Multifrequency mapping catheter and mapping method

Abstract

The present invention is entitled "Multi-frequency mapping catheter and mapping method." The present disclosure provides systems, devices, and methods including a catheter configured to be inserted into an intracorporeal chamber of a patient. An ultrasound transducer array including a plurality of multi-frequency ultrasound transducers can be disposed on the catheter. Each multi-frequency ultrasound transducer of the plurality of multi-frequency ultrasound transducers can be configured to transmit wide-beam ultrasound signals and narrow-beam ultrasound signals, and can be further configured to receive wide-beam echo signals and narrow-beam echo signals. A processor can be configured to detect a free space of the intracorporeal chamber by processing the wide-beam echo signals and the narrow-beam echo signals.

Description

[0001] Incorporated by reference

[0002] This application claims the benefit of U.S. Provisional Application 62 / 908204, filed September 30, 2019, which is incorporated herein by reference as fully illustrated. Technical Field

[0003] This application provides systems, apparatus, and methods for mapping internal cavity spaces. Background Technology

[0004] Medical conditions such as arrhythmias (e.g., atrial fibrillation (AF)) are often treated with in vivo surgery. For example, ablation is used to perform pulmonary vein electrical isolation (PVI) from the left atrium (LA) body for the treatment of AF. PVI and many other minimally invasive catheter insertion procedures require real-time visualization and tracking of the in vivo surface.

[0005] Visualization and tracking of the body's surface can be performed using activation wave mapping, fluorescence microscopy, computed tomography (CT) and magnetic resonance imaging (MRI), as well as other techniques that may require significant time or resources to provide visualization and tracking. Summary of the Invention

[0006] This disclosure provides systems, devices, and methods including catheters configured for insertion into a patient's internal cavity. An array of ultrasound transducers, comprising a plurality of multi-frequency ultrasound transducers, may be disposed on the catheter. Each of the plurality of multi-frequency ultrasound transducers may be configured to transmit a wide-beam ultrasound signal and a narrow-beam ultrasound signal, and each of the plurality of multi-frequency ultrasound transducers may be further configured to receive a wide-beam echo signal and a narrow-beam echo signal, respectively, in response to the wide-beam ultrasound signal and the narrow-beam ultrasound signal.

[0007] In one aspect, the processor is configured to identify free space within the internal cavity by processing the wide-beam echo signal and the narrow-beam echo signal. Attached Figure Description

[0008] A more detailed understanding can be obtained by referring to the accompanying drawings and giving examples, wherein:

[0009] Figure 1 This is an example diagram of a catheter-based cardiac mapping system.

[0010] Figure 2 This is an exemplary illustration of a basket-shaped conduit with an ultrasonic transducer.

[0011] Figure 3 This is an example diagram of projection mapping technology.

[0012] Figure 4A This is a graph of an exemplary narrow beam pattern.

[0013] Figure 4B This is a graph of an exemplary wide beam pattern.

[0014] Figure 5 This is a flowchart used for mapping internal cavity chambers.

[0015] Figure 6A This is a graph of an exemplary wide-beam 1.6MHz analog signal template.

[0016] Figure 6B This is an exemplary graph showing the distance between the detection boundary and the simulated reflector.

[0017] Figure 7 This is an exemplary silicon model of the left atrium.

[0018] Figure 8A An exemplary slice with a recording volume having a signal-to-noise ratio (CNR) is shown.

[0019] Figure 8B An exemplary fitting of the elliptic to the detection boundary data is shown.

[0020] Figure 9 Reconstructed data for left atrial phantom modeling is shown.

[0021] Figure 10A The model created using free-space calibration with a single reflector is shown.

[0022] Figure 10B The model created using free-space calibration with dual reflectors is shown.

[0023] Figure 10C The average absolute summation measurement using a single reflector is shown.

[0024] Figure 10D The mean absolute summation measurement using dual reflectors is shown.

[0025] Figure 10E The coherent summation measurement using a single reflector is shown.

[0026] Figure 10F The coherent summation measurement using a dual reflector is shown.

[0027] Figure 11A A narrow beam pattern according to one embodiment is shown.

[0028] Figure 11B A wide beam pattern according to one embodiment is shown.

[0029] Figure 11CThe amplitudes of the template signal and envelope signal over time are shown.

[0030] Figure 12A The diagram illustrates multiple arrays, elliptical obstacles, and free space according to an imaging technique.

[0031] Figure 12B The average, RMS, and maximum values ​​of the elliptical distances for various element pairs and acceptance angles are shown.

[0032] Figure 13A The pixel intensities around the boundary in a given slice are shown for three different imaging methods.

[0033] Figure 13B It shows Figure 13A The corresponding distribution of imaging methods.

[0034] Figure 14A The reconstruction results based on the first simulation are shown.

[0035] Figure 14B The reconstruction results based on the second simulation are shown.

[0036] Figure 15A An elliptic phantom is shown according to one embodiment.

[0037] Figure 15B A phantom of the left atrium according to one embodiment is shown.

[0038] Figures 16A to 16C Each shows various slices of the detected boundary calculated in different regions along different axes using a free-space algorithm.

[0039] Figure 17A The reconstruction results and errors based on different techniques are shown.

[0040] Figure 17B The reconstruction results and errors based on different techniques are shown.

[0041] Figure 18A and Figure 18B The response to a high-frequency signal from the transducer is shown. Detailed Implementation

[0042] As disclosed herein, these systems, devices, and methods provide instantaneous localization of a large number (e.g., thousands) of points on surfaces such as the endocardial surface. The disclosed subjects provide images and reconstructions with comparable detail to those provided by MRI.

[0043] According to the disclosed implementation scheme, a system and method for generating spatial mappings of internal cavities such as cardiac chambers are provided, which features enhanced contrast and employs a rapid acquisition and reconstruction scheme.

[0044] An array of ultrasound transducers (e.g., 64 transducers) is provided, configured to transmit both wide and narrow beams and receive corresponding scattered wide and narrow beams. The wide beam is used to map the wider portions of internal cavities (such as the surface of the heart chambers). The narrow beam is used to map the narrower portions of internal cavities (such as veins leading to and from the heart chambers).

[0045] As used herein, with respect to the term "wide beam" or "wide beam ultrasonic signal," the term "wide" may have a frequency in the range of 1 MHz to 3 MHz. In one embodiment, the term "wide beam" refers to a signal with a frequency within + / - 50% of 1.4 MHz. As used herein, with respect to the term "narrow beam" or "narrow beam ultrasonic signal," the term "narrow" may have a frequency in the range of 5 MHz to 9 MHz, or 12 MHz to 16 MHz. In one embodiment, the term "narrow beam" refers to a signal with a frequency within + / - 50% of 7.4 MHz.

[0046] Wide-beam or wide-beam ultrasound signals may include beamwidths ranging from 100 to 150 degrees. In one aspect, a wide beam has a width of at least 40 degrees or greater. Narrow-beam or narrow-beam ultrasound signals may include beamwidths ranging from 4 to 12 degrees. Those skilled in the art will understand that the precise values ​​for wide and narrow beams can vary depending on the specific medium in which the beam is transmitted or the geometry being calibrated.

[0047] Although this document describes a 64-element transducer array, those skilled in the art will understand that the number of transducers can vary and may include fewer than 64 or more transducers.

[0048] The disclosed implementations include an ultrasound mapping system and an algorithm for accurately mapping the entire surface of an endocardial chamber in a single scan (e.g., a single heartbeat). This type of mapping, along with the simultaneous mapping of electrical activity at all points on the chamber surface, results in a significant reduction in overall operative time across all scenarios, including complex arrhythmias. The disclosed subject matter provides the ability to reconstruct anatomical structures within several scans by scanning the entire spatial volume with beams of different patterns and rapidly combining the resulting information to locate the chamber boundaries.

