Mobile ultra-wideband radar for monitoring pleural fluid levels and cardiopulmonary function

Abstract

Various examples are provided related to mobile body monitoring using ultra-wideband radar. In one example, a method for determining body characteristics includes: collecting a reflection backscatter data set of an ultra-wideband (UWB) pulse sequence transmitted via an UWB sensor, and corresponding calibration measurements from a calibration channel; determining a reflection coefficient for each tissue interface based on the reflection backscatter data set; and determining a fluid level content of lung tissue based on the reflection coefficient. Based on the reflection backscatter data of the UWB pulse sequence and the corresponding calibration measurements, the reflection coefficient can be determined from a reflection profile. The UWB sensor can include an array of transmit (TX) and receive (RX) antenna pairs located on the user's body. The reflection profile can be associated with a model of tissue layers within the body between the UWB sensor and the lung tissue.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to and the benefit of co-pending U.S. Provisional Application No. 62 / 699,076, filed on July 17, 2018, entitled “MOBILE ULTRA-WIDEBAND RADAR FOR MONITORING PLEXIGLASS FLUID LEVELS AND CARDIO-RESPIRATORY FUNCTION,” which is incorporated herein by reference in its entirety.

[0003] STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT

[0004] This invention was made with government support under Grant U54 EB020404 from the National Institutes of Health and Grant ISS1231577 from the Information Robotics and Intelligent Systems (IIS) Division of the National Science Foundation. The government has certain rights in this invention. Background Art

[0005] Congestive heart failure (CHF) affects nearly 6 million Americans, with 670,000 diagnosed annually. Heart failure is one of the leading causes of hospitalization, readmission, and death in the United States (US) and one of the most costly disease syndromes, with direct and indirect care costs estimated at $34.4 billion per year. Approximately 80% of the high care costs are related to the management of heart failure decompensation in hospitals. Efforts should be aimed at improving heart failure outcomes and reducing care costs. Earlier recognition and treatment of worsening heart failure in the outpatient setting may prevent the development of heart failure exacerbations, which lead to increased morbidity and hospitalizations. Current identifiers of heart failure exacerbations, namely weight gain and dyspnea, are unreliable and often develop too late in the timeline of disease progression to change outcomes. Summary of the invention

[0006] Various aspects of the present disclosure relate to systems, devices, and methods for mobile body monitoring using ultra-wideband radar. In one aspect, a method for determining body characteristics includes: collecting a reflection backscatter data set of an ultra-wideband (UWB) pulse sequence transmitted by an UWB sensor, and corresponding calibration measurements from a calibration channel in the UWB sensor, the UWB sensor including an array of transmit (TX) and receive (RX) antenna pairs located on the user's body; determining a reflection coefficient for each tissue interface based on the reflection backscatter data set, the reflection coefficient being determined from a reflection profile based on the reflection backscatter data of the UWB pulse sequence and the corresponding calibration measurements, the reflection profile being associated with a model of tissue layers within the body between the UWB sensor and lung tissue; and determining a fluid level content of the lung tissue based on the reflection coefficient.

[0007] In one or more aspects, the reflected backscatter data set may include reflected backscatter data obtained for each of the TX and RX antenna pairs in the UWB sensor, and the reflected backscatter data are combined to generate a broadband beamforming signal for each set of reflected backscatter data. The reflection profile can be determined based on sparse deconvolution of the broadband beamforming signal of the reflected backscatter data set using a compensated UWB pulse shape based on the corresponding calibration measurement. The sparse deconvolution of the broadband beamforming signal can be implemented for each of the K frequency bands. In various aspects, the method may include identifying the depth of the lung tissue interface at the top (inhalation), middle and bottom (expiration) points in the respiratory cycle of the lung tissue based on the reflection coefficient; and determining the fluid level content of the lung tissue may include determining the fluid level content at the top, middle and bottom points in the respiratory cycle. In some aspects, the method may include determining the characteristics of a tissue layer located between the UWB sensor and the lung tissue. The characteristics of the tissue layer may include the location of at least one tissue layer interface or the dielectric properties of at least one tissue layer.

[0008] In another embodiment, a mobile body monitoring system includes: an ultra-wideband (UWB) sensor including an antenna array and a calibration channel, the antenna array including a transmit (TX) and receive (RX) antenna pair, the UWB sensor being configured to be positioned on a user's body; a radio frequency (RF) front end including a UWB pulse generator coupled to a TX antenna of the antenna array and a UWB receiver coupled to an RX antenna of the antenna array, wherein UWB pulses generated by the UWB pulse generator are sequentially transmitted into the user's body via the TX antenna, and reflected backscatter signals are received via the RX antenna of the TX and RX antenna pair; a wireless transmitter configured to transmit data associated with the reflected backscatter and corresponding calibration measurements from the calibration channel; and a computing device configured to receive the data and determine a body characteristic of the user based on the reflected backscatter and the corresponding calibration measurements.

[0009] In one or more aspects, the computing device may be configured to: determine a reflection profile based on data associated with reflected backscatter and corresponding calibration measurements of the transmitted UWB pulse sequence, the reflection profile being associated with a model of a tissue layer within the body between the UWB sensor and the target tissue; determine a reflection coefficient based on the reflection profile; and determine a characteristic of the target tissue based on the generated target tissue data. The characteristic of the target tissue may include a depth of an interface with the target tissue or a dielectric property of the target tissue. The target tissue may be lung tissue. The computing device may be configured to identify a measure of lung fluid content based on the characteristic of the lung tissue. The computing device may be configured to simultaneously identify one or more of heart rate, heart rate variability, respiratory rate, or tidal volume. The computing device may be configured to identify the top and bottom depths of the lung tissue interface during a respiratory cycle of the lung tissue. The computing device may be configured to identify dielectric properties at the top and bottom depths and average depth in a respiratory cycle of the lung tissue.

