Body-worn sensor with haptic interface

Abstract

A sensor fully fitted to the patient's body comprises a microphone sensor, a haptic interface, and a processing system. The microphone sensor measures acoustic signals generated by the patient's body. The haptic interface is configured to generate haptic responses. The processing system is programmed to receive acoustic signals from the microphone sensor, process the measured acoustic signals to detect swallowing, breathing, coughing, or a combination thereof, and execute computer code that controls the haptic interface to generate haptic responses in response to the processing of the acoustic signals.

Description

Technical Field

[0001] Cross - reference to Related Applications Claims the priority and benefit of U.S. Patent Application No. 63 / 494,108, filed on April 4, 2023, the entire content of which is hereby incorporated by reference in its entirety.

[0002] The present disclosure generally relates to systems for monitoring and providing care for, for example, patients with Parkinson's disease both in a hospital and at home.

Background Art

[0003] This section of the specification may provide background related to the technical field and / or introduce information from the technical field relevant to the subject matter described in this specification and / or the claims hereinafter. It also provides background information for better understanding various aspects of the present disclosure. This is a discussion of the "related" technical field. That such a technical field is related in no way implies that it is also "prior" art. Related technologies may or may not be prior art. The description in this section of the specification should be read from this perspective and should not be construed as an admission of prior art.

[0004] Swallowing is a major function essential for human survival and is a highly complex coordinated action of rapid and interdependent movements induced through sensory end organs in the oral cavity, pharynx, and larynx. This action not only protects against aspiration into the airway but also plays a role in propelling the ingested substance (hereinafter, "swallowing bolus") throughout the upper gastrointestinal tract (see, for example, Logemann et al., Evaluation and treatment. Folia Phoniatr Logop. 1995;47(3):140 - 64). The importance of swallowing function is further amplified by the millions of adults suffering from swallowing disorders (hereinafter, "dysphagia") related to neurological diseases, head and neck cancers, and digestive and respiratory diseases.

[0005] Diseases and conditions that impair the physiological mechanisms controlling swallowing or directly damage the peripheral structures involved in swallowing can cause dysphagia ranging from mild to significant problems affecting daily life, posing a major rehabilitation challenge (see, e.g., Shaw et al., The normal SWallow: muscular and neurophysiological control. Otolaryngol Clin North Am. 2013;46(6):937-56). Dysphagia may be accompanied by sensory changes that delay the initiation of swallowing, decreased muscle strength, and reduced coordination of range and timing of movement (see, e.g., Martin-HaRRis et al., Breathing and Swallowing dynamics across the adult lifespan. Arch Otolaryngol Head Neck Surg. 2005;131(9):762-70). In addition to the complexity of controlling and executing swallowing, breathing needs to be closely timed with the initiation of swallowing. The reason is that the pharyngeal cavity, common to both functions, alternates between an open state during breathing and a tightly compressed state during swallowing, and generates high positive pressure in the food bolus necessary for it to pass through the upper respiratory tract and digestive tract.

[0006] Patients with diseases affecting swallowing include those suffering from Parkinson's disease, head and neck cancer, Alzheimer's disease and other dementias, myasthenia gravis, amyotrophic lateral sclerosis (ALS), and related disorders. Furthermore, these diseases not only affect swallowing function but also impact patients' speech function (for example, by causing unclear pronunciation or speaking in a voice too quiet for others to hear) and can lead to symptoms such as drooling.

[0007] The coordination of respiration and swallowing is crucial for the safe and efficient transport of food and liquids from the mouth to the esophagus. In healthy adults, the timing of swallowing initiation usually coincides with a pause in the expiratory phase at medium to low lung volume during resting respiration. This coordination pattern is 1) essential for airway protection, 2) promotes physiological phenomena beneficial to the safety and efficiency of swallowing, such as retraction of the tongue base, elevation of the larynx, and opening of the pharyngeal-esophageal segments, and 3) consequently aids in the removal of the food bolus (see, e.g., Hopkins-Rossabi et al. Respiratory-swallow coordination and swallowing impairment in head and neck cancer. Head Neck-J Sci Spec. 2021; 43(5):1398-408). However, in patients with dysphagia, the coordination of respiration and swallowing is significantly impaired, leading to impaired swallowing mechanisms and a significant decline in health and quality of life. Conventional swallowing interventions typically employ a single-system approach that focuses on increasing the strength and range of motion of pharyngeal structures.

[0008] Despite the proven success of these approaches, most patients continue to live with dysphagia. Recently, a novel approach that trains patients with dysphagia to initiate swallowing during the exhalation phase of the respiratory cycle has been shown to reduce aspiration and improve swallowing biomechanics in some patient populations (see, for example, Martin-Harris et al., Respiratory-Swallow Training in Patients With Head and Neck Cancer. Arch Phys Med Rehab. 2015;96(5):885-93). However, swallowing rehabilitation faces three important methodological challenges: 1) the ability to clearly detect the occurrence of swallowing and respiration-coordinated swallowing without the use of expensive and portable imaging and respiratory recording devices; 2) the ability to ensure the accuracy of swallowing interventions and provide real-time visual cues for performance improvement; and 3) the ability to promote the stability of swallowing skills acquired through outpatient monitoring and cue presentation.

[0009] Currently, devices for measuring swallowing are primarily used in research settings. The current gold standard techniques in such devices include the following two technologies: 1) respiratory-inducted plethysmography (hereinafter, "RIP"), which measures the overall expansion of the patient's thoracic and abdominal cavity via an induction coil, and 2) nasal cannula, which monitors the pressure difference within the nasal cavity as an absolute measurement of nasal airflow. Devices that synchronize and capture nasal airflow with thoracic and abdominal movements can accurately evaluate the phase patterning of respiration and swallowing. [Overview of the project] [Problems that the invention aims to solve]

[0010] Considering the above, it is beneficial to have a body-worn sensor optimized for monitoring swallowing events (hereinafter "SW") and their timing relative to the patient's respiratory phase (i.e., inspiration and expiration). Even more beneficial is a sensor that stimulates the patient, for example, with a vibration response (hereinafter "haptic interface") that warns of SWs occurring at an inappropriate time, with the ultimate goal of training the patient to swallow at the appropriate time. Furthermore, patients with dysphagia naturally have a reduced number of swallowing events. Therefore, another beneficial application of the technology disclosed herein includes increasing the number of swallowing events in the patient through a haptic interface, such as a vibration reminder. Yet another benefit is that by activating the haptic interface at the appropriate time, the amount of saliva produced by the patient and the volume of their vocalizations can be increased. [Means for solving the problem]

[0011] In a first embodiment, a sensor fully fitted to a patient's body comprises a microphone sensor, a haptic interface, and a processing system. The microphone sensor measures acoustic signals generated by the patient's body. The haptic interface is configured to generate haptic responses. The processing system is programmed to receive acoustic signals from the microphone sensor, process the measured acoustic signals to detect swallowing, breathing, coughing, or a combination thereof, and execute computer code to control the haptic interface to generate haptic responses in accordance with the processing of the measured acoustic signals.

[0012] In a second embodiment, a method for assisting a patient's swallowing comprises measuring acoustic signals generated by the patient's body; processing the measured acoustic signals to detect swallowing, breathing, coughing, or a combination thereof; and controlling a haptic interface to generate a haptic response in response to the processing of the measured acoustic signals.

[0013] In a third embodiment, the wearable sensor comprises a first physiological sensor, a haptic interface, and a processing system. The first physiological sensor is configured to acquire a first set of physiological signals that may indicate a patient swallowing event related to the patient's respiratory stage. The haptic interface is configured to generate a haptic response. The processing system is programmed to receive motion signals from the motion sensor, process the first set of physiological signals to identify a first signal related to the presence or absence of a swallowing event and a second signal related to the patient's respiration, and generate a haptic response by collectively processing the first signal, the second signal, and a preset parameter indicating the patient's swallowing opportunity.

[0014] In a fourth embodiment, the computer execution method comprises: receiving a first set of physiological signals from a first physiological sensor located in the patient's suprasternal fossa that may indicate a patient swallowing event related to the patient's respiratory phase; processing the first set of physiological signals to identify a first signal related to the presence or absence of a swallowing event and a second signal related to the patient's respiration; processing the first signal, the second signal and a preset parameter indicating the patient's swallowing opportunity; and generating a haptic response via a haptic interface to prompt the patient to swallow.

[0015] In a fifth embodiment, the system comprises a wearable sensor, an external gateway, and a computer system. The wearable sensor is configured to acquire a first set of physiological signals from the patient that may indicate a patient swallowing event related to the patient's respiratory stage, process the first set of physiological signals to identify a first signal related to the presence or absence of a swallowing event and a second signal related to the patient's respiration, and generate a haptic response indicating an opportunity for the patient to swallow. The external gateway is programmed to wirelessly receive the information transmitted from the wearable sensor. The computer system may be capable of storing or accessing the transmitted information. comprises.

[0016] The above provides a simplified overview of the subject matter of the claims in order to provide a basic understanding of some aspects of the invention. This overview is not an exhaustive overview of the subject matter of the claims. Nor is the overview intended to identify key or decisive elements of the invention or to limit the scope of protection. Its sole purpose is to present some concepts in a simplified form prior to the more detailed description that follows.

Brief Description of the Drawings

[0017] In the drawings, like reference numerals generally denote like, functionally similar, and / or structurally similar elements.

