SLEEP-BASED BIOMETRIC TO PREDICT AND TRACK VIRAL INFECTION PHASES

Abstract

An apparatus and method involve leveraging unobtrusive sleep monitoring technologies, even consumer ones, to predict the phase of a viral infection and is usable for guiding patient treatment and tracking treatment effectiveness. An apparatus and method determine a phase of a viral infection based at least in part upon a data set that is input to an algorithm.

Claims

1. A method of determining a phase from among a plurality of phases of a viral infection in a patient, comprising: inputting to an algorithm a data set that comprises a set of parameters that are representative of the patient's current sleep architecture and another set of parameters that are representative of a baseline sleep architecture; and determining the phase of the viral infection based at least in part upon the data set.

2. The method of claim 1 wherein the set of parameters that are representative of the patient's current sleep architecture comprise at least one of a current Non-Rapid Eye Movement (NREM) sleep duration, a current Rapid Eye Movement (REM) sleep duration, a current Wake After Sleep Onset (WASO) duration, and a current Total Sleep Time (TST) duration, and wherein the set of parameters that are representative of the baseline sleep architecture comprise at least one of a baseline NREM sleep duration, a baseline REM sleep duration, a baseline WASO duration, and a baseline TST duration.

3. The method of claim 2, further comprising at least one of: receiving at least one of a current Heart Rate (HR) signal, a current Heart Rate Variability (HRV) signal, a current temperature signal, and a current electroencephalogram signal and deriving therefrom at least a portion of the set of parameters that are representative of the patient's current sleep architecture; and receiving at least one of a baseline HR signal, a baseline HRV signal, a baseline temperature signal, and a baseline electroencephalogram signal and deriving therefrom at least a portion of the set of parameters that are representative of the baseline sleep architecture.

4. The method of claim 2, further comprising employing at least one of a ballistocardiography sensor, a Doppler radar sensor, a photoplethysmography sensor, an electroencephalogram sensor, an actigraphy sensor, and a breathing sensor to derive at least one of: at least a portion of the set of parameters that are representative of the patient's current sleep architecture; and at least a portion of the set of parameters that are representative of the baseline sleep architecture.

5. The method of claim 2, further comprising inputting the set of data into a machine learning device and employing the machine learning device in the determining of the phase of the viral infection.

6. The method of claim 5, further comprising employing the machine learning device to apply a set of thresholds to the set of data in the determining of the phase of the viral infection.

7. The method of claim 2, further comprising: comparing the set of parameters that are representative of the patient's current sleep architecture with the set of parameters that are representative of the baseline sleep architecture to determine a score for the patient; and determining the phase of the viral infection based at least in part upon the score.

8. The method of claim 7, further comprising inputting the data set and the score into a machine learning device and employing the machine learning device in the determining of the phase of the viral infection.

9. An apparatus structured to determine a phase from among a plurality of phases of a viral infection in a patient, comprising: a processor apparatus comprising a processor and a storage; an input apparatus structured to provide input signals to the processor apparatus; an output apparatus structured to receive output signals from the processor apparatus; the storage having stored therein a number of routines which, when executed on the processor, cause the apparatus to perform operations comprising: inputting to an algorithm a data set that comprises a set of parameters that are representative of the patient's current sleep architecture and another set of parameters that are representative of a baseline sleep architecture; and determining the phase of the viral infection based at least in part upon the data set.

10. The apparatus of claim 9 wherein the set of parameters that are representative of the patient's current sleep architecture comprise at least one of a current Non-Rapid Eye Movement (NREM) sleep duration, a current Rapid Eye Movement (REM) sleep duration, a current Wake After Sleep Onset (WASO) duration, and a current Total Sleep Time (TST) duration, and wherein the set of parameters that are representative of the baseline sleep architecture comprise at least one of a baseline NREM sleep duration, a baseline REM sleep duration, a baseline WASO duration, and a baseline TST duration.

11. The apparatus of claim 10, wherein the operations further comprise at least one of: receiving at least one of a current Heart Rate (HR) signal, a current Heart Rate Variability (HRV) signal, a current temperature signal, and a current electroencephalogram signal and deriving therefrom at least a portion of the set of parameters that are representative of the patient's current sleep architecture; and receiving at least one of a baseline HR signal, a baseline HRV signal, a baseline temperature signal, and a baseline electroencephalogram signal and deriving therefrom at least a portion of the set of parameters that are representative of the baseline sleep architecture.

