Multi-channel ambient interference detection

Abstract

Techniques for multi-channel ambient interference detection are described and are implementable to improve accuracy of physiological measurements by wearable devices. In an example, a wearable device attachable to a skin surface of a user includes an optical sensor configured to detect light reflected through the skin surface and a noise detection component configured to sample an environmental condition. A processor of the wearable device receives a first signal from the optical sensor and a second signal from the noise detection component. Physiological data is generated based on the first signal and a noise condition is detected based on the second signal that indicates whether environmental interference is present that impacts accuracy of the physiological data. The wearable device can leverage the noise condition to refine the physiological data, adjust device parameters, or modify data processing to account for the detected interference.

Description

Multi-Channel Ambient Interference DetectionRELATED APPLICATIONS

[0001] This application claims priority to U.S. Provisional Application No. 63 / 703,456, filed October 4, 2024, and titled “Multi-Channel Photoplethysmography System for Noise Rejection” and to U.S. Nonprovisional Application No. 19 / 261,874 filed July 7, 2025, and titled “Multi-Channel Ambient Interference Detection” which are hereby incorporated by reference in their entireties.BACKGROUND

[0002] Optical sensing technologies, such as photoplethysmography (“PPG”), utilize light-based sensors to measure various physiological parameters by detecting changes in light absorption, reflection, or scattering through biological tissues. These optical sensors are often integrated into wearable form factors such as to support continuous health monitoring. However, an accuracy of optical measurements can be affected by environmental factors such as ambient light interference and / or electromagnetic interference. For instance, fluctuations in ambient light conditions (such as those caused by artificial lighting or movement) or proximal electromagnetic fields (such as from nearby electronic devices or power sources) can introduce noise into optical sensor readings. This interference leads degraded signal quality and therefore inaccurate and unreliable physiological data, which can offset the advantages of optical sensing devices for health monitoring applications.FIG. 1 Patents 1 Docket No.: 10030PCT1BRIEF DESCRIPTION OF THE DRAWINGS

[0003] FIG. 1 is a block diagram of a nonlimiting example of an environment that is operable to employ techniques for multichannel ambient interference detection as described herein.

[0004] FIG. 2 depicts a nonlimiting example of a monitoring device.

[0005] FIG. 3 depicts a nonlimiting system in an example implementation of multichannel ambient interference detection showing operation of the noise management system of FIG. 1 in more detail.

[0006] FIGS. 4a and 4b depict nonlimiting examples of multi-channel ambient interference detection in which variations in lighting conditions can cause an ambient light rejection process to fail.

[0007] FIG. 5 depicts a nonlimiting example of multi-channel ambient interference detection in which a wearable device includes a noise detection component configured to sample an ambient condition.

[0008] FIG. 6 depicts a nonlimiting example of multi-channel ambient interference detection that compares photodiode signal data under various ambient light conditions.

[0009] FIG. 7 depicts a nonlimiting example of multi-channel ambient interference detection in which various form factors of wearable devices are depicted.

[0010] FIG. 8 depicts a nonlimiting example of multi-channel ambient interference detection in which a portion of a noise detection circuit is configurable to detect electromagnetic interference.

[0011] FIG. 9 depicts a nonlimiting example of multi-channel ambient interference detection in which a user interface for an interference detection scenario is shown.FIG. 1 Patents 2 Docket No.: 10030PCT1

[0012] FIG. 10 depicts a flow diagram depicting an algorithm as a step-by-step procedure in an example implementation, one or more steps of which are performable by a processing device to determine a noise condition and perform one or more actions based on the noise condition.DETAILED DESCRIPTION

[0013] The use of optical sensing technologies, such as photoplethysmography (“PPG”) in wearable devices has enabled continuous monitoring of various physiological parameters. For instance, these technologies detect changes in light absorption, reflection, or scattering through biological tissues to measure parameters such as heart rate and blood oxygen levels. However, accuracy of optical measurements can be significantly affected by environmental factors, such as ambient light interference and / or electromagnetic interference. Fluctuations in ambient light conditions, such as those caused by artificial lighting or movement, as well as proximal electromagnetic fields from nearby electronic devices or power sources, can introduce noise into optical sensor readings. This interference leads to degraded signal quality and consequently inaccurate and unreliable physiological data, which undermines advantages of optical sensing devices for health monitoring applications.

[0014] Conventional approaches to mitigate these issues are limited. Some conventional approaches attempt to denoise optical sensor data using ambient light rejection techniques. For example, a device that includes a PPG sensor can collect illuminated samples, e.g., with illumination from an integrated LED applied, and nonilluminated samples, e.g., with the LED turned off. Values for the non-illuminatedFIG. 1 Patents 3 Docket No.: 10030PCT1samples, which are intended to capture an impact of ambient light, can then be subtracted from values for the illuminated samples with the goal to isolate the contribution due to the LED.

[0015] However, such ambient light rejection techniques often fail due to an inherently dynamic and high-frequency nature of various light sources. While ambient light rejection techniques are able to denoise optical sensor data when ambient light is relatively constant, such techniques are unable to account for interference due to dynamic lighting conditions or high-frequency light sources. For example, light sources such as fluorescent lamps that “flicker” at relatively high frequencies can generate interference patterns that change more quickly than sampling rates of conventional ambient light rejection and thus these conventional techniques fail to account for high-frequency components of the interference pattern. Further, such techniques do not address electromagnetic interference that may be introduced by nearby electronic devices, power sources, or other electromagnetic field sources. Accordingly, conventional noise mitigation techniques often fail to adequately address the complex and dynamic nature of environmental interference which results in compromised accuracy of physiological measurements in wearable optical sensing devices.

[0016] Accordingly, techniques, methods, and systems for multi-channel ambient interference detection are described that improve environmental noise detection and management relative to conventional approaches. The techniques described herein, for instance, leverage multi-channel sensing components to reliably detect and characterize environmental interference conditions. For instance, by positioning one or more noiseFIG. 1 Patents 4 Docket No.: 10030PCT1detection components proximate to an optical sensor, the techniques described herein enable environmental conditions to be monitored independently from physiological measurements, which provides a robust approach to interference detection that can distinguish between physiological events and environmental artifacts.

[0017] Consider an example in which a smartwatch is equipped with a PPG sensor to collect heart rate and blood oxygen data of a user. The PPG sensor, for instance, emits light into a skin surface of the user and measures an amount of light that is reflected back through the skin surface. The amount of reflected light varies with blood flow, and thus measurements collected by the optical sensor are usable to derive heart rate and blood oxygen levels. However, using conventional techniques, the smartwatch is unable to accurately detect the amount of reflected light due to physiological changes when environmental interference, such as flickering fluorescent lighting or electromagnetic fields from nearby devices, corrupts the optical signal. This results in inaccurate PPG data that may be misinterpreted as physiological events, leading to incorrect health insights and potentially false alerts about a cardiovascular status of the user.

[0018] To overcome these limitations, a wearable device attachable to a skin surface of a user is able to simultaneously collect physiological data via a first channel associated with an optical sensor and interference data via a second channel associated with a noise detection component. The optical sensor, for instance, is configured to detect light reflected through the skin surface and the noise detection component can sample an environmental condition such as ambient lighting or electromagnetic interference. In at least one example, the optical sensor is disposed on a first side of theFIG. 1 Patents 5 Docket No.: 10030PCT1wearable device, such as to face the skin surface, while the noise detection component is disposed on a second side of the wearable device, such as to face “away from” the skin surface however various configurations of the optical sensor and noise detection component are considered.

[0019] The wearable device may also include an illumination component positioned on a same side as the optical sensor. The illumination component, for instance, is configured to emit light to penetrate the skin surface. The optical sensor is then configured to detect an amount of emitted light reflected back through the skin surface. The wearable device further includes a processor that is operable to implement an ambient light rejection process to denoise a signal collected by the optical sensor based on values corresponding to illuminated and non-illuminated conditions.

[0020] However, as described above such denoising is susceptible to failure under environmental conditions that include light sources with a frequency above a threshold and / or include electromagnetic interference. Accordingly, the processor can leverage a signal obtained by the noise detection component to determine a noise condition. The noise condition, for instance, describes properties of detected environmental interference, such as whether environmental interference is present that is capable of causing the ambient light detection process to fail.

[0021] In an example to do so, the processor receives a first signal from the optical sensor and receives a second signal from the noise detection component. Based on the first signal, the processor generates physiological data that is indicative of one or more biological characteristics of the user, e.g., blood oxygen saturation levels, pulse transit time, respiratory rate, etc. The processor further processes the second signal to identifyFIG. 1 Patents 6 Docket No.: 10030PCT1and / or characterize various types of environmental interference that may compromise measurement accuracy.

[0022] The wearable device can include one or more of a variety of noise detection components to detect different types of interference in various environments. In some implementations, the noise detection component includes a photodiode disposed to face away from the skin surface to detect ambient light fluctuations. Additionally or alternatively, the noise detection component includes one or more of a resistor, capacitor, diode, or a particular trace configuration implementable to detect electromagnetic interference. In this way, the wearable device can address multiple sources of environmental interference that could impact measurement accuracy.

[0023] Accordingly, based on the second signal the processor detects the noise condition that indicates whether environmental interference is present that impacts the accuracy of the physiological data. For instance, the processor determines whether the ambient light rejection process is insufficient to effectively remove environmental interference from the physiological data. In various examples, this includes determining that a quantification of the environmental condition (e.g., a frequency of ambient light, a magnitude of electromagnetic interference, etc.) is above a particular threshold.

[0024] By way of example, the processor determines whether ambient light fluctuates at a frequency high enough to cause significant changes in lighting conditions between collection of an illuminated sample and a non-illuminated sample during the ambient noise rejection process. In this example scenario, a sampling interval between collection of the illuminated and non-illuminated samples includes a temporal gap ofFIG. 1 Patents 7 Docket No.: 10030PCT1approximately 10-200 microseconds while an ambient light source proximal to the wearable device operates at a frequency in the kilohertz range. Accordingly, the wearable device detects a noise condition that indicates the variations in light intensity due to the ambient light source occur faster than the sampling interval, and thus compromise effectiveness of the subtraction-based denoising approach.