[0010] In various aspects, a reflection profile may be determined by sparse deconvolution of an average broadband backscatter signal based on data associated with reflected backscatter of a UWB pulse sequence. A computing device may be configured to determine a reflection profile for each of a series of reflected backscatter data sets, each of which includes data associated with reflected backscatter of a transmitted UWB pulse sequence associated with the set. In some aspects, a calibration channel may include a temperature calibration loop having a load of a known impedance located near the antenna array. Changes in transmitted UWB pulses may be compensated based on corresponding calibration measurements. In one or more aspects, a mobile body monitoring system may include a digital signal processing (DSP) circuit configured to obtain and process reflected backscatter signals and corresponding calibration measurements for transmission to a computing device. In various aspects, UWB pulses may be transmitted into the body at a rate of approximately 10,000 per second.

[0011] By studying the following description of the drawings and the specific embodiments, other systems, methods, features and advantages of the present disclosure will become apparent to those skilled in the art. All of these additional systems, methods, features and advantages are included in this specification, are within the scope of the present disclosure, and are protected by the appended claims. In addition, all optional and preferred features and modifications of the described embodiments can be used for all aspects of the disclosure taught herein. In addition, the various features of the dependent claims, as well as all optional and preferred features and modifications of the described embodiments are combinable and can be interchanged with each other. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Many aspects of the present disclosure may be better understood with reference to the following drawings. The components in the drawings are not necessarily drawn to scale, but emphasis is placed on clearly illustrating the principles of the present disclosure. In addition, in the drawings, the same reference numerals represent corresponding parts throughout the several views.

[0013] Figure 1A is a schematic diagram illustrating an example of a mobile body monitoring system according to various embodiments of the present disclosure.

[0014] Figure 1B and 1C According to various embodiments of the present disclosure Figure 1A Image of parts of a mobile body monitoring system.

[0015] Figures 2A-2F The various embodiments of the present disclosure are shown Figure 1A Example of an ultra-wideband (UWB) sensor for a body monitoring system.

[0016] Figure 3A and 3B According to various embodiments of the present disclosure, using Figures 2A-2F The UWB sensor senses the tissue.

[0017] Figures 4A-4C According to various embodiments of the present disclosure, Figures 2A-2F The backscattered response of a UWB sensor to the transmitted UWB pulse.

[0018] Figure 5A and 5B The various embodiments of the present disclosure are shown Figures 2A-2F Multi-layer model and localization of UWB sensors.

[0019] Figure 6 is a schematic diagram illustrating an example of a system imaging model according to various embodiments of the present disclosure.

[0020] Figures 7A-7C Examples of measured backscatter data, recovered sparse reflection profiles, and learned pulse shapes are shown according to various embodiments of the present disclosure.

[0021] Fig. 8A and 8B Examples of frequency bands of backscatter data according to various embodiments of the present disclosure are shown.

[0022] Fig. 9A and 9B An example of phase returns processed from backscatter data according to various embodiments of the present disclosure is shown.

[0023] Fig.10 is a diagram showing various embodiments according to the present disclosure Figure 1A A flow chart of an example of the operation of a body monitoring system.

[0024] Fig.11 It shows that various embodiments of the present disclosure can be used with Figure 1A An example of a computing device for use in conjunction with a body monitoring system.

[0025] Fig. 12A , 12B , 13 and 14A-14D show various embodiments according to the present disclosure, using Figure 1A Example of pilot study results for a body monitoring system.

[0026] Fig.15 An example of reflectance measured using a multi-tissue phantom according to various embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0027] Various examples related to systems, devices, and methods for mobile body monitoring using ultra-wideband radar are disclosed herein. Reference will now be made in detail to the description of the embodiments illustrated in the accompanying drawings, wherein like reference numerals refer to like parts throughout the several views.

[0028] Technological advances in earlier detection of heart failure have centered around the measurement of transthoracic and intrathoracic impedance, as fluid accumulation occurs before symptoms. The concept is based on the fact that conductivity increases with increasing fluid and impedance correspondingly decreases. Current methods, such as the OptiVol Fluid Index obtained from an implantable cardioverter-defibrillator (ICD), facilitate invasive measurement of impedance; however, accuracy in detecting pulmonary edema or predicting hospitalization has been highly variable. Another approach uses dedicated implantable hemodynamic sensors to monitor pulmonary artery pressure. These methods rely on information provided by implanted devices, and their applicability may be limited to all but the most advanced heart failure patients. Therefore, better noninvasive tools that can replicate the utility of these device-based diagnostic systems are needed.

[0029] The present disclosure proposes a non-invasive technology developed for easy body sensing, which measures thoracic fluid levels in addition to measuring heart and lung motion by transmitting ultra-wideband radio frequency pulses and analyzing backscattered waves. The body monitoring system can use a single sensor unit placed in front of the user's chest for measurements. The sensor unit can be placed in other locations to determine other tissue characteristics. Unlike similar technologies, the mobile body monitoring system can be used to assess the volume of fluid in the chest tissue and its spatial distribution, informing the volume inside and outside the blood vessels; potential clinically relevant measurements. In addition, the fast acquisition speed of the body monitoring system allows tracking of the movement of the heart and lungs, enabling continuous monitoring of heart rate, heart rate variability, respiratory rate, and tidal volume. Lung and heart measurements can be correlated to further assess the user's condition. These cardiovascular system status markers, used together with thoracic fluid levels, can provide a comprehensive set of measurements for predicting heart failure events with high sensitivity, low false alarm rates, and sufficient lead time.

[0030] refer to Figure 1A , a schematic diagram illustrating an example of a mobile body monitoring system 100 is shown, which can be controlled by a computing device interface such as, for example, a smartphone interface. The body monitoring system 100 is an ultra-wideband (UWB) radar system that sends short pulses (e.g., 0.3-0.4ns duration, 0.5-3.5GHz for UWB) to the body of a user 103 and records backscatter from the tissue. Radio frequency (RF) sensing is ideal for monitoring the internal movement of fine particles because it can penetrate tissue. For example, each tissue interface at the air / skin, skin / fat, fat / muscle and / or muscle / lung transition provides a reflection point that can be tracked in real time by processing the backscattered echo signal.