12. The apparatus of claim 10, wherein the operations further comprise employing at least one of a ballistocardiography sensor, a Doppler radar sensor, a photoplethysmography sensor, an electroencephalogram sensor, an actigraphy sensor, and a breathing sensor to derive at least one of: at least a portion of the set of parameters that are representative of the patient's current sleep architecture; and at least a portion of the set of parameters that are representative of the baseline sleep architecture.

13. The apparatus of claim 10, wherein the operations further comprise inputting the set of data into a machine learning device and employing the machine learning device in the determining of the phase of the viral infection.

14. The apparatus of claim 13, wherein the operations further comprise employing the machine learning device to apply a set of thresholds to the set of data in the determining of the phase of the viral infection.

15. The apparatus of claim 10, wherein the operations further comprise: comparing the set of parameters that are representative of the patient's current sleep architecture with the set of parameters that are representative of the baseline sleep architecture to determine a score for the patient; and determining the phase of the viral infection based at least in part upon the score.

16. The apparatus of claim 15, wherein the operations further comprise inputting the data set and the score into a machine learning device and employing the machine learning device in the determining of the phase of the viral infection.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0027] FIG. 1 is a conceptual model of the effect of immunity-function activation on sleep;

[0028] FIG. 2 is a three-phase model which depicts classification of disease states and clinical signs associated with at least one viral pathology;

[0029] FIG. 3 is a depiction of a sleep-based biometric that are embodied in an improved method and apparatus in accordance with the disclosed and claimed concept to predict and track viral infection phases;

[0030] FIG. 4 is an exemplary depiction of at least a portion of a viral infection phase decision tree that can be implemented with a machine-learning system in accordance with the improved method and apparatus in accordance with the disclosed and claimed concept.

DETAILED DESCRIPTION OF EXEMPLARY EMBODIMENTS

[0031] As used herein, the singular form of “a”, “an”, and “the” include plural references unless the context clearly dictates otherwise. As used herein, the statement that two or more parts or components are “coupled” shall mean that the parts are joined or operate together either directly or indirectly, i.e., through one or more intermediate parts or components, so long as a link occurs. As used herein, “directly coupled” means that two elements are directly in contact with each other. As used herein, “fixedly coupled” or “fixed” means that two components are coupled so as to move as one while maintaining a constant orientation relative to each other.

[0032] As used herein, the word “unitary” means a component is created as a single piece or unit. That is, a component that includes pieces that are created separately and then coupled together as a unit is not a “unitary” component or body. As employed herein, the statement that two or more parts or components “engage” one another shall mean that the parts exert a force against one another either directly or through one or more intermediate parts or components. As employed herein, the term “number” shall mean one or an integer greater than one (i.e., a plurality).

[0033] Directional phrases used herein, such as, for example and without limitation, top, bottom, left, right, upper, lower, front, back, and derivatives thereof, relate to the orientation of the elements shown in the drawings and are not limiting upon the claims unless expressly recited therein.

[0034] The invention advantageously creates a model to evaluate the viral infection phase with an aggregated score through tracking and analyzing sleep metrics based on the following information. Denote the current duration of NREM sleep as D.sub.NREM; denote the current duration of REM sleep as D.sub.REM; denote the current duration of Wake After Sleep Onset (WASO) as D.sub.WASO; and denote the current total sleep duration as D.sub.TST. Likewise, denote the baseline (e.g. weekly average) of NREM sleep as D.sub.NREMB; denote the baseline (e.g. weekly average) of REM sleep as D.sub.REMB; denote the baseline (e.g. weekly average) of WASO as D.sub.WASOB; and denote the baseline (e.g. weekly average) of total sleep duration as D.sub.TSTB. Additionally, denote S as the aggregated score of a viral condition based on sleep biometrics. S will be in one of the following ranges:

TABLE-US-00001 TABLE 1 Aggregated Score Range and Viral Infection Phases Aggregated Score Range (S) Viral Infection Phase  0 Baseline (no infection)  1-100 Phase 1 101-1000 Phase 2 1001+ Phase 3

