Systems and methods for determining fluid and tissue volume estimates using electrical property tomography
By reconstructing and enhancing the initial impedance image in electrical impedance tomography, identifying the region of interest and determining the correlation parameters, the problem of differential image limitation in the EIT method is solved, and accurate estimation of fluid volume and improvement of diagnostic quality is achieved.
Patent Information
- Application Number
- CN202080052967.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2019-07-30
- Filing Date
- 2020-07-21
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2040-07-21
AI Technical Summary
Existing electrical impedance tomography (EIT) methods usually focus on the generation of differential images, resulting in limitations in medical diagnosis in the absence of absolute characteristics, and reconstruction methods of differential images may overestimate or underestimate the values of certain pixels.
By receiving electrical tomography data, the initial impedance image is reconstructed, the image is enhanced, and the tissue is segmented to identify the region of interest, and the relationship parameters related to the electrical properties within that region are determined, based on these parameters, the fluid volume is estimated.
Accurate estimation of fluid volume is achieved, more accurate absolute impedance images are provided, and diagnostic quality is improved, especially in monitoring respiratory dynamic parameters and detecting pneumothorax.
Smart Images

Figure CN114270397B_ABST
Abstract
Description
[0001] CROSS-REFERENCE TO RELATED APPLICATIONS
[0002] This application is an application entering the national phase of the international patent application PCT / IB2020 / 056851 filed on July 21, 2020, which is destined for the People's Republic of China and was published in English as international patent publication WO 2021 / 014355 A1 on January 28, 2021. According to Article 8 of the Patent Cooperation Treaty, this international patent publication claims the benefits of U.S. patent application serial number 62 / 878,097 filed on July 24, 2019 and U.S. patent application serial number 62 / 880,548 filed on July 30, 2019.
[0003] Pursuant to 35 U.S.C. §119(e), this application claims the benefit of U.S. Provisional Patent Application Serial No. 62 / 878,097, filed on July 24, 2019, and U.S. Provisional Patent Application Serial No. 62 / 880,548, filed on July 30, 2019, the disclosures of each of which are hereby incorporated by reference in their entirety. Technical Field
[0004] The present disclosure relates to systems and methods for determining absolute volume estimates. In particular, the present disclosure relates to systems and methods that include and utilize an electrical property tomography (eg, electrical impedance tomography (EIT)) system to determine absolute volume estimates. Background Art
[0005] Electrical impedance tomography (EIT) is a non-invasive imaging method that can generate an image of a region of interest in a field (e.g., a patient) by collecting data using electrodes arranged along the periphery of the region of interest. One common application of EIT includes clinical applications where tomographic images of the human body may be useful. For example, EIT is routinely used to monitor the cardiopulmonary system, which may be particularly useful for patients being treated in an intensive care unit (ICU) environment.
[0006] During an EIT procedure, an electrical signal (e.g., current) may be injected into the periphery of a region of interest (e.g., a patient's torso) of a domain to be imaged. Electrical properties (e.g., voltage, potential) resulting from the injected electrical signal may be collected at the periphery of the region of interest. Based on the collected data, a map having an estimated value of the electrical property (e.g., impedance) may be generated or reconstructed. EIT systems are generally susceptible to reconstruction artifacts. To mitigate artifacts, reconstruction of EIT images generally employs a differential reconstruction technique in which the generated image (e.g., differential image) utilizes changes between a current property (e.g., impedance) map and a reference impedance map. Although differential images are useful in several scenarios, certain medical diagnoses are limited in the absence of an absolute property (e.g., impedance) map.
[0007] As noted above, current EIT methods generally focus on the generation of difference images. The preference for difference images may be a result of the fact that direct calculation of absolute impedance values from EIT data relies on linear or quasi-linear reconstruction methods, which typically overestimate the values of some pixels in the region of interest and underestimate the values of other pixels. Summary of the invention
[0008] Some embodiments include a method of estimating a fluid volume. The method may include receiving electrical tomography data of a portion of a domain; reconstructing an initial impedance image based at least in part on the electrical tomography data; enhancing the initial impedance image to generate an enhanced impedance image; segmenting the enhanced impedance image to identify one or more tissues depicted within the enhanced impedance image; selecting a region of interest within the enhanced impedance image; determining a relationship parameter that relates an electrical property represented within the region of interest of the enhanced impedance image to one or more properties of the region of interest; and estimating a fluid volume within the region of interest based at least in part on the relationship parameter and the enhanced impedance image.
[0009] Some embodiments include a system for estimating a fluid volume within a domain. The system may include an electrical tomography system and a volume estimation system. The fluid volume estimation system may include at least one processor and at least one non-transitory computer-readable storage medium having instructions stored thereon, the instructions, when executed by the at least one processor, causing the volume estimation system to: reconstruct an initial impedance image based at least in part on received electrical tomography data of the domain, receive prior information associated with the domain, enhance the initial impedance image based at least in part on the received prior information to generate an enhanced impedance image, and generate a volume image of a region of interest of the enhanced impedance image based at least in part on the enhanced initial impedance image, wherein the volume image represents a plurality of values indicative of fluid volume.
[0010] Some embodiments include a system for estimating a fluid volume within a domain. The system may include an electrical tomography system and a volume estimation system. The fluid volume estimation system may include at least one processor and at least one non-transitory computer-readable storage medium having instructions stored thereon, the instructions, when executed by the at least one processor, causing the volume estimation system to: receive prior information associated with the domain, receive an initial impedance image of a portion of the domain from the electrical tomography system, enhance the initial impedance image based at least in part on the received prior information to generate an enhanced impedance image, and generate a volume image of a region of interest of the enhanced impedance image based at least in part on the enhanced initial impedance image, wherein the volume image represents a plurality of values indicative of the fluid volume. BRIEF DESCRIPTION OF THE DRAWINGS
[0011] Figure 1 is a schematic diagram of a portion of an electrical property tomography system according to one or more embodiments of the present disclosure, the schematic diagram showing a plurality of electrodes positioned around a region of interest of a patient;
[0012] Figure 2 is a schematic diagram showing a cross-section of a patient's chest along the plane of an electrode according to one or more embodiments of the present disclosure;
[0013] Figure 3 is a schematic block diagram of an electrical properties tomography system according to one embodiment of the present disclosure;
[0014] Figure 4 A schematic diagram illustrating an environment in which a volume estimation system according to one or more embodiments of the present disclosure may operate;
[0015] Figure 5A is a flow chart of a method of reconstructing an EIT image according to one or more embodiments of the present disclosure;
[0016] Figure 5B yes Figure 5A A schematic block diagram of the method;
[0017] Figure 6 is a flow chart of a method of enhancing an initial impedance image according to one or more embodiments of the present disclosure;
[0018] Fig. 7A is a flow chart of a method of enhancing an initial impedance image according to one or more additional embodiments of the present disclosure;
[0019] Figure 7B yes Fig. 7A A schematic block diagram of the method;
[0020] Figure 8is a schematic block diagram depicting a portion of a Schur complement method for enhancing an impedance image using prior information;
[0021] Fig.9A is a flow chart of a method of determining an estimate of a fluid volume within a region of a body;
[0022] Fig. 9B It is shown Fig.9A A schematic block diagram of the method;
[0023] Fig.10 shows an exemplary segmented and processed image according to one or more embodiments of the present disclosure;
[0024] FIG11 illustrates generated prior information for reconstructing an impedance image using a Gauss-Newton method according to one or more embodiments of the present disclosure;
[0025] FIG. 12A to FIG. 12C Various views of a finite element mesh for estimating electrical property distribution according to one or more embodiments of the present disclosure are shown;
[0026] FIG. 13A to FIG. 13C Various views showing a refined finite element mesh according to one or more embodiments of the present disclosure;
[0027] Fig.14 shows exemplary voltage measurements detected during an EIT process;
[0028] Fig.15 shows an exemplary computed tomography scan obtained from a domain during one or more processes described herein;
[0029] Fig.16 shows a reconstructed initial impedance image reconstructed according to one or more embodiments of the present disclosure;
[0030] FIG. 17A to FIG. 17E showing an additional reconstructed initial impedance image reconstructed according to one or more embodiments of the present disclosure;
[0031] 18A to 18F showing an additional reconstructed initial impedance image reconstructed according to one or more embodiments of the present disclosure;
[0032] FIG. 19A to FIG. 19C showing a depiction of lung regions identified by applying a threshold on the average conductivity distribution of the prior information and removing portions of the reconstructed initial impedance image that do not satisfy the threshold; and
[0033] Fig. 20 A block diagram of an exemplary volume estimation system that may be configured to perform one or more of the processes described above is shown. DETAILED DESCRIPTION
[0034] The illustrations presented herein are not actual views of any EIT system or volume estimation system, but are merely idealized representations for describing exemplary embodiments of the present disclosure. In the following detailed description, reference is made to the accompanying drawings that form a part thereof, in which specific embodiments in which the present disclosure may be practiced are shown by way of illustration. These embodiments are described in sufficient detail to enable one of ordinary skill in the art to practice the present disclosure. However, it should be understood that the detailed description and specific examples, while indicating examples of embodiments of the present disclosure, are given only in an illustrative and not restrictive manner. In accordance with the present disclosure, various substitutions, modifications, additions, rearrangements, or combinations thereof may be made within the scope of the present disclosure, and this will become apparent to one of ordinary skill in the art.
