics unit 616 is input to a decision unit 618.

[0403] The decision unit 618 may optionally generate a decision that VA is determined to be “stable” 622 or a decision that a failure 620 is detected.

[0404] Figure 6 An exemplary classifier algorithm is described, which can perform the following steps based on machine learning (supervised or unsupervised) tools or heuristic rules:

[0405] 1. Data analysis, such as image processing.

[0406] 2. Extract features from one or a group of consecutive images, such as collateral vessel development and VA area.

[0407] Some examples of characteristics include: minimum radius size of the VA body, pulse wave velocity, collateral vein size and density, distance between collateral veins and AV or fistula, etc.

[0408] Features can also be variations in features between consecutive images and / or the rate of variation of features between consecutive images.

[0409] 3. Classify the state of VA based on the extracted features.

[0410] Classification can be based on basic rules, thresholds, and / or statistical models. Statistical models can be based on machine learning algorithms such as Support Vector Machine (SVM), Logistic Regression, Neural Networks, Decision Trees, k-means, etc.

[0411] The classification can be between two levels (with or without intervention) or between more than two levels.

[0412] Non-limiting examples of methods for prediction and / or classification include one or more of the following:

[0413] Parameter value table;

[0414] Regression of patient-related parameter values;

[0415] The K-Nearest Neighbors (KNN) algorithm applied to parameter values;

[0416] Support Vector Machine (SVM);

[0417] Deep learning;

[0418] Neural network.

[0419] In some embodiments, machine learning uses a training set. A non-limiting example of generating and using a training dataset includes repeatedly measuring parameters as described herein for N patients over a period of time.

[0420] During this period, it is recorded which patients developed vascular failure, and / or which patients underwent additional examinations such as physical examinations, ultrasound examinations, and X-ray imaging, and the specific nature of these additional techniques. The parameters and their determination may generate a training dataset that can be used to train the aforementioned machine learning methods, or to generate a KNN dataset.

[0421] In some embodiments, for predictive and / or classification purposes, some or all of the parameters measured for the aforementioned dataset are also measured for the patient.

[0422] It should be noted that the parameters collected over time may optionally include measurements and parameters calculated from the measurements, the first derivative of the parameter values, and the second derivative of the parameter values.

[0423] It is important to note that nurses or physicians performing vascular tests through observation, listening, and sensory methods typically make decisions or classifications based on the values ​​of one or two parameters, whereas the systems and methods described herein may use more parameters and may arrive at more accurate decisions or classifications based on the patient and / or a larger set of patients used to generate the training set.

[0424] Scaling algorithm

[0425] Scaling algorithms calculate image scale (e.g., scaling pixels to millimeters). Scaling can be used to calculate the absolute or relative values ​​of one or more of the following: vessel radius, pulse velocity, collateral vessel size, collateral vessel density, and distance from collateral vessels to the VA.

[0426] Registration algorithm

[0427] Registration algorithms can perform automatic or semi-automatic registration between two or more consecutive images.

[0428] Registration algorithms can align and / or scale two or more images containing the same object at different locations, viewpoints, or fields of view.

[0429] In some embodiments, the input to the registration algorithm includes at least two images, and in the case of semi-automatic registration, optionally, one or more points marked by the user on the two images.

[0430] Registration algorithms can potentially enable a system to measure changes between at least two examinations, regardless of how the arm or another examining organ is positioned during different examinations.

[0431] In some embodiments, registration of at least two images of the same patient containing VA objects may be accomplished optionally by detecting (e.g., segmenting) VA and fitting a VA image in a first image to a VA image in a second image by geometric transformation.

[0432] Vascular pathway (VA) detection algorithm

[0433] In some embodiments, automatic or semi-automatic detection of the location of vascular access bodies in an image is performed.

[0434] In some embodiments, the input to the algorithm for detecting vascular access bodies includes at least one image containing vascular access bodies in an image frame.

[0435] In some embodiments, optional input is a set of one or more points along the blood vessel including the VA body, optionally marked by a doctor / nurse on an image including the VA body.

[0436] The algorithm output can be a set of vascular pathway pixels in an image.

[0437] In some embodiments, computerized detection of VA bodies is based on unique VA shape, size, orientation, position, etc.

[0438] In some embodiments, as a non-limiting example, a device such as the "ELY-1000 Vascular Imaging Instrument for Arterial Puncture" developed by ELYNNSH MEDICAL is used. According to the manufacturer, this device helps medical personnel identify subcutaneous arteries during arterial puncture and can conveniently and quickly display the exact location and orientation of the artery.

[0439] In some embodiments, locations where arteries and veins connect or appear to connect are detected in the image.

[0440] In some embodiments, the blood vessels supplying the VA are surgically elevated towards the skin surface. Due to the depth differences in the vessel segments, the image covering the field-of-view (FOV) includes the VA, which typically appears as a closed contour centroid. The tissue surrounding the VA is usually deeper than the VA itself.

[0441] In some embodiments, the VA body may optionally detect depth differences, which may appear as areas darker than native or surrounding vessels. For example, when using NIR illumination, NIR light is absorbed by blood Hgb, and vessels closer to the surface appear darker than deeper vessels.

[0442] Pulse velocity algorithm

[0443] Refer again Figure 5A and 5B It describes an exemplary method for measuring pulse velocity.

[0444] Pulse velocity is also a commonly used indicator of arteriosclerosis. It can be obtained by measuring the distance between two blood vessels and the pulse propagation time. Pulse velocity can be measured locally, regionally, or systematically.

[0445] The term "locally" is used to refer to the area along the fistula and nearby associated vascular structures.

[0446] Physiologically, there is a relationship between pulse rate, blood flow, and intravascular pressure.

[0447] A pulse wave (caused by the heartbeat) travels from the heart to the arteries and then back to the heart from the veins. As the pulse travels, it causes temporary deformation of blood vessels (such as veins) at discrete points in time.

[0448] For example, the radius of a vein may temporarily increase at a certain point along the vein. This point can be detected by measuring the absorption of light by the blood flowing in the vein—the location of the dilated vein appears as a darker or brighter point along the vein (depending on the measurement method, such as reflection or transmission).

[0449] Pulse velocity can be calculated by detecting points associated with dilated blood vessels in two or more consecutive images, given the time between image captures. Pulse velocity is equal to the distance between the two points divided by the time between the two image captures.

