US 5944680

I'll verify the patent details and check for any 2026 CAFC activity before summarizing.

Let me try to pull the actual claim language to ground the independent-claim summary.

Scope of search and caveats

Limitation up front: I do not have direct API access to USPTO PatentCenter or PACER/CAFC docket systems. I ran web searches against indexed USPTO/EPO/Google Patents data and general web sources. I found no Court of Appeals for the Federal Circuit 2026 docket, appeal, or IPR involving US 5,944,680. Given the patent's status, this is expected (see below), but I cannot positively certify a negative from this tool set — it is an absence of evidence, not a certified clearance.

Identifier-collision warning (I am not auto-correcting this): Search results returned JP 5944680 B2 (特許第5944680号, issued 2016-07-05, KUMAGAI GUMI CO LTD — a concrete formwork / floor underlayment patent). That is a different, unrelated patent sharing the numeric suffix 5944680. It has nothing to do with US 5,944,680 and must not be conflated.

Bibliographic data — US 5,944,680 A

Because the anticipated expiration date passed on 2016-06-26, there is no live patent term on which a 2026 CAFC appeal could plausibly hang.

Abstract

"A method of predicting critical points in patient respiration includes monitoring at least one characteristic of a respiratory effort waveform of a patient to detect a respiratory event. A refractory period is defined that includes a hard refractory period during which time the respiratory event cannot be responded to and a soft refractory period following the hard refractory period. The respiratory event outside of the refractory period is detected as a function of a first set of predetermined parameters for the monitored at least one characteristic and the respiratory event within the soft refractory period is detected as a function of a second set of predetermined parameters for the monitored at least one characteristic. The respiratory event may be inspiration onset and the characteristic of the respiratory effort waveform monitored is at least one of slope and amplitude. The refractory period may be defined based on detection of inspiration offset and further may be defined based on inspiration offset, an average respiratory period, and an average time of inspiration. Further, stimulation may be provided in response to a detected inspiration onset. Another method of predicting critical points includes sampling the amplitude of the respiratory effort waveform of a patient. A sample signal is generated representative of at least one characteristic of the respiratory effort waveform based on each amplitude sample. The sample signals representative of the at least one characteristic of the respiratory effort waveform are monitored and a respiratory event is detected as a function of at least two sample signals. Apparatus and systems for use with such methods are also described."

Independent claims — plain-language overview

Important caveat: the granted claim set was not included verbatim in the full text I retrieved (the fetched document contains the specification and the "Definitions"/summary-of-invention paragraphs, but not the numbered claims). The overviews below are reconstructed from those summary paragraphs, which in this patent track the independent claim structure closely. Exact claim numbering, and whether some of these are separate independents vs. dependent claims, is not verified by the sources I retrieved — treat the numbering as indicative, not authoritative.

1. Method of predicting critical points in patient respiration (likely claim 1). Watch a characteristic of the patient's respiratory effort waveform. Define a refractory period split into a hard refractory period (during which the detected respiratory event cannot be acted upon/responded to) and a following soft refractory period. Detect the event outside the refractory period using a first set of predetermined parameters; detect it within the soft refractory period using a second (different) set of predetermined parameters. The event is preferably inspiration onset, and the monitored characteristic is slope and/or amplitude. The refractory period is defined based on detected inspiration offset — optionally using inspiration offset plus an average respiratory period and an average time of inspiration. Offset itself may be validated by comparing sampled amplitude to a validating offset threshold.

2. Sampling / multi-sample detection method. Sample the amplitude of the respiratory effort waveform; generate a sample signal representative of a characteristic of the waveform from each sample; monitor those sample signals; and detect a respiratory event as a function of at least two sample signals (i.e., requiring consecutive/temporally sustained evidence rather than a single sample).

3. Method of providing stimulation to treat respiratory disorders. Monitor slope and amplitude of the respiratory effort waveform. Define a refractory period with a hard portion (inspiration onset cannot initiate stimulation) and a following soft portion. Detect inspiration onset outside refractory as a function of the waveform slope; detect onset within soft refractory as a function of slope and amplitude. Deliver stimulation in response to a detected onset. Refractory is defined based on detected inspiration offset, where offset is detected from monitored slope and amplitude.

4. Stimulation method with criteria sets and termination logic (likely the broadest therapy claim). Monitor slope and amplitude to detect both inspiration onset and inspiration offset. Define hard + soft refractory periods. Detect onset outside refractory per a first set of slope/amplitude criteria; detect onset inside soft refractory per a second set of slope/amplitude criteria. Provide stimulation on detection. Stimulation terminates as a function of detected inspiration offset or a maximum stimulation time. Notably, the second (soft-refractory) criteria set is set sufficiently sensitive relative to the first such that, for one or more respiratory cycles, stimulation is applied as a function of the maximum stimulation time and the defined refractory period — this is the mechanism that carries therapy through central apnea when no reliable onset/offset is detectable.

5. Apparatus. "Monitoring means" for at least one respiratory-waveform characteristic, plus "respiration detection means" comprising: means for defining hard + soft refractory periods; means for detecting the respiratory event outside refractory per a first parameter set; and means for detecting the event within soft refractory per a second parameter set. Event may be inspiration onset; characteristic may be slope and/or amplitude; may further include means for detecting inspiration offset from monitored slope and amplitude.

6. Stimulation system. A sensor producing a signal characteristic of the respiratory effort waveform; slope monitoring means; amplitude monitoring means; respiration detection means (hard + soft refractory defining means; slope-based onset detection outside refractory; slope-and-amplitude-based onset detection within soft refractory); means for generating a stimulation signal in response to a detected onset; and at least one electrode delivering the stimulation signal to the patient.

Shared inventive core across the independents: the two-tier hard/soft refractory window plus the asymmetric detection criteria — slope alone outside refractory, slope and amplitude inside soft refractory — to filter cardiac and motion artifacts while still triggering predictively (target ≤ ~200 ms after true onset) and to sustain stimulation via a maximum-stimulation-time fallback during central apnea.

