US 10517484

I'll search for this specific patent before summarizing.

Let me pull the claim language and litigation/docket details.

Let me get the remaining independent claims and check for the latest 2026 docket activity.

Bottom line up front

US 10,517,484 is real, it matches the number exactly, and I found authoritative records for it in PTAB, Federal Circuit, and litigation documents. Critically: as of the most recent record I could retrieve, all 23 claims of this patent have been held unpatentable by the PTAB on remand, and Omni MedSci's appeal of that remand decision was pending. Any summary of the patent's "claims" below describes the subject matter as issued/asserted, not as currently enforceable.

1. Bibliographic identification

Sources: Google Patents entry for US10517484B2; PTAB IPR2021‑00453 exhibits (Docket Alarm); Fed. Cir. slip op., Apple Inc. v. Omni MedSci, Inc., No. 2023‑1034 (June 21, 2024).

2. Abstract (verbatim, from the printed patent)

"A wearable device includes a measurement device to measure a physiological parameter adapted to be placed on a wrist or an ear of a user. A plurality of semiconductor light sources such as light emitting diodes generate corresponding output light having an initial light intensity. A receiver includes spatially separated detectors receiving reflected light from the output lights and coupled to analog to digital converters. The receiver is configured to synchronize to the semiconductor source(s). The measurement device improves signal-to-noise ratio of the output signal by increasing light intensity relative to the initial light intensity and by increasing a pulse rate. Further improvement in signal-to-noise ratio is achieved by using change detection, where the receiver compares the signals with light on and with light off."

3. Independent claims — plain language

The patent has 23 claims. PTAB materials identify four independent claims: 1, 7, 15, and 17 ("As to independent claims 1, 7, 15, and 17, Petitioner argues…"). Dependent claims 3–6 depend (directly or indirectly) from claim 1; 8–14 from claim 7; 16 from claim 15; and 18–23 from claim 17.

Claim 1 (verbatim, reproduced in the PTAB FWD and Patent Owner Response) — "A system for measuring one or more physiological parameters and for use with a smart phone or tablet," comprising:

1. A wearable device on the wrist or an ear of a user;
2. A light source of a plurality of semiconductor sources that are LEDs, each generating output optical light at one or more wavelengths;
3. One or more lenses receiving part of the output light and directing a "lens output light" to tissue;
4. A detection system receiving part of the reflected lens output light and generating an output signal having a signal-to-noise ratio, synchronized to the light source;
5. A plurality of spatially separated detectors, with at least one analog-to-digital converter coupled to at least one detector;
6. An amplifier with a gain configured to improve detection sensitivity, coupled to a detector output;
7. A smart phone or tablet with wireless receiver/transmitter, display, speaker, voice input module, buttons/knobs, microprocessor and touch screen, configured to receive, process, store and display the output signal and to transmit part of it over a wireless link;
8. A cloud that receives that output status over the wireless link, processes it into processed data, stores it, and stores a history of the physiological parameter(s) over a specified period;
9. SNR improvement by (1) increasing the light intensity of at least one semiconductor source above an initial light intensity and (2) increasing the pulse rate above an initial pulse rate; and
10. Change detection: generating a first signal with the LEDs off, a second signal with at least one LED on, and increasing SNR by differencing the two.

In plain terms, claim 1 is a three-tier architecture — wrist/ear wearable → smartphone/tablet → cloud with historical storage — where the noise problem is attacked by cranking LED intensity, raising pulse rate, and subtracting an "LEDs-off" baseline frame from an "LED-on" frame.

Claims 7, 15, and 17 are companion independent claims. From the record I can confirm limited specifics but not verbatim text: claim 8 (depending from 7) adds "identify an object" plus threshold comparison, the comparison being performed by the wearable or the phone/tablet; claim 16 (depending from 15) uses the different phrase "detect an object"; and the Board noted claim 15 does not recite the pulse-rate limitation that is central to claim 1's dispute. I do not have authoritative verbatim text for independent claims 7, 15, and 17 (these were not reproduced in the retrieved materials), so I am flagging that as an explicit gap rather than reconstructing them.

4. Post-grant and litigation status (this is the material part)

• IPR2021‑00453, Apple Inc. v. Omni MedSci, Inc.: instituted Aug. 6, 2021 on claims 1–23, grounds based on Lisogurski (US 9,241,676), Tran (US 8,108,036), Carlson (US 2005/0049468) and others.
• Original Final Written Decision (Aug. 3, 2022): claims 1, 2, 7, and 15–23 unpatentable; claims 3–6 and 8–14 not proven unpatentable (the Board distinguished "identify an object" from "detect an object").
• Federal Circuit, No. 2023‑1034 (decided June 21, 2024), panel of Reyna, Cunningham, and Albright (sitting by designation; authored by Judge Albright): "AFFIRMED-IN-PART, VACATED-IN-PART AND REMANDED." The court affirmed the Board's construction of "identify an object" as "to recognize or establish an object as being a particular thing," but held the Board abused its discretion under Axonics, Inc. v. Medtronic, Inc., 75 F.4th 1374 (Fed. Cir. 2023) in refusing to consider Apple's reply-brief "alternative argument."
• Final Written Decision on Remand (Feb. 14, 2025): claims 3–6 and 8–14 unpatentable, and therefore "we determine that claims 1–23 are unpatentable."
• Appeal of the remand decision: Omni noticed appeal; Google Patents lists Court of Appeals for the Federal Circuit case 25‑1646. A PTAB exhibit dated Nov. 12, 2025 ("484 notice of appeal") confirms the remand FWD is on appeal.
• District court / other proceedings listed against this patent include: N.D. Cal. 4:20-cv-00563 (Omni MedSci v. Apple), E.D. Tex. 2:24-cv-01070 (Omni MedSci v. Samsung Electronics et al. / OnePlus / Fossil / Oura — asserting '484 among US 10,874,304, 11,160,455, 9,055,868 and 9,651,533), D. Del. 1:25-cv-00140, and E.D. Tex. 2:26-cv-00289.
• Samsung-side PTAB challenges (filed Aug. 2025) referencing the '484 patent as an exhibit/related patent include IPR2025‑01249, IPR2025‑01250, IPR2025‑01252, IPR2025‑01253, PGR2025‑00063, and PGR2025‑00064.

5. Caveats on this answer

• I could not directly open a live USPTO Patent Center / PatentScope record in this session; the bibliographic data above comes from Google Patents (which mirrors USPTO bibliographic data) plus the printed patent's own [57]/[60]/[65]/[73] fields as reproduced in PTAB exhibits.
• Your instruction to search "CAFC 2026 dockets" could not be fully satisfied. My search on the 2026 Federal Circuit activity for this patent ran out of allotted steps and returned no results. I therefore cannot confirm any 2026 Federal Circuit decision, briefing event, or docket number for US 10,517,484 beyond the Google Patents listing of case 25‑1646 and the Nov. 2025 notice of appeal. Treat any 2026 posture as unverified.
• The Google Patents page labels the legal status "Active," and it also lists a 2026 filing (E.D. Tex. 2:26-cv-00289). That status label appears inconsistent with the PTAB's all-claims-unpatentable determination now on appeal, so I would not rely on the "Active" flag as an enforcement assessment.
• No auto-correction was applied to any identifier; "10517484" was searched and returned US 10,517,484 B2, and the companion identifiers (16/506,885, IPR2021‑00453, 2023‑1034, 25‑1646) are reproduced exactly as they appear in the sources.

I'll research litigation involving US patent 10517484 using multiple sources.

Let me dig deeper into the specific cases and outcomes.