[0031] The mobile body monitoring system 100 can utilize a low-power micro-UWB platform to detect backscattered energy reflected by tissue and its transitions and measure the motion of the heart and lungs, as well as determine other body characteristics, such as pleural fluid levels, which can be used to detect congestion in CHF. Figure 1AAs shown, the mobile body monitoring system 100 can include a UWB pulse generator 106 that generates one or more UWB pulses that are transmitted into the tissue of the user 103 by an antenna 109t coupled through a transmit (TX) switching matrix 112. For example, the UWB pulse generator 106 can generate UWB pulses operating at a frequency of 0.45-3.55 GHz. Backscatter from the tissue interface is received by an antenna 109r coupled to a receive (RX) switching matrix 115. The RX switching matrix 115 directs the received backscattered signal to a UWB receiver 121 through a broadband low noise amplifier (LNA) 124. A compressed sampling scheduler 118 can coordinate switching between different antennas 109 for transmitting UWB radar pulses and receiving backscatter. Multiple-input / multiple-output (MIMO) diversity can be used to focus the signal on a moving source or target area. Figure 1B is an image showing an example of a platform in which the TX switch matrix 112 and the RX switch matrix 115 are coupled to the antenna 109 .

[0032] Digital signal processing (DSP) and wireless transmission circuitry 127 can process the backscattered signal and transmit the signal data wirelessly (e.g., via The DSP circuit 127 may also provide position and / or motion information to the DSP circuit 127, which may also be transmitted to a separate processing unit. Figure 1B FIG. 1 is an example showing a UWB platform, with one quarter used to show its overall size.

[0033] Next reference Figure 2A , an image showing an example of a mobile body monitoring system 100 including a UWB RF sensor having an array of circular antennas 109. Figure 2A As shown, the UWB RF sensor can be placed on the chest of the user with the antenna array close to the skin, directing the transmitted UWB pulses into the tissue and receiving the reflected backscatter. The placement of the UWB RF sensor can be achieved through an interface on a computing device (such as a smartphone, tablet or other mobile device). The sensor can be fixed on the chest of the user and calibration can be initiated through the interface. Feedback can be provided to the user to adjust the position of the sensor if necessary to provide sufficient coupling with the body tissue. The N-pair antenna array can be designed to have good impedance matching over the target broadband frequency. The design can optimize the phase center so that the electromagnetic (EM) transmission of each frequency band occurs at the midpoint of the transmitting antenna to ensure that all frequency bands appear to have the same tissue composition (e.g., thickness, etc.). Figure 2Bis an image of a circular array comprising 6 pairs of antennas 109, Figure 2C An example of a radiation pattern for transmitting EM waves centered on the TX antenna is shown. The TX and RX pairs of antenna 109 can be averaged to find a one-dimensional (1D) cut through the tissue. Larger linear or planar arrays can be used to make 2D and 3D images of the observed tissue (such as fat, skin, muscle, bone, lungs, etc.).

[0034] The measurement of the backscattered signal is sensitive to small variations in the hardware due to temperature and other environmental effects. To compensate for these effects, the UWB RF sensor may include a calibration channel (or loop) in communication with the DSP and wireless transmission circuitry 127. The calibration channel includes a load of known impedance located near the antenna array that is used to obtain loopback measurements that are used to calibrate variations in the transmit pulse, including its timing relative to the digital trigger. Figure 2D is an image showing the payload located on the UWB RF sensor housing which, when assembled, is adjacent to the antenna 109. The loopback signal is processed to extract an instantaneous estimate of the transmit pulse and its timing relative to the trigger, which in turn can be used to estimate tissue profile from the backscattered echoes of the transmit pulse.

[0035] Figure 2E and 2F is an image of a flexible circular array comprising six pairs of antennas 109 that can be used as an ergonomic conformal UWB RF sensor. The UWB RF sensor can be placed on the chest of a user with the antenna array secured to the skin (e.g., using an adhesive patch) to direct transmitted UWB pulses into the tissue and receive reflected backscatter. Placement of the UWB RF sensor can be accomplished, for example, through an interface on a computing device (e.g., a smartphone, tablet, or other mobile device). The sensor can be placed on the chest of the user and calibration initiated through the interface. Feedback can be provided to the user to adjust the position of the sensor if necessary to provide adequate coupling to body tissue. The N-pair antenna array can be designed to have good impedance matching over the target broadband frequency. Figure 2E and 2FThe ergonomic RF patch integrates the antenna element 109 on a flexible substrate that can be easily applied in front, such as on the right chest of a user, and removed after the measurement results are obtained. Connectors allow coupling to each antenna 109. For example, the RF electronics and digital backend can be located in a sensor cabin that connects to the flexible antenna array using a low-profile RF connector. Switching circuits (e.g., TX switching matrix 112 and RX switching matrix 115) and / or UWB pulse generators and samplers can also be integrated into the substrate, or can be provided as part of a connector assembly for coupling to the connector of the antenna 109. Digital signal processing (DSP) and wireless transmission circuitry 127 can also be integrated into the substrate. This can allow real-time, instantaneous lung water measurement and lung water fluid estimation through software (and / or firmware) implemented by a DSP or other integrated processing circuit.

[0036] To couple and focus RF energy into the body, an antenna array consisting of patch antennas balanced with a central ground plane can be used. Figure 2E and 2F As shown in the example of , six circular patches can be arranged in alternating transmit-receive pairs around a circular ground plane. While the technology does not require skin contact, implementing the antenna array in the form of a lightweight adhesive patch on a flexible substrate can support a stable and consistent placement method without the need for an external support device such as a vest or harness. RF patch antennas made of controlled dielectrics can eliminate air gaps and minimize first reflections from the skin, thereby increasing the dynamic range of the measurement.

[0037] Figure 2A and 2B Two examples of realizing antenna arrays on flexible substrates are shown, which can also integrate electronic components (or circuits) such as switching matrices and radar chipsets. Figure 2A Flexible RF laminates with ceramic cores and copper surfaces are shown. These flexible laminates can be processed using a subtractive process, with a milling machine removing the copper to form the antenna surface, signal traces, and integrated circuit (IC) pads. Standard flow soldering can be applied to integrate electronic components onto the antenna patch. This technique allows for easy integration of electronic components, allowing designs where multi-layer rigid IC boards or chips are soldered to a flexible two-layer substrate.

