Temporal and spatial remote sensing system

Abstract

A temporal and spatial remote sensing system comprises a temporal and spatial multi-sensor, a first processor in data communication with the temporal and spatial multi-sensor, and a second processor in data communication with the first processor. The multi-sensor includes a first sensor for collecting a first set of data through a first modality and a second sensor for collecting a second set of data through a second modality. The first modality may involve electrical impedance and the second modality may involve ultrasound. The first processor pre-processes the collected first and second sets of data in the time domain. The second processor processes the pre-processed first and second sets of data in combination to generate tomography data.

Description

TEMPORAL AND SPATIAL REMOTE SENSING SYSTEMTECHNICAL FIELD

[0001] The present disclosure relates generally to tomography, and in particular to a temporal and spatial remote sensing system. Particular embodiments have example applications for medical and / or geophysical imaging.RELATED APPLICATIONS

[0002] This application claims priority from United States Patent Application No. 63 / 542,741 fded on October 5, 2023 entitled “TEMPORAL AND IMPEDANCE MULTI-SENSOR SYSTEM”. For the purposes of the United States, this application claims the benefit under 35 U.S.C. §119 of United States Patent Application No. 63 / 542,741 filed on October 5, 2023 entitled “TEMPORAL AND IMPEDANCE MULTI-SENSOR SYSTEM”. United States Patent Application No. 63 / 542,741 is incorporated herein by reference in its entirety for all purposes.BACKGROUND

[0003] Tomography is an imaging technique used to generate cross-sectional and volumetric images of an object or body. It involves capturing projections of an object from various angles and processing the projections using computer algorithms to construct detailed two-dimensional (2D) or three-dimensional (3D) images of the object. The images can help provide information regarding the internal structure or physical properties of the imaged object.

[0004] There are various different modes of tomography that can be used for various applications. Some example modes of tomography include, but are not limited to, x-ray computed tomography (CT), magnetic resonance imaging (MRI) tomography, ultrasound tomography, electrical impedance tomography (EIT), optical coherence tomography (OCT), positron emission tomography (PET), near-infrared (NIR) tomography, and electron tomography (ET). Each mode of tomography has its advantages and disadvantages.

[0005] As an example, x-ray tomography and MRI tomography are commonly used in medical imaging, which is crucial for determining a patient’s anatomy and physiology and provides doctors with information to make accurate diagnoses. X-ray tomography involves relatively low costs, butrelies on ionizing radiation which can increase the risk of cancer to the patient. MRI tomography has less risks from a health perspective, but involves relatively higher costs.

[0006] In general, existing tomography systems tend to require large equipment that are not mobile. For example, some MRI machines can weigh more than 20,000 lbs and generate magnetic fields of up to 3T, making them dangerous and impractical to transport. Companies have accordingly gone to great lengths to integrate custom magnetically shielded tractor-trailers in order to transport and operate full-sized MRI machines in remote locations. Large philanthropic institutions have also taken aim by funding research into mobile MRI machines that are capable of being wheeled through a hospital for the purposes of bedside imaging. These portable machines tend to be costly to manufacture and are usually not capable of capturing high resolution images.

[0007] With healthcare systems worldwide being challenged to provide care to remote patients rather than anchoring services to conventional brick and mortar hospitals and clinics, first responders must be well-equipped to assess injuries at the point of event. In addition, with hundreds of space missions planned over the next several years, astronauts will soon be tasked with collecting geotechnical information on regolith composition for in situ mining as well as its physical characteristics to assess the suitability of the soil as a base for the lunar construction activities.

[0008] There remains a need for improved tomography systems, especially systems that are suitable for use in medical and / or geophysical imaging applications. There remains a need for improved tomography systems that can generate high resolution images without consuming excessively large amounts of processing power. There remains a need for portable tomography systems that can be deployed in space or other remote and inaccessible communities. Such systems may act as an alternative to the CT, MRI or PET scanners that are currently being used by field medics or astronauts in space stations and habitats in low earth orbit or on the Moon.SUMMARY

[0009] One aspect of the present invention relates to a method for generating tomography images. The method involves collecting a first set of data through a first modality, collecting a second set of data through a second modality, pre-processing the collected first and second sets of data in the time domain, and applying a machine learning model and a filter to the pre-processed first and second sets of data to obtain a combined dataset. The first set of data corresponds to functionaldata, and the second set of data corresponds to structural data. The tomography image is generated from the combined dataset.

[0010] Another aspect relates to a temporal and spatial remote sensing system comprising a temporal and spatial multi-sensor, a first processor in data communication with the temporal and spatial multi-sensor, and a second processor in data communication with the first processor. The multi-sensor comprises a first sensor for collecting a first data set through a first modality and a second sensor for collecting a second data set through a second modality. The first processor is configured to pre-process the collected first and second data sets in the time domain. The second processor is configured to combine the pre-processed first and second data sets to generate tomography data. The first modality may be electrical impedance and the first sensor may be an electrical impedance tomography sensor. The second modality may be ultrasound and the second sensor may be an ultrasonic sensor. The first data set may correspond to functional data, and the second data set may correspond to structural data.