Litigation Involving U.S. Patent No. 10,517,484

Important threshold note on the patent number. US 10,517,484 (application 16/506,885, "Semiconductor diodes-based physiological measurement device with improved signal-to-noise ratio," Omni MedSci, Inc., inventor Mohammed N. Islam, granted Dec. 31, 2019) is referred to in the litigation record as "the '484 patent" or "US10517484B2." Its number should not be confused with the similarly numbered US 10,517,484-family relatives that appear in the same suits (e.g., 10,213,113, 10,517,484's siblings 10,411,776 / 10,386,230 / 10,201,283, and the later '304, '455, '790, '475 patents). Where sources say "'484 patent," they mean US 10,517,484.

Bottom line: Yes — US 10,517,484 has been the subject of substantial litigation and PTAB/appellate proceedings. The docket-family listing on the patent's own Google Patents page identifies the following proceedings tied to this patent: N.D. Cal. 4:20-cv-00563; PTAB IPR2021-00453; E.D. Tex. 2:24-cv-01070; E.D. Tex. 2:26-cv-00289; D. Del. 1:25-cv-00140; and Fed. Cir. Nos. 23-1034 and 25-1646 (https://patents.google.com/patent/US10517484/en, "Family has litigation").

A. District Court Cases

1. Omni MedSci, Inc. v. Apple Inc. — N.D. Cal. No. 4:20-cv-00563

Sources: Stanford NPE Litigation Database, patent 10517484 — "Omni MedSci, Inc. v. Apple Inc., 4:20-cv-00563, N.D. Cal." (https://npe.law.stanford.edu/patent/10517484); filing date confirmed in PTAB record, IPR2020-00029, Paper 7 ("Omni MedSci v. Apple Inc., Case 4:20-cv-00563 YGR (N.D. Cal) filed 1/24/2020"); Google Patents litigation listing for US10517484B2.

2. Related earlier/companion Apple action — N.D. Cal. No. 4:19-cv-05924

Caveat (literal reading, no auto-correction): I am reporting this docket number exactly as it appears in the Samsung IPR petition record ("Omni asserted the '533 and '484 patents in previous litigations, including against Apple Inc. See Omni MedSci, Inc. v. Apple Inc., No. 4:19-cv-05924 (N.D. Cal.)"). Note the chronological tension: No. 4:19-cv-05924 was filed before US 10,517,484 issued on Dec. 31, 2019, so the '484 patent could only have entered that case by amendment (or the citation may be imprecise in the petition). I flag this rather than resolve it.

Sources: PTAB petition record (https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1558539](/patent/1558539)/download-documents...); Omni MedSci, Inc. v. Apple Inc., No. 20-1715 (Fed. Cir.), via CourtListener.

3. Omni MedSci, Inc. v. WHOOP, Inc. — D. Del. No. 1:25-cv-00140

Source: Omni MedSci, Inc. v. Whoop, Inc., No. 1:25-cv-00140 (D. Del.), D.I. 61, Memorandum Opinion and Order (https://law.justia.com/cases/federal/district-courts/delaware/dedce/1:2025cv00140/88127/61/ and https://www.courtlistener.com/opinion/[10860956](/patent/10860956)/omni-medsci-inc-v-whoop-inc/).

4. Omni MedSci, Inc. v. Samsung Electronics Co., Ltd., et al. — E.D. Tex. No. 2:24-cv-01070

Sources: IP-NAVI dispute record (case 2:24-cv-01070, filed 2024.12.20, listing US10517484 | US10874304 | US11160455 | US9055868 | US9651533) (https://ip-navi.or.kr/ipnavi/dispute/daily/dailyDetailPopup.navi?newsSeq=[218401](/patent/218401)); Ex. 2001, Complaint in 2:24-cv-01070 (PTAB IPR2025-01252).

5. Omni MedSci, Inc. v. [defendant(s)] — E.D. Tex. No. 2:26-cv-00289

B. PTAB and Appellate Proceedings Directly on US 10,517,484

6. Apple Inc. v. Omni MedSci, Inc. — PTAB IPR2021-00453

7. Apple Inc. v. Omni MedSci, Inc. — Fed. Cir. No. 23-1034

Sources: Apple Inc. v. Omni MedSci, Inc., No. 23-1034 (Fed. Cir. June 21, 2024) (https://storage.courtlistener.com/pdf/2024/06/21/apple_inc._v._omni_medsci_inc..pdf); VitalLaw IP Law Daily summary.

8. IPR2021-00453 — Final Written Decision on Remand (PTAB)

Source: IPR2021-00453, Paper No. 26 (P.T.A.B. Feb. 14, 2025); reproduced as Ex. 1016 in Samsung's IPR2025-01253.

9. Omni MedSci, Inc. v. Apple Inc. — Fed. Cir. No. 25-1646

Sources: IPR2021-00453, Patent Owner's Notice of Appeal (Apr. 11, 2025); Samsung IPR petition exhibit list identifying "Omni MedSci, Inc. v. Apple Inc., No. 25-1646 (Fed. Cir.)"; Google Patents litigation listing.

C. Proceedings I Checked That Do Not Involve US 10,517,484

To avoid mis-attribution (per your instruction not to return near-number matches):

• IPR2019-00916 (Apple v. Omni MedSci) — the '533 patent (US 9,651,533), not the '484. Claims held unpatentable; affirmed at Fed. Cir. No. 21-1229 (June 8, 2022).
• Fed. Cir. No. 20-1715 / Omni MedSci v. Apple, 7 F.4th 1148 (2021) — standing/ownership appeal concerning US 9,651,533 and US 9,861,286; interesting on assignment law but not the '484 patent.
• IPR2020-00029 / IPR2025-01249 and the other 2025 "Samsung petitions" — these target the '868, '533, '304, '455, '790, and '475 patents. The '484 patent was dropped from the WHOOP case and does not appear in the Samsung 2025 challenge set.
• PGR2025-00063 and PGR2025-00064 (Samsung) — '304/'455-related post-grant reviews; the '484 patent appears in them only as an exhibit/file-history reference (as a parent of US 10,874,304 and US 11,160,604-family continuations), not as the challenged patent.

D. Practical Read-Through

1. US 10,517,484 is currently in an adverse posture. Every claim (1–23) has been held unpatentable by the PTAB (FWD Aug. 3, 2022, plus FWD on Remand Feb. 14, 2025). Its live enforcement value depends entirely on whether Omni prevails in Fed. Cir. No. 25-1646.
2. Enforcement on the '484 patent has effectively narrowed. Omni voluntarily dismissed its '484 claims in the Delaware WHOOP case; the district-court asserters of record are the N.D. Cal. Apple case (4:20-cv-00563) and the E.D. Tex. Samsung multi-defendant case (2:24-cv-01070), with a possible new 2026 E.D. Tex. action (2:26-cv-00289) I could not confirm.
3. Reliability caveat. The PatSnap blog article (patsnap.com) describing "Apple v. Omni MedSci" as a district-court "appeal" with an "affirmed-in-part, vacated-in-part and remanded" verdict is imprecise: No. 23-1034 was an appeal from a PTAB Final Written Decision in IPR2021-00453, not from a district court. I relied on the Federal Circuit opinion and PTAB papers rather than that secondary summary.

Sources consulted: Google Patents US10517484B2 (litigation links, family data); Stanford NPE Litigation Database, patent 10517484; CourtListener (N.D. Cal., D. Del., Fed. Cir.); Justia Dockets (D. Del. 1:25-cv-00140); PTAB papers and exhibit lists via docketalarm.com and ptacts.uspto.gov; IP-NAVI dispute record for E.D. Tex. 2:24-cv-01070. PACER itself was not directly reachable in this session, so filing dates and outcomes are as reflected in the secondary/PTAB-record sources cited. Items I could not verify are flagged as such above.