[0018] [Figure 1] FIG. 1 is a photograph of a patient wearing a body-worn sensor according to one or more embodiments comprising an accelerometer for motion detection and a haptic interface.

[0019] [Figure 2A-2B] FIGS. 2A and 2B are an assembled view and an exploded mechanical view of the body-worn sensor of FIG. 1.

[0020] [Figure 2C] FIG. 2C is a schematic block diagram of the electronic circuit of the body-worn sensor of FIG. 1. [[ID=3%1]]

[0021] [Figure 3A] FIG. 3A is a graph of time-dependent motion signals measured along the x, y, and z axes of the accelerometer within the body-worn sensor of FIG. 1.

[0022] [Figure 3B] FIG. 3B is a graph of time-dependent energy signals calculated from the time-dependent motion signals of FIG. 3A.

[0023] [Figure 4A]Figure 4A is a graph of the processed acceleration signals from the x-axis, y-axis, and z-axis of the accelerometer and the respiration signal derived from the z-axis acceleration signal.

[0024] [Figure 4B] Figure 4B is a diagram showing the signal processing steps and the convolutional neural network (CNN) architecture used to detect SW.

[0025] [Figure 5] Figure 5 is a graph showing the theoretical time-dependent tidal volume measured from a patient, with the time intervals indicating "safe swallowing intervals" overlaid.

[0026] [Figure 6A] Figure 6A is a graph of the respiration signal generated from the time-dependent motion signal measured along the z-axis by the accelerometer in the body-worn sensor of FIG. 1.

[0027] [Figure 6B] Figure 6B is a graph of the time-dependent motion signal used in FIG. 6A, with the respiratory component of the signal removed to show SW.

[0028] [Figure 7] Figure 7 is a flowchart of an algorithm used to detect SW and generate haptic feedback to a patient based on the timing of SW accordingly.

[0029] [Figure 8] Figure 8 is a schematic diagram of a patient wearing the body-worn device of FIG. 1, triggered by an external device to measure SW and generate a haptic response.

[0030] [Figure 9A] Figure 9A is a graph of the time-dependent motion signal measured along the z-axis by the accelerometer in the body-worn sensor of FIG. 1.

[0031] [Figure 9B] Figure 9B is a graph of the time-dependent respiratory signal identified by filtering the time-dependent motor signal in Figure 9A.

[0032] [Figure 10A] Figure 10A is a graph of the time-dependent motion signal measured along the z-axis by the accelerometer in the body-worn sensor shown in Figure 1.

[0033] [Figure 10B] Figure 10B is a graph of time-dependent cardiac signals identified by filtering the time-dependent motion signals in Figure 10A.

[0034] [Figure 11A] Figure 11A is a graph of time-dependent motor signals with respiratory signals and SW occurring in four different "cases," each corresponding to a different phase between the respiratory signal and SW.

[0035] [Figure 11B] Figure 11B is a graph of the time-dependent haptic signal generated by the SW shown in Figure 11A.

[0036] [Figure 12] Figure 12 is a graph showing the average number of swallows per minute measured from a cohort of Parkinson's disease patients with and without haptic feedback from the body-worn sensors shown in Figure 1.

[0037] [Figure 13] Figure 13 shows another configuration example of the sensor in Figure 1, comprising a first part including a sensing element for measuring heart rate (HR) and respiratory rate (RR), a third part including an accelerometer for measuring heart rate variability (SW), and a second part connecting the first and third parts.

[0038] [Figures 13A-13B]Figure 13A shows a photograph of the sensor in Figure 13 in which the first part is positioned in the patient's suprasternal fossa and the third part is positioned in the patient's larynx, and Figure 13B shows a photograph of the sensor in Figure 13 in which the first part is positioned in the patient's chest and the third part is positioned in the patient's SN.

[0039] [Figure 14] Figure 14 shows photographs of patients with a patch-type sensor and pulse oximeter sensor, each attached to the chest and fingertip, respectively, which have physiological sensors and haptic interfaces.

[0040] [Figure 15A-15C] Figures 15A-C show the wearable sensors from Figure 1 attached to the patient's SN, hand, and arm, respectively.

[0041] While various modifications and alternative forms are possible for the disclosed subject matter, the drawings illustrate specific embodiments described in detail. However, it should be understood that the description of specific embodiments herein is not intended to limit the scope to the specific forms disclosed in the claims, but rather to encompass all modifications, equivalents, and alternatives that fall within the spirit and scope of the appended claims. [Modes for carrying out the invention]

[0042] The following discloses exemplary embodiments of the subject matter of the claims of this specification. For clarity, not all features of the actual embodiments are described in each embodiment. It is understood that in developing such actual embodiments, numerous embodiment-specific decisions may be made to achieve the developer's specific goals, such as compliance with system-related and business-related constraints that differ from embodiment to embodiment. Furthermore, it is understood that such development work, even if complex and time-consuming, is a routine undertaking for an ordinary technician of the art who would benefit from this disclosure.

[0043] A wearable sensor according to one embodiment monitors the time-dependent phase of a patient's swallowing response (SW) in relation to exhalation, and the sensor has a haptic interface that alerts the patient to the ideal time for SW. Such sensors address the rehabilitation needs of millions of adults suffering from dysphagia associated with neurological disorders, head and neck cancers, and gastrointestinal and respiratory diseases. In some embodiments, the body-worn sensor simultaneously measures SW and upper chest movement of the patient in relation to respiratory movements (e.g., breathing) when worn near the throat. For most patients, the ideal position for the body-worn sensor is the suprasternal fossa (SN), and second best, the manubrium of the sternum (SM). In patients that only induce small amplitude movements of the throat, the sensor can be positioned above the SN, including a position coinciding with the laryngeal prominence (LP), to increase the amplitude of the signal. The sensor also measures vital signs from the patient, such as heart rate (HR) and respiratory rate (RR), along with signals related to the patient's physical activity, posture, and position.

[0044] Based on the above, in one embodiment, the body-worn sensor is fully attached to the patient's body and has no functional components that are not attached to the body. The body-worn sensor comprises: 1) a motion sensor that measures time-dependent motion signals modulated by the patient's swallowing and respiration; 2) a haptic interface that generates a haptic response; and 3) a processing system programmed to receive motion signals from the motion sensor and to process a first set of physiological signals to identify a first signal related to the presence or absence of a swallowing event and a second signal related to the patient's respiration; and ii) a haptic response generated by collectively processing the first signal, the second signal, and preset parameters indicating the patient's opportunity to swallow.

[0045] In some embodiments, the body-worn sensor is fully attached to the patient's SN. In other embodiments, the second signal is a time-dependent signal indicating the patient's inspiration and expiration. Here, a preset parameter indicates the period during which the patient should exhale, and the processing system generates a haptic response when it determines that no SW occurs during the patient's exhalation.

[0046] In related embodiments, a preset parameter indicates the period during which the patient should exhale, and the processing system generates a haptic response when it determines that SW occurs during the patient's inspiration. Alternatively, a preset parameter indicates the period during which the patient should exhale, and the processing system generates a haptic response when it determines that SW occurs between the patient's inspiration and their exhalation.

[0047] In other embodiments, the second signal is a time-dependent signal indicating the patient's single breath volume. Here, for example, a preset parameter indicates a period during which the patient's single breath volume is approximately 25% of its maximum value, and the processing system generates a haptic response when it determines that SW occurs when the single breath volume exceeds 25% of its maximum value.

[0048] In other embodiments, the motion sensor is an accelerometer. The accelerometer is typically configured to measure time-dependent motion signals along the x, y, and z axes corresponding to the patient, and the processing system is programmed to process at least one time-dependent positive motion signal with a first algorithm to identify a time-dependent "energy" signal. In related embodiments, the first algorithm is programmed to collectively process the time-dependent motion signals measured by the accelerometer along the x, y, and z axes to identify a time-dependent energy signal, and the time-dependent energy signal is expressed, for example, by the following formula

number

[0049] Here, E(t) is the time-dependent energy, x(t) is the time-dependent motion signal measured by an accelerometer along the x-axis corresponding to the patient, y(t) is the time-dependent motion signal measured by an accelerometer along the y-axis corresponding to the patient, and z(t) is the time-dependent motion signal measured by an accelerometer along the z-axis corresponding to the patient.

[0050] In other embodiments, the processing system is further programmed to process time-dependent energy signals, using a second algorithm programmed to count features of the time-dependent energy signal, each feature corresponding to a swallowing event. For example, the second algorithm identifies peaks in time-dependent energy signals induced by SWs associated with a patient. More specifically, the second algorithm can identify features in the peak or a mathematical derivative of the peak, the mathematical derivative of the peak including at least one of i) the maximum value of the peak, ii) the bottom of the peak, iii) the width of the peak, iv) the inflection point where the feature changes from a positive to a negative value, and v) the area under the peak.

[0051] In another embodiment, the processing system is programmed to process time-dependent and frequency-dependent signals by a third algorithm programmed to extract features from time-dependent and frequency-dependent signals corresponding to SW. The third algorithm comprehensively uses x-axis acceleration signals, y-axis acceleration signals, and z-axis acceleration signals, as well as a breathing signal estimated from the z-axis acceleration, to capture relevant characteristics associated with SW. More specifically, the third algorithm computes features by convolving the time-dependent and frequency-dependent signals from the x-axis acceleration, y-axis acceleration, and z-axis acceleration.

