Closed-loop wearable sensor and method

Abstract

Multi-mode neurological and cardiopulmonary monitoring systems may be designed for the especially challenging needs of very fragile patients, such as neonates and infants. Compact, lightweight, time-synchronized wireless sensors may be built into wearable modules. The modules' skin-friendly encapsulations and bio-integrated form factors reduce the risk of iatrogenic injury while permitting natural motion and comforting touch. The system's integrated sensors include electroencephalography, near-infrared spectroscopy, electrocardiography, seismocardiography, photoplethysmography, motion, and temperature. By combining and analyzing the sensors' time-synchronized readings, the system can calculate heart rate, respiratory rate, pulse oxygenation, temperature, perfusion index, tidal volume, pulse wave velocity, and pulse arrival time (a surrogate for blood pressure). Advanced algorithms can further evaluate sleep patterns and detect arrhythmias, strokes, hemorrhaging and pre-seizure events. Other design features facilitate low cost, ruggedness, and ease of use in both hospital and non-hospital settings.

Description

CLOSED-LOOP WEARABLE SENSOR AND METHOD CROSS REFERENCE TO RELATED APPLICATION

[0001] The present application claims the benefit of US provisional application Serial No. 63 / 524,323, filed on June 30, 2023, the content of each of which is hereby incorporated by reference as if set forth in their entirety herein.TECHNICAL FIELD

[0002] Related fields include medical care of neonates, infants, and other especially fragile patients, neurological and cardiopulmonary health monitoring, and remote healthcare.BACKGROUND

[0003] This section of this document introduces information about and / or from the art that may provide context for or be related to the subject matter described herein and / or claimed below. It provides background information to facilitate a better understanding of the various aspects of the present disclosure. This is a discussion of “related” art. That such art is related in no way implies that it is also “prior” art. The related art may or may not be prior art. The discussion in this section of this document is to be read in this light, and not as admissions of prior art.

[0004] In the United States alone, over 480,000 critically ill children enter hospital intensive-care units ("ICUs") every year. Some of the most delicate patients may be premature neonates weighing as little as 500g. A host of different health issues may emerge and develop very quickly. Early detection may be vital to recovery, and continuous monitoring of vital signs can facilitate that early detection.

[0005] Each year, 300,000 neonates may be admitted to neonatal intensive care units (“NICUs”). About 20,000 may require cardiopulmonary bypass surgery. Of these, up to 75% reveal neurologic injuries when examined under magnetic resonance imaging ("MRI") and up to 20% have seizures. Both of these indicators have been linked to adverse neurodevelopmental outcomes.

[0006] More broadly, the first ten years of may be a critical period for neurodevelopment. During this period, cerebral volume may increase four-fold to support structural and functional development promoted by such processes as neurogenesis, pruning, and myelination. These processes may be believed to affect behavior and cognition as well as metabolism and organ development. Neurodevelopment can be adversely impacted by neurologic diseases such as birth trauma leading to hypoxic ischemic encephalopathy, perinatal stroke, and epilepsy.

[0007] Physiological parameters providing useful and timely indications of health conditions and changes may include electrical activity of cerebral neuron populations (which, among other things, indicates seizure tendencies and brain hemorrhages), brain oxygendistribution (which, among other things, may be a marker of intraventricular hemorrhage, cardiac perfusion, and end-organ injury as well as intracerebral hemorrhage), heart rate, respiratory rate, oxygen content of circulating blood, blood pressure, heart valve motions and other myocardial mechanics, body movement, skin temperature, and core temperature. Historically, however, each of these measurements has required one or more sensors attached to the body, connected by a wire to an electronic support "box" supplying power, data processing, display, and various controls. Some information may be lost if the sensor readings cannot easily be time-synchronized with each other. Each additional sensor takes up more surface space on the patient's body (neonates don't have much to begin with), adds another wire, and may add another box to crowd around the crib or incubator. Such equipment may be both delicate and bulky, forming an obstacle course for clinicians performing care functions and all but preventing medically untrained family members from hugging, patting, handholding, or administering other types of therapeutic or social touch.

[0008] Prolonged wired monitoring may be stressful to the patient. Delirium and reduced physical activity have been observed as side effects, even in adults. Infants, at an age where they would ideally be bonding with their families and learning to control their bodies, may suffer lasting damage under similar conditions. In addition, sensors designed for adults, even if scaled down in size, may be incompatible with neonates' fragile skin and unique anatomy. There may be also a heightened risk of iatrogenic skin injuries, infection, scarring, and pain resulting from the use of standard medical adhesives, electrodes, and pastes on infants' and neonates' skin.

[0009] Medical monitoring solutions suited to the unique needs of neonates and infants can be costly. Design constraints may be more numerous and often involve advanced materials. Regulatory requirements may be stringent, and opportunities for economy of scale may be limited. These high costs may slow down implementation of innovative "baby-friendly" monitoring tools in hospitals and clinics, especially in communities with limited financial resources or technological infrastructure. Conversely, these communities often have a higher incidence of complicated births than their more affluent or technology-infused counterparts simply because prenatal care, doctor-patient communications, and transportation to medical facilities may not be readily available to everyone. The same conditions may prevail in regions temporarily affected by natural or man-made disasters.