[0035] As is customary, the various features shown in the drawings may not be drawn to scale. Therefore, the dimensions of the various features may be arbitrarily enlarged or reduced for clarity. In addition, some of the drawings may be simplified for clarity. Therefore, the drawings may not depict all components of a given device or all operations of a particular method.
[0036] The information and signals described herein may be represented using any of a variety of different techniques and skills. For example, data, instructions, commands, information, signals, bits, symbols, and chips that may be referred to throughout the specification may be represented by voltage, current, electromagnetic waves, magnetic fields or magnetic particles, optical fields or optical particles, or any combination thereof. For the purpose of clear presentation and description, some drawings may show multiple signals as a single signal. It will be appreciated by those of ordinary skill in the art that a signal may represent a bus of signals, wherein the bus may have a variety of bit widths, and the present disclosure may be implemented on any number of data signals including a single data signal.
[0037] The various exemplary logic blocks, modules, and circuits described in conjunction with the embodiments disclosed herein may be implemented or executed using a general-purpose processor, a special-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic, a discrete hardware component, or any combination thereof designed to perform the functions described herein. A general-purpose processor may be a microprocessor, but in an alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. When a general-purpose processor executes instructions (e.g., software code) stored on a computer-readable medium, the general-purpose processor may be considered a special-purpose processor. The processor may also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors combined with a DSP core, or any other such configuration.
[0038] In addition, it should be noted that the embodiments can be described according to the process that can be depicted as a flow chart, a flow diagram, a structure diagram or a block diagram. Although the flow chart can describe the operation action as a sequential process, many of these actions can be performed in another order, in parallel or substantially simultaneously. In addition, the order of these actions can be rearranged. The process can correspond to a method, a function, a program, a subroutine, a subprogram, etc. In addition, the method disclosed herein can be implemented in hardware, software or both. If implemented in software, these functions can be stored or transmitted on a computer-readable medium as one or more instructions or codes. Computer-readable media includes both computer storage media and communication media, including any medium that facilitates the transmission of a computer program from one place to another.
[0039] As used herein, the singular forms following “a,” “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise.
[0040] As used herein, the term "may" with respect to a material, structure, feature, or method action indicates that this is intended for use in implementing an embodiment of the present disclosure, and such term is used in preference to the more restrictive term "is" to avoid any implication that other compatible materials, structures, features, and methods that may be used in combination therewith should or must be excluded.
[0041] It should be understood that any reference to elements herein using names such as "first", "second", etc. does not limit the quantity or order of those elements, unless such restrictions are explicitly stated. On the contrary, these names can be used as a convenient method to distinguish two or more elements or two or more instances of elements in this article. Therefore, the reference to the first element and the second element does not mean that only two elements can be adopted here, or that the first element must precede the second element in some way. In addition, unless otherwise stated, a group of elements may include one or more elements.
[0042] As used herein, the term "substantially" with respect to a given parameter, characteristic, or condition means and includes to the extent that one skilled in the art would understand that the given parameter, characteristic, or condition is met with a small degree of variance (such as within acceptable manufacturing tolerances). For example, a substantially met parameter may be at least about 90% met, at least about 95% met, or even at least about 99% met.
[0043] As used herein, the term "about" used with reference to a given parameter is inclusive of the stated value and has the meaning dictated by the context (e.g., it includes the degree of error associated with the measurement of the given parameter, as well as variations due to manufacturing tolerances, etc.).
[0044] Embodiments of the present disclosure include electrical tomography (e.g., electrical impedance tomography (EIT)) devices and / or systems for generating images of regions of a patient's body, and these images can be used to determine absolute volume estimates of biological fluids (e.g., air, blood, water, tissue). For clarity and ease of explanation, reference will be made to EIT systems throughout this disclosure; however, any electrical tomography device may be used in place of or in addition to an EIT system and is within the scope of this disclosure. For example, an electrical tomography device may include a device that measures one or more conductivity, resistivity, dielectric constant, electrical admittance, or any other electrical property. Specifically, EIT is an imaging technique that involves positioning electrodes via an electrode strip placed around a region of a patient's body (e.g., around a patient's chest for lung imaging), injecting an electrical excitation signal through a pair of electrodes, and measuring an induced response signal detected by other electrodes of the electrode strip. The EIT system can generate an image based on voltage measurements indicating an estimated impedance value throughout at least a portion of the region of the patient's body. Compared to other imaging techniques, EIT is non-invasive and does not present exposure risks (e.g., radiation exposure risks) that may limit the number and frequency of monitoring actions (e.g., as with techniques such as X-rays). Therefore, EIT is suitable for continuous monitoring of a patient's condition, and in particular for monitoring a patient's lungs, as the measurements can be used to determine the patient's respiratory and hemodynamic parameters and monitor real-time two / three-dimensional images.
[0045] Embodiments of the present disclosure relate to electrical tomography systems and methods of operating the same that can be used to generate absolute impedance images and / or volume measurements (e.g., fluid volume estimates) from EIT data. For example, embodiments of the present disclosure may include systems capable of performing correction and / or enhancement on absolute impedance images. In some embodiments, correcting and / or enhancing absolute impedance images may include utilizing a priori (e.g., prior information), i.e., a data set including a statistical distribution that associates impedance and EIT data. These systems and methods can be used to provide important diagnostic parameters of respiratory dynamics that are not easily obtained from differential EIT images, for example. Conditions and parameters that can be measured using the systems and methods described herein include, but are not limited to, functional residual capacity (FRC), tidal volume, pneumothorax detection, and cell structure on pleural diffusion.
[0046] Figure 1 1 is a schematic diagram of a portion of an EIT system 100 showing a plurality of electrodes 110 positioned around a region of interest (e.g., a chest) of a patient 105. The electrodes 110 of the EIT system 100 may be physically held in place by an electrode strap 103. The arrangement of the electrodes 110 may be transverse to the craniocaudal axis 104 of the patient. Although the electrodes 110 are positioned in a Figure 1105, but the electrodes 110 may be placed around the entire patient 105, depending on the specific region of interest available for measurement or desired for measurement. In addition, the electrodes 110 may be oriented relative to each other in one or more parallel rows (e.g., planes), one or more zigzag patterns (e.g., one or more lines with alternating sharp turns), or any combination thereof. The electrodes 110 may be operably coupled to a computing system (not shown) that is configured to control the operation of the electrodes 110 and perform reconstruction of the EIT image.
[0047] Figure 2 1 is a schematic diagram showing a cross section of the chest of a patient 105 along the plane of the electrodes. A voltage may be applied to a pair of electrodes 110 (shown by electrodes with + and - symbols) to inject an excitation current into the patient between the pair of electrodes. Thus, the voltages (e.g., V1, V2, V3, ... V n ) may be detected by other electrodes and measured by the EIT system 100. Current injection may be performed according to a circular pattern for a measurement cycle using different electrode pairs to generate the excitation current.
[0048] Figure 3 3 is a schematic block diagram of an EIT system 300 according to one embodiment of the present disclosure. The EIT system 300 may include an electrode belt 310 operably coupled to a data processing system 320. The electrode belt 310 and the data processing system 320 may be coupled together via a wired connection (e.g., a cable) and / or may have a communication module for wirelessly communicating with each other. The data processing system 320 may include a processor 322 operably coupled to an electronic display 324, an input device 326, and a memory device 328. The electronic display 324 may be configured together with the data processing system 320 to form a singular form factor of an EIT device coupled to the electrode belt 310. In some embodiments, the electronic display 324 and the data processing system 320 may be separate units of the EIT device coupled to the electrode belt 310. In yet other embodiments, the EIT system 300 may be integrated within another host system that is configured to perform additional medical measurements and / or procedures, wherein the electrode belt 310 may be coupled to a port of the host system that already has its own input devices, memory devices, and electronic displays. Thus, the host system may have the EIT processing software installed therein. Such software may be built into the host system prior to field use or updated after installation.