[0052] In other embodiments, the haptic interface may be a vibration motor, a mechanical buzzer, a device that supplies current to the patient, a light-emitting diode, a piezoelectric device, and / or a device that generates acoustic sound.

[0053] In other embodiments, some embodiments provide a body-worn sensor comprising: 1) a motion sensor (e.g., an accelerometer or gyroscope) that measures a time-dependent motion signal modulated by the patient's swallowing; 2) a respiratory sensor having a pair of electrodes that measure a time-dependent electrical signal modulated by the patient's breathing; 3) a haptic interface that generates a haptic response; and 4) a processing system that receives an motion signal from the motion sensor and a time-dependent electrical signal from the respiratory sensor, and executes computer code that controls the haptic interface to generate a haptic response by i) processing the motion signal to identify a first signal related to the presence or absence of SW; ii) processing the electrical signal to identify a second signal related to the patient's breathing; and iii) collectively processing the first signal, the second signal, and a preset parameter indicating an ideal time for the patient to swallow.

[0054] In this embodiment, the respiratory sensor is an electrocardiogram (ECG) sensor, and the time-dependent electrical signal is the ECG waveform. Here, the processor is further programmed to analyze the envelope of the ECG waveform to identify a second signal related to the patient's respiration.

[0055] In this embodiment, the impedance sensor comprises at least two detection electrodes and at least two drive electrodes. Here, the two detection electrodes measure the voltage associated with the current injected by the two drive electrodes and the impedance change induced by the patient's respiration.

[0056] In yet another embodiment, the embodiment provides a sensor that is fully fitted to a patient's body and has a microphone sensor configured to measure an acoustic signal modulated by the volume of the patient's speech. The sensor further comprises a haptic interface that generates a haptic response, and a processing system programmed to receive an acoustic signal from the microphone sensor and execute computer code that controls the haptic interface to generate a haptic response by i) processing the acoustic signal to identify a first signal related to the volume of the patient's speech, and ii) collectively processing the acoustic signal and a preset parameter indicating an ideal volume in the patient's speech.

[0057] In some embodiments, the processing system is further programmed to process the amplitude of an acoustic signal or to identify the frequency profile of an acoustic signal. For example, the processing system can identify the amplitude of selected frequency components of the frequency profile. In other embodiments, the processing system can generate a haptic response when the amplitude of the acoustic signal or the amplitude of selected frequency components is less than a predetermined parameter.

[0058] In this embodiment, the selected frequency component is between 20 Hz and 20,000 Hz. In another embodiment, the microphone sensor is a digital microphone or an analog microphone. In yet another embodiment, the microphone sensor is an accelerometer.

[0059] Referring to Figure 1, a wearable sensor 10, which measures swallowing (SW) and has a haptic interface, is attached to the patient's swallowing (SN) 12 to measure swallowing along with physiological signals related to the patient's respiratory rate (RR), heart rate (HR), and stroke volume. Specifically, during use, the sensor 10 detects SW and, more importantly, swallowing absence in patients with Parkinson's disease. As will be described later, ideally, SW occurs in conjunction with the patient's respiratory response, i.e., in a state where the "respiratory response and phase coincide," preferably during the period after inspiration is complete when lung volume is at a medium to low level. Hereinafter, this period will be referred to as the "safe swallowing interval." When detecting the absence (or lack thereof) of SW at an appropriate time, the sensor initiates a haptic signal, such as a vibration signal (e.g., a "ringtone"), to stimulate the patient's SN 12. This alerts the patient to swallow. Over time, this action by the sensor may "train" the patient to swallow at the appropriate time. Furthermore, this sensor-based operation can be used to count the number of swallows and for analysis and / or clinical purposes.

[0060] In general, this disclosure assumes that the sensor 10 includes a physiological sensor, electronic component, and / or computing device that can operate to receive, transmit, process, store, and / or manage patient data and information related to performing the functions of the system as described herein. This assumption includes any suitable processing device adapted to perform computational tasks consistent with the execution of computer-readable instructions stored in memory or a computer-readable recording medium. To further understand the subject matter of the claims and how this assumption is embodied, a specific example of the sensor 10 is described below.

[0061] Figures 2A and 2B show the sensor 10 in more detail. The outer housing 20, made of a soft, stretchable polymer material (e.g., a silicone elastomer such as Silbione®), houses a printed circuit board (hereinafter referred to as "PCB") 21 that functions as the electronic module of the sensor 10. The outer housing 20 is attached to a housing base 40. The housing base 40 is also typically made of a silicone elastomer that adheres to the outer housing 20 to form a waterproof seal. To bond the sensor 10 to the patient, a thin adhesive layer (not shown) is attached to the housing base 40, and then light pressure is applied to the patient's SN to bond the sensor 10 to the patient's SN.

[0062] The PCB 21 typically comprises a combination of rigid and flexible circuits incorporating a first accelerometer 32, a physiological sensor located in an outward-facing portion of the PCB 21, connected to a base portion 26 via a first slender, flexible arm 28 containing a conductive trace. The flexible arm 28 typically features a meandering pattern that allows for extension while maintaining electrical conductivity. The conductive trace is in electrical contact with a conductive pad located beneath the first accelerometer 32, allowing it to be controlled by the circuitry of the PCB 21 as described later. The second accelerometer 22 is mounted on a small glass fiber substrate 25, connected to the base portion 26 via a second slender, flexible arm 23 containing a conductive trace and also featuring a meandering pattern. Both the first and second accelerometers 32 and 22 measure time-dependent signals along the x, y, and z axes. These are modified by physiological events such as respiratory and heart rate responses (which generate RR and HR values, respectively), as well as SN and general patient movement.

[0063] The base unit 26 is further equipped with a power management integrated circuit (hereinafter referred to as "PMIC," not shown in Figures 2A to 2B), which is a chip that extracts voltage input from the rechargeable lithium-ion battery 30. This chip converts the voltage input into appropriate voltages to drive various components mounted on the PCB. A low-power Bluetooth® transceiver (not shown in Figures 2A to 2B) is mounted on the base unit 26 and, as shown in detail in Figure 8, wirelessly transmits processed data and signals from the sensor 10 to an external gateway. In addition to the Bluetooth® transceiver, an embedded microprocessor operates to control the sensor 10, and the embedded microprocessor processes time-dependent waveforms measured by the first accelerometer 32 and the second accelerometer 22 in order to generate a haptic interface in response to identifying signals related to RR, HR, SN, and general motion.

[0064] The haptic motor 24 is connected to the glass fiber substrate 25. Upon startup, the haptic motor 24 generates the haptic interface described above. The haptic motor 24 can take on multiple forms, all of which exhibit vibrational action (i.e., a "ringing sound") upon startup. One form is an "eccentric rotating mass" component, where the haptic motor rotates in a specific direction to generate vibrations felt by the patient, driven by a time-dependent analog or digital signal controlled by an embedded microprocessor. A second form is a "linear resonant actuator," typically shaped like a hockey puck, which vibrates in a similar manner in response to an analog or digital signal. A third type is a "piezoelectric module," which generates vibrations by rapidly expanding and contracting in response to a drive signal (typically a time-dependent analog voltage). Other types of haptic motors or actuators, particularly those easily mounted on the PCB 21, can also be used in this application.

[0065] As described above, the rechargeable lithium-ion battery 30 supplies power to the PCB 21. An inductive coil (not shown in Figure 1) imprinted on the PCB 21 charges the battery 30 when exposed to an electromagnetic field. Alternatively, the PCB 21 may have a port (e.g., a USB port) to connect to a power source for recharging the battery 30.

[0066] The first accelerometer 32 and the second accelerometer 22 measure time-dependent signals along the x, y, and z axes, respectively. In this process, the microprocessor within the Bluetooth® transceiver processes these motion-driven signals as described above to identify respiratory signals, heart rate signals, and fine nerve signals (SNs). For example, such processing may involve the application of algorithms to the signals, such as mathematical differential calculations between the signals, which remove baseline components to further clearly separate certain signals that are too weak to be accurately measured (e.g., signals associated with fine SNs). Alternatively, the microprocessor may determine the "energy" of the signals by measuring the overall amplitude of the signals, which is typically done by squaring each of the signals, adding them together, and then taking the square root. Typically, the most prominent signals measured by either accelerometer correspond to axes directly facing the patient's chest or SN, and in most cases, this is either the y or z axis.

[0067] As shown in Figure 4A, the electronic equipment 33 of the body-worn sensor 10 may include, for example, a sensor interface 34, one or more processors 35, a communication interface 36, a memory 37, and a power supply (or power connection) 38. Communication is performed via an internal bus 39. The sensor interface 34 is implemented in hardware or a combination of hardware and software and is used to connect to the physiological sensor via a wired connection 33a to collect data from the patient 14. The data signals from the physiological sensor may include, for example, sensor data related to the respective physiological data that the physiological sensor is designed and positioned to collect.

[0068] One or more processors 35 may be used to control the general operation of the body-worn sensor 10 and to process sensor data received by the sensor interface 34, as described herein. One or more processors 35 may be any suitable processor-based resource known in the art. One or more processors 35 include, but are not limited to, a central processing unit (CPU), a hardware microprocessor, a multicore processor, a single-core processor, a field-programmable gate array (FPGA), a controller, a microcontroller, an application-specific integrated circuit (ASIC), a digital signal processor (DSP), or any other similar processing unit capable of executing any kind of instructions, algorithms, or software for controlling the operation and performing functions of the body-worn sensor 10. In some embodiments, one or more processors 35 may comprise a processor chipset, for example, one or more coprocessors, but are not limited to.

