AIoT Platform Vetted and Deployed Worldwide Across Medical, Wellness, and Consumer Markets

AIoT end-to-end integration and DevOps, including digital biomarker support, generative AI, machine learning, and firmware, deployed globally.

Fullpower AI Cloud Based Features

AIoT end-to-end integration and DevOps, including digital biomarker support, generative AI, machine learning, and firmware, deployed globally.

AIoT PaaS Bio-Sensing

AIoT platform vetted, compliant, and deployed successfully in 60+ countries.

AIoT Vetted and Deployed Worldwide

Countries Deployed

1. United States, 2. People's Republic of China, 3. Korea, 4. Japan, 5. Mexico, 6. Taiwan, 7. Germany, 8. France, 9. Hong Kong, 10. Chile, 11. Canada, 12. Argentina, 13. Columbia, 14. Malaysia, 15. Austria, 16. Singapore, 17. Poland, 18. Australia, 19. England, 20. Czech Republic, 21. Spain, 22. Peru, 23. Switzerland, 24. Vietnam, 25. Israel, 26. Thailand, 27. Sweden, 28. Philippines, 29. South Africa, 30. Guam, 31. Hungary, 32. Italy, 33. Norway, 34. Belgium, 35. Indonesia, 36. New Zealand, 37. Macau, 38. Jamaica, 39. Netherlands, 40. Finland, 41. Romania, 42. Slovakia, 43. Portugal, 44. Denmark, 45. Ireland, 46. Luxembourg, 47. Dubai, 48. Brazil, 49. Dominican Republic, 50. Tahiti, 51. Iceland, 52. Greece, 53. Mauritius, 54. Turkey, 55. Yugoslavia, 56. Bulgaria, 57. Costa Rica, 58. Virgin Islands, 59. India, 60. Panama, 61. Slovenia, 62. Paraguay, 63. El Salvador, 64. Aruba,

United States
People's Republic of China
Korea
Japan
Mexico
Taiwan
Germany
France
Hong Kong
Chile
Canada
Argentina
Columbia
Malaysia
Austria
Singapore
Poland
Australia
England
Czech Republic
Spain
Peru
Switzerland
Vietnam
Israel
Thailand
Sweden
Philippines
South Africa
Guam
Hungary
Italy
Norway
Belgium
Indonesia
New Zealand
Macau
Jamaica
Netherlands
Finland
Romania
Slovakia
Portugal
Denmark
Ireland
Luxembourg
Dubai
Brazil
Dominican Republic
Tahiti
Iceland
Greece
Mauritius
Turkey
Yugoslavia
Bulgaria
Costa Rica
Virgin Islands
India
Panama
Slovenia
Paraguay
El Salvador
Aruba

AIoT leader, worldwide standard compliant and secure. Includes ISO 27001, GDPR, SOC2, AWS certification.

ISO27001
AICPA SOC2
GDPR
AWS Qualified Software
AWS Standalone Partner

Latest News

Fullpower, Invention in Motion

Fullpower-AI® is the provider of a deep learning generative AI biosensing platform. The AIoT platform is a remote sensing cloud-to-edge solution vetted and successfully deployed in 60+ countries. Fullpower-AI® customers are in life sciences, health, wellness, biotechnology, and industrial automation. A portfolio of 135+ patents backs the platform. Fullpower-AI® is ISO 27001 certified.

Customer successes, including AIoT edge/cloud technology platform deployments.

Fullpower

Fullpower-AI® is the provider of a deep learning generative AI biosensing platform. The AIoT platform is a remote sensing cloud-to-edge solution vetted and successfully deployed in 60+ countries. Fullpower-AI® customers are in life sciences, health, wellness, biotechnology, and industrial automation. A portfolio of 135+ patents backs the platform. Fullpower-AI® is ISO 27001 certified.

Successful B2B examples of Fullpower-AI®
edge/cloud technology platform deployments

Tempur-Pedic

How we use AI in the Tempur Sleeptracker-AI® platform: We use AI to analyze sleep data collected by the Sleeptracker-AI® system to help identify different sleep stages, detect sleep disorders, and provide users with personalized sleep recommendations. With AI data science methodologies, we can match closely gold standard PSG sleep analysis and continue to improve accuracy and applicability in real-world settings. We, of course, in that process carefully consider bias and other ethical issues. Our methodologies include deep learning, supervised learning, and reinforcement learning techniques. To continually generate training data, we operate two clinical PSG sleep labs.