[0049] The processor 322 may coordinate communications between various devices and execute instructions stored in the computer-readable medium of the memory device 328 to direct current excitation, data acquisition, data analysis, and / or image reconstruction. As an example, the memory device 328 may include a finite element mesh library used by the processor 322 to model the patient's body in the region of interest in order to perform image reconstruction. The input device 326 may include devices such as a keyboard, a touch screen interface, a computer mouse, a remote control, a mobile device, or other devices configured to receive information, which the processor 322 may use to receive input from an operator of the EIT system 300. Therefore, for a touch screen interface, the electronic display 324 and the input device 326 that receives user input may be integrated into the same device. The electronic display 324 may be configured to receive the data and output an EIT image reconstructed by the processor for viewing by the operator. The processor 322 may also generate additional data (e.g., numerical data, graphs, trend information, and other information that is considered useful to the operator) using the measured EIT data alone or in combination with other non-EIT data from other equipment coupled thereto. Such additional data may be displayed on the electronic display 324.
[0050] As will be appreciated by one of ordinary skill in the art, the EIT system 300 may include components not shown in the figure but may also be included to facilitate communication with the electrode belt 310 and / or current excitation, such as one or more analog-to-digital converters, signal processing circuits, demodulation circuits, power supplies, etc.
[0051] Figure 4 A schematic diagram of an environment 400 in which a volume estimation system according to one or more embodiments of the present disclosure may operate is shown. As shown, the environment 400 includes an EIT system 402, a volume estimation system 404, a network 406, and one or more additional systems 408. The volume estimation system 404, the EIT system 402, and the additional systems 408 may communicate via the network 406. The network 406 may include one or more networks, such as the Internet, and may use one or more communication platforms or technologies suitable for transmitting data and / or communication signals. Although Figure 4 A particular arrangement of EIT system 402 , volume estimation system 404 , additional systems 408 , and network 406 is shown, but various additional arrangements are possible. For example, volume estimation system 404 may communicate directly with EIT system 402 , thereby bypassing network 406 .
[0052] like Figure 4 As shown, user 410 may interface with volume estimation system 404 to initiate one or more volume estimations and / or any of the methods described herein. User 410 may be an individual (ie, a human user), a company, a group, or any other entity. Figure 4Only one user 410 associated with volume estimation system 404 is shown, but environment 400 may include any number of multiple users that each use a corresponding volume estimation system 404 to interact with environment 400 .
[0053] In some embodiments, volume estimation system 404 may include one or more types of servers, one or more data stores, one or more interfaces (including but not limited to APIs), one or more web services, one or more content sources, one or more networks, or any other suitable components with which, for example, a server can communicate. In this sense, volume estimation system 404 may provide a platform or backbone that other systems (such as additional system 116) may use to initiate fluid volume estimation within a region of interest of a domain (e.g., a patient).
[0054] In one or more embodiments, the volume estimation system 404 can perform reconstruction algorithms, correction algorithms, and / or volume calculation algorithms, as described in detail below. In some embodiments, the volume estimation system 404 can also include an interface system to control electrical signals to and from the electrical leads of the electrode strips of the EIT system 402. In one or more embodiments, the interface system can include, among other things, analog signal generators, analog-to-digital converters, digital-to-analog converters, digital signal processors, filters, and impedance matching circuits to improve signal-to-noise ratios and reduce crosstalk.
[0055] like Figure 4 As shown, in some embodiments, the volume estimation system 404 can include a database 414. As described in more detail below, the volume estimation system 404 can utilize the database 414 to store initial impedance images, enhanced impedance images, and / or fluid volume estimates and values.
[0056] In some embodiments, volume estimation system 404 also includes client application 412 installed thereon. In one or more embodiments, client application 412 can be associated with volume estimation system 404. For example, client application 412 allows user 410, EIT system 402, and / or additional systems 408 to directly or indirectly interface with volume estimation system 404, EIT system 402, and / or additional systems 408. For example, client application 412 can include a web browsing application and / or a specific volume estimation application.
[0057] Additional systems 408 may include additional systems that may interface with and / or provide data to volume estimation system 404. For example, in some embodiments, additional systems 408 may include a computed tomography system (e.g., a CT scanner), an x-ray system, a magnetic resonance imaging system, an echocardiography device, or any other device for generating images and / or data representing an internal portion of a domain (e.g., a patient), a model (e.g., an a priori model), and / or a simulation.
[0058] Volume estimation system 404 may represent various types of computing devices with which a user (e.g., an administrator) may interact. For example, volume estimation system 404 may be a mobile device (e.g., a cell phone, a smart phone, a PDA, a tablet, a laptop, a watch, a wearable device, etc.). However, in some embodiments, volume estimation system 404 may be a non-mobile device (e.g., a desktop computer or a server). Fig. 20 Additional details regarding volume estimation system 404 are discussed.
[0059] Figure 5A is a flow chart of a method 500 of reconstructing an EIT image according to one or more embodiments of the present disclosure. Figure 5B yes Figure 5A A schematic block diagram of a method 500 is provided. Figure 5A and Figure 5B , method 500 may be performed by one or more of volume estimation system 404 , EIT system 402 , additional system 408 , or any combination thereof.
[0060] In some embodiments, method 500 includes receiving data related to voltages (eg, potentials) detected by electrodes of an EIT system (eg, EIT system 402) during an EIT imaging procedure, such as Figure 5A For example, receiving data related to voltage detected by the electrodes of the EIT system 402 may include receiving data related to voltage detected by the electrodes of the EIT system 402 via the above Figures 1 to 3 The method 500 may include data related to the voltage detected after any of the procedures described apply voltage and / or current to a field (e.g., a patient). In some embodiments, the method 500 may include the volume estimation system 404 receiving data related to the detected voltage from the EIT system 402. In one or more embodiments, the volume estimation system 404 may directly detect / measure the voltage by passing over one or more portions of the EIT system 402.
[0061] In some embodiments, method 500 may include reconstructing an initial impedance image using the received data related to the detected voltage, as shown in action 504. In some embodiments, reconstructing the initial impedance image may include reconstructing one or more of an initial absolute impedance image or an initial differential impedance image. For example, one or more of volume estimation system 404, EIT system 402, additional system 408, or any combination thereof may reconstruct the initial impedance image.
[0062] As is known in the art, an absolute impedance image represents electrical characteristics (e.g., represents electrical characteristics via pixels, voxels, etc.) throughout a corresponding two-dimensional or three-dimensional segment of a domain (e.g., a patient) at a given time, and a differential impedance image represents changes in electrical characteristics from a reference state throughout a two-dimensional or three-dimensional segment of a domain (e.g., a patient). For example, each pixel and / or voxel may be associated with and / or may represent an impedance of a corresponding portion of a region of interest. As used herein, an "initial impedance" image may refer to one or more of an absolute impedance image or a differential impedance image.
[0063] Image reconstruction is formulated as an inverse problem (hereinafter referred to as the "reconstruction problem"), which calculates an estimate of the distribution of the internal characteristics (e.g., impedance characteristics) that best represents the measured potential. As discussed in more detail below, reconstructing the initial impedance image (i.e., action 504) may include utilizing an iterative algorithm that attempts to minimize an objective function (e.g., a cost function) and utilizing regularization techniques to reduce error propagation, improve the stability of the iterations (e.g., the iterative algorithm) and / or the conditions of the reconstruction problem. Additionally, in some embodiments, reconstructing the initial absolute impedance image (i.e., action 504) may include utilizing a Bayesian technique, wherein prior information may be used as a regularization of the problem and a maximum a posteriori of a conditional probability density function may be used as an objective function.
[0064] As noted above, in some embodiments, reconstructing the initial impedance image may include reconstructing the initial impedance image via one or more iterative processes. For example, reconstructing the initial impedance image may include reconstructing the initial impedance image via one or more methods such as a Gauss-Newton reconstruction method, a Noser method, a simulated annealing method, a Kalman filter method, a level set method, a Markov chain Monte Carlo method, a black box method, a back projection method, a total variation method, or a nonlinear programming method. In additional embodiments, reconstructing the initial impedance image may include reconstructing the initial impedance image via a Land Weber iteration method.
[0065] In some embodiments, reconstructing the initial impedance image may include reconstructing the initial impedance image via one or more direct methods. In other words, reconstructing the initial impedance image may include reconstructing the initial impedance image without input (e.g., a priori information) from an external data source. For example, reconstructing the initial impedance image may include reconstructing the initial impedance image via one or more methods such as a direct D-bar reconstruction algorithm, a Calderon method, or an algebraic reconstruction technique. In additional embodiments, as described below, reconstructing the initial impedance image may include reconstructing the initial impedance image based at least in part on an external data source (e.g., a priori information).