Supporting specification context (for grounding)

• The waveform is characterized by a negative peak at end-expiration, a positive peak at inspiration offset, and a turning point at inspiration onset (FIG. 2a); slope is derived as ΔV between successive samples every ~10–70 ms.
• Thresholds enumerated in FIG. 14: analog onset 520, ΔV onset 522, analog offset 524, ΔV offset 526, Vref/zero-crossing validation 530, AGC amplitude 528.
• Offset detection requires three consecutive ΔV samples below the ΔV offset threshold, with the first exceeding the analog offset threshold, then validation on falling below Vref; a ~475 ms offset-hysteresis blanking period permits only one offset per respiratory cycle.
• AGC normalizes the signal to ~2.4 V peak-to-peak across 64 exponential gain steps (~5.3% each), updated once per respiratory period or on watchdog timeout (~1.5 × the weighted-sum period).
• Suspension mode and an artifact counter (programmable max artifact count, e.g. 10/20/40/80) drive a therapy delay; practitioners also use sensor tapping as a patient-initiated therapy-termination mechanism.
• Sensor is an implantable dynamic relative pressure sensor (per U.S. Pat. Nos. 4,407,296 and 4,485,813) positioned at the posterior manubrium / suprasternal notch region for intrapleural pressure continuity.

Uncertainty statement

• Verified with high confidence: patent number, title, application number, filing/issue dates, priority date, inventors, assignee, classifications, family members, and the abstract text (corroborated across Google Patents, EPO Global Patent Index, and the Unified Patents portal).
• Not verified: the verbatim granted claim language and the exact number/identity of independent claims. My claim overviews are inferred from the specification's summary paragraphs; confirm against the USPTO PatentCenter "Claims" view or the granted patent PDF before relying on them for any legal or validity purpose.
• No CAFC 2026 activity found for US 5,944,680. I cannot exclude activity in non-indexed docket sources I could not query, though the 2016 expiration makes live appellate activity against this patent highly unlikely.

Sources: https://patents.google.com/patent/US5944680/en · https://data.epo.org/gpi/EP0907926B1 · https://portal.unifiedpatents.com/patents/patent/US-5944680-A · https://radaris.com/f/John/Hammargren/Inventor

I'll research litigation involving US 5,944,680 across patent litigation sources.

Let me search more targeted litigation sources.

Let me search litigation-specific databases directly.

Summary

I could not identify any litigation—district court, Federal Circuit, ITC, or PTAB—involving U.S. Patent No. 5,944,680 ("Respiratory Effort Detection Method and Apparatus," Christopherson et al., Medtronic, Inc.). On the record I was able to retrieve, there appears to be no known asserted litigation for this patent.

I want to be explicit that this is a negative finding based on limited-availability sources, not a certified clearance search, and I flag my confidence level below.

What I searched and what I found

Answer to your specific request

No cases to list. For completeness, the requested fields are empty:

Why the absence of litigation is plausible (context, not proof)

A few non-litigation facts from the patent record are consistent with the patent never having been asserted:

1. It was a Medtronic-owned internal portfolio patent. Google Patents and the Unified Patents page both list Medtronic, Inc. as original and current assignee. There is no recorded assignment to a litigation-holding entity (no NPE/SPV transfer), which is what you'd typically see if the patent had been monetized through suits.
2. It expired June 26, 2016 ("Anticipated expiration"; status "Expired – Lifetime"). Any enforcement window has closed, and inter partes review challenges are of little practical value post-expiration.
3. The reference pattern is prosecution, not litigation. The '680 patent shows up as prior art cited against other applications (e.g., it is cited as an "X" reference against claims 1–8 in WO2017001509A1), and family members (e.g., 5,483,969; 5,540,733; 5,549,655; 5,540,733) appear in the same sleep-apnea stimulation family. Citation as prior art is not evidence of litigation.

Confidence and caveats

• Confidence that there is no reported, indexable litigation: moderate-to-high for the public databases I could reach (Unified Patents, Google Patents legal events, Justia, general web).
• I cannot rule out: (a) suits that settled early and never generated indexed dockets; (b) cases in which '680 was asserted but the docket text is not full-text indexed by these aggregators; (c) confidential arbitrations, licenses, or settlement/consent arrangements (Medtronic's sleep-apnea IP has historically been licensed rather than litigated — a licensing trail would not appear in litigation databases); or (d) PTAB proceedings I could not surface.
• I did not have direct PACER or a commercial docket-analytics account (Lex Machina / Docket Navigator / RPX) in this session, and I was cut off before completing a second round of site-specific litigation searches. Those are the sources most likely to reveal a case the free tools missed.

Recommended verification steps before you rely on this

1. Unified Patents District Court Litigation Portal — search the patent number in the caselist search field, not just the patent page: https://portal.unifiedpatents.com/litigation/caselist
2. CourtListener / RECAP — search the patent number as a phrase in the RECAP docket database (not case law), and separately in the PTAB database.
3. USPTO Patent Trial and Appeal Board — search the patent number in PTAB E2E / the PTO's "Patent Docket" for IPR/PGR/CBM and any interference history (this patent's 1996 priority date makes interference history a real possibility worth checking).
4. USPTO Assignment Center — confirm there is no chain of assignment to a litigation entity: https://assignment.uspto.gov
5. PACER and a commercial docket tool (Lex Machina, Docket Navigator, RPX) for a definitive negative.

If you want, I can attempt the Unified Patents caselist and CourtListener RECAP searches specifically in a follow-up pass now that the broad search has been exhausted.

I'll verify the structured data with web searches before reporting.

Let me run additional targeted searches to confirm the absence of proceedings and check for any related appeals.

PTAB Proceedings Report — US Patent 5,944,680

Patent: US 5,944,680 ("Respiratory effort detection method and apparatus")
Patent Owner / Assignee: Medtronic, Inc. (originally Medtronic, Inc.)
Inventors: Christopherson, Neisz, Hammargren
Filed / Priority: 1996-06-26 · Granted: 1999-08-31 · Anticipated expiration: 2016-06-26
Legal status (Google Patents / as listed): Expired – Lifetime

Proceedings overview

There are no AIA trial proceedings on file for US 5,944,680. The structured USPTO Open Data Portal list is empty (0 active, 0 invalidated, 0 sustained, 0 settled, 0 institution-denied), and my independent web searches surfaced no IPR, PGR, or CBM petition ever naming this patent — no petition number, no institution decision, no Final Written Decision, no appeal. The bottom-line defensive posture is therefore not "hardened by surviving IPR" — it is "never challenged at the PTAB at all," which for a defendant today is a neutral-to-favorable signal that must be read alongside the single most important fact about this patent: it expired 2016-06-26. Because the patent term has run, the only live exposure would be past damages within the § 286 six-year lookback (roughly conduct from ~2010 onward through expiration), and no PTAB proceeding has narrowed or confirmed any claim to shape that exposure.