[0450] Example Implementation - System Description

[0451] The system can measure one or more of the following examples: vessel diameter, pulse velocity, NIR (e.g., 700-1000 nm) reflectance spectrum, appearance and characteristics of collateral veins, such as density, size, distance from the vascular pathway, and oxygen concentration of the vascular pathway.

[0452] In some embodiments, the NIR spectral range is used for vascular imaging. There is a spectral window from approximately 700 nm to approximately 900 nm, during which light can penetrate deep into tissue, and veins absorb more radiation than surrounding tissue.

[0453] For reference Figure 7 This is a simplified block diagram of a system for measuring blood vessels according to an exemplary embodiment of the present invention.

[0454] Figure 7 A top-level block diagram of an example embodiment system 700 is shown. The system 700 may include an imaging / detection unit 702 and a software / computing unit 706.

[0455] The imaging / detection unit 702 may optionally include one or more sensors 710, one or more lenses 712, one or more filters 713, and one or more illuminators 714, 716.

[0456] In some embodiments, the sensor 710 may be a CMOS sensor.

[0457] In some embodiments, the sensor 710 may be a multispectral and / or hyperspectral camera.

[0458] In some embodiments, the sensor 710 may be an NIR sensor or a camera.

[0459] In some embodiments, the lens 712 may optionally be a fixed focal length lens.

[0460] In some embodiments, the lens 712 may optionally be a zoom lens.

[0461] In some embodiments, the filter 713 may optionally include a bandpass or longpass filter.

[0462] In some embodiments, the illuminators 714, 716 may optionally include NIR light-emitting diodes, optionally in the spectral range of 700-1200 nm.

[0463] In some embodiments, the illuminators 714, 716 may optionally include broadband NIR light-emitting diodes.

[0464] In some embodiments, the illuminators 714, 716 may optionally include one or more laser sources, optionally in the near-IR spectral range of 850 nm and 910 nm.

[0465] In some embodiments, the illuminators 714, 716 may optionally include narrowband illumination, optionally in a spectral range of 900 nm.

[0466] In some embodiments, the illuminators 714, 716 may optionally include an array of illuminators.

[0467] In some embodiments, the software / computing unit 706 may optionally include one or more of a graphical user interface 734, an image processing unit 735, a computer vision unit 736, and a machine learning algorithm unit 737.

[0468] In some embodiments, the algorithm unit 737 may optionally include one or more of the following: image processing algorithm, vein segmentation algorithm, collateral vein detection and / or segmentation algorithm, pulse detection algorithm and classifier algorithm – optional machine learning algorithm.

[0469] The system 700 may include additional units, such as a workstation 704, an optional cloud infrastructure 708, etc.

[0470] In some embodiments, the cloud infrastructure 708 may optionally include one or more of a web application 738, a database 740 (optionally including big data analytics capabilities), and an analytics unit 742.

[0471] In some embodiments, the workstation 704 may optionally include one or more of the following:

[0472] A control unit 720 is used to control the operation of the imaging / detection unit 702 and / or one or more components of the imaging / detection unit 702;

[0473] A computer 722;

[0474] One monitor has 724;

[0475] An optional organ fixation surface or device 726 for optionally positioning the organ at a specific location relative to the illumination 714, 716 and / or the sensor 710; and

[0476] A stent 728 is used to place components of the system at a specific location relative to the patient's organ.

[0477] For reference Figure 8A and 8B It is an image of an optical component in a system constructed according to an exemplary embodiment of the present invention.

[0478] Figure 8A and 8B The optical channels of some example systems are shown, which may include, for example Figure 8A As shown:

[0479] A camera 802, optionally a hyperspectral sensor (camera);

[0480] Lens 802, optionally a fixed focal length lens;

[0481] An optional filter holder 806;

[0482] A filter 808, in some embodiments being an optical long-pass filter, and in other embodiments being a filter with a cut-off wavelength of 670 nm; and

[0483] One light source 812.

[0484] In some embodiments, the system includes an optional mechanical adapter 810 for connecting the lighting source 812 to the camera body 802.

[0485] Figure 8B It shows including Figure 8A An assembly unit 814 of the components.

[0486] In some embodiments, the example blood vessel state classification algorithm can be divided into three parts: image processing, feature extraction, and statistical classifier.

[0487] The example algorithm's top-level process can be compared with... Figure 4A Similar to what is shown.

[0488] Image processing: An image processing block may include several steps:

[0489] • Image quality enhancement, such as contrast and brightness enhancement, sharpness, combination of multiple polarization state images, multiple wavelength images (image intensity ratio), multiple exposures, optional high dynamic range (HDR), and contrast limited adaptive histogram equalization (CLAHE).

[0490] In some embodiments, the image intensity ratio shows the pixel-wise ratio between images taken at different wavelengths, as described in the following equation:

[0491]

[0492] in:

[0493] Rij is the pixel at position (i,j) in the ratio image, IM1ij is the pixel at position (i,j) in the first image, and IM2ij is the pixel at position (i,j) in the second image.

[0494] • Image segmentation – Localizing vascular pathway (VA) structures, vascular boundaries, and collateral vessel structures.

[0495] For reference Figure 9 This is a simplified flowchart illustrating the segmentation method according to an exemplary embodiment of the present invention.

[0496] Figure 9 An exemplary segmentation process is illustrated, including a first image 902 as input, segmentation 904 of the first image 902 to produce a second image 906 with optional segmentation lines 907, optional isolation 908 of organs appearing in the second image 906, and producing a third image 910 containing only the isolated organs.

[0497] One or more of the following methods can be used: k-means algorithm, histogram-based methods, edge detection, region growing methods, Mumford-Shah segmentation, CNN (convolutional neural network), etc.

[0498] • Registration between one or more images during early and subsequent checks. The registration step may optionally scale and / or align the new image to a reference image, optionally from the image in the early check.

[0499] For reference Figure 10 , Figure 10 This is a simplified flowchart illustrating the registration method according to an exemplary embodiment of the present invention.

[0500] Figure 10A first image 1002A and a second image 1006A are shown.

[0501] In some embodiments, point detection operation 1004 may be performed on two images.

[0502] In some embodiments, the point detection criterion may optionally be one or more of the following: corner points, intensity-based criteria, such as blob detection, Speeded Up Robust Features (SURF), etc.

[0503] In some embodiments, the similarity between two points is measured by the difference in feature metrics between one or more feature metrics of each of the two points.

[0504] The first image 1002A is marked with specific points detected in the first image 1002A, generating a first new image 1002B on which specific points are marked. The second image 1006A is marked with specific points detected in the second image 1006A, and optionally, a second new image 1006B is generated with specific points marked according to the same criteria used to detect points in the first image 1002A.