[0066] In some embodiments, reconstructing the initial impedance image may include reconstructing the initial absolute impedance image using one or more regularization methods. For example, reconstructing the initial impedance image may include reconstructing the initial impedance image using one or more regularization methods, such as generalized Tikhonov regularization (e.g., Gaussian high pass filter), utilizing prior information (e.g., considering probability density functions, as described below), and / or utilizing regularization of a total variation (TV) functional.
[0067] As noted above, in one or more embodiments, reconstructing the initial impedance image may further include reconstructing the initial impedance image based at least in part on prior information (i.e., one or more priors). As used herein, "prior information" and "prior" may refer to data (e.g., one or more images or scans of a population and / or patient and / or one or more images of other imaging modalities (e.g., CT scans)) that are external to the data provided by the EIT data and that can inform the image generation process (e.g., the reconstruction process) (e.g., provide additional information to the image generation process). For example, the prior information may include one or more computed tomography (CT) scans that can be used to add prior information to the EIT reconstruction algorithm. Specifically, the CT scan may provide anatomically based and / or physiologically based priors, and the initial impedance image may be reconstructed at least in part based on the information provided by the anatomically based priors. For example, the CT scan (e.g., prior) may be used to construct a regularization term of the above-mentioned regularization method. In addition, the prior information may be used to construct the regularization term and determine the approximation error. In some embodiments, the prior information may include images from other imaging modalities and / or data from other sensors (eg, pneumotachographs, plethysmographs, etc.).
[0068] In some embodiments, the prior information may be specific to the domain (e.g., the patient) for which the initial impedance image is to be reconstructed. For example, the prior information may include a CT scan (or other data) of the domain (e.g., a CT scan at or near the location where the EIT data is measured (at or near the electrode belt of the EIT system)). In other embodiments, the prior information may include a CT scan (or other data) of one or more anatomical or physiological models (e.g., a physical model). In further embodiments, the prior information may include data from one or more digital models and / or simulations. In some embodiments, the prior information may be selected and / or obtained based at least in part on the category of the domain (e.g., age, gender, race, weight, etc.). For example, in some embodiments, the prior information may be selected and / or obtained based at least in part on anthropometric measurements. In additional embodiments, the prior information may be selected and / or obtained based at least in part on the type of diagnostic procedure to be performed with the ultimately generated data (e.g., fluid estimates) (e.g., a diagnostic procedure for cellular structures on pneumothorax, FRC, tidal volume, and / or pleural fusion). In further embodiments, the prior information may be selected and / or obtained based at least in part on the type of volumetric image to be generated (eg, air, blood, and / or water volumetric images).
[0069] In some embodiments, utilizing prior information to reconstruct the initial impedance image may include using the prior information (e.g., a CT scan) as an anatomical atlas to reconstruct the initial impedance image. In such embodiments, the probability density function π (ρ) of the resistivity distribution in the region of the body is SW ) can be represented as the following Gaussian distribution:
[0070]
[0071] In the above formula, the average value of the probability density function is and the covariance matrix (Γ sw ) can be used to estimate the initial absolute impedance image. As noted above, the prior information is the regularization term (the second term (i.e., the second line of the above formula)) of the following formula (2):
[0072]
[0073] Where F is a Gaussian high-pass filter, Γ sw is the covariance matrix of the statistical prior interpolated to the finite element mesh (FEM), is the expected vector of resistivity of the statistical prior (which is also interpolated to the FEM mesh), and γ is the regularization parameter of the prior information. If the prior information is sufficiently informative in all directions (e.g., x, y, and z directions), then the third term (i.e., the third line of the above formula) may not be necessary, where the first term (i.e., the first line of the above formula) may not be very informative.
[0074] In formula (2), the first term (i.e., the first row) is derived from probability, the second term (i.e., the second row) is the prior information, and the third term (i.e., the third row) is the smoothing prior. Formula (2) is the relationship between the candidate ρ and the statistical expectation As a non-limiting example, using prior information to reconstruct the initial impedance image may include reconstructing the initial impedance image via any of the methods described in T. de Castro Martins et al., "A review of electrical impedance tomography in lung applications: Theory and algorithms for absolute images," Annual Reviews in Control, Vol. 48, 2019, pp. 442-471, the disclosure of which is incorporated herein by reference in its entirety.
[0075] After reconstructing the initial impedance image, method 500 may include generating and / or outputting the reconstructed initial impedance image, as shown in action 506 of Figure 5. For example, outputting the reconstructed initial impedance image may include generating a data packet including / representing the initial impedance image and / or storing the initial impedance image in a database (e.g., database 414). In some embodiments, the reconstructed initial impedance image may be represented by pixels (e.g., a two-dimensional image), by voxels (a three-dimensional image), by elements of a FEM, or by any combination of the foregoing elements.
[0076] Referring to actions 502 to 506 together, in some embodiments, method 500 may include generating multiple initial impedance images. For example, method 500 may include generating multiple initial impedance images along multiple planes (e.g., slices) or segments along the axial length of a domain (e.g., a patient).
[0077] Still see Figure 5A and Figure 5B, the initial impedance image and the electrical properties represented therein may be underestimated in one or more regions (e.g., due to, for example, the linearity of a Gauss-Newton reconstruction method), may have a relatively low spatial resolution, may include image artifacts, etc. Thus, enhancement (e.g., correction) of the initial impedance image may be required to improve the diagnostic quality of the initial impedance image (e.g., the ability to perform a diagnosis from the initial impedance image or a resulting volume estimate determined from the initial impedance image).
[0078] Figure 6 is an enhanced initial impedance image according to one or more embodiments of the present disclosure (e.g., Figure 5A and Figure 5B Flow chart of method 600 of reconstructing an initial impedance image described herein.
[0079] In one or more embodiments, method 600 may include receiving an initial impedance image, such as Figure 6 As shown in action 602 of . In some embodiments, method 600 may include receiving an initial impedance image from an EIT system (e.g., EIT system 402). In one or more embodiments, method 600 may include volume estimation system 404 receiving the initial impedance image from EIT system 402. In additional embodiments, method 600 may include receiving the initial impedance image from an additional system (e.g., additional system 408), querying a database (e.g., database 414) to retrieve the initial impedance image, or receiving the initial impedance image from an external source. As noted above, the initial impedance image may be represented by pixels (e.g., a two-dimensional image), by voxels (a three-dimensional image), by elements of a FEM, or by any combination thereof.
[0080] In response to receiving and / or retrieving the initial impedance image, method 600 may also include directly enhancing (eg, correcting) the initial impedance image, such as Figure 6 As shown in action 604 of . Specifically, method 600 may include enhancing the initial impedance image without additional information input (eg, a priori information).
[0081] In one or more embodiments, directly enhancing (e.g., correcting) the initial impedance image may include enhancing the initial impedance image by a D-bar method. For example, enhancing the initial impedance image may include enhancing the initial impedance image by any of the methods described in Melody Dodd et al., "A Real-time D-bar Algorithm for 2-D Electrical Impedance," Inverse ProblImaging (Springfield), November 1, 2014; Vol. 8, No. 4, pp. 1013-1031, DOI: 10.3934 / ipi.2014.8.1013, the disclosure of which is incorporated herein by reference in its entirety. In addition, enhancing the initial impedance image may include enhancing the initial impedance image by any conventional D-bar method.
[0082] Enhancement of the initial impedance image via the D-bar method (or any other reconstruction method) may estimate true values or relatively close values of electrical characteristics represented by pixels of the initial impedance image of a region of interest of, for example, a domain (e.g., a patient). Thus, enhancement of the initial impedance image via the D-bar method may correct electrical characteristics represented in the initial impedance image of the region of interest of the domain (e.g., a patient). As a non-limiting example, enhancement of the initial impedance image via the D-bar method may include identifying organs (e.g., lungs, heart, kidneys) and bones from the initial impedance image, and then adjusting associated electrical characteristics represented in the region of the initial impedance identified as the organs and / or bones based on the identified organs and / or bones via the D-bar method.
[0083] After enhancing the initial impedance image, method 600 may include outputting an enhanced impedance image, such as Figure 6 As shown in action 606 of . For example, the volume estimation system 404 may output an enhanced impedance image. In some embodiments, outputting the enhanced impedance image may include outputting an enhanced absolute impedance image. In additional embodiments, outputting the enhanced impedance image may include outputting an enhanced differential impedance image. In one or more embodiments, outputting the enhanced impedance image may include generating a data packet including the enhanced impedance image and / or storing the enhanced impedance image in a database (e.g., database 414). In some embodiments, the enhanced impedance image may be represented by pixels (e.g., a two-dimensional image), by voxels (a three-dimensional image), by elements of a FEM, or by any combination thereof.
