Patent Grant
11759142
The invention relates generally to the collection of neurophysiology and neurovascular data from a wearable device for use in identifying, predicting, and treating neurodegeneration.
Interest in brain fluid transport systems has risen rapidly in recent years with the discovery of the glymphatic system and its role in brain protein clearance involved in neurodegeneration such as amyloid-β. Inhibition of glymphatic flow accelerates protein accumulation and cognitive decline in animal models of Alzheimer's disease, traumatic brain injury and Parkinson's disease. The glymphatic flow is primarily active during sleep and driven by cerebrovascular arterial pulsations, thus, sleep, cerebrovascular integrity and neurovascular coupling are required for clearance of waste products that build up in the awake brain. A reduction in glymphatic flow results in accumulation of protein in the brain that leads to neurodegeneration (known as brain proteinopathy) and which can be detected with neuroimaging or molecular analyses of cerebrospinal fluid or blood plasma. Accordingly, non-invasive techniques for monitoring and analyzing information associated with glymphatic flow and its clearance of protein, with a focus on monitoring the mechanism of glymphatic flow that causes the protein accumulation rather than just reporting the accumulation of protein in the brain as done by neuroimaging or molecular analysis, can be useful in diagnosing and treating neurological degeneration.
The present disclosure is generally directed to collecting neurophysiology and neurovascular data for use in identifying, predicting, and treating neurodegeneration. In one example embodiment, the disclosure is directed to a method, implemented with one or more computer processors, comprising: (a) accessing, by the one or more computer processors, neurophysiological data and neurovascular data recorded during sleep; (b) executing, by the one or more computer processors, a function mapping from said neurophysiological data and neurovascular data to a target that is a marker of glymphatic flow; and (c) outputting, by the one or more computer processors, a target prediction model based on the function mapping.
In another example embodiment, the disclosure is directed to a system comprising: (a) one or more computer processors; (b) a neurophysiological data acquisition module configured to measure neurophysiological data; (c) a neurovascular data acquisition module configured to measure neurovascular data; and (d) a transmission module configured to transmit the electroencephalogram data and the neurovascular data to a second computing device.
The foregoing embodiments are non-limiting examples and other aspects and embodiments will be described herein. The foregoing summary is provided to introduce various concepts in a simplified form that are further described below in the detailed description. This summary is not intended to identify required or essential features of the claimed subject matter nor is the summary intended to limit the scope of the claimed subject matter.
The specific features, aspects and advantages of the present invention will become better understood with regard to the following description and accompanying drawings where:
Referencing the glymphatic flow responsible for the clearance of proteins described above, the perivascular spaces (PVS) of the brain are central to the glymphatic fluid transport system. Fluorescent dye-labeled particles that enable visualization of the perivascular spaces reveals that the system comprises periarterial, pericapillary, and perivenular spaces all interconnected into a single network. The glymphatic flow system clears brain interstitial fluid (ISF) waste products such as the proteinopathy proteins through the periarteriolar and perivenular spaces to meningeal lymphatics. The PVS is a site of great importance that is affected by disease processes such as cerebral amyloid angiopathy characterized by perivascular amyloid beta deposits and present in over 90% of Alzheimer's disease brains. PVS accumulation of p-tau is also observed in Alzheimer's disease and has been demonstrated to be present in arteries, arterioles and veins in the brain. In the glymphatic flow system, cerebrospinal fluid (CSF) enters the periarterial spaces, running in the direction of the blood flow and propelled by arterial wall pulsations. CSF mixes with ISF facilitated by aquaporin-4 (AQP4) water channels that are present at the vascular astrocytic endfeet forming the outer wall of the perivascular spaces. During sleep, arterial pulsations drive CSF bulk flow towards the brain. This process has been shown to be highly dependent on the arousal state and AQP4 expression. With modern neuroimaging techniques, it is now possible to visualize CSF flow and its clearance of proteins in the intact CNS by the second-to-second pulsatile movement of CSF through the ventricular system occurring with each cardiac contraction. Respirations further contribute to this rhythm, adding pulsations at a lower frequency. Superimposed on these acute pulsations is a circadian or diurnal rhythm with maximum CSF production occurring at night during sleep and minimum occurring in the afternoon. MRI studies in animal models using CSF tracers reveal that the glymphatic flow from ISF to subarachnoid CSF is enhanced during NREM sleep.
