Why Use Ultrasensitive Sleep Sensors in Healthcare

🛏️ Revolutionizing Sleep Monitoring: From Wristbands to Real Biomarkers Traditional wearables fall short when it comes to sleep diagnostics—they miss direct respiratory signals, which are vital for identifying sleep stages and disorders like sleep apnea. This new research introduces a low-power, skin-integrated mechanoacoustic (LMA) sensor that changes the game. It doesn't just guess your sleep from motion—it listens to your breath and heartbeat. https://lnkd.in/g_yZW69p 🔬 Key innovations: - Multimodal sensor captures respiratory rate, heart rate variability (HRV), respiration rate variability (RRV), body movement, and more. -Paired with LMA-SleepNet, an interpretable machine learning model that detects sleep stages and apnea events with clinical-grade accuracy. - Uses physiology-based features like baroreflex and muscle tone—giving deeper insights than motion-based trackers. Outperforms other wearables in real-world accuracy. 📊 Why this matters: - Directly measures respiration—a core but missing biomarker in most wearables. - Enables continuous, personalized, and explainable sleep tracking in the home or clinic. - Opens doors for smart OSA detection, snoring tracking, and even future on-body therapeutic interventions. 💡 Bonus: Real-time UTC synchronization allows scalable multi-sensor studies across environments and populations. 🔁 This is more than a device—it’s a complete hardware-software platform for next-gen sleep and health monitoring, with huge potential for precision healthcare, chronic disease management, and behavioral science. 👀 Sleep isn’t just rest—it’s integrated physiological data. And now, we can measure it better than ever. #SleepScience #WearableTech #DigitalHealth #MachineLearning #SleepApnea #RespiratoryHealth #PrecisionMedicine #Bioengineering #HealthTech #HRV #RRV

One of the biggest opportunities in health is moving from episodic measurement to more continuous, real-world understanding. That’s why this new PLOS Digital Health paper is so meaningful. Led by Michael Chee and his team at the National University of Singapore, the study found that nocturnal PPG signals collected from Oura Ring could estimate vascular age with performance comparable to a clinical-grade fingertip sensor in a cohort of 160 healthy adults. Vascular aging is a key marker for cardiovascular risk, but the traditional ways to measure it are often expensive and hard to scale. Research like this helps show how consumer wearables may expand access to longitudinal, real-world health insights in a way that is more practical and more accessible over time. What also stands out to me is that these signals were captured during sleep, when physiological data can be gathered passively and consistently. At Oura, we’ve long known that sleep is one of the clearest windows into overall health, and research like this proves it. It also points to a more proactive model of health—one built not just on isolated clinical snapshots, but on patterns measured over time. Learn more on the Pulse blog: https://lnkd.in/gmi8U5AK  

The first sleep “clinic” like wearable, FDA cleared! Another step in wearables becoming more medical! Sunrise officially launched the Sunrise Air. It is a rechargeable, ultra-lightweight sensor you stick to your chin at night. No wires. No mask. No sleep lab. No disposables. You can wear it for multiple nights in a row and get a clinical-grade diagnosis of sleep apnea. We are about to stop treating sleep like recovery hygiene. And start treating it like the most predictive biomarker for how long, and how well, you live. Here is why this matters: 83.7 million American adults have obstructive sleep apnea. That is 32% of every adult over 20. More than 80% of them have no idea. Untreated, OSA raises the risk of stroke, heart failure, type 2 diabetes, and early death. 72% of type 2 diabetics and 77% of the morbidly obese also have sleep apnea. Until now, the bottleneck was the test. Polysomnography means a night in a lab, wired up, watched by a technician. It costs thousands and you wait months. Existing home tests used disposable kits and gave you one night of data. Sunrise Air breaks that bottleneck. → One 8-gram sensor on the chin → Multi-night testing, fully rechargeable → Measures mandibular jaw movement, airflow, SpO₂, snoring → Differentiates obstructive from central sleep apnea, a clinical first for at-home → AI-supported, clinician-reviewable reports Laurent Martinot, Founder and CEO: "Nobody sleeps the same way twice, yet we have been diagnosing based on a single night. That changes now." Sunrise has 8,000+ patients across 10 peer-reviewed publications, including JAMA Network Open. FDA De Novo in January 2022, recommended in UK NICE guidelines. In September 2025 they raised $29M led by Eurazeo, plus Amazon's Alexa Fund. Total raised: ~$58M. Here is why this matters beyond sleep apnea: Sleep is moving from wellness anecdote to longevity infrastructure. → UK Biobank, 328,850 people: healthy sleep cut the risk of premature end of healthspan by 15%. → 172,321 adult cohort: men with adequate sleep live ~5 years longer, women ~2 years. → OHSU (SLEEP Advances, Dec 2025): insufficient sleep beat diet, exercise, and loneliness as a predictor of life expectancy. Only smoking outranked it. → Resmed projects 77M US adults will have OSA by 2050, 46% of 30 to 69 year olds. We spent the 2010s on glucose. Sleep is next. My bet: by 2028, a multi-night home sleep test will be as normal as a CGM patch. Sunrise Air just made the regulatory case in the US. The real question: which longevity brand bundles sleep diagnostics into its stack first. Oura, Whoop, Function, Levels, Eight Sleep, AG1. One of them is already drafting the partnership memo. Let's build health that works for real life. Sources: BioSpace · Medical Economics · MPO Magazine · Mobihealthnews · JAMA Network Open · Resmed Newsroom · ScienceDirect · OHSU News · Hello Sunrise Longevity Technology