[0084] Fig. 7A is an enhanced initial impedance image according to one or more additional embodiments of the present disclosure (e.g., Figure 5A and Figure 5B Flowchart of method 700 of reconstructing initial impedance described herein. Figure 7B yes Fig. 7A Schematic block diagram of method 700. Method 700 may include utilizing prior information to enhance an initial impedance image.
[0085] In one or more additional embodiments, enhancing (e.g., correcting) the initial impedance image may include enhancing the initial impedance image by a Shul complement method. Specifically, the Shul complement method may utilize the Shul complement property to introduce statistical prior information into the D-bar method and may provide significant improvements in the spatial resolution of the initial impedance image. In other embodiments, the Shul complement may be used as a direct enhancement method, as described above for Figure 6 described. Figure 8 is a schematic block diagram depicting a portion of a Schur's complement method that utilizes a priori information (eg, information from a priori model).
[0086] The prior information and / or the prior model may include a segmented tomographic image of a portion of a body (e.g., a chest) of a domain (e.g., a patient), and the electrical property distribution of the tissue may be used to perform a numerical phantom (described in more detail below). For example, a CT scan may be segmented into different tissues (e.g., bone, ventilated lung, atelectasis, heart, and / or muscle) based on the properties of the corresponding tissue in terms of grayscale. Thus, the electrical property distribution of the tissue (which may have been measured in vivo) may be used to generate (e.g., construct, adjust, and / or correct) the correct electrical property distribution within the initial impedance image.
[0087] See together FIG. 7A to FIG. 8 In some embodiments, the method may include introducing prior information based on the Schur complement characteristic. For example, the method may include applying a correction (e.g., post-processing) to the initial impedance image by maximizing the conditional probability density function of an image consistent with the prior information given the initial impedance image.
[0088] Based on the Schur complement characteristic, the conditional mean μ can be determined by the following formula (3): (x|y) :
[0089] μ x|y =Ay+b (3)
[0090] According to one or more (e.g., a set of) prior samples An estimate of the probability distribution π(x) of the population may be determined. The one or more prior samples σ may be determined by segmenting the tomographic image / scan 3D , and electrical characteristics can be established for each region identified in the resulting segmentation.
[0091] For each obtained electrical characteristic distribution For example, a priori samples of the segmentation, two images can be determined (eg, calculated). The first determined image Can be expressed The electrical characteristics of the domain are oriented on the chest. Gaussian interpolation can be applied to σ 3D And can be considered The second image is determined by Can be expressed To obtain the second image First, a distribution representing an electrical characteristic may be determined (eg, calculated) The vector of the positive problem And secondly, the determined vector (ie, the determined voltage) estimates the second image by a reconstruction method
[0092] The two resulting images and can be converted into vectors and can be represented by x i and i In addition, by connecting x i and i To generate a new vector z i :
[0093]
[0094] In addition, by determining multiple prior samples x i and i The estimated value of , which can determine the expected value μ of the z process z and covariance Γ z According to the expected value μ of the z process z and covariance Γ z , we can get a set of images x i and i Statistics and relationships between them.
[0095]
[0096]
[0097] where μ x and μ y is x i and i The expected value of the image group, Γ xx and Γ yy is the covariance matrix of each set of images, and Γ xy and Γyx is their cross-covariance matrix.
[0098] By using the image group x i and i The prior statistics and the relationship between y i Determine the vector x i The estimated value of .
[0099] μ x|y =μ x +Γ xy Γ yy -1 (y-μ y ) (6)
[0100] By reordering equation (6), the initial impedance image from the reconstruction method can be determined as This correction takes into account the prior distribution of the population and the positive problem model.
[0101] μ x|y =Γ xy Γ yy -1 y+(μ x -Γ xy Γ yy -1 μ y ) (7)
[0102] μ x|y =Ay+b (8)
[0103] In summary, by using the Schur's complement method described herein, a regional mean and conditional covariance can be determined, and the regional mean and conditional covariance can be used to determine (e.g., calculate) a corrected impedance of an initial impedance image. Thus, the above-described Schur's complement method can be used to enhance the initial impedance image. In addition, the initial impedance image can be enhanced using any combination of an alternative reconstruction method enhancement (e.g., a D-bar method) and the above-described Schur's complement method.
[0104] In additional embodiments, as noted above, the initial impedance image may be enhanced via numerical phantom simulation. Specifically, the CT scans may be segmented and converted into electrical property distributions according to the electrical property distributions of each segmented tissue. Subsequently, numerical phantoms may be generated for these electrical property distributions, and reference reconstructed images may be determined based at least in part on these electrical property distributions.
[0105] In addition, based on the comparison between the distribution of electrical properties represented in the initial impedance image and the reference reconstructed image, a relationship function (e.g., a linear function or a nonlinear function) may be determined and the relationship function may be used to adjust the electrical properties within the region of interest of the initial impedance image. In some cases, various machine learning models may be used in the process of enhancing the initial impedance image. For example, enhancing the initial impedance image may include machine learning and / or deep learning techniques, which include providing a training library to a matching learning algorithm or a neural network to train the machine to assist or perform enhancing the initial impedance image or a portion of the initial impedance image. In some embodiments, the volume estimation system 404 may enhance at least a portion of the initial impedance image using one or more of a regression model (e.g., a set of statistical processes for estimating relationships between variables), a classification model, and / or a phenomenon model. In addition, the machine learning model may include quadratic regression analysis, logistic regression analysis, support vector machine, Gaussian process regression, an integrated model, or any other regression analysis. In addition, in yet other embodiments, the machine learning model may include decision tree learning, regression trees, boosted trees, gradient boosted trees, multilayer perceptrons, one-to-the-remainder, naive Bayes, k-nearest neighbors, association rule learning, neural networks, deep learning, pattern recognition, or any other type of machine learning.
[0106] In some embodiments, the enhanced impedance image may include the above Figures 6 to 8 Any combination of the methods described. For example, one or more portions (e.g., pixels and / or regions) of an initial impedance image may be enhanced via a first method, and one or more other portions (e.g., pixels and / or regions) of the initial impedance image may be enhanced via a second, different method.
[0107] After enhancing the initial impedance image, method 700 may include outputting an enhanced impedance image, such as Fig. 7A As shown in action 706 of . For example, the volume estimation system 404 may output an enhanced impedance image. In some embodiments, outputting the enhanced impedance image may include outputting an enhanced absolute impedance image. In additional embodiments, outputting the enhanced impedance image may include outputting an enhanced differential impedance image. In one or more embodiments, outputting the enhanced impedance image may include generating a data packet including / representing the enhanced impedance image and / or storing the enhanced impedance image in a database (e.g., database 414). In some embodiments, the enhanced impedance image may be represented by pixels (e.g., a two-dimensional image), by voxels (a three-dimensional image), by elements of a FEM, or by any combination thereof.
[0108] Fig.9A is a flow chart of a method 900 of determining a fluid volume estimate within a region of a body. Fig. 9B It is shown Fig.9A A schematic block diagram of method 900 is provided.
[0109] See together Fig.9A and Fig. 9B , method 900 may include receiving an enhanced impedance image, such as Fig.9A As shown in action 902. For example, method 900 may include receiving the above Figures 6 to 8 Any enhanced (e.g., corrected) impedance image discussed herein. Specifically, the volume estimation system 404 may receive an enhanced impedance image. As noted above, in some embodiments, the enhanced impedance image may include an enhanced absolute impedance image or an enhanced differential impedance image. In some embodiments, receiving the enhanced impedance image may include receiving a data packet representing the enhanced impedance image and / or querying the database 414 to retrieve the enhanced impedance image.
[0110] In response to receiving the enhanced impedance image, method 900 may include segmenting the enhanced impedance image, such as Fig.9A As shown in action 904 of . For example, method 900 may include applying one or more segmentation processes to the enhanced impedance image. For example, volume estimation system 404 may partition the enhanced impedance image into multiple segments (eg, multiple groups of pixels and / or image objects).
[0111] Segmenting the enhanced impedance image may include locating objects (e.g., lungs, heart, kidneys, bones, and / or other organs) and boundaries (e.g., lines, curves, etc.) within the image. For example, the results of the image segmentation may include a set of segments that together cover the entire enhanced image and / or a set of contours extracted from the image (e.g., edge detection). Segmenting the enhanced impedance image may include applying any conventional image segmentation process to the enhanced impedance image.