Sleep has profound effects on most aspects of brain physiology, including glymphatic flow and brain blood volume. In animal models, the amplitude of brain blood volume in non-rapid-eye movement (NREM) sleep doubles compared to the awake state and the amount of total blood volume in the brain changes significantly in the transition from wake to sleep and back to wake, showing that brain blood volume dynamics is coupled to arousal state. These changes in blood volume during sleep are linked to variations in neural activity which drive fluctuations in the diameter of arterioles. The strength of this neurovascular coupling changes with arousal state, increasing substantially during sleep compared to wakefulness and explains the increase in brain blood volume during sleep when neural activity is lower than in the wake state.
The concentration of amyloid-β and tau in the interstitial space and CSF follow a diurnal pattern with both protein concentrations reaching their peak during wakefulness and trough during sleep. A single night of sleep deprivation results in a significant increase in soluble amyloid-β in the brain ISF demonstrated with positron emission tomography (PET) using an amyloid-binding radiotracer. The net concentration of amyloid-β, tau and other proteins in the CSF and ISF reflect the combined effects of diurnal- and state-dependent variations in CSF production rate, ISF volume, ISF turnover rate, and glymphatic flow. Sleep slow-wave delta waves are associated with glymphatic flow and the amplitudes of slow delta waves responsible for effective waste removal during sleep. Sleep deprivation results in stalled glymphatic flow and the accumulation of waste solutes including amyloid-β solutes and other proteins in the brain.
During sleep, glymphatic flow is increased by the action of astrocytes using AQP4 channels that are responsible for the modulation of phases between sleep and awake states. Using membrane channel protein AQP4 and water flux through AQP4, astrocytes can shrink the cell volume of brain parenchyma to increase ISF space by 23% during sleep enabling water and solutes to diffuse into adjacent perivascular spaces. During slow-wave sleep, the increase in ISF space, the opening of gap junction of astrocyte end-feet and an increase in cerebral hemodynamics creates an environment for exchange of nutrients and elimination of waste between ISF and CSF.
Damage to molecular and cellular components of the glymphatic flow system, to the brain neurovascular integrity and to its neurovascular coupling that occurs with aging, chronic co-morbidities, neurodegenerative diseases and physical brain trauma is closely linked to an increase of tau and amyloid-β in extracellular fluids. The spread of these aggregated proteins in the brain parenchyma leads to worsening clinical symptoms and decline in cognition. Measuring the functional integrity of this system and the pathological accumulation of proteins in the brain ISF requires invasive procedures such as PET scans with radiotracers that bind to the protein of interest or lumbar puncture for CSF extraction and analysis of protein concentrations. The cost and complexity of these procedures limits their use to screen for brain proteinopathies or to monitor their progression. Without access to these tests, practitioners today use clinical assessment and cognitive evaluations to screen, diagnose and monitor neurodegenerative disorders such as Alzheimer's disease. The opportunity to intervene early in the 10-to-20-year pre-clinical period of these disorders is lost as is the ability to evaluate promising new treatments that could slow or stop disease progression. Another challenge with existing invasive test procedures is that they must be done in a hospital or clinic setting by trained technicians. This poses additional patient burden that further reduces the effectiveness of these expensive and time-consuming invasive procedures for monitoring progression of proteinopathy, adjusting care plans and monitoring response to new treatments.
The glymphatic system of protein waste clearance operates during sleep and is dependent on features measurable with sleep electroencephalogram (EEG) and on features that measure neurovascular blood volume, pulsation and coupling. Short term disruption in sleep causes an increase in CSF and ISF waste proteins and long-term disruption leads to formation of ISF protein aggregates with neurodegeneration. A wearable device that can measure the relevant EEG features, brain blood volume changes and arterial pulsations during sleep, transmitting the data to a method and system that can predict CSF and ISF waste protein levels and ISF protein aggregation as can be measured using an existing invasive clinical procedure, would provide a significant advantage to medical screening and monitoring of neurodegenerative proteinopathy and help select appropriate interventions that slow or stop protein accumulation and neurodegeneration. As used herein and as described further below, a “wearable device” is a device worn by a subject comprising one or more sensors for collecting neurophysiological and/or neurovascular data from the subject, a processor, memory, and storage device for storing the collected data, and at least one wireless transmission module enabling the wearable device to transmit the collected data to another computing device. The wireless communication capability of the wearable device allows the wearable device to be self-contained so that it requires no external wiring to other equipment. The wearable device is configured to be worn by the subject while sleeping.
According to the embodiment(s) of the present disclosure, various views are illustrated in
The following detailed description contains many specifics for the purpose of illustration. Any one of ordinary skill in the art will appreciate that many variations and alterations to the following details are within scope of the disclosure. Accordingly, the following example embodiments are set forth without any loss of generality to, and without imposing limitations upon, the claimed invention.
