US 5447489

I'll start by searching for authoritative records on this specific patent number.

Let me check for litigation, CAFC docket activity, and the USPTO record specifically.

Let me check for any litigation history or USPTO assignment records specific to this patent.

US Patent 5,447,489 — Analyst Summary

Search caveat up front: My tool searches returned no CAFC 2026 docket activity and no litigation involving US 5,447,489. A generic search for the bare string also surfaced one irrelevant hit (a Dutch soil-analysis certificate that uses "5447489" as a sample ID), which I discarded as a false positive. I could not directly query USPTO Patent Center/PAIR or the CAFC docket system, so I relied on the Google Patents record (which mirrors USPTO bibliographic data) and the Unified Patents portal. I cannot confirm that no 2026 CAFC proceeding exists — only that none appeared in my searches. Note also that the patent is recorded as Expired – Lifetime with an anticipated expiration of 2012-09-05, so 2026 enforcement activity would be highly unusual.

Bibliographic Data

Prosecution chain: The application is a continuation of Ser. No. 07/834,238, filed 1992-02-13, which was the US national-stage entry of PCT/FR90/00614, filed 1990-08-17. The foreign priority root is French application 89/10987 (1989-08-17).

Abstract (as issued)

A hearing aid has a transmitter element and a receiver-transducer element with a vibrating element. The receiver element is suitable for being placed, preferably movably, in the mouth of the user. The receiver element includes a device for supporting and holding the vibrating element, formed so that when the vibrating element is in place it is in permanent contact either with at least one tooth or with the palate bone, thereby providing sound transmission to the inner ear by bone conduction.

Independent Claim 1 — Plain Language

A hearing aid made of two cooperating pieces:

1. An extra-buccal (outside the mouth) wireless transmitter;
2. An intra-buccal (inside the mouth) wireless receiver-transducer that receives signals from the transmitter and contains at least one vibrating element;
3. The receiver is designed to be removably fixed in the user's mouth;
4. A support/holder positions the vibrating element so that, in use, it makes permanent contact with either a tooth or the palatine bone (or both);
5. Sound received as signals from the transmitter reaches the inner ear exclusively by bone conduction; and
6. The vibrating element is sealed and electrically insulated from the tooth/palatine bone.

Distinguishing features of claim 1 relative to claim 8: the express "wireless" link and the "exclusively by bone conduction" limitation.

Independent Claim 8 — Plain Language

A bone-conduction hearing aid of the same general type (extra-buccal transmitter + intra-buccal receiver-transducer with at least one vibrating element) in which:

1. The receiver is removably fixed in the mouth;
2. A support/holder ensures permanent contact between the vibrating element and a tooth and/or the palatine bone, transmitting sound to the inner ear by bone conduction;
3. The vibrating element is sealed and electrically insulated; and
4. Structurally, the vibrating element is a thin metal plate coated on one of its faces with a piezoelectric ceramic, the coated plate being enclosed in an envelope of biocompatible polymer film.

Claim 8 is the structural/"vibrator-construction" independent claim; claim 1 is the system/"wireless + exclusive bone conduction" independent claim.

Dependent Claims — Brief Orientation

• Claims 2–4 (from claim 1): tooth-contact mounting — hook elements with claws fastening to the tooth and two adjacent teeth, slide-guide elements (U-shaped cross-section), interconnected by a rigid bar against the vestibular face of the tooth.
• Claim 5 (from claim 1): palatine-bone contact via an acrylic-resin palatine plate with a recess holding the vibrator against the palatine bone.
• Claims 6–7 (from claim 1): miniaturized receiver/amplifier/inductively rechargeable battery unit connected by flexible connectors, in a polymer-resin envelope shaped to the vestibular space.
• Claim 9 (from claim 8): truncated contact block in the center of the vibrator, same material as the envelope, contacting the tooth.
• Claims 10–14 (from claim 8): specific tooth-mounting structures (claws, slide guide, two spaced hook elements with a rigid connecting element, U-shaped guides with base connections).
• Claim 15 (from claim 8): palatine plate of a dental prosthesis with a recess.
• Claims 16–17 (from claim 8): rechargeable battery with inductive-coupling recharging; miniaturized receiver/amplifier/battery in a polymer envelope shaped for the vestibular space.