Nu-Skin

Powered by Fullpower-AI® AIoT PaaS, the AgeLOC LumiSPA iO is the winner of the 2023 Beauty Device Awards with hundreds of thousands of units distributed in over 60 countries, including China, Japan, Korea, the EU, and the Americas.

Samsung

Samsung is an investor in Fullpower-AI®. In addition, Samsung and Fullpower collaborate for AIoT solutions.

Nike

Nike and Fullpower-AI® have enjoyed a technology trusted partnership for a decade. Fullpower-AI® builds algorithms and sensing intelligence optimized with machine and deep learning for Nike. "We took great care in evaluating sensing technologies and found the Fullpower-AI® technology platform to be superior." - President of Digital Sport at Nike.

Bryte

Bryte and Fullpower-AI® collaborate on the edge-cloud integration of the Bryte mattress with the Sleeptracker-AI® platform. The Bryte mattress intelligently relieves pressure imbalances for any person in any position, delivering the best sleep experience.

Strategic Partnerships Powering the Platform

Fullpower Partner
Fullpower Partner
Fullpower Partner
Fullpower Partner
Fullpower Partner
Fullpower Partner
Fullpower Partner
Fullpower Partner
Fullpower Partner
Fullpower Partner
Fullpower Partner
Fullpower Partner
ISO27001
AICPA SOC2
GDPR
AWS Qualified Software
AWS Standalone Partner

Company and Leadership

Fullpower-AI® is the provider of a deep learning generative AI biosensing platform. The AIoT platform is a remote sensing cloud-to-edge solution vetted and successfully deployed in 60+ countries. Fullpower-AI® customers are in life sciences, health, wellness, biotechnology, and industrial automation. A portfolio of 135+ patents backs the platform. Fullpower-AI® is ISO 27001 certified.

Fullpower-AI® Leadership

Arthur Kinsolving, Co-Founder and CTO

Arthur Kinsolving, Co-Founder and CTO

Arthur Kinsolving leads the technology and intellectual property development efforts at Fullpower-AI®. Arthur oversees all technology development, patent filings, and intellectual property protection efforts. During his tenure at Fullpower-AI®, Arthur has co-authored over 75 issued patents and filed dozens more. Arthur is passionate about AI, machine and deep learning. Arthur received his Bachelor of Science in Mechanical Engineering from Yale University.
Nathan Tiller, CFO and COO

Nathan Tiller, CFO and COO

Nathan is responsible for Fullpower-AI®'s finance and operations. Nathan has been with Fullpower-AI® since 2015 and has over 20 years of experience in finance and operations in various industries, including software and banking. Nathan was Director of Finance at Serena Software, an international enterprise software company in Silicon Valley. Prior to that, Nathan managed the corporate development function at World Savings. Nathan also served as Director of Finance for New Teacher Center, a national non-profit organization. Nathan studied economics, earning his Bachelor of Arts degree at Rice University, and his Master of Public Affairs degree at University of Minnesota.
Philippe Kahn, Founder and CEO

Philippe Kahn, Founder and CEO

Philippe is the founder of four successful technology companies: Fullpower-AI® (contactless biosensing medical and wellness solutions); LightSurf Technologies (creator of the Camera-Phone); Starfish Software (co-creator of wireless OTA Synchronization); and Borland Software (Leader for Professional Development Tools for C, C++, Prolog, assembler, and Pascal).

Philippe is the inventor of 235+ issued patents, of which 125+ patents are assigned to Fullpower-AI® and the AI, ML, Sleep, and medical Fullpower-AI® Cloud platform: (for a list of Philippe’s patents, follow the link: philippekahn.com/download/philippe-kahn-patents-list, patents.google.com/?inventor=Philippe+Kahn).

Philippe studied at the ETH in Zurich Switzerland and Sofia-Antipolis, France, receiving his Master's in Mathematics. Philippe also earned his Master's in classical flute, with simultaneous minors in composition and chamber music from the Zurich Music Conservatory.