[0069] The communication interface 36 may enable the wearable sensor 10 to communicate directly or indirectly with one or more computing networks and devices, workstations, consoles, computers, monitoring equipment, alarm systems, and / or mobile devices (e.g., mobile phones, tablets, or other portable display devices). The communication interface 36 may include various interfaces, communication channels, clouds, antennas, and / or circuits to enable wireless communication with such computing networks and devices. Essentially, any wireless communication protocol can be used.

[0070] Memory 37 is a single memory device or one or more memory devices located in one or more memory locations, including but not limited to random access memory ("RAM"), memory buffers, hard drives, databases, erasable programmable read-only memory ("EPROM"), electrically erasable programmable read-only memory ("EEPROM"), read-only memory ("ROM"), flash memory, hard disks, various layers of the memory hierarchy, or other non-temporary computer-readable media. Memory 37 may be on-chip or off-chip depending on the implementation of one or more processors 35. Memory 37 may be used to store any type of instructions and patient data related to algorithms, processes, or operations for controlling the general functions and operations of the body-worn sensor 10.

[0071] The power supply 38 may have an internal power supply such as a battery pack, and / or an interface for being powered directly or via a monitor mount from an electrical outlet. The power supply 38 may be a replaceable, removable rechargeable battery. In the case of a rechargeable battery, a small internal backup battery (or supercapacitor) may be provided to supply continuous power to the wearable sensor 10 even when the battery is being replaced. Communication between the components of the wearable sensor 10 in this example may be established using an internal bus 39.

[0072] The data signal received from the physiological sensor may be an analog signal. For example, the data signal may be input to the sensor interface 34. In addition to amplification and filtering circuits, the sensor interface may have an A / D conversion circuit that converts the analog signal to a digital signal using an amplification method, a filtering method, and an analog-to-digital (A / D) conversion method. Therefore, the sensor interface 34 is a component that may be configured to communicate with one or more physiological sensors and to receive sensor data from one or more physiological sensors.

[0073] As further described herein, the processing performed by the data acquisition circuit (not shown individually) within the sensor interface 34 generates analog or digital data waveforms that are analyzed or processed by one or more processors 35 in this particular embodiment. However, other embodiments may use other types of processors as described above. For example, one or more processors 35 may analyze the acquired data as described later. Thus, one or more processors 35 further perform various computer execution methods resulting from the body-worn sensor 10 according to this disclosure.

[0074] Figures 3A and 3B show examples of time-dependent waveforms measured by the accelerometer as described above. Different swallowing actions—namely, swallowing saliva only (indicated in bracket 50), swallowing liquid from a cup (indicated in bracket 52), and swallowing liquid using a straw (indicated in bracket 54)—modulate the waveforms in different ways. Figure 3A shows the waveforms measured along each axis of the accelerometer during these swallowing actions. As is clear from these data, swallowing actions modulate the waveforms, with the most pronounced modulation occurring along the y-axis.

[0075] Figure 3B shows a single processed waveform representing the kinetic energy detected from the patient's suprasternal fossa. The energy is calculated as described above, i.e., by squaring the signals measured along each axis, adding them together, and then calculating the square root of the sum. As is evident from this single waveform, the time-dependent energy shows a series of sharp peaks, each representing SW, indicated by an inverted triangle. Swallowing events are detected under three different conditions as described above, as indicated in parentheses 50, 52, and 54.

[0076] Figure 4A shows an example of a processed time-dependent waveform measured by an accelerometer as described above. SW modulates the waveform in various ways. As is clear from the data, the swallowing motion modulates the waveform, with the most significant modulation occurring along the y-axis acceleration and respiration.

[0077] Figure 4B shows the signal processing procedure and machine learning architecture employed for SW detection. In this architecture, the operating signal is processed sequentially through three steps: signal processing, feature extraction, and classification. The signal processing step prepares the operating signal for feature extraction. This step consists of three parts: baseline removal, bandpass filtering, and Fast Fourier Transform (FFT). In baseline removal, a technique called moving average is used to extract the baseline. This baseline is then removed from the original segment. Next, the resulting signal is passed through a bandpass filter (BPF). After BPF, the filtered signal is duplicated; one copy is used for FFT calculation, and the other copy is directly added to the calculated FFT. The processed time-dependent waveform from Figure 4A is input to a convolutional neural network (CNN) architecture for feature extraction and model training. For feature extraction, two stacked one-dimensional convolutional layers are used. Each convolutional layer is followed by a batch normalization layer. Features are extracted through convolutional layers and then passed through flattening layers and two fully connected (dense) layers for classification. The final output of the classification stage provides the probability of swallowing for a given segment.

[0078] In healthy adults, swallowing typically begins during the expiratory phase of the respiratory cycle, when lung volume is moderate to low. The respiratory phase in which swallowing begins influences the biomechanics necessary for airway protection and the efficiency of pharyngeal passage of food and liquids. Consequently, in patients with dysphagia, airway protection and pharyngeal passage efficiency can be improved by monitoring the respiratory-swallowing coordination—that is, the phase of these elements—and then training them to initiate swallowing during the expiratory phase of breathing.

[0079] Based on this, Figure 5 is a time-dependent plot showing the ideal swallowing pattern in a healthy subject. Figure 5 plots the stroke volume against time, with the peak of the plot (i.e., peak stroke volume) indicating the inspiratory / expiratory transition. The stroke volume can be estimated from the amplitude of the respiratory signal measured by an accelerometer. As the subject begins to exhale, the stroke volume systematically decreases until it reaches its lowest value when lung volume is at its minimum. During this medium-to-low expiratory period, as described above, the subject's airway is theoretically cleared of food and liquid, which represents the "safe swallowing interval" indicated by the shaded box 95 in the figure.

[0080] Figure 6A shows a time-dependent waveform of the respiratory signal identified from the operating signal measured by a sensor similar to the one shown in Figure 1, and Figure 6B shows a time-dependent waveform of the SW identified from the operating signal measured by a sensor similar to the one shown in Figure 1. The operating signal is processed by various band-pass filters, which will be described in detail later, to generate the time-dependent respiratory signal shown in Figure 6A and the SW shown in Figure 6B. The band-pass filters are implemented using embedded computer code executed on a microprocessor within the sensor; alternatively, this type of “digital signal processing” can be performed offline using an external gateway, such as a mobile phone or tablet computer, which will be described in detail later. The respiratory signal includes a low-frequency pulse corresponding to each breath by the patient, each low-frequency pulse including an upward slope indicating inspiration, as shown by the dashed line 100, and a downward slope indicating exhalation, as shown by the dashed line 102. The gray boxes 103a and 103b in the figure highlight the exhalation cycle.

[0081] The time-dependent waveform shown in Figure 6B originates from the same motor signal used to generate the respiratory signal in Figure 6A, and is filtered through a band-pass filter selected to remove the relatively low-frequency respiratory pulses shown in Figure 6A, leaving a signal corresponding to the relatively high-frequency SW, as indicated by the dashed line 104, as shown in Figure 6B. This SW is caused by the fluid motion of the patient's SN and is characterized by high-frequency pulses that are modulated up and down in proportion to the rapid motion of the SN.

[0082] The waveforms shown in Figures 6A and 6B were measured from healthy subjects. They illustrate how patients essentially swallow within safe swallowing intervals through the function of the autonomic nervous system. However, patients with Parkinson's disease, a progressive disorder of the central nervous system, may lack such ability. The technology disclosed herein detects both swallowing (SW) and respiratory events along with the phase between them, and then combines these with a haptic interface that transmits haptic signals to the Parkinson's disease patient when the patient fails to swallow at the appropriate time. By doing this continuously, the embodiment "trains" the Parkinson's disease patient to swallow at ideal timings—i.e., safe swallowing intervals—to improve the Parkinson's disease patient's quality of life.

[0083] Such a process can be driven by algorithm 59, as shown in Figure 7. Algorithm 59 is typically coded using embedded computer code and runs on a microprocessor within the sensor. As shown in Figure 7, algorithm 59 is initiated after the sensor is attached to the patient as shown in Figure 1 (step 60). The sensor does not have an "on / off" button and recognizes that it is attached to the patient. First, high-resolution waveforms are collected along the x, y, and z axes (typically sampled at 500–2,000 Hz) (step 62). These waveforms are stored in the sensor's memory and continuously converted into a time-dependent energy signal such as E(t) described above using a formula as shown below (step 64).

number

[0084] Next, the peak (PN) of the energy signal corresponding to SW is detected using a “beat-picking” algorithm (step 66). Such a peak would look like the peak shown in Figure 3B above. The beat-picking algorithm can take any of several forms. For example, a microprocessor can deploy a form of a conventional algorithm, such as the well-known Pan-Thompkins algorithm used to detect peaks in ECG waveforms. Alternatively, a microprocessor could analyze the E(t) signal using a digital filter (such as a band-pass filter), differentiate the filtered signal, and then analyze the differentiated signal to identify zero-point passes that indicate gradient changes associated with the waveform peaks. Yet another technique can be used to identify the waveform peaks corresponding to SW.