Caveat on the empty result: an empty PTAB record for a 1996-priority, 2016-expired patent is expected, not anomalous. Patents of this vintage were asserted (if at all) before the AIA trial regime matured, and the patent's term had run before the current wave of sleep-apnea-device IPR activity (e.g., the Axonics/Medtronic and Inspire-related filings) began. Silence here is a vintage/expiration signal, not evidence that the claims are strong.

Proceedings — verbatim and claim-level detail

None to report. No proceeding numbers exist to populate the requested per-proceeding template. I will not manufacture IPR/CBM/PGR numbers, panels, or decisions. To be explicit about what I checked and what I found:

• Structured USPTO ODP data (canonical): no AIA trial proceedings indexed.
• USPTO Patent Trial and Appeal Board (PTAB E2E / PTAB Decisions): no petition or decision for this patent surfaced in search.
• Federal Circuit / CourtListener: no appeal involving US 5,944,680 surfaced.
• Third-party aggregators (Unified Patents portal): the Unified Patents portal hosts a routine patent profile page for US-5944680-A (assignee, prior art, NPL). A profile page is not a proceeding. No Unified Patents (or any defensive aggregator) IPR/CBM against this patent was found. Do not misread the profile listing as litigation or PTAB activity.

⚠️ False-positive warning for search-based due diligence

A naive keyword search for "5944680" will surface JP 5944680 B2 — an unrelated Japanese patent to Kumagai Gumi Co., Ltd. titled "Method for forming floor foundation for passage and formwork used therefor" (construction formwork, registered 2016-06-03). This is not a family member and not the U.S. patent. Separately, searches return IPR2022-00988 through IPR2022-00993 (challenging other patents, e.g., an '994 sleep-disordered-breathing patent) and the Medtronic v. Axonics IPR series on Medtronic patents 8,036,756 / 8,626,314 / 9,463,324 / 9,821,112 / 8,738,148 / 8,457,758 / 7,774,069. None of these involve US 5,944,680. They establish that Medtronic both asserts sleep-neurostimulation patents and petitions as a defendant (cf. In re NuVasive, Medtronic's IPRs on spinal implants), but they are not proceedings on this patent.

Strategic summary

Claims status — all UNTESTED; none canceled, none confirmed. Every claim of US 5,944,680 stands exactly as issued, because no tribunal has ever construed, invalidated, or affirmed them. There is no PTAB record to point to saying claims 1–8 (or any subset) survived or fell — the patent's ~this-family's claim set (the family includes counterparts WO1997050049A1, EP0907926B1, JP3194968B2, CA2258759C, AU716513B2) has never been narrowed by an AIA trial in the U.S. For a defendant being asserted against, this cuts both ways: the patent owner cannot invoke any confirmed claim, but you also get no free "claims already canceled" defense. Your invalidity case is a clean slate — build it from scratch on the 1996-era prior art (the patent itself cites Glenn 1978; Hudgel 1992; Remmers; Broniatowski 1985; and U.S. Pat. Nos. 4,757,824; 4,250,884; 4,485,813; 5,174,287; 5,540,733; 5,215,082; 5,123,425; 5,178,156; 4,365,636; 5,190,053; 5,483,969 — a rich prosecution and reference set that a skilled searcher could mine).

Estoppel landscape — § 315(e)(2) is a non-issue here because no petitioner exists. There is no IPR/PGR petitioner and therefore no estoppel attaching to anyone for this patent. A defendant today faces no § 315(e)(2) bar, no prior-art ground that was "raised or reasonably could have been raised," and no frozen record. Practically: because the patent is expired, an IPR would be an unusual (though not impossible) vehicle — the Board can institute on an expired patent, but the primary value of an AIA trial (preempting injunctive/ongoing relief) is largely moot. The real vintage battleground is invalidity under §§ 102/103 and, importantly, § 112 / best-mode / enablement, plus § 286 damages limitation and § 287 marking, in district court or as an affirmative defense, not at the PTAB.

Pattern signals — no pattern, and that is the story. There is no repeat petitioner, no patent-owner PTAB-appeal history, and no defensive aggregator in the chain for this patent. Medtronic is an active, sophisticated PTAB participant on other patents (both as petitioner, e.g., NuVasive, and as patent owner against Axonics), and it retains Unified Patents profile coverage of its portfolio — but that activity has never touched US 5,944,680. Combined with the 2016 expiration, the defensible read is that this patent is legacy/quiet rather than actively asserted. If you have received a demand letter citing US 5,944,680, treat that as a signal worth stress-testing — an expired 1996 patent being asserted today is more often a past-damages or portfolio-threat posture than a live-injunction play.

Recommended next steps

1. Confirm the negative on the record. Pull the patent's full PTAB E2E and PatentCenter prosecution/trial histories directly to lock in the "no proceedings" conclusion:
   • USPTO PTAB E2E (Patent Trial and Appeal Board End-to-End): https://e2e.uspto.gov/
   • PTAB Decisions search: https://www.uspto.gov/patents/ptab/decisions
   • PatentCenter: https://patentcenter.uspto.gov/
   • Patent text: https://patents.google.com/patent/US5944680/en
2. Anchor on the expiration date (2016-06-26). Because the term has run, an assertion can only reach past damages within 35 U.S.C. § 286's six-year lookback (≈2010 onward to expiration). Immediately demand the patent owner's damages model, accused-product timelines, and any § 287 marking/licensing evidence. There is no prospective injunction available, and no § 271(e)/(f) ongoing-royalty posture beyond the statutory window.
3. Build the invalidity case without estoppel constraints. No § 315(e)(2) bar applies, so you may raise any §§ 102/103/112 ground. Start from the patent's own cited art and prosecution file (the references listed above), and consider best-mode and written-description/enablement attacks on the claim limitations (slope+amplitude onset detection, hard/soft refractory periods) — these are not PTAB staples but are powerful district-court defenses for a 1996 software/algorithm-heavy claim set.
4. If your opponent files an IPR against you (or threatens a parallel IPR), monitor these trial-stage mileposts (none are live today, but for planning): institution-decision deadline at 6 months from petition (35 U.S.C. § 314(b)); and Final Written Decision at 12 months from institution (§ 316(a)(11)), extendable up to 6 months for good cause. For an expired patent, weigh whether a stay is even worth seeking.
5. Do not conflate entities. When running your own searches, exclude JP 5944680 B2 (Kumagai Gumi, construction formwork) and the IPR2022-0098x / Medtronic v. Axonics matters — none involve this patent.