[0505] Figure 10 Lines 1007 are shown connecting corresponding specific points in the first new image 1002B and the second new image 1006B.

[0506] In some embodiments, one or both of the first new image 1002B and the second new image 1006B are optionally transformed 1008, using the detection of corresponding marker points to perform the transformation, optionally producing a new combined image 1010. In some embodiments, the transformation 1008 includes one or more of image normalization, image scaling, image rotation, and affine transform performed on one or both of the first new image 1002B and the second new image 1006B.

[0507] In some embodiments, the registration is performed to align and / or scale the first image (e.g., the currently inspected image) to a second image (e.g., a previously inspected image). Registration can be performed using one or more of the following methods: Scale Invariant Feature Transform (SIFT), accelerated robust feature algorithms, optionally for interest point detection, automatic feature detection and matching, and affine transformation computation.

[0508] • Feature extraction: A feature extraction block may consist of several sub-blocks used to analyze data from an image and extract features.

[0509] In some embodiments, the feature extraction may optionally generate a feature vector.

[0510] In some embodiments, the feature extraction may be performed after an image processing step that produces a standardized image.

[0511] The feature vector is a mathematical representation used to characterize data such as images. Several methods can be used to characterize data; some are listed below:

[0512] Extracting features from a pre-trained DNN (A. Krizhevsky, I. Sutskever, GE using deep convolutional neural networks for ImageNet classification, NIPS 2012:1106-1114).

[0513] One approach involves passing images by training a neural network on a large image dataset and using its descriptor layers.

[0514] Another approach is to develop specific descriptors for each phenomenon.

[0515] Blood vessel length and minimum diameter:

[0516] For reference Figure 11 This is a simplified flowchart illustration of a method according to an exemplary embodiment of the present invention.

[0517] Figure 11 This paper describes a method for generating blood vessel length and / or minimum diameter descriptors.

[0518] Figure 11 It is shown that:

[0519] The first image 1102 is used as input;

[0520] The conversion 1104 from the first image 1102 to the binary image 1106;

[0521] The narrowest passage in an organ (blood vessel) is located at position 1114; and

[0522] Tracing 1108 of a center line of an organ (blood vessel) appearing in the binary image 1106 produces a third image 1110, in which the center line 1112 of the organ (blood vessel) is marked on the third image 1110.

[0523] Similar methods can be optionally used to generate descriptors for “pulse velocity”, “collateral vessel development”, and “aneurysm and stenosis”.

[0524] In the above method, the "distance transform" and "local maximum" methods can be used on binary images to detect the centerline and diameter of blood vessels.

[0525] Other potentially useful methods include: pathfinding algorithms - Dijkstra's algorithm and A* search algorithm.

[0526] Mixed arterial and venous oxygen concentrations in VA:

[0527] For reference Figure 12 This is a simplified flowchart illustration of a method according to an exemplary embodiment of the present invention.

[0528] Figure 12 It is shown that:

[0529] The first image 1202 is used as input;

[0530] Histogram unit 1204 is used to generate a histogram 1206 of the first image 1202; and

[0531] The calculation unit 1210 is used to generate a feature vector 1212 associated with the first image 1202.

[0532] In the 740nm to 760nm range, deoxyhemoglobin (deoxy Hb) absorbs light more readily than oxyhemoglobin (Oxy Hb). Therefore, in this range, veins absorb light radiation more readily, while arteries become relatively more transparent.

[0533] In the 850nm to 1000nm range, veins become relatively more transparent, while arteries absorb more radiation.

[0534] The blood in the VA is a mixture of arterial and venous blood, especially when narrowing occurs, which can lead to blood recirculation.

[0535] If the VA function is good, there should be higher arterial blood flow through the VA, which is represented by darker grayscale levels under illumination of 850 nm to 1000 nm. This is achieved by calculating the histogram of the intensity-normalized image, for example... Figure 12 The system can create a feature vector to describe the changes in the blood mixture in the VA, or the rate of change of the blood mixture in the VA, by comparing the histogram of the first image 1202 shown with the histogram of the "reference" image.

[0536] Pulse velocity

[0537] For reference Figure 13 This is a simplified flowchart illustration of a method according to an exemplary embodiment of the present invention.

[0538] Figure 13 A method for calculating pulse velocity is shown.

[0539] Figure 13 show:

[0540] A first image 1302 obtained at time t0 is used as input;

[0541] A second image 1304 obtained at time t1 is used as input; and

[0542] A computing unit 1306 for generating a third image 1308.

[0543] Figure 13 An exemplary method for feature extraction of pulse velocity is shown.

[0544] In some embodiments, optionally, after registration and / or segmentation, two consecutive image frames are fused, for example... Figure 13 Images 1302 and 1304 are used to generate a fused image, such as the third image 1308.

[0545] In some embodiments, the fused image is generated by subtracting one image from another.

[0546] In some embodiments, the fused image is generated by adding one image to another.

[0547] In some embodiments, the centroids of the two brightest points 1312, 1314 are calculated, and the length of the path 1312 along the path 1312 between the centroids of the two brightest points 1312, 1314 is measured.

[0548] In some embodiments, the path 1312 may optionally be the centerline of a blood vessel.

[0549] Statistical classifier model

[0550] For reference Figure 14 This is a simplified flowchart illustration of a classifier method according to an exemplary embodiment of the present invention.

[0551] Figure 14 The input of one or more feature descriptors is shown, such as: a lateral branch vein descriptor 1402, a pulse velocity descriptor 1404, an aneurysm and / or stenosis descriptor 1406, and an arterial and venous blood mixture descriptor 1408.

[0552] In some embodiments, inputs 1402, 1404, 1406, and 1408 are fed into a trend calculation unit 1412. In some embodiments, the trend calculation unit 1412 may optionally accept inputs from historical and / or trend descriptors 1410, optionally from a local or remote database.

[0553] In some embodiments, the trend calculation unit 1412 generates a trend feature vector 1414.

[0554] In some embodiments, the trend feature vector 1414 may optionally be stored in a (local or remote) database.

[0555] In some embodiments, the trend feature vector 1414 is input into a classifier 1416.

[0556] The result of the classifier 1416 may optionally be input to a decision unit 1418, which generates a decision to determine that the VA is determined to be “stable” 1420 or to detect a failure 1422.

[0557] In some embodiments, detecting failure may include estimating a high probability that the VA is about to fail.