[0049] This invention discloses arrays of multi-frequency ultrasound transducers mounted on geometrically shaped catheters (e.g., basket-shaped spherical catheters). These arrays support geometric mapping of the entire endocardial surface, including the pulmonary vein orifices, as well as other veins and small components associated with the endocardial surface. As further disclosed herein, the endocardial surface can be mapped using Radon back projections of nonlinear elliptical extensions.

[0050] In some embodiments, an ultrasound spatial mapping basket catheter is provided. This ultrasound basket catheter may be able to acquire data to map the entire surface of a heart chamber from a single location within the chamber, for example, without requiring rotation or movement throughout the chamber during the process. An improved reconstruction method, used in conjunction with this ultrasound basket catheter, is also provided, capable of reconstructing and presenting spatial mapping maps that enhance visual contrast within the chamber, including the main myocardial surface and narrower surfaces such as veins (e.g., pulmonary veins).

[0051] In some embodiments, the ultrasound basket catheter may be equipped with multiple multi-frequency ultrasound transducers coupled to slats forming the basket shape. The transducers are sparsely distributed on the slats and operate in amplitude mode (A-mode) to generate and acquire echo signals. These echo signals are then processed by a processor for preparing cardiac mapping. In one aspect, the processor uses a free-space method. In another embodiment, the processor uses an elliptic back-projection method to rapidly generate a spatial mapping map. Generally, the elliptic back-projection method is essentially the same as synthetic aperture beamforming, which is the same as delayed and summed beamforming. This type of projection operates based on the principle that if an array is used to examine a point in space, it is possible to transmit from all transducers in the array with a time delay, such that all waves will strike that point simultaneously. This concept involves focusing the beam to that point. The same concept can be used to receive beams or echoes. These events are not physically transmitted and received, but can be performed via software because sound waves are inherently linear. Therefore, an echo or response signal is transmitted from the first transducer and received at all other transducers in the array, and the same operation is performed on the second, third, and so on. Data is collected from all transducers. When focused to a point, a time shift is performed so that all signals strike that point simultaneously or at the same time. The reflectivity value at that point is the sum of these time-shifted signals at the appropriate time value. This concept is incorporated into the Boundary Reflectance Value (BRV), which is described in more detail herein.

[0052] In one aspect, the free-space mapping method disclosed herein is a variation or modification of the ellipsoidal backprojection method. The ellipsoidal backprojection method uses a synthetic aperture to align the beam temporally so that computation can be used to focus the beam, thereby finding the reflectivity of a small region in space. The disclosed method finds an upper bound on the reflectivity, which provides a more robust imaging technique.

[0053] Wide-beam echo signals are used to map the surface of the main myocardium, and narrow-beam signals are used to map narrow surfaces such as veins. The term "multi-frequency transducer" is used herein to refer to a transducer configured to transmit more than one frequency. In one embodiment, the multi-frequency transducer is a dual-frequency transducer, i.e., a transducer with two or more frequencies. In other embodiments, the multi-frequency transducer transmits more than two frequencies.

[0054] As disclosed, the ellipsoidal back-projection method assumes a finite number of transducers transmitting / receiving ultrasound, where each transducer transmits echo signals one at a time, and all transducers receive scattered echo signals. In one aspect, these signals are transmitted over a few microseconds. The acquisition time is determined by the size of the area being imaged. In one embodiment, the parameter setting period is approximately 70 microseconds for a range of approximately 4 cm for each transmission-reception event. The position of the scatterer within the volume and the amplitude of reflections from the scatterer are calculated based on ellipsoidal calculations.

[0055] The accuracy of cardiac chamber spatial mapping benefits from faster acquisition and reconstruction times. In one embodiment, the method disclosed in this invention reduces computation time by generating synthetic spatial maps based on the analysis of a single-dimensional A-mode signal, rather than attempting to perform unnecessary and computationally demanding full image reconstruction. The systems and methods disclosed in this invention combine (i) a basket ultrasound array that instantaneously acquires multiple ultrasound measurements with (ii) an improved reconstruction scheme (including wide and narrow beam signals), which can benefit physicians by enabling them to receive more accurate spatial and functional mappings of intracorporeal chambers such as cardiac chambers. Therefore, the systems and methods disclosed in this invention enable physicians to perform efficient invasive diagnostic and potentially follow-up sessions.

[0056] Figure 1 This is a diagram of a catheter-based cardiac mapping system 20 according to one embodiment, which includes an ultrasound basket catheter 40. It should be understood that while the basket shape is disclosed throughout, the embodiment disclosed herein can be implemented using catheters of any shape including multiple transducers. System 20 includes a catheter 21 having an axis 22 that can be navigated by a physician 30 to the heart 26 of a patient 28 lying on a table 29. Figure 1 As shown, a physician 30 can insert the shaft 22 through the sheath 23 while manipulating the distal end of the shaft 22 using a manipulator 32 near the proximal end of the catheter and / or by deflection from the sheath 23. As shown in illustration 25, a basket catheter 40 can be fitted at the distal end of the shaft 22. The basket catheter 40 can be inserted through the sheath 23 in a collapsed state and then deployed within the heart 26.

[0057] In one embodiment, the basket catheter 40 may be configured to perform spatial mapping of the heart chambers of the heart 26 by transmitting and receiving wide and narrow echo signals reflected from the surface 50 of the heart chambers. The basket catheter 40 within the heart chambers of the heart 26 is shown in an enlarged view. As shown, the basket catheter 40 may include an array of ultrasonic transducers 48 coupled to slats forming a basket shape.

[0058] The proximal end of catheter 21 may be connected to console 24. Console 24 may include processor 41 (such as a general-purpose computer) having suitable front-end and interface circuitry 38 for transmitting and receiving signals to and from catheter 21, as well as other components for controlling system 20. In some embodiments, processor 41 may be further configured to receive multi-frequency (e.g., wide and narrow) echo signals and calculate a mapping of the cardiac chamber surface based on the echo signals. In one aspect, processor 41 is configured to identify a first region of the intracardiac chamber by processing wide-beam echo signals and a second region of the intracardiac chamber by processing narrow-beam echo signals. In other words, wide-beam echo signals are configured specifically for mapping certain portions of the chamber, and narrow-beam echo signals are configured specifically for mapping other portions of the chamber. When used in combination, the two sets of echo signals provide a complete mapping of the chamber.

[0059] In one implementation, the surface surrounding the anatomical structure can be presented to the physician 30 on the display 27, for example, in the form of a mesh diagram 35.

[0060] As described above, processor 41 may include a general-purpose computer that can be programmed with software to perform the functions described herein. This software may be downloaded to the computer electronically via a network, or alternatively or additionally set and / or stored on a non-transitory tangible medium, such as magnetic storage, optical storage, or electronic storage. Figure 1 The exemplary configurations shown are chosen for clarity of concept. The subject matter disclosed herein can be used in a variety of applications, not limited to mapping a patient's heart or mapping anatomical objects. Other system components and setups can be used to similarly apply the techniques disclosed herein. Additionally, system 20 may include additional components, such as those for electrophysiological mapping and / or ablation. Although the illustrated embodiment specifically relates to an ultrasound basket catheter for cardiac mapping, the elements of system 20 and the methods described herein can alternatively be applied to ultrasound mapping using catheters with other multi-arm geometries.