[0025] Based on the noise condition, the processor can perform a variety of actions. For instance, the processor is operable to refine the physiological data by removing noise from the first signal based on the second signal. Additionally or alternatively, the processor can output an indication of the noise condition, such as to present a visual notification, audio alert, or haptic feedback to inform the user about the detected interference. The processor may also adjust parameters of the optical sensor, illumination component, and / or noise detection component based on the noise condition, such as to conserve processing resources during periods of interference when physiological data accuracy may be relatively low. In some examples, the processor can further modify processing of physiological data based on the detected noise condition, such as in real time or for post-processing analysis.

[0026] Accordingly, the wearable device can implement this multi-channel approach to ambient interference detection to detect high-frequency and electromagnetic interference that would otherwise compromise measurement quality, which is not possible using traditional single-channel approaches. Thus, the techniques described herein provide a robust modality to identify when ambient light rejection techniques are likely to fail, allowing for real-time adjustments to maintain data accuracy and enhance a user experience. Further, by quantifying a level of environmental interference, theFIG. 1 Patents 8 Docket No.: 10030PCT1wearable device can implement strategies to mitigate an impact of environmental interference, such as to improve reliability of physiological measurements in various real-world conditions.

[0027] The described techniques additionally support operational advantages for processing devices that leverage these techniques. For instance, a processing device can selectively disable or reduce a sampling rate of the optical sensor during periods of high environmental interference, thereby conserving computational resources that would otherwise be expended processing unreliable data. The noise detection component further enables efficient signal processing by supporting dynamic adjustments to filtering algorithms based on detected interference characteristics which obviates a conventional demand to implement computationally expensive global data filtering techniques. These techniques can also be used to optimize power consumption in processing devices by increasing measurement frequency during periods of lower environmental interference relative to measurement frequency for periods of higher environmental interference, which reduces a need for repeated measurements and postprocessing correction that would otherwise consume additional computational resources.

[0028] In some aspects, the techniques described herein relate to a wearable device attachable to a skin surface of a user including: an optical sensor configured to detect light reflected through the skin surface; a noise detection component configured to sample an environmental condition; and a processor configured to: receive a first signal from the optical sensor and a second signal from the noise detection component; generate physiological data based on the first signal; and detect, based on the secondFIG. 1 Patents 9 Docket No.: 10030PCT1signal, a noise condition that indicates whether environmental interference is present that impacts an accuracy of the physiological data.

[0029] In some aspects, the techniques described herein relate to a wearable device, further including an illumination component positioned on a same side of the wearable device as the optical sensor and configured to emit light to penetrate the skin surface of the user, the optical sensor configured to detect an amount of the emitted light that is reflected back through the skin surface.

[0030] In some aspects, the techniques described herein relate to a wearable device, wherein to generate the physiological data includes an ambient light rejection process to denoise the first signal based on values of the first signal that correspond to an illuminated condition and values of the first signal that correspond to a non-illuminated condition, and wherein to detect the noise condition includes determining that the ambient light rejection process is insufficient to effectively denoise the first signal to remove environmental interference from the physiological data.

[0031] In some aspects, the techniques described herein relate to a wearable device, wherein the noise condition is detected based on one or more of an amplitude, magnitude, frequency, or rate of change of the second signal.

[0032] In some aspects, the techniques described herein relate to a wearable device, wherein the optical sensor includes a first photodiode disposed to face the skin surface and the noise detection component includes a second photodiode disposed to face away from the skin surface and configured to detect ambient light fluctuations.FIG. 1 Patents 10 Docket No.: 10030PCT1

[0033] In some aspects, the techniques described herein relate to a wearable device, wherein the noise detection component includes one or more of a resistor, capacitator, diode, or trace configuration configured to detect electromagnetic interference.

[0034] In some aspects, the techniques described herein relate to a wearable device, wherein the noise condition includes a quantification of the environmental interference.

[0035] In some aspects, the techniques described herein relate to a wearable device, wherein the processor is further configured to refine the physiological data by removing noise from the first signal based on the second signal.

[0036] In some aspects, the techniques described herein relate to a wearable device, wherein the processor is further operable to output an indication of the noise condition, adjust a parameter of the optical sensor or the noise detection component based on the noise condition, or modify processing of the physiological data based on the noise condition.

[0037] In some aspects, the techniques described herein relate to a method implemented by a wearable device attachable to a skin surface of a user, the method including: receiving a first signal from an optical sensor configured to detect light reflected through the skin surface and a second signal from a noise detection component configured to sample an environmental condition; generating physiological data based on the first signal; and presenting a noise condition generated based on the second signal that indicates a presence of environmental interference that reduces an accuracy of the physiological data.

[0038] In some aspects, the techniques described herein relate to a method, wherein the generating the physiological data includes implementing a denoising process basedFIG. 1 Patents 11 Docket No.: 10030PCT1on values of the first signal that correspond to an illuminated condition and values of the first signal that correspond to a non-illuminated condition, and the noise condition indicates that the denoising process is insufficient to remove environmental interference from the physiological data.

[0039] In some aspects, the techniques described herein relate to a method, wherein the presenting the noise condition includes outputting an indication of the noise condition, the indication including one or more of a visual notification, audio alert, or haptic feedback.

[0040] In some aspects, the techniques described herein relate to a method, further including adjusting a parameter of the optical sensor, the noise detection component, or an illumination component of the wearable device based on a quantification of the environmental interference included in the noise condition.

[0041] In some aspects, the techniques described herein relate to a method, wherein the optical sensor includes an illumination component to emit light towards the skin surface and a first photodiode configurable for a photoplethysmography (PPG) task and the noise detection component includes a second photodiode configured to detect ambient light interference.

[0042] In some aspects, the techniques described herein relate to a method, wherein the noise detection component includes one or more of a resistor, capacitator, diode, or trace configuration configured to detect electromagnetic interference.

[0043] In some aspects, the techniques described herein relate to a method, wherein the receiving the first signal and the second signal includes one or more of sampling theFIG. 1 Patents 12 Docket No.: 10030PCT1first signal and the second signal substantially simultaneously or sampling the second signal at a time offset relative to sampling the first signal.

[0044] In some aspects, the techniques described herein relate to a noise management system including: an illumination component configured to cause an illuminated condition via light emission towards a skin surface and cause a non-illuminated condition via cessation of light emission towards the skin surface; a first photodiode configured to detect light reflected through the skin surface; a second photodiode configured to detect a condition of ambient light; and one or more processors configured to: receive a first signal from the first photodiode and a second signal from the second photodiode; generate physiological data based on the first signal by implementing a denoising process based on values of the first signal that correspond to the illuminated condition and values of the first signal that correspond to the non-illuminated condition; and determine a noise condition based on the second signal that indicates that the denoising process is insufficient to account for the condition of ambient light.

[0045] In some aspects, the techniques described herein relate to a noise management system, wherein the condition of ambient light includes a frequency of the ambient light above a threshold.

[0046] In some aspects, the techniques described herein relate to a noise management system, wherein the physiological data includes photoplethysmography (PPG) data used to determine blood oxygen saturation levels.

[0047] In some aspects, the techniques described herein relate to a noise management system, wherein the one or more processors are further configured to: implement a low power mode by deactivating the illumination component; and generateFIG. 1 Patents 13 Docket No.: 10030PCT1photoplethysmography (PPG) data while in the non-illuminated condition based on a relationship between the first signal and the second signal.

[0048] FIG. 1 is a block diagram of a nonlimiting example 100 of an environment that is operable to employ techniques for multichannel ambient interference detection as described herein. The illustrated example 100 includes person 102, who is depicted wearing a monitoring device 104. The illustrated environment also includes an analysis platform 106. The analysis platform 106 may be connected to the monitoring device 104 via one or more wireless connections directly or via one or more wired and / or wireless connections and one or more intermediate devices, such as a computing device associated with the person 102, network routing devices and equipment, server devices, and / or the Internet, to name just a few.

[0049] The monitoring device 104 may be utilized to monitor one or more aspects of the person 102. In some scenarios, for instance, the monitoring device 104 may be provided to record electrical activity of the person 102’s heart over an observation period, e.g., lasting some number of seconds or minutes, lasting multiple days, and so on. By way of example, the person 102 may have a magnitude of his or her heart’s electrical potential monitored over time to produce one or more electrocardiograms, which may be used to predict any of a variety of events. In at least one example, the monitoring device 104 is provided to record photoplethysmography (“PPG”) data over an observation period, such as to collect blood oxygen saturation (“SpO2”) data. Alternatively or in addition, the monitoring device 104 may be used to output measurements 108 (e.g., a time sequence of measurements such as a time sequence ofFIG. 1 Patents 14 Docket No.: 10030PCT1electric potential measurements), which may indicate an observation or be used to generate a prediction of one or more events.

[0050] In connection with the monitoring device, instructions may be provided to the person 102 that instruct the person 102 how to operate the monitoring device 104 and / or how to behave (e.g., sleep, perform activity) while wearing monitoring device 104. In one or more implementations, the instructions may be provided as pail of a kit, e.g., written instructions. Alternately or additionally, the analysis platform 106 may cause the instructions to be communicated to and output (e.g., for display and / or audio output) via a computing device associated with the person 102. In one or more implementations, the analysis platform 106 may wait to provide these instructions for output after a predetermined amount of time of an observation period has lapsed (e.g., two days) while wearing the monitoring device 104 and / or based on patterns in the aspects of the person 102 being measured.

[0051] The monitoring device 104 may be configured in a variety of ways to monitor one or more aspects of the person 102. Moreover, the form factor depicted in FIGs 1 and 2 is just one example form factor, and the form factor of the monitoring device 104 may differ in variations. It is to be appreciated that the monitoring device 104 may be configured with one or more sensors, examples of which include one or more of: a plurality of electrodes (e.g., that can be placed on the skin of the person), an accelerometer, and a pulse oximeter (e.g., to measure and record oxygen saturation (SpO2) and / or produce a pho toplethy smogram of the person 102), to name just a few. Certainly, the monitoring device 104 may be configured with any of a variety of types of sensors without departing from the described techniques.FIG. 1 Patents 15 Docket No.: 10030PCT1

[0052] Although the monitoring device 104 may be configured in a similar manner as monitoring devices used for clinically monitoring patients, in one or more implementations, the monitoring device 104 may be configured differently than the devices used for monitoring and / or diagnosing patients clinically. By way of example, and not limitation, the monitoring device 104 may be configured as a ring, a watch, a patch, and / or a strap, to name just a few form factors. Alternatively or additionally, the monitoring device 104 may have a similar form factor as for clinical settings, but have different functionality, such as functionality that prevents a wearer from viewing the measurements.

