Sleep Staging Algorithm Patents: Top Companies & Trends 2026
- 50.4% of all 1,981 records sit with just five assignees, and 58.6% with the top ten — a field concentrated at the top with a long tail below it.
- Filings grew 34% from 2021 to 2024 (153 to 205), with a 2022 peak of 252, after 2025-2026 figures understate true activity due to publication lag.
- A61B diagnostic sensing covers 77.0% of records while AI-model computing (2.9%) and power-supply integration (2.0%) remain comparatively open.
Filing growth compares 2021 (153 records) with 2024 (205) — a three-year span. 2024 is the most recent year we treat as complete: publication lags filing by roughly 18 months, so 2025 onwards are still filling in and any growth rate that ends there would understate the field. Top-5 share is the combined record count of the five largest assignees divided by all 1,981 records in scope (CR5), not by the ranked leaders only.
What sleep staging algorithm patents actually cover
Sleep staging algorithms classify a sleeper’s physiological state — wake, light, deep, REM — from signals such as EEG, PPG, respiratory effort or heart rate variability. The patent activity tracked here spans 1,981 records filed between 2015 and mid-2026, covering everything from single-sensor wearable staging to full clinical polysomnography scoring systems. Most filings sit inside diagnostic and surgical device classifications, reflecting how staging inventions are typically framed around the sensor and signal-processing pipeline rather than the classification model alone.
Filing activity is concentrated: half of all records in scope belong to just five assignees, and the growth curve, while softened in the most recent one to two years by publication lag, shows real momentum through 2024. The technology mix is uneven too, with sensing hardware claims far outnumbering claims around AI-model architecture or power management, which is where the least-contested claim space currently sits.
Let an AI agent run this analysis on your own technology
Pick a task. Every answer cites the patents behind it.
How sleep staging patent activity has grown and where it concentrates
Filing volume in sleep staging algorithms has climbed steadily since 2017, with technology claims spread across diagnostic sensing, informatics and, to a lesser degree, power and AI-model architecture.
Filing trend, 2017-2026
Annual filings rose from 45 in 2017 to a peak of 252 in 2022, with 2021 to 2024 filings up 34% (153 to 205). The most recent one to two years are understated because publication typically lags filing by about 18 months.
Technology composition by IPC subclass
A61B (diagnosis and surgery) covers 77.0% of the 1,981 records in scope, far ahead of A61M devices for body fluids (19.0%), G16H healthcare informatics (17.4%) and G06F data processing (11.9%). Because records can carry multiple IPC classes, these shares add up to more than 100%.
Shares are the percentage of the 1,981 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.
Go deeper on Sleep Staging Algorithms with Eureka
This page is one run against one query. Ask Eureka your own question about sleep staging algorithms and every answer comes back with the patent numbers behind it.
Try EurekaA closer look at one filing that shapes the claim landscape
Sleep stage classification system (US20200093422A1)
A sleep stage classification system estimates features associated with a user's respiratory effort from the output signal of a PPG sensor, determines respiratory effort from that feature signal, and classifies the user's sleep stage based on the determined respiratory effort — using a single photoplethysmography input rather than a dedicated respiratory sensor.Filed by Koninklijke Philips N.V., published 2020-03-26.
| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | US20050115561A1 | Patient monitoring, diagnosis, and/or therapy systems and methods | 1,106 |
| 2 | US20070118054A1 | Methods and systems for monitoring patients for clinical episodes | 692 |
| 3 | US20040230105A1 | Adaptive prediction of changes of physiological/pathological states using processing of biomedical signals | 541 |
| 4 | US20040193068A1 | Methods and apparatus for monitoring consciousness | 481 |
| 5 | WO2016110804A1 | Mobile wearable monitoring systems | 466 |
| 6 | US20210169417A1 | Mobile wearable monitoring systems | 450 |
| 7 | US20160151603A1 | Methods and systems for sleep management | 413 |
| 8 | US7314451B2 | Techniques for prediction and monitoring of clinical episodes | 404 |
| 9 | US20050124864A1 | System and process for non-invasive collection and analysis of physiological signals | 392 |
| 10 | US7225013B2 | Adaptive prediction of changes of physiological/pathological states using processing of biomedical signals | 388 |
Citation counts favor older records simply because they have had more time to accumulate citations within the searched corpus — read them as a signal of influence, not of current relevance.