A technical breakthrough from Australia is able to track DynamicMicroData (DMD), the basis for Gen-3 Wearables, a major shift in trackers which we predicted recently. A team from the University of New South Wales has created tiny ultra-thin cantilevered sensors that can detect multiple physiological mechano-acoustic signals over an outstanding bandwidth of 15.5 octaves, yes octaves! These sensors are integrated into small adhesive wearables. With a power demand of under 5mW they are able to continuously capture subtle vibrations produced by the heart, lungs, blood flow, an even vocal chords. An AI layer allows these signals to be segregated and analyzed for clinical decision-making. The high sensor bandwidth enables the device to detect signals that are way beyond the capability of today’s trackers. The ability to acquire DMD for example, allows the wearable to ‘listen’ to heart-valves opening and closing, or track the transitions between sleep stages which are rich in information related to central nervous system functionality. As just one example, the attached chart shows details of breathing transitions which are important in diagnosing the occurrence and causes of sleep disturbed breathing, a field of great interest to our team and one which is ripe for new innovations in both diagnoses and therapies. This Australian development is a clear marker for the future of healthcare and a sign that major changes are likely to come faster than originally thought. We can anticipate a time when our key health markers are tracked continuously enabling a shift to early preventative care from late symptom treatment. For those wanting to learn more, access the full Nature report. You will find the future shining bright!