[0053] In one or more implementations, the monitoring device 104 may be configured to offload measurements during the course of the observation period. By way of example, the monitoring device 104 may offload the measurements by transmitting them via a wired or wireless connection to an external computing device, e.g., at predetermined time intervals and / or responsive to establishing or reestablishing a connection with the computing device. In one or more implementations, the measurements 108 and / or other data from the monitoring device 104 may be compressed by the monitoring device 104 for wireless transmission, e.g., using one or more of a variety of data compression techniques. Compression of the sensor data in this way can reduce battery usage of the monitoring device 104 during the observation period and facilitate wear during assessments of various physiological conditions.

[0054] To the extent that the monitoring device 104 may be configured to store the measurements 108 for an entirety of an observation period, in one or more implementations, the monitoring device 104 may be configured without wirelessFIG. 1 Patents 16 Docket No.: 10030PCT1transmission means, e.g., without any antennae to transmit the measurements 108 wirelessly and without hardware or firmware to generate packets for such wireless transmission. Instead, the monitoring device 104 may be configured with hardware to communicate the measurements 108 via a physical, wired coupling. In such scenarios, the monitoring device 104 may be “plugged in” to extract the measurements 108 from the device’s storage.

[0055] Accordingly, the monitoring device 104 may be configured with one or more ports to enable wired transmission of the measurements to an external computing device. Examples of such physical couplings may include micro universal serial bus (USB) connections, mini-USB connections, and USB-C connections, to name just a few. Although the monitoring device 104 may be configured for extraction of the measurements 108 via wired connections as discussed just above, in different scenarios, the monitoring device 104 may alternately or additionally be configured to offload the measurements 108 over one or more wireless connections.

[0056] Once the monitoring device 104 produces the measurements 108, the measurements are provided to the analysis platform 106. As noted above, the measurements 108 may be communicated to the analysis platform 106 over wired and / or wireless connection(s).

[0057] In scenarios where the analysis platform 106 is implemented partially or entirely on the monitoring device 104, for instance, the measurements 108 may be transferred over a bus from the device’s local storage to a processing system of the device. In scenarios where the monitoring device 104 is configured to generate one or more predictions 110 by processing the measurements 108, the monitoring device 104FIG. 1 Patents 17 Docket No.: 10030PCT1may also be configured to provide the generated one or more predictions 110 as output, e.g., by communicating the one or more predictions 110 to an external computing device. In other scenarios, the measurements 108 may be processed by an external computing device configured to generate one or more predictions 110. For example, the measurements 108 and / or other measurements may be processed by a smartphone associated with the user, a smartphone or other dedicated device associated with the monitoring device 104, and / or one or more server computers at a data center or other location that can be utilized by an entity associated with the monitoring device 104, to name just a few. In other words, those other devices may implement at least a portion of the analysis platform 106 and / or a prediction system 114.[0058J In one or more implementations, the monitoring device 104 is configured to transmit the measurements 108 to an external device over a wired connection with the external device, e.g., via USB-C or some other physical, communicative coupling. Here, a connector may be plugged into the monitoring device 104 or the monitoring device 104 may be inserted into an apparatus having a receptacle that interfaces with corresponding contacts of the device. The measurements 108 may then be obtained from storage of the monitoring device 104 via this wired connection, e.g., transferred over the wired connection to the external device. Such a connection may be used in scenarios where the monitoring device 104 is mailed by the person 102 after the observation period, such as to a health care provider, telemedicine service, provider of the monitoring device 104, or medical testing laboratory.

[0059] Alternatively or additionally, the monitoring device 104 may provide the measurements 108 to the analysis platform 106 by communicating the measurementsFIG. 1 Patents 18 Docket No.: 10030PCT1108 over one or more wireless connections. For example, the monitoring device 104 may wirelessly communicate the measurements 108 to external computing devices, such as a mobile phone, tablet device, laptop, smartwatch, other wearable health tracker, and so on. Accordingly, the monitoring device 104 may be configured to communicate with external devices using one or more wireless communication protocols or techniques. By way of example, the monitoring device 104 may communicate with external devices using one or more of Bluetooth (e.g., Bluetooth Low Energy links), near-field communication (NFC), Long Term Evolution (LTE) standards such as 5G, and so forth. Monitoring devices 104 may be configured with corresponding antennae and other wireless transmission means in scenarios where the measurements 108 are communicated to an external device for processing. In those scenarios, the measurements 108 may be communicated to the analysis platform 106 in various manners, such as at predetermined time intervals (e.g., every day, every hour, or every five minutes), responsive to occurrence of some event (e.g., filling a storage buffer of the monitoring device 104), or responsive to an end of an observation period, to name just a few.

[0060] Thus, regardless of where the analysis platform 106 is implemented (e.g., at the monitoring device 104, at a smartphone associated with the person 102, or at a server device), the analysis platform 106 obtains the measurements 108 produced by the monitoring device 104. In one or more implementations, the analysis platform 106 also obtains other measurements produced by the monitoring device 104 and / or any other devices used during the observation period, e.g., a smartwatch, chest strap, etc.FIG. 1 Patents 19 Docket No.: 10030PCT1

[0061] In one or more implementations, the analysis platform 106 may be implemented in whole or in part at the monitoring device 104. Alternately or additionally, the analysis platform 106 may be implemented in whole or in part using one or more computing devices external to the monitoring device 104, such as one or more computing devices associated with the person 102 (e.g., a mobile phone, tablet device, laptop, desktop, or smartwatch) or one or more computing devices associated with a service provider (e.g., a health care provider, a telemedicine service, a service corresponding to the provider of the monitoring device 104, a medical testing laboratory service, and so forth). In the latter scenario, the analysis platform 106 may be implemented at least in part on one or more server devices.

[0062] In the illustrated example 100, the analysis platform includes storage device 112. In accordance with the described techniques, the storage device 112 is configured to maintain the measurements 108 and / or other measurements or information processed by the prediction system 114 to generate one or more predictions 110. The storage device 112 may represent one or more databases and also other types of storage capable of storing the measurements 108 and / or other types of measurements. The storage device 112 may also store a variety of other data, such as personal information, demographic information describing the person 102, information about a health care provider, information about an insurance provider, payment information, prescription information, determined health indicators, account information (e.g., username and password), and so forth. The storage device 112 may also maintain data of other users of a user population.FIG. 1 Patents 20 Docket No.: 10030PCT1

[0063] In the illustrated example 100, the analysis platform 106 also includes the prediction system 114. The prediction system 114 represents functionality to process the measurements 108 to generate the one or more prediction(s) 110. Alternatively or in addition, the prediction system 114 may output one or more time sequences indicating an observation or prediction of one or more events, over time. It is also to be appreciated that the prediction system 114 may output different combinations of multiple predictions in variations.

[0064] In at least one implementation, the prediction system 114 uses machine learning to generate one or more predictions 110. By way of example and not limitation, the prediction system 114 may include one or more neural networks trained based on the historical measurements and the historical outcome data of a user population. The prediction system 114 may include one or multiple machine learning models (e.g., an ensemble of models). Alternatively or additionally, the prediction system 114 may include logic (a machine learning model and / or other types of logic) to pre-process the obtained measurements, such as to extract various cardiovascular and / or other features from the sequences of measurements. The illustrated example 100 also includes prediction(s) 110, which corresponds to the output of the prediction system 114.

[0065] In various examples, the prediction system 114 is representative of and / or includes one or more components of a noise management system 116. In additional or alternative examples, the prediction 110 includes and / or is representative of a noise condition 118. For instance, as further described in more detail below the noise management system 116 is operable to implement multichannel ambient interferenceFIG. 1 Patents 21 Docket No.: 10030PCT1detection techniques to detect, analyze, and / or mitigate various types of environmental interference that may impact an accuracy of physiological measurements, e.g., the measurements 108, collected by the monitoring device 104. In one or more examples, the noise management system 116 is operable to detect, determine, and / or generate the noise condition 118. The noise condition 118, for instance, indicates whether environmental interference is present that would impact (e.g., degrade) an accuracy of the measurements 108.

[0066] FIG. 2 depicts a nonlimiting example 200 of a monitoring device. The illustrated example 200 depicts the monitoring device 104.

[0067] In accordance with the described techniques, the monitoring device 104 includes one or more sensors 202, examples of which include but are not limited to one or more pairs of electrodes, an accelerometer, a pulse oximeter, and sweat sensors, to name just a few. The monitoring device 104 may also include a transmitter 204. In this example 200, the monitoring device 104 further includes one or more adhesive portions 206. In operation, the monitoring device 104 is configured to be applied to the skin via the one or more adhesive portions 206, such that, for example, the one or more sensors 202 are positioned to detect and record the electrical activity of the person 102’ s heart, e.g., to produce an electrocardiogram (ECG and / or EKG). In at least one implementation, the monitoring device 104 may be removed by peeling the one or more adhesive portions 206 off of the skin.

[0068] It is to be appreciated that the monitoring device 104 and its various components are simply one form factor, and the monitoring device 104 and itsFIG. 1 Patents 22 Docket No.: 10030PCT1components may have different form factors without departing from the spirit or scope of the described techniques.

[0069] In one or more implementations, the monitoring device 104 may include a processor and / or memory (not shown). The monitoring device 104, by leveraging the processor, may generate the measurements 108 based on the communications with one or more sensors 202 that are indicative of some aspect of the person 102, such as the person 102’ s heart electrical activity, respiration conditions, changes in blood volume, and so forth. In one or more implementations, the processor further generates one or more communicable packages of data that include one or more of the measurements 108 and / or other measurements. Alternately or additionally, the processor produces and / or causes storage of other data, which may be used for predicting classifications of physiological conditions, e.g., cardiovascular conditions, respiratory conditions, sleep apnea, etc.