Example embodiments of the present disclosure comprise a system and method that enables non-invasive measurement of brain proteinopathy and neurodegeneration. Such measurements are typically done in a medical facility using sophisticated equipment and involve a molecular analysis of cerebrospinal fluid or neuroimaging, and include proteinopathy proteins amyloid, tau or alpha-synuclein levels in a blood plasma or CSF assay, brain amyloid or tau load on a PET scan, brain atrophy on a structural brain MRI, repeated diurnal assessments to measure net change in proteinopathy protein CSF assay or soluble proteinopathy protein burden from neuroimaging. Example embodiments of the present disclosure teach a novel system and method that acquires sleep neurophysiology data synchronized with neurovascular data. The neurophysiology data includes electroencephalogram (EEG) data. The neurovascular data includes continuous transcranial impedance plethysmography (IPG), carotid pulse transit time (PTT), heart rate variability (HRV) and resting heart rate (RHR). The example embodiments can further perform the step of acquiring one of said molecular analyses or neuroimaging test of neurodegeneration following sleep and learning a function mapping from the sleep neurophysiology and neurovascular data to a molecular analyses or neuroimaging marker of neurodegeneration. Learning the function mapping includes using a loss function to determine relevant features in the sleep data, identifying a set of optimal weights that produce a minimum of the loss function, and creating a function mapping, also referred to as a brain proteinopathy and neurodegeneration prediction model, using the optimal weights. The example embodiments can further include performing the step of applying the learned function mapping to new sleep neurophysiology and neurovascular data to screen for, or monitor, said molecular analyses or neuroimaging marker of neurodegeneration. The example embodiments can further include delivering a therapeutic intervention that targets an aspect of sleep neurophysiology or neurovascular either preceding sleep or during sleep, measuring target engagement or effect on the sleep neurophysiology and neurovascular, and further performing the step of applying the learned function mapping to predict the therapeutic effect on brain proteinopathy and neurodegeneration.
Another embodiment of the present disclosure comprises a system and method that enables a person to monitor the effect of a cardiovascular intervention or activity on a molecular analysis or neuroimaging marker of neurodegeneration, which embodiment teaches a novel system and method for performing the step of recording a cardiovascular intervention or activity data that precedes sleep, further performing the step of recording neurophysiology and neurovascular data during sleep and applying a learned function mapping to predict the therapeutic effect on brain proteinopathy and neurodegeneration of the cardiovascular intervention or activity.
Another embodiment of the present disclosure comprises a system and method that enables a person to monitor the effect of a pharmaceutical or neuromodulation intervention on a molecular analysis or neuroimaging marker of neurodegeneration, which embodiment teaches a novel system and method for performing the step of recording a pharmaceutical or neuromodulation intervention either preceding sleep or during sleep, further performing the step of recording neurophysiology and neurovascular data during sleep and applying a learned function mapping to predict the therapeutic effect on brain proteinopathy and neurodegeneration of the pharmaceutical or neuromodulation intervention.
Another embodiment of the present disclosure comprises a system and method that enables a person to monitor the effect of a diet intervention on a molecular analysis or neuroimaging marker of neurodegeneration, which embodiment teaches a novel system and method for performing the step of recording a diet intervention that precedes sleep, further performing the step of recording neurophysiology and neurovascular data during sleep and applying a learned function mapping to predict the therapeutic effect on brain proteinopathy and neurodegeneration of the diet intervention.