[0112] In some embodiments, the image segmentation process may be informed by prior information. For example, in some embodiments, segmenting the enhanced impedance image may include identifying objects (e.g., organs and / or bones) and / or regions of interest within the enhanced impedance image using one or more priors (e.g., CT scans) that have also been segmented. As another non-limiting example, segmenting the enhanced impedance image may include selecting a region of interest as a region of the segmented object using an average contour of the segmented object (e.g., an organ) within the one or more priors (e.g., CT scans).
[0113] After segmenting the enhanced impedance image, method 900 may include selecting a region of interest within the enhanced impedance image, such as Fig.9A As shown in action 906. For example, method 900 may include the above Fig.9AThe enhanced impedance image may be selected in any manner described in action 904 (e.g., using prior information). In some embodiments, selecting the region of interest in the enhanced impedance image may include selecting a region representing one or more organs or one or more bones or any other part of the patient's body. For example, selecting the region of interest in the enhanced impedance image may include selecting a region representing a lung region of the patient. In one or more embodiments, selecting the region of interest in the enhanced impedance image may include selecting a region representing a portion of the body to be analyzed (e.g., a portion of the body where it is desired to measure a fluid volume).
[0114] The selected region of interest of the enhanced impedance image may provide (eg, suggest, represent, define) a distribution of electrical properties through the selected region via values represented by pixels, voxels, and / or FEM elements of the enhanced impedance image.
[0115] In response to selecting a region of interest of the enhanced impedance image, method 900 may include determining a relationship parameter (V ) that relates an electrical characteristic (e.g., impedance) represented in the selected region of interest to one or more other characteristics of the patient in the region of interest. EIT ),like Fig.9A As shown in action 908. For example, the relationship parameter (V EIT ) may include a linear or nonlinear function that relates the fluid content and electrical properties of the region of interest. In some embodiments, the one or more other properties may include the air content in the region (e.g., the total volume of air and / or the volume of air indicated by one or more pixels, voxels, and / or FEM elements), the blood content in the region, the tissue content in the region.
[0116] In some embodiments, the relationship parameter (V EIT For example, the relationship parameter (V EIT For example, the relationship parameter (V EIT ):
[0117] V EIT =Σ(ρ i v i ) (17)
[0118] where ρ i is the resistivity distribution represented in the enhanced impedance image of the region of interest, and v i is the tetrahedral volume of the finite element mesh representation of the region of interest. Alternatively, v i is the pixel area and / or voxel volume represented in the enhanced impedance image. In view of the above, the relationship parameter (V EIT) may be specific to the selected region of interest. In some embodiments, a relationship parameter may be determined for each pixel, voxel, and / or FEM element of the enhanced impedance image.
[0119] In determining the relationship parameter (V EIT ), method 900 may include estimating the volume of fluid within the region of interest (V 流体 (For example, V 空气 )),like Fig.9A As shown in action 910 of . For example, volume estimation system 404 can estimate the volume of the fluid in the region of interest. An exemplary method of estimating the volume of the fluid will be described below.
[0120] The change in resistivity distribution represented in the region of interest in the enhanced impedance image can be substantially linearly correlated with the change in fluid content in the region of interest (e.g., the selected region of interest) of the domain. EIT ), estimate the fluid volume (V 流体 ) may include estimating the fluid volume (V) via a linear model 流体 For example, to estimate the fluid volume (V 流体 ) may include estimating the fluid volume (V) via the following model 流体 ):
[0121] y 流体 =a·V EIT +b (18)
[0122] This can be done by using the air content estimated from a priori information (e.g., CT scans) as described above for Fig.9A 908) and relationship parameters (V EIT ) to calculate (e.g., estimate) variables a and b. FIG. 19A to FIG. 19C and Tables 3 to 6 describe the estimated variables a and b in more detail.
[0123] In some embodiments, estimating the fluid volume (V 流体 ) can be obtained from prior information, such as Fig. 9B For example, in some embodiments, the estimated fluid volume (V 流体) includes determining the fluid content by correlating the Hounsfield unit scale of the tissue of the region of interest (e.g., tissue of the lung) and the fluid volume using one or more priors (e.g., CT scans). As is known in the art, the Hounsfield unit scale is a linear transformation of the initial linear attenuation coefficient measurement to a measurement where the radiodensity of water (e.g., distilled water) at standard pressure and temperature (STP) is defined as zero Hounsfield units (HU), and the radiodensity of air at STP is defined as -1000 HU. In a voxel with an average linear attenuation coefficient μ, the corresponding HU value is thus given by:
[0124]
[0125] where μ 水 and μ 空气 are the linear attenuation coefficients of water and air, respectively. In view of the above, a change of one Hounsfield Unit (HU) represents a 0.1% change in the attenuation coefficient of water, since the attenuation coefficient of air is approximately zero.
[0126] In view of the above, the fluid content (e.g., air content) indicated by a priori information (e.g., CT scan) can be correlated with the relationship parameter (V EIT ). In addition, based on the fluid content indicated by the prior information and the relationship parameter (V EIT ) between them, the relationship parameter (V EIT ), and the fluid volume (V 流体 ) is a final estimate of . In addition, in some embodiments, (V EIT ) and (V 流体 ) can be associated, for example, by linear or nonlinear regression, machine learning, etc. In addition, in some embodiments, the prior can be related to (V EIT ) estimates by minimizing the mean square error between the estimates. In addition, the parameters a and b can be estimated by calculating (V 流体 ) is the maximum a posteriori of the conditional mean, thus estimating (V 流体 ).
[0127] As mentioned above, in some embodiments, the estimated fluid volume (V 流体 ) may include estimating the volume of one or more of air, blood, water, and / or tissue. In response to estimating the fluid volume (V 流体 ), method 900 may include outputting an estimated fluid volume (V 流体 ) (e.g., value, image, etc.), such as Fig.9A As shown in action 912.
[0128] For example, the volume estimation system 404 may output a representation of the estimated fluid volume. In some embodiments, the outputting of the estimated fluid volume representation may include outputting a total value representing the estimated volume of a given fluid. In additional embodiments, the outputting of the estimated fluid volume representation may include outputting a volume image, wherein each pixel, voxel, and / or FEM element represents the volume of a given fluid in an area associated with the pixel, voxel, and / or element. In some embodiments, the outputting of the estimated fluid volume representation may include outputting a representation of an absolute volume estimate. In additional embodiments, the outputting of the estimated fluid volume representation may include outputting a representation of a differential volume estimate. In one or more embodiments, the outputting of the estimated fluid volume representation may include generating a data packet including a representation of the estimated fluid volume and / or storing the estimated fluid volume representation in a database (e.g., database 414).
[0129] The following references Figures 10 to 19C The description of describes a plurality of tests performed by the inventors for estimating the fluid volume in the region of a plurality of domains according to the methods and processes described herein. These experiments include obtaining voltage measurements using an EIT system. The electrodes used to apply and detect voltages are oriented in a zigzag pattern around the domains. In these experiments, the domains included twelve piglets. The applied current patterns included sine, 10 mA, 125 kHz, and adjacent current patterns (jump-3).
[0130] These piglets were used to construct the anatomical atlas (mean and covariance) (eg, to construct the prior information). The flow estimation method described herein was performed on the two remaining piglets that were not included in the prior information.
[0131] Four CT scans were performed on each of the ten piglets used to construct the anatomical atlas. Fig. 7A and Figure 7B Describe the methods for segmenting and processing each CT scan. Fig.10 An exemplary segmented and processed image is shown.
[0132] In addition, small perturbations simulating abnormal lungs are introduced into the prior model (e.g., an artificial model), a CT scan is performed on the prior model, and additional segmented images are artificially generated. Small perturbations are introduced by removing spherical objects from random locations in the modeled lungs and / or the modeled heart. Table 1 shows the electrical properties based on the measured Fig.10 The determined mean and standard deviation of the segmented tissue.
[0133]
[0134] Figure 11 shows the generated prior information for reconstructing the impedance images of the ten piglets using the Gauss-Newton method.The inverse problem was regularized using the anatomical atlas generated from the ten piglets and a high-pass Gaussian filter, and the conductivity distribution was estimated (eg, reconstructed).
[0135] By using FIG. 12A to FIG. 12C The finite element mesh depicted in and the initial impedance image (conductivity distribution) is estimated by regularizing the inverse problem with prior information. FIG. 13A to FIG. 13C The approximate error theory is applied to the refined finite element mesh depicted in .