Practical/Significance Notes

• The patent is a foundational reference in the intra-oral bone-conduction hearing aid field. The Google Patents record shows it being cited by a very large body of later work (60 listings in the summary table; 187 in the expanded "Cited By" set), including the extensive Sonitus Medical / SoundMed family (e.g., US 7,664,277; US 7,724,911; US 8,433,083 — "Dental bone conduction hearing appliance"), Mayur Technologies' tooth-conduction microphone patents, and Audiodontics work. This is consistent with it being prior art of record in later intra-oral hearing-device prosecutions.
• Its own cited prior art (7 references) is largely dental/oral-appliance art plus two Xomed implanted bone-conduction patents (US 4,606,329; US 4,612,915) — i.e., the examiner treated intra-oral placement as the novel departure from the surgical/implanted bone-conduction devices.
• The patent's core commercial/technical idea — a removable, non-surgical, intra-oral vibrator coupled to teeth or the palate (optionally integrated into a dental prosthesis, per the specification) — is articulated in the specification as a bi-functional "dental prosthesis and auditory prosthesis."

Confidence: Bibliographic data, claim text, and status are drawn from the Google Patents record and are high confidence. The absence of any CAFC 2026 docket item and the precise assignee-of-record at grant are the two points where I lack authoritative confirmation and flag accordingly.

I'll search litigation databases for US patent 5447489.

Let me search litigation-specific sources more directly.

Litigation Search: US Patent 5,447,489

Bottom line

No litigation involving U.S. Patent No. 5,447,489 is known.

I searched litigation-focused sources (Unified Patents Portal — both the patent page and litigation/PTAB caselists — Google Patents, and general web for court records) using the exact number "5447489" and "5,447,489." Every hit returned was a patent-citation / prior-art reference, not an infringement action, declaratory judgment, or PTAB proceeding. I found no plaintiff, defendant, jurisdiction, case number, or filing date to report.

What the searches actually returned (and why it is not litigation)

Notably, the many Sonitus Medical / Soundmed / Audiodontics patents that cite 5,447,489 (e.g., US 7,664,277; US 8,433,083; US 7,945,068) show the patent is well-known prior art in the intra-oral bone-conduction space — which is precisely why an infringement suit would be unlikely: the patent expired and its family is old prior art rather than an assertion asset. This is inference on my part, not a documented fact.

Status of the patent itself (context, not litigation)

• Patent: US 5,447,489 A — "Bone conduction hearing aid device"
• Inventors: Robert Issalene, Jean-Francois Lantrua, Bernard Saoli
• Assignee: Individual (no corporate owner of record)
• Priority date: 1989-08-17 (FR 8910987); filed: 1994-03-24 (continuation of Ser. No. 07/834,238, itself PCT/FR1990/00614); granted: 1995-09-05
• Legal status: Expired – Lifetime; anticipated expiration 2012-09-05
• Legal events of record: patent grant (1995), maintenance fees paid (1999, 2003 with surcharge, 2007) — no litigation-related events

Important caveats

1. Not an exhaustive negative. I could not directly query PACER, Docket Navigator, or the full Unified Patents litigation caselist interface in this session; my conclusion rests on the searches above. A paywalled docket database (Docket Navigator, Lex Machina) or a direct PACER party-name search for "Issalene," "Lantrua," or "Saoli" would be the definitive check.
2. Number-specific. I did not sweep similar numbers (e.g., 5,447,489 vs. 5,447,488 or 9,454,748). The "9,454,748" results that appeared belong to Fall Line Patents / Unified Patents — an entirely different patent and irrelevant here.
3. No fabricated case data. Because no case was found, I have deliberately not supplied a plaintiff/defendant/case-number table; doing so would misrepresent the record.

If you need a definitive certification, the most reliable next step is a PACER docket search by party name ("Issalene," "Lantrua," "Saoli") and an assignee-history pull from the USPTO Assignment database, since the patent was individually held and may have been asserted by a successor or exclusive licensee under a name not tied to the patent number.

I'll verify whether any AIA trial proceedings exist for this patent before writing.

Let me run a few more targeted checks to be thorough.