Another embodiment of the disclosure can be a wearable system comprising a wearable computer with sensors and a wireless communication interface to a mobile computer supporting a wireless network interface which communicates with a second computer including a network interface, each computer further including a processor, a memory unit operable for storing a computer program, an input mechanism operable for inputting data into said computer system, an output mechanism for presenting information to a user, a bus coupling the processor to the memory unit, input mechanism and output mechanism, wherein the wearable system includes various executable program modules stored thereon where when executed are operable to perform functions. The wearable computer with sensors can comprise a neurophysiology data acquisition module with sensors capable of recording EEG, a neurovascular data acquisition module with sensors capable of recording neurovascular data, with the data stored on the wearable computer where when executed records to the memory unit of said wearable computer the neurophysiology and neurovascular sensor acquisition data. A transmission module can comprise a wireless network interface (e.g., a Bluetooth or WiFi radio) as well as instructions stored on said wearable computer where when executed transmit through the wireless network interface the recordings stored in the memory unit to said mobile computer. A second transmission module can also be stored on said wearable computer and comprise a wired network interface as well as instructions where when executed transmit through a bus network interface the recordings stored in the memory unit to a local computer station that further transmits the data to a second computer through a wireless network interface. A learning module can be stored on said second computer where when executed learns a function mapping from said transmitted recording to a molecular analysis or neuroimaging marker of neurodegeneration stored on said second computer, uses a loss function to determine relevant features in said recording, identifies a set of optimal weights that produce a minimum for said loss function, and creates said function mapping using said optimal weights. A brain proteinopathy and neurodegeneration prediction module can also be stored on said second computer where when executed applies the learned function mapping, also referred to as the brain proteinopathy and neurodegeneration prediction model, to a new transmitted recording of sleep neurophysiology and neurovascular sensor data from the wearable computer with sensors, to calculate a predicted value for the molecular analysis or neuroimaging marker of neurodegeneration from the new transmitted recordings. Thus, the brain proteinopathy and neurodegeneration prediction model can determine a predicted brain proteinopathy and neurodegeneration value from the new recording of data without requiring the invasive imaging or assay procedures described previously.
Another embodiment of the disclosure can be a wearable system comprising a wearable computer with sensors and a wireless communication interface to a mobile computer supporting a wireless network interface which communicates with a second computer including a network interface, each computer further including a processor, a memory unit operable for storing a computer program, an input mechanism operable for inputting data into said computer system, an output mechanism for presenting information to a user, a bus coupling the processor to the memory unit, input mechanism and output mechanism, wherein the wearable system includes various executable program modules stored thereon where when executed are operable to perform functions. In one embodiment, the wearable system can comprise a data acquisition module with sensors capable of recording cardiovascular activity data, with said data stored on the wearable computer where when executed records to the memory unit of said computer the acquisition data. In another embodiment, the wearable system can comprise a data input module capable of recording diet and nutrition data, with said data stored on the mobile computer where when executed records to the memory unit of said computer the data. In another embodiment, the wearable system can comprise a data input module capable of recording drug or neurostimulation data, with said data stored on the mobile computer where when executed records to the memory unit of said computer the data. In any of the preceding embodiments, a transmission module can comprise a communication interface and associated instructions stored on said wearable or mobile computer where when executed transmit through the wireless network interface the recordings stored in the memory unit to said second computer. In any of the preceding embodiments, a module can be stored on said second computer where when executed applies a learned function mapping to said transmitted recording of sleep neurophysiology and neurovascular sensor data from the wearable computer with sensors, to calculate a value for a molecular analysis or neuroimaging marker of neurodegeneration from the transmitted recordings.
The details of the invention and various embodiments can be better understood by referring to the figures of the drawing.
Referring to
The example embodiment of
In contrast, the brain proteinopathy and neurodegeneration target data acquisition module 800 is an invasive module typically implemented using sophisticated equipment at a medical facility and is responsible for acquiring one or more molecular analysis or neuroimaging markers of neurodegeneration. The marker of interest can be a CSF or plasma assay of one of several brain proteinopathy proteins including amyloid beta 42, tau, alpha-synuclein and neurofilament light that are markers of neurodegenerative disorders, a neuroimaging scan such as PET scan with a radiotracer for the protein of interest, an MM with contrast agent to measure brain atrophy. The CSF assay or PET scan can be taken after sleep to assay an existing level of CSF protein or protein build up in the brain, or taken both before and after sleep to assay the net change in the protein level that occurred during sleep. The marker can also be an MRI with intrathecal contrast agent injected into the CSF prior to sleep and which is cleared via brain interstitial fluid transport into the meningeal lymphatic and cavernous sinus system during sleep, and the uptake and clearance rate measurable using repeat Mill scanning.
The target marker data measured by module 800 is used by the target learning module 900 to create a target prediction module 1000 that maps new neurophysiology data from module 600 and new neurovascular data from module 700 to a predicted molecular or neuroimaging marker of neurodegeneration for longitudinal monitoring without needing to repeat the invasive measurement of module 800. The target learning module 900 and target prediction module 1000 can be implemented as software that uses a machine learning or deep learning algorithm to provide an output. The target learning module 900 and target prediction module 1000 can be implemented as software executed on one or more remote computers such as the second computer described previously. Target prediction module 1000 can also map sleep neurophysiology data from module 600 and sleep neurovascular data from module 700 taken from a new patient to a molecular or neuroimaging marker of neurodegeneration for non-invasive screening that can be used in medical decision making to evaluate the need for the invasive measurement of module 800. The target engagement module 1100 measures the effect that a putative therapeutic intervention has on the sleep neurophysiology or neurovascular data, or degree of target engagement, and how that effect translates into a change in the predicted molecular or neuroimaging marker of neurodegeneration. The intervention can be one of cardiovascular, diet, sleep, pharmacological or neurostimulation.