[0136] Additionally, voltage measurements were obtained for the two remaining piglets. Fig.14 It is shown that in the EIT process (for example, the above Figures 1 to 3 An exemplary average of voltage measurements (e.g., an average of all electrode measurements) for the first of the two piglets obtained during a period of 1-2 hours (described above) and a window during which at least one initial impedance image is reconstructed via any of the above-described means. The same process is applied to the second of the two piglets. In addition, a CT scan is obtained from each of the two piglets. Fig.15 and Fig.16 An exemplary CT scan at the center of the electrode band of one of the first piglets and a reconstructed initial impedance image of the first piglet are shown, respectively. FIG. 17A to FIG. 17E Additional reconstructed impedance images of the first piglet at different applied positive end-expiratory pressure (PEEP) levels are shown. 18A to 18F Reconstructed impedance images of the second piglet at various applied PEEP levels are shown.
[0137] See also Figures 15 to 17E , for the first piglet, the heart was displaced compared to the anatomical atlas information; therefore, the reconstructed initial impedance image followed the true position rather than the a priori information position.
[0138] Table 2 shows the functional residual capacity (FRC) of the first and second piglets determined from CT scans of the first and second piglets.
[0139]
[0140] Lung regions within the reconstructed initial impedance images of the first and second piglets were identified by applying a threshold on the average conductivity distribution of the prior information and removing portions of the reconstructed initial impedance images that did not meet the threshold, as FIG. 19A to FIG. 19C An additional threshold is applied to remove the cardiac region from the reconstructed initial impedance image.
[0141] In addition, the weighted sum model of the lung region is determined via the following formula:
[0142] V EIT =∑(ρ i v i ) (20)
[0143] As mentioned above Fig.9A and Fig. 9B As discussed, ρ i is the resistivity distribution represented in the enhanced impedance image of the region of interest, and v i is the tetrahedral volume represented by the finite element mesh of the region of interest.
[0144] Continue with the above Fig.9A and Fig. 9B The method described above applies a first-order correction to estimate the fluid volume (V 流体 ):
[0145] V 流体 =a·V EIT +b (21)
[0146] The air content (V) can be estimated by correlating the air content (V) estimated from a priori information (e.g., a CT scan of another piglet (e.g., a second piglet) 流体 ) to calculate (e.g., estimate) variables a and b. Specifically, V for the first piglet can be determined EIT(第一只) and a linear correction is applied to it (e.g., variables a and b) so as to correspond to the V determined from the CT scan of the first piglet. 流体(第一只) Then, by using V EIT(第二只) and the linear correction determined by the first piglet (e.g., variables a and b) to determine V 流体(第二只) Additionally, the above process was repeated using a second piglet to determine variables a and b and the associated linear corrections.
[0147] Table 3 shows the air volume estimate for the first piglet when variables a and b were determined using the first piglet, where a = 2.3134e+05 and b = -1.537e+02.
[0148]
[0149] Table 4 shows the air volume estimates for the second piglet when variables a and b were determined using the first piglet.
[0150]
[0151] Table 5 shows the air volume estimate for the second piglet when variables a and b were determined using the second piglet, where a = 2.0492e+05 and b = -9.9857e+01.
[0152]
[0153] Table 6 shows the air volume estimates for the second piglet when variables a and b were determined using the first piglet.
[0154]
[0155] See together Figures 1 to 19C , determining that the methods of volume estimation described herein may be advantageous. Specifically, by enhancing impedance images according to the methods described herein, more accurate absolute impedance images may be generated compared to conventional absolute impedance images. For example, due to the anatomical atlas and prior information and enhancement methods described herein, the enhanced absolute impedance images described herein may be more accurate than conventional absolute impedance images. In addition, the methods described herein using prior information improve organ localization during D-bar, iterative (e.g., Monte Carlo), and Kalman filtering methods.
[0156] Determining the absolute impedance image and fluid volume estimation value in the method described in this article can realize air volume estimation, cellular structure and pneumothorax volume determination on pleural effusion, cardiac stroke volume estimation, whole chest air volume determination, FRC, etc.
[0157] Fig. 20 2 shows a block diagram of an exemplary volume estimation system 2200 that can be configured to perform one or more of the processes described above. It should be understood that one or more computing devices can form the volume estimation system 2200. Fig. 20 As shown, volume estimation system 2200 may include processor 2202, memory 2204, storage device 2206, I / O interface 2208, and communication interface 2210, which may be communicatively coupled via a communication infrastructure. Fig. 20 An exemplary volume estimation system 2200 is shown in FIG. Fig. 20 The components shown are not intended to be limiting. Additional or alternative components may be used in other embodiments. Fig. 20 Those components shown are fewer than those shown. Fig. 20 Components of volume estimation system 2200 are shown.
[0158] In one or more embodiments, the processor 2202 includes hardware for executing instructions, such as those that make up a computer program. As an example and not by way of limitation, to execute instructions, the processor 2202 may retrieve (or fetch) instructions from an internal register, an internal cache, memory 2204, or storage device 2206 and decode and execute these instructions. In one or more embodiments, the processor 2202 may include one or more internal caches of data, instructions, or addresses. As an example and not by way of limitation, the processor 2202 may include one or more instruction caches, one or more data caches, and one or more translation lookaside buffers (TLBs). The instructions in the instruction cache may be copies of instructions in the memory 2204 or storage device 2206.
[0159] The volume estimation system 2200 includes a memory 2204 coupled to the processor 822. The memory 2204 may be used to store data, metadata, and programs for execution by the processor. The memory 2204 may include one or more of a volatile memory and a non-volatile memory, such as a random access memory ("RAM"), a read-only memory ("ROM"), a solid state disk ("SSD"), flash memory, a phase change memory ("PCM"), or other types of data storage devices. The memory 2204 may be an internal memory or a distributed memory.
[0160] The volume estimation system 2200 includes a storage device 2206, which includes a storage device for storing data or instructions. As an example and not by way of limitation, the storage device 2206 may include the non-transitory storage medium described above. The storage device 2206 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a tape, or a universal serial bus (USB) drive, or a combination of two or more of these. Where appropriate, the storage device 2206 may include a removable or non-removable (or fixed) medium. The storage device 2206 may be inside or outside the volume estimation system 2200. In one or more embodiments, the storage device 2206 is a non-volatile solid-state memory. In other embodiments, the storage device 2206 includes a read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or a flash memory, or a combination of two or more of these.
[0161] The volume estimation system 2200 also includes one or more input or output ("I / O") devices / interfaces 2208 (e.g., a touch display) that are provided to allow a user to provide input to the volume estimation system 2200, receive output from the volume estimation system, and otherwise transmit data to and receive data from the volume estimation system. The I / O devices / interfaces 2208 may include a mouse, a keypad or keyboard, a touch screen, a camera, an optical scanner, a network interface, a modem, other known I / O devices, or a combination of such I / O devices / interfaces. The touch screen may be activated with a stylus or a finger.
[0162] I / O devices / interfaces 2208 may include one or more devices for presenting output to a user, including but not limited to a graphics engine, a display (e.g., a display screen), one or more output drivers (e.g., a display driver), one or more audio speakers, and one or more audio drivers. In certain embodiments, I / O interface 2208 is configured to provide graphics data to the display for presentation to the user. Graphics data may represent one or more graphical user interfaces and / or any other graphical content as may serve a particular implementation.
[0163] The volume estimation system 2200 may also include a communication interface 2210. The communication interface 2210 may include hardware, software, or both. The communication interface 2210 may provide one or more interfaces for communication (such as packet-based communication) between the volume estimation system 2200 and one or more other computing devices or networks. As an example and not by way of limitation, the communication interface 2210 may include a network interface controller (NIC) or network adapter for communicating with an Ethernet or other wire-based network, or a wireless NIC (WNIC) or wireless adapter for communicating with a wireless network (such as WI-FI and / or Bluetooth). The volume estimation system 2200 may also include a bus 2212. The bus 2212 may include hardware, software, or both that couples the components of the volume estimation system 2200 to each other.
[0164] Embodiments of the present disclosure also include the following embodiments:
[0165] Embodiment 1. A method for estimating fluid volume, the method comprising: receiving electrical tomography data of a portion of a domain; reconstructing an initial impedance image based at least in part on the electrical tomography data; enhancing the initial impedance image to generate an enhanced impedance image; segmenting the enhanced impedance image to identify one or more tissues depicted within the enhanced impedance image; selecting a region of interest within the enhanced impedance image; determining a relationship parameter that correlates electrical characteristics represented within the region of interest of the enhanced impedance image with one or more characteristics of the region of interest; and estimating the fluid volume within the region of interest based at least in part on the relationship parameter and the enhanced impedance image.
[0166] Embodiment 2. According to the method described in Embodiment 1, the method also includes removing one or more portions outside the region of interest of the enhanced impedance image from the enhanced impedance image.
[0167] Embodiment 3. A method according to embodiment 2, wherein removing the one or more portions includes removing portions of the enhanced impedance image that do not meet a threshold electrical characteristic value.