[0070] In implementations where the monitoring device 104 is configured for wireless transmission, the transmitter 204 may transmit the measurements wirelessly as a stream of data to a computing device. In one or more implementations, for instance, the monitoring device 104 is configured to transfer (e.g., transmit and / or receive) information (e.g., electrical potential measurements) via a Bluetooth Low Energy (BLE) connection. Alternately or additionally, the monitoring device 104 may buffer the measurements (e.g., in memory) and cause the transmitter 204 to transmit the buffered measurements later at various intervals, e.g., time intervals (every second, every thirty seconds, every minute, every five minutes, every hour, and so on), storageFIG. 1 Patents 23 Docket No.: 10030PCT1intervals (when the buffered measurements reach a threshold amount of data), and so forth.Multi-Channel Ambient Interference Detection

[0071] FIG. 3 depicts a nonlimiting system in an example implementation 300 of multi-channel ambient interference detection showing operation of the noise management system 116 of FIG. 1 in more detail. In various examples, the noise management system 116 is representative of, supports functionality of, is implementable by, and / or includes (either partially or wholly) a wearable device, such as the monitoring device 104, attachable to a skin surface of a user.

[0072] The noise management system 116, for instance, is operable to detect and manage environmental interference that may affect physiological measurements. To do so, the noise management system 116 is illustrated to include an illumination component 302, an optical sensor 304, a noise detection component 306, and processors 308 that are operable to detect and classify a noise condition 118. The noise management system 116 is further illustrated to include a condition management module 310 that is implementable to perform various functionality based on the noise condition 118.

[0073] To begin in this example, the illumination component 302 and optical sensor 304 are configurable to interact with one another to collect various optical measurements. For instance, the illumination component 302 is configured to generate emitted light 312 that penetrates the skin surface to interact with subdermal tissue of the user. In various examples, the illumination component 302 may cause an illuminatedFIG. 1 Patents 24 Docket No.: 10030PCT1condition by emitting light towards the skin surface and may cause a non-illuminated condition by ceasing light emission and / or refraining from emitting light.

[0074] In at least one example, the illumination component 302 includes one or more LEDs that can emit light at various wavelengths. The illumination component 302 can be configured as a single LED, an array of LEDs, or other light-emitting elements capable of generating controlled light emissions. In some implementations, the illumination component 302 may include multiple LEDs that emit light at different wavelengths, such as red light (approximately 660 nm) and infrared light (approximately 940 nm), such as to penetrate tissue at variable depths. Particular wavelengths may be selected based on various considerations, such as skin pigmentation characteristics, measurement objectives, power efficiency considerations, and so forth. This is by way of example and not limitation, and various types and configurations of the illumination component 302 are considered.

[0075] The optical sensor 304, for instance, is configured to detect light reflected through the skin surface, e.g., reflectance data 314, which includes an amount of the emitted light 312 that is reflected back through the skin surface. The optical sensor 304 may include one or more of a photodiode, a phototransistor, a photoresistor, a charge- coupled device (CCD), or other light-sensitive components capable of converting light energy into electrical signals. In various implementations, the optical sensor 304 may be configured with one or more spectral sensitivities to detect particular wavelengths of light.

[0076] For instance, an optical sensor 304 configured for a PPG task is configured with a spectral sensitivity to detect red light (approximately 660 nm) and / or infraredFIG. 1 Patents 25 Docket No.: 10030PCT1light (approximately 940 nm). The optical sensor 304 may also include one or more filtering components to reduce noise and / or enhance signal quality, such as optical filters that selectively pass certain wavelengths while blocking others, or electronic filtering circuitry integrated with the sensor. In at least one example, the illumination component 302 and the optical sensor 304 are integrated into a singular component such as an integrated photodiode.

[0077] The optical sensor 304 may be coupled to various components, such as an analog-to-digital circuit to transform detected light signals into digital data that can be processed by the processors 308. Additionally or alternatively, the optical sensor 304 may be configured with variable gain settings or sampling rates that can be dynamically adjusted based on environmental conditions or measurement objectives. In some implementations, multiple optical sensors may be arranged in an array configuration such as to provide spatial diversity in measurements or to enable differential sensing approaches that can further improve signal quality.

[0078] In one or more examples, the illumination component 302 and the optical sensor 304 are positioned on a same side of the wearable device, e.g., adjacent to one another and facing the skin surface of the user. This configuration allows for efficient collection of reflectance data 314 based on interactions of the emitted light 312 with tissue of the user. Additionally or alternatively, the illumination component 302 and / or the optical sensor 304 may be positioned at different locations on the wearable device, such as with the illumination component 302 positioned on a flexible extension that maintains contact with the skin surface while the optical sensor 304 is located on a main body of the device. In some cases, multiple optical sensors 304 may be distributedFIG. 1 Patents 26 Docket No.: 10030PCT1across different areas of the wearable device to provide enhanced measurement coverage and / or to enable comparative analysis between different measurement locations.

[0079] The noise detection component 306 is configured to sample an environmental condition. For instance, the noise detection component 306 collects environmental condition data 316, which may include information about ambient light conditions, electromagnetic interference, and / or other environmental factors that may impact the accuracy of physiological measurements. As further described in more detail below, the environmental condition data 316 is processed to determine a presence and / or characteristics of potential interference sources.

[0080] The noise detection component 306 can be configured to measure a variety of environmental conditions, such as one or more of ambient light fluctuations, electromagnetic interference, radio frequency interference, mechanical vibrations, or other external factors that may impact accuracy of measurements collected by the optical sensor 304. In one or more examples, the noise detection component 306 detects one or more conditions of ambient light, such as a frequency, intensity, flicker rate, spectral composition, polarization, temporal pattern, etc. of the ambient light. In additional or alternative examples, the noise detection component 306 detects one or more properties of an environmental electromagnetic signal, such as a frequency, amplitude, field strength, polarization, phase, temporal pattern, spectral composition, source direction, etc. These properties can be used to characterize electromagnetic interference that may affect the accuracy of physiological measurements collected by the optical sensor 304.FIG. 1 Patents 27 Docket No.: 10030PCT1

[0081] In at least one example, the noise detection component 306 includes a photodiode 318 configured to detect ambient light interference. The photodiode 318 can be disposed to face away from the skin surface of a user, such as to be positioned to collect a signal from one or more environmental light sources. Alternatively or additionally, the photodiode 318 may be positioned on a side surface of the wearable device such as to capture ambient light from one or more lateral directions, and / or integrated within a transparent or translucent portion of the device housing to monitor environmental conditions while maintaining a compact form factor. In some embodiments, multiple photodiodes 318 may be distributed at different orientations around the wearable device such as to provide comprehensive ambient light monitoring from various angles and directions.

[0082] A variety of configurations, types, and geometries of photodiode 318 are considered. In at least one example, the optical sensor 304 includes a first photodiode and the noise detection component 306 includes a second photodiode, e.g., the photodiode 318. The first photodiode and second photodiode may be configured with various spectral sensitivities, response times, sizes, and / or physical orientations based on respective detection objectives. For instance, the first photodiode is configured to detect particular wavelengths of light reflected from tissue, while the second photodiode may be configured with a relatively broader spectral sensitivity such as to capture a variety of ambient light sources.

[0083] The first photodiode and the second photodiode can also be configured with variable sizes. In an example, the second photodiode is smaller than the first photodiode, such as to conserve device space and computational resource expenditureFIG. 1 Patents 28 Docket No.: 10030PCT1while providing sufficient sensitivity to detect ambient light. The photodiodes may have various relative spectral sensitivities (e.g., a range of wavelengths detectable by each respective sensor) and / or variable absolute sensitivities (e.g., a minimum intensity of a stimulus to produce a detectable change) such as based on their respective detection purposes. For example, the first and second photodiode may have similar spectral sensitivity, such as to detect a similar range of wavelengths, however the first photodiode may have a greater absolute sensitivity to detect minute physiological changes. The second photodiode may also be configured with particular filtering characteristics to enhance detection of particular interference patterns or frequencies that are known to impact accuracy of physiological measurements.[0084J The noise detection component 306 may also include one or more components implementable to detect electromagnetic interference. For example, the noise detection component 306 may incoiporate one or more of a capacitor 320, a resistor 322, or a trace configuration 324 that cause measurable changes to one or more metrics or electrical characteristics in response to changes in an electromagnetic field proximate to the wearable device. These components can be arranged in various circuit configurations to detect various properties of electromagnetic interference based on a comparison between expected values and measured responses.

[0085] For instance, the capacitor 320 may be configured to detect changes in electric field strength by measuring variations between a known value for charge accumulation and a measured value for charge accumulation. The resistor 322 may be used to detect changes in an amount of current induced by electromagnetic interference, e.g., one or more fluctuating magnetic fields. The trace configuration 324 may include particularFIG. 1 Patents 29 Docket No.: 10030PCT1routing patterns (e.g., conductive pathways with one or more loop antennas, differential pairs, interdigitations, etc.) that are designed to be sensitive to electromagnetic interferences.

[0086] In at least one example, the trace configuration 324 is configured to detect interference within one or more frequency ranges. For instance, a first routing pattern of the trace configuration 324 is implemented to detect interference within a first frequency range, while a second routing pattern is implemented to detect interference within a second frequency range. In this way, the wearable device is able to detect various characteristics of environmental interference without incorporation of additional discrete hardware components, which conserves device real-estate and supports a form factor with reduced size.

[0087] The noise detection component 306 can be configured with a variety of electromagnetic interference detection characteristics, including but not limited to a frequency response range, sensitivity threshold, directional selectivity, and / or a signal- to-noise ratio. For example, the capacitor 320 may be selected with particular dielectric properties to optimize sensitivity to particular frequency ranges, such as those commonly produced by fluorescent lighting (e.g., 20-60 kHz) or switching power supplies (e.g., 10-300 kHz). Similarly, the resistor 322 may be configured with particular impedance characteristics to maximize sensitivity to induced currents.

[0088] The trace configuration 324 may be designed with various geometric properties, such as trace width, spacing, routing geometry, and / or orientation, such as to enhance coupling with external electromagnetic fields while maintaining a compact form factor within the wearable device. In at least one example, the trace configurationFIG. 1 Patents 30 Docket No.: 10030PCT1324 includes an unterminated trace, e.g., a detection trace, that is positioned adjacent to and conforming to a shape of a trace that corresponds to the optical sensor 304, e.g., a sensor trace. In this way, electromagnetic interference that impacts the sensor trace has a substantially similar impact on the detection trace.