[0011] In some embodiments, the first processor comprises a control circuit for controlling operation of the first and second sensors. The first processor may be configured to implement a time domain solution to reduce noise in the first and second data sets. The first processor may be configured to process the first and second data sets separately. The first processor may be configured to implement a time-gating technique to isolate portions of the first and second data sets for further processing.

[0012] In some embodiments, the second processor comprises an artificial intelligence and machine learning (AM / ML) module for obtaining inverse solutions from the pre-processed first and second data sets. The AM / ML module may be configured to implement a Bayesian machinelearning model pre-trained with experimental data. In some embodiments, the second processor comprises a filtering module. The filtering module may be configured to implement a Kalman filter on the pre-processed first and second data sets. The filtering module may be configured to combine the pre-processed first and second data sets to reduce common noise.

[0013] In some embodiments, the first sensor and the second sensor are configured to collect the first and second data sets simultaneously. In other embodiments, the first sensor and the second sensor are configured to collect the first and second data sets asynchronously. In some embodiments, the system comprises a probe and the first and second sensors are contained in the probe. The probe may comprise a layer of conducting gel for contacting a surface.

[0014] Other aspects relate to geophysical data collection systems comprising the temporal and spatial remote sensing system described herein. Other aspects relate to medical imaging systems comprising the temporal and spatial remote sensing system described herein.

[0015] Additional aspects of the present invention will be apparent in view of the description which follows.BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Various objects, features and advantages of the invention will be apparent from the following description of particular embodiments of the invention, as illustrated in the accompanying drawings in which:

[0017] FIG. 1 is a block diagram of a temporal and spatial remote sensing system according to an example embodiment.

[0018] FIG. 2 is a schematic illustration of a rover equipped with a remote sensing system and collecting geophysical data along a path, in relation to an origin, on a celestial body surface according to another embodiment.

[0019] FIG. 3 is a schematic illustration of the FIG. 1 system being used to collect geophysical data.

[0020] FIG. 4 is a schematic illustration of a remote sensing system used in a medical application according to another embodiment.

[0021] FIG. 5 is a schematic illustration of the FIG. 1 system being used to scan a patient’s body.

[0022] FIG. 6 depicts simulation results of fusing ultrasound data with EIT data.DETAILED DESCRIPTION

[0023] The description which follows and the embodiments described therein are provided by way of illustration of examples of particular embodiments of the principles of the present invention. These examples are provided for the purposes of explanation and not limitation of those principles and of the invention. In some instances, certain structures and techniques have not been described or shown in detail in order not to obscure the invention.

[0024] Aspects of the invention relate to tomography systems which combine two or more different sensing and imaging modalities or techniques to generate or reconstruct images of one or more sections of an imaged object. Such systems may be referred to herein as “hybrid tomography systems”, “hybrid sensors”, “temporal and spatial multi-sensor systems” or “temporal and spatial remote sensing systems”. Similarly, images generated or reconstructed using hybrid tomography systems may be referred to herein as “hybrid tomography images”. With geophysics becoming a new frontier in resource prospecting and exploration, combining various modalities, such as electrical, ultrasound, magnetic, optical, and opportunistic seismic signals (e.g., P-waves, S-waves, Rayleigh waves, and surface waves), will become an attractive means for providing higher quality imaging with better spatial and / or temporal resolution.

[0025] The production or reconstruction of hybrid tomography images may in some cases require the hybrid tomography system to resolve inverse solutions in multi-physics modalities. Hybrid tomography systems described herein may be designed or otherwise configured to resolve these inverse solutions. Hybrid tomography systems described herein may also collect large samples of data and use one or more machine learning techniques to generate hybrid tomography images. Such systems may be used for geophysical imaging, providing a 3D topographical interpretation of the reserves in situ and enhanced capabilities to delimit the various minerals, water, and consumables in the ground using remote processes. Such systems may also be used as early detection diagnosis tools that distinguish between healthy and diseased tissues or for medical imaging, providing 3D tomography, without using magnets or radiation, that can identify inclusions in the human body.

[0026] Unlike state of the art systems which rely on frequency domain (FD) analysis, systems described herein may incorporate a multi -frequency time domain (TD) approach (e.g., time domain gates) to collect additional information related to the impedance and other physical or chemical properties of an object or region of interest (e.g., tissues), thereby enabling more accurate distinction between objects with different properties. In contrast to traditional FD analysis which require multiple single-frequency signals (which greatly extends measurement and data processing times), systems described herein may rely on the use of time domain (TD) signals with multifrequency content, such as the Heaviside step function or a pulse, to improve measurement and processing times and enable more sophisticated real-time solutions. The TD signals may contain transition and boundary information that is typically discarded during FD processing but may be beneficial to the image reconstruction.