[0038] Figure 2BScreen printing of silver ink on polyester film to form an antenna is shown. Silver ink has excellent conductivity and allows printed circuit boards to be manufactured in an additive manner, with the conductive ink and insulating layers deposited on the polyester film. This process is low cost and allows for the formation of multilayer structures (e.g., signal traces and antenna patches), but integrating components is more difficult because the standard alloys used in soldering do not adhere to the conductive ink and the polyester substrate is heat sensitive. Therefore, a conductive epoxy adhesive is used to attach the connector and components. In addition, the conductive ink can become brittle, limiting the amount of shear stress and torsion that can be applied to the patch. However, these may not be a problem for disposable or limited use applications.

[0039] like Figure 2A and 2B As shown in the images, prototype antenna arrays were fabricated using both processes (without integrated circuits). In each case, a network analyzer was used to characterize the antenna array's ability to couple energy into the body. Both designs provided good impedance matching and gain over the wideband frequencies used by the system (e.g., 0.5GHz-3.5GHz). In various embodiments, the radar chipset can be integrated into the antenna layer via an adhesive conformal RF patch that can be attached to the front right chest. A small sensor pod can contain the digital backend, battery, and Transceiver, The transceiver connects to the RF patch using, for example, a self-guiding magnetic connector, powers the RF patch and obtains measurements. To allow patients and caregivers to assess fluid status levels in real time at home and in a point-of-care (POC) environment, lung fluid estimation technology can be implemented within a sensor device or sensor Bode plot with a wireless interface for control and storage on a mobile device. To provide real-time POC measurements without a cloud connection, a processor (e.g., ARM CortexM4F) can be built into the sensor to implement the analysis. This can simplify the data path from the sensor to a smartphone, tablet, or other mobile device and minimize the data rate and associated latency.

[0040] Figure 3A The use of TX and RX antenna pair 109 to sense tissue is shown. UWB radar pulses 303 can be sent from Figure 1A The TX switching matrix 112 of the UWB pulse 303 is coupled to the TX antenna 109t and transmits it into the body. When the UWB pulse 303 propagates through the tissue of the body, the backscatter 306 from the tissue interface is reflected back to the RX antenna 109r. Figure 1AAs can be seen from the cross-sectional image of , the human body is composed of various tissues with different dielectric properties, which affect the UWB pulse 303 and backscatter 306 as it propagates through the body. For example, the relative dielectric constant affects the propagation delay through the tissue, and the loss tangent affects the absorption of RF energy by the tissue. It can be seen that there are multiple tissue interfaces for different layers, such as skin, fat, muscle, bone, lung, etc. Figure 3B The table of provides examples of loss tangents and relative permittivities for some tissues. The backscatter reflected back to and received by the RX antenna 109r contains these overlapping returns, which can be processed to resolve the locations of the various tissue interfaces and the associated complex reflection coefficients, thereby revealing the characteristics of the tissues that make up the interfaces. The high bandwidth and narrow duration (e.g., 0.3-0.4ns) of the UWB pulses allow for higher spatial resolution than, for example, Doppler radar, and the ability to gate returns to target tissue depths.

[0041] During operation of the mobile body monitoring system 100, thousands of pulses per second (e.g., 10,000 per second) may be sent from the TX antenna 106t. Each pulse returns an echo containing several delays, indicating the depth of penetration into the body. Figure 4A As shown in (Radar Principles, N. Levanon, 1988), the backscatter response contains returns from different depths (or ranges) and can be averaged in a short time period (e.g., every 0.1 seconds) or interval (e.g., every 100 pulses). Considering that a 60-second interval contains 15-20 respiratory cycles, averaging the response in such a short time period improves the return signal-to-noise ratio without sacrificing depth information. In this way, the body monitoring system 100 provides 1-D echoes through the tissue multiple times during the respiratory cycle.

[0042] In some embodiments, the body monitoring system 100 may process the backscattered signals to generate a range profile at a rate of 100 Hz. Figure 4B As shown, each range profile, when convolved with the transmitted pulse shape, can indicate the location of the reflection boundary. The pulse signal is filtered to a frequency consistent with cardiac motion (e.g., 0.5-2 Hz) and / or lung motion (e.g., 0.1-0.3 Hz), revealing Figure 4C The structure shown.

[0043] The properties of skin, fat, muscle, lung and / or other tissues are modeled and estimated in order to estimate the dielectric constant of lung tissue which can be used to determine the lung water or fluid content. Consider a multi-layer model of tissue (e.g., skin, fat, and muscle) through which EM waves propagate, such as Figure 5A As shown, lung parameters (eg, thickness and composition) can be estimated. Figure 5BAn example of positioning of a UWB RF sensor on the chest of a user is shown, and the propagation path of the UWB pulse into the body is shown. The reflection / transmission coefficients of the lung tissue can be estimated using broadband measurements from reflected radar pulses (0.5-3.5 GHz) and an estimated multi-layer EM propagation model of the tissue between the UWB RF sensor and the lung. Since the lung tissue dielectric properties change during the breathing cycle, the lung tissue dielectric properties are estimated at three points in the breathing cycle (the bottom of exhalation, the top of inspiration, and the middle (or average) point of the breathing cycle) to provide an overall assessment of the lung tissue.

[0044] The mathematical model of the interface (such as skin, fat, muscle and / or bone) is non-parametric and can be learned from the sensor data itself without requiring a priori information about tissue thickness and order. Assuming that there are K layers (e.g., K=3 or K=4) between the UWB RF sensor and the lung tissue, the thickness and dielectric constant of each layer can be estimated assuming the average value of the loss tangent. Since these parameters can be frequency dependent, the sensor measurements can be divided into M frequency bands with a width of, for example, 500 MHz, within which the tissue properties can be assumed to be constant. The returns from multiple TX and RX antenna pairs can then be combined for each frequency band and the measurements from a calibration channel (or loop) can be used to correct for drift in the trigger delay. The calibration measurements can be used to account for hardware-induced distortion and delays, but not for the transmission interface between the antenna pairs and the body. This transfer function can be accounted for using a system model.