[0010] Therefore, a need exists for a time-synchronized multi-parameter monitoring system with sensors that fit neonates' and infants' bodies, may be less likely to damage delicate skin, and allow more freedom of movement for the patient and tactile access for both functional and social interactions with caregivers and family members. Preferably, versions or parts of the system would be deployable inside or outside of the hospital. Locales with a lowdensity of advanced hospitals could then use stand-alone or satellite systems to monitor patients in small clinics, mobile or pop-up clinics, or homes. Preferably, the systems would provide high-quality information at a widely affordable cost of ownership.SUMMARY

[0011] In a first aspect, a system for monitoring physiological parameters, comprising a wearable sensor subsystem, a clock, a sensor-side wireless transceiver, and a monitor subsystem. The wearable sensor subsystem includes an electroencephalography ("EEG") sensor, a near-infrared spectroscopy ("NIRS") sensor, an electrocardiography ("ECG") sensor, a seismocardiography ("SCG") sensor, a chest wall movement ("CWM") sensor, a chest skin temperature ("CST") sensor, a photoplethysmography ("PPG") sensor; and a limb skin temperature ("LST") sensor. The clock attaches a timestamp to a reading of at least one of the sensors. The sensor-side wireless transceiver transmits the reading with the timestamp and receives a sensor command. The monitor subsystem includes: a monitor-side wireless transceiver, user-interface hardware, and a monitor-side computing system. The monitor-side wireless transceiver receives the reading with the timestamp from, and transmits the sensor command to, the sensor-side wireless transceiver subsystem. The user-interface hardware accepts input from, and provides output to, a human user. The monitor-side computing system incorporates hardware and software to read, analyze and display a series of the time-stamped readings.

[0012] In a second aspect, a head-mounted medical monitoring device, comprises a system-on-chip ("SoC") that includes circuits for storage, data processing, sensor control and wireless communication; a biopotential sensor with a pair of electrodes; a spectroscopic sensor with a light source, a first detector, and a second detector; a receptacle for an expansion connector to additional sensors; a flexible circuit board mechanically supporting and communicatively coupling the SoC to the biopotential sensor, the spectroscopic sensor, and the receptacle; a power management unit for distributing power to components coupled to the flexible circuit board, and a soft, flexible encapsulation surrounding at least the SoC, the flexible circuit board, and the power management unit.

[0013] In a third aspect, a method of monitoring multiple health indicators of an infant or neonate patient, comprises: attaching a soft, flexible, lightweight wireless wearable network of sensors to the patient's skin, including a head-mounted module that combines near-infrared spectroscopy ("NIRS") and electroencephalography ("EEG"); supplying power to a system- on-chip ("SoC") programmed to collect time-synchronized data from the sensors and wirelessly transmit the data to at least one monitor; executing a program by which the data is collected and transmitted; allowing the patient to move enough to maintain comfort, and allowing the patient's relatives or caregivers to touch the patient therapeutically or socially,while the data continues to be collected and transmitted; removing the wearable network of sensors without injuring the patient's skin after the data has been collected and transmitted.

[0014] The above presents a simplified summary of the presently disclosed technique as claimed below in order to provide a basic understanding of some aspects of the presently disclosed technique. This summary is not an exhaustive overview of the presently disclosed technique. It is not intended to identify key or critical elements of the presently disclosed technique or to delineate the scope of the presently disclosed technique. Its sole purpose is to present some concepts in a simplified form as a prelude to the more detailed description that is discussed later.BRIEF DESCRIPTION OF DRAWINGS

[0015] The foregoing and other aspects of the present invention are best understood from the following detailed description when read in connection with the accompanying drawings. For the purpose of illustrating the inventions described herein, the drawings may show present. However, it is to be understood that the protectable aspects of the inventions may not be limited to the specific embodiments disclosed. Included in the drawings are the following figures:

[0016] FIG. 1 is a conceptual overview of a system for monitoring physiological parameters according to some embodiments.

[0017] FIG. 2 is a top view of an example set of wearable sensor modules according to some embodiments.

[0018] FIG. 3 is a front view of some examples of monitors that may be included in the monitor subsystem according to some embodiments.

[0019] FIG. 4 is a schematic of a high-level data architecture for a clinical health-change detection engine according to some embodiments.

[0020] FIG. 5 is a diagram showing the disclosed wearable-sensor subsystem and monitoring subsystem embedded in a large institutional IT backend according to some embodiments.

[0021] FIG. 6 shows a sample of time-synchronized multi-sensor data collected by the system according to some embodiments.

[0022] FIG. 7 shows the base of a head-mounted EEG / NIRS module according to some embodiments.

[0023] FIG. 8 is an electronic block diagram of a head-mounted EEG / NIRS module according to some embodiments.

[0024] FIG. 9 shows features of an example encapsulation for a head-mounted EEG / NIRS module according to some embodiments.

[0025] FIG. 10 shows an example of a laminated flexible adhesive web for expansion accessories for a head-mounted EEG / NIRS module according to some embodiments.

[0026] FIG. 11 shows examples of head-mounted EEG / NIRS module in use with and without expansion accessories according to some embodiments.

[0027] FIG. 12 is a flowchart of an example method of monitoring multiple health indicators of an infant or neonate patient according to some embodiments.

[0028] While the disclosed subject matter is susceptible to various modifications and alternative forms, the drawings illustrate specific implementations described in detail by way of example. It should be understood, however, that the description herein of specific examples is not intended to limit that which is claimed to the particular forms disclosed, but on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the appended claims.DETAILED DESCRIPTION

[0029] Illustrative examples of the subject matter claimed below are disclosed. In the interest of clarity, not all features of an actual implementation are described for every example in this specification. It will be appreciated that in the development of any such actual implementation, numerous implementation-specific decisions may be made to achieve the developers’ specific goals, such as compliance with system-related and business-related constraints, which will vary from one implementation to another. Moreover, it will be appreciated that such a development effort, even if complex and time-consuming, would be a routine undertaking for those of ordinary skill in the art having the benefit of this disclosure.

[0030] Unless otherwise defined, all terms including technical and / or scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present disclosure pertains. In addition, unless otherwise defined, all terms defined in generally used dictionaries may not be overly interpreted.

[0031] Multi-mode neurological and cardiopulmonary monitoring systems may be designed for the especially challenging needs of very fragile patients, such as neonates and infants. Compact, lightweight, time-synchronized wireless sensors may be built into wearable modules. The modules' skin-friendly encapsulations and bio-integrated form factors reduce the risk of iatrogenic injury while permitting natural motion and comforting touch.

[0032] The system's integrated sensors include electroencephalography, near-infrared spectroscopy, electrocardiography, seismocardiography, photoplethysmography, motion, and temperature. By combining and analyzing the sensors' time-synchronized readings, the system can calculate heart rate, respiratory rate, pulse oxygenation, temperature, perfusionindex, tidal volume, pulse wave velocity, and pulse arrival time (a surrogate for blood pressure).