[0085] The ACC signal and / or E(t) can be further processed to identify the patient's RR and to identify more important respiratory waveforms (step 67). Figure 9B shows an example of such a waveform, which includes both inspiratory (rising slope of respiratory evoked peaks indicated by arrows 80a–80c in the figure) and expiratory (descending slope of respiratory evoked peaks indicated by arrows 82a–82c in the figure) periods. By analyzing such a waveform and the peak PN identified by using a beat picker on its characteristic Ravi, it is determined whether the PN occurs at the appropriate time and is in phase with a low to moderate tidal volume during expiratory—i.e., the safe swallowing interval described above, as shown in Figure 5 (step 68). If these are in phase, it indicates that haptic feedback is not needed because the patient is swallowing properly. The algorithm returns to continuous measurement of the ACC waveform along the three axes of the internal accelerometer (step 62). However, if PN occurs outside the safe swallowing interval—that is, if swallowing occurs very early or very late relative to the patient's inhalation and exhalation, or if swallowing does not occur at all—the sensor provides the patient with a haptic signal in the form of a slight vibration or other type of haptic feedback. Once this is complete, the algorithm returns to continuous measurement of the ACC waveform along the three axes of the internal accelerometer (step 62) and continues processing as described above.

[0086] As shown in Figure 8, the sensor 10 is configured to transmit information such as the information described above (e.g., processed numerical values ​​and time-dependent waveforms) to the cloud 77 via the gateway device 72. This allows, for example, a clinician to remotely monitor the patient 14. For example, in a typical use, the sensor 10 is attached to the patient's SN and monitors the patient's physiological responses, such as swallowing behavior along with lung and cardiac signals. This information—i.e., both time-dependent waveforms and numerical values ​​of HR, RR, and PN—is transmitted to the gateway device 72 via Bluetooth®, as indicated by arrow 74. The gateway device 72 is typically a mobile phone or tablet computer (e.g., running on the Android® or iOS operating system) running a custom software application. Once the gateway device 72 and its custom software application receive information from the sensor, they transmit the information to the cloud 77 using a cellular or Wi-Fi® transmitter, as indicated by arrow 75. When the patient-generated information reaches the cloud 77, it is processed in various ways. For example, patient-generated information can be 1) stored in a database, 2) processed using various algorithms, such as machine learning and artificial intelligence-based algorithms, to estimate the patient's physiological state, 3) transmitted to a third-party software application using a web service interface, for example, or 4) transmitted to a hospital information system such as an electronic medical record (EMR). A remote clinician can use this information to manage the patient, i.e., 1) collect information for examination, 2) prescribe medication, 3) initiate a haptic interface, or 4) simply observe the patient for a clinical trial, for example.

[0087] Figures 9A, 9B, 10A, and 10B illustrate methods for extracting patient respiratory and cardiac signals from the ACC waveform, and for using the patient's respiratory and cardiac signals to determine the waveform and processed numerical values ​​such as RR and HR. Figures 9A and 10A show the unfiltered raw ACC waveform measured along the z-axis, in which case the z-axis corresponds to the axis directly toward the patient's chest. The waveform includes a cardiac component (shown as a pair of sharp high-frequency pulses, with the two pulses corresponding to the S1 and S2 heart sounds corresponding to each heartbeat) and low-frequency vibrations, the low-frequency vibrations corresponding to the fluctuations of the patient's chest due to respiration, more specifically, inspiration and expiration.

[0088] When filtering is performed using a band-pass filter, for example, an infinite impulse response (IIR) band-pass filter with a low-frequency cutoff of approximately 0.01 Hz and a high-frequency cutoff of approximately 1 Hz, the ACC waveform shown in Figure 9A is transformed into the waveform shown in Figure 9B. The filter removes high-frequency noise and cardiac signals from the waveform, leaving only the low-frequency vibrations that occur with each breath of the patient, which correspond to inspiration and expiration (arrows 82a, 82b, and 82c) (indicated by arrows 80a, 80b, and 80c). These periods can be extracted from the filtered waveform using well-known signal processing techniques, particularly with reference to Figures 4 and 5, in order to determine the safe swallowing interval described above.

[0089] Figure 10A, like Figure 9A, shows the unfiltered raw ACC waveform. When filtered using an IIR bandpass filter with a low-frequency cutoff of approximately 1 Hz and a high-frequency cutoff of approximately 12 Hz, the ACC waveform shown in Figure 10A is transformed into the waveform shown in Figure 10B. Here, the filter removes the respiratory signals occurring at relatively low frequencies as shown in Figure 9B, leaving only the cardiac signals consisting of high-frequency noise and the S1 and S2 heart sounds. These are indicated by black circles and black squares, respectively. The S1 heart sound corresponds to the closure of the patient's mitral and tricuspid valves, and the S2 heart sound corresponds to the closure of the aortic and pulmonary valves. These signals can be used to identify other cardiac characteristics, such as HR and systolic time interval. Such time intervals can be combined with physiological signals such as ECG, impedance, and photoplethysmography waveforms, which are measured by other embodiments to determine parameters called pulse wave arrival time (PAT) and pulse wave propagation time (PTT). The reciprocals of PAT and PTT are known to correlate with changes in systolic (SYS) blood pressure and diastolic (DIA) blood pressure.

[0090] In related embodiments, the haptic interface can be coupled to vital signs and parameters independent of SYS and DIA values ​​determined by SW, e.g., PAT, PTT, and other methods. For example, the haptic interface is automatically activated when a patient's blood pressure exceeds a preset limit that could be harmful to the patient. Such events may occur, for example, during exercise or stress.

[0091] Patients with Parkinson's disease may have difficulty swallowing regularly, which is a detrimental condition that can have adverse effects. Therefore, increasing the frequency of swallowing is generally considered a beneficial factor for patients suffering from this disease. To clinically validate the concept of using a haptic interface to induce swallowing, a clinical trial was conducted using a wearable sensor as shown in Figure 1, with the addition of four independent vibration motors and a Bluetooth® interface to a gateway device (as shown in Figure 8). The vibration motors are driven by a digital pulse-width modulation (PWM) signal with amplitude shift modulation (ASK), thereby individually setting the vibration intensity of each motor. The vibration response (i.e., the haptic interface) is identified by analyzing the phase of the respiratory and swallowing signals, as described above, and is performed by the gateway device in accordance with the method described above in relation to Figure 7.

[0092] Figure 11A shows a typical time-dependent waveform measured in this study, including a distinct low-frequency signal due to respiration and a relatively high-frequency signal due to SW. Case 1, indicated by the dashed box 110a in the figure, corresponds to SW occurring during exhalation within a safe swallowing interval, as shown in Figures 4 and 5. This represents an ideal SW, and the corresponding haptic pattern, as indicated by the dashed box 110b in Figure 11B, involves all four motors operating synchronously three times at 500-millisecond intervals, which indicates vibrations detected by the accelerometer in the body-worn sensor. For training purposes, this is considered a "positive" response.

[0093] Cases 2-5 in Figure 11A correspond to suboptimal swallowing (SW) and their corresponding "negative" responses. Here, the algorithm running at the gateway detects suboptimal breath-swallowing phase patterns, indicated by dashed boxes 112a, 114a, and 116a, and activates all four vibration motors at 250-millisecond intervals. This haptic interface indicates inappropriate swallowing events to the patient, as shown by dashed boxes 112b, 114b, and 116b in Figure 11B. Over time, the patient perceives these haptic signals as negative responses and potentially adjusts their swallowing pattern to fit safe swallowing intervals (indicated by positive responses). In this way, the haptic interface can "train" the patient to swallow healthily, thereby improving treatment outcomes.

[0094] Referring to Figure 12, a study was conducted with N=20 subjects to demonstrate this concept. The subjects were Parkinson's disease patients diagnosed with dysphagia of any severity by a speech-language pathologist (hereinafter, "SLP"). During the study, each subject wore a sensor as shown in Figure 1 and performed passive activities. SW was monitored by the SLP (which served as a reference device) and the sensor (test device). The sensor independently monitored respiration and used it to generate a haptic interface as described above. More specifically, the algorithm executed in the sensor processed the phases of swallowing and respiratory events to determine a safe swallowing interval, as described above, and initiated haptic feedback when the phases of these events were out of sync. Subjects were measured using the evaluation criteria described above, which indicate improvement outcomes in Parkinson's disease patients, with and without haptic feedback to assess the effectiveness of this approach for increasing SW.

[0095] Figure 12 shows that the haptic interface improved swallowing frequency. This study shows that the swallowing frequency increased from 0.7 times / minute in the group without the haptic interface to 1.1 times / minute in the group with the haptic interface. This represents a 57% increase in swallowing frequency and is considered to be the effect of the technology described herein. Similar effects are expected in other patients with conditions affecting swallowing, such as those with head and neck cancer, Alzheimer's disease and other dementias, myasthenia gravis, ALS, and related diseases.

[0096] Referring to Figures 13, 13A, and 13B, the sensor shown in Figure 1 may employ other configurations, for example, the first part 122 of the sensor comprises a battery and an electronic system (e.g., a power management circuit, a vibration motor for the haptic interface) and a measurement system for measuring physiological signals such as HR and RR (e.g., a first accelerometer, an ECG and / or impedance circuit including electrodes). The first part 122 is connected to a second part 124 which includes a second accelerometer for measuring SW. The connection is made via a third part 126 which includes a conductor.