Philippe is credited for creating the first camera phone solution to share photos instantly over public cell phone networks. During his daughter's birth on June 11th, 1997, Philippe completed his first camera phone prototype by wiring a digital camera to a mobile phone and uploading an image to his home-based webserver to share the photo in real time with several hundred contacts. His "Point, Shoot, Share, Instantly" architecture model for instant-sharing would eventually become the product that was licensed to wireless carriers by the company Philippe and Sonia founded, Lightsurf Technologies, in addition to becoming a blueprint for today's social media. In 2016, Time Magazine included Philippe's first camera phone photo in their list of the 100 most influential photographs of all time.

Philippe is fluent in English, French, Spanish, and German. He is passionate about AI, Machine Learning, and his family. Philippe is also a trustee of the Lee-Kahn Foundation.

Mark Christensen, VP of Engineering

Mark Christensen, VP of Engineering

Mark is the head of engineering at Fullpower-AI®. He has over 15 years of technology development experience. At Fullpower-AI®, Mark is passionate about AI, machine learning, deep learning and data sciences. Mark studied mechanical engineering at the University of Auckland in New Zealand.
Eric Smith, VP of Quality and Security

Eric Smith, VP of Quality and Security

Eric Smith is the head of accuracy and quality assurance at Fullpower-AI®, responsible for the QA and customer support teams. He has been with Fullpower since 2008 and has over 18 years of quality assurance experience. Prior to joining Fullpower-AI®, Eric was a QA Engineer with Seagate and Starfish and a Senior QA Engineer with LightSurf when it was acquired by VeriSign. Eric's team ensures Fullpower-AI® maintains the most accurate, repeatable and reliable end-to-end solutions on the market using sophisticated software tools for testing and tracking development. Eric also manages Fullpower-AI®'s 3D printing technology for wearable & IoT device prototype development and testing. Eric received his Bachelor of Science in Mechanical Engineering from the University of California, Santa Barbara.
Dr. Venkat Easwar, VP of AI and Data Science

Dr. Venkat Easwar, VP of AI and Data Science

Venkat leads the AI and Data Science team at Fullpower-AI®. He has been with Fullpower since 2007 and has over 25 years of product architecture and algorithm development experience.  He has extensive experience in both cloud and edge processing. Prior to Fullpower, Venkat worked at Verisign on Messaging and Voice Response Gateways; at Lightsurf Technologies, on Multimedia Messaging Systems and Media Transcoding; and at Texas Instruments on DSPs, Digital Video, and Imaging. Venkat received his Bachelor’s degree in Electrical Engineering (Electronics) from Indian Institute of Technology, Madras, and his Ph.D. in Electrical Engineering from the University of Southern California. His Ph.D. thesis was in the area of AI and Computer Vision.
Dr. Anil Rama, MD, Medical Advisory Board

Dr. Anil Rama, MD, Medical Advisory Board

Dr. Anil Rama, MD is the founder of Sleep and Brain and serves as Adjunct Assistant Clinical Professor at the Stanford Center for Sleep Sciences and Medicine. He is the recipient of the 2021 Stanford Distinguished Service Award, awarded to one clinical faculty member per year. He is the former Medical Director of Kaiser Permanente’s tertiary sleep medicine laboratory. Dr. Rama is also an editorial board member of the Sleep Science and Practice Journal and has authored several book chapters and seminal peer-reviewed journal articles in sleep medicine. Furthermore, Dr. Rama is a lecturer for the Dental Sleep Medicine Mini-Residency at the University of Pacific, Arthur A. Dugoni School of Dentistry. In addition, Dr. Rama has been an investigator in clinical trials for drugs or devices designed to improve sleep and contributed to stories in national newspapers, local news stations, wellness websites, and health newsletters.
David Shanes, VP of Regulatory Affairs

David Shanes, VP of Regulatory Affairs

David is responsible for managing the Regulatory and Compliance efforts for Fullpower-AI® and advancing its solutions into the healthcare and medical device spaces. David has more than 18 years of medical device experience overseeing Quality and Regulatory, Hardware and Software Engineering, Research and Development, and Manufacturing activities. Prior to joining Fullpower, David held various executive positions at TruMed Systems, Inc., Biocare Medical, LLC., and BioTelemetry, Inc. He holds a BS in Computer Science from the United States Naval Academy and an MS in Computer Science from San Diego State University. David also served in the U.S. Navy for 11 years.
Tom Lewin, VP of Customer Success