[0033] Advanced algorithms can further evaluate sleep patterns and detect arrhythmias, strokes, hemorrhaging and pre-seizure events. Other design features facilitate low cost, ruggedness, and ease of use in both hospital and non-hospital settings.

[0034] A system for monitoring physiological parameters may feature a wearable sensor subsystem. The wearable sensor subsystem may include an electroencephalography ("EEG") sensor, a near-infrared spectroscopy ("NIRS") sensor, an electrocardiography ("ECG") sensor, a seismocardiography ("SCG") sensor, a chest wall movement ("CWM") sensor, a chest skin temperature ("CST") sensor, a photoplethysmography ("PPG") sensor, and a limb skin temperature ("LST") sensor. An on-sensor clock may attach a timestamp to a reading of at least one of the sensors. A sensor-side wireless transceiver may be set up to transmit readings with the timestamp and receive sensor commands.

[0035] The monitor subsystem may include a monitor-side wireless transceiver to receive readings with timestamps from the sensor-side wireless transceiver and transmit sensor commands to the sensor-side wireless transceiver. User-interface hardware may accept input from, and provide output to, a user. A monitor-side computing system may incorporate hardware and software to read, analyze and display a series of the timestamped readings.

[0036] When the wearable sensor subsystem may be arranged to monitor a patient, the EEG sensor and the NIRS sensor may be coupled to the patient's head, while the ECG sensor, the SCG sensor, the CWM sensor, and the CST sensor may be coupled to the patient's chest and the PPG sensor and the LST sensor may be coupled to the patient's limb. If the patient is a neonate, the limb may be a foot. If the patient is an infant, the limb may be a hand or toe.

[0037] The monitor-side computing system may include hardware and software to derive at least one vital sign from the series, examine the series for a warning sign of an unhealthy condition, and notify a clinician if the warning sign is detected. The monitor-side computing system may also include an artificial-intelligence or machine-learning ("AI / ML") model, and the series may be used as training data or test data for the AI / ML model. In the event of a warning sign being detected, the AI / ML model may predict a consequence of the warning sign.

[0038] Some embodiments of the monitor-side computing system may include a central monitoring system communicatively coupled to share data under an interoperable protocol through at least one of a local network and a cloud-based network. Additionally or alternatively, the monitor-side computing system comprises a consumer-grade multi-purpose mobile device programmed to securely receive data from the wearable sensor subsystem.

[0039] A head-mounted medical monitoring device may feature a system-on-chip ("SoC") with circuits for storage, data processing, sensor control and wireless communication; a biopotential sensor with a pair of electrodes; a spectroscopic sensor with a light source, a first detector, and a second detector; a receptacle for an expansion connector to additional sensors; and a flexible circuit board mechanically supporting and communicatively coupling the SoC to the biopotential sensor, the spectroscopic sensor, and the receptacle. A power management unit may distribute power to components coupled to the flexible circuit board, and a soft, flexible encapsulation may surround at least the SoC, the flexible circuit board, and the power management unit.

[0040] In some embodiments, the SoC may be programmed to collect a postsynaptic voltage potential measurement from the electrodes, calculate at least a corresponding electrocortical activity, and wirelessly transmit a result of the calculation. Additionally or alternatively, the SoC may be programmed to collect brain-tissue spectral absorption from the spectroscopic sensor, calculate at least a corresponding regional cerebral oxygenation in both arteries and veins, and wirelessly transmit a result of the calculation.

[0041] An expansion accessory for the device may include an expansion probe integrated in a flexible adhesive web; an expansion connector attached to the flexible adhesive web and detachably engageable with the receptacle; and a stretchable lead integrated in the flexible adhesive web to couple the expansion probe to the expansion connector. The device's pair of electrodes may be arranged to provide two-channel amplitude-integrated electroencephalography ("aEEG"). The expansion accessory may include a set of electrodes arranged such that the device with the accessory may provide 16-channel continuous EEG (cEEG). The SoC selects aEEG or cEEG mode of operation and attaches a mode-identifying tag when wirelessly transmitting aEEG or cEEG data.

[0042] At least part of the expansion accessory may be consumable or disposable. Additionally or alternatively, the expansion accessory or the receptacle may include spring- loaded pins. In some embodiments, the encapsulation and the flexible adhesive web may be removable from human skin using a peel force of 0-3 N / mm for a 90° peel angle.

[0043] In some embodiments, the flexible circuit board may be bendable to a radius of 2- 3cm without plastic deformation. The power management unit includes wireless charging or wireless power.

[0044] A method of monitoring multiple health indicators of an infant may include attaching a soft, flexible, lightweight wireless wearable network of sensors to the infant's skin; including a head-mounted module that combines near-infrared spectroscopy ("NIRS") and electroencephalography ("EEG"); supplying power to a system-on-chip ("SoC") programmedto collect time-synchronized data from the sensors and wirelessly transmit the data to at least one monitor; executing a program by which the data may be collected and transmitted; allowing the patient to move enough to maintain comfort; and allowing the patient's relatives or caregivers to touch the patient therapeutically or socially while the data continues to be collected and transmitted; and finally removing the wearable network of sensors without injuring the patient's skin after the data has been collected and transmitted.

[0045] FIG. 1 is a conceptual overview of a system for monitoring physiological parameters according to some embodiments. System 100 includes wearable sensor subsystem 102 (shown in situ on infant patient 101) and monitor subsystem 103, communicatively coupled by wireless link 104. Wireless link 104 may be a near-field communication ("NFC") link such as Bluetooth® or Bluetooth Low Energy®, a local high- quality link such as WiFi®, a cellular data link of any locally supported generation, a satellite link, or any other suitable wireless communication technology, alternate hospital wireless communications may include, without limitation, wireless medical telemetry service (“WMTS”), which can operate at specified frequencies (e.g., 1.4 GHz). Other wireless communication connections can include wireless connections that operate in accordance with, but are not limited to, IEEE802.11 protocol, a Radio Frequency For Consumer Electronics (“RF4CE”) protocol, ZIGBEE® protocol, and / or IEEE802.15.4 protocol.