Proceedings overview

Total AIA trial proceedings on file for U.S. Patent No. 5,447,489: zero — no IPRs, no PGRs, no CBMs, no derivation proceedings, and therefore zero proceedings in any status bucket (0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denied). The bottom line for a defendant is not "the patent is hardened by surviving IPRs" and not "the claims are canceled" — it is that the patent left the PTAB's reachable universe before the AIA trial regime applied to it, and today there is nothing left to challenge and nothing left to enforce: its statutory term ran out on 1995-09-05 + 17 years = 2012-09-05, eleven days before IPR practice became available on 2012-09-16. Any PTAB-based defense is moot, but so is the infringement threat, because an expired patent cannot be infringed and carries no forward damages exposure (pre-expiration damages would require conduct that is now ~14+ years stale).

No proceeding entries to report

I am deliberately not creating per-proceeding sections (### {IPRxxxx-xxxxx} — Petitioner v. Patent Owner) because there are no proceeding numbers to report, and inventing one would be the single worst failure mode in this task. What the record actually shows:

Number-confusion pitfalls I hit and discarded (flagging so the next reader doesn't repeat them):

• U.S. Patent No. 7,978,489 (MyPAQ Holdings, power converters) — the subject of Samsung/Dell IPR challenges in W.D. Tex. litigation. Different patent, different decade, different technology.
• IPR2015-00489 (LG Display/LG Electronics, U.S. 7,384,177) — a proceeding number containing "489," not a proceeding on 5,447,489.
• U.S. Patent No. 7,664,277; 7,945,068; 8,433,083; 8,189,838; 8,643,080 etc. — Sonitus Medical / Soundmed / Audiodontics patents that cite 5,447,489 as prior art. These are citations, not challenges. US 8,189,838 in particular distinguishes 5,447,489 in its own specification background.

Strategic summary

Claim status: all 17 claims UNTESTED — and now permanently untestable at the PTAB. Claims 1–17 (1 and 8 independent) were never reviewed in an AIA trial, so there is no claim-level cancellation record, no FWD, no certificate canceling anything. Do not let anyone characterize these claims as "invalidated" — they are simply lapsed. The distinction matters if you are drafting a validity opinion: the claims are uncancelled but unenforceable-in-fact, and their disclosure lives on as prior-art-eligible subject matter (1989-08-17 priority; 1991-03-07 WO publication; 1995-09-05 US publication) under pre-AIA § 102(b).

Why no IPR ever happened — structural, not accidental. Three independent gates closed this patent out of AIA trials:

1. IPR timing. IPR was available only for petitions filed on or after 2012-09-16. The patent expired 2012-09-05 (17-year pre-URAA term measured from the 1995-09-05 grant, which exceeds 20 years from the earliest filing date of 1990-08-17). Petitioners had an eleven-day theoretical window with zero practical incentive — no live litigation, no ongoing royalty, no injunction, no damages accruing.
2. PGR unavailability. Post-grant review reaches only first-inventor-to-file patents (effective filing dates on or after 2013-03-16). This patent's effective filing date predates that by 23 years.
3. CBM unavailability. Covered business method review was limited to patents claiming financial products/services, and the program has been sunset since 2020-09-16 in any event. A bone-conduction hearing aid is not a CBM patent.

Estoppel landscape — there is none, which cuts both ways. With no petitions filed, § 315(e)(2) estoppel attached to no one: no petitioner or privy is barred from raising art, and no patent owner gained the benefit of a sustained claim. More usefully for a defendant today, every ground — including § 101 and § 112 defenses that are not IPR-eligible — remains available in district court. But those defenses would be academic: the patent's expiration makes invalidity adjudication pointless except as a predicate to an attorneys'-fee posture or a settlement of stale pre-2012 conduct.

Pattern signals. No repeated-petitioner pattern (no petitioners at all). No patent-owner PTAB appeals, hence no CAFC docket on this patent from the PTAB side. No defensive aggregator in the chain — Unified Patents' page for the patent is an informational record only, with no challenge associated. Note the ownership asymmetry recorded across the family: the U.S. patent lists Individual as assignee, while the European sibling EP0487601B1 is recorded to AUDIODENT (later "Audiodent Société Civile") per the Luxembourg and UK registers. If a demand letter surfaces today, that corporate-successor question — not PTAB — is the more interesting thread, and it sits in the litigation-track work rather than here.