[0061] Figure 2This is an example diagram of a basket-shaped conduit 40 equipped with an ultrasonic transducer 48 according to one embodiment of the invention. As shown, the transducer 48 may be coupled to a slat 49 and may be formed into a shape, such as a basket. The transducers 48 may be sparsely distributed (e.g., with a large gap between each of two adjacent transducers, such that the basket-shaped surface defines a large portion of the surface of the slat 49 without transducers). The transducers 48 may be distributed in an approximately spherical pattern outside the distal end of the axis 22. An array of transducers 48 can achieve the desired coverage and detail of cardiac chamber features, although the transducers are sparse, because each transducer 48 may have a sufficiently large transmit and receive angle, and calculations performed based on the signals from and to the transducers can be optimized to utilize the sparse array, as further disclosed herein.

[0062] Figure 2 The catheter 40 shown is merely an example. The number and arrangement of transducers 48 can vary. Additional elements such as electrodes can be disposed on the slats 49. In one embodiment, there are sixty transducers and the slats are formed as flexible PCB slats glued to a nylon balloon. Other catheter geometries (e.g., spiral arms, balloons, etc.) can be provided.

[0063] In one aspect, an apparatus is provided comprising: a catheter 40 configured for insertion into a patient's internal cavity; and an ultrasound transducer array comprising a plurality of multi-frequency ultrasound transducers 48 disposed on the catheter 40. Each transducer 48 is configured to transmit a wide-beam ultrasound signal and a narrow-beam ultrasound signal, and each transducer 48 is configured to receive a wide-beam echo signal and a narrow-beam echo signal in response to the wide-beam ultrasound signal and the narrow-beam ultrasound signal. A processor 41 is configured to detect free space of the internal cavity by processing the wide-beam echo signal and the narrow-beam echo signal. In one aspect, the processor 41 is configured to detect free space by determining a boundary reflection value (BRV), and the BRV indicates whether a specific point in the space within the internal cavity is in free space. The processor 41 is configured to identify free space based on at least one of signal directivity or signal strength, both of which will be described in more detail herein. A monitor or display 27 is configured to show the free space.

[0064] In one aspect, the narrow-beam ultrasonic signal has a frequency in the range of 12 MHz to 16 MHz, and the wide-beam ultrasonic signal has a frequency in the range of 1 MHz to 3 MHz. In another aspect, the wide-beam ultrasonic signal has a beamwidth of at least 40 degrees, and the narrow-beam ultrasonic signal has a beamwidth in the range of 4 degrees to 12 degrees.

[0065] This document also discloses a method comprising inserting a catheter 40 into a patient's internal cavity. The catheter 40 includes an array of ultrasound transducers comprising a plurality of multi-frequency ultrasound transducers 48. The method includes transmitting a wide-beam ultrasound signal and a narrow-beam ultrasound signal from each of the multi-frequency ultrasound transducers 48, and receiving a wide-beam echo signal in response to the wide-beam ultrasound signal and a narrow-beam echo signal in response to the narrow-beam ultrasound signal. The method includes identifying the free space of the internal cavity by processing the wide-beam echo signal and the narrow-beam echo signal.

[0066] According to one embodiment, a two-dimensional intraatrial ultrasound array (e.g., a basket-mounted array) may include N A collection of transmitting / receiving elements Ω ,like Figure 2 The array is shown via transducer 48. It may operate in a synthetic aperture-like mode, which includes N transmit-receive events, where a single element transmits, followed by all elements jointly receiving the transmission from the single element. As further disclosed herein, such an implementation can lead to the identification of the location and reflection coefficient of a scatterer at the endocardial surface, where scattering can occur at a point on the endocardial surface.

[0067] Each transmitter s can be located at position x. s ∈ R 3 Location, and can be of the form Px s (r, t) = p0(t) 1 / 4 The source of the spherical pressure wave rδ(r − ct), where r is the distance from the transmitter, c is the speed of sound in the medium, R is the set of real numbers, and p0(t) is the signal waveform. As used herein, the terms “transmitter,” “transmitter-receiver,” or “transmitter-receiver pair” are generally used to refer to the transducer.

[0068] Waves can come from x sc The point scatterer sc at that location scatters (also called echo) and is scattered by x re The transducer element re at point X receives the signal. Angle Θ(el, X) can be the angle between the normal to the surface of element el ∈ Ω and the vector from the center of the element to point X. Dir(Θ, ka) can be a directional factor that attenuates the signal as a function of Θ, wavenumber k, and element radius a. The signal strength can be expressed as... sc The measurement response received at the receiver when a scatterer is present. This response is a function of time t at x. re The signal strength measured at this location will be:

[0069] (Formula 1)

[0070] In the above formula, P0(s) is the peak envelope (absolute value of complex envelope) of the signal transmitted from s, P1 is the scattering coefficient of the scatterer at sc, c is the speed of sound in the medium, and D(x, y) = ||x – y|| represents the Euclidean distance between two points x and y.

[0071] For the transmitter-receiver pair s, respectively at x s ∈ R 3 and x re ∈ R 3 re ∈ Ω at a given location satisfies the formula c·t s,sc,re = D (x s , x sc )+ D (x sc , x re All points x) sc ∈ R 3 Given an ellipse E ∈ R 3 Such that its two s and re lie in its length c·t s,sc,re On its long axis.

[0072] Based on experiments, as further disclosed herein, it has been shown that from x s All signals transmitted and scattered from all scatterers sc ∈ E(s, re, sc) will be in x re Simultaneous and coherent reception.

[0073] Additionally, data is transmitted from s and in t s,re,sc The maximum signal envelope S(s, re, sc) received at re after 1 second (which has been scattered from all reflectors sc ∈ E(s, re, sc)) is given by the following integral:

[0074] (Formula 2)

[0075] In the above formula, P 0( s ) is assumed to be positive and is s irrelevant from s The maximum value of the envelope of the transmitted signal, and P 1( X )yes X The reflection coefficient at the location. Equation 2 provides the signal strength based on all scatterers located on a given ellipsoid, such that the signal strength value provided by Equation 2 corresponds to the signal strength actually measured at the receiver at a given time.

[0076] According to the implementation disclosed herein, free space is a set F of all points X with respect to P1(X) = 0. Any region including the blood pool will exhibit minimum reflectivity, such that, as further described herein, after calculating the boundary reflectance values, a segmentation technique can be applied to remove such blood pool regions from the surface calculation. The segmentation technique can be applied to the BRV to detect free space (e.g., blood pool regions) based on the surface calculation, as further provided herein.

[0077] Regarding BRV, the disclosed subject matter provides the ability to integrate different beam patterns (i.e., directivity) to image and detect openings or holes (i.e., pulmonary veins) using narrow beams, and to use beams of different widths for better shape observation. Certain aspects of the imaging technique are substantially disclosed in U.S. Patent Publication 2019 / 0209089, which is incorporated herein by reference as if shown in its entirety herein. Generally, BRV examines a collection of time-shifted signals, but focuses on the minimum of the envelope (i.e., energy) of all signals at the appropriate time rather than calculating the sum, which is used in elliptic back-projection techniques. The principle of BRV is based on the concept that if even if a transducer / transmitter-receiver pair does not receive an echo (i.e., a noise level signal) at a particular time, then all points with corresponding transit times (i.e., from the transducer to a point in space and back to the receiver) are in free space.

[0078] The segmentation technique can be any applicable technique (e.g., region growing) that distinguishes blood pools with low signal intensity values ​​from surfaces with high signal intensity values. Given that the entire set Ω of transducer elements is in free space and since the acoustic impedance of myocardial tissue is greater than that of blood, P1(X) ≥ 0 for all scatterers. The triplet (s, re, sc); sc ∈ E(s, re, sc); and s, re ∈ Ω define all elliptic bodies passing through the point sc of all transducer elements whose focus is Ω. Given the above, since the integral in Equation 1 is non-negative, if S(s, re, sc) = 0, then P1(x) ≥ 0. sc ) = 0, such that sc ∈ F, where point sc is in free space.