[0046] Similarly, wireless link 104 may be simply set up by a sensor-side transceiver and a monitor-side transceiver, or it may include repeaters or entire intervening networks. For security and privacy, the link may be encrypted or otherwise protected from unauthorized access.

[0047] Wearable sensor subsystem 102, as illustrated, may be constituted by lightweight, soft, flexible modules that hold the sensors in gentle contact with skin surfaces of patient 101 . Head-mounted, or “he”, module 105, chest-mounted module 106, and limb-mounted module 107 are shown as examples, but other sensors on other parts of the body may be used in addition or as alternatives. In some embodiments, head-mounted module 105 may include one or more electroencephalography ("EEG") sensors to measure electrical activity in the brain and / or near-infrared spectroscopy ("NIRS") sensors to measure cortical tissue oxygenation. Some embodiments of chest-mounted module 106 may include an electrocardiography ("ECG") sensor, a seismocardiography ("SCG") sensor, chest wall movement ("CWM") sensor, and a chest skin temperature ("CST") sensor. Some embodiments of limb-mounted module 107 may include a photoplethysmography ("PPG") sensor and a limb skin temperature ("LST") sensor. As illustrated, limb-mounted module 107 is attached to patient 101's foot, typical when monitoring a neonate. If patient 101 is older orlarger than is customary for a neonate, limb-mounted module 107 may be coupled to a toe or a hand.

[0048] The multiple readings may be most informative if they are synchronized in time, especially where multiple readings combine to indirectly measure an additional parameter or suggest an underlying cause for a detected event. For example, the chest and limb thermometer readings together may yield a central I peripheral skin temperature differential. The pulse arrival time ("PAT") or pulse transit time ("PPT") between the detection of a pulse by the ECG and / or SCG sensors of the pulse module and the PPG sensor of the limb module may yield a continuous cuffless measurement of diastolic and systolic blood pressure. As another example, if the measured body motion and vocalization decrease, the patient might be falling asleep, or there could be a different cause. If, however, a heart rate, respiratory rate, and brainwave frequencies associated with healthy sleep are observed at the same time, confidence in the conclusion is increased. Time synchronization may involve adding a timestamp from a local clock to the various readings before a sensor-side wireless transceiver (not shown in this figure) transmits them through wireless link 104 to monitor-side computing system 108, or any other suitable process may be used.

[0049] Monitor subsystem 103 may take any of a wide variety of forms from a consumergrade multi-purpose mobile device to a ruggedized portable system to a network of physical or virtual servers linked to a central monitoring system. Each embodiment may run a different set of executable instructions tailored to its capabilities. Monitor-side computing system 108 receives the timestamped readings through a monitor-side wireless transceiver (not shown in this figure), may perform additional calculations, then presents time-synchronized multi-sensor data 111 to a user, such as a clinician or other caregiver, via user interface output 109. Monitor-side computing system 108 also accepts commands from the user through user interface input 110.

[0050] Monitor-side computing system 108 may be as simple as a SoC or chipset embedded in a consumer-grade multi-purpose mobile device running a user-friendly application, as advanced as one or more networked physical or virtual servers running advanced artificial intelligence or machine learning ("AI / ML") models, or any suitable intermediate solution. Time-synchronized multi-sensor data 111 may be raw data, results of calculations on the series of readings from one or more of the sensors in wearable sensor subsystem 102, classifications and predictions from AI / ML models, recalls of previously stored data, or any combination.

[0051] User interface output 109 is illustrated as a desktop display screen, but may optionally include a virtual reality ("VR") headpiece, glasses-mounted heads-up display,projected image, or other display type. It may additionally or alternatively include a non-visual information output, such as a user's earpiece receiving audio updates or a user's wearable haptic device sending signals when, for instance, readings change suddenly or in an unusual way. The latter can be useful if a facility is short-staffed or swamped with concurrent emergencies. Hard copies of important traces may also be printed or plotted if, for example, storage or power are subject to unscheduled downtime. User interface input 110 is illustrated as a keyboard, but may additionally or alternatively include pointing devices such as a mouse, trackball, touchpad, touchscreen, joystick, or gestural wearable; a microphone with speech recognition or a camera with lip-reading or gestural recognition; or any other suitable way to communicate with monitor-side computing system 108 or wearable sensor subsystem 102.

[0052] FIG. 2 is a top view of an example set of wearable sensor modules according to some embodiments. Shown are head-mounted module 205, chest-mounted module 206, limbmounted module 207, and optional expansion accessory 215. All the modules may be soft, flexible, stretchable, and very lightweight. They have distinctly different shapes, making them identifiable at a glance even for a newly trained caregiver. Adhesives, where used, may be selected for low peel strength to make the modules easy to remove without damaging the patient's skin.

[0053] Head-mounted module 205 may include biopotential sensing electrodes 201 (for EEG or other electrically detectable parameter measurements), connect spectroscopic light source 202 and spectroscopic detectors 203 (for NIRS and other oxygen distribution measurements), expansion connector receptacle 204 to connect optional expansion accessory 215, and optionally at least one indicator 208, shown here as an outward-facing visible light but also realizable as an audible alert. Although expansion connector receptacle 204 and indicator 208 are shown as parts of head-mounted module 205, some implementations may include either or both features on chest-mounted module 206 and I or limb-mounted module 207.

[0054] Chest-mounted module 206 may include ECG electrodes 209 to measure heart rate ("HR"), respiratory rate ("RR"), and bioimpedance; high-frequency accelerometer 210 to measure chest-wall movement, heart sounds, vocalization, and body movements; and chest thermometer 211 to monitor skin temperature on the thorax.