I'll research this. The structured block indicates no ODP-indexed proceedings, but the patent's own Google Patents page flags IPR2021-00453 — let me verify with search.

Good — the ODP block is empty but Google Patents flags IPR2021-00453. Let me pin down petitioner identity, the appeal chain, and whether Samsung has since filed on the '484.

Now let me check the status of the pending CAFC appeal and confirm no newer petition targets the '484.

Proceedings overview

One AIA trial proceeding has been filed against US 10,517,484 — Apple Inc. v. Omni MedSci, Inc., IPR2021-00453 — and it is a total loss for the patent owner: the Board has held all 23 claims unpatentable, with the invalidation of claims 1, 2, 7, and 15–23 now final and unappealed, and claims 3–6 and 8–14 invalidated on Federal Circuit remand in a decision currently on appeal (No. 25-1646); the bottom-line defensive posture is that this patent has been extinguished, not hardened — any demand letter or infringement count resting on the '484 is resting on claim language that the PTAB has thrice-considered and canceled.

⚠️ Data-source flag (required): The structured "PTAB proceedings on file" block in this prompt returned nothing from the USPTO ODP API. That is a false negative. Google Patents' own litigation block for this patent flags IPR2021-00453 (FWD issued), and the proceeding is fully documented in public PTAB/CAFC records. Treat the ODP block as incomplete; do not tell a client "no PTAB activity." (For completeness: the portal.unifiedpatents.com link on the Google Patents page is Unified Patents' data portal, i.e., the source of Google's litigation feed — Unified Patents is not the petitioner here. Apple Inc. is, and it is the sole real party-in-interest. Per. 1.)

IPR2021-00453 — Apple Inc. v. Omni MedSci, Inc.

• Type: Inter Partes Review (35 U.S.C. §§ 311–319; pre-AIA patent, priority 2012-12-31, so IPR — not PGR — is the correct vehicle)
• Filed: 2021-01-22 (Paper 1). Apple's petition expressly noted the '484 was asserted against it in Omni MedSci, Inc. v. Apple Inc., No. 4:20-cv-00563 (N.D. Cal.) — i.e., the § 315(b) one-year clock drove the filing date.
• Status: Final Written Decision on remand determining all challenged claims unpatentable; appealed — PTAB docket reflects "Terminated 2025-02-14" and "Final Written Decision – Appealed."
• Judge panel: Grace Karaffa Obermann, Brian J. McNamara, and Sharon Fenick, APJs. McNamara authored the FWD on remand. (Same panel for both merits decisions.)
• Petition grounds: § 103 obviousness against all of claims 1–23, with U.S. Pat. No. 9,241,676 (Lisogurski) as the primary reference, combined with U.S. Pat. Pub. No. 2005/0049468 (Carlson) and U.S. Pat. No. 8,108,036 (Tran). Support: Declaration of Brian W. Anthony, Ph.D. (Ex. 1003). The FWD separately addressed claims 1, 7, 15, and 17 as obvious over Lisogurski + Carlson, and claims 3, 8, and 16 over Lisogurski + Tran + Carlson. The petition exhibit list also included Valencell-093, Hanna, and Mannheimer; I cannot confirm from the available record which claims each of those supported, so I do not attribute them to a specific ground.
• Institution decision: Instituted 2021-08-06 (Paper 7) — all claims, all grounds ("On August 6, 2021, we instituted an inter partes review of all claims on all challenge grounds"). No claim was carved out at institution; Apple is the sole RPI.
• First Final Written Decision — 2022-08-03 (Paper 22):
  • Unpatentable: claims 1, 2, 7, and 15–23 (this sweep includes independent claims 1, 7, and 15 and every claim depending from them, including claim 16).
  • Not shown unpatentable: claims 3–6 and 8–14.
  • Reasoning, briefly: the merits fight was narrow. The panel wrote that "much of the dispute concerns claim limitation 1(l)(2), the 'pulse rate limitation,'" and that "Patent Owner does not contest Petitioner's contentions that the remaining claim limitations [are] obvious over the cited prior art." The surviving dependent claims turned on construction: the Board held that "identify an object" (claims 3 and 8) is distinguished from "detect an object" (claim 16), and found Lisogurski's probe-off/error-signal teaching sufficient for the detecting limitation but not the identifying limitation.
• Federal Circuit appeal (Apple): No. 23-1034, decided 2024-06-21, nonprecedential, panel Reyna, Cunningham, and Albright (D.J., sitting by designation), opinion by Judge Albright:
  • Affirmed-in-part the Board's construction of "identify an object" = "to recognize or establish an object as being a particular thing."
  • Vacated-in-part and remanded, holding the Board abused its discretion under Axonics, Inc. v. Medtronic, Inc., 75 F.4th 1374 (Fed. Cir. 2023), by refusing to consider Apple's reply-brief alternative argument that Lisogurski and Tran disclose the identifying limitation even under Omni's construction. (Citing SAS Inst. v. ComplementSoft and Hamilton Beach.)
  • Rehearing (panel + en banc) denied 2024-09-06; mandate issued 2024-09-13.
  • Opinion link: https://www.courtlistener.com/pdf/2024/06/21/apple_inc._v._omni_medsci_inc..pdf
• Final Written Decision on remand — 2025-02-14 (Paper 26):
  • Verdict, claim-level: "we have reviewed the merits and determine that claims 3–6 and 8–14 are unpatentable. Having determined that claims 3–6 and 8–14 are unpatentable, and recognizing that all other claims have previously been determined to be unpatentable, we determine that claims 1–23 are unpatentable."
  • The Board also entered an order denying post-remand briefing (37 C.F.R. § 42.5) after a telephonic hearing on 2024-10-29. The judgment is styled as a "Final Written Decision on Remand Determining All Challenged Claims Unpatentable" under 35 U.S.C. §§ 144, 318(a).
• Settlement / termination: None. This proceeding was decided on the merits at every stage; there was no adverse judgment, no abandonment, and no settlement.
• Second appeal (Omni): Omni filed a Notice of Appeal on 2025-04-11 (Paper 27) directed at the remand FWD's treatment of claims 3–6 and 8–14. Docketed as No. 25-1646 (Fed. Cir.), docketed 2025-04-16; Omni's corrected opening brief was filed 2025-09-11 (Dkt. 17). I could not confirm a merits disposition as of today (2026-09-28) — verify the live docket before relying on finality of the claims 3–6/8–14 holdings. Docket: https://dockets.justia.com/docket/circuit-courts/cafc/25-1646
• Defensive value: Decisive, with one caveat. Claims 1, 2, 7, and 15–23 are dead and were never appealed — any infringement theory built on the '484's independent claims (or claim 16) is not merely weak, it is inconsistent with a final, unchallenged administrative adjudication, and Omni is precluded from taking positions inconsistent with it. Claims 3–6 and 8–14 have been held unpatentable twice over (FWD on remand), but that holding rides on the pending 25-1646 appeal. Note also the practical tell: the '484 still shows legal status "Active" at the PTO with anticipated expiration 2033-12-17, because no § 318(b) certificate of cancellation can issue while the appeal is pending. "Active" in Patent Center does not mean enforceable claims remain.

Strategic summary

Claim status. Every one of the '484's 23 claims has been adjudicated unpatentable. Unpatentable and final (never appealed): claims 1, 2, 7, 15, 16, 17, 18, 19, 20, 21, 22, 23 — that is all three independent claims (1, 7, 15) plus their dependents. Unpatentable subject to pending appeal: claims 3, 4, 5, 6, 8, 9, 10, 11, 12, 13, 14. There are no sustained claims and no untested claims remaining in the patent. The appellate tether is narrow: Omni's 25-1646 appeal concerns "just one limitation across those claims" (the "identify an object" limitation) per the petitioners' own characterization in the parallel Samsung filings. If that appeal fails, the patent is a carcass.