Referring now to
One or more of the EEG data elements illustrated in
Referring to
During the same night and for the same individual, simultaneous recordings were made using an ear device comprising modules 100, 200 and 300 illustrated in
Referring to
For the same individual during the same night of EEG recording,
One or more of the neurovascular data elements illustrated in
Referring to
An interval of cerebral perfusion pulsation during sleep for this patient is illustrated in 732 and the corresponding ECG signal in 734. The cerebral perfusion stroke volume is the difference in impedance between the peak and the trough of a pulsation. A larger difference signifies a greater change in fluid, or water, composition resulting from the heart beat and correlates directly with cerebral stroke volume.
Referring to
Referring to
In another embodiment 830, a PET scan can be used with a radiotracer specific for a particular protein of interest and relative standard uptake value (SUVr) that quantifies the proteinopathy build up in the brain. In another embodiment, the radiotracer binds to a soluble form of the proteinopathy protein and the PET scan SUVr quantifies the change in concentration of the soluble protein following sleep.
In another embodiment, MM 850 can be used to measure changes in brain volume, or degree of neurodegeneration, in anatomical regions of interest such as the hippocampus, entorhinal cortex, thalamus, orbitofrontal, parietal, temporal, anterior, and posterior cingulate and the precuneus regions over two time points separated by several months to provide an target rate of neurodegeneration for an individual.
In another embodiment 870, spinal fluid concentration of a proteinopathy protein is assayed either in the morning or both before and after sleep. The absolute concentration levels normed to a population or the diurnal difference in concentration before and after sleep provide a target molecular marker for that protein that measures glymphatic flow as net clearance of the protein and also serves as a marker of brain proteinopathy and neurodegeneration.
In another embodiment 890, plasma concentration of a proteinopathy protein is assayed either in the morning or both before and after sleep. The absolute concentration levels normed to a population or the diurnal difference in concentration before and after sleep provide a target molecular marker for that protein that measures glymphatic clearance function as net glymphatic clearance of the protein and also serves as a marker of brain proteinopathy and neurodegeneration.
The target marker data acquisition module 800 can be implemented using one of embodiment 810, 830, 850, 870, or 890. Alternatively, the target marker data acquisition module 800 can be implemented using a plurality of the embodiments illustrated in
Referring to
In 903, a model and fitting procedure are selected. The model can be one of many known machine learning and deep learning models, including those used in commercially available software packages and services, and the fitting procedure is selected based on the model selected. Random forest, a general-purpose machine learning model, works very well to identify the patterns and ranges of the neurophysiology and neurovascular feature values that best separate, or predict, the target glymphatic flow, molecular or neuroimaging markers. Acquiring the patient training data is expensive and requires an investigational review board approved clinical study. Random forest has the advantage over deep learning models in that it provides good results on a relatively small number of patient data examples. Once the model is learned on a population of patients representative of the target application, predictions of brain glymphatic flow, proteinopathy or neurodegeneration in new, unseen patients can be made from the neurophysiology and neurovascular sleep data acquired by the device illustrated in
Once a model such as random forest is selected, the neurophysiology and neurovascular data are input features and the target marker data is the target output in the model cross validation 905. The patient data of 901 is first randomly split into a number of folds, such as five non-overlapping sets. The model in 903 is fit to the data of the first four folds and the fitted model is used to predict the target marker data of the patients in the 5th fold. This procedure is repeated four more times, each time leaving out a new fold of patients and their data, resulting in model predictions of the complete target marker data from the input neurophysiology and neurovascular feature data. In 907 the model is evaluated using a goodness-of-fit metric that can be the root-mean-square between the prediction in 903 and the target marker data. Operations 901-907 can be repeated using several different models to determine which provides the most accurate prediction of the target marker data. The best target prediction model is promoted to 909 where holdout patients and their data, including target marker data that has never been seen by the model, is tested and the prediction model and its performance on the holdout data is outputted in 911.