[0055] Limb-mounted module 207 may include PPG light source 212, PPG photodetector 213, and limb thermometer 214. Limb-mounted module 107's PPG may supply both SpO2 (arterial oxygen saturation) and arterial tonometry measurements. Alone or in conjunction with chest-mounted module 206, the readings indicate the quality of circulation to the extremities, which can affect the infant's development of strength and coordination in the hands and feet.

[0056] Expansion accessory 215 may include stretchable, breathable, flexible adhesive web 217 and expansion connector 218. Expansion connector 218 may be designed to detachably mate with an expansion connector receptacle on a host module, such as receptacle 204 pictured here on head-mounted module 205. Expansion connector 218 is attached to the conformable material of the adhesive web 217, for example by an adhesive or one or more snap-fit or press-fit features. Like the mating between expansion connector 218 and expansion connector receptacle 204, the attachment of expansion connector 218 to flexible adhesive web 217 may be detachable, allowing re-use of expansion connector 218 with a fresh flexible adhesive web 217 or a different style of, expansion accessory 215.

[0057] Other types of expansion accessory may include different types of expansion connector. Flexible adhesive web 217 incorporates one or more expansion probes 216 to act as extra sensors for the connected module. For example, expansion probes 216 may be electrodes for electrical measurements such as EEG or ECG. Conductive leads 219 couple expansion probes 216 to the expansion connector 218 and thereby to the host module which operates them as extra sensors. In some embodiments, an expansion accessory 215 or part of it may be consumable or disposable.

[0058] For example, a host module such as head-mounted module 205 may operate two built-in electrodes as an amplitude-integrated electroencephalography ("aEEG") sensor, or if the host's on-board processor detects a connected expansion accessory 215, it can instead perform a different type of EEG measurement that makes use of some or all of the expansion probes. If at least 16 electrodes are made available, the host module can measure 16-channel continuous electroencephalography ("cEEG"), which is a common measurement protocol in the US. In other places, standard protocols may use multiples of 10 or 20 electrodes; additional expansion accessories may be designed to meet those needs.

[0059] FIG. 3 is a front view 300 of some examples of monitors that may be included in the monitor subsystem according to some embodiments. Once outside the environment of an optimally resourced hospital, purpose-built medical monitors may be onerously expensive, bulky to store when not in use, difficult to safeguard from damage while moving from place to place, and challenging to upgrade or otherwise modify. In contrast, widely available mass- produced consumer-grade computing systems, mobile or otherwise, have steadily increased performance factors while reducing price. Android® devices are one example, with many potential users already familiar with the operating system.

[0060] Widely available laptop or notebook computer 302, widely available desktop computer 303, widely available large tablet 304, and widely available small tablet or mobile phone 305 each have advantages for different outside-the-hospital environments, and somemay be suited for use inside the hospital as well. Using screen-size-flexible approaches to software development such as responsive design, mobile-first design, emphasis on accessibility, fluid typography, and flexible image formats such as WebP® may facilitate optimized display and navigation of time-synchronized multi-sensor data 311 on a variety of different devices.

[0061] FIG. 4 is a schematic of a high-level data architecture for a clinical health-change detection engine according to some embodiments. In architecture 400, cloud-connected systems 401 collect and share real-time monitored sensor readings 406 to a raw signal database 407 and associated device information with device management server 408. Cloudbased systems 401 are accessed by system administrators through central monitoring system 403. Analytics engine 410 hosts AI / ML models that classify detected events in the raw or processed data and predict outcomes. During training of the model, each layer's weights are iteratively updated to meet a global objective, for example, estimating patient health.

[0062] The global objective allows the model to learn different levels of feature representation as data flows through the layers of the network. For example, once optimized for the global objective, some layers may learn transformations related to simpler signal patterns, while other layers will learn semantic representations of the data. This training process allows superior performance by generating novel features from signal combinations without a priori knowing what signal attributes will be important for achieving the global objective.

[0063] Stored data from the real-time monitoring may be used as training data for the AI / ML model, while new incoming data may be used as test data. If the data collected by monitoring an individual patient includes a warning sign of an unhealthy condition, the model looks up in contacts a clinician or other caregiver associated with the patient. Analytics engine 410 may include one or more optional neural networks or deep-learning models, as well as other types of Al and machine learning. The learning may be supervised, semi-supervised, or unsupervised. Where appropriate, classical non-adaptive algorithms may also run on analytics engine 410.

[0064] Clinicians and other caregivers are contacted over secure bi-directional wireless link 409, which carries alerts or alarms to their local computers, such as laptop or notebook computers 402, large tablets 404, or small tablets or mobile phones 405. In return, the clinicians or other caregivers can send questions and other messages back over secure bidirectional wireless link 409. In addition, information about the caregivers' computers may be tracked and updated by device management server 408. The wireless links in this system maybe similar to any of the versions of wireless link 104 discussed with reference to FIG. 1 , any custom-built alternative, or any suitable type adopted in the future.

[0065] FIG. 5 is a diagram showing the disclosed wearable-sensor subsystem and monitoring subsystem embedded in a large institutional IT backend 500 according to some embodiments. Here, wearable sensor subsystem 502 and monitoring subsystem 503 join a variety of other diagnostic and treatment devices in a network that may operate as a secure Internet (or Intranet) of Things ("loT"). AI / ML server 510 may be a centralized structure as illustrated, or may be a distributed system. AI / ML server 510 may optionally include any alternative aspects of analytics engine 410 as discussed with reference to FIG. 4.

[0066] FIG. 6 shows a sample of time-synchronized multi-sensor data collected by the system according to some embodiments. Referring to Fig. 12A, the plot shows representative time-synchronized readings of NIRS 601 , ECG 610, PPG 603, EEG 606, SCG 609, respiration 612, limb temperature 615, and chest temperature 618 that can be collected from embodiments of the wearable sensor subsystem and displayed, raw or transformed, by embodiments of the monitor subsystem.