[0097] Such a spatially distributed system uses separate, uncoupled sensors (i.e., motion-detecting accelerometers) to measure respiration and swallowing, thus simplifying the analysis of these data to independently measure respiratory and swallowing waveforms and ultimately determine a safe swallowing interval. By coupling with a haptic interface as described above, this can improve the accuracy with which this parameter can be determined.

[0098] Figures 13A and 13B illustrate, for example, how this sensor can be applied to a patient. In Figure 13A, the first part 122 of the sensor is directly attached to the patient's chest, which is the site where accurate measurement of RR is most easily performed. As described above, this parameter can be measured by one or more different detection methods, for example, by an accelerometer, ECG and / or impedance circuit, the impedance circuit will be discussed later. The third part 126 is positioned in the proximal part of the patient's larynx to measure SW more accurately. The isolation and dispersion characteristics of this configuration result in relatively higher measurement accuracy for RR and SW. When coupled to a haptic interface, this increases the sensor's influence on the patient, for example, its ability to increase the patient's SW frequency.

[0099] Figure 13B shows yet another configuration of the sensor 120 attached to the patient. Here, the first part (hidden by the patient's shirt in the figure) is attached to a lower position on the patient's chest. This position—directly above the patient's heart and lungs—typically allows for relatively accurate measurements of HR and RR. The second part 124, connected to the first part via a third part 126, is positioned directly in the patient's suprasternal fossa. As mentioned above, this is the ideal position for measuring SW.

[0100] As shown in Figures 13, 13A, and 13B, yet another combination of the first part 122, the second part 124, and the third part 126 is also included in the following claims.

[0101] In other embodiments, sensors for detecting other parameters such as vital signs including HR, RR, pulse oximetry (here, “SpO2”), temperature, and SYS / DIA blood pressure may include haptic interfaces as described herein. The haptic interface may be used to form a “feedback loop” that prompts the patient to change their activity (e.g., movement, exercise, stress, or sleeping in a particular position) when vital signs fall outside a preset range. Referring, for example, to Figure 14, a sensor having a haptic interface has a patch 152 or an oximeter 154 that is attached to the chest or finger of a patient 150, respectively. Here, the patch 152 measures both ECG and impedance waveforms to calculate HR and RR values. The oximeter measures SpO2 values ​​along with HR and RR. Both sensors may additionally have a temperature sensor for measuring skin temperature.

[0102] For patch 152, disposable electrodes secure the sensor to the patient's chest. The electrodes may include a pair of “sensing” and “driving” electrodes for detecting bioelectrical signals that generate the ECG and impedance waveforms described above after processing. For impedance measurement, the “driving” electrode pair is configured to inject a low high-frequency current into the patient's chest. The current can be injected at multiple frequencies from approximately 5 to 1000 kHz, and the current typically has an amplitude of approximately 0.1 to 1.0 mA. The sensing electrode pair measures bioelectrical signals that generate time-dependent ECG and impedance waveforms after processing. Upon further processing, such waveforms generate HR, RR, stroke volume, stroke volume, and cardiac output.

[0103] Patch 152 may also have a reflective optical sensor equipped with LEDs that emit red and infrared wavelengths. In an embodiment, a circular array of photodetectors surrounds the LEDs. A thin film of Kapton® with embedded electrical wiring surrounds the photodetectors and LEDs and generates heat when a voltage is applied. For this purpose, a closed-loop system is used to gently warm the skin to 41°C to 42°C, thereby increasing blood flow and amplifying the corresponding light waveform, thereby improving the accuracy of SpO2 measurement.

[0104] The patch 152 may have a temperature sensor (not shown) and a thermally conductive metal post that connects to the patient's skin during measurement. This allows the patch 152 to measure skin temperature. The patch 152 is powered by a rechargeable lithium-ion battery that can be charged via a small USB port or by a built-in transformer that performs wireless charging. In yet another embodiment, the patch 152 has an acoustic sensor configured to measure heart sounds at S1 and S2.

[0105] Referring to Figures 15A, 15B, and 15C, in yet another embodiment, a sensor 10 having an accelerometer, as shown in Figure 1 (and again in Figure 15A), is attached to different parts of the patient 12, such as the patient's hand (Figure 15B) and upper arm (Figure 15C). More specifically, in Figure 15B, the sensor 10b is attached to the patient's hand 12b. In this configuration, the haptic interface is activated during a scratching event and may be used to "train" the patient to reduce their response to scratching and other types of skin irritation. In Figure 15C, the sensor 10c is attached to the patient's arm 12c, and the haptic interface is used to reduce arm movement or to prevent the patient from raising their arm above a certain level.

[0106] In other embodiments, the sensors described herein can count physiological events that can be easily measured by an accelerometer, such as coughing, sneezing, and aspiration, and can apply a haptic interface to the patient when these events exceed a preset level.

[0107] In other embodiments, the sensors described herein can be used in combination with other devices used in hospitals and homes, such as feeding tubes, for example, feeding tubes coupled to nerve stimulators. For example, the sensors and their haptic interface can work in combination with a nerve stimulating feeding tube to induce swallowing at the optimal time. In other embodiments, the sensors and their haptic interface also allow for external manual adjustment of nerve stimulation when coupled to a feeding tube. For example, the sensors may have a button, which, when pressed, can induce internal nerve stimulation from the feeding tube.

[0108] In other embodiments, the haptic interface can be timed to reduce the amount of saliva produced by the patient. The body-worn sensor may also have a microphone that can be used to measure sounds produced by the patient and to reduce slurred speech or to encourage the patient to speak louder if their voice is too weak.

[0109] Furthermore, those skilled in the art who benefit from this disclosure will understand that the physiological sensor may be implemented as a motion sensor such as an accelerometer or one or more electrocardiogram leads, a respiratory sensor, or an acoustic sensor such as a microphone sensor, as described above. In embodiments in which the physiological sensor is implemented as a microphone sensor, the microphone sensor may be used to collect other acoustic signals, such as sounds produced by the patient's body through swallowing or breathing (i.e., through respiration).

[0110] In a first embodiment, a sensor fully fitted to a patient's body comprises a microphone sensor, a haptic interface, and a processing system. The microphone sensor is configured to measure acoustic signals generated by the patient's body. The haptic interface is configured to generate haptic responses. The processing system is programmed to process the measured acoustic signals to detect swallowing, breathing, coughing, or a combination thereof, and to execute computer code that controls the haptic interface to generate haptic responses in response to the processing of the acoustic signals.

[0111] In the second embodiment, the sensor of the first embodiment generates the measured acoustic signal by the patient's body through swallowing, breathing, coughing, or a combination thereof.

[0112] In a third embodiment, the sensor of the first embodiment responds to a preset parameter indicating the time when the patient swallows.

[0113] In the fourth embodiment, the sensor of the first embodiment further comprises an accelerometer.

[0114] In the fifth embodiment, the sensor of the fourth embodiment is programmed to process at least one time-dependent motion signal using a first algorithm in order to identify a time-dependent energy signal.

[0115] In the sixth embodiment, the sensor of the fifth embodiment is configured to collectively process time-dependent motion signals measured along the x-axis, time-dependent motion signals measured along the y-axis, and time-dependent motion signals measured along the z-axis by the accelerometer in order to identify a time-dependent energy signal, and the first algorithm uses the following equation to identify the time-dependent energy signal:

number

[0116] E(t) is the time-dependent energy, x(t) is the time-dependent motion signal measured along the x-axis corresponding to the patient by an accelerometer, y(t) is the time-dependent motion signal measured along the y-axis corresponding to the patient by an accelerometer, and z(t) is the time-dependent motion signal measured along the z-axis corresponding to the patient by an accelerometer.

[0117] In the seventh embodiment, the sensor of the first embodiment includes processing the acoustic signal by applying a convolutional neural network to the acoustic signal.

[0118] In the eighth embodiment, the sensor of the first embodiment is further programmed to process the amplitude of the acoustic signal.

[0119] In the ninth embodiment, the sensor of the first embodiment is further programmed to identify the frequency profile of the acoustic signal.

[0120] In the tenth embodiment, the sensor of the ninth embodiment is further programmed to identify the amplitude of selected frequency components within the frequency profile.

[0121] In the eleventh embodiment, a method for assisting a patient's swallowing comprises measuring acoustic signals generated by the patient's body; processing the measured acoustic signals to detect swallowing, breathing, coughing, or a combination thereof; and controlling a haptic interface to generate a haptic response in response to the processing of the measured acoustic signals.

[0122] In the twelfth embodiment, the method of the eleventh embodiment generates the measured acoustic signal by the patient's body through swallowing, breathing, coughing, or a combination thereof.

[0123] In the 13th embodiment, the method of the 11th embodiment responds to a preset parameter indicating the time when the patient swallows.

[0124] In the fourteenth embodiment, the method of the eleventh embodiment further comprises measuring the patient's movements.

[0125] In a 15th embodiment, the wearable sensor comprises a first physiological sensor, a haptic interface, and a processing system. The first physiological sensor is configured to acquire a first set of physiological signals that may indicate a patient swallowing event related to the patient's respiratory stage. The haptic interface is configured to generate a haptic response. The processing system is programmed to receive motion signals from the motion sensor, process the first set of physiological signals to identify a first signal related to the presence or absence of a swallowing event and a second signal related to the patient's respiration, and generate a haptic response by collectively processing the first signal, the second signal, and a preset parameter indicating the patient's swallowing opportunity.