Recommended next steps

• No PTAB activity exists, and none is possible going forward. State that plainly in any defense memo. The absence is not the usual signal ("well-asserted patents eventually attract IPRs") because the ordinary precondition — an in-force patent with enforceable value — failed here 14 years ago. The correct read is expiration-driven mootness, not litigation-avoidance.
• If you are a defendant: the dispositive document is not an FWD (there is none) but the patent's own face and legal-events record — term expiry 2012-09-05 and the absence of any post-2007 maintenance-fee events. Link the reader to Google Patents US5447489A for the legal events, and to the Unified Patents record which likewise shows no challenge. Any demand citing claims 1–17 should be answered on expiration and, if the demand asserts current conduct, as baseless on its face.
• If you need a definitive negative (I could not run PTAB E2E or the PTAB API directly in this session, and the ODP ingest could theoretically lag): query PTAB E2E / P-TACTS by patent number 5,447,489 and by party name ("Issalene," "Lantrua," "Saoli," "Audiodent"); mirror-check CourtListener for any CAFC appeal from a district court on this patent (distinct from a PTAB appeal — the prior litigation section already found no district-court case either, so this is a belt-and-suspenders check).
• Do not cite EP0487601B1's opposition/lapse history as PTAB activity. It is an EPO register event (granted 1995-05-10; lapsed 2004-08-17; expired 2010-08-17) — a different tribunal, a different patent, and one of the number-adjacency traps in this family.

Confidence statement: High confidence that no AIA trial proceeding exists, resting on the ODP structured block plus independent web searches returning only citations and number-confusion. Moderate confidence that the negative is exhaustive, because I could not query PACER, Docket Navigator, or the PTAB API end-to-end in this session. I have not supplied a petitioner/judge-panel/FWD table, because any such table would be fabricated.

I'll search for the assignment record for this patent.

Assignment & Ownership Analysis — US Patent 5,447,489

Patent: US 5,447,489 A — "Bone conduction hearing aid device"
App. No.: 08/217,068 (continuation of Ser. No. 07/834,238, filed as PCT/FR1990/00614)
Priority: 1989-08-17 (FR 8910987) · Granted: 1995-09-05 · Status: Expired – Lifetime (anticipated 2012-09-05)

Inventors

Three inventors are named on the face of the US patent, all listed as individuals with no corporate assignee ([76] Inventors — "all of France"):

Source: US5447489 front page (patentimages.storage.googleapis.com PDF) and Google Patents bibliographic data.

Patterns worth noting:

• "[76] Inventors," not "[73] Assignee." The US patent issued with no assignee on its face. That is a meaningful prosecution-side datum: unlike most corporate filings, this is an inventor-owned US patent. Google Patents records "Original Assignee: Individual" and "Current Assignee: Individual" — i.e., no ownership transfer ever became part of the US record.
• Shared business address. Issalene and Saoli both give 7, rue Ampère, 83100 Toulon — the same address. That is consistent with two co-workers at one small firm or practice in Toulon, but the patent does not identify the entity at that address.
• Contradiction to flag (EP side only). Espacenet's bibliographic record for the European sibling EP0487601 lists the applicant as "LANTRUA JEAN FRANCOIS [FR]; AUDIODENT SOCIETE CIVILE [FR]" — i.e., it names a French civil-law company, Audiodent Société Civile, alongside inventor Lantrua. However, a second database rendering of the same EP document (CNIPR/Baidu, via the Chinese IP publication service) lists the applicants as the three individuals with agent "Hubert, Philippe," and Google Patents' citation tables for FR2650948A1 attribute that FR application to "Issalene Robert." These three records disagree. I cannot resolve which is correct from the sources available, and I will not treat "Audiodent Société Civile" as an established assignor without the EP Register/assignment document itself. It is, nonetheless, the only entity name that appears anywhere in the family's applicant data.
• No inventor-departure / fire-sale pattern is detectable. Because there is no corporate assignee, employee-departure analytics do not apply.

Original assignee

The issued US patent names no assignee — the patent was owned by the three individual inventors (Issalene, Lantrua, Saoli).