Tom Lewin, VP of Customer Success

Tom is the VP of Customer Success and has spent valuable time on the Fullpower-AI® R&D team, Quality Assurance, Customer Support, and Development teams. Tom is responsible for in-house demos, customer onboarding, assists in the sales process, and manages the Tempur-Pedic relationship. Tom is passionate about delivering a quality experience to end users and business partners and has worked diligently to achieve the ratings success you see with our apps and products. Tom has been with the company since 2008 after studying at Old Dominion University.

Contact Information and Headquarters

Silicon Valley

Fullpower Technologies, Inc.
1200 Pacific Avenue, Suite 300
Santa Cruz, CA 95060
USA
ISO27001
AICPA SOC2
GDPR
AWS Qualified Software
AWS Standalone Partner

AIoT Biosensing Case Study: Globally vetted and deployed AIoT medical and wellness PaaS for PSG-grade contactless sleep, vital signs, and remote patient monitoring, including clinical trials, compliance, and regulatory.

Fullpower-AI® Sleep Research & Technology Expertise

Over 100,000 New Daily Live Home Sleep HD Recordings With 5-Years Of History & 250+ Million Nights Of Sleep
Recorded with high fidelity, analyzed, and processed by our sleep experts using AI and machine learning, our tools analyze differences and changes over time in detailed sleep patterns day-after-day.
Real-Time Dashboard Tools
Remotely monitor patients or all trial subjects with event alerts and notifications. This includes remote vital signs, respiratory events, and sleep patterns of patients.
AI-powered, Highly Accurate & Cloud-Based
Exceeds 90% accuracy of gold-standard Polysomnography in many key metrics. Large anonymized dataset of demographically diverse subjects ideal for use as a statistical reference to back trials and remote monitoring.
Complete Integrated Machine Learning Modeling Tools With a Complete Set of AI-Powered Analytical Tools
Supports supervised and unsupervised learning, deep analysis, and infographics with statistical backing.
125+ Patents, 10+ Years Of Sleep-Science And AI Leadership
Spanning AI, Machine Learning, Biosensing, Health, Cognitive Behavioral Science, Sleep Science, and more.
Two Sleepers With Correlations
The data shows that two out of three beds have two sleepers. Two sleepers can be monitored and correlated to measure some of the effects of one sleeper on the other and build meaningful models.
In sleep science, for example, we use AI to analyze sleep data collected by the Sleeptracker-AI system to help identify different sleep stages, detect sleep disorders, and provide users with personalized sleep recommendations. With AI data science methodologies, we can match closely gold standard PSG sleep analysis and continue to improve accuracy and applicability in real-world settings. We, of course, in that process carefully consider bias and other ethical issues. Our methodologies include deep learning, supervised learning, and reinforcement learning techniques. To continually generate training data, we operate two clinical PSG sleep labs.

Polysomnographic validation of the Sleeptracker-AI® solution
in estimating sleep architecture and obstructive sleep apnea in adults

Polysomnographic validation of the Sleeptracker-AI® solution in estimating sleep architecture and obstructive sleep apnea in adults, in collaboration with Stanford Medical Research, as presented by Dr. Clete Kushida at the 2022 Sleep World Congress in Rome (Italy).

Reference: Ding, F., Cotton-Clay, A., Fava, L., Easwar, V., Kinsolving, A., Kahn, P., Rama, A., & Kushida, C. (2022). Polysomnographic validation of an under-mattress monitoring device in estimating sleep architecture and obstructive sleep apnea in adults.Sleep medicine, 96, 20-27.
https://www.sciencedirect.com/science/article/pii/S1389945722001368

End-To-End Use Case: Globally Deployed AIoT PaaS for Contactless Sleep and Remote Patient Monitoring

We use AI to analyze sleep data anonymously collected by the Sleeptracker-AI® system to help identify different sleep stages, detect sleep disorders, and provide users personalized sleep recommendations. We can closely match in-the-home, contactless, gold-standard PSG sleep analysis in real-world settings with machine and deep learning models and statistical inference techniques. We, of course, in that process carefully consider bias and other ethical issues. Our methodologies include deep learning, supervised learning, self-supervised learning, and reinforcement learning techniques. To continually generate training data, we operate multiple clinical PSG sleep labs.

