Multisensory wearable technology to monitor sleep, circadian rhythms, and… | MedLink Neurology (original) (raw)

Introduction

Overview

There is an increasing awareness of the importance of optimized sleep. To achieve supreme sleep—in quality, quantity, and architecture—many folks are turning to devices to better realize their goal. It is understood that there are established benefits in terms of quality of life when focusing on enhanced functionality and performance. Certainly, daytime performance hinges on the integrity of nocturnal sleep. In this era of deep learning, algorithms, and “AI,” the sensors developed within the industry have made tremendous gains toward the advancement of sleep monitoring, especially in the setting of at-home sleep assessment.

This article discusses certain types of monitors currently available on the market. The article does not attempt to be complete as the product line is ever evolving. Many are considered “wellness products.”

Many devices strive toward FDA clearance. Newer sensors require validation testing, which is cumbersome but necessary. The traditional polysomnogram is the “gold standard” for accuracy of interpretation of achieved physiological data. Many outfits do not disclose their “black box” proprietary formulas. Transparency is poor, and validation is skewed.

Obstructive sleep apnea is considered a significant and independent risk factor regarding cardiovascular illnesses and metabolic diseases and neurodegenerative processes. The swell of interest in a “good night’s sleep” increases demand for novel, efficient, affordable, precise, and validated diagnostic sleep medicine services (02). It is underrecognized and undertreated in everyday clinical practice. The pandemic of individuals with unhealthy body mass is a major burden on society. General detection of obstructive sleep apnea in the setting of concurrent obesity is key. This cannot be sufficiently accomplished by way of in-lab polysomnography studies: the volume is simply too great.

The wearables are also valuable in the diagnosis of insomnia, circadian rhythm disorders, and, quite recently, neurohormonal changes. The data include time in bed as well as other parameters, including total sleep time, wake after sleep onset, sleep onset latency, and periodic limb movements.

Key points

| | • Software (SaMD or Software as Medical Device) is playing an increasingly important and critical role in healthcare with many clinical and administrative purposes. There is a veritable digital health revolution, with the ability to capture different biosignals with the generation of large datasets (Big Data) and the potential of offering unprecedented insight into the user’s health status. The importance of measuring sleep as an integrated component of wellness is understood. | | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | | • Masses of people suffer from significant sleep disorders. These can be detected with reasonable accuracy and efficiency via FDA-cleared wearable multisensory monitors that are now available on a large scale. | | | • There are distinct advantages to nonclinical monitoring outside of the in-lab sleep setting. Validation studies are indicated and are typically compared to polysomnography. | | | • Many sleep monitoring devices are reviewed in this article. No endorsements are made. They are marketed in many forms, including ring, patch, chin, headband, button, in-ear, and mask, among others. |

A conservative caveat stems from the 2024 JCSM article that compares the latest devices: “These novel obstructive sleep apnea wearables may emerge as primary diagnostic tools for patients at risk for moderate-to-severe obstructive sleep apnea without significant comorbidities” (02).

Discussion

There have been tremendous advances in the arena of portable electronics for sleep monitoring. This article reviews certain varied brands of “gadgets” after providing an overview of the expanding field of home sleep monitoring.

Approximately 12% of the U.S. population suffers from obstructive sleep apnea, which is often undiagnosed and undermanaged. The symptoms are not particularly noticed during sleep. The apnea/hypopnea with deoxygenation creates cortical arousal and sleep disruption. Obstructive sleep apnea is a major economic burden due to loss in productivity and the increased cost of likely comorbidities (such as hypertension, hyperlipemia, stroke, dementia, mood disorders, and type 2 diabetes mellitus).

Currently, there is a movement toward home sleep testing. Monitoring could focus on the elderly, snorers, obese candidates, brain workers, and those invested in aviation or outdoor sports.

The advantages of home sleep testing (nonclinical setting) over in lab polysomnography are defined in key points in a study by Go and Thaler and include the following (04):

| | • Compact, less equipment, noninvasive | | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | | • Ambulatory, ease of use, no “first night effect,” per se | | | • More representative of personal sleep given the home bedroom environment | | | • Less expensive: cost-effective | | | • Less personnel consuming, conducive to public health considerations | | | • More scale and efficiency in order to meet increasing demand | | | • Comfortable, accurate* (+/-), private, convenient | | | • Outperforms standard actigraphy | | | • Bluetooth compatible, less workload, streamlined management | | | • Touted as less accurate than on-site lab testing, leading to possible misdiagnosis (limited access to raw data; difficulty with dark skin/tattoos; limited standardization and validation; diminished accuracy if sleep is fragmented or if bed is shared with pet or partner) | | | • In contradistinction, Teixeira purports that in-lab polysomnography worsens obstructive sleep apnea (via artificial increase of severity) by changing body position compared to home sleep tests, where bodies can move more naturally, unfettered by head electrodes and the like (09). |

The American Academy of Sleep Medicine (AASM) offers a position statement regarding home sleep testing (05):

| | Consumer sleep technologies (CSTs) cannot be utilized for the diagnosis and/or treatment of sleep disorders at this time. … It is the position of the AASM that CST must be FDA cleared and rigorously tested against current gold standards if it is intended to render a diagnosis and/or treatment. Given the unknown potential of CST to measure sleep or assess for sleep disorders, these tools are not substitutes for medical evaluation. However, CSTs may be utilized to enhance the patient-clinician interaction when presented in the context of an appropriate clinical evaluation. | | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | | A recent disclosure dampens the AUTO-interpretation of home sleep tests a bit. Wireko and Morgenthaler pointed to a clinical case whereby high-level sound perhaps indicating “true” loud snoring via home sleep test was, instead, emanating from a mechanical heart valve, ie, artifact (10). |

Meltzer and colleagues state: “Despite its low cost and ease of use for consumers, neither sleep-recording mode of the Fitbit Ultra accelerometer provided clinically comparable results to PSG. Further, pediatric sleep researchers and clinicians should be cautious about substituting these devices for validated actigraphs, with a significant risk of either overestimating or underestimating outcome data including total sleep time and sleep efficiency” (06).