[0558] Classifying a system as "stable" or "early failure detection" can be accomplished using a statistical classifier model, such as SVM, logistic regression, or neural networks.

[0559] In some embodiments, the extracted features 1402, 1404, 1406, 1408 of each phenomenon are optionally collected into a “feature” vector 1414.

[0560] In some embodiments, the feature vector 1414 may optionally be stored in a database.

[0561] In some embodiments, the feature vector 1414 and a "historical feature vector" 1410 may optionally be sent to a "trend calculation" unit 1412.

[0562] In some embodiments, the output of the "trend calculation" unit 1412 is a "new trend feature vector" 1414, which may be stored in a database and / or sent to a classifier unit 1416.

[0563] In some embodiments, the output from the classifier unit 1416 can be detected as "early failure detection" or "stable".

[0564] In some embodiments, maturity or maturity rate after VA surgery is classified.

[0565] The maturity rate of a fistula can be expressed as the maturity of X% after Y days.

[0566] Now for reference Figure 15A -C, which shows three different images of the same patient's arm according to an exemplary embodiment of the present invention.

[0567] Figure 15AAn image of a patient's arm at human visible wavelengths is shown, taken at a distance of approximately 40 centimeters from the arm.

[0568] Figure 15B An image of a patient's arm in near-infrared wavelengths is shown. Figure 15B The image shows that near-infrared imaging was used to improve the visibility of blood vessels, such as superficial veins 1512.

[0569] Figure 15C An image of a patient's arm is displayed, where point of interest 1522 is automatically generated (through image analysis) at the location of blood vessels.

[0570] When a nurse or doctor examines a patient's blood vessels, they typically use a three-step procedure: observation, listening, and feeling.

[0571] In some embodiments, the system described herein performs observation, listening, and sensing based on illuminating and imaging a patient's limb and analyzing data collected from the imaging.

[0572] In some embodiments, the methods described herein are based on illuminating and imaging a patient's limb and analyzing data collected from the imaging to perform observation, listening, and sensing.

[0573] For reference Figure 16A It is a form that shows the process by which medical staff examine patients for vascular stenosis or thrombosis.

[0574] Figure 16A The purpose is to show instructions to humans. However, it is well known that differences between humans can predictably affect such examinations.

[0575] It is worth noting that automated checks may provide better repeatability for such checks.

[0576] It should be noted that automated screening may provide faster testing while reducing the involvement of medical staff.

[0577] For reference Figure 16B This is a simplified flowchart illustrating a method for examining a patient according to an exemplary embodiment of the present invention.

[0578] The method in Figure 16 includes:

[0579] A device observes (1622) a patient’s body by taking one or more images of the body and using image analysis on the images;

[0580] The device listens to (1624) the patient's body by capturing vibrations and analyzing vibrations at human audible frequencies; and

[0581] The device senses (1624) the patient’s body by analyzing bodily vibrations below frequencies audible to humans.

[0582] In some embodiments, capturing one or more images of the body may be performed by capturing images at near-IR wavelengths.

[0583] In some embodiments, as described elsewhere herein, body vibrations may optionally be captured by laser speckle imaging.

[0584] In some embodiments, vibrations of the body may be captured via a microphone that contacts the patient's body and / or via a microphone connected to a stethoscope that contacts the patient's body.

[0585] It should be noted that in some embodiments, automated screening may be able to provide such screening without human contact with the patient, and may be used in situations where medical distancing is required, such as when the patient may be carrying an infectious disease.

[0586] In some embodiments, the systems and methods described herein may optionally “observe” (i.e., analyze images of blood vessels), “listen” (i.e., analyze vibrations of the patient’s body at human hearing frequencies), and “feel” (i.e., analyze vibrations of the patient’s body at low frequencies, down to frequencies below typical audio frequencies).

[0587] In some embodiments, non-contact monitoring tools are provided to supplement and / or replace physical examinations of vascular access (VA). Such monitoring has the potential to enable early detection of stenosis, possibly earlier than manual examination.

[0588] In some embodiments, the monitoring tool does not contact the patient's fistula and / or the patient's limb, even if the limb is optionally located in a device that can locate the fistula within the device's field of view.

[0589] In some embodiments, monitoring the recorded and monitored parameters may enable early detection and / or prediction of stenosis, potentially earlier than a human examination.

[0590] In some embodiments, monitoring can be performed without human contact, for example at a distance greater than 10, 20, 30, 40, or 50 centimeters from the VA location.

[0591] In some embodiments, the system and method may optionally be able to acquire all parameters typically obtained through observation, sensation, and listening during a human physical examination.

[0592] In some embodiments, training personnel to operate the monitoring VA using the methods described herein is easier than using human senses.

[0593] Using the embodiments described herein, value can be added by recording and using historical data of the same patient and tracking changes.

[0594] Using the embodiments described herein, it may be possible to conduct pre-session and / or post-session examinations in clinics without physical contact.

[0595] Using the embodiments described herein may support healthcare during COVID-19.

[0596] Care can be provided in a home setting that may be operated by a patient, using the embodiments described herein.

[0597] For reference Figure 17 This is a simplified block diagram illustration of a method for examining a patient according to an exemplary embodiment of the present invention.

[0598] Figure 17 A method is shown, including:

[0599] Received an examination from a patient (1702);

[0600] Patient (1704) was measured using an embodiment of the present invention;

[0601] Collect data from sensors (1706);

[0602] Analyze the data (1708); and

[0603] Optional decision-making regarding the patient's fistula status (1710).

[0604] The status of a patient's fistula may optionally include determining the medical condition and / or patency of the fistula. The medical condition may optionally be determined as healthy and / or functionally normal or with a probability of deterioration. In some embodiments, a probability of deterioration exceeding a certain threshold may optionally generate a recommendation to refer the patient for additional examination, such as Doppler ultrasound or angiography.

[0605] For reference Figure 18A and 18B These are images of the patient's fistula taken at two different times.

[0606] Figure 18A An ink mark 1802 shows the outline of the fistula. Figure 18A It also shows the physical characteristics visible on the patient's skin 1804.

[0607] Figure 18B These are images of the fistula taken at different times. Figure 18BThe ink mark 1802 is shown to have changed shape due to changes in the shape and / or size of the fistula. Figure 18 also shows that the physical feature 1804 appears to have moved relative to the outline of the fistula or relative to the ink mark 1802.

[0608] Figure 18A and 18B This is an image of a 36-year-old male with a right cephalobrachial fistula created in 2004. The fistula contains multiple aneurysms 1801A and 1801B.

