WO2007001352A9 - Ultrasonic technique for assessing wall vibrations in stenosed blood vessels - Google Patents
Ultrasonic technique for assessing wall vibrations in stenosed blood vesselsInfo
- Publication number
- WO2007001352A9 WO2007001352A9 PCT/US2005/031135 US2005031135W WO2007001352A9 WO 2007001352 A9 WO2007001352 A9 WO 2007001352A9 US 2005031135 W US2005031135 W US 2005031135W WO 2007001352 A9 WO2007001352 A9 WO 2007001352A9
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- tissue
- vibration
- vibrations
- signal
- stenosis
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 84
- 210000004204 blood vessel Anatomy 0.000 title abstract description 21
- 238000002604 ultrasonography Methods 0.000 claims abstract description 179
- 208000031481 Pathologic Constriction Diseases 0.000 claims abstract description 130
- 238000012545 processing Methods 0.000 claims abstract description 37
- 230000017531 blood circulation Effects 0.000 claims abstract description 36
- 238000001845 vibrational spectrum Methods 0.000 claims abstract description 21
- 210000003484 anatomy Anatomy 0.000 claims abstract description 8
- 210000001519 tissue Anatomy 0.000 claims description 219
- 238000001228 spectrum Methods 0.000 claims description 151
- 208000037804 stenosis Diseases 0.000 claims description 100
- 230000036262 stenosis Effects 0.000 claims description 100
- 230000033001 locomotion Effects 0.000 claims description 81
- 238000000354 decomposition reaction Methods 0.000 claims description 39
- 239000011159 matrix material Substances 0.000 claims description 21
- 230000000747 cardiac effect Effects 0.000 claims description 15
- 238000001914 filtration Methods 0.000 claims description 15
- 230000006870 function Effects 0.000 claims description 15
- 230000010351 cardiac pulsation Effects 0.000 claims description 10
- 230000004044 response Effects 0.000 claims description 9
- 238000002156 mixing Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 6
- 210000004872 soft tissue Anatomy 0.000 claims description 4
- 230000000004 hemodynamic effect Effects 0.000 claims description 2
- 238000003384 imaging method Methods 0.000 abstract description 65
- 238000002592 echocardiography Methods 0.000 abstract description 42
- 238000006073 displacement reaction Methods 0.000 abstract description 25
- 230000003595 spectral effect Effects 0.000 description 40
- 238000001514 detection method Methods 0.000 description 26
- 230000000875 corresponding effect Effects 0.000 description 24
- 239000000523 sample Substances 0.000 description 24
- 230000035945 sensitivity Effects 0.000 description 22
- 210000001367 artery Anatomy 0.000 description 20
- 239000008280 blood Substances 0.000 description 17
- 229910052704 radon Inorganic materials 0.000 description 17
- SYUHGPGVQRZVTB-UHFFFAOYSA-N radon atom Chemical compound [Rn] SYUHGPGVQRZVTB-UHFFFAOYSA-N 0.000 description 17
- 210000004369 blood Anatomy 0.000 description 16
- 230000002123 temporal effect Effects 0.000 description 13
- 238000004088 simulation Methods 0.000 description 12
- 210000003462 vein Anatomy 0.000 description 12
- 230000001133 acceleration Effects 0.000 description 11
- 238000013459 approach Methods 0.000 description 11
- 239000000835 fiber Substances 0.000 description 9
- 238000001727 in vivo Methods 0.000 description 9
- 210000004351 coronary vessel Anatomy 0.000 description 8
- 208000029078 coronary artery disease Diseases 0.000 description 7
- 230000000541 pulsatile effect Effects 0.000 description 7
- 230000029058 respiratory gaseous exchange Effects 0.000 description 7
- 238000010183 spectrum analysis Methods 0.000 description 7
- 230000002861 ventricular Effects 0.000 description 7
- 210000001105 femoral artery Anatomy 0.000 description 6
- 230000000740 bleeding effect Effects 0.000 description 5
- 238000009556 duplex ultrasonography Methods 0.000 description 5
- 210000003191 femoral vein Anatomy 0.000 description 5
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 238000012360 testing method Methods 0.000 description 5
- 201000000057 Coronary Stenosis Diseases 0.000 description 4
- 230000007423 decrease Effects 0.000 description 4
- 230000001419 dependent effect Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 239000013307 optical fiber Substances 0.000 description 4
- 230000003534 oscillatory effect Effects 0.000 description 4
- 238000010200 validation analysis Methods 0.000 description 4
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 4
- 238000007792 addition Methods 0.000 description 3
- 238000002555 auscultation Methods 0.000 description 3
- 230000001427 coherent effect Effects 0.000 description 3
- 230000001934 delay Effects 0.000 description 3
- 238000003745 diagnosis Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 239000012530 fluid Substances 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 230000002107 myocardial effect Effects 0.000 description 3
- 230000010355 oscillation Effects 0.000 description 3
- 238000012216 screening Methods 0.000 description 3
- 238000012285 ultrasound imaging Methods 0.000 description 3
- 238000012800 visualization Methods 0.000 description 3
- 238000007476 Maximum Likelihood Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 210000001715 carotid artery Anatomy 0.000 description 2
- 208000006170 carotid stenosis Diseases 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000013016 damping Methods 0.000 description 2
- 238000013480 data collection Methods 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000003205 diastolic effect Effects 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000009472 formulation Methods 0.000 description 2
- 230000003902 lesion Effects 0.000 description 2
- 230000004807 localization Effects 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 230000001575 pathological effect Effects 0.000 description 2
- 208000030613 peripheral artery disease Diseases 0.000 description 2
- 238000009611 phonocardiography Methods 0.000 description 2
- 238000012805 post-processing Methods 0.000 description 2
- 238000011002 quantification Methods 0.000 description 2
- 230000002792 vascular Effects 0.000 description 2
- 206010002329 Aneurysm Diseases 0.000 description 1
- 206010007687 Carotid artery stenosis Diseases 0.000 description 1
- 206010008479 Chest Pain Diseases 0.000 description 1
- 206010011089 Coronary artery stenosis Diseases 0.000 description 1
- 208000009087 False Aneurysm Diseases 0.000 description 1
- 208000035211 Heart Murmurs Diseases 0.000 description 1
- 241000282412 Homo Species 0.000 description 1
- 235000019687 Lamb Nutrition 0.000 description 1
- 201000003099 Renovascular Hypertension Diseases 0.000 description 1
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000008602 contraction Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000001066 destructive effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 238000011049 filling Methods 0.000 description 1
- 238000005111 flow chemistry technique Methods 0.000 description 1
- 210000005003 heart tissue Anatomy 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 210000004165 myocardium Anatomy 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000010363 phase shift Effects 0.000 description 1
- 230000004962 physiological condition Effects 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 210000002027 skeletal muscle Anatomy 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000035488 systolic blood pressure Effects 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
- 230000008733 trauma Effects 0.000 description 1
- 230000006496 vascular abnormality Effects 0.000 description 1
- 208000019553 vascular disease Diseases 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/02007—Evaluating blood vessel condition, e.g. elasticity, compliance
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/02—Measuring pulse or heart rate
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/08—Detecting organic movements or changes, e.g. tumours, cysts, swellings
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S15/00—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems
- G01S15/88—Sonar systems specially adapted for specific applications
- G01S15/89—Sonar systems specially adapted for specific applications for mapping or imaging
- G01S15/8906—Short-range imaging systems; Acoustic microscope systems using pulse-echo techniques
- G01S15/8979—Combined Doppler and pulse-echo imaging systems
- G01S15/8981—Discriminating between fixed and moving objects or between objects moving at different speeds, e.g. wall clutter filter
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/52—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00
- G01S7/52017—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00 particularly adapted to short-range imaging
- G01S7/52023—Details of receivers
- G01S7/52025—Details of receivers for pulse systems
- G01S7/52026—Extracting wanted echo signals
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/52—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00
- G01S7/52017—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00 particularly adapted to short-range imaging
- G01S7/52023—Details of receivers
- G01S7/52034—Data rate converters
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/52—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00
- G01S7/52017—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00 particularly adapted to short-range imaging
- G01S7/52023—Details of receivers
- G01S7/52036—Details of receivers using analysis of echo signal for target characterisation
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/52—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00
- G01S7/52017—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00 particularly adapted to short-range imaging
- G01S7/52053—Display arrangements
- G01S7/52057—Cathode ray tube displays
- G01S7/52071—Multicolour displays; using colour coding; Optimising colour or information content in displays, e.g. parametric imaging
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7253—Details of waveform analysis characterised by using transforms
- A61B5/7257—Details of waveform analysis characterised by using transforms using Fourier transforms
Definitions
- vascular sounds associated with stenoses, aneurysms, arteriovenous fistulae and pseudoaneurysms are produced by the forces exerted on vessel walls by eddies produced when blood flows from a high-pressure region to a low-pressure region through a narrow orifice.
- the luminal area is reduced by a stenosis; therefore the flow velocity in the throat of the stenosis is increased.
- This local increase in the flow velocity creates a post-stenotic jet, if the post-stenotic expanse region is not sufficiently streamlined to prevent flow separation (i.e., the stenosis is not a "venturi tube").
- Regions of high fluid shear are produced due to the difference in velocity across the boundary of the jet. These shear forces produces eddies in the flow, which in turn produce a resistance to the laminar fluid flow.
- the presence of eddies cause fluctuations in the flow velocity and pressure in the post-stenotic region, which cause a corresponding motion in the vessel walls.
- the energy produced by the pressure drop across the stenosis is therefore dissipated through these mechanical vibrations of the vessel wall as well as minor heating of the blood.
- the local vibrations in the vessel wall and surrounding tissue manifest either as audible "bruits” and “murmurs” or palpable "thrills" when they reach the skin surface.
- the elasticity of the artery walls also introduces a capacitance into the circuits.
- the post-stenotic capacitance causes the pressure distal to the stenosis to rise during systole.
- the pressure drop across the stenosis is significant only during the peak systolic pressure phase, thus the Son lasts only during the systolic phase.
- the pressure drop distal to the stenosis is low, thus a significant pressure gradient exists across the stenosis even during the diastolic phase causing the Son to extend into the early diastolic phase.
- the power spectrum of the vibration exhibits a frequency peak called the "break frequency" that is directly related to the diameter of the orifice and the local flow velocity through the Strouhal number. Ia other words, the break frequency is inversely proportional to the residual lumen diameter at the stenosis.
- Phonoangiography and phonocardiography were developed to quantify the spectral content of bruits and murmurs recorded with a sensitive microphone, and carotid phonoangiography has been successfully used to estimate the degree of carotid artery stenosis in multiple clinical trials.
- auscultation and phonoangiography lack sensitivity and specificity because they are limited to diagnosing high-intensity vibrations that reach the skin surface, and the origin of the vibrations cannot be clearly resolved.
- Ultrasonic tissue Doppler imaging has been used for assessment of abnormal wall motion in the cardiac wall as well as in arteries.
- tissue vibrations from abnormal blood flow produce characteristic speckled artifacts in the surrounding tissue. These artifacts indicate tissue vibrations and are useful for recognizing stenoses. However, they are difficult to interpret and are not quantitative.
- tissue vibration detection and imaging modes for ultrasound instruments in which vibrations produced by stenosed blood vessels can be detected and color-coded according to their amplitude and frequency and overlaid on a B-mode and/or a color-flow image in real time.
- the tissue vibration-imaging mode might then be used for locating the origin of the vibration more precisely, relative to the patient's anatomy and/or for obtaining simultaneous information about vibrations and the underlying stenosis.
- the concepts disclosed herein were developed to detect, localize and quantify arterial stenoses by imaging tissue vibrations associated with such stenoses. Sounds caused by these vibrations (bruits and murmurs) are sometimes audible using a stethoscope, or palpable at the skin surface, and are indicative of various physiological conditions, including internal bleeding and arterial stenoses.
- the techniques disclosed herein employ algorithms that process an ensemble of received ultrasound echoes for detecting tissue vibrations, imaging tissue vibrations in a relatively large region of interest, and quantifying the hemodynamic properties of the stenosis based on the measured properties of the tissue vibrations.
- the algorithms disclosed herein have been implemented in a programmable ultrasound system to study the usefulness of tissue vibrations in real-time localization of stenoses in peripheral arteries and coronary arteries in humans.
- the vibration imaging algorithms described herein use an ensemble of 2D ultrasound data acquired during conventional ultrasonic imaging and the clutter signal (which is normally suppressed in conventional color-flow imaging) associated with such data, to detect and characterize tissue vibrations.
- Various signal processing algorithms have been developed that are suitable for this purpose, including three primary algorithms, based on parametric modeling of vibrations and the criteria to distinguish between clutter, blood flow, and vibrations.
- a first primary algorithm is based on phase decomposition
- a second primary algorithm is based on using an estimation of complex exponentials in noise
- a third primary algorithm is based on autoregressive modeling.
- Another set of algorithms utilize a larger ensemble of received ultrasound echoes (typically 64-512) from a small region of interest (e.g., a Doppler range gate) near the site of the stenosis to confirm the presence of tissue vibrations, and to measure the properties of the tissue vibrations with improved accuracy.
- a small region of interest e.g., a Doppler range gate
- Various signal processing algorithms have been developed that are suitable for this purpose, including a two-dimensional (2D) Fourier transform utilizing both the slow time variations in the received ultrasound echoes as a result of motion as well as variation in the motion of tissue at neighboring locations along the direction of the ultrasound beam.
- a first algorithm for identifying vibrations is based on the 2D Fourier transform of the quadrature-demodulated received echo and utilizes the Radon transform to identify spectral peaks corresponding to vibrations.
- a second algorithm for identifying vibrations utilizes a multi-frequency average to identify spectral peaks corresponding to vibrations, and suppress other sources of noise.
- a method for localizing and grading arterial stenoses using such algorithms is further disclosed herein.