A two-year+ study on the COVID-19 effect on sleep

The COVID-19 pandemic has affected sleep in multiple ways. Technological advances www.sleeptracker.com in-home sleep monitoring has provided the opportunity to analyze sleep-wake patterns on a scale much larger than previously imaginable. This study compares the estimated sleep-wake patterns in the time before and after the start of the pandemic in a large U.S. sample.

Sleep parameters [estimated total sleep time (TST), bedtime (BT), and morning rise time (RT)] were analyzed. The device passively monitors sleep using piezo-electric sensors that register the forces exerted through the mattress by features such as the individual's motion, respiration, heartbeats, and snoring vibrations. The de-identified data obtained from the devices were analyzed, following review and exemption of the study (#57681) from the Stanford University IRB. Data from the calendar years 2019 and 2020, from 62,152 individuals with 20,255,441 recorded nights, were available. Individuals who had at least 300 nights of sleep each year were included in the analytic dataset, with Sunday nights through Thursday nights analyzed since sleep on weekend nights was expected to have more variability.

Irregular sleep-wake schedules and the impact on health

Cardiovascular disease, increased blood pressure, stroke, and insulin resistance are associated with Irregular sleep-wake schedules

This very large study is the first extensive home study with more than 4 million nights of accurately recorded sleep patterns.

Interestingly, this study finds a clear age-dependent trend, with older age corresponding to more regular sleep-wake schedules.

The importance of REM-rebound

REM sleep is an essential component of sleep. "REM-rebound” is an evolutionary mechanism to help us get some of that precious REM-sleep back.

Most of us are typically sleep-deprived during weekdays and attempt to recover our lost sleep on the weekends. This first-of-a-kind, extensive study of more than 6.8 million nights of sleep as per Sleeptracker-AI explores the weekly rapid-eye-movement (REM) sleep deprivation-recovery cycle. This study validates the repetitive REM sleep deprivation-recovery cycle. Individuals are, on average, partially sleep-deprived starting Sunday night, which leads to a progressive REM sleep rebound that transitions into a REM recovery cycle on Friday and Saturday nights.

Background: After early research connected rapid eye movement with dreaming and established that it made up about 20% of normal human sleep, experimenters started depriving test subjects of only REM sleep to test its unique importance. Whenever a subject's electroencephalogram and eye movements indicated the beginning of REM sleep, the experimenter would thoroughly wake them for several minutes. As this "dream deprivation" continued, the tendency to initiate REM increased and the subjects were woken up more and more times each night. As a result, the subjects became irritable, anxious, and hungry, and several left the study early. Finally, after five nights, the remaining issues were allowed to sleep undisturbed and showed a significant increase in the percentage of sleep devoted to REM: from an average of 19.4% to an average of 26.6%. These effects were significant in comparison with a control group woken up on an equal number of occasions each night, at random times.

Further reading:

What is REM sleep?
https://www.sleepassociation.org/about-sleep/stages-of-sleep/rem-sleep/

The importance of REM rebound and CPAP compliance
https://www.sciencedirect.com/science/article/abs/pii/S1389945712001864

How bed-partners impact each other's sleep quality

A joint study between Stanford Medical Research and Fullpower-AI.

Sharing the bed with a partner is common among adults and is likely to impact sleep in multiple ways. However, gold standard polysomnograms are performed without a bed partner, and objective data on co-sleeping couples is scarce.

This large study of over 5,000 users investigates the effects of a bed partner's presence on objective sleep parameters.

This detailed, science-based study suggests that when the bed partner is absent, an individual's sleep architecture shows, on average, a higher sleep efficiency, with less micro-awake time and total sleep, more minutes spent in deep and REM sleep, and less in light sleep. This suggests a less interrupted night, perhaps due to fewer disruptions from the partner, where the individual has enough continuity in their sleep to transition to deeper stages. Just like with pets sharing our bed, it's challenging to quantify the counter-balancing positive emotional impact of having a bed partner.