Bottom line for a defendant: There are zero PTAB proceedings on US 5,944,680 — no canceled claims to hand you a free win, but also no IPR estoppel and no PTAB-sanctioned claim construction to fight. Your decisive facts are (a) the patent expired 2016-06-26, capping exposure to a past-damages window, and (b) a clean-slate invalidity defense unencumbered by any prior PTAB record. The absence of any IPR on a 1996 sleep-apnea neurostimulation patent is itself informative: well-asserted patents of the modern era attract IPRs, and this one never did — consistent with a legacy patent surfacing in a demand letter rather than an active litigation campaign.

Nothing in this report should be read as a representation that no proceeding exists in some non-public or unindexed form; it reflects the USPTO ODP structured data provided and the public web sources available as of 2026-09-28. Verify against PTAB E2E before relying on it.

I'll verify the assignment record before writing anything.

Let me check for post-issuance transfers and any litigation/assertion activity.

Let me verify whether the Medtronic sleep-apnea family was later transferred (Inspire Medical Systems is a candidate) and check the EP/JP family status.

Ownership & Assignment Reconstruction — US 5,944,680

Scope note before the findings: I was able to pull the patent's full text and legal-events data plus third-party and SEC sources, but I could not load the live USPTO Assignment Center / assignment.uspto.gov record pages for this patent (search results did not return the reel/frame detail view). Several fields the brief asks for — reel/frame and correspondent of record — are therefore not verifiable from what I retrieved, and I have flagged them as such rather than filling them in. Anything below that is not directly sourced is labeled as unverified.

Inventors

The grant names all three as inventors, with Medtronic, Inc. (Minneapolis, MN) as assignee. Employer at filing is determinable for all three by the same-day assignment to Medtronic, Inc. (see below) and by the subject matter of the application, which is directed to Medtronic's implantable sleep-apnea stimulation program (ITREL II Model 7424 / ITREL III Model 7425 platforms are named in the specification as the base stimulator hardware).

Unusual pattern noted — partial, not portfolio-wide: first-named inventor Mark A. Christopherson re-surfaces as an inventor on Inspire Medical Systems patents (e.g., EP 2,331,201, "System for treating sleep apnea transvenously," filed 30 Sep 2009, applicant Inspire Medical Systems, Inc.). Inspire states it was formed in 2007 as a spin-out of Medtronic's technology portfolio and patents. This is inventor continuity with a corporate spin-out rather than a fire-sale signal — but it is worth recording because it ties the human capital of this family to the entity that ultimately commercialized the therapy. I found no evidence that Neisz or Hammargren departed Medtronic within 12 months of filing, and I did not confirm departure dates for any of the three. Do not read this as a fire-sale precursor.

Original assignee

• Entity on the issued patent: Medtronic, Inc., Minneapolis, MN (per the grant front page and the Google Patents record; "Current Assignee (listed assignees may be inaccurate): Medtronic Inc").
• Primary line of business: Implantable medical devices — cardiac rhythm management, neurostimulation, diabetes, and surgical technologies. Publicly traded (NYSE: MDT). Following the 2015 Covidien acquisition, Medtronic, Inc. sits under Medtronic plc; the US operating entity remains in Minneapolis. Status: operating (ongoing).
• Did they ship a product embodying the claims? No — not as a commercial product. Medtronic developed and clinically evaluated the inspiration-synchronized implantable sleep-apnea system (the "Inspire 1" feasibility study, 1998–2001, was run under Medtronic-origin IDE G950075 using the Model 3024 IPG, Model 4323 pressure sensing lead and Model 4063 stimulation lead), and the specification describes that hardware in detail. But the commercial product was brought to market by Inspire Medical Systems, Inc., which describes itself as founded in 2007 when "the Inspire technology, as well as the technology portfolio and patents of Medtronic (NYSE: MDT) were spun off," and which obtained FDA PMA approval for the Inspire Upper Airway Stimulation system in 2014. So: Medtronic built and tested it; Inspire shipped it.
• Status of original assignee's rights in this family: unclear. If a US assignment had been recorded to Inspire, Google Patents' "Current Assignee" and legal-events list would normally reflect it. They do not. See Qualifications below.

Assignment timeline

Important qualification up front: the USPTO Assignment Center / Google Patents legal-events record for this patent shows only one recorded assignment — the original inventor-to-Medtronic assignment. No post-issuance assignment (to Inspire Medical Systems or to any other entity) is recorded in the authoritative source I retrieved. I found no NPE-typical chain for this patent.

1. 1996-06-26 (executed) / recorded 1996-06-26 — Reel/frame not exposed in the sources retrieved

• Conveyance: Assignment
• Assignor: John Hammargren; Johann J. Neisz; Mark A. Christopherson (individually)
• Assignee: Medtronic, Inc. (Minneapolis, MN)
• Correspondent: not captured — the Google Patents legal-events entry does not expose the correspondent of record, and the Assignment Center record page itself did not load in my session. Recorded as "ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)." Please verify the correspondent directly at the Assignment Center URL in the Verdict section.
• Context: Standard pre-issuance employment assignment of a corporate R&D invention — filed the same day as the application (App. No. 08/674,494), years before any spin-out.