[0027] FIG. 1 is a block diagram of a multi-modality temporal and spatial remote sensing system 100 in accordance with an example embodiment of the invention. Compared to traditional imaging systems, system 100 is generally more portable and can be operated at lower power levels, making it suitable for use in remote environments and for space operations. System 100 leverages the time domain in post processing, including noise-based filtering and the knowledge of a prior, and artificial intelligence and machine learning (AI / ML) to reconstruct 3D images. Compared to traditional tomography systems, system 100 can provide greater resolution, while consuming less processing power. System 100 may be designed or otherwise configured for Earth or other celestial body use. In some embodiments, the payload of system 100 may be designed or otherwise configured to enable more efficient future in-situ resource utilization (ISRU) operations and planetary exploration.

[0028] System 100 comprises a temporal and spatial multi-sensor 110, which is a hybrid sensor operable to collect tomography data from two or more different modalities. In the illustrated embodiment, multi-sensor 110 comprises a first sensor 101 operable to collect data from a first modality and a second sensor 102 operable to collect data from a second modality. First sensor 101 and second sensor 102 may be synchronized by a processor 103 of system 100 to collect data from the two different modalities simultaneously or at different times. As described in more detail below, first sensor 101 and second sensor 102 may be encapsulated or otherwise contained within one or more probes. As an example, multi-sensor 110 may comprise four probes and each probe of multisensor 110 may contain a first sensor 101 and a second sensor 102.

[0029] First sensor 101 may be a type of sensor used to collect functional data (e.g., electrical impedance tomography, positron emission tomography, light diffraction, etc.). Second sensor 102 may be a type of sensor used to collect structural data (e.g., ultrasound, X-ray, computed tomography, etc.). In some embodiments, the first sensor 101 is an electrical impedance tomography (EIT) sensor measuring impedance across the time domain and the second sensor 102 is an ultrasonic sensor. For such embodiments, system 100 may also be referred to herein as a temporal and impedance prospecting system (TIPS) and multi-sensor 110 may also be referred to herein as a temporal and impedance multi-sensor (TIMS).

[0030] EIT sensor 101 may collect and transmit raw EIT data 101A as a first data stream to a first data processor 103 of TIPS 100. The raw EIT data 101A may consist of voltage measurements recorded from one or more surface electrodes of EIT sensor 101 in response to a known injected current. Similarly, ultrasound sensor 102 may collect and transmit raw ultrasound data 102A as asecond data stream to first data processor 103 as shown in FIG. 1. The raw ultrasound data 102A may consist of time-series signals that represent the intensity of sound waves received over time by one or more transducers of ultrasound sensor 102.

[0031] First data processor 103 comprises a control circuit and means for processing the various modes of data collected by the various sensors of multi-sensor 110. In the illustrated embodiment, first data processor 103 is configured to process raw EIT data 101 A and raw ultrasound data 102A in the time domain to generate one or more sets of processed data 101A’, 102A’. The data processing methods implemented by first data processor 103 may use a time domain solution to reduce noise and / or maximize signal from the two sensors 101, 102. In some embodiments, data processor 103 is configured to implement one or more of the following: averaging repeated measurements, transformations (e.g., extracting signal power), filtering stages (e.g., bandpass, matched filters), Independent Component Analysis (ICA), and wavelet denoising or a combination of any of the foregoing.

[0032] Raw EIT data 101 A and raw ultrasound data 102A are typically processed separately by first data processor 103. The raw data may be processed by one or more central processing units (CPUs), programmable gate arrays, and / or graphics processing units (GPUs) of first data processor 103. The raw EIT data 101 A could, for example, be processed to validate measurements, normalize and calibrate signals, correct drifts, and / or format data for enhancing quality. The raw ultrasound data 102A could, for example, be processed using filtering (e.g., bandpass filtering, matched filters, etc.), gain adjustment, time gating, time windowing, averaging, and other signal enhancement techniques to remove noise and artifacts. The processed data 101A’, 102A’ are then transmitted or otherwise provided to a second data processor 109 of TIPS 100 for further data processing.

[0033] As shown in FIG. 1, in one embodiment, second data processor 109 comprises an artificial intelligence or machine learning (AI / ML) module 104 for finding inverse solutions and a filtering module 105. The modules may be implemented by diverse computing hardware, software combinations, programmable gate arrays, or computers, including dedicated digital processing systems, executing software code. For example, the modules may be implemented as one or more sets of computer-readable instructions stored in a memory device and executable by the second data processor 109.

[0034] Unless context dictates otherwise, the term “inverse solution”, as used herein, refers to the process of reconstructing a 2D image or a 3D model from the data collected by multi-sensor 110.The process may involve solving for both a direct problem and an inverse problem, with the goal of determining the internal material or structural properties of an object or region of interest based on the measurements taken from sensors 101, 102 at different angles. The direct problem involves projecting the measured quantities (e.g., voltages in EIT, sound wave magnitudes in ultrasound, etc.) through the internal structure of the object or region of interest, such as a series of 2D images or slices or a 3D volume. The inverse problem involves reconstructing or generating a depiction of the original material or structural properties (e.g., impedances in EIT) in 2D or 3D. Such properties are responsible for generating the measured quantities, and are estimated in the direct problem.