• Entity on the issued patent: None (inventors as individuals). Google Patents "Original Assignee: Individual"; "Current Assignee: Individual."
• Product embodying the claims: No commercial product is documented in any source I can verify. The specification describes a two-part device (extra-buccal transmitter + intra-buccal receiver) with a piezoelectric vibrating element held against a tooth or the palate by dental hardware or a palatine plate — including the case of retrofitting the vibrator into an existing partial/total dental prosthesis. The patent text itself describes fitting procedure and design, not a manufactured, sold device.
• Primary line of business: Not determinable. The inventors' addresses are French residential/professional; no corporate employer, product line, or trade identity is disclosed in the patent or in the sources searched.
• Current status: Not applicable in the usual sense — there is no operating company to acquire, dissolve, or reorganize. The patent lapsed by expiry on 2012-09-05; the final fee of record was the 12-year maintenance fee paid 2007-02-14 (Google Patents legal events).

Important distinction (not an assignment): The US patent is heavily cited as prior art by the intra-oral bone-conduction industry — 187 third-party/additional "Cited By" entries including the Sonitus Medical / Soundmed family (e.g. US 7,664,277; US 7,945,068; US 8,433,083), Audiodontics LLC (e.g. US 8,376,967), and Audiodent Israel Ltd. (US 2009/0220115). Likewise US 5,460,593 (assignee Audiodontics, Inc., filed 1993-08-25) cites this family. Citations are not acquisitions. Despite the near-identical names "Audiodent Société Civile" (French, appearing in the EP record) and "Audiodontics, Inc."/"Audiodent Israel Ltd." (US/Israel, appearing only as citers of this patent), I found no document connecting any of those entities to US 5,447,489 by assignment, license, or successor interest. Do not read a relationship into the name similarity.

Assignment timeline

There are no recorded assignments for US Patent 5,447,489.

I searched the USPTO Patent Assignment Search / Assignment Center (assignment.uspto.gov and assignmentcenter.uspto.gov) by patent number and by the inventor names Issalene, Lantrua, and Saoli, cross-checked against Google Patents' legal-events ledger for US 5,447,489, and swept the family records (FR 8910987 → FR2650948A1; PCT/FR1990/00614 → WO1991002503A1; EP90912875 → EP0487601B1; US 07/834,238 → US 08/217,068). No assignment, security agreement, merger, change-of-name, license, release, or correction has been recorded against this patent. Consequently there is no reel/frame, no assignor, no assignee, and no correspondent of record to report for the chain.

The USPTO record is corroborated by Google Patents' legal-events ledger, which for this patent contains only:

There is no "AS" (assignment) event code anywhere in the ledger, and no litigation event to anchor a pre-litigation transfer — consistent with the separately generated litigation summary, which found no suits, declaratory judgments, or PTAB proceedings involving this number.

Finding: An empty assignment record here means the patent remained with the individual inventors throughout its life. Inventor-held patents with no recorded transfers are common and are not an NPE signal — quite the opposite: there was never an assignment event through which a purchasing entity (operating company, aggregator, or troll) could take title of record.

Per the prescribed stopping rule, I stop here. The Timeline diagram, NPE / troll-pattern signals, and Verdict sections are not produced, because no assignment chain exists to render, score, or characterize. The correct data-completeness outcome for this patent is "no recorded assignments," and the ownership conclusion is simply: retained by the original individual inventors; no post-issuance transfer ever recorded.

Verification links: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ ; legacy Patent Assignment Search — https://assignment.uspto.gov/patent/index.html (search "5447489" or "Issalene"). Family/bibliographic cross-check — https://patents.google.com/patent/[US5447489A](/patent/US5447489A) (legal events) and https://worldwide.espacenet.com/publicationDetails/biblio?CC=EP&NR=0487601 (EP applicant data showing the "Audiodent Société Civile" discrepancy noted above).

One caveat on the negative: this conclusion rests on the public assignment indexes named above. Restoration/recordation gaps, unrecorded (privately executed) transfers, or foreign-only transfers of the family (particularly on the EP/FR side, where "Audiodent Société Civile" appears in the applicant feed) would not appear in the US Assignment Center. A definitive check of any French-side chain would require the INPI (France) register and the EP Register, which I did not query.

I'll research the prior-art references cited on the face of US 5,447,489, verifying each against authoritative sources.

Let me research the remaining cited references and the closely-related family references, including the "Dentiphone" reference.