𝗕𝗥𝗘𝗔𝗞𝗜𝗡𝗚: Just one night of sleep data linked to long-term health. A new study in 𝘕𝘢𝘵𝘶𝘳𝘦 𝘔𝘦𝘥𝘪𝘤𝘪𝘯𝘦 used AI to explore just how informative sleep can be. They trained an AI model on a 𝗛𝗨𝗚𝗘 𝗱𝗮𝘁𝗮𝘀𝗲𝘁: 👉 585,000+ hours of gold-standard sleep lab data from 65,000+ people. From a 𝘴𝘪𝘯𝘨𝘭𝘦 𝘯𝘪𝘨𝘩𝘵 𝘰𝘧 𝘴𝘭𝘦𝘦𝘱, the model predicted the future registration of 𝟭𝟯𝟬 𝗵𝗲𝗮𝗹𝘁𝗵 𝗰𝗼𝗻𝗱𝗶𝘁𝗶𝗼𝗻𝘀! These spanned a wide range: • dementia, • heart disease, • stroke, • kidney disease, • and even all-cause mortality. Impressive. But the real story is 𝘸𝘩𝘢𝘵 𝘬𝘪𝘯𝘥 of sleep information this model is actually using. Critically, this isn’t about “how long you slept” or how much REM or deep sleep you got. The model learns directly from raw physiological signals: • EEG from the brain, • heart rhythms, • breathing patterns, • muscle activity and, crucially, • how these systems interact across the night. Think:  – arousal burden  – autonomic regulation  – fragmentation  – cross-talk between brain, heart, and breathing Not sleep stages alone. There’s also important nuance in how to interpret this. What the model is likely capturing is a 𝗺𝗶𝘅 𝗼𝗳 𝘁𝗵𝗿𝗲𝗲 𝘁𝗵𝗶𝗻𝗴𝘀: 1. 𝗧𝗿𝘂𝗲 𝗹𝗼𝗻𝗴𝗲𝗿-𝘁𝗲𝗿𝗺 𝘃𝘂𝗹𝗻𝗲𝗿𝗮𝗯𝗶𝗹𝗶𝘁𝘆 2. 𝗩𝗲𝗿𝘆 𝗲𝗮𝗿𝗹𝘆 𝘀𝗶𝗴𝗻𝘀 𝗼𝗳 𝗱𝗶𝘀𝗲𝗮𝘀𝗲 𝗮𝗹𝗿𝗲𝗮𝗱𝘆 𝗱𝗲𝘃𝗲𝗹𝗼𝗽𝗶𝗻𝗴 (𝗯𝗲𝗳𝗼𝗿𝗲 𝗱𝗶𝗮𝗴𝗻𝗼𝘀𝗶𝘀) 3. 𝗛𝗼𝘄 𝘁𝗵𝗲 𝗵𝗲𝗮𝗹𝘁𝗵𝗰𝗮𝗿𝗲 𝘀𝘆𝘀𝘁𝗲𝗺 𝗱𝗲𝘁𝗲𝗰𝘁𝘀/𝗿𝗲𝗰𝗼𝗿𝗱𝘀 𝗱𝗶𝘀𝗲𝗮𝘀𝗲 (𝘮𝘢𝘯𝘺 𝘤𝘰𝘯𝘥𝘪𝘵𝘪𝘰𝘯𝘴 𝘦𝘹𝘪𝘴𝘵 𝘭𝘰𝘯𝘨 𝘣𝘦𝘧𝘰𝘳𝘦 𝘵𝘩𝘦𝘺’𝘳𝘦 𝘧𝘰𝘳𝘮𝘢𝘭𝘭𝘺 𝘥𝘪𝘢𝘨𝘯𝘰𝘴𝘦𝘥) And one more thing matters. These recordings cam from one night in a sleep lab—an unfamiliar bed, sensors attached, a mildly stressful environment. So one possibility is that the model isn’t just reading “sleep quality,” but how effectively the brain and body can downshift and recover when conditions aren’t ideal. Almost like an 𝗼𝘃𝗲𝗿𝗻𝗶𝗴𝗵𝘁 𝘀𝘁𝗿𝗲𝘀𝘀 𝘁𝗲𝘀𝘁:  👉 How quickly does arousal settle?  👉 How well does the brain disengage?  👉 How coordinated are brain, heart, and breathing under challenge? That ability to relax and recover, even in a mildly stressful situation, is increasingly linked to long-term health. So no, one night of sleep doesn’t predict your destiny. But this study reinforces something important: 𝗦𝗹𝗲𝗲𝗽 𝗶𝘀𝗻’𝘁 𝗷𝘂𝘀𝘁 𝗿𝗲𝘀𝘁. It’s a window into regulation, resilience, and recovery across the whole body. Now it's just waiting until we have a similar dataset with consumer sleep tracker data....! Speaking of which: my next newsletter will focus on sleep tracker data and whether it's a useful tool to have.

Some of the most important signals in health may come from a place we’ve traditionally overlooked: during sleep. A new study, published by Michael Chee and his team in PLOS Digital Health, looked at whether nighttime photoplethysmography from Oura Ring could be used to estimate vascular aging, a meaningful marker of cardiovascular risk. The researchers found that Oura Ring performed comparably to a clinical-grade fingertip sensor in estimating vascular age from pulse waveforms collected overnight. One concrete detail that stood out to me: the deep learning model achieved a mean absolute error of about 7.25 years using ring data, versus 6.28 years with the clinical-grade device, with no statistically significant difference between them. This is a significant proof point that scalable, longitudinal, low-friction sensing may help us understand cardiovascular health earlier, more continuously, and in the context of daily life. If we want to move healthcare upstream, we need tools that meet people where they are — and sometimes that means learning from the physiology we can observe at night. To me, that’s where digital health becomes genuinely useful: not louder, but earlier, steadier, and more actionable. Read more on the Pulse blog: https://lnkd.in/eRyNd_qT Full paper: https://lnkd.in/eT_DgE5P