[0067] Each of the sensors in the disclosed system measures a core vital sign and at least one advanced parameter derivable from the vital sign. By way of non-limiting example:

[0068] An EEG sensor directly measures cerebral electrical activity, but detailed examination of the readings may also detect, or provide information about, seizures, hemorrhages, and stroke. Here, the noted "sleep spindle" indicates a period of non-rapid-eye- movement ("NREM") sleep

[0069] A NIRS sensor directly measures regional cerebral oximetry (rSO2), but detailed examination of the readings may also detect, or provide information about, hemorrhage and functional localization.

[0070] An ECG sensor directly measures heart rate and respiratory rate, but detailed examination of the readings may also detect, or provide information about, arrhythmia detection, ST segment, QT interval, and heart rate variability.

[0071] A temperature sensor directly measures skin temperature, but detailed examination of the readings may also detect, or provide information about, core body temperature.

[0072] A PPG sensor directly measures SpO2, but detailed examination of the readings may also detect, or provide information about, perfusion index.

[0073] An accelerometer, paired with a gyroscope to function as am internal measurement unit ("IMU"), directly measures chest wall movement, but detailed examination of the readingsmay also detect, or provide information about, vocalizations, respiratory rate, falls, body position and gait.

[0074] A bioimpedance sensor directly measures respiratory rate, but detailed examination of the readings may also detect, or provide information about, tidal volume estimation.

[0075] A time-synchronized (chest) ECG and (limb) PPG sensor directly measure pulse arrival time (a surrogate for continuous blood pressure), but detailed examination of the readings may also detect, or provide information about, pulse wave velocity.

[0076] Further time-synchronized combinations of readings may yield an even wider range of information once AI / ML and other advanced computation tools and techniques sift through large collections of data that includes outcomes. Because these sensors are non- invasive, further applications in which they take the place of invasive tools or methods may be attractive and likely to be pursued.

[0077] FIG. 7 shows the base of a head-mounted EEG / NIRS module according to some embodiments. Time-synchronized NIRS optical absorption measurements and EEG electrical biopotential measurements may reveal a variety of cranial events and conditions, and also reflect developments elsewhere in the body. In some embodiments, head-mounted module base 700 is power-optimized and reusable.

[0078] EEG measures electrical activity, or biopotential, of populations of neurons in the brain. It may provide, for example, detailed information about sleep states or early warning of epileptic seizures. As shown in image (a), head-mounted module base 700 includes a built-in pair of EEG electrodes 701 , and one or more expansion connector receptacles 702. Base 700 can use EEG electrodes 701 to collect aEEG data on its own, or an expansion accessory, such as expansion accessory 215 in FIG. 2, may be coupled to one or both of expansion connector receptacles 702 to add more EEG electrodes. For example, base 700 may operate all the expansion probes of expansion accessory 215 to provide 16-channel cEEG, or select a subset of them to concentrate on a particular area. In some embodiments, the receptacles or their mating connectors may be spring-loaded by including, for example, pogo pins.

[0079] NIRS is a noninvasive optical measurement of regional tissue oxygenation. Besides assessing oxygen saturation of the brain, its use has been expanded to evaluate oxygenation of other tissues as well, and also as an estimate of systemic venous saturation. Besides cerebral oxygenation (ScO2), NIRS can measure cerebral vascular tone, SpO2 similarly to a PPG, and heart rate via local pulse measurements. Base 700 includes multiwavelength NIRS light source 703, NIRS short-path detector 704, and NIRS long-path detector 705. NIRS light source 703 may include one or more LEDs or other suitable light-emittingelements. The wavelengths of most interest for detecting hypoxia, ischemia, and other potentially harmful phenomena in the neonate and infant brain are in or around the 700-900nm range. Narrow bands may be selected within the range: usually at least one red and one infrared wavelength, but more may be chosen.

[0080] The illustration shows one NIRS short-path detector 704 and one NIRS long-path detector 705. However, in some embodiments there may be two short-path detectors spaced equal short distances from light source 703, and two long-path detectors spaced equal long distances from light source 703.

[0081] Because the detectors need to measure faint reflections and backscattering, stray light (i.e., any light that did not traverse the path of interest for the measurement) is preferably excluded. NIRS stray-light excluder 706 may include a black silicone layer or other absorbing material, or one or more deflecting baffles. The windows protecting the detector from the environment may include polydimethylsiloxane ("PDMS") a nonpolar elastomer. Skin-contact surface 707 is a soft, conformable material such as Silbione RTV4420™ by Elkem® and may be wholly or partially coated with a skin-safe, low-peel-force (e.g., 0-3 N / mm for a 90° peel angle) adhesive, such as Mepitac® by Safetac®.

[0082] Image (b) shows the sensor elements of image (a) in use on a patient's forehead. Built-in aEEG electrodes 701 collect electrical signals 708 from groups of cranial neurons while NIRS short-path detector 704 and NIRS long-path detector 705 collect reflected and backscattered light 709 originally emitted by NIRS light source 703.

[0083] FIG. 8 is an electronic block diagram of a head-mounted EEG / NIRS module according to some embodiments. Built-in EEG electrodes 801 and expansion connector(s) 802 may be coupled to EEG analog front-end ("AFE") 806, while NIRS light source 803, NIRS short-path detector 804, and NIRS long-path detector 805 may be coupled to NIRS analog front-end 807. AFEs may be hardware circuits that receive analog signals (such as brainwaves captured by EEG or a variety of other in-vivo physiological signals) and modify them to enter a digital circuit. An AFE may amplify, filter, condition, and digitize a received reading. Because different physiological processes may yield signals that differ dramatically in amplitude and frequency, each type of AFE may be tailored to the defined range of input signal characteristics associated with a specific sensor type and / or measurement process.

[0084] In some embodiments, head-mounted base module 800 may include an accelerometer or I MU 808 for sensing motion and orientation of the patient's body in space. These measurements may also be time-synchronized with the EEG and NIRS measurements. Power management unit ("PMU") 809 supplies power to the system and may use an algorithm to minimize overall power consumption, both to extend battery life and reduce the productionof waste heat. Wireless charging 810 using a bult-in rectifier and regulator, and / or optionally wireless power (not shown in this figure), may also be included in the PMU.