[0035] To find the inverse solution, second data processor 109 models the relationship between 2D projections and 3D structures to combine the direct problem solutions from first and second sensors 101, 102, and applies a reconstruction technique (e.g., filtered back projection, single-step reconstruction, iterative reconstruction methods, algebraic reconstruction techniques, etc.) to solve the inverse problem and obtain tomography data 106 corresponding to a 2D or 3D model. The reconstructed model may then be validated against known data, and additional measurements or refinements may be made to improve accuracy.

[0036] The AI / ML module 104 of second data processor 109 may execute algorithms pre-trained with experimental data, synthetic data, and / or additional data (e.g., in situ data or human / animal / estimation data), based on prior Bayesian statistics, in order to further reduce the noise. The training may involve, for example, providing experimental data to one or more machine learning models and iteratively repeating the process to improve the predicted model output based on input data. The filtering module 105 of second processor 109 may implement, for example, noise abating, gradient transfer methods, Gaussian processes, and / or a Kalman filter. Filtering module 105 may also mix or combine the processed EIT data 101 A’ and the processed ultrasound data 102A’ together, reducing common and correlated noise between the two sets of data. Mixing high-precision measurements from one modality (e.g., EIT) with low-quality measurements from another modality (e.g., ultrasound), and processing the mixed measurements through a Bayesian machine learning algorithm and / or Kalman filter can improve the inverse solution. By combining functional data obtained from first sensor 101 and structural data obtained from second sensor 102, the fusion of data sets by second processor 109 may be accelerated to produce higher-quality tomography.

[0037] After processing the data 101A’, 102A’ that has been pre-processed by first data processor 103, second data processor 109 outputs tomography data 106. Tomography data 106 may includeboth two-dimensional and three-dimensional tomography data, thereby providing a more accurate version of the tomography image.

[0038] In some embodiments, system 100 may be configured to implement one or more timegating techniques using first data processor 103 and / or second data processor 109. Some of these techniques may involve isolating a portion of the data for further analysis. For example, first data processor 103 may be configured to isolate portions of raw EIT data 101 A and raw ultrasound data 102A, and the isolated portions of the raw data may be analyzed by second data processor in the process of generating tomography data 106. Illustratively, time-gating enables discriminated trajectory-dependent signals, thus enabling measurements of the 3D flow of signals and time-of- flight (e.g., ballistic, diffracted and reflected electrons, photons and phonons) data for the stochastic transmitted and conducted signals on the imaged object (e.g., soil and rock material). Time-gating also allows signals with similar trajectory delay to be measured and used to improve the inverse solution. Further advantages may be achieved by time-resolving the measurements of impedance, using time-gating, thereby fine-tuning the system’s capacity to topographically image various layers, provide better resolution in the inverse reconstruction of the impedance, and / or better discriminate various objects.

[0039] The use and placement of multiple electrodes and acoustic sensors combined will also benefit the refinement of the inverse solutions required for tomography. In some embodiments, multiple wavelengths (e.g., four or more wavelengths) are used to over-determine the solution. This may result in the accumulation of large amounts of data that may be fed to a machine learning algorithm.

[0040] Illustratively, combining EIT with ultrasound can beneficially yield additional information on the tomography and composition when analyzed through, for example, the time domain methods described above. The impedance at different frequencies and the ultrasound at different frequencies can over-define the solution such that the solution to the inverse problem will result in a higher definition tomography with better compositional and functional separation from the data. This may result in or lead to additional AI / ML or classic inverse solutions.

[0041] In general, the angular resolution of a system may be approximated by the ratio between the wavelength probed (i.e., lambda) and the diameter of the aperture of the instrumentation (i.e., D). In addition, smaller wavelengths or higher frequencies can improve spatial resolution, with trade-offs such as higher signal losses at greater distances. By collecting data across multiplefrequency bands, system 100 can identify different characteristics of the inclusions and / or subsurface material being probed to thereby improve the resolution of the tomographic image. By incorporating modalities which complement each other (e.g., electrical impedance and acoustic impedance), system 100 can identify different material properties (or gradients thereof) of the imaged object or area of interest and reconstruct higher-definition images.

[0042] In addition to the exemplary aspects described above, two applications of system 100 are described in the following Examples, which are set forth to aid in the understanding of the invention and should not be construed to limit in any way the scope of the invention as defined in the claims which follow thereafter.EXAMPLE 1 - GEOPHYSICAL IMAGING

[0043] Systems of the type shown in FIG. 1 can be used for geophysical imaging, including for the geophysical imaging of the Earth’s surface or subsurface or the geophysical imaging of an extraterrestrial body.