[0126] In the sixteenth embodiment, in the wearable sensor of the fifteenth embodiment, the first physiological sensor is a motion sensor, and the first set of physiological signals includes time-dependent motion signals modulated by the patient's swallowing and breathing, and the preset parameter is time.

[0127] In the 17th embodiment, in the wearable sensor of the 15th embodiment, the first physiological sensor is a respiratory sensor, and the first set of physiological signals includes a time-dependent signal modulated by the patient's respiration, and a preset parameter is time.

[0128] In the 18th embodiment, the wearable sensor of the 15th embodiment further comprises a second physiological sensor configured to acquire a second set of physiological signals that may indicate a patient's swallowing event in relation to the patient's respiratory stage.

[0129] In the 19th embodiment, in the wearable sensor of the 18th embodiment, the second physiological sensor is a respiratory sensor, and the second set of physiological signals includes a time-dependent electrical signal modulated by the patient's respiration.

[0130] In the 20th embodiment, the sensor of the 15th embodiment is a time-dependent signal indicating the patient's inhalation and exhalation.

[0131] In the 21st embodiment, the sensor of the 20th embodiment has a preset parameter that indicates the time period during which the patient should be exhaling, and the processing system generates a haptic response when it determines that no swallowing event occurs during the patient's exhalation, a swallowing event occurs during the patient's inhalation, or a swallowing event occurs between the patient's inhalation and exhalation.

[0132] In the 22nd embodiment, the wearable sensor of the 15th embodiment, the first physiological sensor is a microphone sensor, and the first set of physiological signals includes an acoustic signal modulated by the patient's speech volume, and a preset parameter indicates the patient's speech volume.

[0133] In the 23rd embodiment, the wearable sensor of the 15th embodiment is a first physiological sensor, and the first set of physiological signals includes acoustic signals generated by swallowing, breathing, or a combination thereof by the patient's body.

[0134] In the 24th embodiment, the computer execution method comprises: receiving a first set of physiological signals from a first physiological sensor located in the patient's suprasternal fossa that may indicate a patient swallowing event related to the patient's respiratory phase; processing the first set of physiological signals to identify a first signal related to the presence or absence of a swallowing event and a second signal related to the patient's respiration; processing the first signal, the second signal and a preset parameter indicating the patient's swallowing opportunity; and generating a haptic response via a haptic interface to prompt the patient to swallow.

[0135] In the 25th embodiment, the computer execution method of the 24th embodiment, the first physiological sensor is a motion sensor, a respiration sensor, or a microphone sensor.

[0136] In the 26th embodiment, the computer execution method of the 24th embodiment further comprises a second physiological sensor configured to acquire a second set of physiological signals that may indicate a patient swallowing event in relation to the patient's respiratory stage.

[0137] In the 27th embodiment, the computer execution method of the 26th embodiment, the second physiological sensor is a respiratory sensor, and the second set of physiological signals includes time-dependent electrical signals modulated by the patient's respiration.

[0138] In the 28th embodiment, the system comprises a wearable sensor, an external gateway, and a computer system. The wearable sensor is configured to acquire a first set of physiological signals from the patient that may indicate a patient swallowing event related to the patient's respiratory stage, process the first set of physiological signals to identify a first signal related to the presence or absence of a swallowing event and a second signal related to the patient's respiration, and generate a haptic response indicating an opportunity for the patient to swallow. The external gateway is programmed to wirelessly receive the information transmitted from the wearable sensor. The computer system may be used to store or access the transmitted information.

[0139] In the 29th embodiment, the system of the 28th embodiment further comprises: a wearable sensor configured to acquire a first physiological sensor which may indicate a patient swallowing event related to the patient's respiratory stage; a haptic interface configured to generate a haptic response; and a processing system programmed to control the haptic interface which receives motion signals from a motion sensor and processes the first set of physiological signals to identify a first signal related to the presence or absence of a swallowing event and a second signal related to the patient's respiration, and generates a haptic response by collectively processing the first signal, the second signal and a preset parameter indicating the patient's swallowing opportunity.

[0140] In the 30th embodiment, in the system of the 29th embodiment, the first physiological sensor is a motion sensor, a respiration sensor, or a microphone sensor.

[0141] In the 31st embodiment, the system of the 29th embodiment further comprises a second physiological sensor configured to acquire a second set of physiological signals that may indicate a patient swallowing event in relation to the patient's respiratory stage.

[0142] In the 32nd embodiment, in the system of the 31st embodiment, the second physiological sensor is a respiratory sensor, and the second set of physiological signals includes a time-dependent electrical signal modulated by the patient's respiration.

[0143] In the 33rd embodiment, the sensor of the first embodiment processes the measured acoustic signal to determine if the swallowing frequency exceeds a threshold parameter, and a haptic response is generated accordingly.

[0144] In the 34th embodiment, the sensor of the first embodiment is a microphone sensor.

[0145] In this specification, unless expressly defined using the phrase "in this specification" or similar wording, the meaning of any term is not intended to be limited beyond its plain or ordinary meaning. Where any term is referred to in this specification in a way that corresponds to one meaning, this is done for clarity only and is not intended to limit the claim term to that single meaning. Finally, unless a claim element is defined by the phrase "means" and a description of a function without a description of a structure, the scope of any claim element is not intended to be construed under Section 112(f) of the United States Patent Act.

[0146] Multiple embodiments of wearable sensors disclosed herein are described as “configured to” perform several functions. In some contexts, the expression “configured” means that a design selection has been made to provide a certain function. For example, a wearable sensor configured to measure an acoustic signal means that the wearable sensor includes a microphone sensor. In other contexts, the expression “configured” may indicate that a programmable electronic component has been programmed to perform a given function. Thus, as another example, a wearable sensor having a processing system or a processing system itself may be configured to perform several functions by appropriately programming the processing system. A person skilled in the art who benefits from this disclosure can easily construct various embodiments of wearable sensors from the context provided herein.

[0147] Expressions such as “includes” and “may include” as used herein indicate the presence of disclosed functions, operations, and components, and do not limit the presence of one or more additional functions, operations, and components. In this specification, terms such as “includes” and / or “have” may be interpreted as indicating a particular characteristic, number, operation, component, part or combination thereof, but should not be interpreted as excluding the possibility of the presence or addition of one or more other characteristics, number, operation, component, part or combination thereof.

[0148] In this specification, the article “a” is intended to have its usual meaning in the patent art, i.e., “one or more.” In this specification, the term “about” applied to a value generally means within the tolerance range of the apparatus used to produce the value, or in some examples, ±10%, ±5%, or ±1%, unless otherwise specified. Furthermore, the term “substantially” as used in this specification means, for example, a majority, almost all, all, or a quantity ranging from about 51% to about 100%. Also, the examples in this specification are for illustrative purposes only and are presented for discussion purposes and are not limiting.

[0149] In this specification, “to provide” means to own and / or control an article. This includes, for example, forming (or assembling) part or all of an article from the materials that constitute the article, and / or acquiring ownership and / or control of an article that has already been formed.

[0150] Unless otherwise defined, all terms used herein, including technical and / or scientific terms, have the same meaning as those generally understood by those skilled in the art relating to this disclosure. Furthermore, unless otherwise defined, all terms as defined in commonly used dictionaries should not be interpreted more than necessary. Details are provided below to further illustrate the examples. However, those skilled in the art will understand that the examples can be carried out without these specific details. In other examples, well-known structures and apparatus are shown in block diagrams or schematic diagrams rather than in detail, so as not to obscure the examples. Furthermore, unless otherwise stated, features of different examples described below can be combined with each other. For example, a variation or modification described in relation to one example is applicable to other examples unless otherwise stated.

[0151] Furthermore, equivalent or similar elements, or elements with equivalent or similar functions, will be indicated by equivalent or similar reference numbers in the following descriptions. In the drawings, identical or functionally equivalent elements are assigned the same reference number, so redundant descriptions of elements with the same reference number may be omitted. Therefore, descriptions of elements with the same or similar reference numbers are interchangeable.

[0152] When it is stated that one element is “connected” or “joined” to another element, it should be understood that the element is either directly connected or joined to the other element, or that an intermediate element may exist. Conversely, when it is stated that one element is “directly connected” or “directly joined” to another element, no intermediate element exists. Other terms describing relationships between elements should be interpreted similarly (e.g., “between” vs. “directly between,” “adjacent” vs. “directly adjacent,” etc.).

[0153] In this disclosure, expressions containing ordinal numbers such as "first," "second," etc., may modify various elements. However, such elements are not limited by the above expressions. For example, the above expressions do not limit the order and / or importance of the elements. The above expressions are simply used to distinguish one element from another. For example, the first box and the second box are both boxes, but they refer to different boxes. To give a further example, the first element may be referred to as the second element, and similarly, the second element may be referred to as the first element, and these do not deviate from the scope of this disclosure.

[0154] A sensor refers to a component that converts a physical quantity being measured into an electrical signal (e.g., an electric current signal or a voltage signal). Physical quantities include, but are not limited to, electromagnetic radiation (e.g., photons of infrared or visible light), magnetic fields, electric fields, pressure, force, temperature, electric current, and voltage.