- a new tissue vibration imaging mode for ultrasound instruments in which soft-tissue vibrations produced due to impact of blood flow eddies are detected and color-coded according to their amplitude and frequency, and overlaid on the B-mode and/or color-flow image in real time.
- the tissue vibration imaging mode can be used for locating the origin of vibration more precisely relative to the anatomy, and/or for obtaining simultaneous information about vibrations and the underlying blood flow.
- Real-time tissue vibration imaging has been implemented at frame rates, for example, of 10 frames/second, on an ultrasound system with a software-programmable signal and image processing back-end.
- the preliminary results confirm that vibrations produced as a result of arterial stenoses can be detected and imaged using such techniques.
- the vibration amplitude is expected to be the largest near a site downstream of the stenosis, and this fact can be used to localize a stenosis quickly and non-invasively.
- the strong backscattered ultrasonic echoes from tissue vibrations can improve detection of stenoses that are otherwise hard to detect using Doppler blood velocity-based methods due to weak scattering from blood. Potentially, this new tissue vibration imaging technology could be useful in a variety of devices and clinical settings.
- a low-cost portable screening device with tissue vibration detection functionality could be beneficially employed by general practitioners for diagnosing and/or screening patients with coronary and peripheral artery disease, or by paramedics and trauma centers to evaluate patients with chest pain.
- a tissue vibration imaging mode on high-end ultrasound systems can augment duplex ultrasound for enhanced diagnostic capability, which could be beneficially employed by imaging centers, cardiology clinics, and hospitals for diagnosing stenoses in patients.
- the detected tissue vibrations indicative of a stenosis could be presented as an audible signal in a manner recognizable to a person trained to listen to Sons using a stethoscope or as a palpable signal recognizable to a person trained to detect palpable thrills.
- One aspect of the concepts disclosed herein is directed to a method for detecting and localizing arterial stenoses using an ensemble of 2D ultrasound data by detecting and characterizing tissue vibrations caused by blood flow eddies downstream of a stenosis.
- the method includes the step of processing an ensemble of 2D ultrasound data to produce a tissue motion spectrum signal of a site being imaged.
- the tissue motion spectrum signal is then processed to produce a tissue vibration signal, from which any contribution to the tissue motion from a source other than vibrations at the stenosis has been substantially minimized.
- a vibration image is displayed using the tissue vibration signal and indicates a location of the stenosis at the site.
- One approach for processing the ensemble of 2D ultrasound data comprises the steps of estimating a correlation matrix from the ultrasound data, and carrying out an eigen decomposition of the correlation matrix to identify a signal subspace and a noise subspace. A frequency of the dominant vibration components in the signal subspace and the noise subspace is then estimated, and based upon that estimate, a vibration amplitude estimate and a vibration frequency estimate are determined. At least one of the vibration amplitude estimate and the vibration frequency estimate comprises the tissue vibration signal.
- a second approach for processing the ensemble of 2D ultrasound data for imaging vibrations associated with stenoses includes the step of computing the reflection coefficients of an autoregressive model of an ensemble of received ultrasound echoes. Linear prediction filter coefficients are computed from the reflection coefficients.
- a power spectrum is estimated, and the peaks in the power spectrum are detected. Based upon the estimate of the power spectrum and the peaks, a vibration amplitude estimate and a vibration frequency estimate are determined. At least one of the vibration amplitude estimate and the vibration frequency estimate again comprises the tissue vibration signal.
- a mean clutter velocity is estimated from the ultrasound data using autocorrelation.
- the ensemble of 2D ultrasound data is down-mixed with the mean clutter velocity, producing a down-mixed signal.
- a phase of the down-mixed signal and a mean phase of the down- mixed signal are determined, and the mean phase is subtracted from the phase of the down-mixed signal, producing a residual phase.
- the residual phase is then decomposed into its dominant components.
- the step of decomposing the residual phase preferably comprises the steps of estimating a correlation matrix from the residual phase, and performing an eigen decomposition of the correlation matrix to determine the dominant components.
- the step of filtering preferably comprises the step of filtering out clutter and noise at frequencies that are substantially lower than an expected frequency range of tissue vibrations corresponding to a stenosis at the site, and also preferably includes the step of filtering out noise that is at frequencies, which are substantially higher than an expected frequency range of tissue vibrations corresponding to the stenosis at the site.
- This step also encompasses differentiating between blood flow and tissue vibrations, by utilizing the statistical properties of the signals.
- the method can include the step of confirming that vibrations displayed in the vibration image correspond to a stenosis at the site by placing a Doppler sample volume at a location of the tissue vibration determined from the tissue vibration image, hi this step, the tissue vibration spectrum determined from relatively larger ensembles (typically, ensembles including 64-512 pulses) can be employed to confirm the tissue vibration detected from relatively smaller ensembles (typically, ensembles including 6-16 pulses) of 2D ultrasound data.
- relatively larger ensembles typically, ensembles including 64-512 pulses
- relatively smaller ensembles typically, ensembles including 6-16 pulses
- the step of displaying the vibration image preferably comprises the step of displaying at least one of a vibration amplitude image and a vibration frequency image of the site. Because of its efficiency, the method can include the step of displaying the vibration image in connection with an underlying anatomy of the site (i.e., the B-mode grayscale image), substantially in real time.
- the apparatus includes an ultrasound transducer for transmitting ultrasound pulses toward the internal site and receiving ultrasound data from scatterers at the internal site, including tissue that is vibrating due to a stenosis.
- a front-end system controls the ultrasound pulses produced by the ultrasound transducer and demodulates the echoes received by the ultrasound transducer, producing a signal having both in-phase and quadrature components.
- the apparatus also includes a back-end system to receive the signal from the front-end system and a tissue vibration processor.
- the front and back- end systems can optionally be combined into a single unit, or one or more parts of these systems can be operating remotely from other parts of the systems.
- the tissue vibration processor processes the ultrasound signal to estimate tissue vibrations caused by a stenosis, producing a tissue vibration signal.
- the tissue vibration signal is converted to an image signal by the back-end system.
- a display is coupled to the back-end system to receive the image signal, to display a tissue vibration image in which a stenosis at the internal site is indicated.
- the display could be remote from the tissue vibration detection and identification apparatus.
- the display can be physically located in a hospital, while the tissue vibration detection and identification apparatus is physically located at another location, such as in an ambulance carrying a patient on which the apparatus is being used.
- the result of tissue vibration detection and identification can be presented as an audible or a palpable output indicating tissue vibrations.
- the tissue vibration signal can also be interpreted by an automated algorithm to indicate a stenosis, and the result of the automated interpretation can be presented as an electronic readout.
- the functions performed by the apparatus are consistent with the steps of the method described above.
- FIGURE 1 is a functional block diagram of an exemplary ultrasound system that is suitable for carrying out tissue vibration imaging
- FIGURE 2 is a graph of an expected Doppler spectrum (i.e., frequency vs. power) from a vibrating sample volume;
- FIGURE 3A is a flow chart showing the logical steps of an exemplary s ⁇ bspace-based algorithm for creating a vibrating tissue image in which a stenosis is evident
- FIGURE 3B is a flow chart showing the logical steps of an alternative exemplary algorithm that uses autoregression for creating a vibrating tissue image in which a stenosis is evident;
- FIGURE 3 C is a flow chart showing the logical steps of yet another exemplary alternative algorithm that uses phase decomposition for creating a vibrating tissue image in which a stenosis is evident;
- FIGURE 4 is a flow chart showing the logical steps of a method for localizing and grading arterial stenoses using a vibrating tissue image generated using ultrasound;
- FIGURE 5 A graphically illustrates both a 2D Fast Fourier Transform (FFT) spectrum and a conventional Doppler spectrum of simulated received ultrasound echoes when scatterers responsible for the echoes are moving with constant velocity of 20 mm/s, the scatterer motion being along the axis of the ultrasound beam, with the Doppler spectrum being disposed to the left of the vertical axis, and the 2D FFT spectrum being disposed on the right of the vertical axis;
- FFT Fast Fourier Transform
- FIGURE 5B graphically illustrates both a 2D FFT spectrum and a conventional Doppler spectrum of simulated received ultrasound echoes when scatterers responsible for the echoes are moving with a constant velocity of 20 mm/s along the axis of the ultrasound beam, and are also vibrating with a frequency of 300 Hz and an amplitude of 5 ⁇ m, with the Doppler spectrum being disposed to the left of the vertical axis, and the 2D FFT spectrum being disposed on the right of the vertical axis;
- FIGURE 6A graphically illustrates both a 2D FFT spectrum and a conventional Doppler spectrum of received ultrasound echoes (where the center frequency of the transmitted ultrasound signal is 5 MHz) when scatterers responsible for the echoes are moving with a constant velocity of 200 mm/s along the axis of the ultrasound beam, and also vibrating with a frequency 300 Hz and an amplitude of 5 ⁇ m, with the Doppler spectrum being disposed to the left of the vertical axis, and the 2D FFT spectrum being disposed on the right of the vertical axis;
- FIGURE 6B graphically illustrates both a 2D FFT spectrum and a conventional Doppler spectrum of received ultrasound echoes (where the center frequency of the transmitted ultrasound signal is 2 MHz) when scatterers responsible for the echoes are moving with a constant velocity of 200 mm/s, and also vibrating with a frequency of 300 Hz and an amplitude of 5 Dm, with the Doppler spectrum being disposed to the left of the vertical axis, and the 2D FFT spectrum being disposed on the right of the vertical axis;
- FIGURE 7A graphically illustrates both a 2D FFT spectrum and a conventional Doppler spectrum of simulated received ultrasound echoes from a group of scatterers accelerating from 10 mm/s to 30 mm/s, with an acceleration of 5 m/s 2 , with the Doppler spectrum being disposed to the left of the vertical axis, and the 2D FFT spectrum being disposed on the right of the vertical axis;
- FIGURE 7B graphically illustrates both a 2D FFT spectrum and a conventional Doppler spectrum of simulated received ultrasound echoes from a group of scatterers accelerating from 10 mm/s to 30 mm/s, with an acceleration of 5 m/s 2 , when the scatterers are also vibrating with a frequency of 300 Hz and an amplitude of 5 ⁇ m, with the Doppler spectrum being disposed to the left of the vertical axis, and the 2D FFT spectrum being disposed on the right of the vertical axis;
- FIGURE 8A graphically illustrates both a 2D FFT spectrum and a conventional Doppler spectrum of simulated received ultrasound echoes when scatterers responsible for the echoes are moving with a constant velocity of 20 mm/s, the scatterer motion being perpendicular to the axis of the ultrasound beam, with the Doppler spectrum being disposed to the left of the vertical axis, and the 2D FFT spectrum being disposed on the right of the vertical axis;
- FIGURE 8B graphically illustrates both a 2D FFT spectrum and a conventional Doppler spectrum of simulated received ultrasound echoes when scatterers responsible for the echoes are moving with a constant velocity of 20 mm/s, the scatterer motion being perpendicular to the axis of the ultrasound beam, where the scatterers are also vibrating with a frequency of 300 Hz and an amplitude of 5 ⁇ m with the Doppler spectrum being disposed to the left of the vertical axis, and the 2D FFT spectrum being disposed on the right of the vertical axis;
- FIGURE 9A graphically illustrates a Radon transform of the 2D FFT spectrum of FIGURE 5A;
- FIGURE 9B graphically illustrates a Radon transform of the 2D FFT spectrum of FIGURE 5B;
- FIGURE 9C graphically illustrates a spectral estimate computed from the Radon transform of FIGURE 9A;
- FIGURE 9D graphically illustrates a Doppler spectral estimate computed from the Radon transform of FIGURE 9B;
- FIGURE 1OA graphically illustrates a multifrequency spectral estimate computed from the 2D FFT spectrum of FIGURE 5A
- FIGURE 1OB graphically illustrates a multifrequency spectral estimate computed from the 2D FFT spectrum of FIGURE 5B;
- FIGURE 1 IA schematically illustrates a 3-D simulation model of scatterer distributions used to validate the use of ultrasound to image vibrations associated with a stenosis, as described herein;
- FIGURE 1 IB graphically illustrates modeled clutter motion with vibrations indicated by a boxed-in region
- FIGURE 11C graphically illustrates a modeled blood flow profile
- FIGURE 12A is a vibration amplitude image of a simulation model used to validate the vibration imaging techniques disclosed herein;
- FIGURE 12B schematically illustrates masks used for computing the sensitivity and specificity of vibration detection
- FIGURE 13A graphically illustrates sensitivity versus threshold curves for phase-decomposition-based vibration detection
- FIGURE 13B graphically illustrates specificity versus threshold curves for phase-decomposition-based vibration detection
- FIGURE 13C graphically illustrates receiver-operating characteristic curves for phase-decomposition-based vibration detection
- FIGURE 13D graphically illustrates sensitivity versus threshold curves for root-MUSIC-based vibration detection
- FIGURE 13E graphically illustrates specificity versus threshold curves for root-MUSIC-based vibration detection
- FIGURE 13F graphically illustrates exemplary receiver-operating characteristic curves for root-MUSIC-based vibration detection
- FIGURE 14 graphically illustrates the robustness of sensitivity to increasing vibration bandwidth for the phase-decomposition and root-MUSIC algorithms disclosed herein;
- FIGURE 15 schematically illustrates an experimental setup in which an ultrasound probe is used to image the vibration of a plate, such a setup having been used to generate empirical evidence in support of using ultrasound to image tissue vibrations, as disclosed herein;
- FIGURE 16A is a B-mode ultrasound image of the vibration phantom of FIGURE 15, overlaid with vibration amplitude;
- FIGURE 16B is a B-mode ultrasound image of the vibration phantom of FIGURE 15, overlaid with vibration frequency;