2. 2007 — no USPTO recording surfaced

• Conveyance: Assignment and License Agreement (per Inspire Medical Systems' own description of its 2007 founding transaction with Medtronic)
• Assignor: Medtronic, Inc.
• Assignee: Inspire Medical Systems, Inc. (or a related acquisition vehicle)
• Correspondent: not surfaced
• Context: Corporate carve-out / spin-out of Medtronic's sleep-apnea technology portfolio and patents to a then-startup.
• Evidentiary basis and limits: Inspire's SEC filings state "In 2007, we entered into an assignment and license agreement with Medtronic ('the Assignment and License Agreement')." Inspire's S-1 Exhibit 10.1 contains a patent schedule that includes EP0907926B1 (the European counterpart of US 5,944,680 — same priority, same family), JP3194968B2, WO9750049A1, and US 6,021,352 (a sibling US patent from the same specification/diagnostic-self-test disclosure). The foreign members of this exact family therefore appear on Inspire's schedule. I did not confirm US 5,944,680 itself on that schedule, and no USPTO assignment to Inspire for the US patent was recorded. Treat the US patent's post-2007 ownership as unresolved: consistent with either (a) a non-recorded assignment or an assignment-plus-license structure that left record title with Medtronic, or (b) a license-only structure for the US member with only foreign members assigned outright.

3. 2016-06-26 — Anticipated expiration

• Patent term ended (per Google Patents legal status: "Expired – Lifetime"; anticipated expiration 2016-06-26). No maintenance-fee-lapse or adverse event was surfaced.

Timeline diagram

timeline
    title Ownership of US 5944680
    1996 : Filed by Medtronic Inc
         : Inventors assign rights to Medtronic
    1999 : Patent issued
    2007 : Sleep apnea IP spun out to Inspire
    2014 : Inspire UAS system wins FDA approval
    2016 : Patent expires

NPE / troll-pattern signals

1. Shell-entity transfer — not present. No assignment to any entity with an "IP / Patents / Licensing / Holdings / Ventures" suffix was recorded. The only recorded assignee is Medtronic, Inc., a large operating manufacturer. No registered-agent-service address appears anywhere in the chain I retrieved.

2. Known asserter in the chain — not present. Neither recorded assignee (Medtronic, Inc.) nor the uncovered spin-out counterparty (Inspire Medical Systems, Inc.) appears on the Acacia / Marathon / IV / IPNav / Wi-LAN / Mosaid / Conversant / Vringo / Pendrell / Round Rock / Spangenberg-type lists, and neither is surfaced as a high-frequency plaintiff in the Unified Patents portal entry for this patent, which lists the assignee simply as "Medtronic Inc." Inspire Medical Systems is itself a frequent defendant-side participant in IPR practice, i.e., an operating company, not an asserter.

3. Repeat correspondent across the chain — not present (and unverifiable). Only one assignment is recorded, so there is no chain across which a correspondent could recur. The correspondent of record for the 1996-06-26 assignment was not exposed in any source I retrieved. This is a genuine gap in this analysis; if a later recording exists that my sources missed, the correspondent on it is the single most valuable field to pull.

4. Cascading transfers — not present. Zero post-issuance transfers are recorded. There is no chained-LLC pattern, no <24-month succession, and no shared correspondent address to compare.

5. Pre-litigation transfer — not present. No infringement suit naming US 5,944,680 was surfaced by any search. The patent is cited extensively as prior art in later sleep-apnea filings (Medtronic, Inspire, and third-party applications), but prior-art citation is not assertion. Absent a suit, there is no pre-litigation transfer to date.

6. Bankruptcy fire-sale — not present. Medtronic has never filed Chapter 7 or 11 in a manner affecting this portfolio, and no §363 sale or reorganized-debtor transfer involving this patent was surfaced.

7. Privateering — not present (with one nuance worth stating precisely). The 2007 Medtronic→Inspire transaction is a spin-out to an operating company, not privateering: Inspire designs, manufactures, and markets its own device (FDA PMA, 2014) and does not generate revenue from licensing this family against competitors. Medtronic's relationship to Inspire is that of an equity holder/strategic partner in a spin-out, not that of a manufacturer using a shell to assert against rivals. If anything, the direction of the 2007 deal is the reverse of privateering — the operating company shed the program.

8. Defensive aggregator (anti-NPE) — not present. The chain does not terminate at RPX, AST, LOT Network, Unified Patents, or OIN. No defensive aggregation or neutralizing acquisition is recorded.

Verdict

Insufficient data — the USPTO record for this patent contains only the original inventor-to-Medtronic assignment executed and recorded 1996-06-26, with no reel/frame or correspondent data retrievable in my sources; and the patent expired 2016-06-26, so there is no live assertion to characterize. Nothing in the record resembles an NPE chain: the sole recorded assignee is Medtronic, Inc., a large operating manufacturer, and the only uncovered downstream event is the 2007 Medtronic spin-out of its sleep-apnea portfolio to Inspire Medical Systems, Inc. (Inspire's SEC filings; EP0907926B1, the EP counterpart of this patent, appears on Inspire's S-1 Exhibit 10.1 patent schedule) — an operating-company carve-out, not a shell-entity transfer. Caveat on the verdict: if a recorded assignment to Inspire Medical Systems or a related vehicle exists for the US patent and my session simply failed to retrieve it, the verdict would shift to "operating-company transfer," not to NPE — so I do not expect the gap to change the disposition.

Verify / complete the record here:

• USPTO Assignment Center search by patent number: https://assignment.uspto.gov/patent/index.html#/patent/search?q=[5944680](/patent/5944680) (and https://assignmentcenter.uspto.gov/)
• Google Patents legal events for this patent (source of the single recorded assignment above): https://patents.google.com/patent/US5944680/en#legal-events
• Unified Patents portal entry (assignee of record shown as Medtronic Inc): https://portal.unifiedpatents.com/patents/patent/US-5944680-A
• Inspire Medical Systems S-1 Exhibit 10.1 (Assignment and License Agreement schedule listing family member EP0907926B1): https://getfilings.com/sec-filings/[180406](/patent/180406)/Inspire-Medical-Systems-Inc_S-1/a2235179zex-10_1.htm

Fields I could not verify and did not invent: the 1996 assignment's reel/frame number and correspondent of record; whether US 5,944,680 (as opposed to its EP/JP/PCT siblings) was included in the 2007 Medtronic→Inspire schedule; and current record title in the US patent.

I'll research US 5,944,680 on the USPTO and related databases to identify its cited prior art references.