[0044] By way of background, NASA’s Artemis Mission is on its way to establishing a permanent lunar base by 2030 and the international community is mobilizing to create technology and solutions that will ensure the mission’s success. There are also hundreds of other missions planned to the Moon over the next several years, which include both missions funded by space agencies and missions funded by the commercial space sector. A primary objective is the development of technologies to enable in-situ resource utilization (ISRU) for rich deposits of precious metals, minerals, water, and oxygen from the lunar surface. Furthermore, there exists a need for geotechnical applications and medical applications on Earth as well.

[0045] Studies have found computer processing to be essential for all stages of data manipulation in archaeological prospection. It has been shown that the workflow of satellite imagery, aerial photography, and geophysical prospection can be broken down into measurements, acquisition, processing, visualization and interpretation, and that the advantages of digital data manipulations can be explored with individual examples based on these categories. It has also been shown that informatics can greatly assist with the final archaeological analysis of the measurements. Accordingly, there is a general desire for data collection and analysis systems which aid human experience and assessments to arrive at meaningful interpretations of geophysical data.

[0046] FIG. 2 schematically shows a top view of a vehicle or rover 600 (e.g. , a lunar rover) moving along a path 120, in relation to a notional origin point 121, on a celestial body surface 198 (e.g., surface of the Earth, surface of the Moon, etc.). Rover 600 is equipped with or otherwise carries one or more remote sensing systems 100 configured to collect geophysical information and generate tomography images therefrom. Optionally, rover 600 may also be equipped with one or more magnetic field detectors enabling measurement of magnetite composition around the probes. The magnetic field detectors can help collect ground truth data.

[0047] System 100 of vehicle 600 is light in weight, requires relatively low power consumption, and has the capacity to perform near-surface and sub-surface imaging of sufficiently high resolution to identify usable resources and enable long duration operation on the moon or another celestial body. Because of its size and nimble resource requirements, system 100 can be borne directly by rovers 600 and / or used during extravehicular (EVA) crew activities. Furthermore, system 100 enables geotechnical surveys for the safe positioning of habitats, structures, and infrastructure on the surface of planetary bodies. When placed on permafrost water ice, infrastructure will collapse due to the inopportune pressure melting of sub-lunar ice. Fusing the geological data to electro-optic soil imagery provides an additional avenue of improvements.

[0048] In the illustrated embodiment, system 100 comprises four probes 111, 112, 113, 114. Depending on the desired configuration, each probe may house or support its own multi-sensor 110, or each probe may house or support part of a common single multi-sensor 110. Each probe may comprise means for penetrating into the celestial body surface 198 (e.g., a lunar surface) and / or past a regolith of the celestial body surface 198. For example, probes 111, 112, 113, 114 may penetrate the surface 198 using tip vibration and nominal downward force, rod tip in a “drill bit” shape, percussive methods, a separate drill to create the hole, or other similar means.

[0049] System 100 may be operated or configured to collect geophysical data at discrete time intervals. For example, at time instances tl, t2, .... tn, vehicle 600 may be parked or stationarily positioned, at pl, p2, . . .pn, and the probes 111, 112, 113, 114 of system 100 may be inserted into the celestial body surface 198 or regolith to collect data. Each probe may comprise a first sensor 101 and a second sensor 102. The sensors 101, 102 may collect raw data at time instances tl, t2, .... tn, corresponding to positions at pl, p2, ...pn, and transmit or otherwise provide the data to processors 103, 109 to generate a 3D tomography mapping 106 of the Earth or celestial body.

[0050] FIG. 3 is a schematic illustration of system 100 being used for collecting geophysical data on Earth or another planetary body. As shown in FIG. 3, the data processor and control circuit 103 of system 100 provide control signals to probes 111, 112, 113, 114. Each probe may be in physical contact with and / or at least partially penetrate through a surface of the ground, permafrost, soil, ore body, or regolith. Each of the probes 111, 112, 113, 114 may collect data via two or more different modes of tomography (e.g., EIT, ultrasound, seismic, infrasound, etc.) and generate data streams 101A, 102A. The data streams 101A, 102A are then provided to the data processor and control circuit 103. The probe data streams 101 A, 102A may be provided to processor 103 via either wired or wireless transmission.

[0051] In some cases, the tomography data 106 obtained from system 100 will be processed in situ by one or more supplementary systems. The processing may involve obtaining positionally accurate location data of vehicle 600 in relation to the origin 121. For example, vehicle 600 may traverse along path 120 to map the environment by obtaining tomography data 106 from system 100 and positional data (i.e., data corresponding to locations pl, p2. . .pn) using other systems. The inverse solution of the time-gated data may be produced locally in system 100. The tomography data 106 may be stored in system 100 or a supplementary system in the form of voxelized images. Back-up copies of tomography data 106 (e.g. , a FEM voxel model) may be stored locally to further accumulate location level data.

[0052] The processing may also involve combining tomography data 106 with other sources of data, such as JMARS surface data. The fusion of ultrasound assisted extraction process (UAEx) data and data obtained from system 100 may be performed on the ground using acoustic and / or EIT finite element (FEM) models. Spurious information arising from bolide impact on the moon as seismic S-Waves may also be recorded and incorporated within the FEM model.