[0085] Indicator 811 may be a selectable-color LED with an infrared option and / or a haptic actuator as illustrated, but may also include an audio source. Wireless communication module 812 is shown here as NFC (e.g., Bluetooth or Bluetooth Low Energy) able to transmit to, and receive from, an NFC reader. Alternatively, however, the design may be adapted to any suitable wireless protocol. Data processing component 813 is shown here as a dual-core microcontroller unit (MCU), but other processors capable of time-synchronizing the sensor readings (e.g., using an internal clock), performing any other desired on-board processing, and controlling the other components such as the sensors (via, for example Serial Peripheral Interface ("SPI") bus 814 or other suitable bus type), AFEs, indicator(s), and wireless transceiver(s). In some embodiments, wireless communication module 812, SPI buses 814, and data processing component 813 are integrated as a system-on-chip (SoC).

[0086] FIG. 9 shows features of an example encapsulation for a head-mounted EEG / NIRS module 900 according to some embodiments. The properties of the encapsulation facilitate a safe and comfortable mounting of the wearable wireless multi-sensor base on a fragile neonate’s skin. As shown in image (a), encapsulation layers 901 may be a soft, conformal but durable RTV such as Silbione RTV4420® by Elkem®, a 2-part silicone with a lifelike feel. Encapsulation filling 902 may be a platinum-catalyzed silicone such as Ecoflex 00-10® by Smooth-on®. At the core of the base, flexible printed circuit board ("FPCB") 903 is flexible to a bending radius of 2.5 cm without plastic deformation. As extremely premature 24-week neonates' foreheads have radii of about 3.5 cm, the head-mounted module is expected to readily fit them without pulling. In image (c), stress-strain graph 904 compares the stress-strain curves for the base device and skin. Image (d) demonstrates the base's twist tolerance 906.

[0087] FIG. 10 shows an example of a laminated flexible adhesive web for expansion accessories for a head-mounted EEG / NIRS module according to some embodiments. This "adhesive stack" may provide a flexible support structure for expansion probes and associated leads like those of FIG. 2.

[0088] Expansion accessory lamination 1003 may be positioned between the bottom of head-mounted base 1001 and patient's skin 1005. When using an expansion accessory, EEG electrodes 1002 may make dry contact with expansion connector top contacts 1004, which may be fabricated from low-impedance conductive film or metal and penetrate through to the bottom of expansion accessory lamination 1003. In effect, they extend the reach of base EEG electrodes 1002 through the adhesive stack.

[0089] The first layer below the top may include silicone gel 1006. Below that is a layer of fluid-resistant foam 1007 over a wicking material 1008. These work together to trap sweat. The skin-facing layer 1009, just above the expansion connector bottom contacts 1010, may include a porous gel such as Mepitac® by Safetac® or any other suitable adhesive safe for fragile neonatal skin.

[0090] In some embodiments there is a lateral displacement "d" between expansion connector top contacts 1004 and expansion connector bottom contacts 1010. Because there is an electrical connection between top conductive film (from which top contacts 1004 are formed) and the bottom conductive film (from which bottom contacts 1010 are formed), a wide range of such displacements may be designed into different expansion connectors. This may be an easy and inexpensive way to vary the effective placement and spacing of the pair of aEEG electrodes on the skin, rather than being limited by the fixed locations of EEG electrodes 1002 in the base.

[0091] In some embodiments, the connector or receptacle may include spring-loaded components, such as pogo pins. If the patient's hair, for example, prevents the adhesive- backed electrode from making acceptable electrical contact, the pins may provide an alternative connection to a traditional wired EEG electrode, which would attach to the patient's skin through a layer of paste. This gives the head-mounted multi-sensor module another way to adapt to the uniqueness of individual patients.

[0092] FIG. 11 shows examples of head-mounted EEG / NIRS module in use with and without expansion accessories according to some embodiments. In image (a), a neonate in an incubator wears a base module 1101 , and may also wear chest and limb modules (not visible here because of the body position), yet is free to move, neither weighed down by bulky sensors nor tethered by wires. Some embodiments may include an incubator monitor 1102 that may fold down or detach from the incubator wall when not in use. In image (b), an infant wears head-mounted base module 1101 (and possibly other sensor modules not in view) while benefiting from therapeutic or social touch that would not be possible with larger, heavier, and / or wired sensors. In image (c), an infant wears an expansion accessory 1103 similar to expansion accessory 215 in FIG. 2. The soft, lightweight, sweat-resistant laminate is comfortable to wear.

[0093] FIG. 12 is a flowchart of an example method 1200 of monitoring multiple health indicators of an infant or neonate patient according to some embodiments.

[0094] At step 1201 , a wireless head-mountable EEG / NIRS module in soft, flexible encapsulation is provided. At step 1202, the module is coupled to a patient's skin as part of a wearable wireless network of sensors.

[0095] At step 1203, power is supplied to the SoC in the wearable network. At step 1204, a program is executed on the SoC.

[0096] At step 1205, time-synchronized data is collected from the sensors per the program's instructions. At step 1206, the data is wirelessly transmitted to at least one monitor per program instructions.

[0097] Concurrently with steps 1205 and 1206, step 1207 allows the patient to move to maintain comfort, and step 1208 lets caregivers touch the patient therapeutically or socially, all while continuing to collect sensor data.

[0098] After all the desired monitoring data is found to have been collected and transmitted in step 1209, the process ends with step 1210, where the wearable network of sensors is removed without injuring the patient's skin.

[0099] The subject matter of the present disclosure is provided as examples of apparatus, systems, methods, circuits, and programs for performing the features described in the present disclosure. However, further features or variations are contemplated in addition to the features described above. It is contemplated that the implementation of the components and functions of the present disclosure can be done with any newly arising technology that may replace any of the above-implemented technologies.