Let me get the specific citations list from the USPTO/Google Patents "Patent Citations" section for 5,944,680.

Let me retrieve the full "Patent Citations" and "Cited By" tables from the Google Patents record.

Prior Art Analysis — U.S. Patent No. 5,944,680

Context: This builds on the earlier litigation summary, which established that 5,944,680 is Medtronic's Respiratory Effort Detection Method and Apparatus (Christopherson, Neisz, Hammargren), App. No. 08/674,494, filed 1996-06-26, granted 1999-08-31, expired 2016-06-26. That finding is consistent with what follows. I do not repeat it here.

1. Patent identified (literal match — no auto-correction)

I treated 5944680 strictly: the near-number collisions flagged in the prior section (JP 5944680 B2; Steuben Foods US 6,481,468; even Boehringer's US 5,844,680 color-measurement patent, which appears in the same citation neighborhoods) were excluded.

2. U.S. patent references (the "Patent Citations" / References-Cited table)

Retrieved from the Google Patents "Patent Citations" table for US5944680A
(https://patents.google.com/patent/US5944680 — citations block). These are the references cited against/considered for this patent during prosecution.

Foreign patent document cited:

3. Patents referenced in the specification text of '680

These are expressly named in the '680 description (and are the references one would expect in the References-Cited section):

Note the strategic point: items #2–#4 of the citation table and USSN 08/310,177 share the same 1994-09-21 filing date family — i.e., '680 was prosecuted against its own corporate siblings.

4. Which claims each reference potentially anticipates

⚠️ Important limitation: the full claim set of '680 was not fully retrievable in this session (the authoritative text I was given truncates before the claims). The mapping below is therefore by claim category, inferred from the '680 abstract/summary language, not a verified line-by-line construction of literal claim language. Treat it as an analytical lead, not a conclusion.

Claim categories in '680 (per its own summary):

• (A) "Predicting critical points" method — refractory period with hard + soft phases; detect the event (inspiration onset) outside vs. inside soft refractory using a first vs. a second parameter set (slope; slope+amplitude).
• (B) Stimulation method — monitor slope & amplitude; refractory; stimulate on onset; terminate on offset or max-stim time; second criterion set more sensitive than the first.
• (C) Sampling method — sample amplitude; generate sample signals; detect a respiratory event as a function of at least two sample signals.
• (D) Apparatus — monitoring means + respiration-detection means (with refractory-defining means).
• (E) System — sensor + slope-monitoring means + amplitude-monitoring means + detection means + stimulation-generating means + electrode.

Bottom line on anticipation: On the record retrievable, none of the cited references appears to disclose the hard-refractory/soft-refractory two-tier onset-detection scheme or the "second parameter set applied only within soft refractory" limitation. That is the apparent inventive core. Every reference above is therefore better characterized as § 103 (obviousness) art or as background, not as clear § 102 anticipatory art — which is consistent with the examiner allowing the claims over these references.

5. Confidence, caveats, and what I could not verify

1. Source coverage. I worked from the Google Patents "Patent Citations" table (mirrors USPTO citation data) plus EPO GPI family data. I did not directly complete a query of USPTO Patent Public Search (PatFT/PPUBS) or Patent Center in this session, so I cannot certify that the citation list above is exhaustive. The retrievable citation rows for US 5,215,082 and US 5,540,733 showed the number, dates, and assignee but not the title; I have flagged those rather than guess.
2. Claim text. Because the '680 claims were not fully available, the § 102 claim mapping is by category, not literal claim number. Do not cite it as a claim-by-claim invalidity position without pulling the actual claim set.
3. Strict-number rule applied. 5944680 was matched literally; the near-number patents surfaced during searching (JP 5944680 B2; US 5,844,680; US 6,481,468) were excluded and are not reported here as references to '680.
4. § 102 regime. '680 has a 1996 filing date, so pre-AIA 35 U.S.C. § 102 governs. The distinction matters here: US 5,540,733 and US 5,549,655 issued after the '680 filing and are available only under § 102(e), while EP 0 765 631 A2 is not available at all as § 102 art.
5. Recommendation. For a defensible prior-art table, pull the "References Cited" section directly from the USPTO Patent Center / Patent Public Search record for 5,944,680 (and the PCT/EP equivalents WO 97/50049, EP 0907926 B1) to capture any cited rows not surfaced here, and obtain the granted claim set for a literal § 102 mapping.

Most relevant prior art, ranked: (1) US 5,549,655 (Erickson) and (2) US 5,540,733 and US 5,483,969 (Testerman) — the same-family, same-purpose references; then (3) US 5,215,082 for monitoring architecture. None of these, on the retrievable record, appears to anticipate the charactering hard/soft-refractory detection claims.

I'll gather the prior art cited on the US 5,944,680 Google Patents page to ground the obviousness analysis.

Let me pull the cited prior art list and claim text from secondary sources.

Let me try to retrieve the actual granted claim text to ground the obviousness mapping.

Obviousness Analysis — US 5,944,680 under 35 U.S.C. § 103

0. Scope, sourcing, and caveats (read first)

What I could and could not ground.

• I built this analysis on the prior-art references that appear on the Google Patents "Prior Art" record for US 5,944,680 — i.e., the references cited within the '680 specification and the references cited on the face of the granted patent / its international search report and patent family. Source of truth for the disclosure: https://patents.google.com/patent/US5944680/en (fetched 2026-09-26), plus the cited references themselves.
• I was unable to retrieve the verbatim granted claim text (the fetched document contains the specification and the summary-of-invention paragraphs, but not the numbered claims). The claim scope used below is therefore reconstructed from the specification's summary paragraphs, which in this patent track the independent claims closely, and is consistent with the previously generated "Independent claims — plain-language overview." Treat claim numbering and the independent/dependent split as indicative, not verified. Any real §103 opinion must be re-run against the PatentCenter "Claims" view or the granted PDF.
• Per your strict rule, I have not auto-corrected any identifier. Note the standing collision flagged earlier: JP 5944680 B2 (Kumagai Gumi, concrete formwork) is a different patent and is excluded here. Likewise US 5,944,680 must not be confused with the Medtronic sibling filings made the same day (1996-06-26), e.g., US 5,895,360; US 6,021,352; US 6,099,479; US 6,132,384; US 6,572,543.
• A material legal caveat appears in §VI below: the most on-point prior art is commonly owned by Medtronic, which raises a pre-AIA §103(c) disqualification issue for art that qualifies only under §102(e). I flag it explicitly because it can defeat grounds that look strong on the merits.