[0609] Now for reference Figure 19A and 19B This is a simplified diagram of a system for monitoring vascular access (VA) and / or fistulas according to two exemplary embodiments of the present invention.

[0610] Figure 19A A system 1900 is shown, comprising a head 1902 and a base 1906. The head 1902 optionally includes a projector and an imaging system. In some embodiments, the base 1906 optionally includes a shape configured to support an arm or leg at a specific position relative to the head 1902. In some embodiments, the base 1906 optionally includes straps configured to support an arm or leg at a specific position relative to the head 1902.

[0611] Figure 19A A system 1910 is shown, comprising a projector 1912 and an imager 1914. In some embodiments, the projector 1912 includes an optional cover 1916. In some embodiments, particularly where the projector includes a laser, the cover 1916 may be required and / or needed for safety reasons.

[0612] In some embodiments, the imager 1914 is optionally capable of imaging frames at a rate higher than the standard video rate (selectively 60 frames per second (FPS), higher than 60 FPS, higher than 100 FPS, higher than 150, 200, 300, 400, 500, and 600 FPS).

[0613] High frame rates enable the detection of high-frequency vibrations in a patient's body, such as those known in the art: Shannon's Law.

[0614] In some embodiments, a NIR fast camera may be used, and a frame rate of 150 FPS or higher may be selected.

[0615] In some embodiments, an off-the-shelf camera, such as the FLIRFL3 U3 camera, can be used to image at a frame rate of 150 FPS and a full frame size of 1.3 megapixels.

[0616] In some embodiments, the camera is used to capture frame rates exceeding 160 FPS, up to 600 FPS, 620 FPS, etc.

[0617] In some embodiments, off-the-shelf cameras may be used, capable of imaging at frame sizes of 1.3-2 megapixels or more.

[0618] In some embodiments, off-the-shelf cameras may be used, capable of imaging at higher frame rates when imaging at lower frame sizes. By way of some non-limiting examples, the camera may optionally image at sizes such as 10x20 pixels, 10x10 pixels, etc.

[0619] In some embodiments, the imager captures small frames (smaller than the maximum frame size and optionally as small as the aforementioned frame size) of a specific location of interest on the patient's body at the fistula location or the VA point of interest location.

[0620] In some embodiments, the projector may be selected to project light onto a location of interest so that a user can accurately position the patient's body.

[0621] In some embodiments, the location of interest is the patient's fistula.

[0622] In some embodiments, multiple points are illuminated simultaneously.

[0623] In some embodiments, the location of interest for which one point is illuminated is the patient's fistula, and the other location of interest for which one point is illuminated is a location adjacent to but not at the fistula.

[0624] In some embodiments, the location of interest for which one point is illuminated is the fistula aneurysm, and the other location of interest for which one point is illuminated is a location adjacent to the fistula aneurysm but not at the fistula aneurysm.

[0625] In some embodiments, the projector is a digital light processing projector.

[0626] In some embodiments, the projector is a laser projector.

[0627] In some embodiments, a location of interest, such as a fistula or aneurysm, or a swollen area of ​​the body, may be identified using structured light and image analysis, and the projector is controlled (optionally automatically) to illuminate the location of interest. In some embodiments, a digital light processing projector and / or a laser projector may be controlled to illuminate the location of interest.

[0628] In some embodiments, a doctor or nurse controls the lighting at a location of interest.

[0629] In some embodiments, a doctor or nurse controls the laser to be directed at the location of interest.

[0630] In some embodiments, the projector may optionally be capable of projecting light in multiple modes. These modes include two or more of the following:

[0631] Project uniform (or near-uniform) illumination onto an area or a restricted point on a patient’s body that may be sufficient to image collateral veins.

[0632] Projecting structured light, optionally including light stripes of a specific width, equal or unequal width, such as programmed or other patterns; and

[0633] Projecting one or more coherent laser spots can be used to measure one or more of vibrations, micro-vibrations, and pulses, for example, by laser speckle interferometry.

[0634] In some embodiments, the projector can switch between any of three different lighting modes: uniform, structured light, and point light.

[0635] In some embodiments, the projector is capable of providing a spot size between 0.5 mm and 5 mm in diameter on a patient's limb. For example, the spot size is approximately 1 mm.

[0636] In some embodiments, the projector includes one or more LED and / or laser light sources, optionally at near-IR wavelengths.

[0637] In some embodiments, the projector may optionally be a digital light processing projector.

[0638] In some embodiments, the projector may optionally include nanomirrors to shape light.

[0639] In some embodiments, the projector may optionally include a microelectromechanical system (MEMS) mirror for shaping light.

[0640] In some embodiments, the projector may optionally include a digital mirror driver (DMD).

[0641] In some embodiments, the projector and the camera are packaged in a single package.

[0642] Now for reference Figure 20 The images are of a system for monitoring vascular access (VA) and / or fistulas according to an exemplary embodiment of the present invention.

[0643] Figure 20 A system including a projector 2004, an imager 2006 and an optional processor 2002 is shown.

[0644] Now for reference Figure 21A -C, which shows three different images of the same patient's arm.

[0645] Figure 21A The image shows the patient's arm remaining below the level of the patient's heart. Two inflatable needle insertion points, 2102 and 2104, are shown on the fistula in the patient's arm.

[0646] Figure 21B The image shows the patient's arm held above the level of the patient's heart, taken after the patient raised their arm to the elevated position. The first insertion point 2102 is shown deflated, while the second insertion point 2104 remains inflated.

[0647] Figure 21C The image shows the patient's arm raised, the image is larger than Figure 21B The images were taken later. Both needle insertion points 2102 and 2104 show deflation and collapse.

[0648] Elevation test inspection

[0649] In some embodiments, the systems and methods described herein may be optionally used to measure and quantify in elevation tests, i.e., to measure and quantify once or multiple times while keeping the body or limb below heart level and to measure and quantify once or multiple times while keeping the body or limb in an elevated position above heart level.

[0650] like Figure 21A As shown in -C, in some cases, a fistula may not drain when held in one location, but may drain when held in one or more other locations.

[0651] The differences between locations are related to the medical conditions of the fistula, such as the ratio between inflow and outflow rates and / or pressures.

[0652] The rate and manner of fistula drainage, as well as the differences between drainage patterns at different heights, are related to the medical condition of the fistula. In some embodiments, the drainage rate and / or pattern are measured, optionally by generating one or more 3D curves depicting the external shape of the fistula and / or tracing changes between said curves along said set.