[0079] Figure 3 An exemplary illustration of an elliptical Radon projection is shown, where point sc 350 is the actual scattering element, and sc'351 lies in free space. The two ellipses 310 and 320, induced by the pairs (s 301, re 302) and (s 301, re' 303), are in... sc' 351 places instead sc There are 350 intersections.

[0080] Formula 3 below takes into account the distance and directionality at a specific point:

[0081]

[0082] (Formula 3)

[0083] In Formula 3, when s, re ∈ Ω, then P1(X) ≤ For all X, s, re.

[0084] For all points X, BRV can be defined as:

[0085]

[0086] (Formula 4)

[0087] As defined above, BRV is measured as an A / D voltage proportional to the received acoustic pressure. BRV can be calculated from a segment of volume (e.g., the detected free space). For example, BRV can be calculated by dividing the volume (10 cm) centered in the middle of a conduit (e.g., a basket conduit). 3 The calculation is performed using a 3D 2mm mesh for sampling. The volume can be divided into connected free spaces using a region growing algorithm (e.g., Equation 2), with the seed located at the center of the volume. Voxels at the periphery of the resulting free space can also be referred to as the detected boundary points.

[0088] In other words, similar to CT or MRI, the BRV is calculated in space using a voxel grid to generate a 3D image. This image is used to segment the space of interest into free space and tissue, which gives a set of points representing the surface of the chamber on the boundary. These can then be processed using additional high-level algorithms, such as model-based fast anatomical mapping (mFAM), smoothing methods, or neural networks, to generate realistic surface images.

[0089] BRV assigns a low value (i.e., noise value) for always free locations in the blood pool and a higher value for potentially occupied locations (i.e., at least a single scatterer P1). These values ​​will vary due to relative distance, directionality, etc. They are also affected by system parameters such as signal-to-noise ratio (i.e., frequency voltage, averaging, filtering, etc.) and the reflective tissue itself. In some applications, the signal-to-noise ratio (SNR) can be 10, meaning a signal of approximately 500-1000 with a noise level less than 100.

[0090] The left atrial boundary is centered at 10*10*10cm in the middle of the catheter. 3 Starting with a volume of 2*2*2mm, and using 3Sampling is performed using a 3-D mesh with voxel resolution. In one implementation, the BRV is first calculated for each voxel in the volume. Then, in another implementation, the volume is partitioned into connected free space using a region growing algorithm, with the seed located at the center of the volume. The growing algorithm is used to define a locally adaptive threshold for the BRV. A region is considered free space if it falls below a predetermined threshold. Voxels at the periphery of the resulting free space are referred to as detected boundary points.

[0091] The disclosed embodiments of the subject matter can utilize both wide and narrow beams to map internal cavity spaces. Each element can be configured to transmit both a wide and narrow beam and receive the corresponding wide and narrow beams. The wide beam may have a first directivity or be within a range of a first directivity, and the narrow beam may have a second directivity or be within a range of a second directivity. A narrow beam can be used in addition to a wide beam because a wide beam may be too wide to distinguish between orifices (e.g., pulmonary veins (PV)) and the surfaces of internal cavity spaces (e.g., endocardial surfaces), as may be the case in situations where there is no elliptic body completely intersecting the interior of the PV.

[0092] At a general level, the free-space method essentially involves using a highly sparse array of ultrasound elements (i.e., catheter 40 and transducer 48) to transmit signals, and then determining whether the transducer is in free space (i.e., in the blood or not in contact with cardiac structures) or in occupied space (i.e., in contact with cardiac structures) based on the echo or reflected signals. Wide-beam echo signals will have one set of characteristics and narrow-beam echo signals will have another set of characteristics.

[0093] The free-space method combines extended nonlinear beamforming and surface rendering. This invention provides an analysis scheme to determine the number and frequency of array elements required to achieve the desired segmentation accuracy. The catheter 40 disclosed herein simultaneously captures hundreds of surface points or scans. Using three frequencies, each scan takes less than 16 ms, providing real-time tracking of the endocardial surface. The boundaries representing the chamber surfaces are determined by segmenting the BRV sampled in a voxel grid using a region growing algorithm. Boundaries can be tracked in a timely manner to provide real-time motion imaging.

[0094] Additional information about free space can be acquired using the dependence of the transmitter directivity pattern as a function of wavelength. Based on this configuration, both narrow and wide beams can be used to map or identify chamber regions smaller than the main chamber. The directivity of the piston model is proportional to the following formula:

[0095]

[0096] (Formula 5)

[0097] In Equation 5, J1 is a first-order Bessel function. Directivity can produce a wide beam suitable for elliptic back projection or a narrow beam providing laser-like scanning access to narrower portions of internal cavities (e.g., PV). Beams with intermediate widths can add information to elliptic back projection.

[0098] Figure 4A An example of a narrow beam pattern (at 13 MHz) with a directivity of 26.5 ka is shown. Figure 4B An example of a wide beam pattern (at 1.8 MHz) with a directivity of 3.67 ka is shown. Normalized intensity values ​​are shown on the right side of these figures. A narrow beam can be used in the first echo detection mode, such that a given space is free until it strikes the first echo. The transmission and reception response information from such a narrow beam is added to a wide beam scan to obtain mapping information of narrower portions of the intracardiac cavity, such as veins (e.g., PV).

[0099] In one aspect, a narrow beam is used for the first echo mode. The system then checks for the first instance or time when the signal exceeds a noise level threshold and sets that instance or time as an impact. This time corresponds to the distance. The transducer position, orientation, and beam shape are all known variables. Regarding the beam shape, it is assumed that the beam is a narrow cone with a known angle, which is calculated using directivity (i.e., the angle is defined as 6 dB downwards from the center, which is approximately 10 degrees for 13 MHz). All spaces in the cone from the transducer are considered free until an impact occurs, while the area around the impact is occupied. This information is added to the segmentation mapping, where free spaces detected by the narrow beam are given higher priority than those by the wide beam. In other words, spaces considered occupied by the wide beam and free for the narrow beam are ultimately determined to be free. This takes into account the various smaller pathways or veins mapping to and from the chamber.

[0100] Beam directivity depends on wavelength. Shorter wavelengths emit narrower beams, while longer wavelengths emit wider beams. Additional information is obtained or determined based on this configuration, as a wide beam is typically insufficient to detect pores in the endocardial surface (i.e., pulmonary vein orifices), which would require the elliptic to exist within fully contained free space. Using the techniques disclosed herein, points in the space where the directivity attenuation exceeds 6 dB are ignored for either the transmitter or receiver, and therefore their contribution to the signal is negligible. Continuously varying the beamwidth between beams for the elliptic approach to narrow the beam provides a laser-like scan into the pulmonary veins, thus providing improved imaging. Beams with a mid- or intermediate width can also be used to provide additional information compared to wide and narrow beams.

[0101] Figure 5A flowchart for mapping an intravascular cavity according to an embodiment disclosed herein is shown. As shown, at step 510, a catheter including an array of ultrasound transducers (e.g., Figure 2 A basket-shaped catheter (40) is inserted into the patient. An ultrasound transducer array is inserted into a cavity within the body, such as a heart chamber. At step 520, transducers within the ultrasound transducer array transmit wide and narrow beam echo signals. At step 530, after the echo signals are reflected from free-space scatterers and actual scatterers, the echo signals are received at one or more transducers within the ultrasound transducer array, including the main myocardial surface and narrower surfaces such as veins (e.g., pulmonary veins).