1. The claim scope being tested

From the summary paragraphs, the independent claims break into five substantive clusters:

Structural observation that drives the whole analysis: every independent claim is a combination of individually known building blocks — (i) respiratory-effort waveform monitoring, (ii) slope-based onset detection, (iii) amplitude-based corroboration, (iv) a refractory/blanking window, (v) averaging of respiratory period, (vi) synchronous stimulation — assembled in a particular hierarchy. That is exactly the fact pattern the Supreme Court addressed in KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), i.e., predictable combination of known elements.

2. The prior art of record

3. Obviousness grounds

Ground 1 — Claims 1–2 (Cluster A / E): Testerman '655 in view of Testerman '733 and further in view of '862/'969

Element mapping (Cluster A):

Motivation to combine. Both '655 and '733 are Medtronic sleep-apnea stimulation patents in the same family, filed the same day (1994-09-21), sharing inventors. A POSITA seeking to (a) trigger stimulation predictively at inspiratory onset and (b) avoid false triggers would combine '655's slope/amplitude turn-point detector with '733's baseline-average/limit logic, because both references are directed to the same problem (synchronizing upper-airway stimulation to inspiration) and their combination yields only the predictable aggregation of the two detection criteria. KSR, 550 U.S. at 416–17 (predictable use of prior-art elements according to their established functions). The strong suggestion comes from the references themselves: '655 already teaches using slope AND amplitude together to declare a turn point — the '680's differencing of "outside-refractory" vs. "inside-soft-refractory" criteria is a mere rearrangement of '655's two parameters into two tiers.

Ground 2 — Claims 1, 3–5 (Clusters A, C, E): Any of the Testerman references in view of the cardiac-pacing refractory paradigm (also evidenced by Shannon '425 and EP '977)

The '680's point of novelty is the hard refractory / soft refractory split with asymmetric sensing criteria. This concept is not only suggested — it is inherently available from the well-known refractory-period model of implantable-cardiac-device art, which the '680 inventors themselves concede ("a refractory period, i.e., a blanking period after inspiration has occurred," and "The refractory period (R) … can be approximated as described below"). Evidence:

• EP 0 702 977 (Medtronic) discloses a timing circuit that "notifies … the time window in which patient 10 is within an inspiration cycle" — a time window that gates when sensing/stimulation may occur, the functional equivalent of a refractory/blanking gate.
• US 5,123,425 (Shannon) teaches gating stimulation on both "electrical activity of the diaphragm suggests that an inspiration cycle is in progress" and "the pressure sensors show an abnormal pressure differential" — i.e., a two-condition gate analogous to the soft-refractory "slope AND amplitude" requirement.
• Cardiac pacemakers universally use refractory periods subdivided into absolute/hard and relative/soft portions with different sensing thresholds — a POSITA in implantable neurostimulator design is presumed to know this art.

Motivation. The specification's own stated problem is artifact rejection: cardiac artifacts "produce slope changes very similar to onset and offset slope changes," and motion artifacts (FIG. 15, element 542) can satisfy an onset criterion. A POSITA would reach for the pacemaker refractory model — hard refractory to absolutely blank the physiologically impossible window, and a soft refractory with a more demanding threshold (slope AND amplitude) to permit early-but-real onsets — because that is the art's standard tool for exactly this problem. This is a classic obvious-to-try situation under KSR, 550 U.S. at 421.

Ground 3 — Claim (Cluster B): Testerman '655 (claims 7, 10) in view of AAPA

Motivation. '655's explicit teaching of digitized sampling and successive-sample averaging supplies the "at least two sample signals" limitation directly. The '680 spec itself explains why one would require ≥2 samples: "short duration (higher frequency) noise or cardiac artifacts can be rejected," with the acknowledged "tradeoff … that delay is added." A POSITA would adopt two-or-more-sample corroboration precisely because single-sample triggering is artifact-prone — a predictable use of a known technique to improve a known device. This claim is arguably anticipated by '655 claim 7 read with the '680's own admitted background; at minimum it is obvious.

Ground 4 — Clusters C and D: Testerman '655 in view of '733, in further view of '862 and/or Glenn (1978)

Cluster C/D add: (i) stimulation provided on detected onset; (ii) refractory defined from detected offset; (iii) slope for out-of-refractory detection vs. slope+amplitude in soft refractory; (iv) termination on offset or maximum stimulation time; (v) the second criteria set "sufficiently sensitive" so stimulation runs on max-stim-time + refractory for one or more cycles (the central-apnea carry-through).

• (i), (iii) are met by '655 claims 1 and 13–14 (commencing stimulation on the detected turn point) combined with '733 (stimulate on waveform-derived event).
• (ii) Offset-based timing: '862 and '655 both characterize the waveform by its end-inspiration positive peak and end-expiration negative peak — the very definition of a detectable offset. Deriving the refractory start from the detected offset is the natural, and in cardiac/respiratory pacing the conventional, choice.
• (iv) "or maximum stimulation time": This is the standard failsafe timeout in any stimulation controller. EP '977's timed inspiration window and the cardiac-pacing precedent supply it. Critically, the rationale is spelled out by the '680 itself: during central apnea, "offset … is not detected" and "maximum stimulation time is used to terminate stimulation."
• (v) The "sufficiently sensitive second set" is simply the comparative-tuning instruction required to make (iv) operative — the specification states the object plainly: stimulation is applied "as a function of the maximum stimulation time and the defined refractory period … until the patient's periodic respiration returns." Framing an obvious timeout as "sensitive enough to run the timeout" is a result-oriented characterization of a known mode, not a patentable difference. In re Kao, 639 F.3d 1057 (Fed. Cir. 2011) (a result cannot be claimed by wording that merely describes the result).

Motivation to combine. Glenn (1978) and '862 establish respiratory-cycle-synchronized stimulation as the field's baseline; '655 supplies the detector; '733 supplies amplitude/averaging; the timeout is ubiquitous. Combining them to handle central apnea (a recognized failure mode in the same clinical population, and detected per '969) is an obvious design choice driven by patient-safety and device-comfort considerations — the very interests KSR recognizes as supplying the "motivation."