- FIGURE 16C graphically illustrates a MUSIC pseudo-spectrum of I-Q ensemble data extracted from a portion of the ultrasound image of FIGURE 16B;
- FIGURE 17A graphically illustrates a vibration amplitude of the plate from the experimental setup of FIGURE 15 obtained using both ultrasound and an optical fiber micrometer;
- FIGURE 17B graphically illustrates a vibration frequency of the plate from the experimental setup of FIGURE 15 obtained using both ultrasound and an optical fiber micrometer;
- FIGURE 17C graphically illustrates the differences between the vibration amplitude of the plate from the experimental setup of FIGURE 15 obtained using both ultrasound and an optical fiber micrometer;
- FIGURE 17D graphically illustrates the differences between the vibration frequency of the plate from the experimental setup of FIGURE 15 obtained using both ultrasound and an optical fiber micrometer
- FIGURE 18 schematically illustrates an experimental setup in which an ultrasound probe is used to image a stenosis in an ex vivo artery, such a setup having been used to generate empirical evidence in support of using ultrasound to image tissue vibrations associated with a stenosis, as disclosed herein;
- FIGURE 19 is an image of the flow in the ex vivo artery of FIGURE 18;
- FIGURES 2OA and 2OB are respectively vibration amplitude images of an ex vivo artery obtained using the experimental setup of FIGURE 18, with FIGURE 2OA representing a simulated stenosis of 30%, and FIGURE 2OB representing a simulated stenosis of 42%;
- FIGURE 21 graphically depicts vibration spectra from a stenosis simulated using the experimental setup of FIGURE 18, the vibration spectra being generated using both pulsed-wave Doppler ultrasound and a fiber optic micrometer;
- FIGURE 22A is a Color-Doppler image from a stenosed vein graft in a human subject
- FIGURE 22B is a vibration amplitude image of the stenosed vein graft of FIGURE 9A;
- FIGURE 23 A is a color power ultrasound image of a stenosed femoral vein graft including perivascular artifacts
- FIGURE 23B is a vibration amplitude image of the stenosed femoral vein graft of FIGURE 23 A
- FIGURE 23C is a vibration frequency image of the stenosed femoral vein graft of FIGURE 23A;
- FIGURE 24A graphically illustrates instantaneous vessel wall position estimated using phase decomposition of pulsed wave Doppler data from a stenosed femoral vein graft
- FIGURE 24B is a motion periodogram of the signal used to generate FIGURE 24A;
- FIGURE 24C graphically illustrates a cross-sectional profile of the spectrum of FIGURE 24B at a particular point in time
- FIGURE 24D graphically illustrates a motion pseudo-spectrum computed using the MUSIC algorithm for 10 ensembles of color-flow ultrasound data at the same location
- FIGURE 25 A graphically illustrates an arterial wall displacement spectra of a normal femoral artery obtained in vivo using the techniques disclosed herein;
- FIGURE 25B graphically illustrates an arterial wall displacement spectra of a stenosed femoral bypass vein graft obtained in vivo using the techniques disclosed herein;
- FIGURE 25 C graphically illustrates an arterial wall displacement spectra of a different stenosis present in the same patient as the stenosis represented in FIGURE 25B;
- FIGURE 25D graphically illustrates an arterial wall displacement spectra of a stenosis present in another patient
- FIGURE 26 A graphically illustrates a time- varying wall vibration spectra of a normal artery obtained using the techniques disclosed herein;
- FIGURES 26B-D graphically illustrate time-varying wall vibration spectra of stenosed blood vessels obtained using the techniques disclosed herein;
- FIGURE 27A is a Doppler spectrum computed using a 2D FFT method from an ultrasound image of the myocardium of a patient who has coronary artery disease, symmetric double-sided peaks being indicative of vibrations observed in the late ventricular ej ection phase;
- FIGURE 27B graphically illustrates a time course of a wall velocity during ventricular ejection, a boxed-in region indicating high-frequency vibrations that appear include harmonic components;
- FIGURE 28A is an angiographic image of a patient who has coronary artery disease, acquired in the left anterior oblique proj ection with caudal angulation
- FIGURE 28B is a vibration amplitude image overlaid on an apical two- chamber view of the patient of FIGURE 28A;
- FIGURE 29 A is an angiographic image of a patient who has coronary artery disease, acquired in right anterior oblique projection with cranial angulation; and FIGURE 29B is a vibration amplitude image overlaid on the apical two- chamber view of the patient of FIGURE 29A.
- FIGURE 1 is a block diagram illustrating an ultrasound system 10, which is generally similar to a conventional ultrasound system, but which has been modified to include tissue vibration imaging and is thus usable in practicing the concepts disclosed herein.
- Ultrasound system 10 includes an ultrasound transducer 12 that transmits a signal, which is modulated with a carrier frequency, typically 1 MHz - 15 MHz, using multiple cycles (i.e., 2-20 cycles). The transmitted signal is reflected by scatterers (not shown) along the beam path and is received after a time delay, the duration of which depends upon the distance of the scatter from the transducer.
- the acoustic echoes received from the tissue are converted to electrical signals by the transducer, and these signals are digitized by analog-to-digital converters (not separately shown).
- a front-end subsystem 14 includes a beam former 16 that performs dynamic focusing, apodization, and steering of both transmitted and received ultrasonic waveforms. Also included in front-end system 14 are a time-gain-compensation (TGC) circuit 18 that amplifies signals with a variable gain that is proportional to the depth within tissue, and a radio frequency (RF) demodulator and decimator 20 that digitally removes the high frequency carrier by quadrature demodulation and decimation, providing both in-phase (T) and quadrature (Q) samples, which may be represented as a complex quantity, I(t) + jQ(t).
- TGC time-gain-compensation
- RF radio frequency
- the acquired quadrature ensemble (or color-flow) data are then processed in a back-end subsystem 22, depending on the one (or more) ultrasound mode(s) that is/are selected, e.g., B-mode, color-flow mode, tissue vibration mode, and Doppler mode.
- ultrasound mode(s) e.g., B-mode, color-flow mode, tissue vibration mode, and Doppler mode.
- the signal of interest is the envelope of I(t) + jQ(t).
- the time delay introduced by the scatterers is reflected in the phase of the complex quantity I(t) + jQ(t).
- the phase of the complex received signal provides an estimate of the instantaneous position of the scatterer. By monitoring the change of phase over time, the displacement and velocity of the scatterer can be estimated.
- multiple pulses (commonly from 6 to 16 pulses) are transmitted and received along each scan line at a rate known as the pulse repetition frequency (PRF).
- PRF pulse repetition frequency
- a collection of received temporal samples from each spatial location is thus called an "ensemble.”
- a color-flow processor 26 estimates the blood flow velocity from the ensemble of data by estimating the phase difference between the adjacent temporal samples, typically using an autocorrelation algorithm.
- a 2D image is created by acquiring multiple samples from different spatial locations.
- Doppler mode which is implemented with a Doppler processor 30, scanning is performed along a single scan line, and a spectrum of the blood velocity from a single spatial location is estimated from a substantially larger ensemble of data (typically, data from 64-512 pulses).
- scan conversion is performed by a scan converter circuit 32, which converts the acquired ultrasound data from polar coordinates to the Cartesian coordinates used by the raster display.
- Post processing may optionally be applied by a post-processing circuit 34, to improve the quality of the displayed image, as well as to combine the anatomy and flow images on the display.
- a tissue vibration processor 28 that is used to process the ultrasound data in one exemplary embodiment is shown in FIGURE 1.
- the quadrature data ensembles are input to the tissue vibration processor.
- the tissue vibration processor estimates the instantaneous displacement of the scattering tissue from the phase of the complex received signal.
- This tissue motion is referred to as clutter in conventional color-flow imaging and is suppressed using clutter filters.
- cardiac pulsation, respiration and transducer motion each can contribute to an observed displacement or motion of tissue. Such motion is at a low frequency of a few Hertz or less.
- the tissue surrounding the stenosis vibrates locally with a frequency ranging from a few tens of Hertz to more than a 1000 Hertz.
- the tissue vibration processor performs this analysis by decomposing the tissue motion into the dominant motion components and identifying any motion components that appear to be at a frequency higher than that of cardiac pulsation .
- tissue vibration processor 28 can be implemented as an additional fixed-function circuit board or an application specific integrated circuit (ASIC) for use in conventional ultrasound machines.
- the tissue vibration processor can be combined with color-flow processor 26, since both process the same data ensemble.
- a standalone tissue vibration imaging device can be implemented with front-end subsystem 14, B-mode processor 24, tissue vibration processor 28, and scan converter 32.
- DSPs digital signal processors
- the tissue vibration processor can be implemented in software/hardware using one or more digital signal processors (DSPs) or alternatively, in an ASIC, or even on a conventional general purpose processor chip that accesses machine language instructions stored in a memory accessed by the processor to carry out the processing steps of the tissue vibration processor.
- the velocity of blood flow is estimated by computing the average phase difference between multiple ultrasound echoes (typically 6-16 pulses) that are received from a sample volume. Echoes backscattered from moving tissue tend to have a significantly higher signal strength (typically 40 dB-60 dB higher), compared to the weak scattering from blood, and also have lower velocities. This high amplitude and low frequency tissue signal is commonly referred to as clutter and tends to bias the estimated blood flow velocity. Thus, clutter is suppressed using appropriate filters in conventional color flow imaging. The main components of clutter are cardiac pulsation, respiration, and transducer movement.
- any local tissue vibrations e.g., those caused by the blood flow eddies in stenosed blood vessels, will also be part of this clutter and would normally be suppressed in conventional ultrasound processing systems.
- the tissue vibrations are separated from the remaining clutter and flow signals. In achieving this function, it was recognized that the tissue vibrations and clutter produce statistically independent signals that have different frequency content. While clutter due to cardiac pulsation and breathing typically occurs at 1 Hz or less, tissue vibrations typically occur at 50 Hz or more. Other noise sources are at substantially higher frequencies. Scattering from tissue is typically more coherent compared to the scattering from blood, because the tissue scatterers are more tightly bound together and tend to move as a group.
- the blood flow signal typically has a much greater frequency bandwidth. Due to its weak signal strength and greater bandwidth, blood flow signals may be considered as noise compared to the stronger and more coherent tissue vibration signals for purposes of this approach. Therefore, tissue vibrations can be distinguished from clutter and blood flow based on spectral analysis. Spectral analysis of the phase of the received ultrasound echo can be used to separate the components of the scatterer motion, ignoring the scattered signal strength, whereas spectral analysis of the complex ultrasound echo considers both the signal strength and the motion components.
- eigen decomposition-based spectral estimation which models the signal as an optimum set of orthogonal components
- autoregressive spectral estimation which models the signal as the output of an autoregressive linear prediction filter driven by white Gaussian noise.
- three signal processing algorithms were developed for isolating tissue vibrations associated with stenoses (two based on eigen decomposition and one based on autoregression).
- the first algorithm is based on an eigen decomposition-based spectral analysis of the phase of the received ultrasound echo; the second algorithm is based on an eigen decomposition-based spectral analysis of the complex ultrasound echo; and, the third algorithm is based on an autoregressive spectral analysis of the complex ultrasound echo. Since eigen decomposition is a computationally-intensive operation, an approximate eigen decomposition utilizing iterative QR factorization is used as a computationally-efficient algorithm.
- ⁇ d r (t),d ⁇ (t),d ⁇ i) denote the displacement as a function of time. If the scattering from the sample volume is uniform with ⁇ as the average scattering coefficient, then the scattering function of the sample volume is ⁇ v(r, i) .
- the complex received signal from the sample volume, y( ⁇ ,t) can then be modeled as a convolution of the pulse echo spatial impulse response, h pe (r, ⁇ ) , of a single point scatterer, the temporal response of the transducer, x( ⁇ ), and the scattering function, ⁇ v(f,t) .
- y( ⁇ , t) h pe (r, ⁇ )* x( ⁇ )* ⁇ v(r, t)+n( ⁇ , t)
- A( ⁇ ) is the complex amplitude of the scattered signal. It is apparent that the complex received signal is phase modulated with the instantaneous radial displacement.
- the tissue displacement due to cardiac pulsation, breathing, and other tissue movement (i.e., clutter or noise) relative to the transducer is d tiss (f).
- This motion will hereinafter be referred to as the "clutter motion.”
- T PRF the pulse repetition interval.
- the Fourier transform of the phase-modulated complex received signal is a Bessel series:
- FIGURE 2 illustrates a typical power spectrum 40 of the ultrasound signal when a tissue vibration is present.
- the spectrum includes multiple copies of the clutter spectrum separated by the vibration frequency, as indicated by Eq. (8), which is presented below.
- a low frequency peak 42 at f t ⁇ ss corresponds to the clutter spectrum, while symmetric peaks /, and f + indicated respectively by reference numbers 44 and 46 correspond to vibration, andp tiss ,p + , andj!?., are the corresponding peak powers.
- the frequency peaks at f + and/ are referred to herein as a "matching pair.”
- higher-order terms can be ignored; thus, most of the spectral energy will be present in the three frequency peaks, /, f t ⁇ ss , and f + , respectively. Since ⁇ ) « £- , the ratio of the power in the frequency peaks can
- the vibration frequency and amplitude may be estimated from the power spectrum as follows:
- the vibration frequency and amplitude may be estimated from the residual phase ⁇ (k) ⁇ of the ultrasound signal after suppressing the effects of clutter motion.
- a coarse computationally-efficient estimate of the frequency of the dominant components, f vlb can be obtained by counting the zero crossings, ⁇ zero , in the residual phase. This estimate can be further refined by interpolating the residual phase to compute the mean period of oscillation.
- the vibration amplitude may be estimated by the variance of the residual phase.
- zero-crossing frequency estimator and are respectively referred to herein as the "zero-crossing frequency estimator” and the “phase variance amplitude estimator.”
- One method of producing the complex ultrasound signal in Eq. (4) is a quadrature demodulation of the received ultrasound signal.