[0035] The aggregated score S can be defined as a function of relevant sleep metrics that at least includes, but not limited to, NREM, REM, TST, and WASO. In the exemplary embodiment described herein, the aggregated score S can be represented as a function of percentage changes of these metrics,

S=W.sub.1×F.sub.1+W.sub.2×F.sub.2+W.sub.3×F.sub.3 where

[0036] F.sub.1 is a function of % change of current NREM sleep as compared with baseline, e.g. F1=100×((D.sub.NREM/D.sub.TST)−(D.sub.NREMB/D.sub.TSTB));

[0037] F.sub.2 is a function of % change of current REM sleep as compared with baseline, e.g. F.sub.2=100×((D.sub.REMB/D.sub.TSTB)−(D.sub.REM/D.sub.TST));

[0038] F.sub.3 is a function of % change of current WASO as compared with baseline, e.g. F.sub.3=100×((D.sub.WASO/D.sub.TST) (D.sub.WASOB/D.sub.TSTB)); and

[0039] where W.sub.1, W.sub.2, and W.sub.3 are the corresponding weights that are applied to each function. A data set formed of the aforementioned parameters D.sub.NREM; D.sub.REM; D.sub.WASO; D.sub.TST; D.sub.NREMB; D.sub.REMB; D.sub.WASOB; D.sub.TSTB for a given patient could thus be used to determine a score for the patient which describes the phase of the viral illness in the patient. This can then be output to a medical team, or the output could be in the form of an instruction to provide a certain type of medical care, such an instruction to administer pharmacotherapy in the event that the score S indicates that the patient is early in the course of the illness, but or to provide anti-inflammatory therapy in the event that the score S indicates that the patient is later in the course of the illness.

[0040] In some embodiments, the weights can be pre-determined to distinguish the impact of each corresponding function. For instance, is W.sub.1=100; W.sub.2=1000; W.sub.3=10000, this will result in the estimated score ranges being as defined in Table 1. To improve the accuracy of the estimation, a moving average of % changes (e.g. past 3 nights) instead of % change on a single night could be used for evaluation.

[0041] In one implementation, the disclosed and claimed concept is embodied in a decision tree based upon a machine learning algorithm to determine the viral infection phase through tracking and analyzing sleep metrics. An example of such a decision tree is depicted generally in FIG. 4 with the notations set forth elsewhere herein. With the decision tree, the estimated score is a discrete value that maps to a specific viral infection phase as in Table 2:

TABLE-US-00002 TABLE 2 Estimated Score and Viral infection Phase Estimated Score (S) Viral Infection Phase 0 Baseline (no infection) 1 Phase 1 2 Phase 2 3 Phase 3

[0042] The initial percentage change thresholds α, β, and δ for the corresponding sleep metrics are selected heuristically. They can be improved through continuous learning approach (e.g. minimize the aggregated cost function) as the model is being used continuously in the large population.

[0043] By way of example, each patient could be characterized as a data set in the exemplary form of a vector, i.e., an array of values, such as the nine parameters D.sub.NREM; D.sub.REM; D.sub.WASO; D.sub.TST; D.sub.NREMB; D.sub.REMB; D.sub.WASOB; D.sub.TSTB; and S. The decision tree might ask a series of yes/no questions of the vector, and each answer would lead either to another question or to a decision. For instance, the questions might begin with something like “is D.sub.NREM greater than 20?” If the answer to that question is “yes”, the next question in the decision tree might be something like “is D.sub.REM less than 30?” The answer to this question would then lead to another question, and so forth, until a determination is made by the decision tree about what is the phase of the patient's viral illness. Responsive to this determination, the output can be in the form of the patient's score, the phase of the patient's viral illness, and/or an instruction to provide a certain therapy to the patient.

[0044] In an alternative embodiment, the presence of additional biometric data points, like respiratory rate, heart rate, body temperature, SpO.sub.2 etc. can be used by the invention to provide early detection of a viral infection. If, in additional to the change in sleep architecture described elsewhere herein, at least one additional biometric data point falls beyond a normal range for a predetermined duration, it can be inferred that the user is at a higher susceptibility for a viral infection. The user is then alerted to take further action of either contacting their healthcare provider or consulting for a check-up to confirm the presence an infection. The previously stated embodiments can be used to determine the phase of infection and drive the appropriate treatment decisions.