The target prediction can be a predicted value for glymphatic flow, a level of a protein accumulation such as those described previously or a degree of neurodegeneration measured as changes in whole-brain volume or regional volume using structural MM. The predictions can be continuous values of marker values used to monitor for disease progression or response to intervention, or categorical values to determine the presence or absence of a disease stage. The target predictions of disease stage classification can be submitted for regulatory approval with the appropriate regulatory validation study and reporting of sensitivity, specificity, positive and negative predictive values compared to a gold-standard target such as a molecular analysis or neuroimaging test.
The prediction model 911 can be executed using data inputs 1110 and 1130 to output a prediction of the therapeutic intervention's expected impact on glymphatic flow, brain proteinopathy or neurodegeneration in the short term or if sustained in the long term. The glymphatic flow, brain proteinopathy or neurodegeneration prediction that is output can be used to show the effect of the intervention on the glymphatic flow, molecular or neuroimaging marker of neurodegeneration. In other embodiments, operation 1110 can record cardiovascular, diet, pharmaceutical or neuromodulation interventions, which data is then input into 1130 for execution computation of target engagement at operation 1150. As such, the effect of the interventions on the neurophysiology and neurovascular data can be ascertained and using the prediction model 911 the intervention's effect on a marker of neurodegeneration determined.
Referring to
In another embodiment, an input intervention can be a cardiopulmonary intervention such as a Valsalva maneuver or hyperventilation. The transcranial impedance tracing following these two cardiopulmonary interventions are illustrated in 1136 and 1138. In 1136 the patient performed a 45 second Valsalva maneuver starting at the 1-minute mark. The Valsalva maneuver increases intrathoracic pressure thereby decreasing venous return from the head. The decrease in venous return results in venous engorgement and increase in cerebral fluid volume. This is demonstrated by the decrease in the transcranial impedance during the Valsalva in tracing 1136. In 1138, the same patient started hyperventilating lightly at 1 minute and progressed to full hyperventilation at 1 minute and 15 seconds. Each inspiration decreases intrathoracic pressure and increases venous return, causing a decrease in cerebral fluid volume and an increase in transcranial impedance, as shown in tracing 1138. In both these interventions, monitoring a neurovascular tracing demonstrated target engagement of the intervention with the neurovascular measure, in this case cerebral fluid volume.
Referring to
In addition to the forgoing, the various embodiments of the present disclosure include, but are not limited to, the embodiments set forth in the following clauses.
Clause 1. A method implemented with one or more computer processors, the method comprising:
Clause 2. The method of clause 1, further comprising:
Clause 3. The method of clause 1, wherein the neurophysiological data are electroencephalogram recordings, and sleep macrostructure and sleep microstructure features are extracted from the electroencephalogram recordings.
Clause 4. The method of clause 1, wherein the neurovascular data comprise one or more of electrical impedance recordings taken from a patient's head, photoplethysmography measured at the patient's head, and inertial measurement unit acceleration measured in a patient's ear, wherein the neurovascular data are used to compute sleep cerebral fluid volume changes, cerebral perfusion pulsation, heart rate variability, resting heart rate, pulse transit time, pulse wave velocity and respiratory rate.
Clause 5. The method of clause 1, wherein the neurophysiological data and neurovascular data are acquired from a wearable device.
Clause 6. The method of clause 5, wherein sensors of the wearable device are inserted into an ear of a patient and measurements are taken from the ear or ear canal.
Clause 7. The method of clause 6, wherein the wearable device has a bladder that is pressurized inside the ear canal to increase the interfacial surface contact of the sensors to the canal a wall of the ear canal.
Clause 8. The method of clause 2, wherein the molecular analysis marker of neurodegeneration is a CSF or blood plasma assay of one of β-amyloid, tau, p-tau, α-synuclein, and neurofilament light.
Clause 9. The method of clause 2, wherein the neuroimaging marker of neurodegeneration is a PET scan with a radiotracer that binds to one of β-amyloid, tau, or glucose.
Clause 10. The method of clause 2, wherein the neuroimaging marker of neurodegeneration is an MM scan.
Clause 11. The method of clause 1, further comprising:
Clause 12. The method of clause 1, further comprising:
Clause 13. A system comprising:
Clause 14. The system of clause 13, wherein the one or more computer processors, the neurophysiological data acquisition module, the neurovascular data acquisition module, and the transmission module are disposed in a wearable device.
Clause 15. The system of clause 14, wherein the wearable device is configured to attach to an ear of a patient.
Clause 16. The system of clause 15, wherein sensors of the wearable device are inserted into the ear and measurements are taken from the ear or ear canal.
Clause 17. The system of clause 16, wherein the wearable device comprises a bladder that is pressurized inside the ear canal to increase interfacial surface contact of the sensors to a wall of the ear canal.