[0102] At step 540 of the flowchart, the processor calculates the reflection amplitude based on the techniques disclosed herein to segment the boundaries of internal chambers (e.g., the endocardium). In one aspect, the processor is configured to combine multiple scans during catheter movement. Using these scans, a model-based rapid anatomical mapping (mFAM) can be created by matching a statistical model with data collected from the catheter. In one aspect, a combination of imaging techniques is used, where free space takes precedence over occupied space. In another aspect, for each point in the space of the sampling grid, a minimum operator is applied to the BRV values ​​from all scans. A region growing algorithm is used to provide boundary voxels between free space and chamber tissue. To further improve this imaging technique, anatomically aware algorithms, such as mFAM or neural network imaging tools, can be used.

[0103] At step 550, the segmented internal cavity is presented via a display. The display, monitor, or other type of output device may be configured to receive and display the segmented internal cavity data.

[0104] Based on computer simulation experiments, a 64-element transducer with a radius of 30 mm was inserted into a left atrial model and placed at the center of a 10 cm side cube. The cube was divided into 503 voxels using a 2 mm 3D mesh. The reflector was assigned to all grid points within a 6 mm thick shell surrounding the atrial boundary of the heart chamber. Assuming a Gaussian envelope... . 6MHz and 7 . A signal 611 consisting of several cycles of a 4MHz sine function, such as... Figure 6A As shown in Figure 610. For each transducer element pair, the response of each reflector is appropriately time-shifted and summed to generate an analog signal according to Equation 1. The analog signal is used to reconstruct the volumes of wide and narrow responses according to the embodiments disclosed herein. Figure 6BThe curve 620 shows the distance between the detection boundary and the simulated reflector, which describes the two-sided normalized distance probability. When the distribution is skewed, the median and median absolute deviation (MAD) of the distance (in mm) for the distance from the detection boundary to the reflector 621 respectively produce a 1 . 99 and 0 . 983, and 3 are generated respectively for the reflector to the boundary 622. . 377 and 1 . 26. The detected distances from the boundary to the reflector and from the reflector to the boundary are estimated by errors from the model surface.

[0105] An array of 64 piezoelectric transducers mounted on a twelve-slatted printed circuit board (PCB) basket was tested in vitro. Each transducer element comprised two resonant frequencies at approximately 1.6 MHz and 7.4 MHz, respectively. The elements were connected to an acquisition system capable of generating pulses and recording scans of sequential transmissions for all elements in real time. A catheter was placed within a phantom chamber, including a water bath to simulate blood, and encapsulated in an elliptical plastic package to simulate the walls of the heart chambers. The walls of the phantom chambers were forced to move continuously by pumping water inward and outward to simulate the natural wall motion of the heart. Long-term averages were then subtracted from all time signals to eliminate constant signals caused by reflections from the slats themselves. In all simulations, the transducer positions were assumed to be known. Figure 7 An exemplary left atrial silicon phantom is shown that can be used to test the embodiments disclosed herein.

[0106] Based on the ellipsoidal phantom experiment, a basket-shaped conduit was placed in a container made of silicon and water with a radius of 35 mm and 32 mm. . 5mm and 22 . In a 5mm elliptical phantom, as described in this paper, a wide beam is used for acquisition. The volume is dynamically reduced using a square root function, ensuring that the square root of each acquired voxel within a given volume is calculated. The elliptical is then fitted to the boundaries detected by the algorithm. Figure 8A Slices of the recorded volume are shown, where the signal-to-noise ratio (CNR) was calculated for a group of pixels 6 mm around the boundary. Element 810 shows a CNR of 1.88 (slice along the height axis) with a Z value of 15, element 820 shows a CNR of 1.62 with a Z value of 25, and element 830 shows a CNR of 1.82 with a Z value of 35. Figure 8A The value to the right of the value is the intensity value.

[0107] Figure 8B The fit and deviation of the detection boundary are shown. Figure 8B The right side shows the distance values ​​in mm. As shown in the figure, ellipse 850 is fitted to the detected boundary data. Figure 8BOnly the first half of the volume is shown. (Example) Figure 8B As shown in elliptic 850, the bottom portion of the fit and deviation are outliers because the phantom used for this experiment is not a complete elliptic. Therefore, the bottom portion causes the error distribution, median, and MAD (which pass through approximately 1) to be outliers. . The factor estimate of 5 (standard deviation of the Gaussian distribution) is skewed. Based on this experiment, radii of 36 and 33 are obtained. . 4 and 25 . 5, while the deviation from the median of the elliptic in the data is 1. . 4mm and MAD is 0 . 6283mm.

[0108] Based on the left atrial phantom experiment, narrow and wide beams were used to scan such... Figure 7 The silicon left atrial phantom is shown. The detected boundary points, along with six manually marked anatomically identified points of interest (POIs), were used as input to an anatomically aware, model-based reconstruction algorithm. The reconstructed anatomical structure was aligned with a reference CAD mesh, and the difference from each reconstructed vertex to the reference mesh was calculated. The resulting mean, standard deviation, median, MAD, and RMS were 4.2 mm, 4.4 mm, 2.6 mm, 1.77 mm, and 6.1 mm, respectively. Figure 9 The reconstruction results and errors are shown in the figure. Figure 9 As shown, the right side indicates the distance in mm. Figure 9 Element 910 shows a top view of the detected boundary-based model's distance (in millimeters) from the atrial phantom. Figure 9 Element 920 shows a side view of the detected boundary-based model's distance (in millimeters) from the atrial phantom. Figure 9 Element 930 shows a left PV view of the detected boundary-based model's distance (in millimeters) from the atrial phantom. As shown, the resulting model visually approximates the phantom, and the detected PV is consistent with the CAD. In the silicon phantom of the left atrial shape, using the systems, apparatus, and methods disclosed herein, images are accurate to within at least 4.0 mm.

[0109] In one aspect, the array operates in synthetic aperture mode, comprising multiple (N) transmit-receive events. In this mode, each element in the array transmits sequentially once, while all elements receive simultaneously. This generates a set comprising (N(N+1)) / 2 distinct RF signals. According to the embodiment disclosed herein, the scatterer x sc Position and reflection coefficient P1(x) scThe catheter is located at the endocardial surface boundary and is located using this signal set. In one aspect, all signals in the set are sampled to define a specific location in space. In other words, if one signal does not receive an echo from a specific location, it is determined or assumed that free space exists in that region. If a low-value signal is generated, it is determined or assumed that the catheter is located in the blood.

[0110] Generate a graph showing the relative distances to the signal amplitudes. For example... Figure 18A and Figure 18B As shown, the RF response of the high-frequency signal from the transducer is measured. Figure 18A Indicates the RF response after applying a median filter to fifteen signals. Figure 18A The arrow in the image indicates the excitation signal. Figure 18B The absolute value of the amplitude when the excitation signal is removed is shown. Figure 18B The circled area indicates the first impact, and the arrow indicates the dead zone.

[0111] At time t d After that, in x re Received at point P1(x) sc From x sc The point reflector reflects the x-axis. s The signal emitted by the transducer element at that location, the timing of which is defined by the following formula 6:

[0112]

[0113] (Formula 6)

[0114] In Formula 6, c = 1.54 mm / µs, which is the speed of sound in blood, and t is the value of the following term. d (s, re, sc) Delayed replicas:

[0115]

[0116] (Formula 7)

[0117] In Equation 7, P0 is the transmit / receive peak amplitude for transmit and receive directivity correction, as defined by Equation 8 below:

[0118]

[0119] (Formula 8)

[0120] Equation 8 corresponds to the directionality factor that attenuates the signal as a function of the incident angle Θ with respect to the transducer surface. In Equation 8, the wavenumber k = (2πf / c), J1 is a first-order Bessel function, and the element size a = 0.5 mm. According to Equation 8, these values ​​are divided by the distance between the transmitter scatterer and the receiver and multiplied by the scattering coefficient P1(x). sc ).