- An alternative method is to compute the time delays producing the phase variations in Eq. (4) by processing the received RF ultrasound data using a cross correlation technique. Vibration Imaging Using Estimation of Complex Exponentials in Noise Using the inverse Fourier transform of the Bessel expansion in Eq. (7),
- the ultrasound signal can be modeled as a sum of complex exponentials embedded in noise.
- vibrations correspond to matching pairs of complex exponentials.
- the complex exponentials corresponding to clutter motion will typically not have such matching pairs of frequencies.
- the frequencies (f tm , f + , f) may be estimated using the root-MUSIC and ESPRIT algorithms (disclosed by P. Stoica and R. Moses in "Introduction to Spectral Analysis," Upper Saddle River, NJ: Prentice-Hall, 1997).
- Vibrations may then be detected using a matching peak criterion
- the steps of the algorithm are described in more detail below, in regard to FIGURE 3A. Based on this criterion, vibrations can be detected and distinguished from clutter motion. Any blood flow signals maybe considered as part of the noise spectrum.
- FIGURE 3A illustrates a flow chart 50 that shows the logical steps involved in a first algorithm for estimating the tissue vibrations based upon a pair of complex exponentials in clutter or noise, which are normally excluded from color-flow processing.
- the procedure begins with a quadrature-demodulated ensemble of 2D ultrasound data 52.
- a step 53 low frequency clutter due to cardiac pulsation, pulsatile blood vessel wall motion, and respiration is suppressed.
- this step may include down mixing with the mean clutter velocity estimated using the autocorrelation method.
- the low frequency motion may be suppressed by filtering.
- the motion of the surrounding tissue may be used to form an estimate of the low frequency tissue motion by analyzing the principal motion components.
- a correlation matrix is estimated from the color-flow data.
- the correlation matrix is employed to carry out an eigen decomposition, producing a signal subspace 58 and a noise subspace 60.
- the frequency of the dominant components is estimated in a step 62, by employing the root-MUSIC and ESPRIT algorithms, as noted above.
- a decision step 64 determines if each dominant component is double-sided, while a step 66 estimates the vibration amplitude and frequency of each dominant component using Eq. (8).
- decision step 64 returns a "zero," while if the dominant component is double-sided, the decision step return a one.
- a multiplier 68 then multiplies the output of decision step 64 by the vibration amplitude and frequency estimate for the dominant component, yielding a null if the dominant component is not a tissue vibration component, and otherwise returning the estimate of vibration amplitude and frequency of the dominant component.
- a median filter 70 then filters isolated falsely-detected vibrations and other undesired noise from the results, so that the remaining vibration image indicating a stenosis site is displayed in a step 72.
- Vibration Imaging Using an Autoregressive Signal Model. The ultrasound signal from vibrations can be modeled as the output of a
- linear prediction coefficients can be computed using either a least-squares minimization of the prediction errors or using the computationally-efficient
- the presence of symmetric matching pairs of frequency peaks in the power spectrum around the clutter motion peak may be detected as a vibration.
- any flow signals may be regarded as noise.
- the second alternative algorithm also begins with quadrature-demodulated ensemble data set 52.
- step 53 low frequency clutter due to cardiac pulsation, pulsatile blood vessel wall motion, and respiration is suppressed.
- this step may include down mixing with the mean clutter velocity estimated using the autocorrelation method.
- the low frequency motion may be suppressed by filtering.
- the motion of the surrounding tissue may be used to form an estimate of the low frequency tissue motion by analyzing the principal motion components.
- a step 82 reflection coefficients are computed for each ensemble of the quadrature-demodulated data.
- linear prediction coefficients are determined in a step 84.
- decision step 64 determines if the peaks thus identified are for tissue vibration by determining if they are double-sided and returning a zero if not, and a one, if so.
- step 66 provides for estimating the vibration amplitude and frequency at each of these peaks, and the results from decision step 64 are multiplied by the estimated amplitude and frequency in multiplier 68.
- Median filter 70 is then applied to the results, and the filtered image data are displayed as a vibration image, in step 72.
- a third algorithm for detection and imaging of vibrations can be based on the phase ultrasound signal.
- vibrations will produce an oscillatory signature in the phase, which will typically not be present in the case of clutter motion.
- flow signals may have oscillatory phase
- the echoes from vibrating tissue are expected to be more coherent than those from flow.
- their phase may be modeled by a smaller number of dominant components.
- a vibration detection algorithm can also be based on decomposition of the phase of the ultrasound signal into its dominant components and testing for oscillatory phase.
- the phase can be estimated from RF ultrasound data by estimating the time delays between a pair of RF ultrasound data.
- Any linear time-varying motion is first suppressed by down mixing the ensemble of 2D ultrasound data with the mean clutter velocity, estimated using the conventional autocorrelation method.
- the phase of the ensemble of 2D ultrasound data is then computed, and the mean phase is subtracted to suppress the effect of the stationary echo.
- the residual phase is then decomposed into its dominant components using a method similar to principal component analysis.
- the first step of the decomposition involves the estimation of the correlation matrix of the residual phase using the modified covariance method (Marple, 1987).
- An approximate eigen decomposition can then be performed using iterative QR factorization of the correlation matrix.
- the approximate eigen values, ⁇ i may be estimated by the diagonal elements of the upper triangular matrix Rk after the ⁇ * iteration.
- the eigenvectors are arranged in order of decreasing eigen values.
- the eigen values are a measure of the signal energy contributed by the corresponding eigenvector.
- the fraction of the total signal energy contained in the p dominant components
- F t hreshoi d is chosen so that at least one half of one period of the vibration is contained in an ensemble.
- the vibration frequency and amplitude may be estimated using Eq. (10). The steps of this algorithm are described in more detail below, in connection with FIGURE 3 C. While developed to image vibrations associated with internal bleeding, this exemplary algorithm has also been shown to be effective in imaging vibrations associated with stenosed blood vessels, as is discussed in detail.
- a flow chart 90 illustrates the logical steps of the third algorithm.
- step 53 suppresses low frequency clutter due to cardiac pulsation, pulsatile blood vessel wall motion, and respiration.
- this step may include down mixing with the mean clutter velocity estimated using the autocorrelation method.
- the low frequency motion may be suppressed by filtering.
- the motion of the surrounding tissue may be used to form an estimate of the low frequency tissue motion by analyzing the principal motion components.
- a step 94 provides for computing an unwrapped phase of the quadrature ensemble or color flow data, and then subtracting the mean clutter velocity from the unwrapped phase, resulting in a residual phase.
- the phase can be determined from RF ultrasound data by estimating time delays between a pair of RF ultrasound data.
- a step 96 estimates a correlation matrix, which is then used to carry out a QR factorization in a step 98, yielding an eigen value estimate 100, an eigenvector estimate 102, and a vibration amplitude and frequency estimate 104, which are determined using Eq. (10), as noted above.
- a decision step 106 determines if the total energy contained in the;? dominant component is greater that a predefined threshold, T. If so, decision step 106 returns a zero if not, and a one if so. Similarly, a decision step 108 determines if the estimate eigenvector has a frequency that is greater than a predefined threshold, F. If so, decision step 108 returns a one, and if not, a zero.
- FIGURE 4 illustrates a flow chart 120 that shows the logical steps involved in using ultrasound to localize and quantify arterial stenoses.
- a blood vessel is imaged using color-Doppler ultrasound.
- PRF pulse repetition frequency
- a vibration amplitude image is generated in real time, generally as described above.
- a Doppler sample volume is placed at the Son proximate the maximum vibration amplitude.
- a vibration spectrum is generated from the pulsed-wave Doppler data (again, using the techniques described above).
- the "break" frequency of the vibration spectrum is noted, while in a step 136 the stenosis is quantified. Detectable Vibration Amplitudes and Frequencies
- tissue vibrations with a peak amplitude of about 1 ⁇ m have been accurately detected.
- the minimum detectable vibration amplitude depends upon the noise level and dynamic range of the phase of the received ultrasound echo.
- the phase can have a dynamic range of 96 dB or more (for 16-bit quadrature-demodulated data) and the signal typically exceeds the electronic and thermal noise level by 80 dB or more. Therefore, from Eq. (4), vibrations as small as 50 nm may theoretically be detected using a 5 MHz ultrasound transducer. Practically, the attenuation of the ultrasound signal will reduce the dynamic range and limit the minimum detectable amplitude in deep tissue to -0.5 ⁇ m.
- the detectable vibration frequencies depend upon the choice of PRF, i.e., on Fp RF .
- a PRF that is too low compared to the vibration frequency would lead to aliasing, while selecting a PRF that is too high will fail to detect low-frequency vibrations.
- a vibration can be detected only if at least half of one vibration cycle is captured within the temporal window corresponding to an ensemble.
- the maximum detectable frequency is — ⁇ SL.
- E vibration are included in an ensemble.
- a simulation and phantom experiments that were carried out indicate that reliable detection may be performed using only one half to six vibration periods during the interrogation period. For example, with a PRF of 1 kHz and an ensemble size of 16 periods/pulses, vibrations with frequency between 31.3 Hz and 375 Hz maybe reliably detected. Quantification of Residual Lumen Diameter
- the Strouhal number (S) relates the break frequency of turbulent fluctuations (f vit ) Xo the length scale of the turbulence (the residual lumen diameter at the stenosis (D) and the mean downstream blood velocity in the unobstructed vessel (U)) according to:
- the flow velocity can be estimated using pulsed-wave Doppler.
- the break frequency can be then used to quantify the residual lumen diameter at the stenosis, assuming the Strouhal number remains constant at a value of l.
- the ability to directly measure the amplitude of the vibrations enables a stenosis to be graded.
- the energy in the eddies (E) and thus, the amplitude of the tissue vibrations (a v ») > is directly proportional to the flow rate, as follows:
- Vibrations are transient, with typical durations of lO ms-lOO ms. Thus, there is a possibility that some vibrations may not be interrogated. Since the vibrations typically have a relatively large spatial extent and repeat every cardiac cycle, it is unlikely that the vibrations will be missed entirely; however, the spatial extent of the vibrations visible in the image may be only a part of the true spatial extent. By appropriately choosing the PRF and the region of interest, such discrepancies may be minimized. Other artifacts may be falsely detected as vibrations.
- Transducer motion may introduce additional frequency peaks in the clutter spectrum and may cause false detections; however, using a trained sonographer to perform the scanning may minimize these false detections.
- Vibrations in the tensed skeletal muscle of the sonographer, and any ambient vibrations may be detected in the vibration image.
- the high- resolution spectral estimation methods may produce spurious peaks that can be falsely detected as vibrations.
- Such artifacts can be easily distinguished from pathological vibrations, which are expected to be correlated with the anatomy and periodic with every cardiac cycle. These artifacts can be also avoided if additional temporal samples are available. Any vibrations displayed in the vibration image should therefore be confirmed with the vibration spectrum by placing a Doppler sample volume at the location of the peak intensity.
- the computational requirements of the proposed algorithms are shown below in Table 2.
- the subspace-based algorithms (MUSIC/ESPRIT) have a computational requirement that is highly dependent on the choice of model order.
- eigen decomposition is the most computationally-intensive task.
- the autoregression-based algorithm is less computationally intensive, and the computational requirement is less dependent on the model order.
- the computation of the FFT for spectral estimation is the most computationally-intensive task.
- the phase decomposition method is the least computationally intensive, since it involves operations on real signals only. Thus, the phase-decomposition algorithm is most suitable for real-time implementation.
- FIGURE 5A graphically illustrates both a 2D Fourier (2D FFT) spectrum and a conventional Doppler spectrum of simulated received ultrasound echoes when scatterers responsible for the echoes are moving with constant velocity of 20 mm/s, with the Doppler spectrum being disposed to the left of the vertical axis, and the 2D FFT spectrum being disposed on the right of the vertical axis.
- the motion of the scatterers represented here is axially along a direction of an ultrasound beam. Note the spectrum of the received pulse lies on a line that passes through the origin with slope:
- FIGURE 5B graphically illustrates both a 2D Fourier (2D FFT) spectrum and a conventional Doppler spectrum of simulated received ultrasound echoes when scatterers responsible for the echoes are moving with constant velocity of 20 mm/s and also vibrating with a frequency 300 Hz and an amplitude of 5 ⁇ m, with the Doppler spectrum being disposed to the left of the vertical axis, and the 2D FFT spectrum being disposed on the right of the vertical axis. From Eq.
- the presence of a vibration causes a Bessel modulation of the received signal, which generates multiple copies of the spectrum illustrated in FIGURE 5 A, on lines parallel to a line through the origin, and offset with respect to each other by f ⁇ .
- the conventional Doppler spectrum i.e., on the left of the vertical axis in FIGURE 5B
- FIGURE 6A which includes both a 2D FFT spectrum and a conventional Doppler spectrum of received ultrasound echoes (where the center frequency of the transmitted ultrasound signal is 5 MHz) when scatterers responsible for the echoes are moving with constant velocity of 200 mm/s and also vibrating with a frequency 300 Hz, at an amplitude of 5 ⁇ m.
- FIGURE 6B depicts the same situation when a lower ultrasound center frequency of 2 MHz is used for interrogation (i.e., for both a 2D FFT spectrum and a conventional Doppler spectrum of received ultrasound echoes) and when scatterers responsible for the echoes are moving with constant velocity of 200 mm/s, and also vibrating with a frequency 300 Hz, at an amplitude of 5 ⁇ m.
- a lower ultrasound center frequency of 2 MHz is used for interrogation (i.e., for both a 2D FFT spectrum and a conventional Doppler spectrum of received ultrasound echoes) and when scatterers responsible for the echoes are moving with constant velocity of 200 mm/s, and also vibrating with a frequency 300 Hz, at an amplitude of 5 ⁇ m.