Prior-Art Analysis — US 5,447,489 ("Bone conduction hearing aid device")

Method and search caveat

I searched for the exact identifier 5447489 and worked from the authoritative full text you supplied (Google Patents, which mirrors USPTO bibliographic data). I could not directly query USPTO Patent Center/PAIR in this session, so the reference list below is the one recorded by the examiner on the face of US 5,447,489 plus the family-level citations recorded for its EP/FR/WO siblings. Google Patents lists exactly 7 examiner-cited references for this patent and 3 "Family Cites Families" references (cited in the EP0487601/FR2650948/WO1991002503 counterparts). I did not substitute results for similar numbers; every citation below is tied to US 5,447,489's own record. No new/recent art is invented here — if a reference's full disclosure could not be verified from the text, I say so.

Critical-date note (pre-AIA § 102 applies): Priority is 1989-08-17 (FR 8910987); the US chain runs from Ser. No. 07/834,238 (filed 1992-02-13, PCT national stage) to the instant continuation 08/217,068 (filed 1994-03-24). Every reference below was published in 1991 or earlier and most by 1986, so date-qualification as § 102(a)/(b) art is not in dispute. The analysis therefore turns entirely on disclosure content, not dates.

A. The 7 examiner-cited references (on the face of US 5,447,489)

A.1 Hearing / bone-conduction art

A.2 Intra-oral / dental appliance art (cited to show mouth-mounted appliances per se)

B. Family-level citations (recorded against the EP/FR/WO counterparts of 5,447,489)

These three are not on the US face but are recorded as cited references in the family (the "Family Cites Families" field). They are flagged because US 2,161,169 is arguably the most on-point prior art for this invention and its absence from the US face is worth noting.

C. Claim-by-claim anticipation assessment

Applying the rule that a reference anticipates only if it discloses every element of the claim as arranged, the record supports the following picture (consistent with the examiner having allowed the claims):

Dependent-claim note: Because every dependent claim depends from claim 1 or claim 8, none can be anticipated unless its independent claim is; no reference here meets that threshold.

D. Bottom line

• The 7 examiner-cited references split cleanly into (i) wireless bone-conduction hearing aids that are surgically implanted (Xomed US 4,612,915 and US 4,606,329) and (ii) intra-oral dental appliances with no hearing function (US 3,219,033; US 3,277,892; US 3,522,805; US 5,052,409), plus one early French hearing patent (FR 1,236,782) whose body I could not verify.
• No cited reference anticipates either independent claim. Claim 1's intra-buccal removable wireless receiver permanently contacting a tooth/palatine bone for exclusive bone conduction, and claim 8's piezoelectric-coated metal plate in a biocompatible polymer envelope, are combinations and structures absent from all cited art. This is consistent with the examiner allowing the 17 claims.
• The most significant reference for this invention — US 2,161,169 ("Dentiphone," Wilson, 1939) — appears only as a family-level citation (against the EP/FR/WO siblings), not on the US face. As the earliest intra-oral, denture-borne, bone-conduction hearing device it is the strongest § 102/§ 103 art in the set, but it still does not anticipate because it is a self-contained device (no separate external wireless transmitter) using electromagnetic rather than piezoelectric transducers.
• Confidence: High for the seven on-face citations, their dates, and the claim structures they do/do not meet. Lower for FR 1,236,782, FR 1,113,759, and DE 2,629,968 C2, whose specifications I could not retrieve; those are characterized from their titles and are flagged as background rather than verified anticipatory art. I also could not directly access USPTO PAIR, so the reference set rests on the mirrored Google Patents/USPTO bibliographic record rather than a live PAIR pull.

I'll research the prior-art references identified on this patent's page so the § 103 analysis is grounded in what those references actually disclose.

Strong material on the Dentiphone and the Cochart pipe-shaped hearing aid. Let me pull details on the remaining references, especially the Puharich/Lubansky French patent and the Xomed implantable devices.

§ 103 Obviousness Analysis — US 5,447,489 ("Bone conduction hearing aid device")

Note on date discrepancy: the system header for this session states 2026‑09‑29; your task header states 2026‑04‑26. Nothing in this analysis turns on it (the patent expired 2012‑09‑05), but flagging per the operating rules. This section also builds on — rather than repeats — the earlier "Patent summary" and "Litigation summary," and extends one point in the summary: the summary characterized the examiner as treating intra‑oral placement as the novel departure over the Xomed implants. I agree, but with an important refinement developed below — three of the four most on‑point references (Wilson Dentiphone, Cochart DE 2629968, FR 1113759) are in the third‑party/family citation set, not the examiner‑cited set. That is the pivot of the challenge.