[0121] Imaging system quality is typically measured by the response to a single-point reflector. In one aspect, the response to the reflector is the point spread function (PSF). In one experiment, a point spherical reflector was placed in front of a planar rectangular array of nine elements. The distance from the array to the reflector was four to five times the distance between individual array elements. Figures 10A to 10F The PSF and responses of two reflectors were compared between the free space (FS) and envelope and coherent summation (DAS) methods. Figures 10A to 10F The value on the right indicates the intensity value.

[0122] Figure 10A and Figure 10B Corresponding to free space calibration, Figure 10C and Figure 10D Corresponding to the average absolute summary standard, and Figure 10E and Figure 10F Corresponding to coherent summation calibration. Figure 10A , Figure 10C and Figure 10E Corresponding to a single reflector arrangement, and Figure 10B , Figure 10D and Figure 10F This corresponds to a dual-reflector arrangement. For example... Figure 10A and Figure 10B As shown, the free-space method has a PSF with minimal array processing artifacts and distinguishes between two reflectors. Figures 10C to 10F As shown, the DAS experiences up to 33% of the reflector signal across the entire space, and also up to 40% blur between two reflectors. If the element's positional accuracy is less than the subwavelength of a non-rigid catheter, only the absolute summation is feasible and results in relatively large artifacts. In other words, it is assumed that the left atrium and other chambers are bodies that can be closely approximated as ellipsoids. Therefore, this assumption provides a relatively accurate approximation of the summation performance of the actual system.

[0123] Regarding the analysis framework, the implementation scheme disclosed in this paper calculates the free space and measures the boundary distance error based on a circular array in a plane within the elliptical center of the obstacle. For low-curvature chambers, the cross-section of a uniformly distributed array can be approximated using the following method. For other types of chambers, integration of several viewpoints can be used.

[0124] A transmitter-receiver pair defines a receiving area. Within this area, certain echoes can be detected based on directionality and relative positioning, while others are considered noise for a particular transmitter-receiver pair. Free space is typically identified or detected via the transmitter-receiver pair by determining the timing of the first echo from an obstacle, which reflects off or within the boundary of the receiving area. The transmitter-receiver pair splits into two narrow beams for a very narrow receiving angle, resulting in the number of intersections with obstacles being the same as the number of elements (i.e., the number of transducers). For a very wide receiving angle, the number of associated ellipses for the element pair is approximately proportional to the square of the number of elements or transducers.

[0125] Regarding the 2D mathematical framework, the implementation disclosed herein relies on a 2D array of n elements uniformly placed around a circle of radius r. In one implementation, the radius is 15 mm. Those skilled in the art will understand that the radius can vary depending on different parameters and conditions. Each element receives an angle of φ. Points outside the circle are discarded. The array lies within an ellipse E0 with major axes a and b. The process is based on the following values ​​and parameters calculated as the union of geometric objects from the free space detected by the array. Each element or transducer e in the array... i A slice s in a plane defined by the accepting angle of the element. i,ϕ The plane is substantially perpendicular to the surface of the element. Each element in the array is paired with p. i,j Generate corresponding elements to form a wedge that receives reflections from them. The wedge consists of two slices s i,ϕ and s j,ϕ, The intersection points and the following formulas define the constraints:

[0126]

[0127] (Formula 9)

[0128] When i=j, the element pair is the element itself and the wedge is the slice s. j,ϕ The sector. In order to determine the sector by p i,j The free space for detection must be determined by p. i,j The time of the first echo detected. Time and pair combination to define the ellipse E. P And the free space of this pair is wedge W. i,j And ellipse E P The intersection point.

[0129] Based on the method disclosed in this paper, the following observations are provided (Observation 1). The first echo can be received from one of two possible cases: it is located at point x. ∈ E P At location E P Tangent to E0, or located at ∂W i,j ∩ EP The extreme point.

[0130] Observation 1 describes the need to be located in order to find the point that can be reached by p. i,j The points in free space are found. This is solved using the following set of formulas.

[0131]

[0132] (Formula 10)

[0133] Formula 10 requires that the intersection point belongs to the ellipse defined by the transmitter-receiver pair and the obstacle ellipse.

[0134] Equation 11 below indicates that the obstacle ellipse and the transmitter-receiver pair are tangent.

[0135]

[0136] (Formula 11)

[0137] In formulas 10 and 11, p1 and p2 are the positions of the transmitter-receiver pair, and (u, v) is the average value, i.e., the center of the ellipse. Elements a and b are the major axes of the obstacle ellipse. θ b θ It is the major and minor axes of the ellipse defined by the transmitter-receiver pair and the first echo point. Angle θ = ∠ (p1– p2), and angle θ is the rotation angle of the transmitter-receiver ellipse from the positive x-axis.

[0138] Elements x0, α, and a θ There are three equations, which are solved numerically. The sign (i.e., the positive and negative roots of the fundamental elliptic equations) is chosen to generate the region W that accepts the equation. i, j The smallest intersecting ellipse, and use observation 1 to calculate the free space. If the first echo is inside the wedge, then it originates from E. P The normal to the obstacle ellipse is equal at the point of tangency, and the curvature of the obstacle ellipse must be less than E. P The curvature of . This relationship applies to any smooth reflective surface. Based on the above relationship, the following theory is provided: Let x be a point on the first echo of Pi,j, and if x ∈ Wi ,j and x ∉ ∂W i,j Then Δx = ΔE P (x) and Δ 2 x ≤ Δ 2 E P (x) (hereinafter referred to as the "Theorem").

[0139] In one aspect, the disclosed subject matter is configured to identify free space based on at least one of spherical pressure waves, directivity, or signal strength. Generally, free space is at the system noise level, while the actual echo is at least five times larger. This relationship and calculation are determined by the system's SNR. All transducers emit spherical pressure waves. This wave attenuates in certain directions due to directivity. There is no attenuation at all along a line starting from the transducer center and tangent to its normal (i.e., the transducer axis). If this line rotates, points in space along the line of rotation will feel less pressure from the wave compared to those points on the axis. This change is a result of directivity. As used in this aspect, signal strength is defined as the measured signal received from the transducer. Signal strength is a direct result of the pressure wave reflected from one or more reflectors in space, where reflectors are typically organizations in this application.

[0140] The complete form of the signal is P1(x) sc ),in:

[0141]

[0142] (Formula 12)

[0143] The signal transmitted from s and received by RE is calculated based on the sum of the magnetoresistances from all scatterers, as shown in the following formula:

[0144]

[0145] (Formula 13)

[0146] In both the classical delay (i.e., the typical DAS beamformer) and the summation (DAS) beamformer P1, x is estimated by summing ψ(s, re, t) over the (s, re) signal for all (N(N+1)) / 2 transducer elements at point sc. sc Thus, the following formula is obtained:

[0147]

[0148] (Formula 14)

[0149] Equation 14 represents the sum of all contributions from the transmitter and receiver pair to a specific point in space within a predetermined time period. The peak envelope of the signal operator s is denoted as Ɛ(s). The envelope is calculated using the absolute value of the signal and taking the maximum value within a relatively decreasing time window ΔW (wavelength width). The transducer / receiver pair (s, re) and the scatterer (sc) define the path through point x. sc Let there be an elliptic E(sc, s, re) with foci at x. s and x reSince the acoustic impedance of blood is lower than that of myocardium, all points sc ∈ V within the blood pool or at the blood pool / myocardial interface are considered acoustically imperceptible. BP , P1(x sc ) ≥ 0. It is assumed that all reflectors are speckled (i.e., speckle is ignored or neglected). Therefore, if no signal is received at time T at a predetermined pair (s, re) (i.e., below the noise level), then all points T are defined by Equation 15, as shown below.