Estoppel landscape. § 315(e)(2) estoppel runs against Apple and its privies/real parties-in-interest only — they cannot relitigate in district court any ground raised or reasonably raisable in IPR2021-00453, which as a practical matter is academic because Apple already won outright on claims 1, 2, 7, and 15–23. A new defendant is not barred by § 315(e)(2) from filing its own petition, but that is a bad trade: (i) the entire claim set is already adjudicated unpatentable, so the better play is collateral estoppel / issue preclusion against Omni on the finally-invalidated claims, exactly the posture Samsung took in the sibling proceedings (citing Samsung Elecs. Co. v. Netlist, Inc., IPR2025-00002, Paper 17 (PTAB May 15, 2025)); and (ii) a fresh petition would face § 314(a)/§ 325(d) discretion under General Plastic and Advanced Bionics in a family that has already been picked over. The estoppel worth asserting is one-way: Omni is bound by the '484 FWD findings (37 C.F.R. § 42.73(d)(3)) and, per petitioners' arguments, cannot enforce later continuation claims "not patentably distinct" from the invalidated '484/'533 claims.

Pattern signals. (1) Not a defensive-aggregator case — Apple, a commercial adversary, filed the only petition on the '484, and it is Apple's second campaign against this family (Apple filed IPR2019-00910 through -00917 against the earlier Omni patents, e.g. IPR2019-00916 on the '533). (2) The patent owner litigates appeals aggressively — Omni has taken adverse FWDs up repeatedly (Fed. Cir. Nos. 20-1715, 21-1229, 21-2213, 23-1034, and now 25-1646), losing at least two of them. (3) The '484 is the family's lead domino. Omni asserted the '484 in Omni MedSci v. Apple (N.D. Cal. 4:20-cv-00563), in Omni MedSci v. Samsung et al., No. 2:24-cv-01070 (E.D. Tex.) (per the NPE database / case reporting), and it appears in the patent's litigation feed for a 2026 E.D. Tex. filing (2:26-cv-00289) and the Delaware WHOOP case (1:25-cv-140) — where Omni voluntarily dismissed its '484 claims (D. Del. C.A. No. 25-140, Mem. Op. & Order). (4) No new petition targets the '484. Samsung/Fossil/Oura/OnePlus filed IPR2025-01249–01254 and PGR2025-00063/00064 on '868, '533, '304, '455, '790, and '475 — expressly not the '484, because Apple's IPR already killed it (Samsung/Fossil/Oura Motion to Stay, E.D. Tex., 2025-08-27: defendants petitioned on "all asserted claims of the Asserted Patents, except for the '484 patent"). Those parallel proceedings nonetheless lean heavily on the '484 FWD as their estoppel anchor, so the '484 record keeps doing work for defendants across the family.

Recommended next steps

1. If you are a defendant being asserted on the '484: lead with the FWD, not with a new petition. Cite Apple Inc. v. Omni MedSci, Inc., IPR2021-00453, Paper 22 (PTAB Aug. 3, 2022) for claims 1, 2, 7, 15–23 and Paper 26 (PTAB Feb. 14, 2025) — "we determine that claims 1–23 are unpatentable" — and move to stay and/or for summary judgment of invalidity on the finally-adjudicated claims. PTAB E2E: https://ptacts.uspto.gov/ptacts/home (search IPR2021-00453); Google Patents record: https://patents.google.com/patent/US10517484/en.
2. Reconcile your docket with 25-1646 now. Check the Federal Circuit docket for the disposition of Omni's appeal of claims 3–6 and 8–14. Until it lands, the invalidation of those claims is binding on Omni but technically non-final; the claims have not been canceled by certificate. If the appeal is resolved against Omni, demand the § 318(b) certificate and treat all 23 claims as canceled.
3. Fix the ODP gap in your workflow. The emptiness of the ODP block here would have produced a materially wrong answer ("no PTAB activity"). Cross-check ODP against Google Patents' litigation block and the PTAB docket before advising on this patent family.
4. Don't waste money on a new IPR on the '484. With all claims already held unpatentable, institution would be discretionary-denial bait, and you would surrender the much stronger district-court estoppel argument. If you need offensive art coverage on the live continuation patents in this family ('790, '475, '455, '304), note that the Samsung/WHOOP petitions are already pending and were referred for merits review (Samsung: '533, '304, '455, '790, '475; denial on the '868), with institution decisions clustered in Q1 2026 and § 316(a)(11) FWDs one year later — coordinate rather than duplicate.
5. Watch the collateral-estoppel ripple. Because Samsung is arguing Omni is estopped from asserting claims "not patentably distinct" from the invalidated '484 claims, an early win on that theory in the Texas case could knock out a large slice of the family — the '484 FWD is the exhibit to build that motion around.

I'll research the assignment record for US 10517484 before writing anything.

Assignment & Ownership Analysis — US 10,517,484 B2

Verification caveat up front. The USPTO Assignment Center / Patent Assignment Search is a JavaScript application that I could not query directly in this session, and no secondary source I reached publishes the reel/frame for this patent's records. I therefore cannot state reel/frame numbers for US 10517484 and I will not invent them. Everything below is sourced from (a) Google Patents legal events for the patent (patents.google.com/patent/US10517484/en), (b) the Federal Circuit record in Omni MedSci, Inc. v. Apple Inc., 7 F.4th 1148 (Fed. Cir. 2021), and (c) PTAB/district-court dockets. To confirm reel/frame, run the patent number at assignment.uspto.gov. Treat every reel/frame field below as unverified.

Inventors

Unusual pattern — flag this. The classic tell ("all inventors leave the assignee within 12 months") does not apply, because the sole inventor is the assignee's founder and principal. What is unusual is the inverse: the inventor's university employer asserted ownership of the same inventions, which the inventor's own transfer to Omni MedSci had to defeat. UM's Office of Technology Transfer denied Islam's 2013 ownership request, and Apple's standing motion turned on it; the Federal Circuit held UM Bylaw 3.10 ("shall be the property of the University") was only a promise to assign, not a present assignment, so Islam's assignment to Omni MedSci survived. For an ownership-chain review, this means the recorded chain rests on judicially validated title, but the title was clouded at the moment of the original 2013 assignment.

Original assignee

Omni MedSci, Inc., Ann Arbor, MI (original assignee; also the current assignee of record). Founded ~2012 by Dr. Islam to commercialize broadband near-/short-wave-infrared super-continuum laser sources and non-invasive blood-constituent sensing (the '484 specification is squarely SWIR spectroscopy hardware — laser diodes, lens arrays, IR camera, smartphone/cloud back end).

• Products embodying the claims: I could not confirm any shipping product embodying the '484 claims. Public record shows Omni MedSci monetizing through patent assertion, not through a marketed device.
• Primary line of business (as observed): patent licensing / enforcement of an NIR-SWIR sensor portfolio. Third-party litigation databases classify the plaintiff entity as an NPE (IP-Navi dispute record, E.D. Tex. 2:24-cv-01070 lists "원고명 Omni MedSci Inc (미국 / NPE)").
• Current status: Active. Still the named plaintiff in 2024–2026 suits; Google Patents legal status shows "Active," with anticipated expiration 2033-12-17.

Assignment timeline

Two events are documented. The first is family-level (it covers the priority applications, not yet the '484, which did not exist). The second is the assignment of the '484's own application, 16/506,885.