Further reading:

The effects of sleep continuity disruption on positive mood and sleep architecture in healthy adults
https://academic.oup.com/sleep/article/38/11/1735/2662282

Snoring and sleep architecture
https://www.atsjournals.org/doi/abs/10.1164/ajrccm/143.1.92

Reference: Zitser, J., Cotton-Clay, A., Baron, S., Easwar, V., Kinsolving, A., Kahn, P., & Kushida, C. (2022). 0353 Estimated Sleep-Wake Patterns Obtained from a Large US Sample by Home-Based Under-Mattress Monitoring Devices. Sleep, 45(Supplement_1), A159-A159.
https://academic.oup.com/sleep/article/45/Supplement_1/A159/6592308

Evaluation of Sleep-Related Respiratory Events

In an extensive longer-term study, the data shows that more than 14% show moderate to severe apnea (AHI>15) as presented at the Athens Greece sleep conference, September 2022.

This is the first known large-scale continuous Sleep Apnea study. Sleep Apnea is a potentially serious sleep disorder. It causes breathing to stop and repeatedly start during sleep. Studies have found a direct correlation between High Blood Pressure, Diabetes, Stroke, heart attack, and Apnea. In a large study of over 75,000+ sleepers, over 14 million nights of sleep. The data shows that more than 14% experienced serious sleep apnea at least one night a week. There are effective therapies for apnea. The diagnosis must come first. There are several types of sleep apnea, but the most common is obstructive sleep apnea. This type of apnea occurs when your throat muscles intermittently relax and block your airway during sleep. A noticeable sign of obstructive sleep apnea is snoring. Population studies have estimated the prevalence of sleep-related respiratory events characteristic of obstructive sleep apnea (OSA) and reported night-to-night variability in OSA severity. Still, these have been constrained by the inability to obtain continuous nightly data on a large scale. The current study is the largest to date for evaluating these events' prevalence and night-to-night variability. The de-identified data from 2021-04-01 to 2022-03-31, in 76,769 individuals with 14,296,394 recorded nights. Individuals with at least 300 nights of recordings were included in the analytic dataset.

Sleep Duration Effect on Heart and Respiratory Rate

This new, large long-term study by Stanford and Fullpower-AI may show that we should strive to sleep between 6 and 9 hours. Not less and probably not more. As presented at the Athens, Greece, 2022 Sleep summit two weeks ago.

Intuitively sleep deprivation is unhealthy. This very large study shows that sleeping too much, hence 9 hours or more, may not be beneficial. This study looks at the length of actual sleep as opposed to the time spent in bed. This correlation could mean that 6.5 to 8 hours of sleep may be optimal for health. That's because resting heart and respiration rate are generally considered signposts of wellness.

Sleep data from 76,769 users, with 14,296,394 total recorded nights from 2021-04-01 to 2022-03-31; only subjects with at least 300 nights of recordings during the period were included. In total 18,252 individuals (40% female, 13% unspecified gender, mean age 49) with 5,846,745 recorded nights met this inclusion criterion. Estimated total sleep time (TST) was categorized as: <5 hours, 5-6 hours, 6-7 hours, 7-8 hours, 8-9 hours, and >=9 hours. Normalized Heart Rate (HR) and Respiration Rate (RR) for a recording were taken to be the mean HR and RR for that recording as a percentage of the average over all recordings for that subject. Excess HR and RR for a recording were taken to be the excess/deficit of the normalized HR and RR over 100%. Subjects had lower HR and RR than average on nights when they slept 7-8 hours. Interestingly, their HR was higher than average on nights when they slept <6 hours or >=9 hours.

Notably, the American Academy of Sleep Medicine recommends >=7 hours of sleep without an upper limit. Furthermore, these findings may inform on the relationship between extreme sleep duration as a risk factor for cardiovascular events.

Here is an article from Mayo Clinic about Cognitive Behavioral Therapy for sleep.

Discuss on LinkedIn

The number of nights needed for OSA detection

Obstructive Sleep Apnea (OSA) is a sleep disorder that occurs when breathing is repeatedly interrupted during sleep. It's often caused by a blockage in the airway, leading to loud snoring, choking, and gasping during sleep. OSA can cause daytime sleepiness, fatigue, and other health problems, and getting a proper diagnosis and treatment is important. OSA is typically diagnosed through a sleep study, which can be conducted in a sleep lab or at home. During a sleep study, various measurements are taken to evaluate your sleep, including breathing patterns, oxygen levels, and brain activity.