Can you imagine getting woken up every hour? All night long. That is what happens to patients in the hospital. Nighttime wakefulness tripled. From 10.7% at home to 34.8% in a hospital bed. The researchers placed multiple remote sensors on cardiac surgery patients to study sleep-wake rhythms and cognitive decline in the hospital environment. What they found is no surprise.  But knowing it and measuring it are different things. -At least one visit per hour through the night.  -Sound levels averaged 51.9 dBA   -Ambient temperature stayed flat. No diurnal drop to signal the body it was time to sleep. -And 31% of patients developed transient cognitive impairment. Every nurse knows it. Every patient lives it. And as a physician, none of this is surprising. The irony: the person walking in to make sure you are okay is the very thing causing harm. Disrupting sleep, fragmenting cognition, slowing recovery. And then there are the alarms. The average ICU patient is exposed to hundreds of alarms per day. Most are false. Staff learn to tune them out and it is one of the most documented safety risks in hospital medicine. The very system designed to alert caregivers becomes background noise. Another layer of disruption no one sleeps through. But I love this study because the sensors made the invisible visible. I am not advocating leaving a sick patient unattended. We have the technology to monitor patients untethered and remote. We have used it for years at home and we did it during Covid. But decades of learned behavior keep hospital protocols centered on interruption. It is time to design controlled trials and protocols treating the hospital environment itself as a modifiable risk factor. We know what is breaking our patients' sleep. We have the tools to fix it. It is Time to Let Them Sleep. References Carsten Skarke Nadim El Jamal Skarke, C., El Jamal, N., Genuardi, M.V. et al. Quantifying sleep wake rhythms in the hospital environment with digital technologies. npj Digit. Med. (2026)

𝐌𝐨𝐬𝐭 𝐝𝐢𝐬𝐜𝐮𝐬𝐬𝐢𝐨𝐧𝐬 𝐚𝐛𝐨𝐮𝐭 𝐀𝐈 𝐢𝐧 𝐡𝐞𝐚𝐥𝐭𝐡𝐜𝐚𝐫𝐞 𝐟𝐨𝐜𝐮𝐬 𝐨𝐧 𝐝𝐢𝐚𝐠𝐧𝐨𝐬𝐭𝐢𝐜𝐬 𝐨𝐫 𝐚𝐝𝐯𝐚𝐧𝐜𝐞𝐝 𝐦𝐨𝐝𝐞𝐥𝐬. 𝐁𝐮𝐭 𝐚 𝐥𝐨𝐭 𝐨𝐟 𝐯𝐚𝐥𝐮𝐞 𝐜𝐨𝐦𝐞𝐬 𝐟𝐫𝐨𝐦 𝐦𝐮𝐜𝐡 𝐬𝐢𝐦𝐩𝐥𝐞𝐫 𝐬𝐲𝐬𝐭𝐞𝐦𝐬 𝐚𝐩𝐩𝐥𝐢𝐞𝐝 𝐢𝐧 𝐭𝐡𝐞 𝐫𝐢𝐠𝐡𝐭 𝐩𝐥𝐚𝐜𝐞. A recent pilot in Dorset care homes used AI-based sound and motion sensors to monitor residents at night. The system detects patterns that may indicate a fall, distress, or unusual activity and alerts staff in real time. The outcomes were quite concrete: fewer falls, a strong reduction in ambulance callouts, and fewer hospital transfers. What makes this case interesting is the shift from periodic checks to continuous awareness. Instead of relying on scheduled rounds, staff can respond based on actual events. That changes both the quality of care and how resources are used. It also highlights a broader point: in many settings, the impact of AI is less about complex prediction and more about timely detection and response within existing workflows. If you’re looking at similar questions – where better visibility could improve decisions or response time – it’s worth exploring what can be done with the data you already have. 𝐏.𝐒. 𝐈’𝐦 𝐀𝐥𝐞𝐱𝐞𝐢 𝐓𝐮𝐫𝐜𝐡𝐚𝐤, 𝐅𝐨𝐮𝐧𝐝𝐞𝐫 𝐚𝐧𝐝 𝐂𝐓𝐎 𝐚𝐭 𝐀𝐫𝐢𝐬𝐭𝐞𝐤. 𝐈 𝐬𝐡𝐚𝐫𝐞 𝐢𝐧𝐬𝐢𝐠𝐡𝐭𝐬 𝐨𝐧 𝐀𝐈 𝐚𝐧𝐝 𝐭𝐞𝐜𝐡 𝐢𝐧𝐧𝐨𝐯𝐚𝐭𝐢𝐨𝐧. 𝐅𝐨𝐥𝐥𝐨𝐰 𝐦𝐞 𝐟𝐨𝐫 𝐮𝐩𝐝𝐚𝐭𝐞𝐬 𝐨𝐫 𝐜𝐨𝐧𝐭𝐚𝐜𝐭 𝐦𝐞 𝐢𝐟 𝐲𝐨𝐮’𝐫𝐞 𝐞𝐱𝐩𝐥𝐨𝐫𝐢𝐧𝐠 𝐚𝐩𝐩𝐥𝐢𝐞𝐝 𝐀𝐈 𝐢𝐧 𝐲𝐨𝐮𝐫 𝐛𝐮𝐬𝐢𝐧𝐞𝐬𝐬.