[0653] In some embodiments, the drainage rate and pattern can optionally be estimated by evaluating the volume encapsulated by a 3D shape and / or one or more curves and tracing the change of the total volume over time, potentially providing the drainage level and rate.

[0654] In some embodiments, the spatial curvature of the curve can be estimated, and the drainage pattern can be characterized by analyzing the change of curvature over time.

[0655] In some embodiments, the curvature smoothness of each curve and the change in curvature smoothness over time during drainage can be used to estimate the drainage pattern.

[0656] In some embodiments, the correlation between any flow-related parameters estimated based on measured vibration analysis and any parameters estimated from the drainage pattern or rate based on 3D shape or curve shape can optionally be used to estimate fistula health.

[0657] Now for reference Figure 22A It displays a graph of the vibrational power spectrum measured by analyzing images generated by laser speckle imaging.

[0658] Figure 22A A graph 220 is displayed, with the X-axis 2202 showing the frequency range or interval and the Y-axis 2204 showing the relative power spectrum in units of measurement.

[0659] This graph was generated by sampling from two groups of patients. Group 1, 2206, had a blood flow velocity (FV) greater than 500 mL / min, while Group 2, 2208, had an FV less than 500 mL / min.

[0660] Figure 2200 shows that the maximum values ​​in the power spectra of the two groups are located at approximately 140 Hz. This leads us to suspect that listening to the pitch of the two blood flow signals may not be a good way to distinguish them. However, analysis of the power spectra of the two groups reveals differences:

[0661] The first group 2206 appears to have a higher amplitude at the position of the maximum value than the second group 2208;

[0662] The second group, 2208, appears to have a flatter or wider curve than the first group, 2206.

[0663] like Figure 22A As shown, the vibrations analyzed in the power spectrum are caused by blood flow and / or turbulence through the blood vessels.

[0664] Flow and turbulence vary over time and are influenced by local physical conditions within and around the blood vessel where the flow occurs. These physical conditions may include pressure gradients, vessel diameter, vessel wall compliance, and the properties of the vessel's inner surface.

[0665] The blood flow power spectrum measured at the VA / fistula location may be related to the physical and / or clinical blood flow conditions at these locations. Changes in this power spectrum characteristics over time may be associated with deterioration of fistula health. Analyzing changes in the power spectrum obtained from the VA / fistula location may be an early predictor of fistula deterioration.

[0666] Referring to the early predictions described in this article, it should be noted that such predictions may lead to percutaneous transluminal angioplasty (PTA) being performed earlier than based on existing medical examinations.

[0667] In some embodiments, the power spectrum is measured by measuring the intensity of light reflected from the patient's body. The intensity is expected to vary with frequency in relation to the body's vibrational frequency.

[0668] In some embodiments, the power spectrum is measured by measuring the intensity of light reflected from a point of illumination on the patient's body. In such embodiments, the vibration is actually measured specifically at the point of illumination.

[0669] In some embodiments, the power spectrum is measured by measuring the differences between successive images of the body, such as minute changes in patterns on the body. These patterns can be moles on the skin, structured light, movement of light spots, movement of laser speckle, and similar movements.

[0670] For reference Figure 22B This is a simplified flowchart illustration of a method for converting data from an image stream into a spectrum according to an exemplary embodiment of the present invention.

[0671] Figure 22B The methods include:

[0672] Receive an image stream of a patient's body (2222);

[0673] Optionally, one or more pixels (2224) that have a high variance of intensity over the duration of the image stream can be selected;

[0674] A strength vector (2226) is generated over the duration;

[0675] The intensity vector is transformed into a spectrum vector (2228).

[0676] In some embodiments, the transformation is performed using a Fast Fourier Transform.

[0677] In some embodiments, the power spectrum may optionally be normalized before analysis. As some non-limiting examples, the normalization factor may optionally be calculated from the following: total spectral energy, peak value, peak-to-baseline ratio, energy in a specific bandwidth, etc.

[0678] In some embodiments, a reference spectrum measured at a remote location (e.g., away from the fistula) is used as a reference. Both spectra may or may not be normalized, and the measured spectra are replaced by differences between the spectra at different locations.

[0679] In some embodiments, the skewness or kurtosis of the measured power spectrum or power spectrum difference may optionally be used to estimate the flow rate.

[0680] In some embodiments, the measured power spectrum is first fitted to a model. In some embodiments, it is assumed that one or more hidden model mixers (as a non-limiting example, a Poisson-Gaussian mixture) and model parameters are used as correlators to flow.

[0681] In some embodiments, energy within a specific frequency range is used to estimate the flow rate.

[0682] Perform observation, listening, and sensing.

[0683] We also describe some aspects of the invention based on the “observation, listening, and feeling” procedures used by medical personnel.

[0684] In some embodiments, the “observation, listening, and sensing” procedure may optionally be performed by an embodiment of the system described herein.

[0685] In some embodiments, the system described herein performs observation, listening, and sensing based on illuminating and imaging a patient's limb and analyzing data collected from the imaging.

[0686] In some embodiments, the methods described herein are based on illuminating and imaging a patient’s limb and analyzing data collected from the imaging to perform observation, listening, and sensing.

[0687] A fistula murmur, also known as a vascular murmur, is an indicator of how the dialysis access is functioning.

[0688] An arteriovenous fistula (AVF) is a type of passage formed by connecting an artery to a vein under the skin (usually located in the upper arm, lower arm, or leg). (i) The high blood flow from the artery to the vein makes the fistula larger and stronger. A healthy AV fistula has a murmur (a rumbling sound that humans can hear), a thrill (a rumbling sound that humans can feel), and good blood flow velocity.

[0689] In some embodiments, the "observation" aspect may optionally be performed by imaging the body or limbs and analyzing one or more images to quantify vascular structures and / or fistula structures.

[0690] In some embodiments, the "observation" aspect may optionally be performed by imaging the body or limb using structured light and generating a 3D shape of the fistula.

[0691] In some embodiments, the "listening" aspect may optionally be performed by measuring and analyzing vibrations of the body or limbs to quantify parameter values ​​related to the medical condition of the fistula. In some embodiments, the "listening" aspect includes analyzing vibrations within a frequency range that is audible to humans.

[0692] In some embodiments, the "sensory" aspect may optionally be performed by measuring and analyzing vibrations of the body or limbs to quantify parameter values ​​related to the medical condition of the fistula. In some embodiments, the "sensory" aspect includes analyzing vibrations, optionally in frequency ranges even beyond and / or below the human audible range.