- the Doppler spectrum is disposed to the left of the vertical axis, and the 2D FFT spectrum is disposed on the right of the vertical axis). Note that since the Doppler shift of the 2 MHz ultrasound frequency of FIGURE 6B is lower, the spread in the Doppler spectrum in FIGURE 6B is reduced as compared to the spread in the Doppler spectrum of FIGURE 6A, and the symmetric vibration signature is partially visible. Therefore, for analyzing vibrations in rapidly moving tissue, such as the cardiac wall, a lower frequency should be chosen for the transmit pulse of the interrogating ultrasound wave.
- FIGURE 7A graphically illustrates both a 2D Fourier (2D FFT) spectrum and a conventional Doppler spectrum of simulated received ultrasound echoes from a group of scatterers accelerating from 10 mm/s to 30 mm/s within the interrogation window (i.e., with an acceleration of 5 m/s 2 ), with the Doppler spectrum being disposed to the left of the vertical axis, and the 2D FFT spectrum being disposed on the right of the vertical axis.
- 2D FFT 2D Fourier
- FIGURE 7B graphically illustrates both a 2D FFT spectrum and a conventional Doppler spectrum of simulated received ultrasound echoes from a group of scatterers accelerating from 10 mm/s to 30 mm/s within the interrogation window (i.e., with an acceleration of 5 m/s 2 ), when the scatterers are also vibrating with a frequency 300 Hz and an amplitude of 5 ⁇ m, with the Doppler spectrum being disposed to the left of the vertical axis, and the 2D FFT spectrum being disposed on the right of the vertical axis.
- FIGURE 7B indicates that when a vibration is present in accelerating tissue, the high acceleration can significantly obscure the vibration spectrum. Thus, it is important to preprocess the received ultrasound data to suppress the effect of tissue acceleration.
- the phase of the received ultrasound signal as described in Eq. (5) is influenced primarily by the axial component of the displacement. If the tissue were perfectly homogeneous, the phase would remain unchanged for any motion orthogonal to the axial direction. However, due to the non homogeneous nature of many tissues, there is a change in the phase as well as the amplitude of the received signal, even for the lateral and elevation components of motion. Thus, off-axis motion components do affect the received signal. The presence of transverse velocity components will result in a broadening of the spectrum, which is proportional to the magnitude of the transverse velocity component. Transverse vibration components will cause a similar broadening of the spectrum.
- FIGURE 8A graphically illustrates both a 2D FFT spectrum and a conventional Doppler spectrum of simulated received ultrasound echoes when scatterers responsible for the echoes are moving with a constant velocity of 20 mm/s, the scatterer motion being perpendicular to the axis of the ultrasound beam, with the Doppler spectrum being disposed to the left of the vertical axis, and the 2D FFT spectrum being disposed on the right of the vertical axis.
- FIGURE 8B graphically illustrates both a 2D FFT spectrum and a conventional Doppler spectrum of simulated received ultrasound echoes when scatterers responsible for the echoes are moving with a constant velocity of 20 mm/s, the scatterer motion being perpendicular to the axis of the ultrasound beam, where the scatterers are also vibrating with a frequency of 300 Hz, at an amplitude of 5 ⁇ m.
- the Doppler spectrum is disposed to the left of the vertical axis
- the 2D FFT spectrum is disposed on the right of the vertical axis.
- the conventional Doppler spectrum can have a large spectral variance depending upon the Doppler shift and tissue acceleration.
- Another artifact in conventional Doppler processing is a granular speckle pattern that is produced due to random phase shifts produced by constructive and destructive interference of scattering from multiple scatterers in the sample volume. Speckle artifacts and large variance can mask the underlying harmonic spectral signatures that are associated with vibrations. Such effects are more pronounced when the tissue motion and acceleration are large, such as in the case of cardiac wall motion, which is an important limitation of conventional Doppler processing with respect to analyzing vibrations in the cardiac wall.
- Wideband Doppler estimation techniques can reduce the inherent spectral broadening introduced by conventional Doppler spectral processing. As indicated in FIGURES 7 A and 7B, the spectral variance is primarily due to the bandwidth of the transmitted signal. Wideband estimation techniques utilize the bandwidth of the transmitted signal to estimate the Doppler shift and thus can reduce the spectral broadening. Such wideband estimates can also reduce the speckle noise, since the contributions from the sample volume are analyzed separately.
- WLE Wideband Maximum Likelihood Estimator
- WCE Wideband Cross-correlation Estimator
- the Radon transform For estimating blood velocity, the Radon transform has been proposed to estimate the slope of the line in the 2D Fourier transform domain (Munk and Jensen. "A new approach for the estimation of axial velocity using ultrasound," Ultrasonics, 2000; 37:661-5).
- this method is adapted for identifying vibrations in the tissue surrounding stenosed blood vessels.
- the Radon transform domain Based on FIGURES 5A-5B, the Radon transform domain can be interpreted as a mapping between the normalized Doppler shift - ⁇ - , and the frequency shift jQ ⁇ -
- FIGURE 9A graphically illustrates a Radon transform of the 2D FFT spectrum of FIGURE 5A.
- a peak is seen at the normalized Doppler shift of 2.59 x 10 ⁇ 5 , at a zero frequency shift.
- FIGURE 9B graphically illustrates a Radon transform of the 2D FFT spectrum of FIGURE 5B. Multiple copies of the peak at 2.59 x 10 ⁇ 5 can now be seen at different frequency shifts.
- the corresponding Doppler spectra can be extracted from the Radon transform by plotting the intensities corresponding to all the frequency shifts at the peak Doppler shift.
- FIGURES 9C and 9D graphically illustrate the corresponding Doppler spectra computed from the Radon transforms of FIGURES 9 A and 9B.
- the spectral spread is significantly reduced.
- the vibration peaks occur at the frequency of ⁇ 278 Hz (a 7.3% error compared to the actual value of 300 Hz), as indicated by arrows 29 in FIGURE 9D.
- the difference between the tissue motion peak (indicated by an arrow 31) and the first vibration peak (arrows 29) is 4.9 dB, which corresponds to an estimated vibration amplitude of 7.9 ⁇ m (a 58% error compared to the actual value of 5 ⁇ m).
- An alternative blood velocity estimator can be derived from the 2D FFT spectrum by combining appropriately scaled Doppler spectra corresponding to the whole range of frequencies in the transmitted bandwidth (Loupas and Gill, "Multifrequency Doppler: Improving the quality of spectral estimation by making full use of the information present in the backscattered RF echoes," IEEE Trans Ultrason Ferroelect Freq Contr., 1994; 42:672-88).
- the estimated velocity spectra from this method is given by
- fo is the center frequency
- BW is the bandwidth of the fransmitted pulse
- ⁇ and foop are the spatial and temporal frequency in the 2D FFT domain, respectively
- Ytfw, foop) and Y O ([R F ) are the Fourier transforms defined in Eq. (7).
- This estimate is referred to as the "multifrequency estimate.”
- the multifrequency estimate improves the velocity resolution, since the large statistical fluctuations introduced by the integration over a sample volume are avoided. Ih the present disclosure, these techniques have been adapted for identifying vibrations in the tissue surrounding a stenosed blood vessel.
- the multifrequency estimates computed from FIGURES 5A and 5B using 1-MHz bandwidths are respectively illustrated in FIGURES 1OA and
- the tissue motion peak is at 125 Hz, as indicated by an arrow 33 (in
- FIGURES 1OA and 10B which corresponds well with the expected Doppler shift of 129 Hz.
- the first vibration peaks are at -172 Hz and 422 Hz, respectively (as indicated by arrows 35 in FIGURE 10B), resulting in an estimated vibration frequency of 297 Hz, according to Eq. (13) (a 1% error compared to the actual value of 300 Hz).
- the difference between the low-frequency peak and the first vibration peak is 6.5 dB, which corresponds to an estimated vibration amplitude of 5.48 ⁇ m (a 9.6% error compared to the actual value of 5 ⁇ m).
- the multifrequency estimate provides a more accurate estimate of the vibration amplitude and frequency as compared to the Radon transform.
- the Radon transform estimate automatically corrects for the Doppler shift due to mean tissue motion, whereas this Doppler shift is preserved in the multifrequency estimate.
- the spectral spread of the multifrequency estimate is similar to that of the Radon transform estimate.
- the additional computational burden of computing the Radon transform makes the estimate based on the Radon transform less desirable than the multifrequency estimate.
- FIGURE IlA schematically illustrates the simulation model (i.e., a 3D model of scatterer distributions) used in the validation.
- the ultrasound simulator Field II (Jensen 1996) was used to compute the pulse echo spatial impulse response, h pe (-), and the transducer temporal response, xo(-).
- the scattering amplitudes, ⁇ %, and mean positions, r ⁇ j, were randomly assigned from a Gaussian distribution- with the scattering strength from the vessel wall 40 dB higher than that from blood.
- the instantaneous scatterer positions, v(r ⁇ ,t), were estimated using the phase of the Doppler ultrasound signal from the vessel wall of a normal human femoral artery, as is graphically illustrated in FIGURE HB.
- the motion was defined as being in a direction perpendicular to the vessel wall with a peak displacement of 0.08 mm. Vibrations were generated in one region of the vessel wall, with motion in a direction perpendicular to the vessel wall, with the peak amplitude of 5 ⁇ m and a frequency of 100 Hz.
- the clutter motion with vibrations is graphically illustrated in a box 180 of FIGURE HB.
- the vibration, d ⁇ h (t) was modeled as a Gaussian-weighted sinusoid with additive white Gaussian noise at different signal-to-noise ratios (SNR), as follows:
- ⁇ sm is the SNR of the white Gaussian noise n(t).
- Gaussian noise simulates broadband vibrations expected to be produced by blood flow eddies and turbulent flow.
- the vibration frequency was 100 Hz and ⁇ sm . was varied from 0 to 2.
- the vibration detection performance was evaluated with different threshold values to measure the sensitivity and specificity.
- the threshold value, Et hres h indicates the % of energy in the dominant components for a signal to be considered as vibrations.
- the threshold value, i ⁇ hres b indicates the maximum difference in frequency of a matching pair of complex exponentials. Simulations were performed with different threshold values and different model orders, and receiver- operating characteristic (ROC) curves were generated to evaluate the detector performance. The ROC curves can then be used as a guideline for choosing the appropriate threshold setting and model orders.
- ROC receiver- operating characteristic
- FIGURE 12A is a vibration amplitude image overlaid on a B-mode image using a black-green colormap. The colormap is calibrated according to the values of the estimated amplitude.
- FIGURES 13A-13F The sensitivity, specificity and ROC curves for the two primary algorithms discussed above (the phase decomposition algorithm and the algorithm based on estimating complex exponentials in noise) using different model orders are graphically illustrated in FIGURES 13A-13F.
- FIGURE 13A indicates that, for the phase-decomposition algorithm shown in FIGURE 3 A, the sensitivity decreases with the increasing threshold value for all model orders, because more true vibrations are rejected with larger threshold values.
- Lower model orders have higher sensitivity, because the correlation matrix is smaller; thus, a better estimate can be obtained using the limited number of temporal samples.
- FIGURE 13B shows that the specificity is quite similar for all the model orders and increases with increasing threshold value, because a larger threshold leads to better noise rejection.
- FIGURE 13 C The ROC curves for different model orders are graphically illustrated in FIGURE 13 C, which indicates that a sensitivity of 96% and a specificity of 98% can be achieved with a second-order model.
- an appropriate threshold value an operating point is selected in the ROC curve. The corresponding threshold value can then be found from FIGURES 13A or 13B.
- FIGURE 13D shows that, for the root-MUSIC-based algorithm shown in FIGURE 3A, the sensitivity increases with the increasing frequency threshold value for all model orders, as more true vibrations can be detected if the frequency threshold is increased.
- FIGURE 13E shows that the specificity decreases with increasing threshold values, because more false detections occur with increased frequency threshold.
- the fourth-order model has slightly better sensitivity and specificity due to better modeling of the clutter space.
- the ROC curves for the root-MUSIC-based algorithm are graphically illustrated in FIGURE 13F.
- the third-order algorithm a sensitivity of 97% and a specificity of 98% are achievable, whereas, for the fourth-order algorithm, the sensitivity can be increased to 98%, with a specificity of 99%.
- the sensitivity decreases for the root-MUSIC algorithm (line 202) because it is based on modeling the vibrations as complex exponentials with narrow bandwidth.
- the phase-decomposition algorithm (line 200) is more robust to the vibration bandwidth because it makes no a priori assumptions about the vibration bandwidth characteristics.
- Tissue Vibration Imaging System For tissue vibration imaging to be clinically useful, real-time visualization of vibrations is quite desirable.
- Programmable ultrasound signal and image-processing systems using high-performance multimedia processors to support all the conventional processing modes, such as B, M, color flow, and Doppler are available in software.
- the main strength of a programmable system is the ease of development of new modes and applications without the need for new hardware or making hardware modifications to conventional ultrasound machines.
- Such programmable ultrasound machines allow access to and processing of internal raw color-flow and pulsed-wave (PW) Doppler quadrature data, and facilitate implementing the phase- decomposition algorithm discussed above for tissue vibration imaging in real time.
- PW pulsed-wave
- the phase-decomposition algorithm has already been empirically tested in a software-programmable ultrasound system for online visualization of vibrations during 2D ultrasound scans.
- tissue vibration imaging can be achieved at 9.1 frames/s for 32 scan lines with an ensemble size of 10 and 256 samples per scan line.
- the computational power of ultrasound machines has increased significantly in recent years, benefiting from advances in processor technology, and this trend is expected to continue in the future.
- Many modern processors targeted for multimedia applications have specialized instructions that can perform complex multiplications and additions with the same computational overhead as real multiplications and additions. Using such processors, the computational burden to support the algorithms disclosed herein can be reduced by a factor of three or four. Thus, the additional computational burden of the tissue- vibration imaging algorithms disclosed herein can be reasonably supported in modern ultrasound machines.