[0089] The processors 308 are implemented by the noise management system 116 to perform a variety of functionality, such as by leveraging one or more of the components of the noise management system 116, e.g., the illumination component 302, optical sensor 304, noise detection component 306, and / or condition management module 310 to generate the noise condition 118. In various examples, the processors 308 include and / or are representative of one or more microcontrollers, amplifiers, and / or other signal processing elements. The processors 308 may further support multi-channel input / output, such as to receive the reflectance data 314 via a first channel associated with the optical sensor 304 and environmental condition data 316 via a second channel associated with the noise detection component 306.

[0090] Further, it should be understood that the noise management system 116 may include various additional or alternative hardware components usable to perform one or more aspects of multi-channel ambient interference detection, such as but not limited to electromagnetic shielding elements, analog filtering circuits, digital signal processors, field-programmable gate arrays (FPGAs), application- specific integrated circuits (ASICs), analog-to-digital converters (ADC’s) with variable sampling rates, optical filters, spectral sensors, temperature sensors to compensate for thermal drift, motion sensors to detect movement artifacts, dedicated power management circuitry, and so forth.FIG. 1 Patents 31 Docket No.: 10030PCT1

[0091] In the illustrated example, the processors 308 are configured to receive a first signal 326 from the optical sensor 304 that includes the reflectance data 314 and receive a second signal 328 from the noise detection component 306 that includes the environmental condition data 316. The processors 308 can generate physiological data 330 based on the first signal 326 and can generate interference data 332 based on the second signal 328. As described in the following discussion, the processors 308 are operable to detect a noise condition 118 based on one or more of the physiological data 330 and / or the interference data 332 that characterizes environmental interference, such as to indicate whether environmental interference is present that impacts an accuracy of the physiological data 330.

[0092] The physiological data 330 may include various types of biological measurements derived from the first signal 326. For example, the physiological data 330 may include heart rate measurements, blood oxygen saturation levels, pulse transit time, respiratory rate, blood pressure estimates, cardiac rhythm patterns, and / or other cardiovascular parameters. In some cases, the physiological data 330 may also include derived metrics such as heart rate variability, perfusion index, or other calculated health indicators based on the reflectance data 314.

[0093] The processors 308 can generate the physiological data 330 by processing the first signal 326 using various signal processing techniques, such as filtering operations to remove noise, amplification to enhance signal strength, analog-to-digital conversion to digitize the optical measurements, and / or algorithmic processing to extract physiological features from the digitized signals. In some implementations, the processors 308 may apply one or more denoising algorithms, such as an ambient lightFIG. 1 Patents 32 Docket No.: 10030PCT1rejection process that compares illuminated and non-illuminated samples to isolate physiological signals from environmental interference.

[0094] In an example to do so, the processors 308 cause the illumination component 302 to alternate between illuminated and non-illuminated conditions while synchronously sampling the first signal 326 using the optical sensor 304. During an illuminated condition, the optical sensor 304 generates the reflectance data 314 based on both an amount of the emitted light 312 that is reflected from tissue as well as ambient light present in the environment. During a non-illuminated condition, the reflectance data 314 is based on the ambient light while the illumination component 302 is inactive. The processors 308 may then subtract the non-illuminated sample values from the illuminated sample values to attempt to isolate the physiological signal component by removal of the ambient light contribution from the physiological data 330.

[0095] However, as described above, the ambient light rejection is susceptible to failure under various conditions. Accordingly, the processors 308 generate the interference data 332 by analyzing characteristics of the second signal 328 to identify patterns, frequencies, amplitudes, and other properties indicative of environmental interference that may compromise measurement accuracy. Accordingly, the interference data 332 may include one or more qualitative or quantitative metrics such as numerical values and / or statistical measures that characterize a magnitude and / or properties of detected interference.

[0096] In various implementations, the processors 308 may calculate one or more of these metrics by applying various signal processing techniques, e.g., Fast FourierFIG. 1 Patents 33 Docket No.: 10030PCT1Transforms (“FFT”) to identify dominant frequency components, time-domain analysis to detect rapid fluctuations, and / or comparative analysis between expected baseline values and measured values to determine deviation magnitudes. In some implementations, the processors 308 may implement machine learning techniques to analyze the second signal 328 to generate the interference data 332, such as to detect particular interrelated data features included in the environmental condition data 316. In some examples, the processors 308 may classify the interference data 332 into different categories based on interference type (e.g., ambient light fluctuations, electromagnetic interference, or mechanical artifacts), severity levels (e.g., low, medium, or high), or temporal characteristics (e.g., continuous, intermittent, or periodic), and so forth.

[0097] In various implementations, the processors 308 may coordinate timing of sampling between the first signal 326 and the second signal 328 such as to enhance interference detection accuracy. For example, the processors 308 may sample the first signal 326 and the second signal 328 substantially simultaneously to concurrently capture environmental conditions and physiological measurements. Alternatively or additionally, the processors 308 may implement a time-offset sampling approach to sample the second signal 328 at predetermined intervals offset from collection of the first signal 326, such as to monitor environmental conditions continuously while collecting physiological data at different rates. The sampling timing may also be dynamically adjusted based on detected interference characteristics, such as increasing the sampling frequency of the noise detection component 306 when high-frequencyFIG. 1 Patents 34 Docket No.: 10030PCT1interference is detected and / or synchronizing sampling with known interference patterns to improve characterization of environmental conditions.

[0098] Based on the interference data 332 and / or the physiological data 330, the processors 308 determine a noise condition 118. The noise condition 118, for instance, represents a characterization of environmental interference that may affect the accuracy or reliability of the physiological data 330. The noise condition 118 may include various types of information about detected interference, such as but not limited to a binary indication of whether interference is present above a threshold level, a quantitative measure of interference magnitude or severity (e.g., a percentage, value, etc.), a qualitative measurement (e.g., “high”, “medium”, “low” classifiers), a classification of interference type (e.g., ambient light flicker, electromagnetic interference, motion artifact, etc.), a frequency spectrum analysis of detected interference, a temporal pattern description (e.g., continuous, periodic, intermittent, etc.), a confidence level associated with the interference detection, a predicted impact on measurement accuracy, and / or a recommendation for mitigation actions.

[0099] The processors 308 may implement various strategies to determine the noise condition 118. In some implementations, the processors 308 may compare a characteristic of the interference data 332 against a threshold to determine whether interference levels exceed acceptable limits for reliable physiological measurements. For instance, if one or more metrics such as a frequency of ambient light, magnitude of electromagnetic interference, amplitude variations, temporal pattern characteristics, or signal-to-noise ratio exceed a respective threshold, the processors 308FIG. 1 Patents 35 Docket No.: 10030PCT1generate the noise condition 118 to indicate that environmental interference is present capable of decreasing accuracy of the physiological data 330.

[0100] In at least one example, the threshold is based in part or in whole on a sampling interval (e.g., a temporal gap between collection of an illuminated sample and a nonilluminated sample) used by the processors 308 in an ambient light rejection process. For instance, if a frequency of ambient light causes fluctuations that occur faster than the sampling interval, the ambient light rejection process will fail to effectively denoise the reflectance data 314, leading to inaccurate physiological data 330. Accordingly, in this example the threshold is based on an inverse of a temporal gap between collection of an illuminated sample and a non-illuminated sample during the ambient noise rejection process.

[0101] In some examples, the processors 308 generate the noise condition 118 based on two or more metrics. By way of example, the processors 308 may determine that ambient light has a frequency above a threshold that would otherwise indicate interference, however a magnitude of the ambient light fluctuations is sufficiently low such that the impact on the physiological data 330 would be relatively insignificant. Accordingly, the processors 308 can generate the noise condition 118 to indicate that while high-frequency interference is present, the interference level does not significantly compromise measurement accuracy.

[0102] In one or more examples, the processors 308 may implement a weighted scoring system that combines multiple interference metrics to generate a composite interference assessment. For instance, the processors 308 may calculate a weighted sum for multiple interference characteristics, e.g., frequency characteristics, amplitudeFIG. 1 Patents 36 Docket No.: 10030PCT1variations, and temporal stability, that each contribute to an overall interference score. Accordingly, the noise condition 118 in this example is based on a multi-metric assessment of interference characteristics. This is not possible using conventional techniques that are constrained to detection of a single interference metric.

[0103] In some examples, the noise condition 118 is determined based on the physiological data 330 as well as the interference data 332. For instance, the processors 308 may determine the noise condition 118 based on a relationship between a type of physiological data 330 and a type of interference data 332. By way of example, heart rate measurements derived from PPG may be particularly susceptible to ambient light flicker in the 50-60 Hz range, while blood oxygen saturation measurements may be adversely impacted by electromagnetic interference from devices operating in the 2.4 GHz band. Accordingly, the processors 308 may weigh different interference characteristics differently based on a particular physiological parameter being measured by the physiological data 330, such that a high-frequency ambient light condition may trigger a noise condition 118 for heart rate monitoring but may not significantly impact respiratory rate calculations that rely on relatively lower frequency signal components.

[0104] The processors 308 may also implement correlation analysis between the first signal 326 and second signal 328 to identify instances where environmental interference is directly impacting the physiological measurements, such as when both signals exhibit similar noise patterns that indicate a common interference source. For instance, the processors 308 may detect a particular oscillation pattern in the physiological data 330 that corresponds to a specific frequency component identified in the interference data 332. In this way the processors 308 can confirm that observed physiological signalFIG. 1 Patents 37 Docket No.: 10030PCT1artifacts are attributable to environmental interference rather than physiological events. Accordingly, the processors 308 can generate the noise condition 118 based on a variety of considerations to indicate various characteristics of detected interference as it relates to the physiological data 330 which is subsequently usable to perform a variety of functionality.

[0105] For instance, the condition management module 310 may receive the noise condition 118 from the processors 308 and implement various responsive actions, such as to manage detected environmental interference or maintain execution of system operation. These actions may include one or more of implementing a processing adjustment 334, a hardware reconfiguration 336, or a system operation modification 338 and can be based on various aspects of the noise condition 118, such as a type of interference detected, a magnitude of the interference, and / or additional properties of the interference. In various examples, the condition management module 310 is able to leverage one or more components of the noise management system 116 to perform the various functionality.