[0693] In some embodiments, vibration analysis may be performed optionally in a frequency range less than 1,000 Hz. In some embodiments, vibration analysis may be performed optionally in a frequency range less than that of typical human speech, such as less than 4,000 Hz.

[0694] By way of some non-limiting examples, the aspect of "feeling" includes one or more of the following:

[0695] Measuring a human pulse typically involves analyzing vibrations at frequencies of 40 beats per minute or higher. This requires analyzing vibrations with frequencies of 1 Hz or even lower. When performing this analysis by analyzing image frames from a video sequence, analyzing the image frames at approximately twice the frequency being measured is sufficient; that is, for example, approximately 2 frames per second or higher.

[0696] jitter is typically measured in the range of 50-250 Hz or 50-750 Hz. When performing this analysis by analyzing image frames of a video sequence, it is sufficient to analyze the image frames at approximately twice the frequency being measured, that is, for example, approximately 100 frames per second or higher.

[0697] Analyzing and quantifying the power spectrum of vibrations, for example, as referenced above. Figure 22A and 22B As described.

[0698] In some embodiments, “observation, listening, and sensing” are performed without physical contact with the patient through image analysis and / or by using specific lighting patterns.

[0699] For reference Figure 23 This is a simplified flowchart illustrating a method for monitoring vascular function according to an exemplary embodiment of the present invention.

[0700] Figure 23 The methods include:

[0701] Illuminate one or more blood vessels through a patient's skin (2302);

[0702] Take at least one image of the blood vessel (2304);

[0703] Analyze the at least one image (2306); and

[0704] Parameters related to vascular function were calculated based on image analysis (2308).

[0705] For reference Figure 24 This is a simplified flowchart illustration of a method for replacing a physical examination performed by a doctor to monitor vascular function, according to an exemplary embodiment of the present invention.

[0706] Figure 24 The methods include:

[0707] Generate at least one image of a patient's organ (2402);

[0708] Analyze the at least one image (2404); and

[0709] Generate parameter values ​​related to vascular function (2406).

[0710] Analyzing and quantifying the power spectrum of vibrations, for example, as referenced above. Figure 22A and 22B As described.

[0711] In some embodiments, “seeing, hearing, and feeling” are performed without physical contact with the patient through image analysis and / or by using specific lighting patterns.

[0712] For reference Figure 23 This is a simplified flowchart illustrating a method for monitoring vascular function according to an exemplary embodiment of the present invention.

[0713] Figure 23 The methods include:

[0714] Illuminate one or more blood vessels through the patient's skin (2302);

[0715] Capture at least one image of the blood vessel (2304);

[0716] Analyze at least one image (2306); and

[0717] Parameters related to vascular function were calculated based on image analysis (2308).

[0718] Now for reference Figure 24 This is a simplified flowchart illustrating a method for replacing a physical examination performed by medical personnel to monitor vascular function, according to an exemplary embodiment of the present invention.

[0719] Figure 24 The methods include:

[0720] Generate at least one image of the patient's organ (2402);

[0721] Analyze at least one image (2404); and

[0722] Generate parameter values ​​related to vascular function (2406).

[0723] Although specific embodiments have been disclosed in detail herein, they are provided by way of example for illustrative purposes only and are not intended to limit the scope of the appended claims. In particular, various substitutions, changes, and modifications are contemplated without departing from the spirit and scope of this disclosure as defined by the claims. In other aspects, advantages, and modifications are considered to be within the scope of the appended claims. The proposed claims represent the embodiments and features disclosed herein. Other unstated embodiments and features are also contemplated. Therefore, other embodiments are within the scope of the appended claims.

[0724] It is anticipated that many related image processing algorithms will be developed during the validity period of this patent application; the scope of the terminology used herein is intended to include all such new technologies of priori.

[0725] As used in this article, when referring to quantities or values, the term “approximately” means “within ±20%”.

[0726] The terms “comprising,” “including,” “having,” and their variant forms refer to “including but not limited to.”

[0727] The term "consisting of" means "including but not limited to".

[0728] The term "essentially consisting of" means that a composition, method, or structure may include additional ingredients, steps, and / or components, but only if the additional ingredients, steps, and / or components do not substantially alter the essential or novel characteristics of the claimed composition, method, or structure.

[0729] As used herein, the singular forms “a,” “an,” and “the” include plural references unless the context clearly specifies otherwise. For example, the terms “a unit” or “at least one unit” can include multiple units, including combinations thereof.

[0730] The terms “example” and “exemplary” as used herein mean “serving as an example, illustration, or description.” Any embodiment described as “example” or “exemplary” is not necessarily to be construed as preferred or superior to other embodiments and / or as excluding combinations of features of other embodiments.

[0731] The term "optionally" as used herein means "provided in some embodiments but not in others." Any particular embodiment of the invention may include a number of "optional" features unless such features conflict with each other.

[0732] Throughout this application, various embodiments of the invention may exist in a range format. It should be understood that this range format is merely for convenience and brevity and should not be construed as a rigid limitation on the scope of the invention. Therefore, it should be assumed that the range description specifically discloses all possible sub-ranges and single numerical values ​​within that range. For example, a range description from 1 to 6 should be assumed to specifically disclose sub-ranges such as from 1 to 3, from 1 to 4, from 1 to 5, from 2 to 4, from 2 to 6, from 3 to 6, etc., and single digits within the range, such as 1, 2, 3, 4, 5, and 6, regardless of the range itself.

[0733] Whenever a range of numbers is indicated herein (e.g., “10-15”, “10 to 15”, or any pair of numbers connected by such other ranges), unless the context clearly indicates otherwise, it means to include any number (fraction or integer) within the range indicated. “The range between the first and second indicated numbers” and “the range from the first to the second indicated number” are used interchangeably herein and refer to the range including the first and second indicated numbers, and all fractions and integers in between.

[0734] Unless otherwise stated, the numbers used herein and any ranges of numbers based thereon are approximations within the range of accuracy that would be understood by those skilled in the art to be reasonable for measurement and rounding errors.

[0735] It is understood that certain features of this invention, described in separate embodiments for clarity, may also be provided in combinations of a single embodiment. Conversely, for brevity, various features described in a single embodiment may also be provided separately, in any suitable sub-combination, or in embodiments applicable to any other description of the invention. Specific features described in the various embodiments are not considered essential features of those embodiments unless the embodiment would not function without those elements.