[0045] This invention is advantageously applicable to numerous known products. For instance Philips's Smart Sleep Deep Sleep Headband can provide numerous of the inputs described in FIG. 3. Existing Philips offerings track patients both in-hospital and at home. This solution can be utilized while monitoring patients during and after respiratory disease treatments, both in and out of hospital. This invention alone or in conjunction with patient monitoring can be utilized to advantageously provide important insights to customers (consumers, patients, and/or healthcare providers) in early detection and phase prediction of viral infections and in monitoring of recovery from viral infections. This invention is particularly useful for viral conditions that affect the respiratory system as these are characterized by inflammatory responses mediated by cytokines, the latter of which are well-known for regulating sleep.

[0046] An improved apparatus 4 in accordance with the disclosed and claimed concept is depicted in a schematic fashion in FIG. 3. Apparatus 4 can be employed in performing an improved method 100 that is likewise in accordance with the disclosed and claimed concept and at least a portion of which is depicted in a schematic fashion in FIG. 3. Apparatus 4 can be characterized as including a processor apparatus 8 that can be said to include a processor 12 and a storage 16 that are connected with one another. Storage 16 is in the form of a non-transitory storage medium that has stored therein a number of routines 20 that are likewise in the form of a non-transitory storage medium and that include instructions which, when executed on processor 12, cause apparatus 4 to perform certain operations such as are mentioned elsewhere herein.

[0047] Apparatus 4 further can be said to include an input apparatus 24 that provides input signals to processor 12 and to further include an output apparatus 28 that receives output signals from processor 12. For instance, input apparatus 24 can include any one or more of an activity monitor 32, a sonar device 36, a radar device 40, a force sensor 44, etc., without limitation, among any of a variety of other input devices such as monitors for Heart Rate (HR), Heart Rate Variability (HRV), temperature, daytime activity, and the like. The waveforms during sleep are useful for the purposes of the disclosed and claimed concept. In other words, it is not strictly necessary to calculate first sleep architecture. Output apparatus 28 includes any of a variety of output devices such as visual displays and the like, without limitation, that can visually or otherwise output instruction, the score S, the phase of the viral infection in the patient, etc.

[0048] The various input devices of input apparatus 24 provide sleep macro-architecture and micro-architecture of the patient to a number of algorithms 48 that can be considered to be among the routines 20 and which can include a machine learning algorithm. The sleep-based biometric data is processed using, for instance, a machine learning algorithm 52 or a signal processing algorithm 56, or both, to result in any one or more of the parameters set forth hereinbefore, such as TST, WASO, N3, N2, REM, and the like without limitation. These parameters can then be used as inputs to determine S in order to determine the phase of the viral infection, or the parameters, potentially also including S, can be input into the machine learning decision tree algorithm depicted generally in FIG. 4 to result in a decision as to the phase of the viral infection. It is understood that the parameters are evaluated over time and thus, over time, can include a set of baseline parameters and a set of current parameters.

[0049] Apparatus 4 displays and communicates information about inflammation levels related to the viral infection, suggests a treatments (e.g. pharmacological, anti-inflammatory) or additional tests by considering sleep changes. Further investigation involves tracking and determining a correlation between the sleep architecture and the output recovery prediction.

[0050] In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word “comprising” or “including” does not exclude the presence of elements or steps other than those listed in a claim. In a device claim enumerating several means, several of these means may be embodied by one and the same item of hardware. The word “a” or “an” preceding an element does not exclude the presence of a plurality of such elements. In any device claim enumerating several means, several of these means may be embodied by one and the same item of hardware. The mere fact that certain elements are recited in mutually different dependent claims does not indicate that these elements cannot be used in combination.

[0051] Although the invention has been described in detail for the purpose of illustration based on what is currently considered to be the most practical and preferred embodiments, it is to be understood that such detail is solely for that purpose and that the invention is not limited to the disclosed embodiments, but, on the contrary, is intended to cover modifications and equivalent arrangements that are within the spirit and scope of the appended claims. For example, it is to be understood that the present invention contemplates that, to the extent possible, one or more features of any embodiment can be combined with one or more features of any other embodiment.