• 2013-12-17 (executed) / recorded 2013-12 — Reel/frame NOT VERIFIED (Google Patents lists the priority claims from US 14/109,007, US 14/108,986 and the two PCT applications dated 2013-12-17; the Federal Circuit record confirms "On December 17, 2013, Dr. Islam assigned his patent rights to Omni MedSci; the assignments were recorded at the USPTO.")

  • Conveyance: Assignment (inventor → company)
  • Assignor: Mohammed N. Islam
  • Assignee: Omni MedSci, Inc. (Ann Arbor, MI)
  • Correspondent: not verified from the assignment record; prosecution correspondence of record is Brooks Kushman P.C., 1000 Town Center, 22nd Floor, Southfield, MI 48075 (Andrew B. Turner / David S. Bir). Flag: Brooks Kushman appears across essentially every Omni MedSci filing I reviewed, but that is prosecution correspondence, not the assignment-recording correspondent — do not treat the two as the same without the reel/frame.
  • Context: Founder-to-company transfer of the SWIR portfolio at formation; contemporaneous with UM's refusal to release ownership, which later produced the Fed. Cir. standing decision.

• 2019-08-15 (recorded; execution date not shown in my sources) — Reel/frame NOT VERIFIED

  • Conveyance: Assignment of Assignors' Interest (Google Patents: "Assigned to OMNI MEDSCI, INC. … Assignors: ISLAM, MOHAMMED N.")
  • Assignor: Mohammed N. Islam
  • Assignee: Omni MedSci, Inc.
  • Correspondent: not verified.
  • Context: Routine confirmatory assignment of newly filed continuation application 16/506,885 (filed 2019-07-09; granted 2019-12-31) into the same company that had owned the family since 2013 — internal to the existing owner, not a transfer to a third party.

No post-issuance assignments are recorded. Omni MedSci has owned the '484 continuously since the application stage; every 2020–2026 assertion names Omni MedSci as plaintiff. There is no shell-LLC link, no security interest, no merger re-record, and no change-of-name in the chain.

Timeline diagram

timeline
    title Ownership of US 10517484
    2012 : Provisionals filed by Islam
         : Omni MedSci founded in Ann Arbor
    2013 : Islam assigns family rights to Omni MedSci
         : UM refuses to release ownership
    2019 : Continuation app 16506885 filed
         : Assignment recorded to Omni MedSci
    2020 : First suit naming 484 filed in N D Cal
    2024 : Suits filed against Samsung OnePlus Fossil Oura
    2025 : Suit filed against WHOOP in Delaware

NPE / troll-pattern signals

1. Shell-entity transfer — not present. The only assignee in the chain is Omni MedSci, Inc., the original assignee, at an operating address (Ann Arbor, MI), founded by the inventor. No "IP/Holdings/Ventures" successor, no registered-agent service address, no single-purpose LLC appears in the record.
2. Known asserter in the chain — not present (against the enumerated lists), but qualified. Omni MedSci does not match Acacia, Marathon, IV, IPNav, Wi-LAN/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, or Spangenberg entities. However, it is a repeat plaintiff: 4:20-cv-00563 (N.D. Cal.), 2:24-cv-01070 (E.D. Tex.), 1:25-cv-00140 (D. Del., WHOOP), 2:26-cv-00289 (E.D. Tex.), plus IPR2021-00453 (Apple). Treat this as conduct-based, not chain-based evidence.
3. Repeat correspondent across the chain — unclear / not verifiable. The chain has a single link with a verified event, so recurrence within the chain cannot be assessed. Brooks Kushman P.C. recurs across Omni MedSci prosecution filings (Andrew B. Turner, Reg. No. 63,121; David S. Bir), and the attorney of record on the underlying applications is consistent — but without the reel/frame correspondent-of-record fields I will not assert that Brooks Kushman filed the assignment recordings.
4. Cascading transfers — not present. One recorded family assignment (2013) and one confirmatory assignment (recorded 2019-08-15). No chained LLCs, no shared-address successors, no transfers within 24 months of each other.
5. Pre-litigation transfer — not present in substance. Timing is superficially close (assignment recorded 2019-08-15; the '484 was first put in suit in N.D. Cal. 4:20-cv-00563 after its 2019-12-31 issuance), but the assignee was already the owner of the family since 2013. This was an inventor-to-owner papering of a new continuation, not a venue/standing orchestration between unrelated parties.
6. Bankruptcy fire-sale — not present. No bankruptcy of Omni MedSci or of the inventor appears in any record I reviewed.
7. Privateering — not present. Omni MedSci is not a proxy asserting on behalf of a larger operating company; it asserts on its own account as the original assignee.
8. Defensive aggregator — not present. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. To the contrary, Unified Patents' portal is tracking litigation against Omni MedSci.

Verdict

NPE — moderate confidence.

Justification: the recorded chain itself is unremarkable and shows no NPE assignment pattern — a single inventor-to-company assignment executed 2013-12-17 and a confirmatory assignment of application 16/506,885 recorded 2019-08-15, with no shell entity, no cascade, and no post-issuance transfer. The NPE character comes from the assignee's conduct rather than from the record: Omni MedSci, Inc. has no product I could verify embodying the '484 claims while asserting it against Apple, Samsung, OnePlus, Fossil, Oura, and WHOOP through 2026, and at least one public litigation database classifies it as an NPE. Confidence is capped at moderate precisely because the two strongest rubric signals (known-asserter match and repeat assignment correspondent) cannot be confirmed without the reel/frame entries.

Verify the underlying records at the USPTO Assignment Center: https://assignmentcenter.uspto.gov/ (search "10517484" or "Omni Medsci").

I'll research US Patent 10517484 and its cited prior art references.

Let me search for the specific cited references on the patent's front page and the IPR proceedings.

I have strong information on the IPR grounds. Let me get the full details of the key prior art references and the patent's front-page citations.

Prior Art Analysis — US Patent No. 10,517,484 B2

1. Patent Identification (verified)

Critical procedural note: The most consequential prior-art challenge to this patent is IPR2021-00453 (Apple Inc. v. Omni MedSci, Inc.) and the later PGR2025-00063 / PGR2025-00064 (Samsung Electronics Co., Ltd.). The IPR was decided on 35 U.S.C. § 103 (obviousness) — not § 102 (anticipation). This distinction matters and I flag it below.

2. Important Caveat on the § 102 Question

You asked which reference "potentially anticipates under 35 U.S.C. § 102." Based on the records I could retrieve:

• No reference was found to anticipate (i.e., disclose every limitation of a claim in a single reference under § 102). The Patent Trial and Appeal Board (PTAB) rejected the patent on obviousness (§ 103) grounds, using combinations of at least three references.
• This is technically significant: because no single reference disclosed all limitations, a strict § 102 anticipation theory was not the basis for invalidation. Below I identify the references and the claims to which each was applied, and I explicitly flag that these are § 103 combinations, not § 102 anticipations.

If you need a genuine single-reference § 102 analysis, that would require a limitation-by-limitation mapping against the '484 claims that is not present in the public IPR record.

3. Primary Prior Art — the IPR2021-00453 Grounds

The PTAB's Final Written Decision (Paper 22, Aug. 3, 2022) and the Federal Circuit decision (Apple Inc. v. Omni MedSci, Inc., No. 23-1034, Fed. Cir. June 21, 2024) applied the following grounds:

3.1 Lisogurski — U.S. Patent No. 9,241,676 B2

• Full citation: Lisogurski, Daniel, and Clark R. Baker, Jr. "Methods and Systems for Power Optimization in a Medical Device." U.S. Patent No. 9,241,676 B2. Assignee: Covidien LP, Mansfield, MA. Granted January 26, 2016 (priority ~May 30, 2012).
• Brief description: A photoplethysmography (PPG)/oximetry system that varies light-drive-signal parameters to optimize power — including light intensity, firing rate, and duty cycle — synchronized to the cardiac cycle. It discloses adjusting emitter intensity and firing (pulse) rate, and using a light source plus photosensitive detector to determine physiological parameters (blood oxygen saturation, hemoglobin, pulse rate, etc.). Source: https://www.freepatentsonline.com/9241676.html
• Relevance: This is the primary reference (the "base" reference) applied to essentially every claim. It maps to the '484 limitations of (a) a light source with adjustable intensity/firing rate and (b) synchronization of the driving electronics with physiological signals. All 23 challenged claims were analyzed starting from Lisogurski.
• Claims implicated: 1, 7, 15, 17 (Ground 1) and by extension all claims in Grounds 2–4.