[0736] While the invention has been described in conjunction with specific embodiments thereof, it will be apparent to those skilled in the art that many alternatives, modifications, and variations will be readily apparent. Therefore, it is intended to include all alternatives, modifications, and variations falling within the scope of the appended claims.

[0737] All publications, patents, and patent applications mentioned in this specification are incorporated herein by reference in their entirety, to the same extent as if each individual publication, patent, or patent application were specifically and individually identified and incorporated herein by reference. Furthermore, any references cited or indicated should not be construed as an admission that such references are prior art to the present invention. Heading portions in this application are used herein to facilitate understanding of the specification and should not be construed as necessary limitations. Additionally, any priority documents of this application are incorporated herein by reference in their entirety.

Claims

1. A system for monitoring vascular access of a patient receiving hemodialysis treatment through automated visual, auditory, and sensory checks, characterized in that: The system includes: A camera, configured to image a patient's body to obtain the geometry of blood vessels; A light source; and A processor configured to receive images from the camera; The system is configured as follows: Image analysis was used to obtain the shape of a body organ at a location of a vascular pathway in the patient's body; and Image analysis was used to analyze the vibrations of the patient's body at a location on the patient's body, including the blood vessels; The system is programmed to automatically observe, listen to, and perform sensory checks on the patient's body in the following ways: The observation is achieved by performing image analysis on multiple images obtained from imaging. By analyzing the multiple images obtained through imaging, and analyzing vibrations at frequencies audible to humans, listening can be achieved; and By performing image analysis on the multiple images obtained from imaging, vibrations below the human audible frequency are detected, thus achieving sensation.

2. The system of claim 1, wherein: The system further includes a classifier configured to combine two or more features extracted from observation, listening, and sensation to classify the state of the vascular pathway.

3. The system of claim 1, wherein: Obtaining the shape of the body organ further includes calculating a three-dimensional shape of the patient's fistula.

4. The system of claim 1, wherein: The system further includes a classifier configured to classify the state of the vascular pathway based on two or more feature combinations generated from image analysis of the plurality of images.

5. The system of claim 4, wherein: The classifier is based on using machine learning to classify the state of the vascular pathway.

6. The system of claim 1, wherein: The system further includes a classifier configured to classify the patient's condition as either suitable for dialysis or at risk of stenosis.

7. The system of claim 2, wherein: The system is configured to record features measured during monitoring and predict the likelihood of narrowing based on changes along the monitoring period.

8. The system as described in claim 1, characterized in that: The lighting source is configured to switch between at least two of the following three projection light modes: uniform illumination, structured light illumination, and point illumination.

9. The system of claim 1, wherein: The camera is configured to capture images at a frame rate greater than 150 frames per second.

10. The system of claim 1, wherein: The camera is configured to capture images at a resolution lower than the camera's maximum resolution and at a frame rate exceeding 500 frames per second.

11. The system of claim 1, wherein: The lighting source and the camera are encapsulated in a container.

12. The system of claim 1, wherein: The lighting source includes a digital light processing projector.

13. The system of claim 1, wherein: The system includes a laser for illuminating the patient's body using a laser speckle interferometer.

14. The system of claim 8, wherein: The illumination source includes a light source with a near-infrared wavelength.

15. The system of claim 13, wherein: The system is configured to perform the laser speckle interferometer illumination based on a position of the shape of the patient's body obtained using image analysis.

16. The system of claim 1, wherein: The system is configured to calculate the emptying rate of a patient's fistula based on changes in the three-dimensional shape of the fistula.

17. The system of claim 16, wherein: The calculations were performed during the elevation test.

18. The system as claimed in claim 1, characterized in that: The system is configured to calculate one or more parameters related to vascular function based on the image analysis.

19. The system of claim 18, wherein: The system is configured to calculate an estimate of the probability of vascular function failure based on one or more of the parameters.

20. The system of claim 18, wherein: The system is configured to calculate the time when a blood vessel may fail to function based on one or more of the parameters.

21. The system of claim 18, wherein: The system is configured to calculate an estimate of the probability of vascular function failure based on the rate of change of one or more of the parameters.

22. The system of claim 18, wherein: The system is configured to calculate an estimate of the maturity of a vascular pathway based on the rate of change of one or more parameters.

23. The system of claim 18, wherein: The system is configured to calculate one or more parameters that indicate the development of one or more collateral vessels.

24. The system of claim 18, wherein: The system is configured to use one or more parameters based on the calculation of the count of multiple collateral vessels.

25. The system of claim 18, wherein: The system is configured to automatically detect collateral veins by counting the number of veins in a specific image region in different images taken at different times.

26. The system of claim 1, wherein: The system is configured to automatically detect the location of a fistula in at least one image using image analysis.

27. The system as claimed in claim 1, characterized in that: The system is configured to analyze vibrations of the patient's body by analyzing the light intensity at a specific location in an image of the patient's body containing blood vessels.

28. The system of claim 27, wherein: The system is configured to generate a vibration spectrum, which is generated by generating a light intensity vector at the specific location and by converting the intensity vector into a frequency vector.

29. The system of claim 28, wherein: The system is configured to analyze a vibration spectrum within a vibration frequency range.

30. The system of claim 29, wherein: The vibration spectrum is within the range of frequencies that are audible to humans.

31. The system of claim 30, wherein: The vibration spectrum extends into a frequency range below the human audible frequency range.

32. The system of claim 1, wherein: The vibrations of the patient's body were analyzed by analyzing images captured at a frame rate greater than 150 frames per second.

33. The system of claim 1, wherein: The vibrations of the patient's body were analyzed by analyzing images captured at a frame rate greater than 500 frames per second.

34. The system of claim 1, wherein: The vibrations of the patient's body are analyzed by examining selected pixels in the captured images.

35. The system of claim 1, wherein: The system is configured to measure a pulse parameter by detecting the pulse position in two images taken at different times and comparing the pulse positions in the two images.

36. The system of claim 1, configured to calculate a number of collateral vessels, wherein: The system is configured as follows: The camera is used to image a patient's body to obtain the geometry of blood vessels; and Calculate the number of collateral vessels.

37. The system of claim 36, wherein: Calculating the number of collateral vessels involves automatically detecting collateral veins by counting the number of veins in a specific image region in different images taken at different times.

38. The system as described in claim 36, characterized in that: The system is configured to automatically detect the location of a vascular access by detecting a junction of a vein and an artery.

39. The system of claim 1, wherein: The system is configured to compute one or more parameters based on passing the image through a trained neural network and using the descriptor layer of the trained neural network.

Images