[0100] The detailed description is made with reference to the accompanying drawings and is provided to assist in a comprehensive understanding of various example embodiments of the present disclosure. Changes may be made in the function and arrangement of elements discussed without departing from the spirit and scope of the disclosure. Various embodiments may omit, substitute, or add various procedures or components as appropriate. For instance, features described with respect to certain embodiments may be combined in other embodiments. In addition, descriptions of well-known functions and constructions may be omitted for clarity and conciseness. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the examples described herein can be made without departing from the spirit and scope of the present disclosure.

[0101] Various modifications to the disclosure will therefore be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other variations without departing from the spirit or scope of the present disclosure. Throughout the present disclosure the terms “example,” “examples,” or “exemplary” indicate examples or instances and do not imply or require any preference for the noted examples. Thus, the present disclosure is not to be limited to the examples and designs described herein but is to be accorded the widest scope consistent with the principles and novel features disclosed.

Claims

CLAIMSWhat is claimed is:

1. A system for monitoring physiological parameters, comprising: a wearable sensor subsystem including: an electroencephalography ("EEG") sensor; a near-infrared spectroscopy ("NIRS") sensor; an electrocardiography ("ECG") sensor; a seismocardiography ("SCG") sensor; a chest wall movement ("CWM") sensor; a chest skin temperature ("CST") sensor; a photoplethysmography ("PPG") sensor; and a limb skin temperature ("LST") sensor; a clock to attach a timestamp to a reading of at least one of the sensors; a sensor-side wireless transceiver to transmit the reading with the timestamp and receive a sensor command; and a monitor subsystem including; a monitor-side wireless transceiver to receive the reading with the timestamp from, and transmit the sensor command to, the sensor-side wireless transceiver subsystem; user-interface hardware to accept input from, and provide output to, a human user; a monitor-side computing system incorporating hardware and software to read, analyze and display a series of the time-stamped readings.

2. The system of claim 1 , wherein, when the wearable sensor subsystem is arranged to monitor a patient: the EEG sensor and the NIRS sensor are mounted on the patient's head; the ECG sensor, the SCG sensor, the CWM sensor, and the CST sensor are mounted on the patient's chest; and the PPG sensor and the LST sensor are mounted on the patient's limb.

3. The system of claim 2, wherein the patient is a neonate and the patient's limb is a foot.

4. The system of claim 2, wherein the patient is an infant and the patient's limb is a hand or toe.

5. The system of claim 1 , wherein the monitor-side computing system further comprises hardware and software to: derive at least one vital sign from the series, examine the series for a warning sign of an unhealthy condition; and notify a clinician if the warning sign is detected.

6. The system of claim 5, wherein the monitor-side computing system further comprises an artificial-intelligence or machine-learning ("AI / ML") model, and uses the series as training data or test data for the AI / ML model.

7. The system of claim 6, wherein the warning sign is detected, and a consequence of the warning sign is predicted, by the AI / ML model.

8. The system of claim 1 , wherein the monitor-side computing system comprises a central monitoring system communicatively coupled to share data under an interoperable protocol through at least one of a local network and a cloud-based network.

9. The system of claim 1 , wherein the monitor-side computing system comprises a widely available consumer-grade multi-purpose mobile device programmed to securely receive data from the wearable sensor subsystem.

10. A head-mounted medical monitoring device, comprising: a system-on-chip ("SoC") with circuits for storage, data processing, sensor control and wireless communication; a biopotential sensor with a pair of electrodes; a spectroscopic sensor with a light source, a first detector, and a second detector; a receptacle for an expansion connector to additional sensors; a flexible circuit board mechanically supporting and communicatively coupling the SoC to the biopotential sensor, the spectroscopic sensor, and the receptacle; a power management unit for distributing power to components coupled to the flexible circuit board, and a soft, flexible encapsulation surrounding at least the SoC, the flexible circuit board, and the power management unit.

11. The device of claim 10, wherein the SoC is programmed to collect a postsynaptic voltage potential measurement from the electrodes, calculate at least a corresponding electrocortical activity, and wirelessly transmit a result of the calculation.

12. The device of claim 10, wherein the SoC is programmed to collect brain-tissue spectral absorption from the spectroscopic sensor, calculate at least a corresponding regional cerebral oxygenation in both arteries and veins, and wirelessly transmit a result of the calculation.

13. The device of claim 10, further comprising an expansion accessory including: an expansion probe integrated in a flexible adhesive web; an expansion connector attached to the flexible adhesive web and detachably engageable with the receptacle; and a stretchable lead integrated in a flexible adhesive web to couple the expansion probe to the expansion connector.

14. The device of claim 13, wherein: the pair of electrodes is arranged to provide two-channel amplitude-integrated electroencephalography ("aEEG"), the expansion accessory comprises a plurality of electrodes arranged such that the device with the accessory provides 16-channel continuous EEG (“cEEG”); and the SoC selects aEEG or cEEG mode of operation and attaches a mode-identifying tag when wirelessly transmitting aEEG or cEEG data.

15. The device of claim 13, wherein at least part of the expansion accessory is consumable or disposable.

16. The device of claim 13, wherein the expansion accessory or the receptacle includes spring-loaded pins.

17. The device of claim 13, wherein the encapsulation and the flexible adhesive web are removable from human skin using a peel force of 0-3 N / mm for a 90° peel angle.

18. The device of claim 10, wherein the flexible circuit board is bendable to a radius of 2- 3cm without plastic deformation.

19. The device of claim 10, wherein the power management unit includes wireless charging or wireless power.

20. A method of monitoring multiple health indicators of an infant or neonate patient, comprising: attaching a soft, flexible, lightweight wireless wearable network of sensors to the patient's skin, including a head-mounted module that combines near-infrared spectroscopy ("NIRS") and electroencephalography ("EEG"); supplying power to a system-on-chip ("SoC") programmed to collect time-synchronized data from the sensors and wirelessly transmit the data to at least one monitor; executing a program by which the data is collected and transmitted; allowing the patient to move enough to maintain comfort, and allowing the patient's relatives or caregivers to touch the patient therapeutically or socially, while the data continues to be collected and transmitted; removing the wearable network of sensors without injuring the patient's skin after the data has been collected and transmitted.

Images