Clinical sleep studies typically rely on one night for OSA detection and diagnosis. However, uncertainty exists regarding the degree of AHI stability across different nights. Population studies collecting continuous nightly data on a large scale enable the detection of night-to-night variability in OSA severity; this study is the largest to date for evaluating the number of nights to achieve high sensitivity/specificity for OSA detection. This extensive study led by Dr. Clete Kushida of Stanford Medicine builds on 96,228 individuals with 19,148,323 recorded nights.

Reference: Kushida, C., Cotton-Clay, A., Fava, L., Easwar, V., Kinsolving, A., & Kahn, P. (2023). 0502 Number of Nights to Achieve High Sensitivity/Specificity for Detecting OSA Using a Large US Sample by Home Under-Mattress Devices. Sleep, 46(Supplement_1), A222-A223.
https://academic.oup.com/sleep/article/46/Supplement_1/A222/7182096

Reference: Kushida, C., Cotton-Clay, A., Baron, S., Fava, L., Easwar, V., Kinsolving, A., & Kahn, P. (2023). Detection and Prevalence of OSA in Men and Women using a Continuous Large US Sample by Home Under-Mattress Devices. CHEST, 164(4), A6286-A6287.
https://journal.chestnet.org/article/S0012-3692(23)05084-5/fulltext

New very large Sleeptracker-AI® Sleep Study in partnership with Stanford Sleep Medicine

Fullpower-AI®, in a very large-scale sleep study, presented new findings at the World Sleep Congress in Brazil. This study shows the first large-scale quantification of declines in deep and REM sleep with age. Practically speaking, this is one of the reasons why maintaining enough deep sleep can be vital for overall health.

With the Sleeptracker-AI® platform, which leverages deep learning and generative AI, and in collaboration with Stanford Sleep Medicine, Fullpower-AI® studied 24,850,420 nights of sleep and presented findings at the World Sleep Congress in Brazil.

While it's recommended that adults sleep at least seven hours per night, this large-scale study provides new insights into how sleep duration and architecture vary with age. Most people undergo 75-80% of their sleep in NREM (non-REM sleep) and the rest in REM. The study shows that the middle-aged population tends to sleep less, and as people age, there's a significant decrease in deep sleep consistent across ages. In REM sleep, steep declines begin in late middle age. Notably, those with sleep apnea experience an even more significant decrease in deep sleep regardless of age, emphasizing the impact of sleep apnea on overall health and the need for remediation. 

What This All Means for Improved Health

Understanding sleep patterns and architecture is crucial for overall health. These findings emphasize the importance of adequate sleep quality, especially as one ages. The decline in deep sleep, particularly for those with sleep apnea, indicates the need for proper diagnosis and treatment. By addressing these sleep disturbances, individuals can work towards achieving better health and reducing risks associated with sleep deficiencies.

Fullpower-AI® New Findings in Major Apnea Study

Dr. Clete Kushida of Stanford presents a new extensive study, collaborating with Stanford Sleep Medicine and Fullpower-AI®, at the World Sleep Congress 2023 in Rio de Janeiro.

The study shows the health benefits of consistent CPAP use by people with apnea in 720 female and male patients, using 58,000+ nights of sleep monitored in high fidelity with Sleeptracker-AI® in their homes. Sleeptracker-AI® can monitor sleep during CPAP usage, including when it is not used. The findings show significant decreases in breathing anomalies when CPAP is used.

Sleeptracker-AI® uses AI to analyze sleep data anonymously collected to help identify different sleep stages, detect sleep disorders, and provide users with personalized sleep recommendations. It closely matches in-the-home, contactless, gold-standard PSG sleep analysis in real-world settings with machine and deep learning models and statistical inference techniques while carefully considering bias and other ethical issues. Its methodologies include deep, supervised, self-supervised, and reinforcement learning techniques. To continually generate training data, multiple clinical PSG sleep labs are in operation.

Whether one has Obstructive Sleep Apnea (OSA) or Central Sleep Apnea, the data shows that consistent use of a CPAP can make a difference in sleep quality and overall health.