Your patients are walking into appointments with six months of sleep data on their wrist. Most dental offices don't know what to do with it. HRV trending down. Deep sleep under 15%. Resting heart rate elevated overnight. Respiratory rate creeping up. These aren't vanity metrics. They're the body's attempt to communicate something it can't put into words and for the first time in history, patients have access to this data before any clinician orders a single test. The question is: who in their care team is equipped to have that conversation? Five years ago, sleep wearables were niche. Today they're mainstream. And the patients wearing them are noticing patterns poor sleep, low energy, elevated overnight heart rate  without anyone explaining what those patterns mean or what to do about them. That's the gap airway-focused providers are uniquely positioned to fill. Because Low HRV + poor deep sleep + elevated resting heart rate overnight isn't just "bad sleep data." It's a clinical signal pointing toward airway dysfunction. And airway dysfunction has a structural cause one that a proper evaluation can identify and a treatment plan can address. If you're a dental provider and you're not yet asking patients what their wearable is showing start asking. If you're a patient who's been staring at concerning data without answers bring it to someone who knows what it means. And if you're a provider who wants the framework to have this conversation confidently in your own practice: Foundations of Airway Dentistry → https://lnkd.in/dJGyFSEG

𝗬𝗼𝘂𝗿 𝗕𝗼𝗱𝘆 𝗦𝗵𝗼𝘄𝘀 𝗜𝘁𝘀 𝗙𝘂𝘁𝘂𝗿𝗲 𝗪𝗵𝗶𝗹𝗲 𝗬𝗼𝘂’𝗿𝗲 𝗔𝘀𝗹𝗲𝗲𝗽   𝘠𝘰𝘶𝘳 𝘣𝘰𝘥𝘺 𝘬𝘯𝘰𝘸𝘴 𝘸𝘩𝘢𝘵’𝘴 𝘤𝘰𝘮𝘪𝘯𝘨 𝘭𝘰𝘯𝘨 𝘣𝘦𝘧𝘰𝘳𝘦 𝘺𝘰𝘶 𝘥𝘰. 𝘔𝘦𝘥𝘪𝘤𝘪𝘯𝘦 𝘶𝘴𝘶𝘢𝘭𝘭𝘺 𝘧𝘪𝘯𝘥𝘴 𝘰𝘶𝘵 𝘵𝘰𝘰 𝘭𝘢𝘵𝘦.   In early 2026, Stanford Medicine researchers unveiled SleepFM—an AI model trained on nearly 600,000 hours of polysomnography data from 65,000 people. By reading a single night of sleep—brain waves, heart rhythm, breathing, and micro-movements—the model detects biological patterns associated with elevated long-term risk across 100+ diseases, including heart disease, dementia, cancer, and stroke.   This exposes a structural flaw in modern healthcare:  • We wait for symptoms instead of signals.   𝗦𝗹𝗲𝗲𝗽𝗙𝗠 𝗽𝗼𝗶𝗻𝘁𝘀 𝘁𝗼 𝗮 𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝘁 𝗿𝗲𝗮𝗹𝗶𝘁𝘆:  • Disease progression leaves measurable fingerprints years before diagnosis  • Sleep captures system-level physiology no clinic visit ever sees  • One night of high-resolution data can be more informative than years of sporadic checkups  • AI’s real power is not automation, but early truth extraction  • This is not about better sleep scores or optimization culture.  • It’s about shifting healthcare from reaction to anticipation.   The future of medicine won’t start in hospitals. It will start in places where the body stops performing and starts revealing. Source & details: Stanford Medicine SleepFM study (Nature Medicine, 2026): https://lnkd.in/gxBYePw5 https://lnkd.in/gnYB53sJ https://lnkd.in/gurVfR3c #ArtificialIntelligence #MultimodalAI #FoundationModels #BrainyNeurals #AIinHealthcare #ClinicalAI #ComputerVision #EdgeAI #MultimodalAI #SleepScience #PreventiveMedicine #EarlyDetection #Neuroscience #FutureOfMedicine