3.2 Carlson — U.S. Patent Application Publication No. 2005/0049468 A1

• Full citation: Carlson, et al., "U.S. Patent Publication No. 2005/0049468." U.S. Patent Pub. 2005/0049468 A1. Published March 3, 2005.
• Brief description (as characterized in the IPR record): A pulse oximetry/optical sensing reference. In Apple's petition it supplies the wearable/optical-sensor architecture elements missing from Lisogurski. (I could not retrieve the full text of this reference in retrieval; the characterization here is limited to its role in the IPR grounds.)
• Claims implicated: 1, 7, 15, 17 (Ground 1); 1–4, 7–12, 15–22 (Ground 2); 5, 13 (Ground 3); 6, 14, 23 (Ground 4).

3.3 Tran — (U.S. Patent, "Tran")

• Citation: Referenced as "Tran" in the petition exhibit list (Ex. 1011 is Lisogurski; Tran is a separate U.S. patent used in Ground 2). I was unable to retrieve the full patent number/date for Tran in this session — I am flagging this as unverified rather than guessing a number.
• Relevance: Per the Board's decision, Tran teaches techniques for measuring blood oxygen saturation and other blood constituents. The dispute over Tran centered on whether "measuring" a constituent satisfies the claim's "configured to identify an object" limitation (Board found it did not for claims 3, 8; the Federal Circuit affirmed that construction but remanded for Apple's reply-brief argument).
• Claims implicated: 1–4, 7–12, 15–22 (Ground 2); 5, 13 (Ground 3); 6, 14, 23 (Ground 4).

3.4 Isaacson — (U.S. Patent, "Isaacson")

• Citation: Referenced as "Isaacson" in Ground 3. Full number/date not verified in this session.
• Relevance: Supplied structural/optical elements the petitioner argued a skilled person would incorporate into the Lisogurski+Carlson+Tran system.
• Claims implicated: 5, 13 (Ground 3); 6, 14, 23 (Ground 4).

3.5 Valencell-093 — U.S. Patent Application Publication No. 2012/0197093

• Full citation: "Valencell-093," U.S. Patent Pub. No. 2012/0197093. Assignee: Valencell, Inc. Published August 2, 2012.
• Brief description: A wearable physiological-monitoring reference (Valencell is a well-known ear-based/wrist PPG developer). Used to supply earbud/wearable form-factor and monitoring features.
• Claims implicated: 6, 14, 23 (Ground 4).

4. Other References in the IPR Exhibit Record (context / secondary)

These were submitted with the petition as supporting or background references, not as the core invalidating combination:

• U.S. Patent No. 6,505,133 ("Hanna") — Ex. 1007
• U.S. Patent No. 5,746,206 ("Mannheimer") — Ex. 1008
• U.S. Patent No. 9,596,990 ("Park") — Ex. 1010
• U.S. Patent No. 7,278,966 ("Hjelt") — Ex. 1032
• U.S. Patent Pub. No. 2012/0041767 ("Hoffman") — Ex. 1031
• U.S. Provisional App. No. 61/350,673, "Opticoustic Sensor," Massi Joe E. Kiani, filed June 2, 2010 — cited on the '484 front page
• Non-patent literature: Bronzino, The Biomedical Engineering Handbook (1995) (Ex. 1019); Kranz et al., Pervasive and Mobile Computing (2012); Patel et al., J. Neuroengineering & Rehabilitation (2012); Omre, J. Diabetes Sci. & Tech. (2010); Lister et al., J. Biomedical Optics (2012); Schubert, Light-Emitting Diodes (Cambridge, 2nd ed.)

5. Front-Page "References Cited" (U.S. Patent Documents)

The '484 patent's front page lists a substantial "References Cited" section (U.S. Patent Documents from the 1970s–2000s). Examples I verified in the retrieved front-page text include:

• US 4,063,106 A (12/1977) — Ashkin et al.
• US 4,158,750 A (6/1979) — Sakoe et al.
• US 4,221,997 A (9/1980) — Flemming
• US 6,285,897 B1 (9/2001) — Kilcoyne
• US 6,301,271 B1 and 6,301,273 B1 (10/2001) — Sanders et al.
• US 6,325,978 B1 (12/2001) — Labuda et al.
• US 6,374,006 B1 and 6,381,391 B1 (2002) — Islam et al.
• Also cited: CA 2 458 123 (5/2012) and CN 101849821 B (7/2013)

Limitation: I was unable to retrieve the complete front-page "References Cited" list (the OCR snippets were partial). I therefore cannot give you an exhaustive itemization of every cited U.S. patent document with per-claim anticipation mappings — that data is in the granted patent PDF (available at http://knobbemedical.com/wp-content/uploads/2021/08/US-Patent-No..pdf and via USPTO PatentCenter). The complete list also appears as Exhibit 1082 in PGR2025-00064.

6. Outcome Summary (for accuracy)

• IPR2021-00453 Final Written Decision (Paper 22, Aug. 3, 2022): Held claims 1, 2, 7, 15–23 unpatentable as obvious; declined to hold claims 3–6 and 8–14 unpatentable (because "detecting" ≠ "identifying").
• Fed. Cir. No. 23-1034 (June 21, 2024): Affirmed-in-part, vacated-in-part, remanded — affirmed the claim construction; remanded for the Board to consider Apple's reply-brief obviousness argument.
• Remand decision (Paper 26, Feb. 14, 2025): Held claims 3–6 and 8–14 unpatentable as obvious. Thus, all 23 claims of the '484 patent were held unpatentable by the Board — on § 103 grounds, not § 102 anticipation.
• Appeal: Pending (Fed. Cir. No. 25-1646) as of the most recent record retrieved.

7. Direct Answer to Your § 102 Question

• Most relevant prior art: U.S. Patent No. 9,241,676 B2 (Lisogurski, Covidien LP) — the base reference applied to every claim.
• Which claims "potentially anticipate": Strictly speaking, none of the identified references was used as a standalone § 102 anticipation. They were applied in § 103 combinations (Lisogurski + Carlson + Tran + Isaacson + Valencell-093) that collectively rendered claims 1–23 obvious.
• If you want a § 102 framing, the closest candidate for a single-reference attack is Lisogurski, but the record shows the petitioner needed multiple additional references to reach the '484 wearable/optical/wireless limitations — so a pure anticipation theory is not supported by the retrieved record.

Confidence flags: (a) Full citations for "Tran" and "Isaacson" and the complete front-page reference list were not fully verified in this session — I have deliberately not fabricated numbers for them. (b) The Carlson characterization is limited to its role in the IPR grounds, not a full text review. To close those gaps you would need the granted patent PDF (USPTO PatentCenter) and the IPR2021-00453 petition exhibits.

I'll pull the prior-art references actually cited on this patent's page and the IPR record before building the § 103 analysis.

Good — I have the IPR ground structure. Let me nail down the remaining references and their identities.

Let me pin down how the key SNR limitation was mapped, and check the newest Samsung challenges for additional art.