There is limited validation of the currently available technology in wearable devices, which is a primary factor curbing their large-scale use. Commonly, home sleep tests would overestimate total sleep time along with sleep efficiency. For example, “quiet wakefulness” during bouts of insomnia were problematic. Inter-device reliability and non-standardization are at issue as well.

The inter-scorer reliability – with senior scorers of polysomnography – is not well established. Younes and colleagues concluded that manual scoring of nonrapid eye movement sleep stages is highly unreliable among highly trained, experienced technologists (11). This inherently confounds the attempt to use polysomnography interpretation as the gold standard for validation of burgeoning technology when it is, itself, quite imperfect. Parenthetically, the AASM offers a sleep facility interscorer reliability requirement in which there must be 12 polysomnography tests per year, compared by epoch.

The most commonly used at-home test (equipment type) for sleep apnea is type 3. By contrast, the formal in-lab study is deemed type 1 – the “gold standard.” The type 3 test tracks the following, by definition: breathing effort and rate, airflow, snoring, and blood oxygen saturation.

Body position, heart rate and variability, nasal pressure, and apnea hypopnea index are also understood, along with sleep onset, total sleep time, sleep efficiency, and wake after sleep onset.

Type 1 testing equipment includes that of type 3 as well as heart rate/heart rate variability, rhythm, and sleep stages, along with carbon dioxide output.

Wearable sensors with portable electronics include sleep stages as well. This is a key component in the field of sleep medicine. The sleep relevant metrics related to EEG include sleep stages: W, NREM, REM or wake, non-REM (stages I, II, and III) and REM. Sleep cycling and continuity, beyond sleep architecture, are vitally important parameters.

A valuable screening tool toward the detection of obstructive sleep apnea is the STOP-BANG questionnaire.

S Snoring
T Tired
O Observed apnea
P high blood Pressure
B Body mass index > 35
A Age 50 and older
N Neck circumference wider than 16 inches
G Gender (men are at higher risk of obstructive sleep apnea)

The latest wearable devices rely not only on 3D accelerometers but also on reflective photoplethysmography. Optical photoplethysmography sensors allow an optimal measure of the changes in blood pressure and blood volume at the periphery. Of note, when undertaking a home sleep test, remember no hand cream, fingernail polish, jewelry, nor tattoo marks. Wear snugly for improved signal-to-noise ratio.

Finally, wearable sleep monitoring devices to detect obstructive sleep apnea in the setting of “globesity” is aimed at remediation. Pharmacotherapy to curb cravings may prevent obesity and, thus, obstructive sleep apnea and downstream pathologies (eg, stroke, heart attack, migraine, dementia, T2DM, lipemia, hypertension, mood disorders, accidents). Short of discipline and nutrition/dietetics and exercise, GLP-1 receptor agonists and SGLT-2 inhibitors may play a major role (07).

As the field gains in sophistication, the Department of Defense will express interest as most soldiers are sleep restricted. In vulnerable adults, this sleep loss impacts performance on the next day. One measure of disrupted sleep continuity is the accumulation of cortical microarousals. Microarousals are brief (less than 15 seconds) shifts in EEG power (usually theta ranges) and correlate with altered daytime functioning and negative health outcomes across a sleep period (01).

Regarding various modalities, as marketed, this article will focus on an array of products, including headband, ring, armband, patch, topic phonic device, wristband, and mask.

The Dreem headband monitors both sleep-related physiological signals and accurately processes them into sleep stages. The study sample was small and homogeneous, with the need for a more diverse population.

The FitBit and Jawbone UP3 are defunct without a companion app. Consumers were disgruntled over the extinction.

A non-exhaustive list of home sleep testing services includes the following: MedBridge Healthcare, ApneaMed, betternight, SleepViewDIRECT, Gem Sleep, General Sleep, Ognomy, Onera, dreamclear, SleepCareonline, the SLEEP DOCTOR, snap diagnostics, Virtuox, ZOLL Itamar, Z Machine, NightOwl, WatchPat-One, and iButton, among others. Each holds their own platform feature.

The OURA ring is under assessment for the detection of the onset of menstruation or pregnancy, and/or prediction of pregnancy outcomes, at a university-based level of study. The physiological changes during hormonal shifts with sustained increase in nighttime body temperature (via rapid sampling rates of > 1/minute with detection of small changes) reveals the working impression. This product is marketed, including membership, as a smart ring to encourage individuals to be proactive in their health and “anti-aging.” Beyond being informative in reference to sleep, it offers insight into stress, activity (steps; calories expended), work productivity/training trends, and performance recovery by accurately measuring some 20 biometrics.

EEG-based wearables were systematically and comprehensively reviewed, and the conclusions, or practice points, were as follows (03):

• EEG-based wearables demonstrate impressive accuracy in sleep staging, competing with polysomnography, while offering more comfort and less intrusion.

• Potential users are encouraged to select a user-friendly system based on personal needs and cost.

The university-based AISAP study, artificial intelligence home sleep apnea testing, tested the performance of the “WVU-device” (08). The technology is patented. They shared evidence of demonstration of good accuracy in predicting respiratory event index when compared to an approved home sleep apnea test device, even in patients with darker skin tones.