[0168] Embodiment 4. A method according to any one of embodiments 1 to 3, wherein the one or more characteristics include one or more of air content, blood content, or tissue content.
[0169] Embodiment 5. According to the method described in any one of embodiments 1 to 4, the method also includes: receiving prior information; determining the fluid content indicated in the prior information; associating the fluid content indicated in the prior information with a relationship parameter; and estimating the fluid volume in the area of interest based at least in part on the received prior information.
[0170] Embodiment 6. A method according to embodiment 5, wherein the received prior information includes one or more CT scans of the domain.
[0171] Embodiment 7. A method according to embodiment 5, wherein the prior information includes image data of one or more anatomical and physiological models.
[0172] Embodiment 8. A method according to embodiment 5, further comprising selecting the prior information to be received based on one or more of the anthropometric metrics of the domain.
[0173] Embodiment 9. The method according to embodiment 5 further includes segmenting the enhanced impedance image to identify objects depicted in the prior information.
[0174] Embodiment 10. A method according to any one of embodiments 1 to 9, wherein estimating the volume of the fluid within the region of interest includes estimating the volume of air within at least a portion of the lungs in the region.
[0175] Embodiment 11. A method according to embodiment 10, wherein estimating the air volume includes estimating the absolute air volume.
[0176] Embodiment 12. A method according to embodiment 11, wherein reconstructing the initial impedance image includes reconstructing the initial absolute impedance image.
[0177] Embodiment 13. A method according to embodiment 11, wherein reconstructing the initial impedance image includes reconstructing an initial differential impedance image.
[0178] Embodiment 14. A method according to any one of embodiments 1 to 13, the method also includes outputting a representation of the estimated fluid.
[0179] Embodiment 15. A method according to claim 14, wherein the representation includes a volumetric image having multiple elements of a plurality of pixels, voxels, or a finite element grid, wherein each pixel, voxel, or each element of the finite element grid represents a fluid volume.
[0180] Embodiment 16. A system comprising: an electrical tomography system; a volume estimation system, the volume estimation system comprising: at least one processor; and at least one non-transitory computer-readable storage medium having instructions stored thereon, the instructions, when executed by the at least one processor, causing the volume estimation system to: reconstruct an initial impedance image based at least in part on received electrical tomography data of a domain; receive prior information associated with the domain; enhance the initial impedance image based at least in part on the received prior information to generate an enhanced impedance image; and generate a volume image of a region of interest of the enhanced impedance image based at least in part on the enhanced initial impedance image, wherein the volume image represents multiple values indicating a volume of a fluid.
[0181] Embodiment 17. A system according to claim 16, wherein the volumetric image is constructed from a plurality of pixels, voxels, or a plurality of elements of a finite element mesh, wherein each pixel, voxel, or each element of the finite element mesh represents a fluid volume.
[0182] Embodiment 18. A system according to any one of embodiments 16 or 17, wherein reconstructing the initial impedance image includes reconstructing an initial absolute impedance image.
[0183] Embodiment 19. A system according to any one of embodiments 16 to 18, the system also including instructions that, when executed by the at least one processor, cause the volume estimation system to segment the image in the prior information.
[0184] Embodiment 20. A system comprising: an electrical tomography system; a volume estimation system, the volume estimation system comprising: at least one processor; and at least one non-transitory computer-readable storage medium having instructions stored thereon, the instructions, when executed by the at least one processor, causing the volume estimation system to: receive prior information associated with a domain; receive an initial impedance image of a portion of the domain from the electrical tomography system; enhance the initial impedance image based at least in part on the received prior information to generate an enhanced impedance image; and generate a volume image of a region of interest of the enhanced impedance image based at least in part on the enhanced initial impedance image, wherein the volume image represents multiple values indicating a volume of a fluid.
[0185] Although the present disclosure has been described herein with respect to certain illustrated embodiments, those of ordinary skill in the art will recognize and appreciate that it is not limited thereto. Rather, many additions, deletions, and modifications may be made to the illustrated embodiments without departing from the scope of the claimed invention (including its legal equivalents). Additionally, features from one embodiment may be combined with features of another embodiment while still being encompassed within the scope of the present disclosure as contemplated by the inventor. Furthermore, embodiments of the present disclosure have utility for a variety of different tool types and configurations.
Claims
1. A method for estimating a volume of a fluid, the method comprising: receiving electrical tomography data of a portion of a field; reconstructing an initial impedance image based at least in part on the electrical tomography data; enhancing the initial impedance image to generate an enhanced impedance image; segmenting the enhanced impedance image to identify one or more tissues depicted within the enhanced impedance image; selecting a region of interest within the enhanced impedance image; determining a relationship parameter that relates an electrical characteristic represented within the region of interest of the enhanced impedance image to one or more characteristics of the region of interest; as well as A fluid volume within the region of interest is estimated based at least in part on the relationship parameter and the enhanced impedance image. 2 . The method of claim 1 , further comprising removing one or more portions outside the region of interest of the enhanced impedance image from the enhanced impedance image. 3 . The method of claim 2 , wherein removing the one or more portions comprises removing portions of the enhanced impedance image that do not satisfy a threshold electrical characteristic value.
4. The method of any one of claims 1 to 3, wherein the one or more characteristics include one or more of air content, blood content, or tissue content.
5. The method according to any one of claims 1 to 3, further comprising: Receiving prior information; determining a fluid content indicated in the a priori information; associating the fluid content indicated in the a priori information with the relationship parameter; as well as The fluid volume within the region of interest is estimated based at least in part on the received a priori information. The method of claim 5 , wherein the received a priori information comprises one or more CT scans of the domain.
7. The method of claim 5, wherein the prior information comprises image data of one or more anatomical and physiological models.
8. The method of claim 5, further comprising selecting the prior information to be received based on one or more anthropometric values of the domain.
9. The method of claim 5, further comprising segmenting the enhanced impedance image to identify objects depicted in the prior information.
10. The method of any one of claims 1 to 3, wherein estimating the fluid volume within the region of interest comprises estimating the volume of air within at least a portion of a lung of the region. The method of claim 10 , wherein estimating the air volume comprises estimating an absolute air volume.
12. The method of claim 11, wherein reconstructing the initial impedance image comprises reconstructing an initial absolute impedance image.
13. The method of claim 11, wherein reconstructing the initial impedance image comprises reconstructing an initial differential impedance image.
14. A method according to any one of claims 1 to 3, further comprising outputting a representation of the estimated fluid volume.
15. The method of claim 14, wherein the representation comprises a volumetric image having a plurality of pixels, voxels, or a plurality of elements of a finite element grid, wherein each pixel, voxel, or each element of the finite element grid represents a fluid volume.
16. A system, comprising: Electrical tomography system; A volume estimation system, the volume estimation system comprising: at least one processor; and at least one non-transitory computer-readable storage medium having instructions stored thereon, The instructions, when executed by the at least one processor, cause the volume estimation system to: reconstructing an initial impedance image based at least in part on the received electrical tomography data of the domain; receiving a priori information associated with the domain; enhancing the initial impedance image based at least in part on the received prior information to generate an enhanced impedance image; determining a relationship parameter correlating an electrical characteristic of the domain of the enhanced impedance image to one or more characteristics of the domain; and A volumetric image of a region of interest of the enhanced impedance image is generated based at least in part on the enhanced impedance image and the relationship parameter, wherein the volumetric image represents a plurality of values indicative of a volume of a fluid.
17. The system of claim 16, wherein the volumetric image is constructed from a plurality of pixels, voxels, or a plurality of elements of a finite element mesh, wherein each pixel, voxel, or each element of the finite element mesh represents a fluid volume.
18. The system of claim 16, wherein reconstructing the initial impedance image comprises reconstructing an initial absolute impedance image.
19. The system of any one of claims 16 to 18, further comprising instructions that, when executed by the at least one processor, cause the volume estimation system to segment the image in the a priori information.
20. A system, comprising: Electrical tomography system; A volume estimation system, the volume estimation system comprising: at least one processor; and at least one non-transitory computer-readable storage medium having instructions stored thereon, The instructions, when executed by the at least one processor, cause the volume estimation system to: receiving prior information associated with the domain; receiving an initial impedance image of a portion of the domain from the electrical tomography system; enhancing the initial impedance image based at least in part on the received prior information to generate an enhanced impedance image; determining a relationship parameter correlating an electrical characteristic of the domain of the enhanced impedance image to one or more characteristics of the domain; and A volumetric image of a region of interest of the enhanced impedance image is generated based at least in part on the enhanced impedance image and the relationship parameter, wherein the volumetric image represents a plurality of values indicative of a volume of a fluid.