Clause 18. The system of clause 13, wherein the neurophysiological data are electroencephalogram recordings, and sleep macrostructure and sleep microstructure features are extracted from the electroencephalogram recordings.
Clause 19. The system of clause 13, wherein the neurovascular data comprise one or more of electrical impedance recordings taken from a patient's head, photoplethysmography measured at the patient's head, and inertial measurement unit acceleration measured in a patient's ear, wherein the neurovascular data are used to compute sleep cerebral fluid volume changes, cerebral perfusion pulsation, heart rate variability, resting heart rate, pulse transit time, pulse wave velocity and respiratory rate.
Clause 20. The system of clause 13, wherein the second computing device is one of a mobile telephone, a local computing device, and a remote computing device.
Clause 21. The system of clause 13, wherein the second computing device comprises a machine learning algorithm configured to input the neurophysiological data and the neurovascular data and configured to output a target prediction model.
Clause 22. The system of clause 21, wherein the target prediction model is configured to receive as input new neurophysiological data and new neurovascular data and configured to output a predicted marker of neurodegeneration.
Clause 23. The system of clause 21, wherein the target prediction model is configured to receive as input intervention data, new neurophysiological data, and new neurovascular data, and configured to output a predicted marker of brain proteinopathy or neurodegeneration, and a determination of an effect of the intervention data on the predicted marker of brain proteinopathy or neurodegeneration.
Clause 24. A method implemented with one or more computer processors, the method comprising:
Clause 25. A method implemented with one or more computer processors, the method comprising:
Clause 26. A non-transitory computer-readable medium comprising computer-executable instructions performing the operations recited in clause 1.
Clause 27. A non-transitory computer-readable medium comprising computer-executable instructions performing the operations recited in clause 24.
Clause 28. A non-transitory computer-readable medium comprising computer-executable instructions performing the operations recited in clause 25.
The present application is a continuation application of and claims priority to U.S. patent application Ser. No. 17/713,329 filed Apr. 5, 2022, which claims the benefit of U.S. Provisional Patent Application No. 63/244,080 filed Sep. 14, 2021. The entire content of the foregoing applications is incorporated herein by reference.
| Number | Name | Date | Kind |
|---|---|---|---|
| 6640121 | Telischi et al. | Oct 2003 | B1 |
| 7209775 | Bae et al. | Apr 2007 | B2 |
| 8884767 | Kidmose | Nov 2014 | B2 |
| 9025800 | Kidmose et al. | May 2015 | B2 |
| 9210517 | Pontoppidan et al. | Dec 2015 | B2 |
| 9408552 | Kidmose et al. | Aug 2016 | B2 |
| 9918650 | Kilsgaard et al. | Mar 2018 | B2 |
| 9949045 | Kure et al. | Apr 2018 | B2 |
| 10052065 | Boesen | Aug 2018 | B2 |
| 10194228 | Hirsch | Jan 2019 | B2 |
| 10275027 | Segal | Apr 2019 | B2 |
| 10433788 | Boesen | Oct 2019 | B2 |
| 10542904 | Durand | Jan 2020 | B2 |
| 10575777 | Ayers et al. | Mar 2020 | B2 |
| 10606354 | Segal | Mar 2020 | B2 |
| 10617297 | Turner | Apr 2020 | B2 |
| 10728642 | Solum et al. | Jul 2020 | B2 |
| 10743121 | Pontoppidan et al. | Aug 2020 | B2 |
| 10765337 | Andersen | Sep 2020 | B2 |
| 10798487 | Hviid | Oct 2020 | B2 |
| 10835145 | Prevoir et al. | Nov 2020 | B1 |
| 10841682 | Masaki et al. | Nov 2020 | B2 |
| 10856063 | Hviid | Dec 2020 | B1 |
| 10856071 | Verdooner et al. | Dec 2020 | B2 |
| 20070112277 | Fischer et al. | May 2007 | A1 |