[0053] System 100 may be used to identify metallic oxides, anorthite and water in one or more phases on the surface or on the immediate near-surface of the lunar regolith material. While techniques for measuring soil resistivity are similar regardless of the purpose of the measurement, the interpretation of the recorded data can vary considerably, especially where soils with non- uniform resistivities are encountered. The added complexity caused by nonuniform soils is common, and in only a few cases are the soil resistivities constant with increasing depth. System 100 provides a novel approach to survey large areas of lunar terrain or terrain on Earth in situ, through the use of amalgamated geophysics sensors working in harmony.

[0054] On Earth, both EIT and UAEx can be used to measure the resistivity and natural frequencies of various deposited materials in, for example, the Arctic regions (e.g., permafrost deposits, to size and locate water ice of all forms). Frequency domain EIT techniques are effective for providing a rough tomographic image, but do not support a time domain solution corresponding to a path dependent slice of the photon's trajectory. This combination of EIT and UAEx provided by system 100 will add further location abilities to usual EIT and allow the identification and positioning of ISRU resources capable of producing water from the proton-laiden regolith and / or from lenticular or granular ice deposited and maintained in permanently shadowed regions on the moon.

[0055] Table 1 below shows a comparison of different sensor types based on objective requirements. As shown in Table 1, the time domain EIT / UAEx system (e.g., TIPS) provides better mass payback ratio and is the least heavy geophysical system for locating shallow and deeper underground (dependent on the separation of receivers) resources and water on the moon. Furthermore, TIPS 100 provides higher value compared to ground penetrating radar, less artefacts, better identification of resources and at least the same or better imagery resolution compared to other systems.Table 1: Comparison of different sensor types based on objective requirements.

[0056] Compared to scanning techniques, like electro optic (E / O) techniques, that can resolve to a higher accuracy (e.g, accuracies in the mm, cm range), TIPS 100 is advantageous because it is not limited to collecting 2D surface information. Compared to ground penetrating radar (GPR) which can image quite deep, TIPS 100 is advantageous because it is not limited by the sidelobe of the GPR or the ground surface reflection noise and does not need to be physically moved to obtain additional volumetric information. Compared to neutron spectrometers which may give good information about the elements contained within the first 20 cm, TIPS 100 is advantageous because it can distinguish between various poly-element compounds (e.g., economic ore) and provide information about the tomography of the ground substructure.

[0057] The technology described herein can also be used on Earth to provide data lateral to geophysics data with higher resolution tomography, thereby providing improved data andresolution of identification of various metal deposits and the water body locations. For example, TIPS 100 can enable a quicker solution and reach further depth than ground penetrating radar, opening new vistas for applications on Earth and other extraterrestrial bodies.

[0058] In general, the TIPS technology described herein can be used to provide both geophysical and geotechnical analysis for Earth and space, providing transformative localization and resolution, and subsurface tomography at sub kilometer scale depths in 3D. In space, novel facets of the TIPS technology include the use of ultrasonic for the in-situ mapping of composition of ground properties, and volatile content density mapping. The TIPS technology may be used to assess the polarization of signals in near-surface regolith or soil to determine optimal transmissivity and to estimate water content of regolith and quantify the reflectance of agglomerated regolith surfaces.EXAMPLE 2 - MEDICAL IMAGING

[0059] Systems of the type shown in FIG. 1 can also be used for medical imaging, including for various industrial and medical situations that require the non-invasive extraction of information from the inside of bodies. Examples of medical applications include, but are not limited to, functional neurological imaging, cancer screening, and monitoring lung ventilation, strokes, congestive heart failure, muscle health, bone fracture and healing.

[0060] By way of background, medical imaging is usually carried out through tomographic methods that generate images based on internal body properties. However, the image reconstruction involves an ill-posed mathematical inverse problem, for which accurate resolution demands proper boundary conditions for convergence, large computation time, and large computation capacity. Studies have explored the use of machine learning to develop an accurate solver for reconstructing medical images in real-time, including studies which compare machine learning results with other methods like the Iterative Gauss-Newton method and the Primal-Dual Interior-Point method for detecting internal body features, the latter being based on accurately predicting internal property distributions. However, there are currently no hybrid tomography systems that leverage EIT, ultrasound, and machine learning to reconstruct internal body properties with high accuracy and in a way that replaces more complex and slower alternatives.

[0061] EIT (z. e, Electrical Impedance Tomography) is a non-invasive and cost-effective method for medical imaging which utilizes non-ionizing (i.e., radiation-free) electrical properties such as electrical conductivity, resistance, reactance, and impedance to collect data of body tissues andorgans. Systems described herein, such as TIPS 100, may be designed or otherwise configured to implement various and different types of EIT, including absolute EIT (a-EIT), time difference EIT (td-EIT), and / or frequency difference EIT (fd-EIT). a-EIT relies on the free ion content in cells to determine and differentiate tissues. td-EIT compares two or more physiological states to reconstruct images. fd-EIT compares the states using input currents at two or more frequencies. a-EIT, fd-EIT and td-EIT may be used for different purposes. For example, a-EIT may be used to image biological tissue, while td-EIT may be used to image the movement of fluids or gases within the tissues. td- EIT may be used to compare a state of interest to a baseline condition or system. Unlike current medical imaging technologies, EIT systems do not require large, sophisticated equipment, leading to low-cost products which can be easily transported and medical imaging systems that are more accessible for rural or remote communities. EIT systems also provide high temporal resolution, which may enable real-time imaging.