- FIGURE 15 schematically illustrates a test system including a physical phantom (i.e., a vibrating plate) used for validating the vibration-imaging algorithms discussed above.
- a piezoelectric plate 181 is incorporated into a test vessel including walls 183 and a plastic base 181.
- the piezoelectric plate is logically coupled with a function generator 194.
- the test vessel is filled with water 184.
- An ultrasound probe 190 is an ultrasound probe 190
- a fiber optic micrometer 188 (logically coupled to an oscilloscope 192) are disposed in the water bath (i.e., in the test vessel filled with water) proximate the piezoelectric plate, which was vibrated at frequencies between 100 Hz and 800 Hz using a sinusoidal signal from a function generator 194.
- the amplitude of the vibrating plate was calibrated using fiber optic micrometer 188 for different drive voltages corresponding to peak plate displacements of 1 to 7 ⁇ rn.
- Piezoelectric plate 181 was then imaged using a programmable ultrasound system with a 5-MHz linear transducer and an ensemble size of 10 at different PRFs.
- the position of the plate was also measured using the fiber optic micrometer, enabling a comparison to be made with the data obtained using ultrasound.
- FIGURE 16A is a vibration amplitude image of the plate phantom (i.e., FIGURE 18), while FIGURE 16B is a vibration frequency image of the plate phantom, indicating that vibrations have been correctly detected at the location of the piezoelectric plate. Because the edges of the plate are attached to the base, the maximum vibration amplitude is expected at the center of the plate, with zero displacement at the edges, which indeed corresponds to what is shown in FIGURE 16 A. The estimated vibration frequency at the center of the plate is between 450 and 500Hz.
- the MUSIC pseudo- spectrum is shown in FIGURE 16C, which indicates that the zero-frequency peak corresponds to stationary echo. A prominent double-sided peak is observed at ⁇ 500 Hz, corresponding to the vibration frequency of the plate.
- FIGURE 17A graphically illustrates the ultrasonically estimated vibration amplitude (y-axis) versus the independently measured values using the fiber optic micrometer of FIGURE 15 (x-axis) for different drive voltages used to displace the piezoelectric plate.
- the amplitude and frequency were estimated using the estimators defined in Eqs. (8) and (9).
- the fiber optic amplitude measurements were made at the center of the plate.
- FIGURE 17B graphically illustrates the ultrasonically estimated frequency (y-axis) versus the function generator frequency (x-axis).
- a solid line 204 with a slope of unity is shown in both plots. The difference between the estimated and measured values is plotted against the corresponding measured value in FIGURE 17C for amplitude and in FIGURE 17D for frequency.
- the maximum difference between the detected and measured values is less than 1 ⁇ m for amplitude and less than 50 Hz for frequency for both estimators. Some of the differences in amplitude can be attributed to variability in the location on the plate at which the fiber optic measurements were made.
- FIGURE 18 schematically illustrates the experimental setup, which includes a computer-based controller 140 logically coupled with an oscilloscope 142 and a fiber optic micrometer 148, a Doppler capable ultrasound-imaging machine 144 and an imaging probe 146.
- Fiber optic micrometer 148 and imaging probe 146 are positioned proximate to an in vivo artery sample disposed in a water bath 150. Pulsatile flow mimicking human arterial flow is created through the artery sample using a pulsatile pump 154.
- pulsatile pump 154 (a PulsatronTM pump, available from Pulsafeeder Inc, Punta Gorda, FL) is connected to the in vivo artery sample through a damping column 156. Adjusting the height of fluid in the damping column controls the overall flow impedance, such that the Doppler flow profile in the artery can be made to appear visually similar to that of human arterial flow, as shown in FIGURE 19.
- the time-varying wall displacement is measured using ultrasound as well as the fiber-optic micrometer.
- the output of the micrometer is digitized using the oscilloscope, and the data are acquired using the computer-based controller.
- the raw ultrasound data are acquired digitally from inside the ultrasound machine.
- the power spectra of the wall displacement determined using both methods can then be compared.
- Stenoses are simulated by partially ligating one part of the artery to reduce the effective lumen diameter. Power spectra measurements are then repeated using both ultrasound and fiber-optic methods.
- FIGURES 2OA and 2OB are vibration amplitude images from two different ex vivo arteries with simulated stenoses. In FIGURE 2OA, a 30% stenosis is simulated in a 3.4 mm diameter artery.
- FIGURE 2OB The wall vibrations occur downstream, at a distance of approximately four artery diameters.
- FIGURE 2OB a 42% stenosis is simulated in a 3.1 mm diameter artery, and the wall vibrations occur closer, approximately three diameters downstream.
- FIGURES 2OA and 2OB indicate that using the technique described above, it is possible to distinctly image the location of the artery wall vibrations (and hence, the location of the stenosis).
- the stenosis sites are indicated by solid arrows 141, while dash arrows 143 indicate the direction of flow.
- the region of interest in each image is indicated by a box 145.
- FIGURE 21 graphically illustrates the mean vibration spectra determined using ultrasound, indicated by an arrow 147, and the fiber-optic micrometer (see FIGURE 18, discussed above), indicated by an arrow 149, during peak flow acceleration.
- the vertical bars indicate the standard deviations. Both spectra have a similar shape, and a break frequency can be observed beyond which the energy drops off rapidly. This result indicates that it is possible to assess the wall vibration spectrum using ultrasound.
- FIGURE 3 A (preferably employing the ESPRIT method) was implemented on a programmable ultrasound machine, the Hitachi HiVision 5500TM, which is manufactured by Hitachi Medical Systems America, Twinsburg, OH. All of the signal and image processing on this machine is performed by software, thus providing the flexibility to easily incorporate new algorithms.
- This system programmed to use the algorithms disclosed above, enables vibrations to be visualized in real time, facilitating the evaluation of the technique described above during an in vivo procedure.
- FIGURE 23 A is a color power image of a stenosed vein graft
- FIGURE 23B is a vibration amplitude image of the same stenosed vein graft
- FIGURE 23C is a vibration frequency image of the same stenosed vein graft.
- An arrow 206 indicates the location of the stenosis.
- a perivascular artifact is visible in the color power image, as indicated by an arrow 208, while the vibration amplitude image clearly shows the origin of the Son downstream of the stenosis, as indicated by arrows 210.
- the vibration amplitude is highest close to the vessel wall and decreases farther away from the vessel wall.
- FIGURE 24A graphically illustrates instantaneous vessel wall position estimated using the phase of pulsed-wave Doppler data from a stenosed femoral vein graft, with vibrations being indicated in boxes 212.
- FIGURE 24B is a motion periodogram of the signal from FIGURE 24A.
- the displacement spectrum in FIGURE 24B shows significant energy up to 200 Hz, and repeats with each cardiac cycle.
- FIGURE 24C A cross section of the spectrum in FIGURE 24B at a time of 1.25 seconds is graphically illustrated in FIGURE 24C.
- a peak is observed at the break frequency of about 90Hz, as indicated by an arrow 214), beyond which the energy decays with increasing frequency.
- FIGURE 24D graphically illustrates the pseudo-spectrum estimated from only 10 ensembles of color-flow data at the same location using the MUSIC algorithm, generally as described in connection with FIGURE 3A.
- a prominent spectral peak is observed at the break frequency, as indicated by an arrow 216.
- the MUSIC pseudo-spectrum does not reflect the full spectral characteristics, but may be used to estimate the spectral peaks. This case study shows that in vivo tissue vibrations caused by blood flow eddies can be detected using only a short temporal record, demonstrating the feasibility of real-time vibration imaging.
- FIGURE 25A graphically illustrates the wall displacement spectrum from a normal femoral artery computed using the estimated displacement from pulsed-wave Doppler data.
- the spectral energy rapidly decays within a few tens of Hz, and the spectral energy beyond 100 Hz is comparable to the noise level.
- FIGURE 25B graphically illustrates the spectrum from a stenosed bypass vein graft. The spectral energy decays more gradually, and a significant energy is present, even at several hundred Hz.
- An arrow 160 indicates the break frequency, beyond which the energy decays with increasing frequency.
- FIGURE 25C graphically illustrates the spectrum from a second stenosis in the same patient. A peak in the spectrum can be observed at the break frequency, as indicated by an arrow 162.
- FIGURE 25D graphically illustrates the spectrum from a vein-graft stenosis in a different patient. Again, a prominent spectral peak can be observed at the break frequency, as indicated by an arrow 164.
- FIGURES 26A-26D graphically illustrate such a time-varying wall vibration spectrum.
- the spectral energy beyond 45 Hz is comparable to the noise level, m case of stenoses, significant spectral energy is present in the higher frequencies, as graphically illustrated in FIGURES 26B-26D.
- the vibration occurs just after the peak systolic wall motion.
- the break frequency can be determined visually from this time-varying vibration spectral display, as shown by arrows 166, 168, and 170. Automatic detection of these break frequencies can also be performed in real time.
- FIGURE 27A is a Doppler spectrum computed using the 2D FFT method described above from a range placed on the myocardial wall of a patient with angiographicaUy confirmed coronary artery disease in the left anterior descending (LAD) artery and the right coronary artery (RCA).
- the range gate was placed in the vicinity of the RCA.
- the four phases of myocardial wall motion corresponding to isovolumetric contraction (IVC), ventricular ejection (VE), isovolumetric relaxation (IVR), and ventricular filling (VF) are indicated.
- IVC isovolumetric contraction
- VE ventricular ejection
- IVR isovolumetric relaxation
- VF ventricular filling
- a clear harmonic spectrum indicative of high- frequency narrowband vibrations can be observed during the latter part of the ventricular ejection phase.
- the vibrations have continuously decreasing frequency and appear as oblique bands, and repeat in two consecutive cardiac cycles.
- the symmetric double- sided peaks are indicative of vibrations observed in the late
- FIGURE 27B graphically illustrates a detailed time course of the wall velocity during ventricular ejection, estimated using the autocorrelation method discussed in detail above.
- the velocity shows oscillatory components indicative of vibrations, as indicated in a boxed enclosed region.
- the duration of the oscillation is approximately 85 ms, and the oscillation appears to have harmonic components.
- FIGURE 28A is an angiographic image of a right coronary artery of the patient imaged in FIGURES 27 A and 27B, acquired in the left anterior oblique projection with caudal angulation.
- a diffuse 20% stenosis in the proximal RCA, a tubular 20% stenosis in the mid RCA, and a 40% stenosis in the distal RCA can be identified.
- FIGURE 28B is a vibration amplitude image overlaid on an apical two- chamber view in diastole from the patient of FIGURES 27A, 27B and 28A. Vibrations in the posterior left-ventricular wall can be seen near the mid and distal portions of the RCA. The vibrations appear to be localized in two regions, which could correspond to the two different lesions in the distal RCA.
- FIGURE 29A is an angiographic image of the patient imaged in FIGURES 27 A, 27B, 28 A and 28B, acquired in the right anterior oblique projection with cranial angulation.
- the proximal LAD is moderately calcified.
- FIGURE 29B is a vibration amplitude image overlaid on the apical two- chamber view of the patient of FIGURE 29A. Myocardial vibrations can be observed in the mid-LAD section.
- Differentiating Tissue Vibrations ArisinR from A Stenosis from Other Sources Vibrations are produced due to pressure differences across an orifice.
- Stenoses represent a relatively common physiological features including orifices where such pressure differences exist.
- other physiological features such as punctured blood vessels, also include orifices with pressure differences which can generate vibrations. The following provides a description of how the vibration imaging techniques disclosed herein can distinguish stenoses from other sources of tissue vibrations.
- tissue vibration processor 28 can be configured to determine the timing of the vibrations in the cardiac cycle using electrocardiograph signals. In another embodiment, the tissue vibration processor can be configured to determine the timing of the vibrations during the cardiac cycle using the periodicity of tissue motion due to cardiac pulsation.
- Empirical evidence demonstrates the feasibility of real-time ultrasound imaging of low-intensity local vibrations in the vessel wall and surrounding tissue associated with stenosed blood vessels.
- Several algorithms based on parametric signal decomposition and spectral estimation have been developed for imaging small-amplitude tissue vibrations using as few as 10 temporal samples. Simulations show that these algorithms have high sensitivity (96 to 98%) and specificity (98 to 99%) for detecting vibrations in the presence of clutter as well as blood flow, and are robust even when broadband vibrations are present.
- the vibration amplitude and frequency can be estimated accurately, and real-time tissue vibration imaging has been implemented on an ultrasound machine with a software-programmable subsystem. Vibrations were observed in stenosed bypass vein grafts and from coronary arteries in human subjects.
- Tissue vibration imaging can provide additional diagnostic information that is currently not available to the clinician using conventional tools.
- An ultrasound device with tissue vibration imaging capability can become a useful screening and diagnostic tool for the assessment of stenoses and other vascular abnormalities traditionally associated with Sons that are otherwise hard to diagnose using conventional duplex ultrasound.
- the ultrasonic vibration imaging techniques disclosed herein are attractive because of their potential to visualize small-amplitude vibrations at their origin.
- the vibration spectra can be used to compute the break frequency, which is directly related to the residual lumen diameter at the stenosis.
- An important application of tissue vibration imaging will likely be the noninvasive diagnosis of coronary artery stenoses.
- Conventional duplex ultrasound is limited by the difficulty in visualizing coronary arteries and the poor scattering strength from coronary blood flow.