| 20120165695 | Kidmose et al. | Jun 2012 | A1 |
| 20120302858 | Kidmose et al. | Nov 2012 | A1 |
| 20130085356 | Schlottau | Apr 2013 | A1 |
| 20140018649 | Jespersen | Jan 2014 | A1 |
| 20160310028 | Kidmose et al. | Oct 2016 | A1 |
| 20170224949 | Action | Aug 2017 | A1 |
| 20170251950 | Piron et al. | Sep 2017 | A1 |
| 20170311870 | Bardakjian et al. | Nov 2017 | A1 |
| 20170311881 | Jensen et al. | Nov 2017 | A1 |
| 20180014741 | Chou | Jan 2018 | A1 |
| 20180055398 | Westermann et al. | Mar 2018 | A1 |
| 20180177421 | Kilsgaard et al. | Jun 2018 | A1 |
| 20180220903 | LeBoeuf et al. | Aug 2018 | A1 |
| 20180228381 | LeBoeuf et al. | Aug 2018 | A1 |
| 20180235540 | Kirszenblat | Aug 2018 | A1 |
| 20180271428 | Takagi | Sep 2018 | A1 |
| 20180277238 | Boesen | Sep 2018 | A1 |
| 20180277245 | Boesen | Sep 2018 | A1 |
| 20180353095 | Boesen | Dec 2018 | A1 |
| 20180368718 | Koziol | Dec 2018 | A1 |
| 20190174238 | Lunner et al. | Jun 2019 | A1 |
| 20190209038 | Saab et al. | Jul 2019 | A1 |
| 20190223747 | Chou | Jul 2019 | A1 |
| 20190253793 | Pedersen et al. | Aug 2019 | A1 |
| 20190269336 | Perkins et al. | Sep 2019 | A1 |
| 20190282119 | Andersen et al. | Sep 2019 | A1 |
| 20190380597 | Howard | Dec 2019 | A1 |
| 20200085331 | Chou | Mar 2020 | A1 |
| 20200225750 | Segal | Jul 2020 | A1 |
| 20200268265 | Walsh et al. | Aug 2020 | A1 |
| 20200288253 | De Haan et al. | Sep 2020 | A1 |
| 20200306493 | Lee | Oct 2020 | A1 |
| 20210241909 | Bulut | Aug 2021 | A1 |
| 20210282698 | Garcia Molina | Sep 2021 | A1 |
| 20220000421 | Ippolito | Jan 2022 | A1 |
| Number | Date | Country |
|---|---|---|
| 2019169142 | Sep 2019 | WO |
| 2020035852 | Feb 2020 | WO |
| 2020061209 | Mar 2020 | WO |
| 2021 | Jun 2020 | WO |
| 2020210813 | Oct 2020 | WO |
| WO-2020210813 | Oct 2020 | WO |
| Entry |
|---|
| Meghdadi, Amir H., et al. “Transcranial impedance changes during sleep: a rheoencephalography study.” IEEE Journal of Translational Engineering in Health and Medicine 7 (2019): 1-7. (Year: 2019). |
| Guermandi, Marco, et al. “Active electrode IC for EEG and electrical impedance tomography with continuous monitoring of contact impedance.” IEEE transactions on biomedical circuits and systems 9.1 (2014): 21-33. (Year: 2014). |
| Goverdovsky et al., “In-Ear EEG From Viscoelastic Generic Earpieces: Robust and Unobtrusive 24/7 Monitoring”, IEEE Sensors Journal, Jan. 1, 2016, pp. 271-277, vol. 16, No. 1. |
| Goverdovsky et al. “Hearables: Multimodal physiological in-ear sensing”, Scientific Reports, Jul. 31, 2017, pp. 1-10, 7: 6948 | DOI:10.1038/s41598-017-06925-2. |
| R. Kusche et al., “An in-ear pulse wave velocity measurement system using heart sounds as time reference”, Current Directions in Biomedical Engineering 2015; 1:366-370, De Gruyter. |
| Kwon, Shinjae, Hojoong Kim, and Woon-Hong Yeo; “Recent Advances In Wearable Sensors and Portable Electronics For Sleep Monitoring”; iScience 24.5 (2021): 102461 (Year: 2021). |
| Hablitz, Lauren M., et al.; “Increased Glymphatic Influx Is Correlated With Hight EEG Delta Power and Low Heart Rate In Mice Under Anesthesia”; Science Advances 5.2 (2019); eaav5447; (Year: 2019). |
| Tanabe, Sean, et al. “Cohort Study of Electroencephalography Markers of Amyloid-Tau-Neurodegeneration Pathology.”; Brain Communications 2.2 (2020): fcaa099. (Year:2020). |
| International Search Report and Written Opinion issued in PCT Application No. PCT/US2022/023409, dated Jul. 25, 2022, 9 pages, Korean Intellectual Property Office. |
| Number | Date | Country | |
|---|---|---|---|
| 20230080140 A1 | Mar 2023 | US |
| Number | Date | Country | |
|---|---|---|---|
| 63244080 | Sep 2021 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 17713329 | Apr 2022 | US |
| Child | 17937952 | US |