[0045] Figure 6 is a schematic diagram showing an example of a system model for RF imaging. The model can be expressed as:

[0046]

[0047] Among them, y i is the radar return (or backscatter) of frame i, x i is the estimated reflection (or reflectivity) profile of frame i, p(t) is the radar impulse response, Q m is the bistatic projection matrix of the mth channel, H(p) represents the convolution with the transmitted pulse p i The Toeplitz structure matrix, G m is the antenna / body transfer function.

[0048] Similarly, the reference channel response can be expressed as:

[0049] r i =H(p i ) i +n i

[0050] First, the reference channel can be inverted using a sparse deconvolution inversion algorithm (or other regularized inversion) to obtain an estimate of the transmitted pulse in the presence of temperature and other environmental factors, imposing constraints on power and band-limited frequency support, and using the l1-norm to enforce a sparse set of reflections in the reference channel (ideally a single reflection, but in practice there will be several due to imperfect connector mismatches)

[0051] Outer passband

[0052] Although sparse deconvolution inversion is used here to sharpen the reflection profile, other regularized inversion methods can also be used to achieve this result. For example, regularized inversion methods such as Tikhonov regularization, TV (total variation) norm regularization, Lp norm regularization, and machine learning-based inversion methods such as generative inversion networks or deep neural networks can be used to sharpen the range profile. Next, the estimated pulse from the reference channel It can be used to estimate a sparse set of reflectors corresponding to tissue interfaces and antenna transfer functions. The hybrid L21-norm imposes group sparsity, encoding the knowledge that tissue boundary locations are stationary relative to the range bins over short time frames, but their complex amplitudes can vary based on breathing and other internal motions.

[0053] G m It is unit power and bandwidth limited

[0054] Note that due to internal reflections, a K-layer model will typically generate a larger number of different returns than K. Tissue / fluid estimation can focus only on the first return from each tissue interface.

[0055] The solution of these two optimization problems can be achieved by alternatively minimizing multiple convex problems corresponding to various constraints and leads to the complex reflection coefficient Absolute measurements of the antenna enable wideband (in our case more than 3 GHz bandwidth) calibration for pulse distortion and for changes in the antenna transfer function due to antenna replacement and body detuning.

[0056] Fig. 7A shows an example of measured backscatter data, Figure 7B and 7C The recovered sparse reflection profile and the learned pulse shape are shown, respectively. The determination of the reflection profile can be greatly improved by using loopback measurements from a calibration circuit to adjust for distortion and delay in the transmission response.

[0057] For example, the returns from the two bands are Fig. 8A and 8BThe center frequencies are 1.25 GHz and 2 GHz respectively. The depth (or range) into the body is given on the y-axis and the time of the breathing cycle is given on the x-axis. In a one minute breathing cycle, the top, bottom and middle of the breathing cycle can identify different points in the breathing cycle and the average return.

[0058] The effect of a multi-layered tissue model (comprising, for example, skin, fat, muscle and / or bone) can then be estimated and removed from the measurement, leaving only the reflected and transmitted returns of the lung tissue. Fig. 9A shows the phases returned from all tissues, Fig. 9B The effect of removal of interfaces (eg, skin, fat, muscle, and / or bone) is shown.

[0059] Next, the reflection coefficient can be determined from the reflection profile and the lung response over the depth (or extent) corresponding to the lung contents can be aggregated to provide a measure of the lung water or fluid content. It should be noted that due to propagation delays through tissue and interface locations, the reflection coefficient is a complex value, containing both amplitude and phase information about the backscattered signal. A backscatter-based monitoring system is unique in its ability to resolve tissue based on delays, thus informing where (in which tissue) changes in fluid volume occur in addition to the amount of fluid. This is not possible with alternative systems that use through-the-lens measurements, which use transmitters and receivers placed at the back and front of the body.

[0060] refer to Fig.10 , a flow chart illustrating an example of the operation of the mobile body monitoring system 100 is shown. As described above, the body monitoring system 100 includes a UWB pulse generator 106, which generates UWB pulses for transmission into the tissue of the user 103, such as Figure 1A As shown. Beginning at 1003, backscatter data samples are collected using the TX-RX antenna pair 109 in the UWB sensor located on the user's chest. Backscatter is collected for each of the N spatial channels of the TX-RX antenna pair 109. The integrated sampling scheduler 118 can control the TX switching matrix 112 to direct the generated UWB pulses to each TX antenna and control the RX switching matrix 115 to receive the backscatter reflected by the corresponding RX antenna in each pair. At 1006, the captured backscatter data obtained by the UWB receiver 121 from the different spatial channels is collected by the DSP circuit 127 ( Figure 1A ) are delayed and summed to generate an average broadband beamformed signal, which is then divided into K frequency bands at 1009.

[0061] After completing the TX-RX cycle through each antenna pair 109 at 1003, a UWB pulse may be directed from the UWB pulse generator 106 through a calibration circuit (or loop) to obtain a loopback measurement that may be used to account for distortion and delay generated by hardware and temperature effects at 1012. The measured calibration signal is then divided into K frequency bands at 1015. Using the frequency band information of the measured calibration signal, a computing device (or DSP circuit) may determine an instantaneous UWB pulse that is distorted and delayed by the circuit hardware at 1018.

[0062] At 1021, a computing device (or DSP circuit) may use the instantaneous UWB pulse to guide the determination of a sparse set of reflectors and corresponding reflection coefficients of the organizational layer of the beamforming channel signal. As described above, sparse deconvolution may be used to identify the UWB pulse shape and the reflection profile of the K-band. The reflection profiles of the bands may be combined to determine an average reflection profile. The use of the instantaneous UWB pulse determined at 1018 compensates for temperature effects on the UWB RF sensor during operation, which improves the accuracy and consistency of the determined reflection profile. The reflection coefficient may be extracted from the reflection profile.