[0062] FIG. 4 schematically shows a TIPS 100 that has been configured for use in non-invasive medical tomography according to one embodiment. Compared to the TIPS used for geophysical applications, the TIPS used for medical applications is usually operated at lower voltages and currents to reduce the risks of inadvertently harming the patient. TIPS 100 comprises a TIMS 110 which utilizes a combination of ultrasound and EIT with time domain analysis to map the density, type, and distribution of human tissue and bone underneath the skin surface. TIPS 100 may be configured to apply an alternating current at a single frequency and / or multiple frequencies (i.e., multi frequency-EIT (also known as EIT spectroscopy) to differentiate normal tissues from abnormal tissue (e.g., cancers, abnormal tissue in the brain, abnormal tissue in the pancreas, etc.).

[0063] TIMS 110 comprises or otherwise supports a plurality of probes 211 that are arranged in spatial relation to an origin 221 during operation. For example, probes 211 may be disposed in a grid configuration on a skin surface of a portion of the human body 299 as depicted in FIG. 4. By taking continuous surface electrode measurements using the probes 211, a tomographic figure of a patient’s body may be inferred and reconstructed in real time

[0064] In the embodiment illustrated in FIG. 4, TIPS 110 comprises a TIMS 110 with eight probes 211 that are attachable to the skin or surface of the body 299. The probes 211 do not penetrate the skin surface and may include a layer of conducting gel for contacting the skin surface. In operation, the probes 211 may be spaced apart at locations (e.g., pl, p2, ... pn) relative to the origin 221 that are suitable for each probe 211 to collect data along a notional X-Y grid to generate a 3Dtomography mapping of the body portion 299. The notional X-Y grid may be in spatial relation to the origin 221 .

[0065] FIG. 5 is a block diagram of TIPS 100 being used to scan a patient’s body according to another embodiment, where the data processor and control circuit 103 provides electrical signals to the probes 211 that are in physical contact with the skin of the body portion 298. Each of the probes 211 generates probe data streams which are then provided to the data processor and control circuit 103 in either a wired or wireless manner. For example, a strap with integrated EIT and ultrasound measurement probes may be provided with cables connecting to a computing device for handling the parameter settings, collection of data, and visualization.

[0066] To enhance biomedical image resolution, TIPS 100 may be configured to combine the time domain electrical signal with structural information from an auxiliary imaging method such as tomographic ultrasound. Such ultrasound fusion features may be implemented based on ultrasound image gradients. This can help reduce the risks associated with reconstruction algorithm development. The fusion of simulated ultrasound data with simulated EIT data is depicted in FIG. 6.

[0067] Systems of the type described herein may be combined with other systems for use in and beyond geophysical and medical applications. By way of example, various teachings of the present application may be readily combined with various teachings of any of the following documents, which are herein incorporated by reference: the paper entitled “Towards a Fast and Accurate EIT Inverse Problem Solver: A Machine Learning Approach” by Xose Fernandez-Fuentes, the paper entitled “Remote Sensing and Geophysical Prospection” by Schmidt, A., United States Patent Application Publication No. 20050187478, and United States Patent Application Publication No. 20030085338.

[0068] While the above description describes features of example embodiments, it will be appreciated that some features and / or functions of the described embodiments are susceptible to modification without departing from the spirit and principles of operation of the described embodiments. For example, the various characteristics which are described by means of the represented embodiments or examples may be selectively combined with each other. Accordingly, what has been described above is intended to be illustrative of the claimed concept and nonlimiting. It will be understood by persons skilled in the art that other variants and modifications may be made without departing from the scope of the invention as defined in the claims appendedhereto. The scope of the claims should not be limited by the preferred embodiments and examples, but should be given the broadest interpretation consistent with the description as a whole.Interpretation of Terms

[0069] The terms "including," "comprising," and variations thereof mean "including but not limited to," unless expressly specified otherwise. “Connected,” “coupled,” or any variant thereof means any connection or coupling, either direct or indirect, between two or more elements; the coupling or connection between the elements can be physical, logical, or a combination thereof. “Herein,” “above,” “below,” and words of similar import, when used to describe this specification shall refer to this specification as a whole and not to any particular portions of this specification. “Or” in reference to a list of two or more items, covers all of the following interpretations of the word: any of the items in the list, all of the items in the list, and any combination of the items in the list. A listing of items does not imply that any or all of the items are mutually exclusive, unless expressly specified otherwise. The terms "a," "an," and "the" mean "one or more," unless expressly specified otherwise.