[0106] For example, the condition management module 310 can implement a processing adjustment 334 based on the noise condition 118. The processing adjustment 334, for instance, includes a modification to how data is processed by the one or more processors 308. For example, the processing adjustment 334 may include one or more of refinement of the physiological data 330 based on the noise condition 118, such as by removing noise from the first signal 326 based on the second signal328.FIG. 1 Patents 38 Docket No.: 10030PCT1

[0107] In some implementations, the condition management module 310 may implement the processing adjustment 334 in real-time during data collection, such as by dynamically adjusting filtering parameters, modifying signal processing algorithms, and / or applying adaptive noise cancellation techniques as environmental conditions change. For instance, when a particular noise condition 118 indicates high-frequency ambient light interference, the condition management module 310 may adjust a signal processing parameter to reduce an impact of detected interference on ongoing physiological measurements.

[0108] Additionally or alternatively, the processing adjustment 334 may include operations for post-processing analysis. For instance, the condition management module 310 may flag, weight, or filter flag portions of the physiological data 330 that correspond to temporal intervals where interference is present. In this way, the condition management module 310 is able to exclude potentially inaccurate data from physiological parameter calculations in real-time and / or during post-processing data analysis.

[0109] The condition management module 310 may also implement a hardware reconfiguration 336. The hardware reconfiguration 336 may involve adjusting hardware settings or parameters of one or more components of the noise management system 116. For, the condition management module 310 can adjust parameters of the illumination component 302, optical sensor 304, noise detection component 306, and / or various circuit components based on the noise condition 118. By way of example, the condition management module 310 may increase an intensity of the illuminationFIG. 1 Patents 39 Docket No.: 10030PCT1component 302 when the noise condition 118 indicates high ambient light interference, such as to improve a signal-to-noise ratio of the first signal 326.

[0110] In some implementations, the condition management module 310 may adjust a sampling rate of the optical sensor 304 and / or the noise detection component 306 based on detected interference characteristics. For example, the condition management module 310 may increase a sampling rate of the noise detection component 306 when interference is detected, such as to provide enhanced characterization of the interference. In this example, the condition management module 310 may also increase a sampling frequency of the optical sensor 304, such as to overcome an impact of the interference and increase accuracy of the physiological data 330.

[0111] In an alternative or additional example, the condition management module 310 may suspend or reduce a frequency of physiological data collection when the noise condition 118 indicates that environmental interference would render the physiological data 330 unreliable, such as to conserve computational resources. For instance, when ambient light interference or electromagnetic interference above a threshold is detected, the condition management module 310 may pause data collection operations to avoid processing and storing inaccurate measurements that would consume processing power without providing meaningful physiological insights. In some implementations, the condition management module 310 may reduce a rate of data collection during interference periods to allow the noise management system 116 to maintain a reduced level of monitoring while conserving battery life and computational resources that would otherwise be expended on processing noisy data. This is by way of example andFIG. 1 Patents 40 Docket No.: 10030PCT1not limitation, and various hardware reconfigurations 336 that correspond to various scenarios are considered.

[0112] The condition management module 310 is further operable to implement one or more system operation modifications 338 based on the noise condition 118. The system operation modification 338 may involve changing system operations of the noise management system 116. For example, the system operation modification 338 may include temporarily suspending PPG measurements to conserve power when noise levels above a threshold are detected. In some implementations, the system operation modification 338 involves varying a sampling rate of noise detection independently from a main PPG sampling rate.

[0113] In some examples, the system operation modification 338 includes presentation of an indication of the noise condition 118. The indication can include various content, such as one or more of a visual notification, audio alert, or haptic feedback. The indication may be presented to a user via a display and / or other output component of a device incorporating the noise management system 116. For example, the condition management module 310 may cause display of a warning icon on a device screen when high-frequency ambient light is detected, and / or cause a wearable device to vibrate when electromagnetic interference exceeds a threshold level.

[0114] In at least one example, the system operation modification 338 includes implementation of a low-power mode responsive to the noise condition 118. For example, the condition management module 310 deactivates the illumination component 302 and ambient light rejection circuitry, such that both the optical sensor 304 and noise detection component 306 measure illumination due to ambient lightFIG. 1 Patents 41 Docket No.: 10030PCT1independent of emitted light 312 from the illumination component 302. The condition management module 310 can then process a signal from the optical sensor 304 based on a signal from the noise detection component 306 (e.g., divide the first signal 326 by the second signal 328) to generate physiological data 330, e.g., a low-power PPG reading. In this way, the noise management system 116 conserves computational resources while maintaining physiological monitoring capabilities during periods of environmental interference.

[0115] In various examples, the action performed by the condition management module 310 are based on properties of the interference included in the noise condition 118, such as a type, magnitude, temporal interval, rate, etc. of the interference. For instance, the noise condition 118 may include categorizations for the detected interference, such as low, medium, or high interference levels. Each respective categorization may trigger a corresponding response from the condition management module 310.

[0116] Alternatively or additionally, the action may be based on a type of interference detected. For instance, the noise management system 116 can perform a first action based on detection of ambient light interference and can perform a second action different from the first action based on detection of electromagnetic interference. In this way, the noise management system 116 is able to dynamically provide tailored responses to manage various types of interference, which is not possible using conventional noise detection approaches.FIG. 1 Patents 42 Docket No.: 10030PCT1

[0117] FIGS. 4a and 4b depict nonlimiting examples 400a and 400b of multi-channel ambient interference detection in which variations in lighting conditions can cause an ambient light rejection process to fail.

[0118] For instance, FIG. 4a illustrates an example 400a of a wearable device 402 positioned adjacent to a skin surface 404 and tissue 406 of a user. The wearable device 402 includes an illumination component 302 and an optical sensor 304, as well as various hardware to support operation of the illumination component 302 and optical sensor 304 which is not depicted.

[0119] The illumination component 302 is configured to generate emitted light 312 that penetrates through the skin surface 404 into the tissue 406. The optical sensor 304 is positioned to detect reflected light 408 that emerges back through the skin surface 404 after interaction with the tissue 406. For instance, the optical sensor 304 includes a first photodiode positioned adjacent to the illumination component 302 that faces the skin surface 404 to detect the reflected light 408.

[0120] Ambient light 410 from external sources also illuminates the skin surface 404 and tissue 406, which contributes to the reflected light 408 detected by the optical sensor 304. In this example, the ambient light 410 is due to low intensity sunlight from the sun that has a relatively constant magnitude with a relatively low frequency. Thus, a detected signal from the optical sensor 304 is influenced by both the emitted light 312 and the ambient light 410, which may complicate isolation of physiological data 330 from environmental interference. Conventional techniques may implement ambient light rejection processes that attempt to denoise a signal obtained by the optical sensor 304.FIG. 1 Patents 43 Docket No.: 10030PCT1

[0121] For instance, FIG. 4b illustrates an example 400b of a timing and sampling process used in a conventional denoising approach, such as an ambient light rejection process for PPG measurements. The example 400b continues from FIG. 4a and includes a graph that represents how a wearable device 402 modifies states of an illumination component 302 and an optical sensor 304 for denoising using ambient noise rejection. The example 400b depicts this process under a first lighting condition 412, e.g., with a relatively constant ambient light source with low frequency, and a second lighting condition 414, such as a light source with relatively high frequency ambient light.

[0122] In the illustrated example, the “LED State” indicates when the illumination component 302 is active (“ON”) or inactive (“OFF”), with two distinct illumination periods labeled as a first illumination period 416, e.g., during a portion of the first lighting condition 412, and a second illumination period 418, e.g., during a portion of the second lighting condition 414. The “Optical Sensor Sampling State” portion of the graph demonstrates sampling windows during which measurements are taken by the optical sensor 304. These sampling windows alternate between illuminated samples and ambient samples.

[0123] For instance, under the first lighting condition 412, a first illuminated sample 420 is taken during the first illumination period 416 when the illumination component 302 is active, e.g., “ON”. This is followed by a first ambient sample 422 taken during a non- illuminated condition when the illumination component 302 is inactive, e.g.,“OFF”. Under the second lighting condition 414, a second illuminated sample 424 isFIG. 1 Patents 44 Docket No.: 10030PCT1taken during the second illumination period 418, followed by a second ambient sample426 during a non-illuminated condition.

[0124] Illuminated samples are separated from corresponding ambient samples by a temporal sampling gap 428. For instance, the temporal sampling gap 428 represents a time between collection of the first illuminated sample 420 and the first ambient sample 422 by the optical sensor 304, and a time between collection of the second illuminated sample 424 and the second ambient sample 426. In various examples, the temporal sampling gap 428 is between 10-200 microseconds.

[0125] Using conventional denoising approaches, e.g., ambient noise rejection, values of the ambient samples (e.g., the first ambient sample 422 and the second ambient sample 426) are subtracted from values of corresponding illuminated samples (e.g., the first illuminated sample 420 and the second illuminated sample 424) such as to attempt to isolate a contribution of the emitted light 312 from the illumination component 302 from an impact of the ambient light.

[0126] However, this process is reliant on consistency of ambient light conditions between the illuminated and ambient samples. Accordingly, conventional denoising approaches break down when ambient light fluctuates rapidly, such as during the second lighting condition 414 where the ambient light has a relatively high frequency (e.g., greater than or equal to 1 kHz) and a relatively short wavelength 430. For instance, during the second lighting condition 414 the temporal sampling gap 428 is greater than the wavelength 430.

[0127] Accordingly, the ambient light conditions change significantly between the illuminated and ambient samples during the second lighting condition 414. Thus,FIG. 1 Patents 45 Docket No.: 10030PCT1subtraction of the ambient sample from the illuminated sample may not accurately remove ambient light interference. This limitation of conventional denoising modalities causes residual noise to be included in signals obtained by optical sensors, such as when the wearable device 402 is used in environments with rapidly changing light conditions or in the presence of high-frequency light sources, leading to inaccurate physiological measurements.

[0128] Accordingly, the multichannel ambient interference detection techniques as described herein overcome the limitations of conventional techniques by incorporating a signal from a noise detection component 306 to detect properties of ambient interference to determine when ambient light rejection techniques are insufficient to accurately generate the physiological data 330.

[0129] For instance, FIG. 5 depicts a nonlimiting example 500 of multi-channel ambient interference detection in which a wearable device includes a noise detection component configured to sample an ambient condition.

[0130] In this example, the wearable device 502 is positioned adjacent to a skin surface 404 overlying tissue 406 of a user. In various implementations, the wearable device 502 may be implemented as a smartwatch, fitness band, medical patch, or ring. The wearable device 502 is illustrated to include a noise management circuit 504, an illumination component 302, an optical sensor 304, and a noise detection component 306.