Obviousness Analysis — US 10,517,484 B2 under 35 U.S.C. § 103

Analyst note on scope and sourcing. The Google Patents "Prior Art" layer for US 10,517,484 surfaces two distinct bodies of art, and I treat them separately below: (i) the ~100+ U.S. references cited on the face of the patent (the "(56) References Cited" list, e.g., Flemming US 4,221,997; Lasar US 4,275,266; Howard US 4,374,618; Tanikawa US 4,403,605; Chance US 4,972,331; Kwoh US 5,078,140; Kilcoyne US 6,285,897; Smart US 6,337,462; Islam US 6,374,006; and the "Prior Art Keywords" — light, signal, output, wearable device, light emitting), and (ii) the far smaller, dispositive cluster that the Office actually adjudicated in IPR2021‑00453. The § 103 analysis that matters is (ii). Where I rely on a source I retrieved, I cite it; where I am inferring, I say so.

Critically, this is not a hypothetical. A § 103 obviousness determination on these exact claims has already been made, twice, and is largely final. I fold that record into the analysis rather than speculating about what might be obvious.

1. The legal frame I am applying

• Graham v. John Deere Co., 383 U.S. 1 (1966): scope/content of the prior art → differences between prior art and claims → level of ordinary skill → secondary considerations.
• KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007): a combination is obvious where the prior art elements were known, the combination was "according to known methods," and it yielded predictable results; an explicit teaching, suggestion, or motivation is not required.
• Rationale (MPEP 2143 / KSR): combining prior-art elements to achieve an improved property or result (here, signal-to-noise ratio) is an enumerated, acceptable rationale — and "SNR improvement" is the stated raison d'être of every reference in the cluster.
• Omni's counter-doctrine (MPEP 2143.01(VI), In re Ratti, 270 F.2d 810, 813 (CCPA 1959)): a proposed modification is improper if it would change the principle of operation of the reference being modified. Omni raised this; it ultimately did not carry the day. This is the single most important contested issue and I address it in § 7.

2. The claims to be analyzed

Claim 1 (reproduced verbatim in the Petition at § III and in the Board's decision, per fedcircuitblog.com/wp-content/uploads/2024/07/Omni-Opinion-Below.pdf) is a three-tier architecture: wrist/ear wearable (LED light source + lens + synchronized detection system + spatially separated detectors + ADC + amplifier) → smartphone/tablet (wireless receiver/transmitter, display, speaker, voice input, buttons/knobs, microprocessor, touch screen) → cloud (receive "output status," process, store, and store a history over a specified period). The invalidity-relevant core is the last three limitations:

• (l) increase SNR (1) by increasing light intensity above an initial intensity and (2) by increasing a pulse rate above an initial pulse rate;
• (m)–(o) change detection: a first signal with the LEDs off, a second signal with at least one LED on, and SNR increased by differencing the two.

Four independent claims: 1, 7, 15, 17. Dependents: 3–6 (from 1), 8–14 (from 7), 16 (from 15), 18–23 (from 17).

Key dependent-claim texture that generated the whole appeal:

• Claims 3 and 8 require a wearable "identify an object" and compare a property of the output signal to a threshold (compare performed by the wearable or the phone/tablet).
• Claim 16 uses the different verb "detect an object." The Federal Circuit affirmed that these are not interchangeable — "identify" = "to recognize or establish an object as being a particular thing."
• Claim 5/13 adds multiple emitters/detectors at varying separations.
• Claim 23 adds a reflective surface positioned to reflect light reflected from the tissue.

3. The prior-art cluster and what each reference teaches

4. The four grounds, and the mapping

The Petition (knobbemedical.com/wp-content/uploads/2021/08/Petition-for-Inter-Partes-Review.pdf) asserted four grounds, all under § 103:

Element-by-element mapping (claim 1):

Gaps I will not paper over: I did not retrieve verbatim text for independent claims 7, 15, 17, nor for claims 2, 4, 9–12, 18–22. I therefore do not map those limitations line-by-line. What the record does establish is that the Board found them all unpatentable (Ground 1 for 1/7/15/17; Ground 2 for the remainder). Likewise, the precise paragraph-level citations for the (m)–(o) change-detection mapping in Ground 1 are not in my retrieved material — I infer Carlson's modulated/off-state sampling plus Lisogurski's off-periods, and flag that as inference.

5. Motivation to combine — the required showing, and the record's answer

A § 103 combination needs an articulated reason with a reasonable expectation of success. For each pairing the record supplies reasons of the strongest KSR type — same field, same problem, known technique, predictable result:

(a) Lisogurski + Carlson. Both are portable/wearable pulse-oximetry devices; the Board has repeatedly characterized them as recognizing the same problem — ambient-light (sunlight) noise — "and the need to offset its negative impact on the signal-to-noise ratio." Lisogurski itself invites the modification: it expressly identifies improving SNR as an objective. Carlson supplies two distinct, immediately applicable techniques: (i) lenses that focus light onto the skin to raise optical signal power per unit electrical power — a predictable, purely additive improvement (Petition: "increase optical power and improve signal to noise ratio," '484 FWD at 23, 26–27); and (ii) pulse-frequency selection/shifting outside the 0–120 Hz ambient band. Combining a known SNR-improvement technique into a device whose stated goal is SNR improvement is the paradigm KSR case; expectation of success is high because the mechanism (moving the signal band away from the noise band) is a well-understood, near-deterministic result.

(b) + Tran. Tran contributes the threshold/alert logic that supplies claims 3/8/16's "compare a property … to a threshold" and the object-identification context. Motivation: Tran addresses the same class of device (monitoring blood-oxygen/health information from a wearable sensor) and the same user-facing purpose (flagging when a measured property crosses a limit). Adding alert thresholds to an existing physiological monitor was routine, and the Petition argued that measuring blood constituents inherently requires identifying/quantifying what is being measured — i.e., the threshold comparison is a small, conventional software increment on Lisogurski's monitor, not a change to the sensor's physics.

(c) + Isaacson. Isaacson's emitter/detector pairs at varying separation add depth-selectivity (shallow vs. deep tissue sampling) — desirable in the exact same wrist/hand/ear geometries Lisogurski contemplates, where skin interference (the '484 patent's own central problem — see its Fig. 9/10 and the differential-probe discussion) dominates. Motivation: depth discrimination directly attacks the interfering skin contribution, i.e., it furthers the same SNR objective shared by Lisogurski and Carlson. Predictable result: greater source-detector spacing → greater sampled depth is a known, monotonic relationship.

(d) + Valencell‑093. Valencell‑093 is a wearable (earbud/wristband/armband/headband) photo-plethysmographic monitor — an analog in structure and application. It supplies two limitations that the others supply only implicitly: (i) raising emitter intensity to increase the ratio of physiological scatter to sunlight (the "increase light intensity" half of limitation (l)), and (ii) the on-state vs. off-state subtraction that maps essentially 1:1 onto the (m)–(o) change-detection limitation. Both are, again, SNR techniques, and the Board found Valencell‑093 "relevant to the subject matter" and sufficient to motivate a POSITA.

Level of ordinary skill. The Petition was supported by the declaration of Brian W. Anthony, PhD (Ex. 1003), and Omni countered with Dr. MacFarlane (Ex. 2122). I did not retrieve the Petition's verbatim statement of the POSITA definition, so I will not fabricate a degree-years formulation. The Board evidently applied a definition under which the mapping above was cognizable.