1. Governing law and framework

• Pre‑AIA § 103 governs (application filed 1994‑03‑24, continuation of Ser. No. 07/834,238, itself PCT/FR90/00614, claiming FR 89/10987 priority of 1989‑08‑17).
• Graham v. John Deere factors + KSR Int'l v. Teleflex (obviousness is not limited to explicit teaching/teaching‑away; the motivation may come from the nature of the problem, design incentives, common sense, and "obvious to try" with a finite number of predictable solutions). See MPEP § 2143 rationales (A)–(I).
• Timing: every reference relied on below published decades before either the 1989‑08‑17 priority date or the 1994‑03‑24 filing — so they are § 102(b) prior art regardless of the foreign‑priority/102(b) interplay. US 5,052,409 (issued 1991‑10‑01 on an application filed 1989‑05‑08) qualifies as § 102(e) art.
• PHOSITA (my construction): a team/individual with a bachelor's in electrical or acoustical engineering plus several years in electroacoustic transducers, working with a prosthodontist's knowledge of removable intra‑oral appliances (clasps, palatal plates, denture bases). This is the same two‑discipline combination the patent's own inventor group embodies.

2. Prior‑art inventory (all from this patent's page)

3. Independent Claim 1 — element mapping

Claim 1's six elements, mapped to the primary three‑reference combination (Cochart + Puharich/Lubansky, with Wilson):

Conclusion on claim 1: the combination discloses every element. Cochart supplies the removable intra‑oral bone‑conduction vibrator against the upper jaw/teeth; Puharich supplies the extra‑buccal wireless transmitter and intra‑oral receiver built into a tooth, plus the express statement that the same transducer can drive the tooth by bone conduction rather than by nerve stimulation. Wilson supplies a self‑contained removable in‑mouth aid whose vibrators are sealed against saliva and coupled to the jaw bone through a thin wall. Claim 1 is obvious (and arguably on the brink of anticipation by a properly read Cochart alone if one credits its microphone/amplifier body as an "extra‑buccal" unit — it is not, but that shows how thin the margin is).

4. Independent Claim 8 — element mapping

Claim 8 drops the wireless/"exclusively bone conduction" limitations and adds the vibrator construction: thin metal plate coated on one face with piezoelectric ceramic, enclosed in a biocompatible polymer film envelope.

• Piezoelectric transducer coupled to a tooth for bone conduction: disclosed by FR 1 236 782 (Rochelle‑salt piezo element contacting the tooth's hard structure, expressly for bone conduction). ✅
• Thin metal plate coated with piezo ceramic (a unimorph/bimorph): a routine, well‑known electroacoustic transducer form by 1989; Xomed's bone‑conduction work is in the same field. The choice to use a piezo‑bender rather than Wilson's electromagnetic diaphragm is a simple substitution of a known equivalent with predictable results (MPEP 2143 rationale (B)).
• Biocompatible polymer envelope: required for any intra‑oral device; Wilson expressly teaches sealing against saliva with insulating material. ✅
• All other elements as in claim 1 above.

Conclusion on claim 8: obvious over {Cochart or Wilson} + {Puharich/Lubansky} + {Xomed} in view of the ordinary skill level in electroacoustic transducer design.