[0063] like Fig.10 As shown, for each group of N spatial channels of the TX-RX antenna pair 109, the process (1003 to 1021) is repeated multiple times within a defined time period. For example, backscattered data samples (and corresponding measurements by calibration circuits) can be collected for P series of transmission pulses using the TX-RX antenna pair 109. Data can be collected for a predetermined number of data sets or within a predetermined time period. Using the reflection profile and reflection coefficient determined for the P groups of backscattered data, at 1024, the computing device can track the lung position and characteristics in the respiratory cycle based on the determined information. For example, the change in the depth of the lung tissue interface produced by inhalation and exhalation and the lung tissue characteristics can be determined over a period of time. The results that have been identified as being at the top of inspiration or the bottom of exhalation can be averaged to provide a better measure of lung tissue characteristics. In addition, the results of the intermediate (or average) points in the respiratory cycle can be determined and averaged to provide a common point for evaluating lung tissue.

[0064] At 1027, the reflectance coefficient can be converted into an estimate of the fluid level in the lung tissue. By using average data from the bottom, top, and middle of the breathing cycle, the accuracy of tissue location and characteristics can be improved. In addition to lung tissue, information about surrounding tissue (e.g., skin, fat, muscle, bone, and heart) can also be determined from the reflectance coefficient. In some cases, correlations between different tissues can be analyzed and evaluated. This information can be converted to be displayed in real time (or near real time) by a computing device.

[0065] It will be appreciated that the processing of the backscatter data may be performed by the DSP circuit 127 ( Figure 1A ) and a computing device. For example, the backscatter data and calibration measurements can be processed by the DSP circuit to provide frequency band information (1003 to 1015), which can then be transmitted to the computing device for subsequent processing and determination of organizational information (1018 to 1027). In other embodiments, the DSP circuit 127 can be used to perform additional processing before transmission to the computing device.

[0066] Reference now Fig.11 , an example of a computing device 1103 that may be included in the mobile body monitoring system 100 is shown. The computing device 1103 may be one or more computing devices 1103 that include at least one processor circuit, for example, having a processor 1109 and a memory 1112, both of which are coupled to a local interface 1115. To this end, the computing device 1103 may include, for example, a computer, a laptop, a smart phone, a tablet computer, or other mobile processing units that provide computing capabilities. The computing device 1103 may include, for example, one or more display devices, such as a cathode ray tube (CRT), a liquid crystal display (LCD), a flat panel display based on gas plasma, an LCD projector, or other types of display devices. The computing device 1103 may also include, for example, various peripheral devices. Specifically, the peripheral device may include an input device, such as a keyboard, a keypad, a touch pad, a touch screen, a microphone, a scanner, a mouse, a joystick, or one or more buttons. Although the computing device 1103 is singular, it should be understood that multiple computing devices 1103 can be used in various arrangements as described above. It is understood that the local interface 1115 may include, for example, a data bus with an accompanying address / control bus or other bus structure.

[0067] Stored in the memory 1112 are data and several components that can be executed by the processor 1109. In particular, stored in the memory 1112 and executable by the processor 1109 are body monitoring applications 1118 and potential other applications. Data storage 1121 and other data can also be stored in the memory 1112. For example, the data stored in the data storage 1121 is associated with the operation of various applications and / or functional entities described below. For example, the data storage can contain data samples, reflection profiles, and other data or information that can be understood. In addition, an operating system 1124 can be stored in the memory 1112 and can be executed by the processor 1109. The data storage 1121 can be located in a single computing device, or can be dispersed among many different devices.

[0068] The body monitoring system 100 may be communicatively coupled to the computing device 1103 via a wireless communication link or a network. In some embodiments, the body monitoring system 100 may be directly connected to the computing device 1103.

[0069] Components executed on the computing device 1103 include, for example, a body monitoring application 1118 and other systems, applications, services, processes, engines, or functions not discussed in detail herein. It is understood that there may be other applications stored in the memory 1112 and executable by the processor 1109. Where any of the components discussed herein are implemented in software, any of a variety of programming languages ​​may be used, such as, for example, C, C++, C#, Objective C, Java, JavaScript, Perl, PHP, Visual Basic, Python, Ruby, Delphi, Flash, or other programming languages.

[0070] Many software components are stored in the memory 1112 and are executable by the processor 1109. In this regard, the term "executable" refers to a program file in a form that can ultimately be run by the processor 1109. Examples of executable programs may be, for example, a compiled program that can be translated into machine code in a format that can be loaded into the random access portion of the memory 1112 and run by the processor 1109, source code that can be expressed in a suitable format, such as object code that can be loaded into the random access portion of the memory 1112 and executed by the processor 1109, or source code that can be interpreted by another executable program to generate instructions in the random access portion of the memory 1112 to be executed by the processor 1109, etc. The executable program may be stored in any portion or component of the memory 1112, including, for example, random access memory (RAM), read-only memory (ROM), a hard drive, a solid-state drive, a USB flash drive, a memory card, an optical disk such as a compact disk (CD) or a digital versatile disk (DVD), a floppy disk, a magnetic tape, or other storage component.

[0071] Memory 1112 is defined herein as including volatile and non-volatile memory and data storage components. Volatile components are components that do not retain data values ​​when power is off. Non-volatile components are components that retain data when power is off. Therefore, memory 1112 may include, for example, random access memory (RAM), read-only memory (ROM), hard disk drive, solid-state drive, USB flash drive, memory card accessed via a memory card reader, floppy disk accessed via an associated floppy disk drive, optical disk accessed via an optical drive, magnetic tape accessed via an appropriate tape drive, and / or other memory components, or any combination of two or more of these memory components. In addition, RAM may include, for example, static random access memory (SRAM), dynamic random access memory (DRAM) or magnetic random access memory (MRAM) and other such devices. ROM may include, for example, programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM) or other similar storage devices.

[0072] In addition, the processor 1109 may represent multiple processors 1109, and the memory 1112 may represent multiple memories 1112 respectively operating in parallel processing circuits. In this case, the local interface 1115 may be an appropriate network that facilitates communication between any two of the multiple processors 1109, between any processor 1109 and any memory 1112, or between any two memories 1112. The local interface 1115 may include additional systems designed to coordinate the communication, including, for example, performing load balancing. The processor 1109 may be electrical or some other available structure.