Each row carries its publication number; clicking a row searches Eureka by that number.
Put your own technology through the same analysis
Eureka on the web
When you want the answer in the next five minutes.
The agent works the prompt against patents and technical literature, citing every source.
Run your analysis now →MCP server & REST API
When it has to run inside your own pipeline.
Patent search, landscape analysis and assignee resolution as MCP tools. Drop them into any agent framework, or call REST directly.
Browse MCP servers →What the filing data says about direction of travel
Three patterns stand out once family counts, IPC classes and citation data are read together: concentration at the top, a real but understated growth curve, and a narrow set of branches still open to new claims.
A small group controls half the field
The five leading assignees together hold 999 of the 1,981 records in scope, and the top ten extend that to 58.6%. New entrants competing head-on in the dominant sensing claim space are competing against a handful of well-established portfolios, not a fragmented market.
Activity is still rising through the last complete year
Filings climbed from 153 in 2021 to 205 in 2024, with an interim peak of 252 in 2022. Because publication lags filing by roughly 18 months, the softer-looking 2025-2026 counts do not indicate a slowdown — they indicate incomplete data.
Sensing and diagnostic claims dominate
A61B is present in more than three-quarters of the 1,981 records, well ahead of A61M (19.0%) and G16H (17.4%). Records can carry multiple IPC classes, so this reflects claim density in the sensing layer rather than a simple category split.
Power and AI-model claims are the thinnest branches
H02J (power supply integration) and G06N (AI-model computing) sit at 2.0% and 2.9% of records respectively, the smallest of the eight subclasses tracked. That leaves room for claims combining on-device model efficiency with power-aware sensing.
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to sleep staging algorithms, with the prior art for and against each one.
Who is filing, and how positions are shifting
The ranked field runs 100 companies deep, but the shape of the competition is set by a small leading group whose filing pace has already started to cool, alongside branches of the technology that remain largely unclaimed.
A single wearable-sleep specialist sets the pace
The top-ranked assignee holds 576 of the 1,981 records in scope, more than five times the fifth-place holder's 39. That gap is unusually wide for this kind of field and suggests an entrenched claim position around consumer wearable staging.
Competition is a portfolio review, not a market scan
With 58.6% of all records held by the top ten assignees, most freedom-to-operate work in this space reduces to checking a short list of portfolios rather than surveying a fragmented market of small filers.
Filing pace is slowing even at the top
The leading assignee's latest-year filing count fell 63% year-over-year, and several other leading assignees show flat or zero latest-year filings. That slowdown at the top, set against overall field growth through 2024, points to claim space maturing faster for incumbents than for the field as a whole.
Joint filings are rare and tightly clustered
Only 10 co-assignee pairs appear across the dataset, several of them pairing a corporate entity with a named inventor, which suggests most staging patents are filed by a single organisation rather than through joint ventures or research partnerships.
| Assignee | Recent year | YoY |
|---|---|---|
| Oura Health Oy | 35 | -63% |
| ResMed Pty Ltd | 2 | -50% |
| Koninklijke Philips N.V. | 1 | -88% |
| ResMed Sensor Technologies Ltd | 0 | -100% |
| EARLYSENSE | 0 | — |
| Beth Israel Deaconess Medical Center Inc | 0 | — |
| Cardiac Pacemakers Inc | 0 | — |
| WIDEMED | 0 | — |
Where to take this analysis
The filing data points to a field that is concentrated at the top but still has real open space in specific technical branches. The next steps depend on whether the goal is freedom-to-operate, portfolio strategy, or identifying a defensible new claim.
Run a freedom-to-operate check against the leading portfolios
With 58.6% of records held by ten assignees, a targeted review of those portfolios covers most of the practical risk in this field, rather than a broad sweep of the full 1,981-record set.