Ground 5 — Cluster E (system/apparatus): Shannon '425 in view of Testerman '655 and Anderson '296 / '813 and Testerman '438

Motivation. All elements are known, each for its known purpose, all in the same field of treating obstructive sleep apnea by nerve stimulation. KSR, 550 U.S. at 416. The combination — standard "means-plus-function" claim drafting over an assemblage of prior elements — is a classic obviousness scenario; the apparatus claims rise or fall with the method claims because the "means" recite only the same functional blocks.

4. The "motivation to combine" themes (consolidated)

1. Same field, same problem, same assignee/inventors. '655, '733, '862, '969 (and the sensor/'438 references) are Medtronic sleep-apnea-stimulation references, several filed the same day (1994-09-21). References sharing a common origin and a common objective supply a strong TSM-type rationale. In re Kahn, 441 F.3d 977 (Fed. Cir. 2006).
2. The '680's stated problem was already recognized. Artifact rejection (cardiac/motion) and predictive onset timing (≤ ~200 ms) are the '680's own framing of the problem; '655's slope AND amplitude two-parameter criterion and the pacing-art refractory window are the known tools for it.
3. Predictable aggregation, not synergy. Combining slope detection ('655) + amplitude corroboration ('655/'733) + a refractory gate (pacing art / EP '977) yields only the expected sum of their functions — KSR at 416–17; no unexpected result is asserted in the specification beyond the intended improvement.
4. Design incentives / known tradeoffs. The '680 spec admits the multi-sample tradeoff ("delay is added") and the TI/T ratio "generally known (0.30–0.40)" — evidence that the parameter choices were within the ordinary skill level, i.e., routine optimization.
5. Failsafe/timeout is ubiquitous. The "maximum stimulation time" and the central-apnea carry-through are safety-motivated design choices recognized in the art (Glenn; cardiac-device practice).

5. Why the claims are not saved by their asserted points of novelty

6. Anticipated rebuttals, and the two real weaknesses in the obviousness case

(a) §103(c) common-ownership disqualification — the single biggest threat to Grounds 1–4.
The on-point Testerman references ('655, '733, '862, '969) and the sensor references ('296, '813) and '438 are Medtronic-owned, as is the '680. Under pre-AIA 35 U.S.C. §103(c), subject matter that qualifies as prior art only under §102(e), (f), or (g) cannot be used to establish obviousness where the subject matter and the claimed invention were commonly owned (or subject to an obligation of assignment to the same person) at the time the invention was made.

• '655 (issued 1996-08-27) and '733 (issued 1996-07-30) issue after the '680's 1996-06-26 filing date, so they are §102(e) art (prior art as of their 1994-09-21 filing dates) — potentially disqualified by §103(c).
• '862 (issued 1996-06-04) and '969 (issued 1996-01-16) issue before the '680 filing and may independently qualify under §102(a)/(b), which §103(c) does not reach — these are therefore the safer grounds.
• Practical implication: a challenger should anchor on the non-commonly-owned or §102(a)/(b) art — EP 0 702 977 (Testerman/Medtronic, but a published foreign application → §102(a)/(b), not §102(e)); WO 92/19318 (Cyberonics); US 5,097,830 (Eikefjord); EP 0 671 687 (Spacelabs); Shannon '425; and the AAPA — and use the commonly-owned Testerman patents only where they independently qualify. This is a genuine limitation of the grounds above, not a formality.

(b) Teaching away / non-analogous-art arguments.
The patent owner could argue (i) the cardiac-pacing refractory model is non-analogous art, and (ii) '733's averaging is directed to apnea detection, not onset detection, so it teaches away from continuous onset-synchronized triggering. Both are weak: the fields are "reasonably pertinent to the particular problem" (sleep apnea stimulation is an implantable-stimulation problem, In re Bigio, 381 F.3d 1320), and '733 and '655 share inventors and a common disclosure.

(c) Secondary considerations (must be developed with evidence).
No unexpected-results, commercial-success, or industry-praise evidence has been identified in the materials I retrieved. Any such argument would require a nexus between the evidence and the claimed combination (not merely the commercial product), In re GPAC Inc., 57 F.3d 1573 (Fed. Cir. 1995). None is currently on the record.

7. Bottom line

On the record I can retrieve, US 5,944,680 reads as an obvious assembly of known respiratory-stimulation building blocks, most of them produced by the same assignee in the same 1994–1996 window:

1. Primary combination: Testerman '655 + Testerman '733 (with '862/'969), optionally further in view of the cardiac-pacing hard/soft refractory model and EP '977's inspiration time window, renders Clusters A, C, D, and E obvious.
2. Cluster B (≥ two sample signals) is obvious over '655 claims 7/10 + AAPA, and is arguably anticipated.
3. The "predictive ≤200 ms," artifact-rejection, and central-apnea carry-through benefits are stated objectives already present in the art, not unexpected results.

But two things prevent a clean opinion:

• I do not have the verbatim granted claims, so claim-specific mappings (especially any unstated dependent limitations) are unverified; and
• pre-AIA §103(c) common ownership can knock out the strongest references ('655, '733) because they issue after the '680 filing and may qualify only under §102(e). A defensible invalidity position must therefore lead with the §102(a)/(b) art ('862, '969, EP '977, Shannon '425, WO 92/19318) and the applicant's own admissions.

Recommended verification before relying on this: (1) pull the verbatim claims from USPTO PatentCenter / the granted PDF; (2) confirm the §102 category and common-ownership status of each Testerman reference as of the '680 invention date; (3) confirm the front-page "References Cited" list on the granted patent (which I could not fully retrieve) so no cited reference is overlooked.

Key sources: https://patents.google.com/patent/US5944680/en · https://www.freepatentsonline.com/5549655.html · http://www.everypatent.com/comp/pat5540733.html · https://uspto.report/patent/grant/5522862 · https://patentimages.storage.googleapis.com/38/62/68/a9ebeed18451d9/EP0702977A3.pdf · https://patentimages.storage.googleapis.com/37/b2/84/379e44289aea6e/US5549655.pdf