Abstract
Description
Claims
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2007530355A JP5529378B2 (en) | 2004-08-31 | 2005-08-31 | An ultrasonic technique for evaluating wall vibration in stenotic vessels |
EP05858095.2A EP1784130A4 (en) | 2004-08-31 | 2005-08-31 | Ultrasonic technique for assessing wall vibrations in stenosed blood vessels |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US60616204P | 2004-08-31 | 2004-08-31 | |
US60/606,162 | 2004-08-31 |
Publications (3)
Publication Number | Publication Date |
---|---|
WO2007001352A2 WO2007001352A2 (en) | 2007-01-04 |
WO2007001352A9 true WO2007001352A9 (en) | 2007-02-22 |
WO2007001352A3 WO2007001352A3 (en) | 2007-05-03 |
Family
ID=37595565
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/US2005/031135 WO2007001352A2 (en) | 2004-08-31 | 2005-08-31 | Ultrasonic technique for assessing wall vibrations in stenosed blood vessels |
Country Status (6)
Country | Link |
---|---|
US (1) | US7736314B2 (en) |
EP (1) | EP1784130A4 (en) |
JP (1) | JP5529378B2 (en) |
KR (1) | KR101264442B1 (en) |
CN (1) | CN100548224C (en) |
WO (1) | WO2007001352A2 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103020474A (en) * | 2012-12-27 | 2013-04-03 | 北京航天福道高技术股份有限公司 | Cluster model determining method based on cluster statistical distribution model and power spectrum model |
Families Citing this family (102)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9066679B2 (en) * | 2004-08-31 | 2015-06-30 | University Of Washington | Ultrasonic technique for assessing wall vibrations in stenosed blood vessels |
US7520860B2 (en) * | 2005-04-13 | 2009-04-21 | Marie G. Johnson | Detection of coronary artery disease using an electronic stethoscope |
US8211024B2 (en) * | 2005-06-06 | 2012-07-03 | Siemens Medical Solutions Usa, Inc. | Medical ultrasound pressure gradient measurement |
JP4987295B2 (en) * | 2005-12-26 | 2012-07-25 | 株式会社東芝 | Ultrasonic diagnostic equipment |
WO2007092054A2 (en) | 2006-02-06 | 2007-08-16 | Specht Donald F | Method and apparatus to visualize the coronary arteries using ultrasound |
GB0613739D0 (en) * | 2006-07-11 | 2006-08-23 | Univ Belfast | Improvements relating to the assessment of blood velocity signals |
EP2088932B1 (en) | 2006-10-25 | 2020-04-08 | Maui Imaging, Inc. | Method and apparatus to produce ultrasonic images using multiple apertures |
US8050476B2 (en) * | 2006-11-22 | 2011-11-01 | General Electric Company | Heart rate demodulation of periodic movement in ultrasound data for object identification |
US8460190B2 (en) * | 2006-12-21 | 2013-06-11 | Siemens Medical Solutions Usa, Inc. | Automated image interpretation with transducer position or orientation sensing for medical ultrasound |
JP5346440B2 (en) * | 2007-02-15 | 2013-11-20 | 富士フイルム株式会社 | Ultrasonic diagnostic apparatus and data measurement program |
US9005126B2 (en) * | 2007-05-03 | 2015-04-14 | University Of Washington | Ultrasonic tissue displacement/strain imaging of brain function |
US8323199B2 (en) * | 2007-09-28 | 2012-12-04 | The University Of British Columbia | Method and apparatus for imaging the mechanical properties of tissue from an endocavity |
US9282945B2 (en) | 2009-04-14 | 2016-03-15 | Maui Imaging, Inc. | Calibration of ultrasound probes |
DE102008027384A1 (en) * | 2008-06-09 | 2009-12-10 | Ge Inspection Technologies Gmbh | Improved non-destructive ultrasound examination with coupling control |
US8187187B2 (en) * | 2008-07-16 | 2012-05-29 | Siemens Medical Solutions Usa, Inc. | Shear wave imaging |
JP5666446B2 (en) * | 2008-08-08 | 2015-02-12 | マウイ イマギング,インコーポレーテッド | Image forming method using multi-aperture medical ultrasonic technology and synchronization method of add-on system |
JP5485373B2 (en) * | 2009-04-14 | 2014-05-07 | マウイ イマギング,インコーポレーテッド | Multiple aperture ultrasonic array alignment system |
JP2012523920A (en) * | 2009-04-14 | 2012-10-11 | マウイ イマギング,インコーポレーテッド | Universal multi-aperture medical ultrasound probe |
DK200900527A (en) * | 2009-04-24 | 2010-10-25 | Region Nordjylland Aalborg Syg | Device for holding an imaging probe and use of such device |
KR101117900B1 (en) * | 2009-04-30 | 2012-05-21 | 삼성메디슨 주식회사 | Ultrasound system and method for setting eigenvectors |
US8306296B2 (en) | 2009-04-30 | 2012-11-06 | Medison Co., Ltd. | Clutter signal filtering using eigenvectors in an ultrasound system |
US20110002518A1 (en) * | 2009-07-01 | 2011-01-06 | General Electric Company | Method and system for processing ultrasound data |
JP5355327B2 (en) * | 2009-09-18 | 2013-11-27 | 日立アロカメディカル株式会社 | Ultrasonic diagnostic equipment |
US20110137210A1 (en) * | 2009-12-08 | 2011-06-09 | Johnson Marie A | Systems and methods for detecting cardiovascular disease |
JP5398514B2 (en) * | 2009-12-21 | 2014-01-29 | 株式会社東芝 | Color Doppler ultrasound system |
JP6274724B2 (en) | 2010-02-18 | 2018-02-07 | マウイ イマギング,インコーポレーテッド | Point source transmission and sound velocity correction using multi-aperture ultrasound imaging |
EP2380498B8 (en) * | 2010-04-26 | 2015-01-28 | Samsung Medison Co., Ltd. | Adaptive clutter filtering in an ultrasound system |
KR101239809B1 (en) * | 2010-05-07 | 2013-03-06 | 삼성메디슨 주식회사 | Ultrasound imaging device and method for clutter filtering |
WO2011155168A1 (en) * | 2010-06-07 | 2011-12-15 | パナソニック株式会社 | Malignant tissue tumor detection method and malignant tissue tumor detection device |
KR101232796B1 (en) * | 2010-07-22 | 2013-02-13 | 삼성메디슨 주식회사 | Ultrasound imaging device and method for clutter filtering |
US8315812B2 (en) * | 2010-08-12 | 2012-11-20 | Heartflow, Inc. | Method and system for patient-specific modeling of blood flow |
CN102370499B (en) | 2010-08-26 | 2014-05-07 | 深圳迈瑞生物医疗电子股份有限公司 | Method and system for simultaneously displaying Doppler image, B-type image and colored blood flow image |
EP3563768A3 (en) | 2010-10-13 | 2020-02-12 | Maui Imaging, Inc. | Concave ultrasound transducers and 3d arrays |
WO2012051305A2 (en) | 2010-10-13 | 2012-04-19 | Mau Imaging, Inc. | Multiple aperture probe internal apparatus and cable assemblies |
KR101313220B1 (en) * | 2010-11-23 | 2013-09-30 | 삼성메디슨 주식회사 | Ultrasound system and method for providing color doppler mode image based on qualification curve |
WO2012070588A1 (en) * | 2010-11-25 | 2012-05-31 | 株式会社日立メディコ | Ultrasound moving image processing method, device and program |
JP5787527B2 (en) | 2011-01-18 | 2015-09-30 | キヤノン株式会社 | Signal processing circuit and ultrasonic diagnostic apparatus |
US8874390B2 (en) * | 2011-03-23 | 2014-10-28 | Hach Company | Instrument and method for processing a doppler measurement signal |
KR101312312B1 (en) * | 2011-09-09 | 2013-09-27 | 서강대학교산학협력단 | Method for processing ultrasound images using nonlinear property, Apparatus thereof |
WO2013059303A1 (en) * | 2011-10-17 | 2013-04-25 | University Of Washington Through Its Center For Commercialization | Methods and systems for imaging tissue motion using optical coherence tomography |
WO2013082455A1 (en) | 2011-12-01 | 2013-06-06 | Maui Imaging, Inc. | Motion detection using ping-based and multiple aperture doppler ultrasound |
CN104080407B (en) | 2011-12-29 | 2017-03-01 | 毛伊图像公司 | The M-mode ultra sonic imaging of free routing |
JP6438769B2 (en) | 2012-02-21 | 2018-12-19 | マウイ イマギング,インコーポレーテッド | Determination of material hardness using multiple aperture ultrasound. |
US8834376B2 (en) * | 2012-02-28 | 2014-09-16 | Her Majesty The Queen In Right Of Canada As Represented By The Minister Of Health | Dispersive ultrasound technology as a diagnostic device for traumatic brain injuries |
IN2014DN07243A (en) | 2012-03-26 | 2015-04-24 | Maui Imaging Inc | |
US10667790B2 (en) | 2012-03-26 | 2020-06-02 | Teratech Corporation | Tablet ultrasound system |
US9877699B2 (en) | 2012-03-26 | 2018-01-30 | Teratech Corporation | Tablet ultrasound system |
KR102101875B1 (en) * | 2012-07-20 | 2020-04-17 | 삼성전자주식회사 | Apparatus and method for generating tomography image |
JP6270843B2 (en) | 2012-08-10 | 2018-01-31 | マウイ イマギング,インコーポレーテッド | Calibration of multiple aperture ultrasonic probes |
WO2014031642A1 (en) | 2012-08-21 | 2014-02-27 | Maui Imaging, Inc. | Ultrasound imaging system memory architecture |
US10495725B2 (en) | 2012-12-05 | 2019-12-03 | Origin Wireless, Inc. | Method, apparatus, server and system for real-time vital sign detection and monitoring |
US10735298B2 (en) | 2012-12-05 | 2020-08-04 | Origin Wireless, Inc. | Method, apparatus, server and system for vital sign detection and monitoring |
US11012285B2 (en) | 2012-12-05 | 2021-05-18 | Origin Wireless, Inc. | Methods, apparatus, servers, and systems for vital signs detection and monitoring |
US9414752B2 (en) | 2012-11-09 | 2016-08-16 | Elwha Llc | Embolism deflector |
US9702972B2 (en) * | 2012-11-28 | 2017-07-11 | B-K Medical Aps | Angle independent velocity spectrum determination |
CN104936531B (en) * | 2013-01-22 | 2017-06-09 | 东芝医疗系统株式会社 | Diagnostic ultrasound equipment, image processing apparatus and image processing method |
AU2014209143B2 (en) | 2013-01-24 | 2018-05-24 | Alio, Inc. | Method and apparatus for measuring flow through a lumen |
CN103961135B (en) | 2013-02-04 | 2017-04-12 | 通用电气公司 | System and method for detecting guide pipe position in three-dimensional ultrasonic image |
US9510806B2 (en) | 2013-03-13 | 2016-12-06 | Maui Imaging, Inc. | Alignment of ultrasound transducer arrays and multiple aperture probe assembly |
WO2014153265A1 (en) * | 2013-03-18 | 2014-09-25 | PhonoFlow Medical, LLC | Spectrum analysis of coronary artery turbulent blood flow |
US9883848B2 (en) | 2013-09-13 | 2018-02-06 | Maui Imaging, Inc. | Ultrasound imaging using apparent point-source transmit transducer |
KR20150069830A (en) * | 2013-12-16 | 2015-06-24 | 삼성전자주식회사 | Method for providing blood vessel analysis information using medical image and apparatus providing blood vessel analysis information using medical image |
CN105395214B (en) * | 2014-02-20 | 2018-09-21 | 飞依诺科技(苏州)有限公司 | Ultrasonic imaging processing method and system based on RF data |
JP6282942B2 (en) * | 2014-06-18 | 2018-02-21 | キヤノンメディカルシステムズ株式会社 | Ultrasonic diagnostic apparatus, image processing apparatus, and image processing program |
CN104207803B (en) * | 2014-08-13 | 2016-03-02 | 四川省人民医院 | Based on the cardiac flow whirlpool Motion Adaptive visable positioning method of color doppler image information |
JP6722656B2 (en) | 2014-08-18 | 2020-07-15 | マウイ イマギング,インコーポレーテッド | Network-based ultrasound imaging system |
US10809365B2 (en) * | 2014-08-25 | 2020-10-20 | Texas Instruments Incorporated | Vibration parameters monitoring using FMCW radar |
JP6315196B2 (en) * | 2014-09-19 | 2018-04-25 | パナソニックIpマネジメント株式会社 | Biological information detection device and bed |
JP6006769B2 (en) * | 2014-10-16 | 2016-10-12 | 株式会社日立製作所 | Ultrasonic diagnostic equipment |
KR102389866B1 (en) * | 2014-12-30 | 2022-04-22 | 삼성전자주식회사 | Method for Generating a Ultrasound Image and Image Processing Apparatus |
KR102387708B1 (en) | 2015-01-30 | 2022-04-19 | 삼성메디슨 주식회사 | Ultrasound System And Method For Providing Guide To Improve HPRF Doppler Image |
US20160256107A1 (en) * | 2015-03-06 | 2016-09-08 | GraftWorx, LLC | Detection of stenosis in a prosthesis using break frequency |
US9924905B2 (en) | 2015-03-09 | 2018-03-27 | Graftworx, Inc. | Sensor position on a prosthesis for detection of a stenosis |
JP5936734B1 (en) * | 2015-03-11 | 2016-06-22 | 日立アロカメディカル株式会社 | Ultrasonic diagnostic equipment |
GB201506420D0 (en) * | 2015-04-15 | 2015-05-27 | Univ Newcastle | Improved blood pressure sensor |
US10206651B2 (en) * | 2015-09-30 | 2019-02-19 | General Electric Company | Methods and systems for measuring cardiac output |
CN105286919B (en) * | 2015-10-13 | 2018-06-22 | 广州丰谱信息技术有限公司 | Blood vessel state detection method and device based on heart point fluctuation transport properties |
GB201519985D0 (en) * | 2015-11-12 | 2015-12-30 | Respinor As | Ultrasonic method and apparatus for respiration monitoring |
CN108778530B (en) | 2016-01-27 | 2021-07-27 | 毛伊图像公司 | Ultrasound imaging with sparse array probe |
US11457890B2 (en) * | 2016-02-23 | 2022-10-04 | Mayo Foundation For Medical Education And Research | Ultrasound blood flow imaging |
CN108778106B (en) * | 2016-03-11 | 2021-09-07 | 欧利景无线有限公司 | Method, device, server and system for vital sign detection and monitoring |
JP6888915B2 (en) * | 2016-03-28 | 2021-06-18 | パイオニア株式会社 | Presentation control device, control method of presentation control device, computer program and recording medium |