6. What Omni argued, and why it ultimately failed

1. "Neither Lisogurski nor Carlson discloses increasing SNR by increasing a pulse rate." Omni's core reading was that Carlson teaches modulating at a single, fixed carrier frequency (f_c, e.g., 1000 Hz), not increasing a rate, and that Lisogurski's cardiac-cycle-modulation tracks the heartbeat (0.5–3 Hz) for reasons unrelated to SNR. This argument won the day only in the narrow, pre-Axonics sense — and was defeated on the merits of the increasing limitation by (a) Lisogurski's own text that decreasing the "off" periods = increasing firing rate "relates to an increased sampling rate" and yields "more accurate and reliable physiological information," and (b) the admission of Omni's own expert in a related proceeding that "generally speaking, the faster the modulation, the faster the pulse rate, the lower the background noise" (Ex. 1060, 37:13–38:3).
2. "Change of principle of operation" (MPEP 2143.01(VI); In re Ratti). Omni argued on appeal from the parent '533 patent that driving Lisogurski's LED from 1 Hz CCM to 1000 Hz "completely changes the principle of operation." The rebuttal, which carried: Lisogurski itself expressly teaches drive-cycle modulation (DCM) at 1,000 Hz and even simultaneous CCM + DCM (Ex. 1011, 25:58‑65, 35:10‑30, 37:18‑22) — so raising the firing rate to ~1,000 Hz is within Lisogurski's own disclosure and does not alter its principle of operation.
3. Hindsight / selective excision ("components selectively culled from … the monitor" to build a "wearable"/"personal device"). This is the classic Cheese Systems / anti-hindsight refrain. It failed because the trend toward integration of front-end circuitry into wearables was itself known, and because Lisogurski's sensor unit is already placed on the body.
4. The one argument that actually worked — and it was a claim-construction, not prior-art, argument. Omni proposed "identify an object" = "to recognize or establish an object as being a particular thing." The Board adopted it; the Federal Circuit affirmed it (holding the presumption that "identify" ≠ "detect" was unrebutted). On that construction, the Board's original FWD found claims 3–6 and 8–14 not proven unpatentable, because the Petition's art (Lisogurski's sensor-fall-off error signal; Tran's alert thresholds) showed detection of presence, not identification of what the object is.

7. Procedural turn that restored the § 103 rejection of the remaining claims

The Federal Circuit found the Board abused its discretion under Axonics, Inc. v. Medtronic, Inc., 75 F.4th 1374 (Fed. Cir. 2023), in refusing to consider Apple's reply-brief alternative argument — which was made in response to Omni's newly raised construction. Axonics holds that where a patent owner first proposes a construction in its POR, the petitioner "must be given the opportunity in its reply to argue and present evidence of anticipation or obviousness under the new construction," at least where it relies on the same embodiments. The court affirmed-in-part, vacated-in-part, and remanded (Fed. Cir. No. 2023‑1034, June 21, 2024; opinion authored by Judge Albright sitting by designation; https://storage.courtlistener.com/pdf/2024/06/21/apple_inc._v._omni_medsci_inc..pdf).

On remand (Paper 26, Feb. 14, 2025), the Board found claims 3–6 and 8–14 unpatentable — i.e., claims 1–23 are unpatentable.

8. Bottom-line § 103 conclusion, and how final it is

• Claims 1, 2, 7, and 15–23: held unpatentable in the original FWD (Aug. 3, 2022). Omni's Notice of Appeal (dated April 11, 2025) identifies only claims 3–6 and 8–14 as the issues on appeal. Because the earlier adverse determination as to claims 1, 2, 7, 15–23 was not appealed and Omni did not cross‑appeal the relevant finding, those claims are, for practical purposes, finally and non‑appealably unpatentable. (The Petition/Reply record confirms Omni "did not cross-appeal" Lisogurski's teaching of the increasing-SNR limitation.)
• Claims 3–6 and 8–14: found unpatentable on remand (Feb. 14, 2025); on appeal at the Federal Circuit (Google Patents lists No. 25‑1646). The live dispute has narrowed to a single limitation — "identify an object" across those claims (per WHOOP's parallel petition: "Omni's appeal as to claims 3–6 and 8–14 of the '484 patent remains pending, and involves just one limitation across those claims").
• Ground-level conclusion: Every claim is obvious over Lisogurski + Carlson, alone or in further view of Tran, Isaacson, and/or Valencell‑093, because each reference is drawn from the same field (wearable optical physiological monitoring / pulse oximetry), addresses the same problem (ambient-light and motion noise degrading SNR), and supplies a known, predictable SNR-improving technique — the precise rationale MPEP 2143(A)(1) and KSR endorse. Notably, Omni did not dispute that the references disclosed nearly all limitations, and did not challenge motivation to combine as to the majority of them (a point successive petitioners pressed as near-concession).

9. Caveats, uncertainties, and one flagged discord

• Tran's patent number is unverified in this session. The exhibit-list snippet I retrieved truncates at "1064 | U.S. …". The prior generated section states Tran = US 8,108,036; I could not independently confirm that number here, so treat it as unconfirmed rather than adopting it silently.
• Newer, parallel art. Later Samsung-side challenges (IPR2025‑01249/01250/01252/01253; PGR2025‑00063/64) assert Debreczeny (a pulse-oximeter reference with spatially symmetric, additive red LEDs and a "spatial symmetry"/motion-robustness teaching) against at least claim 3, in a ground styled "Lisogurski in view of Carlson and Debreczeny." That is a different combination from IPR2021‑00453's Grounds 1–4 — but it is aimed at the same surviving claims, and it reinforces the obviousness of the "identify an object"/threshold-adjacent subject matter. Other exhibits in the '484 record include Hanna US 6,505,133 (Ex. 1007) and Mannheimer US 5,746,206 (Ex. 1008). I flag Debreczeny as sourced from the 2025 petition strand, not from IPR2021‑00453.
• Contradiction to flag (repeating and sharpening the prior section): Google Patents labels the patent "Active," with an anticipated expiration of 2033‑12‑17 and a 2026 E.D. Tex. filing (2:26‑cv‑00289). That "Active" flag is a status field, not an enforceability assessment, and it is in tension with the Board's determination that claims 1–23 are unpatentable (now largely final as to claims 1, 2, 7, 15‑23). Do not read "Active" as "valid."
• What I did not verify: verbatim text of claims 7, 15, 17 and of all dependents except as quoted above; the Petition's exact POSITA definition; and paragraph-level citations for the (m)–(o) change-detection mapping in Ground 1. I have marked each of these as a gap rather than reconstructing it.

Sources: Google Patents, https://patents.google.com/patent/US10517484/en; Apple IPR2021‑00453 Petition, https://knobbemedical.com/wp-content/uploads/2021/08/Petition-for-Inter-Partes-Review.pdf; Fed. Cir. slip op. No. 2023‑1034, https://storage.courtlistener.com/pdf/2024/06/21/apple_inc._v._omni_medsci_inc..pdf; Board decision below, http://fedcircuitblog.com/wp-content/uploads/2024/07/Omni-Opinion-Below.pdf; Patent Owner's Notice of Appeal (Apr. 11, 2025), https://www.docketalarm.com/cases/[PTAB](/ptab)/IPR2025-01253/SAMSUNG_ELECTRONICS_CO._LTD._v._OMNI_MEDSCI_INC/docs/10-14-2025-Patent_Owner/Exhibit-2012-Apple_Inc_v_Omni_Medsci,_Inc,_No_IPR2021_00453,_Paper_27.pdf; PGR2025‑00063 Ex. 1091, https://www.docketalarm.com/cases/PTAB/PGR2025-00063/SAMSUNG_ELECTRONICS_CO._LTD_v._OMNI_MEDSCI_INC/11-12-2025-Petitioner/Exhibit-1091-Ex_1091___484_notice_of_appeal/; WHOOP '304 Petition, https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1558539](/patent/1558539)/...