[0070] The terms "an embodiment," "embodiment," "embodiments," "the embodiment," "the embodiments," "one or more embodiments," "some embodiments," and "one embodiment" mean "one or more (but not all) embodiments of the present invention(s)," unless expressly specified otherwise.

[0071] Where a component is referred to above, unless otherwise indicated, reference to that component should be interpreted as including as equivalents of that component any component which performs the function of the described component (i.e., that is functionally equivalent), including components which are not structurally equivalent to the disclosed structure which performs the function in the illustrated exemplary embodiments of the invention.

[0072] Furthermore, it will be appreciated that for simplicity and clarity of illustration, where considered appropriate, reference numerals may be repeated among the figures to indicate corresponding or analogous elements. In addition, numerous specific details are set forth in order to provide a thorough understanding of the example embodiments described herein. However, it will be understood by those of ordinary skill in the art that the example embodiments described herein may be practiced without these specific details. In other instances, well-known methods, procedures, and components have not been described in detail so as not to obscure the exampleembodiments described herein. Also, the description is not to be considered as limiting the scope of the example embodiments described herein.

[0073] Specific examples of systems, methods and apparatus have been described herein for purposes of illustration. These are only examples. The technology provided herein can be applied to systems other than the example systems described above. Many alterations, modifications, additions, omissions, and permutations are possible within the practice of this invention. This invention includes variations on described embodiments that would be apparent to the skilled addressee, including variations obtained by: replacing features, elements and / or acts with equivalent features, elements and / or acts; mixing and matching of features, elements and / or acts from different embodiments; combining features, elements and / or acts from embodiments as described herein with features, elements and / or acts of other technology; and / or omitting combining features, elements and / or acts from described embodiments.

[0074] Although the present invention has been described and illustrated with respect to preferred embodiments and preferred uses thereof, it is not to be so limited since modifications and changes can be made therein which are within the full, intended scope of the invention as understood by those skilled in the art.

Claims

CLAIMS1. A tomography system comprising: a temporal and spatial multi-sensor comprising a first sensor for collecting a first data set through a first modality and a second sensor for collecting a second data set through a second modality; a first processor in data communication with the temporal and spatial multi-sensor, the first processor configured to pre-process the collected first and second data sets in the time domain; and a second processor in data communication with the first processor, the second processor configured to combine the pre-processed first and second data sets to generate tomography data.

2. The system of claim 1, wherein the first modality is electrical impedance and the first sensor is an electrical impedance tomography sensor.

3. The system of claim 1 or claim 2, wherein the second modality is ultrasound and the second sensor is an ultrasonic sensor.

4. The system of claim 1, wherein the first data set corresponds to functional data, and wherein the second data set corresponds to structural data.

5. The system of any one of claims 1 to 4, wherein the first sensor and the second sensor are configured to collect the first and second data sets simultaneously.

6. The system of any one of claims 1 to 4, wherein the first sensor and the second sensor are configured to collect the first and second data sets asynchronously.

7. The system of any one of claims 1 to 6, wherein the first processor comprises a control circuit for controlling operation of the first and second sensors.

8. The system of any one of claims 1 to 7, wherein the first processor is configured to implement a time domain solution to reduce noise in the first and second data sets.

9. The system of any one of claims 1 to 8, wherein the first processor is configured to process the first and second data sets separately.

10. The system of any one of claims 1 to 9, wherein the second processor comprises an artificial intelligence and machine learning (AM / ML) module for obtaining inverse solutions from the pre-processed first and second data sets.

11. The system of claim 10, wherein the AM / ML module is configured to implement a Bayesian machine-learning model pre-trained with experimental data.

12. The system of any one of claims 1 to 11, wherein the second processor comprises a filtering module.

13. The system of claim 12, wherein the filtering module is configured to implement a Kalman filter on the pre-processed first and second data sets.

14. The system of claim 12, wherein the filtering module is configured to combine the pre- processed first and second data sets to reduce common noise.

15. The system of any one of claims 1 to 14, wherein the first processor is configured to implement a time-gating technique to isolate portions of the first and second data sets for further processing.

16. The system of any one of claims 1 to 15, comprising a probe, wherein the first sensor and the second sensor are contained in the probe.

17. The system of claim 16, wherein the probe comprises a layer of conducting gel for contacting a surface.

18. A geophysical data collection system comprising the system of any one of claims 1 to 17.

19. A medical imaging system comprising the system of any one of claims 1 to 17.

20. A method for generating a tomography image comprising: collecting a first set of data through a first modality, the first set of data corresponding to functional data; collecting a second set of data through a second modality, the second set of data corresponding to structural data; pre-processing the collected first and second sets of data in the time domain; applying a machine learning model and a filter to the pre-processed first and second sets of data to obtain a combined dataset; and generating the tomography image from the combined dataset.

21. Systems having any new and inventive feature, combination of features, or subcombination of features as described herein.

22. Methods having any new and inventive feature, combination of features, or subcombination of features as described herein.

Images