JP6651405B2 (en) * | 2016-04-19 | 2020-02-19 | キヤノンメディカルシステムズ株式会社 | Ultrasound diagnostic apparatus and program |
EP3484353B1 (en) * | 2016-07-12 | 2020-09-09 | GraftWorx, Inc. | System for measuring blood flow parameters in a blood vessel having an endovascular prosthesis |
EP3278735A1 (en) | 2016-08-03 | 2018-02-07 | PI-Harvest Holding AG | A system and method for non-invasive measurement of pressure inside a body including intravascular blood pressure |
KR20190031567A (en) | 2016-08-03 | 2019-03-26 | 피아이-하베스트 홀딩스 아게 | System and method for non-invasive measurement of intra-body pressure including intra-arterial blood pressure |
WO2018049412A1 (en) | 2016-09-12 | 2018-03-15 | Graftworx, Inc. | Wearable device with multimodal diagnostics |
TWI633870B (en) | 2017-05-12 | 2018-09-01 | 新漢股份有限公司 | Portable device of monitoring hemodialysis access status |
CN108939185B (en) * | 2017-05-19 | 2021-09-14 | 新汉股份有限公司 | Portable blood vessel access state monitoring device |
KR101996025B1 (en) * | 2017-08-14 | 2019-07-03 | 국방과학연구소 | An information processing apparatus and method for processing a reflection signal for a target object |
JP7216720B2 (en) * | 2017-09-25 | 2023-02-01 | コーニンクレッカ フィリップス エヌ ヴェ | Method and system for filtering acoustic clutter and random noise |
EP3745097B1 (en) * | 2018-01-23 | 2023-11-08 | Kyocera Corporation | Fluid measurement device, fluid measurement method, and program |
KR102648307B1 (en) * | 2018-02-20 | 2024-03-15 | 삼성메디슨 주식회사 | Method and ultrasound diagnostic apparatus for displaying a doppler image |
EP3725232A1 (en) * | 2019-04-17 | 2020-10-21 | Koninklijke Philips N.V. | A patch sensor for a medical device |
CN110010229B (en) * | 2019-04-09 | 2020-06-02 | 西南石油大学 | Ultrasonic image simulation method based on generation of countermeasure network |
US20220183573A1 (en) * | 2019-04-18 | 2022-06-16 | Msheaf Health Management Technologies Limited | Non-invasive method and system for measuring motion characteristics of myocardial tissue |
CN110575627B (en) * | 2019-09-24 | 2021-04-06 | 黄晶 | Physical mapping device for rapidly acquiring target nerve treatment energy delivery site |
JP7077433B1 (en) | 2021-02-10 | 2022-05-30 | ジーイー・プレシジョン・ヘルスケア・エルエルシー | Control program for ultrasonic diagnostic equipment, ultrasonic diagnostic system and ultrasonic diagnostic equipment |
CN112998754B (en) * | 2021-02-20 | 2022-07-19 | 苏州大学 | Noninvasive blood flow detection method, detection system and detection equipment |
US11928811B2 (en) | 2021-03-30 | 2024-03-12 | Rtx Corporation | System and method for structural vibration mode identification |
CN113223069B (en) * | 2021-06-07 | 2023-02-10 | 西安电子科技大学 | Color image background clutter quantification method based on target and background difference |
CN113288218B (en) * | 2021-06-24 | 2023-04-07 | 无锡祥生医疗科技股份有限公司 | Echo processing method, imaging method and device for ultrasonic blood flow imaging |
Family Cites Families (33)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4770184A (en) * | 1985-12-17 | 1988-09-13 | Washington Research Foundation | Ultrasonic doppler diagnostic system using pattern recognition |
JPH04279154A (en) * | 1991-03-07 | 1992-10-05 | Toshiba Corp | Ultrasonic diagnostic apparatus |
US5194291A (en) * | 1991-04-22 | 1993-03-16 | General Atomics | Corona discharge treatment |
JP3453415B2 (en) * | 1992-12-22 | 2003-10-06 | コーニンクレッカ フィリップス エレクトロニクス エヌ ヴィ | Apparatus and method for measuring elasticity of artery by ultrasonic echography |
EP0674185A1 (en) * | 1994-03-25 | 1995-09-27 | Laboratoires D'electronique Philips S.A.S. | Method and system for detecting and characterising a segment of a blood vessel by ultrasonic echography |
US5534232A (en) * | 1994-08-11 | 1996-07-09 | Wisconsin Alumini Research Foundation | Apparatus for reactions in dense-medium plasmas |
US5638823A (en) * | 1995-08-28 | 1997-06-17 | Rutgers University | System and method for noninvasive detection of arterial stenosis |
US5824277A (en) * | 1995-12-06 | 1998-10-20 | E. I.Du Pont De Nemours And Company | Plasma oxidation of an exhaust gas stream from chlorinating titanium-containing material |
US5935339A (en) * | 1995-12-14 | 1999-08-10 | Iowa State University | Decontamination device and method thereof |
JP3652791B2 (en) * | 1996-06-24 | 2005-05-25 | 独立行政法人科学技術振興機構 | Ultrasonic diagnostic equipment |
IL119713A0 (en) * | 1996-11-28 | 1997-02-18 | Diasonics Israel Ltd | Coronary stenosis measurements |
EP0961577A1 (en) * | 1997-11-18 | 1999-12-08 | Koninklijke Philips Electronics N.V. | Method for the processing of signals relating to an object having moving parts and echographic device for carrying out this method |
US5919139A (en) | 1997-12-19 | 1999-07-06 | Diasonics Ultrasound | Vibrational doppler ultrasonic imaging |
US6406759B1 (en) * | 1998-01-08 | 2002-06-18 | The University Of Tennessee Research Corporation | Remote exposure of workpieces using a recirculated plasma |
WO1999035893A2 (en) * | 1998-01-08 | 1999-07-15 | The University Of Tennessee Research Corporation | Paraelectric gas flow accelerator |
US6210332B1 (en) * | 1998-03-31 | 2001-04-03 | General Electric Company | Method and apparatus for flow imaging using coded excitation |
JP2000041983A (en) * | 1998-05-25 | 2000-02-15 | Hitachi Medical Corp | Ultrasonograph |
US6036650A (en) * | 1998-09-15 | 2000-03-14 | Endosonics Corporation | Ultrasonic imaging system and method with ringdown reduction |
BR9804529A (en) * | 1998-10-23 | 2000-04-25 | Newmed Importadora Ltda | Device for performing secretion aspiration in intubated or trachenostomized patients and kit containing a device for performing secretion aspiration in intubated or tracheostomized patients. |
JP2000175915A (en) * | 1998-12-11 | 2000-06-27 | Hitachi Medical Corp | Ultrasonograph |
EP1055133A1 (en) * | 1998-12-15 | 2000-11-29 | Koninklijke Philips Electronics N.V. | Ultrasound process and apparatus for the determination of the location of a parietal surface in a tissue and of the absolute radius of an artery |
JP3398080B2 (en) | 1999-02-10 | 2003-04-21 | 科学技術振興事業団 | Vascular lesion diagnostic system and diagnostic program storage medium |
US6706892B1 (en) * | 1999-09-07 | 2004-03-16 | Conjuchem, Inc. | Pulmonary delivery for bioconjugation |
WO2001041648A1 (en) * | 1999-12-07 | 2001-06-14 | Koninklijke Philips Electronics N.V. | Ultrasonic image processing method and system for displaying a composite image sequence of an artery segment |
US6955648B2 (en) * | 2000-09-29 | 2005-10-18 | New Health Sciences, Inc. | Precision brain blood flow assessment remotely in real time using nanotechnology ultrasound |
JP2004520870A (en) * | 2000-11-28 | 2004-07-15 | アレズ フィジオニックス リミテッド | Non-invasive physiological evaluation system and method |
US6730030B2 (en) * | 2001-03-02 | 2004-05-04 | Yoram Palti | Method and apparatus for detecting arterial stenosis |
US7744537B2 (en) * | 2001-08-20 | 2010-06-29 | Japan Science And Technology Agency | Ultrasonic method and system for characterizing arterial tissue using known elasticity data |
US6709407B2 (en) * | 2001-10-30 | 2004-03-23 | Mayo Foundation For Medical Education And Research | Method and apparatus for fetal audio stimulation |
US20040059265A1 (en) * | 2002-09-12 | 2004-03-25 | The Regents Of The University Of California | Dynamic acoustic focusing utilizing time reversal |
US20050065436A1 (en) * | 2003-09-23 | 2005-03-24 | Ho Winston Zonh | Rapid and non-invasive optical detection of internal bleeding |
US20050182319A1 (en) * | 2004-02-17 | 2005-08-18 | Glossop Neil D. | Method and apparatus for registration, verification, and referencing of internal organs |
GB2445322B (en) * | 2004-08-13 | 2008-08-06 | Stichting Tech Wetenschapp | Intravasular ultrasound techniques |
-
2005
- 2005-08-31 CN CNB2005800292507A patent/CN100548224C/en not_active Expired - Fee Related
- 2005-08-31 JP JP2007530355A patent/JP5529378B2/en not_active Expired - Fee Related
- 2005-08-31 US US11/218,292 patent/US7736314B2/en not_active Expired - Fee Related
- 2005-08-31 WO PCT/US2005/031135 patent/WO2007001352A2/en active Search and Examination
- 2005-08-31 EP EP05858095.2A patent/EP1784130A4/en not_active Withdrawn
- 2005-08-31 KR KR1020077007438A patent/KR101264442B1/en not_active IP Right Cessation
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103020474A (en) * | 2012-12-27 | 2013-04-03 | 北京航天福道高技术股份有限公司 | Cluster model determining method based on cluster statistical distribution model and power spectrum model |
Also Published As
Publication number | Publication date |
---|---|
EP1784130A4 (en) | 2014-11-19 |
JP5529378B2 (en) | 2014-06-25 |
CN100548224C (en) | 2009-10-14 |
WO2007001352A3 (en) | 2007-05-03 |
CN101022765A (en) | 2007-08-22 |
KR101264442B1 (en) | 2013-05-14 |
KR20070056137A (en) | 2007-05-31 |
EP1784130A2 (en) | 2007-05-16 |
US20060079782A1 (en) | 2006-04-13 |
JP2008511415A (en) | 2008-04-17 |
US7736314B2 (en) | 2010-06-15 |
WO2007001352A2 (en) | 2007-01-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7736314B2 (en) | Ultrasonic technique for assessing wall vibrations in stenosed blood vessels | |
US9066679B2 (en) | Ultrasonic technique for assessing wall vibrations in stenosed blood vessels | |
US7803116B2 (en) | Transcutaneous localization of arterial bleeding by two-dimensional ultrasonic imaging of tissue vibrations | |
Hasegawa et al. | Improving accuracy in estimation of artery-wall displacement by referring to center frequency of RF echo | |
JP4842933B2 (en) | Diagnostic imaging equipment | |
US9883852B2 (en) | Ultrasound systems, methods and computer program products for estimating tissue deformation with harmonic signals | |
EP1637081B1 (en) | Ultrasonograph | |
Sahani et al. | Automated system for imageless evaluation of arterial compliance | |
WO2008038615A1 (en) | Ultrasonographic device | |
Huang et al. | Continuous measurement of arterial diameter using wearable and flexible ultrasonic sensor | |
JP2017538539A (en) | System and method for measuring arterial parameters | |
JP5534674B2 (en) | Ultrasonic method and apparatus for measuring or detecting non-sinusoidal periodic flow effects | |
Hasegawa et al. | Modification of the phased-tracking method for reduction of artifacts in estimated artery wall deformation | |
JP2009525117A5 (en) | ||
Bazan et al. | Power spectral estimation of high-harmonics in echoes of wall resonances to improve resolution in non-invasive measurements of wall mechanical properties in rubber tube and ex-vivo artery | |
Thorne et al. | In vivo Doppler ultrasound quantification of turbulence intensity using a high-pass frequency filter method | |
Sikdar et al. | Ultrasonic technique for imaging tissue vibrations: preliminary results | |
Tao et al. | An ultrawideband radar based pulse sensor for arterial stiffness measurement | |
JP2009000444A (en) | Ultrasonic diagnostic equipment | |
Xu | Local measurement of the pulse wave velocity using Doppler ultrasound | |
Bonnefous | Blood flow and tissue motion with ultrasound for vascular applications | |
Sikdar et al. | Ultrasonic techniques for assessing wall vibrations in stenosed arteries | |
van den Bos-van de Steeg et al. | In Vivo Comparison of Pulse Wave Velocity Estimation Based on Ultrafast Plane Wave Imaging and High-Frame-Rate Focused Transmissions | |
Lee et al. | An arrayed-range-gate data acquisition for spatial distribution analysis of myocardial tissue vibration from stenosis in coronary Doppler vibrometry | |
Tavoni | Development of New Techniques for Clinical Applications of Jugular Venous Pulse with Ultrasound Devices |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 2005858095 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 1132/DELNP/2007 Country of ref document: IN |
|
WWE | Wipo information: entry into national phase |
Ref document number: 200580029250.7 Country of ref document: CN Ref document number: 2007530355 Country of ref document: JP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWE | Wipo information: entry into national phase |
Ref document number: 1020077007438 Country of ref document: KR |
|
WWP | Wipo information: published in national office |
Ref document number: 2005858095 Country of ref document: EP |
|
DPE1 | Request for preliminary examination filed after expiration of 19th month from priority date (pct application filed from 20040101) |