[0073] Although the body monitoring application 1118 and the various other systems described herein may be implemented in software or code executed by general purpose hardware as described above, alternatively, they may also be implemented in dedicated hardware or a combination of software / general purpose hardware and dedicated hardware. If included in dedicated hardware, each may be implemented as a circuit or state machine using any one or combination of a variety of technologies. These technologies may include, but are not limited to, discrete logic circuits with logic gates for implementing various logic functions when one or more data signals are applied, application specific integrated circuits with appropriate logic gates, or other components, etc. These technologies are generally well known to those skilled in the art and are therefore not described in detail herein.

[0074] Fig.10The flowchart illustrates the functions and operations of the implementation of the portion of the body monitoring application 1118. If embodied in software, each block may represent a module, code segment, or code portion that includes program instructions that implement a specified logical function. The program instructions may be implemented in the form of source code, which includes human-readable statements written in a programming language, or machine code, which includes digital instructions that can be recognized by a suitable execution system (such as a processor 1109 in a computer system or other system). The machine code may be converted from the source code, etc. If implemented in hardware, each block may represent a circuit or multiple interconnected circuits to implement a specific logical function.

[0075] although Fig.10 The flowchart of the embodiment shows a specific order of execution, but it should be understood that the order of execution may be different from the order described. For example, the order of execution of two or more blocks may be scrambled relative to the order shown. In addition, Fig.10 3 and / or 6 can be performed simultaneously or partially simultaneously. In addition, in some embodiments, one or more frames shown in Fig. 3 and / or 6 can be skipped or omitted. In addition, for the purpose of enhancing utility, accounting, performance measurement or providing troubleshooting help, any number of counters, state variables, warning signals or messages can be added to the logic flow described herein. It should be understood that all these changes are within the scope of the present disclosure.

[0076] In addition, any logic or application including software or code described herein including the body monitoring application 1118 may be included in any non-temporary computer-readable medium for use by an instruction execution system (such as, for example, a processor 1109 in a computer system or other system) or in combination with an instruction execution system. In this sense, logic may include, for example, statements including instructions and declarations, which may be taken out of a computer-readable medium and executed by an instruction execution system. In the context of the present disclosure, a "computer-readable medium" may be any medium capable of containing, storing, or maintaining logic or applications described herein that are used by or in combination with an instruction execution system. A computer-readable medium may include any of a number of physical media, such as, for example, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor media. More specific examples of suitable computer-readable media include, but are not limited to, tapes, magnetic floppy disks, magnetic hard drives, memory cards, solid-state drives, USB flash drives, or optical disks. In addition, a computer-readable medium may be a random access memory (RAM), including, for example, a static random access memory (SRAM) and a dynamic random access memory (DRAM), or a magnetic random access memory (MRAM). Furthermore, the computer readable medium may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or other types of storage devices.

[0077] A pilot study of the mobile body monitoring system 100 was conducted on patients with a primary diagnosis of acute decompensated heart failure. The patients were evaluated by body monitoring technology to correlate pleural fluid measurements with the clinical picture of congestive heart failure. The mobile body monitoring system 100 was able to provide the patient with a personalized measurement that could be used to help determine how close the patient was to a "dry" state. Fluid levels obtained by the body monitoring system 100 were compared to the total net fluid volume loss during the hospital stay. The patients were evaluated daily to correlate pleural fluid measurements with the clinical picture of congestive heart failure. Fig. 12A The raw sensor readings captured as an image are shown, with the depth of the echo-generating interface shown along one axis and time (T) along the second axis; the motion of the tissue interface toward and away from the transducer (M) (similar to the TM mode of ultrasound). Note that the motion of the lung tissue caused by breathing is visible, helping to identify relevant tissue transformations. Fig. 12B is a graph showing net fluid volume loss (in liters) compared to the standardized pleural fluid metric provided by the mobile body monitoring system 100. Data was collected from study subjects over a period of eight days.

[0078] Fig.13A graph showing an example of the average phase of the reflectance over the depth (or extent) of the body. Curve 1303 being above zero indicates dryness of the lungs, especially in both compartments. A controlled repeat experiment was also performed using a rotating disk. Fig.14A and 14B The raw data of backscatter amplitude and phase are shown separately. The amplitude shows the hierarchical structure of the tissue, while the phase shows the temporal changes caused by breathing. The hierarchical structure amplitude and phase of the average hyperpnea recovery are shown in Fig. 14C and 14D As can be seen, both the amplitude and phase show the changes in fluid levels deep within the body. The peaks and troughs of the breathing cycle can also be recovered.

[0079] To test the validity of the measurements provided by the mobile body monitoring system, a multilayer phantom consisting of three tissue layers (skin, bone, and muscle) was created and placed on a foam layer of known dielectric constant. The dielectric constant (dielectric constant and conductivity) of the simulated tissue layers were adjusted using polyethylene powder (PEP) and sodium chloride, respectively. Agar was used to self-form the mixture into a solid layer, and TX-151 powder was used to increase the viscosity of the mixture. The dielectric constant of the simulated tissue was verified using an Agilent 85070E dielectric probe kit. The measured dielectric constant of the simulated tissue was compared with the reference values, and the measured conductivity and dielectric constant were found to be consistent with the reported values ​​for these tissue types. Fig.15 Depth vs. reflection amplitude measurements are shown (similar to ultrasound A-mode). The observed delay and reflection amplitude can be used to estimate the dielectric properties of multi-layer tissue sections. For three layers (skin, bone, muscle), the estimated dielectric constants are (41, 12, 55) with an average error of 4.5%.

[0080] It should be emphasized that the above-mentioned embodiments of the present disclosure are merely possible examples of implementations set forth for a clear understanding of the principles of the present disclosure. Many changes and modifications may be made to the above-mentioned embodiments without departing substantially from the spirit and principles of the present disclosure. All of these modifications and variations are intended to be included within the scope of the present disclosure and are protected by the appended claims.

[0081] The term "substantially" means that deviations from the descriptive term are permissible without adversely affecting the intended purpose. A descriptive term is implicitly understood to be modified by the word substantially, even if the term is not explicitly modified by the word substantially.