[0155] In one or more embodiments, the use of expressions such as “can,” “is possible,” “operable,” or “configured” refers to a device, logic, hardware, and / or element designed to be used in a particular manner. In one or more embodiments, the use of the expression “exceeds” indicates that a measured value is higher than a predetermined threshold (e.g., an upper threshold) or lower than a predetermined threshold (e.g., a lower threshold). Where a predetermined threshold range (defined by an upper and lower threshold) is used, the use of the expression “exceeds” in one or more embodiments may also indicate that a measured value is outside the predetermined threshold range (e.g., higher than the upper threshold or lower than the lower threshold). The subject matter of this specification is provided as examples of devices, systems, methods, circuits, and programs for performing the features described herein. However, in addition to the features described above, other features or variations are conceivable. Implementation of the components and functions of this specification is conceivable to be carried out using newly emerging technologies that may replace any of the technologies implemented above.

[0156] Detailed descriptions are made with reference to the accompanying drawings and are provided to help provide a comprehensive understanding of the various embodiments of this disclosure. Modifications to the function and arrangement of the elements discussed can be made without departing from the spirit and scope of the disclosure. Various embodiments may omit, replace, or add various procedures and components as appropriate. For example, features described in relation to a particular embodiment can be combined in other embodiments. Furthermore, descriptions of well-known functions and configurations may be omitted for clarity and brevity. Therefore, those skilled in the art will recognize that various changes and modifications are possible to the examples described herein without departing from the spirit and scope of this disclosure.

[0157] Accordingly, those skilled in the art will readily understand various modifications to this disclosure, and the general principles defined herein may be applied to other modifications without departing from the spirit or scope of this disclosure. Throughout this disclosure, the terms “example,” “exemplary,” or “exemplary” are illustrative and do not imply or require any preference for any example described herein. Accordingly, this disclosure should be construed in the broadest possible sense to be consistent with the disclosed principles and novel features, and should not be limited to the examples or designs described herein.

Claims

1. A sensor that is fully attached to the patient's body, A microphone sensor configured to measure acoustic signals generated by the patient's body, A haptic interface configured to generate haptic responses, The microphone sensor receives an acoustic signal, To detect swallowing, breathing, coughing, or a combination thereof, the measured acoustic signals are processed, A processing system programmed to execute computer code that controls the haptic interface in order to generate the haptic response in response to the processing of the measured acoustic signal, A sensor equipped with the following features.

2. The sensor according to claim 1, wherein the measured acoustic signal is generated by the patient's body through swallowing, breathing, coughing, or a combination thereof.

3. The sensor according to claim 1, wherein the haptic response responds to a preset parameter indicating the time when the patient swallows.

4. The sensor according to claim 1, further comprising an accelerometer.

5. The sensor according to claim 4, wherein the processing system is programmed to process at least one time-dependent motion signal using a first algorithm in order to identify a time-dependent energy signal.

6. The first algorithm is configured to collectively process time-dependent motion signals measured along the x-axis, time-dependent motion signals measured along the y-axis, and time-dependent motion signals measured along the z-axis by the accelerometer in order to identify the time-dependent energy signal. The first algorithm identifies the time-dependent energy signal using the following equation [Math 1] Or using its mathematical transformation, The sensor according to claim 5, wherein E(t) is time-dependent energy, x(t) is a time-dependent motion signal measured by the accelerometer along the x-axis corresponding to the patient, y(t) is a time-dependent motion signal measured by the accelerometer along the y-axis corresponding to the patient, and z(t) is a time-dependent motion signal measured by the accelerometer along the z-axis corresponding to the patient.

7. The sensor according to claim 1, wherein the processing of the acoustic signal comprises applying a convolutional neural network to the acoustic signal.

8. The sensor according to claim 1, wherein the processing system is further programmed to process the amplitude of the acoustic signal.

9. The sensor according to claim 1, wherein the processing system is further programmed to identify the frequency profile of the acoustic signal.

10. The sensor according to claim 9, wherein the processing system is further programmed to identify the amplitude of selected frequency components within the frequency profile.

11. A method for assisting a patient's swallowing, Measuring acoustic signals generated by the patient's body, Processing the measured acoustic signals to detect swallowing, breathing, coughing, or a combination thereof, Controlling the haptic interface to generate a haptic response in accordance with the processing of the measured acoustic signal, A method for providing this.

12. The method according to claim 11, wherein the measured acoustic signal is generated by the patient's body through swallowing, breathing, coughing, or a combination thereof.

13. The method according to claim 11, wherein the haptic response responds to a preset parameter indicating the time when the patient swallows.

14. The method according to claim 11, further comprising measuring the patient's movements.

15. A first physiological sensor configured to acquire a first set of physiological signals that may indicate a patient's swallowing event related to the patient's respiratory stage, A haptic interface configured to generate haptic responses, Receive motion signals from motion sensors, To identify a first signal related to the presence or absence of a swallowing event and a second signal related to the patient's respiration, the first set of physiological signals is processed. A processing system programmed to control a haptic interface in order to generate the haptic response by collectively processing the first signal, the second signal, and a preset parameter indicating the patient's swallowing opportunity, A wearable sensor equipped with [feature / feature].

16. The first physiological sensor is a motion sensor, The first set of physiological signals includes time-dependent operating signals that are modulated by the patient's swallowing and breathing. The wearable sensor according to claim 15, wherein the aforementioned preset parameter is a time interval.

17. The first physiological sensor is a respiratory sensor, The first set of physiological signals includes time-dependent signals modulated by the patient's respiration. The wearable sensor according to claim 15, wherein the aforementioned preset parameter is a time interval.

18. The wearable sensor according to claim 15, further comprising a second physiological sensor configured to acquire a second set of physiological signals that may indicate a patient's swallowing event in relation to the patient's respiratory stage.

19. The second physiological sensor is a respiratory sensor, The wearable sensor according to claim 18, wherein the second set of physiological signals includes a time-dependent electrical signal modulated by the patient's respiration.

20. The wearable sensor according to claim 15, wherein the second signal is a time-dependent signal indicating the patient's inhalation and exhalation.

21. The sensor according to claim 20, wherein the preset parameters indicate a period of time during which the patient should be exhaling, and the processing system generates the haptic response when it determines that no swallowing event occurs during the patient's exhalation, that the swallowing event occurs during the patient's inhalation, or that the swallowing event occurs between the patient's inhalation and the patient's exhalation.

22. The first physiological sensor is a microphone sensor, The first set of physiological signals includes an acoustic signal modulated by the patient's speech volume. The wearable sensor according to claim 15, wherein the preset parameter indicates the volume of the patient's speech.

23. The first physiological sensor is a microphone sensor, The wearable sensor according to claim 15, wherein the physiological signals of the first set include acoustic signals generated by swallowing, breathing, or a combination thereof by the patient's body.

24. A first set of physiological signals that may indicate patient swallowing events related to the patient's respiratory phase are received from a first physiological sensor located in the patient's suprasternal fossa, To identify a first signal related to the presence or absence of a swallowing event and a second signal related to the patient's respiration, the first set of physiological signals is processed, Processing the first signal, the second signal, and a preset parameter indicating the patient's swallowing opportunity, To generate haptic responses that encourage swallowing in patients via a haptic interface, A computer execution method comprising the following features.

25. The computer execution method according to claim 24, wherein the first physiological sensor is a motion sensor, a respiratory sensor, or a microphone sensor.

26. The computer execution method according to claim 24, further comprising a second physiological sensor configured to acquire a second set of physiological signals that may indicate a patient's swallowing event in relation to the patient's respiratory stage.

27. The second physiological sensor is a respiratory sensor, The computer execution method according to claim 26, wherein the second set of physiological signals includes time-dependent electrical signals modulated by the patient's respiration.

28. It is a wearable sensor, A first set of physiological signals from the patient that may indicate patient swallowing events related to the patient's respiratory stage are obtained, To identify a first signal related to the presence or absence of a swallowing event and a second signal related to the patient's respiration, the first set of physiological signals is processed. A wearable sensor configured to generate haptic responses indicating opportunities for the patient to swallow, An external gateway programmed to wirelessly receive information transmitted from the wearable sensor, A computer system capable of storing or accessing the transmitted information, A system equipped with these features.

29. The aforementioned wearable sensor is A first physiological sensor configured to acquire a first set of physiological signals that may indicate a patient's swallowing event related to the patient's respiratory stage, A haptic interface configured to generate haptic responses, Receive motion signals from motion sensors, To identify a first signal related to the presence or absence of a swallowing event and a second signal related to the patient's respiration, the first set of physiological signals is processed. A processing system programmed to control a haptic interface in order to generate the haptic response by collectively processing the first signal, the second signal, and a preset parameter indicating the patient's swallowing opportunity, The system according to claim 28, further comprising the above.

30. The system according to claim 29, wherein the first physiological sensor is a motion sensor, a respiratory sensor, or a microphone sensor.

31. The system according to claim 29, further comprising a second physiological sensor configured to acquire a second set of physiological signals that may indicate a patient's swallowing event in relation to the patient's respiratory stage.

32. The second physiological sensor is a respiratory sensor, The system according to claim 31, wherein the second set of physiological signals includes a time-dependent electrical signal modulated by the patient's respiration.

33. By processing the measured acoustic signal, it is determined that the swallowing frequency exceeds the threshold parameter. The sensor according to claim 1, wherein the haptic response is generated in accordance with the determination.

34. The sensor according to claim 1, wherein the microphone sensor is a microphone.

Images