Search assignee portfolios in EurekaScope a claim in the under-filed branches
AI-model computing and power-supply integration remain the thinnest IPC categories in the dataset, and are worth a closer prior-art check before drafting.
Explore white space in EurekaTrack filing momentum before committing R&D budget
Several leading assignees show slowing latest-year filing counts even as the field overall grew 34% from 2021 to 2024 — worth monitoring before assuming any one competitor's position is fixed.
Set up momentum tracking in EurekaCommon questions about sleep staging algorithm patents
One assignee holds 576 of the 1,981 records in scope, a clear lead over the rest of the ranked field. The top 5 assignees together account for 50.4% of all records, and the top 10 account for 58.6%, so filing activity is concentrated among a small group of wearable and medical-device companies rather than spread evenly across many filers. Beyond that group there is a long tail of single- or few-filing entrants.
Diagnosis and surgery claims under IPC class A61B dominate, covering 77.0% of the 1,981 records in scope, reflecting how most sleep-staging inventions are framed around the sensing and diagnostic hardware. Healthcare informatics (G16H, 17.4%) and general data processing (G06F, 11.9%) are the next largest groups, covering the software and workflow layers built on top of that sensing. Smaller categories like AI-model computing (G06N, 2.9%) and power supply integration (H02J, 2.0%) are comparatively under-filed.
Filings grew from 45 in 2017 to a peak of 252 in 2022, and the 2021-2024 span shows a 34% increase (153 to 205 filings). Filing counts for 2025 and 2026 appear lower in the raw data, but that reflects publication lag of roughly 18 months rather than an actual slowdown, since recent applications have not all published yet. The underlying trend through the last complete year, 2024, is one of continued growth.
US20200093422A1, assigned to Koninklijke Philips N.V. and published in March 2020, claims a sleep-stage classification system that derives a respiratory-effort feature from a PPG sensor signal and uses it to determine sleep stage. This narrows the design space specifically for PPG-only devices that infer respiratory effort as an intermediate step before classification. Systems relying primarily on EEG signals, actigraphy, or a dedicated respiratory sensor rather than a PPG-derived estimate sit outside this particular claim scope, though a full freedom-to-operate review should still check neighboring filings from the same assignee family.
The smallest IPC branches in the dataset, AI-model computing (G06N, 2.9% of records) and power-supply integration (H02J, 2.0%), represent the most under-claimed territory relative to the dominant A61B sensing claims. This suggests opportunity in claims that tie specific on-device classifier architectures to power-budget constraints, an area barely touched by the current filing base. Recent-year momentum data also shows some leading assignees slowing their filing pace, which can open room for new claim strategies in adjacent branches rather than head-on competition in the crowded sensing core.
Research Sleep Staging Algorithms in depth with Eureka
Go past this page: query the whole sleep staging algorithms corpus yourself, in your own scope.
Every answer comes back with patent numbers you can open.
Disclaimer. This page is generated from Patsnap Eureka data drawn from a limited snapshot of global patent and scientific-literature records, and is provided for general information and reference only.
Patent data carries inherent limitations: recent filings (typically the most recent 18–24 months) are under-counted due to standard publication lag; counts may be reported at either a patent-family or a patent-record basis and are not always directly comparable; classification, applicant-name, and citation data may contain errors, duplicates, or omissions; and the underlying search query defines and constrains the scope shown. As a result, the analysis may be incomplete or inaccurate and may not reflect the full technology landscape.
Nothing on this page constitutes an exhaustive prior-art, novelty, freedom-to-operate, or validity search, nor does it constitute legal, financial, investment, or professional advice, and it should not be relied upon as such. Any patent, commercial, or strategic decision should be verified independently and reviewed with qualified patent, legal, and domain professionals. Patsnap makes no warranties, express or implied, as to the accuracy, completeness, or fitness for any particular purpose of the information presented.
Machine translation. Assignee and organisation names originally recorded in Chinese, Japanese or Korean have been rendered into English by an AI translation step so that the tables stay readable. These renderings are best-effort and may not match a company’s registered English name; the original name is what the underlying patent record carries, and it is what any Eureka query launched from this page uses.