[0150]

[0151] (Formula 15)

[0152] In Formula 15, T corresponds to the elliptic E ( All points on (s, re) with zero reflectivity or in free space. At t=t d When , Ɛ(φ) = P0 and Ɛ(S) is defined as above. Based on these relationships, the following formula 16 is provided:

[0153]

[0154] (Formula 16)

[0155] For all points sc ∈ V BP And the transmitter-receiver pair (s, re), the signal is defined by the following formula:

[0156]

[0157] (Formula 17)

[0158] Since Ɛ(S) is positive by definition, and part of Equation 14 is generated by neglecting or ignoring speckle, the following formula is also true:

[0159]

[0160] (Formula 18)

[0161] For all points BRV is limited as follows:

[0162]

[0163] (Formula 19)

[0164] Figure 11A A narrow beam pattern with a frequency of 7.4 MHz is shown, and Figure 11B A wide beam pattern with a frequency of 1.4 MHz is shown. Figure 11A and Figure 11B The value on the right indicates the normalized intensity. Narrow beams are typically used in first-echo detection mode. In other words, the space being navigated is free until the beam strikes the first echo. Add this information to the information about wide-beam scans. Narrow beams provide greater clarity and analysis within the pulmonary veins. Figure 11C The amplitudes of the template signal and envelope signal over time are shown. Figure 11C In this case, the signal bandwidth is 77%.

[0165] Figure 12A Multiple arrays, elliptical obstacles (represented as outer oval rings), and free space (represented as shaded areas) are shown. The arrays are confined to the innermost point defined by the free space and are indicated by dots. Figure 12A Each subset of the images in the diagram indicates the number of elements as the first value in parentheses, and indicates the accepted angle as the second or subsequent value in parentheses. Figure 12B The average, RMS, and maximum values ​​of the elliptical distance for each element pair and the receiving angle are shown. (Refer to...) Figure 12A and Figure 12B Simulations were performed using arrays with 15 mm radii for 8, 10, and 12 elements operating at angles of π / 3, π / 4, and π / 6. The obstacle ellipse has a major axis ranging from 50 mm to 25 mm. This ratio of the major axis is larger than that of a typical left atrium, but demonstrates the capability of the disclosed subject. Distances (referred to as elliptical distances) were measured from the discovered free space to sampling points along the obstacle ellipse, each less than 0.25 mm apart. The 0.25 mm value is the sampling choice that defines the error measurement. Those skilled in the art will understand that this value can be smaller or larger. Figure 12A The resulting free-space region detected by each array and receiving angle pair is shown in the shaded area, and the union of all three receiving angles of the 10-element array is also shown. This configuration illustrates an example of multi-frequency operation. Figure 12B In the middle, the horizontal line corresponds to approximately 3.5 mm and indicates the accuracy of these configurations.

[0166] In the simulation experiment, the free-space method was compared with the conventional method in silicon. To provide a coherent summation, it was assumed that the positions of the elements were precisely known. A spherical array with 64 elements and a radius of 15 mm within the left atrium was obtained from a computed tomography scan. CT) Segmentation. A cube with a side length of 10cm is divided into 50 segments using a 2mm 3D mesh. 3 Individual pixels. The reflector is sampled in a grid within a 6mm thick shell surrounding the atrial boundary. A template detector records the signal (i.e., 1.4MHz, 77% bandwidth) for simulation, and also includes a time-scaled version for higher frequencies (7.4MHz). This corresponds to signals from... Figure 11CThe value of is not used to evaluate optimal algorithm performance. The template signal is delayed and summed for each transmitter-receiver pair and reflector according to Equation 12 below. Those generated RF signals are used to calculate the BRV for wide-beam and the first echo for narrow-beam. DAS and envelope DAS are also used to calculate the imaging volume. The envelope is calculated as the maximum absolute value within a window corresponding to the loop length. For example, Figure 13A The three methods (which are free space methods) shown collect pixel intensities around the boundary of a given slice within a 4mm radius and compare them as follows: Figure 13B The corresponding distribution is shown. For example... Figure 13A As shown, the closed loop forms the boundary of each slice. The free space boundary is less affected by array processing artifacts and has a much higher signal-to-noise ratio (CNR) compared to the DAS and DAS envelope methods. Figure 13A The value at the bottom is the intensity value.

[0167] Figure 14A and Figure 14B Reconstruction results from two simulations using the free-space method are shown. The reconstruction is close to the left atrial body, and the location of the pulmonary veins is detectable given a sphere generated by a narrow beam. The mesh reconstruction is modeled based on the distance to the reflector. An outer mesh is shown, which represents various patches at different distances from the reconstruction. All distances are clipped at 15 mm. The median and median absolute deviation (MAD) of the distances (in mm) are provided. In Simulation 1, the median distance from the detected boundary to the reflector is 2.27 mm and the MAD is 1.3 mm. In Simulation 2, the median distance from the detected boundary to the reflector is 2.78 mm and the MAD is 1.8 mm. From the reflector to the boundary, Simulation 1 produces a median distance of 3.78 mm and a MAD of 1.68 mm, while Simulation 2 produces median distances of 3.54 mm and 1.53 mm. The distance from the reflector to the boundary is calculated from the middle of the atrial wall shell, which is 2 mm from the endocardium.

[0168] In another aspect, additional in vitro experiments were conducted using a distributed array of 64 piezoelectric (PZT-5H) transducers. In these experiments, each element had dimensions of 1 × 1 × 0.3 mm and was mounted on slats of a spherical basket, as shown. Figure 2 As shown. Each transducer element has two resonant frequencies: a first resonant frequency of approximately 1.4 MHz and a second resonant frequency of approximately 7.4 MHz. The positions of the elements in the array were measured by an optical system with an accuracy of approximately 1.0 mm. The array was operated using a National Instruments (NI) acquisition system configured to generate excitations of arbitrary pulse shapes and simultaneously receive echoes from all elements in the array. The catheter was placed in a water bath designed to simulate a blood pool and encapsulated in an elliptical plastic package designed to simulate the walls of a heart cavity. Figure 15A An elliptic phantom was shown and Figure 15B A phantom of the left atrium is shown. Heart-wall-like motion is generated by continuously pumping water into and out of the chamber. Direct transmission received signals that do not experience reflections from the moving walls are eliminated by subtracting the long-term signal average. The signals from multiple acquisitions are averaged. In one aspect, acquisitions occur approximately fifteen times for low frequencies and three times for high frequencies. This generates an SNR value of at least four.

[0169] In another aspect, additional elliptic phantom experiments were conducted using the disclosed subject matter. In this experiment, a conduit was placed within a water-filled silicon elliptic-shaped phantom. The phantom had radii of 35 mm, 32.5 mm, and 22.5 mm. In one aspect, acquisition was performed using a 1.4 MHz wide beam (i.e., 2.5 cycles). The term "cycle" as used herein refers to a transmitted signal consisting of multiple sinusoidal cycles. Square root signal compression was used to adjust the signal dynamic range. The elliptic was then fitted to the boundaries detected by the algorithm. Figures 16A to 16C This includes multiple slices of the detected boundary calculated by a free-space algorithm at different points along different axes. Figures 16A-16C The result of the BRV operator is shown. Figures 16A to 16C In each of them, the detection boundary is shown as a closed loop around a shaded patch roughly at the center. Figures 16A to 16C Each of the figures shows a series of slices passing through the recorded volume at 15mm, 25mm, and 35mm, perpendicular to the x, y, and z axes. The CNR is calculated for a 6mm sub-volume around the boundary. Because the phantom is not a perfect ellipsoid, the error distribution is skewed. The median and MAD are calculated. MAD is estimated by a factor of approximately 1.5 in a Gaussian distribution representing the standard deviation. The measured radius is relative to... Figure 16A It is 36mm, for Figure 16B It is 33.4mm, and for Figure 16C The diameter is 25.5 mm. The median deviation of the elliptic is 1.4 mm and the MAD is 0.6283 mm.