[0131] The noise management circuit 504, for instance includes various components to support operation of the noise management system 116, such as one or more processors 308, analog-to-digital converters, bioamplifiers, and so forth. In one or moreFIG. 1 Patents 46 Docket No.: 10030PCT1examples, the optical sensor 304 is connected to a first channel of the noise management circuit 504 and the noise detection component 306 is connected to a second channel of the noise management circuit 504.

[0132] The illumination component 302 is configured to emit light, e.g., emitted light 312, that penetrates the skin surface 404 and tissue 406. The optical sensor is positioned to detect reflected light 506 that emerges back through the skin surface 404 after interaction with the tissue 406. For instance, the optical sensor 304 includes a first photodiode adjacent to the illumination component 302 that faces the skin surface 404 to detect the reflected light 506.

[0133] Ambient light 508 from external sources also illuminates the skin surface 404 and tissue 406, contributing to the reflected light 506 detected by the optical sensor 304. In this example, the ambient light 508 represents relatively high frequency light, such as flicker from artificial lighting sources. Accordingly, an amount of reflected light 506 detected by the optical sensor 304 is based on both the emitted light 312 from the illumination component 302 and the ambient light 508 from the environment.

[0134] The noise detection component 306 is oriented to detect the ambient light 508 from the environment. In some implementations, the noise detection component 306 includes a second photodiode, e.g., an ambient photodiode, that faces away from the skin surface 404 to monitor ambient conditions. In this example, the components of the wearable device 502 are arranged such that the illumination component 302 and optical sensor 304 are positioned on the same side of wearable device 502 facing the skin surface 404, while the noise detection component 306 faces away from the skin surfaceFIG. 1 Patents 47 Docket No.: 10030PCT1404 to monitor ambient conditions. This configuration allows for simultaneous collection of reflectance data 314 and environmental condition data 316.

[0135] In accordance with the techniques described above, the noise management circuit 504 processes signals from these components, e.g., the first signal 326 and the second signal 328, to determine whether conventional denoising processes are insufficient to account for the ambient light 508. For instance, the noise management circuit 504 determines that a wavelength of the ambient light 508 is shorter than a sampling gap used for an ambient noise rejection process, and thus the noise condition 118 indicates that conventional denoising techniques would fail.

[0136] Accordingly, the multichannel approach described herein overcomes limitations of conventional ambient light rejection techniques. By including multiple channels to simultaneously detect physiological measurements as well as ambient interference, the wearable device 502 can identify high-frequency noise conditions, such as flicker from artificial lighting, that may impact the accuracy of the physiological data 330.

[0137] In response to detection of the noise condition 118, the noise management circuit 504 may perform various actions such as adjusting signal processing parameters, modifying sampling rates, or triggering alerts to indicate potential measurement inaccuracies. The noise management circuit 504 may also leverage the data from the noise detection component 306 to refine the physiological data collected by the optical sensor 304, such as to improve an accuracy of the physiological data 330 in the presence of environmental interference.FIG. 1 Patents 48 Docket No.: 10030PCT1

[0138] FIG. 6 depicts a nonlimiting example 600 of multi-channel ambient interference detection that compares photodiode signal data under various ambient light conditions.

[0139] For instance, a first graph 602 depicts photodiode signal data with no ambient flicker present, while a second graph 604 shows photodiode signal data with a 5 kHz ambient flicker present. In this example 600, a wearable device includes a first photodiode of the optical sensor 304 and a second photodiode of the noise detection component 306. The first photodiode captures the skin facing photodiode signal, while the second photodiode captures the ambient photodiode signal.

[0140] As shown in the first graph 602, the ambient photodiode signal indicates that there is little to no ambient light interference present. The first graph 602 further depicts a relatively consistent periodic waveform while no interference is present. This waveform, for instance, represents typical physiological variations, such as “healthy” changes in blood volume associated with the cardiac cycle.

[0141] In contrast, the second graph 604 demonstrates significant signal distortion in both the skin-facing photodiode and ambient photodiode signals when exposed to 5 kHz ambient flicker. For instance, the ambient photodiode signal depicts a waveform with pronounced fluctuations (e.g., relatively high magnitude and short wavelengths) due to the 5 kHz ambient lighting. The skin-facing photodiode signal exhibits sporadic changes in readings, with irregularly spaced data peaks and valleys, such as due to interference from the 5kHz lighting.

[0142] Whereas conventional single-channel denoising approaches may incorrectly interpret the irregular signal patterns shown in the second graph 604 as abnormalFIG. 1 Patents 49 Docket No.: 10030PCT1physiological events, the multichannel techniques described herein analyze the skinfacing and ambient photodiode signals concurrently to determine when abnormal signal characteristics are due to interference rather than unusual physiological events. This capability enables the system to maintain measurement accuracy in the presence of variable environmental conditions.

[0143] FIG. 7 depicts a nonlimiting example 700 of multi-channel ambient interference detection in which various form factors of wearable devices are depicted including a first device configuration 702 and a second device configuration 704.

[0144] In the first device configuration 702, an amplifier 706 is positioned at a first end within a main body of the wearable device. The amplifier 706, for instance, is representative of a dual-channel photodiode amplifier chip such as to be used in PPG measurement. The wearable device in this example further includes a flexible extension 708 that extends from main body of the wearable device along a skin surface 404.

[0145] An optical sensor 304 and an illumination component 302 are located towards an end of the flexible extension 708, e.g., distal from the main body of the wearable device. The optical sensor 304 is connected to the amplifier 706 via a first channel, such as via a trace that runs within the flexible extension 708 towards the main body. The wearable device further includes a noise detection component 306 that is positioned along a top surface of the main body and is connected to the amplifier 706 via a second channel.

[0146] The first device configuration 702 may provide various advantages for physiological monitoring applications. The flexible extension 708 may allow the optical sensor 304 and illumination component 302 to maintain consistent contact withFIG. 1 Patents 50 Docket No.: 10030PCT1the skin surface 404 during movement or activities that can cause displacement. Additionally, positioning the noise detection component 306 on the top surface of the main body may provide exposure to ambient light sources while maintaining physical separation from the skin-facing optical sensor 304, such as to conserve space within the flexible extension 708.

[0147] In the second device configuration 704, the amplifier 706 remains positioned within the main body of the wearable device and the flexible extension 708 includes the optical sensor 304 and the illumination component 302. However, in this arrangement, the noise detection component 306 is also positioned along the flexible extension 708, rather than on the top surface of the main body. Accordingly, the noise detection component 306 is positioned to be subjected to similar conditions as the optical sensor 304 based on the spatial proximity of the noise detection component 306 to the optical sensor 304. In this way, the second device configuration 704 supports enhanced localized detection of environmental interference near the noise detection component 306.

[0148] FIG. 8 depicts a nonlimiting example 800 of multi-channel ambient interference detection in which a portion of a noise detection circuit is configurable to detect electromagnetic interference.

[0149] In the illustrated example 800, a portion of a noise detection circuit includes an illuminated photodiode 802 that is connected to a first trace 804, and a noise detection component 306 that is connected to a second trace 806. The illuminated photodiode 802, for instance, includes the illumination component 302 and the optical sensor 304 in an integrated component and is configured to face a skin surface of a user such as toFIG. 1 Patents 51 Docket No.: 10030PCT1collect physiological measurements. The noise detection component 306 in this example includes a termination component (e.g., a capacitor, diode, or a resistor) coupled to the second trace 806. The noise detection circuit leverages the noise detection component 306 to detect environmental interference, such as by comparing measured electrical property values with expected electrical property values.

[0150] The first trace 804 extends from the illuminated photodiode 802 towards an amplifier, e.g., to serve as a first channel input. The second trace 806 extends from the noise detection component such as to serve as a second channel input. The first trace 804 and the second trace 806 are positioned adjacent to one another in a parallel configuration.

[0151] The particular configuration and layout of the first trace 804 and the second trace 806 enables effective detection of electromagnetic interference. By routing the traces in close proximity to each other, external electromagnetic fields that couple into the second trace 806 and cause detectable changes are likely to couple similarly into the first trace 804. The parallel arrangement thus facilitates identification and measurement of electromagnetic interference that may impact accuracy of physiological measurements.

[0152] FIG. 9 depicts a nonlimiting example 900 of multi-channel ambient interference detection in which a user interface 902 for an interference detection scenario is shown.

[0153] The user interface 902 includes various elements that provide visual feedback about environmental interference that may impact sensor accuracy as well as guidance for improving measurement quality. For instance, an interference alert 904 is displayedFIG. 1 Patents 52 Docket No.: 10030PCT1at a top portion of the user interface 902. The interference alert 904 includes a warning message that indicates interference has been detected that may affect sensor accuracy.

[0154] Below the interference alert 904, a sensor graph 906 is displayed. The sensor graph 906 presents physiological measurement data over time. In this example, the physiological measurements are erratic, such as based on the detected interference. In some cases, the sensor graph 906 may include visual indicators that highlight periods of detected interference, to allow users to correlate interference events with particular time periods or measurement anomalies.

[0155] A guidance section 908 is positioned beneath the sensor graph 906. The guidance section 908 provides information about the detected environmental interference and suggests actions the user can take to improve measurement accuracy. For example, the guidance section 908 recommends for the user to move to a location with different lighting conditions or adjust a position of the wearable device to reduce interference. Accordingly, the user interface 902 enables users to understand when environmental factors may be impacting measurement accuracy and take appropriate steps to improve data quality.

[0156] FIG. 10 depicts a flow diagram depicting an algorithm as a step-by-step procedure 1000 in an example implementation, one or more steps of which are performable by a processing device to determine a noise condition 118 and perform one or more actions based on the noise condition 118.

[0157] To begin, a first signal from an optical sensor and a second signal from a noise detection component of a wearable device attachable to a skin surface of a user are received (block 1002). In various examples, the optical sensor 304 may be configuredFIG. 1 Patents 53 Docket No.: 10030PCT1to detect light reflected through the skin surface 404, while the noise detection component 306 may be configured to sample an environmental condition. The first signal 326 and the second signal 328 may be sampled substantially simultaneously or the second signal 328 may be sampled at a time offset relative to sampling the first signal 326.