[0082] It should be noted that ratios, concentrations, amounts and other numerical data can be expressed in range format herein. It should be understood that such range format is used for convenience and brevity, and therefore, should be interpreted in a flexible manner as not only including the numerical values ​​explicitly listed as limits of the range, but also including all single numerical values ​​or sub-ranges included in the range, as if each numerical value and sub-range were explicitly listed. For illustration purposes, the concentration range of "about 0.1% to about 5%" should be interpreted as not only including the concentration of about 0.1wt% to about 5wt% explicitly listed, but also including single concentrations (e.g., 1%, 2%, 3% and 4%) and sub-ranges (e.g., 0.5%, 1.1%, 2.2%, 3.3% and 4.4%) within the indicated range. The term "about" may include traditional rounding according to the significant figures of the numerical value. In addition, the term "about 'x' to 'y'" includes "'x' to 'y'".

Claims

1. A method for determining a physical characteristic, comprising: collecting a reflection backscatter dataset of a UWB pulse sequence transmitted by an ultra-wideband (UWB) sensor, and corresponding calibration measurements from a calibration channel in the UWB sensor, the UWB sensor comprising an array of transmit TX and receive RX antenna pairs located on a user's body; determining a reflection coefficient for each tissue interface based on the reflection backscatter data set, the reflection coefficient being determined from a reflection profile associated with a model of tissue layers within the body between the UWB sensor and lung tissue based on the reflection backscatter data for the UWB pulse sequence and the corresponding calibration measurements; and determining a fluid level content of the lung tissue based on the reflectance, wherein the reflected backscatter data set includes reflected backscatter data obtained for each of the TX and RX antenna pairs in the UWB sensor, the reflected backscatter data being combined to generate a broadband beamforming signal for each set of reflected backscatter data, and wherein the reflection profile is determined based on sparse deconvolution of the wideband beamformed signal of the reflected backscatter data set using a compensated UWB pulse shape based on the corresponding calibration measurement, Wherein, the compensated UWB pulse shape compensates for temperature changes.

2. The method according to claim 1, wherein: The sparse deconvolution of the wideband beamformed signal is performed for each of K frequency bands.

3. The method according to any one of claims 1 or 2, comprising: identifying depths of a lung tissue interface at top, middle, and bottom points in a respiratory cycle of the lung tissue based on the reflectance; as well as Wherein determining the fluid level content of the lung tissue includes determining the fluid level content of the top, middle and bottom points in the breathing cycle.

4. The method of any of claims 1 or 2, comprising determining characteristics of a tissue layer located between the UWB sensor and the lung tissue.

5. The method according to claim 4, wherein: The characteristics of the tissue layers include the location of at least one tissue layer interface or a dielectric property of at least one tissue layer.

6. A mobile body monitoring system comprising: An ultra-wideband (UWB) sensor comprising an antenna array and a calibration channel, wherein the antenna array comprises a transmit TX and a receive RX antenna pair, and the UWB sensor is configured to be positioned on a user's body; a radio frequency (RF) front end comprising a UWB pulse generator coupled to the TX antenna of the antenna array and a UWB receiver coupled to the RX antenna of the antenna array, wherein UWB pulses generated by the UWB pulse generator are sequentially transmitted into the user's body through the TX antenna and reflected backscattered signals are received by the RX antenna of the TX and RX antenna pair; a wireless transmitter configured to transmit data associated with the reflected backscatter and corresponding calibration measurements from the calibration channel; and a computing device configured to receive the data and determine a physical characteristic of the user based on the reflected backscatter and corresponding calibration measurements; Wherein, the computing device is configured as: determining a reflection profile based on the data associated with the reflected backscatter and the corresponding calibration measurements of the transmitted UWB pulse sequence, the reflection profile associated with a model of a tissue layer within the body between the UWB sensor and target tissue; determining a reflection coefficient based on the reflection profile; and determining a characteristic of the target tissue according to the generated target tissue data; wherein, based on data associated with the reflected backscatter of the UWB pulse sequence, Determining the reflection profile by sparse deconvolution of the average broadband backscattered signal; and wherein the reflection profile is determined based on using a compensated UWB pulse shape based on the corresponding calibration measurement, wherein the compensated UWB pulse shape compensates for temperature variations.

7. The mobile body monitoring system according to claim 6, wherein: The characteristic of the target tissue includes a depth of an interface with the target tissue or a dielectric property of the target tissue.

8. The mobile body monitoring system according to claim 7, wherein: The target tissue is lung tissue.

9. The mobile body monitoring system according to claim 8, wherein: The computing device is configured to identify a measure of lung fluid content based on the characteristic of the lung tissue.

10. The mobile body monitoring system according to claim 9, wherein: The computing device is configured to simultaneously identify one or more of heart rate, heart rate variability, respiratory rate, or tidal volume.

11. The mobile body monitoring system according to claim 8, wherein: The computing device is configured to identify top and bottom depths of a lung tissue interface during a breathing cycle of the lung tissue.

12. The mobile body monitoring system according to claim 11, wherein: The computing device is configured to identify dielectric properties at the top and bottom depths and an average depth in the breathing cycle of the lung tissue.

13. The mobile body monitoring system according to claim 6, wherein: The computing device is configured to determine a reflection profile for each of a series of reflection backscatter data sets, each of the reflection backscatter data sets comprising data associated with the reflection backscatter of the transmitted UWB pulse sequence associated with the set.

14. The mobile body monitoring system according to claim 6, wherein: The calibration channel includes a temperature calibration loop having a load of known impedance located proximate to the antenna array.

15. The mobile body monitoring system of claim 14, wherein: Variations in the transmitted UWB pulses are compensated based on the corresponding calibration measurements.

16. The mobile body monitoring system of any one of claims 6-15, comprising digital signal processing circuitry configured to obtain and process the reflected backscatter signal and the corresponding calibration measurement for transmission to the computing device.

17. The mobile body monitoring system according to any one of claims 6 to 15, wherein: The UWB pulses are transmitted into the body at a rate of approximately 10,000 per second.

Images