[0170] In another aspect, additional silicon phantom experiments were conducted to create the shape of the left atrium, such as... Figure 15B As shown. In these experiments, narrow and wide beams were used to obtain scans, and a free-space algorithm was used to detect the boundary points of the phantom. These points were used as input to a model-based fast anatomical mapping smoothing algorithm (mFam).

[0171] The reconstructed anatomy was manually aligned with the reference CAD mesh. The distance from each vertex of the smooth surface to the corresponding reference mesh was calculated. The resulting average distance was 4.2 ± 4.4 mm (compared to 3.5 for FAM), the median was 2.6 mm, the MAD was 1.77 mm, and the RMS was 6.1 mm.

[0172] exist Figure 17A and Figure 17B The reconstruction results and errors are shown in the figure. Figure 17A and Figure 17B The model-based smoothing is shown using BRV, segmentation on BRV, and using the segmentation boundary as input to the mFam algorithm. Figure 17A and Figure 17B The left-handed image in the figure represents the detected boundary-based model's distance from the CAD (in mm). Figure 17A and Figure 17B The right image in the diagram is a CAD reference image with the distance to the reconstruction. The resulting model closely approximates the phantom, and the detected pulmonary veins conform to the CAD mesh, but their size estimation is insufficient, leading to large errors in those areas. This problem can be addressed by performing additional acquisitions. Some errors experienced in left atrial reconstruction are caused by the non-fixed distance between sensors, which can result in errors in sensor position up to several millimeters, which are then translated into reconstruction errors. The subject matter disclosed in this invention improves the calibration of these distances to sub-millimeter accuracy.

[0173] In summary, the elements in a sparse array are not spaced at subwavelength distances and therefore do not pass the Nyquist criterion. Determining element positions to subwavelength accuracy is difficult, leading to additional errors, particularly in the coherent summation. The distinction between the processes, systems, apparatus, and methods disclosed in this invention and DAS is described above relative to experiments (such as in-computer experiments) and by way of array processing artifacts that are readily apparent in the PSF and used to distinguish nearby reflectors. This is in Figures 10A to 10F As shown in the image.

[0174] The subject matter and envelope signal disclosed in this paper significantly improve the detection results, such as at least Figure 13A and Figure 13B As shown. In one aspect, to match the performance of a clinical system, at least ten elements operating at a wide angle of π / 3 are required, as present at any given 2D slice. Unlike other systems that use only narrow beams and perform linear sampling of the endocardial surface, the system disclosed herein samples at a quadratic rate because it examines the received echo by the transducer while obtaining additional normal and curvature information through the theorems disclosed herein. Compared to existing systems that require one minute per region to achieve 3mm accuracy, the disclosed subject accurately maps the entire endocardium within fewer acquisitions performed in real time. In one embodiment, the scan described herein takes less than 16ms.

[0175] The disclosed subject matter is not limited to the left atrium or any heart chamber. The disclosed subject matter can be used in a variety of applications, including applications that utilize arrays of elements to analyze the characteristics of objects such as chambers.

[0176] In summary, in one aspect, a catheter is provided that allows for the detection of electrical activity by real-time imaging and tracking of the propagation of electrocardiogram waves, which enables the mapping of complex arrhythmias and reduces the time required to provide reliable images and data.

[0177] Generally, a specific time in the signal corresponds to the transit time of a catheter within a chamber, which in turn defines the entire elliptic in space. Due to the unique structure of the heart, narrow chambers such as the pulmonary veins are not included in the elliptic and are therefore difficult to detect using known mapping methods. Using narrow beam emission overcomes this problem due to its size, which easily fits within narrower sections of the heart's structure. Additional beams with different directional patterns rely on the same elliptic concept, and slices at acceptable angles are defined by beam directionality. Thus, beam combination provides additional information about the boundary shape with a relatively small acquisition time. For example, each element may require an additional acquisition time of 70 microseconds.

[0178] Any of the functions and methods described herein can be implemented in a general-purpose computer, processor, or processor core. By way of example, suitable processors include general-purpose processors, special-purpose processors, conventional processors, digital signal processors (DSPs), multiple microprocessors, one or more microprocessors associated with a DSP core, controllers, microcontrollers, application-specific integrated circuits (ASICs), field-programmable gate array (FPGA) circuits, any other type of integrated circuit (IC), and / or state machines. Such processors can be manufactured by configuring a manufacturing process using hardware description language (HDL) instructions for processing and the results of other intermediate data, including netlists (such instructions can be stored on a computer-readable medium). The result of such processing can be a maskwork, which is subsequently used in a semiconductor manufacturing process to manufacture a processor implementing the features of this disclosure.

[0179] Any of the functions and methods described herein may be implemented in computer programs, software, or firmware incorporated into a non-transitory computer-readable storage medium for execution by a general-purpose computer or processor. Examples of non-transitory computer-readable storage media include read-only memory (ROM), random access memory (RAM), registers, cache memory, semiconductor memory devices, magnetic media (such as internal hard disks and removable disks), magneto-optical media, and optical media (such as CD-ROMs and digital multi-purpose discs (DVDs)).

[0180] It should be understood that many variations are possible based on the disclosure herein. Although features and elements have been described above in specific combinations, each feature or element may be used alone without other features and elements, or in various combinations with or without other features and elements.

Claims

1. A medical device comprising: A catheter, configured for insertion into a patient's internal cavity; An ultrasonic transducer array, comprising a plurality of multi-frequency ultrasonic transducers arranged on the catheter. Each of the plurality of multi-frequency ultrasonic transducers is configured to transmit wide-beam ultrasonic signals and narrow-beam ultrasonic signals, and Each of the plurality of multi-frequency ultrasonic transducers is configured to receive a wide-beam echo signal and a narrow-beam echo signal in response to the wide-beam ultrasonic signal and the narrow-beam ultrasonic signal. as well as A processor configured to detect the free space of the internal cavity by processing a combination of the wide-beam echo signal and the narrow-beam echo signal.

2. The medical device of claim 1, wherein the processor is configured to detect the free space by determining a boundary reflection value (BRV), and the boundary reflection value (BRV) indicates whether a specific point in the space within the internal cavity is in the free space.

3. The medical device of claim 1, wherein the ultrasonic transducer array comprises at least 64 multi-frequency ultrasonic transducers.

4. The medical device of claim 1, wherein the narrow-beam ultrasound signal has a frequency in the range of 12MHz to 16MHz.

5. The medical device of claim 1, wherein the wide-beam ultrasound signal has a frequency in the range of 1 MHz to 3 MHz.

6. The medical device of claim 1, wherein the wide-beam ultrasound signal has a beamwidth of at least 40 degrees.

7. The medical device of claim 1, wherein the narrow-beam ultrasound signal has a beamwidth in the range of 4 to 12 degrees.

8. The medical device of claim 1, wherein the internal cavity comprises a vein.

9. The medical device of claim 1, further comprising a monitor configured to display the free space.

10. The medical device of claim 1, wherein the internal cavity is a cardiac chamber.

11. The medical device of claim 1, wherein the processor is configured to identify the free space based on at least one of signal directionality or signal strength.

Images