[0158] Physiological data is generated based on the first signal that is indicative of one or more biological characteristics of the user (block 1004). The generation of physiological data 330 may involve processing the first signal 326 to extract relevant biological information. In at least one example, the physiological data 330 includes PPG data used to determine blood oxygen saturation levels. In various examples, generation of the physiological data 330 includes implementation of a denoising process, such as an ambient light rejection process.

[0159] The second signal is processed to generate interference data indicative of one or more environmental conditions (block 1006). Generation of the interference data 332, for instance, includes analysis of the second signal 328 to identify patterns or characteristics that suggest a presence of environmental interference. In one or more examples, the interference data 332 may include one or more qualitative or quantitative metrics such as numerical values and / or statistical measures that characterize a magnitude and / or properties of detected interference.

[0160] A noise condition 118 is detected based on the interference data that indicates whether environmental interference is present to impact an accuracy of the physiological data (block 1008). The detection of the noise condition 118 may involve comparing the interference data 332 to predetermined thresholds or patterns known toFIG. 1 Patents 54 Docket No.: 10030PCT1be associated with environmental interference. The noise condition 118 can include one or more of a binary indication of whether interference is present above a threshold level, a quantitative measure of interference magnitude, a classification of interference type and / or severity, a predicted impact on measurement accuracy, and so forth. In some examples, the noise condition 118 is further based on the physiological data 330, such as by analyzing patterns or characteristics in the physiological data 330 that may indicate presence of environmental interference that impacts measurement accuracy.

[0161] One or more actions are then performed based on the noise condition (block 1010). These actions may include adjustment to parameters of the wearable device (e.g., properties of the illumination component 302, optical sensor 304 or noise detection component 306), modification of processing of the physiological data 330 (e.g., in real time or in post processing), or output of an indication of the noise condition 118. In some examples, the actions include refining the physiological data 330 based on the interference data 332, such as by removing noise from the first signal 326 based on the second signal 328.

[0162] In various implementations, based on the noise condition 118 the wearable device may implement a low power mode by deactivating an illumination component 302 of the wearable device. During the low power mode, the wearable device may generate physiological data 330, e.g., PPG data, while in a non-illuminated condition based on a relationship between the first signal 326 and the second signal 328. This approach allows for continued monitoring of physiological parameters while conserving power.FIG. 1 Patents 55 Docket No.: 10030PCT1

[0163] The procedure 1000 may be repeated continuously or at regular intervals to provide ongoing noise management for the wearable device. By implementing this procedure 1000, the wearable device can adapt to changing environmental conditions and maintain accuracy of physiological measurements.

[0164] It should be understood that many variations are possible based on the disclosure herein. Although features and elements are described above in particular combinations, each feature or element is usable alone without the other features and elements or in various combinations with or without other features and elements.

[0165] Clause 1. A wearable device attachable to a skin surface of a user comprising: an optical sensor configured to detect light reflected through the skin surface; a noise detection component configured to sample an environmental condition; and a processor configured to: receive a first signal from the optical sensor and a second signal from the noise detection component; generate physiological data based on the first signal; and detect, based on the second signal, a noise condition that indicates whether environmental interference is present that impacts an accuracy of the physiological data.

[0166] Clause 2. The wearable device of clause 1, further comprising an illumination component positioned on a same side of the wearable device as the optical sensor and configured to emit light to penetrate the skin surface of the user, the optical sensor configured to detect an amount of the emitted light that is reflected back through the skin surface.

[0167] Clause 3. The wearable device of clause 1 or clause 2, wherein to generate the physiological data includes an ambient light rejection process to denoise the first signal based on values of the first signal that correspond to an illuminated condition andFIG. 1 Patents 56 Docket No.: 10030PCT1values of the first signal that correspond to a non-illuminated condition, and wherein to detect the noise condition includes determining that the ambient light rejection process is insufficient to effectively denoise the first signal to remove environmental interference from the physiological data.

[0168] Clause 4. The wearable device of any preceding clause, wherein the noise condition is detected based on one or more of an amplitude, magnitude, frequency, or rate of change of the second signal.

[0169] Clause 5. The wearable device of any preceding clause, wherein the optical sensor includes a first photodiode disposed to face the skin surface and the noise detection component includes a second photodiode disposed to face away from the skin surface and configured to detect ambient light fluctuations.

[0170] Clause 6. The wearable device of any preceding clause, wherein the noise detection component includes one or more of a resistor, capacitor, diode, or trace configuration configured to detect electromagnetic interference.

[0171] Clause 7. The wearable device of any preceding clause, wherein the noise condition includes a quantification of the environmental interference.

[0172] Clause 8. The wearable device of any preceding clause, wherein the processor is further configured to refine the physiological data by removing noise from the first signal based on the second signal.

[0173] Clause 9. The wearable device of any preceding clause, wherein the processor is further operable to output an indication of the noise condition, adjust a parameter of the optical sensor or the noise detection component based on the noise condition, or modify processing of the physiological data based on the noise condition.FIG. 1 Patents 57 Docket No.: 10030PCT1

[0174] Clause 10. A method implemented by a wearable device attachable to a skin surface of a user, the method comprising: receiving a first signal from an optical sensor configured to detect light reflected through the skin surface and a second signal from a noise detection component configured to sample an environmental condition; generating physiological data based on the first signal; and presenting a noise condition generated based on the second signal that indicates a presence of environmental interference that reduces an accuracy of the physiological data.

[0175] Clause 11. The method of clause 10, wherein the generating the physiological data includes implementing a denoising process based on values of the first signal that correspond to an illuminated condition and values of the first signal that correspond to a non-illuminated condition, and the noise condition indicates that the denoising process is insufficient to remove environmental interference from the physiological data.

[0176] Clause 12. The method of clause 10 or clause 11, wherein the presenting the noise condition includes outputting an indication of the noise condition, the indication including one or more of a visual notification, audio alert, or haptic feedback.

[0177] Clause 13. The method of any one of clauses 10-12, further comprising adjusting a parameter of the optical sensor, the noise detection component, or an illumination component of the wearable device based on a quantification of the environmental interference included in the noise condition.

[0178] Clause 14. The method of any one of clauses 10-13, wherein the optical sensor includes an illumination component to emit light towards the skin surface and a first photodiode configurable for a photoplethysmography (PPG) task and the noiseFIG. 1 Patents 58 Docket No.: 10030PCT1detection component includes a second photodiode configured to detect ambient light interference.

[0179] Clause 15. The method of any one of clauses 10-14, wherein the noise detection component includes one or more of a resistor, capacitor, diode, or trace configuration configured to detect electromagnetic interference.FIG. 1 Patents 59 Docket No.: 10030PCT1

Claims

CLAIMSWhat is claimed is:

1. A wearable device attachable to a skin surface of a user comprising: an optical sensor configured to detect light reflected through the skin surface; a noise detection component configured to sample an environmental condition; and a processor configured to: receive a first signal from the optical sensor and a second signal from the noise detection component; generate physiological data based on the first signal; and detect, based on the second signal, a noise condition that indicates whether environmental interference is present that impacts an accuracy of the physiological data.

2. The wearable device of claim 1, further comprising an illumination component positioned on a same side of the wearable device as the optical sensor and configured to emit light to penetrate the skin surface of the user, the optical sensor configured to detect an amount of the emitted light that is reflected back through the skin surface.FIG. 1 Patents 60 Docket No.: 10030PCT13. The wearable device of claim 1 or claim 2, wherein to generate the physiological data includes an ambient light rejection process to denoise the first signal based on values of the first signal that correspond to an illuminated condition and values of the first signal that correspond to a non-illuminated condition, and wherein to detect the noise condition includes determining that the ambient light rejection process is insufficient to effectively denoise the first signal to remove environmental interference from the physiological data.

4. The wearable device of any preceding claim, wherein the noise condition is detected based on one or more of an amplitude, magnitude, frequency, or rate of change of the second signal.

5. The wearable device of any preceding claim, wherein the optical sensor includes a first photodiode disposed to face the skin surface and the noise detection component includes a second photodiode disposed to face away from the skin surface and configured to detect ambient light fluctuations.

6. The wearable device of any preceding claim, wherein the noise detection component includes one or more of a resistor, capacitor, diode, or trace configuration configured to detect electromagnetic interference.

7. The wearable device of any preceding claim, wherein the noise condition includes a quantification of the environmental interference.FIG. 1 Patents 61 Docket No.: 10030PCT18. The wearable device of any preceding claim, wherein the processor is further configured to refine the physiological data by removing noise from the first signal based on the second signal.

9. The wearable device of any preceding claim, wherein the processor is further operable to output an indication of the noise condition, adjust a parameter of the optical sensor or the noise detection component based on the noise condition, or modify processing of the physiological data based on the noise condition.

10. A method implemented by a wearable device attachable to a skin surface of a user, the method comprising: receiving a first signal from an optical sensor configured to detect light reflected through the skin surface and a second signal from a noise detection component configured to sample an environmental condition; generating physiological data based on the first signal; and presenting a noise condition generated based on the second signal that indicates a presence of environmental interference that reduces an accuracy of the physiological data.FIG. 1 Patents 62 Docket No.: 10030PCT111. The method of claim 10, wherein the generating the physiological data includes implementing a denoising process based on values of the first signal that correspond to an illuminated condition and values of the first signal that correspond to a non-illuminated condition, and the noise condition indicates that the denoising process is insufficient to remove environmental interference from the physiological data.

12. The method of claim 10 or claim 11, wherein the presenting the noise condition includes outputting an indication of the noise condition, the indication including one or more of a visual notification, audio alert, or haptic feedback.

13. The method of any one of claims 10-12, further comprising adjusting a parameter of the optical sensor, the noise detection component, or an illumination component of the wearable device based on a quantification of the environmental interference included in the noise condition.

14. The method of any one of claims 10-13, wherein the optical sensor includes an illumination component to emit light towards the skin surface and a first photodiode configurable for a photoplethysmography (PPG) task and the noise detection component includes a second photodiode configured to detect ambient light interference.FIG. 1 Patents 63 Docket No.: 10030PCT115. The method of any one of claims 10-14, wherein the noise detection component includes one or more of a resistor, capacitor, diode, or trace configuration configured to detect electromagnetic interference.FIG. 1 Patents 64 Docket No.: 10030PCT1

Images