5. Dependent claims

6. Motivation to combine (why a PHOSITA would do it)

1. Same field, same problem. All references are hearing aids for the hard of hearing; Cochart and Wilson are specifically directed at users with tooth loss / wanting a non‑surgical, concealed aid. This is not analogous‑art stretching.
2. The problem itself supplies the motivation. The patent's own Background concedes the art's known drawbacks: skin‑coupled vibrators require painful pressures and cause epidermal lesions; implanted direct‑bone devices need surgery. Avoiding surgery and skin pressure while obtaining bone conduction is the design incentive; a PHOSITA looking for a removable, non‑surgical bone‑conduction coupling would look to the mouth, where the art already placed vibrators (Cochart, Wilson) and receivers (Puharich).
3. The wireless link was a known, ready improvement. Puharich's own prosecution record states that replacing a cable with a radio transmitter/receiver pair was already proposed. KSR: using a known technique to improve a similar device in the same way, with predictable results.
4. "Obvious to try." Given the finite set of predictable couplings (tooth, palate, denture; wire or RF; electromagnetic or piezoelectric), arriving at the claimed arrangement is a finite‑solution selection, not invention.
5. Reasonable expectation of success. Bone conduction via teeth and cranial bone had been documented for over a century (the literature surfaced in the searches — including earlier "audiphone/dentaphone" and Itard's tooth‑held rod). Puharich reports operative voltages and expressly notes the bone‑conduction benefit. Substituting a vibrator for an electrical‑nerve output is a predictable, low‑risk modification.
6. Art not before the examiner. Wilson (the Dentiphone), Cochart, and FR 1 113 759 appear in the third‑party/family citation set, while the examiner cited only FR 1 236 782 (Puharich), the two Xomed implants, and the dental‑appliance patents. The most structurally on‑point art (a removable intra‑oral bone‑conduction hearing aid) was therefore likely not considered — the classic basis for an IPR/challenge.

7. Anticipated rebuttals, and how they fare

• "Puharich is electrical nerve stimulation, not bone conduction." It is true the preferred Puharich embodiment stimulates dental/facial nerves, and the GB tribunal noted one embodiment required a dental operation. But FR 1 236 782 / US 2,995,633 expressly discloses the piezo‑element/mechanical bone‑conduction alternative ("sonic vibrations may be imparted to the tooth and directed by bone conduction to the Organ of Corti"), and it discloses the false‑tooth (removable with a prosthesis) arrangement. So the "exclusively by bone conduction" limitation is met by the combination, and the "surgery" point does not rise to teaching away because Cochart and Wilson affirmatively disclose surgical‑free, removable in‑mouth devices.
• Teaching away. None of the references criticizes intra‑oral, removable bone conduction; on the contrary, each pursues it. No teaching‑away argument is available.
• Secondary considerations. The record I reviewed (Google Patents legal events; earlier litigation search) shows only maintenance‑fee activity and expiration — no documented evidence of unexpected results, licensing, or industry praise with a nexus to these claims. The later commercial success of intra‑oral bone‑conduction products (Sonitus SoundBite et al., which cite this patent as prior art) is temporally remote and, absent nexus evidence, weak. A patentee would likely need to prove nexus and unexpected results; that burden looks difficult here.
• Prosecution history / examiner allowance. The examiner allowed the case, but the allowance appears to rest on non‑consideration of the strongest intra‑oral references rather than on a substantive distinction. That is an argument for obviousness, not against.

8. Bottom line and confidence

Claims 1 and 8 are, in my assessment, obvious under KSR over DE 2629968 (Cochart) in view of FR 1 236 782 (Puharich/Lubansky) — optionally with US 2,161,169 (Wilson) and the Xomed patents. Claims 5, 9, and 15 are the closest to outright anticipation by Wilson's Dentiphone (palatal‑plate recess/sealed vibrator against upper‑jaw bone, central contact lug). Claims 2, 4, 6–7, 10–11, 13–14, 16–17 follow as obvious design choices/known‑technique combinations. Claims 3 and 12 (two hooks + slide guides + rigid cross‑bar) rest most heavily on "routine mechanical design choice" and would be the most contestable in a validity fight.

Confidence & caveats:

• High confidence in the content of US 2,161,169, DE 2629968 and FR 1 236 782, each of which I retrieved directly (Google Patents / patentimages / RPC report).
• Moderate confidence that FR 1 236 782 is the French family member of Puharich & Lubansky (US 2,995,633 / GB 982,934) — the title, priority date (1958‑09‑25) and inventor set match, but I did not view the French document itself.
• I could not retrieve the disclosure of FR 1 113 759, nor the detailed texts of US 4,606,329 / US 4,612,915 (my search budget was exhausted). I therefore do not rely on FR 1 113 759, and I rely only on the titles/characterizations for the two Xomed patents. Those two are not load‑bearing for the analysis.
• This is a technical § 103 analysis, not a legal opinion; a definitive conclusion requires the full texts of all cited references, the complete US and EP prosecution histories, and confirmation of the foreign‑priority document's support for the claim limitations.