Here's why:

OSA: With OSA, the airway gets blocked during sleep, causing breathing interruptions and sleep disturbances. A CPAP machine delivers a continuous pressurized airflow, keeping the airway open and ensuring stable oxygenation (SpO2).

Central Apnea: For those dealing with Central Sleep Apnea (about 10% of the population diagnosed with apnea), it's neurological, and the brain's signal to breathe seems to be temporarily disrupted. A CPAP with adaptive settings can help stabilize breathing patterns.

Sleeptracker-AI® studies also show the benefits of consistent CPAP use:

  • Improved sleep quality: wake up feeling refreshed and energized.
  • Increased alertness: better concentration and daytime alertness.
  • Reduced health risks: Lowered risk of heart disease, stroke, and other sleep apnea-related health issues.
  • Enhanced mood and mental outlook.

The data clearly shows that with apnea, the key is a consistent use of the CPAP.

Sleeptracker-AI® research is now validated by Stanford University, Division of Sleep Medicine.

Fullpower-AI® is the leader in Person/Patient-Generated
Sleep Data with our Sleeptracker-AI® Platform (PGHD [1])

Sleep is one-third of our lives; wearables are invasive. Yet, sleep is a crucial signpost for health and changes in health. All of an individual's sleep experience is outside of a sleep lab. However, clinicians and researchers fly blind to this aspect of an individual's sleep and change over time. Sleeptracker-AI's network of sleepers is highly motivated to participate in managing their health. We complement their active engagement with the passive deep analysis of their anonymized data with their consent.

A significant fraction of individuals over the age of 30 show breathing anomalies during sleep, with estimates ranging up to 50%, including some of the more severe varieties [2][3][4]. This ranges from habitual snoring to life-threatening COPD and sleep apnea (including Central and Obstructive). These conditions often correlate with diabetes, hypertension, stroke, and heart attack risks. The Sleeptracker-AI platform delivers the first in-home, non-invasive, automatic, long-term sleep analysis solution, together with all the necessary data science tools and analytical dashboards powered by AI.

  1. Patient-Generated Health Data, HealthIT.gov https://www.healthit.gov/topic/scientific-initiatives/pcor/patient-generated-health-data-pghd
  2. Garvey JF, Pengo MF, Drakatos P, Kent BD. Epidemiological aspects of obstructive sleep apnea. J Thorac Dis. 2015;7(5): 920-929.
  3. Heinzer R, Vat S, Marques-Vidal P, et al. Prevalence of sleep-disordered breathing in the general population: the HypnoLaus study. Lancet Respir Med. 2015;3(4):310-318.
  4. Adeloye D, Chua S, et al. Global Health Epidemiology Reference Group (GHERG). Global and regional estimates of COPD prevalence: Systematic review and meta-analysis. J Glob Health. 2015 Dec;5(2):020415.

Fullpower-AI® Person/Patient Generated Sleep Data serve as Synthetic Control Arms,
saving time and money in clinical trials

Fullpower-AI® synthetic control arms use validated, real-world person/patient-generated sleep data as comparators for clinical trials instead of collecting data from patients recruited for a trial who have been assigned to the control arm. This halves the number of participants needed for clinical trials, speeding up trials and decreasing their costs.[1]

  1. Synthetic Control Arms can save time and money in clinical trials, StatNews.com
    https://www.statnews.com/2019/02/05/synthetic-control-arms-clinical-trials/

Sample Analytics from the Sleeptracker-AI Live Dataset

Loading...

Patents

The Fullpower-AI® platform is backed by a patent portfolio of 135+ patents

AI, Machine Learning, Sensor Fusion

AI, Machine Learning, & Sensor Fusion

IoT, Cloud, MACS (Monitor Alert, Control Share)

IoT, Cloud, & MACS
(Monitor Alert, Control Share)

User Experiences

User Experiences

Cognitive Behavioral Science and Sleep Science

Cognitive Behavioral Science & Sleep Science

Digital Sports, Health Monitoring, Medical

Digital Sports, Health Monitoring, & Medical

Cloud and wearable

Cloud & Wearable

Technology Fully Covered
Now and In the Future

Fullpower Fully Covered Now and In the Future