US 12251320

I'll search for authoritative information on US 12251320 from USPTO and litigation sources.

Patent Identification: US 12,251,320

Literal identifier check: Searched for 12251320 / "12,251,320" as an exact string. Results consistently resolve to US 12,251,320 B2. No results surfaced under any near-neighbor number; the confirmation below is based on the exact number only.

Bibliographic data

Sources: the patent text supplied in your message (authoritative), cross-checked against HT Syndication's USPTO issue notice ("United States Patent no. 12,251,320, issued on March 18, was assigned to Carlsmed Inc. (Carlsbad, Calif.)… invented by Niall Patrick Casey… and Michael J. Cordonnier") and the Google Patents continuity data.

Abstract (as issued)

"A personalized fixation system includes a surgical planning software tool configured to adjust relationships of relevant anatomy of a subject, at least one bone anchor, a plate having a shape that does not conform to a single plane, the plate configured to accept the at least one bone anchor, wherein the shape of the plate is at least partially determined by the surgical planning software tool, and wherein the plate includes at least one node having a hole configured to receive the at least one bone anchor, and a locking element configured to connect the at least one bone anchor to the plate, wherein the plate is manufactured using additive manufacturing."

Important observation: There is a mismatch between the abstract and the granted claims. The abstract (and the specification's SUMMARY) is directed to a fixation hardware system — non-planar additive-manufactured plate, nodes, bone anchors, locking element. The granted independent claims, however, are computer-implemented design-method claims reciting spinopelvic predictive guidelines. Anyone construing scope should rely on the claims, not the abstract.

Independent claims — plain language

Claim 1 — Computer-implemented method of designing patient-specific spinal implants. Steps:

1. Generate a 3D virtual model of at least a portion of the patient's spinal anatomy.
2. Receive user-proposed changes to that anatomy, specifically including changes to the spatial relationship between at least two vertebrae.
3. Adjust the virtual anatomy to reflect those changes.
4. Measure the adjusted anatomy to obtain values for at least one of: pelvic tilt, sagittal alignment, or lumbar lordosis.
5. Compare those measurements against "predictive guidelines" — predetermined target values for pelvic tilt, sagittal alignment, or lumbar lordosis — to decide whether the adjusted anatomy conforms.
6. If it conforms, design patient-specific implant(s) to achieve the corrected spinal geometry when implanted.

The novelty center of gravity is step 5/6: the software gate that only proceeds to implant design once a virtual correction satisfies pre-set spinopelvic parameter thresholds.

Claim 9 — Computer-implemented method of designing patient-specific spinal implants. Essentially the same pipeline as claim 1 (3D model → receive proposed changes → adjust → measure pelvic tilt/sagittal alignment/lumbar lordosis → compare to predictive guidelines → design implants), with two differences visible:

• The "receiving" step is recited without the claim 1 limitation that the changes include a change to the spatial relationship between at least two vertebrae (broader on that point).
• The final limitation adds that the patient-specific implants include a spinal implant and a bone anchor configured to anchor the spinal implant to a bony structure.

Caveat on scope: The claim set supplied to me is truncated mid-claim-9 ("…a bone anchor con…"), even though the patent states 23 claims. I can confirm claims 1 and 9 are independent, but I cannot rule out one or more further independent claims (e.g., apparatus/system claims) beyond claim 9. Treat claims 10–23 as unverified from my sources.

Dependent claims 2–8 (all depend from claim 1)

• 2: Pelvic tilt measured; guideline = pelvic tilt < 20°.
• 3: Measurement relates to sagittal alignment.
• 4: Guideline = sagittal vertebral axis (SVA) ≤ 5 cm.
• 5: Guideline includes a lordosis value.
• 6: Lordosis value tied to a subset of vertebral levels.
• 7: Measurements include lumbar lordosis and pelvic incidence; guideline = |pelvic incidence − lumbar lordosis| < 10°.

(These mirror the specification's three predictive correction guidelines at ¶ step 208, credited to Pimenta, "Current Surgical Strategies to Restore Proper Sagittal Alignment," J. Spine 2015, 4(4), incorporated by reference.)

• 8: The patient-specific implants include a spinal implant plus a bone anchor configured to anchor it to bone.

Litigation review (USPTO / CAFC)

• CAFC 2026 dockets: I found no Federal Circuit docket, appeal, or decision naming US 12,251,320. Targeted searches for "12251320" CAFC docket 2026 returned zero results. I cannot affirmatively state no such docket exists — only that nothing authoritative surfaced.
• PTAB: No IPR/PGR/CBM proceeding involving this number surfaced in my searches.
• District court: Carlsmed, Inc. filed a patent infringement suit against DeGen Medical, Inc. in the U.S. District Court for the District of South Carolina, Florence Division, announced September 15, 2026, asserting eight Carlsmed patents "relating to its implants, methods and systems," seeking injunctive relief and damages (GlobeNewswire release; ODT Magazine). The public reporting does not identify which eight patent numbers are asserted, so I cannot confirm or deny that US 12,251,320 is among them. Given the '320 claims are method claims on virtual surgical planning and the press release language ("implants, methods and systems"), it is plausible but unverified.

⚠️ Date discrepancy, flagged per operating rules: your message states the current date is April 26, 2026, but my session date is September 17, 2026, and the DeGen filing above is dated September 15, 2026 — i.e., after the date in your prompt. I have reported the litigation as found rather than suppressing it, but you should reconcile which date governs your analysis.

Related family members (for context, exact numbers as reported)

• Ser. No. 16/383,215 → US 11,439,514 (parent of the '320).
• Ser. No. 16/352,699 → US 11,432,943 (interbody implant with deployable tooth — different disclosure).
• Ser. No. 18/754,101 → US 12,245,952 (continuation, issued March 11, 2025).
• Ser. No. 19/041,891 → US 2025/0235324 A1 (continuation published July 24, 2025).
• Ser. No. 17/875,699 → US 12,491,085 (issued December 9, 2025).

Confidence statement

High confidence on title, number, application number, inventors, assignee, filing/priority/issue dates, and the text of claims 1–9. Low/undetermined on claims 10–23 and on whether this specific patent is asserted in the Carlsmed v. DeGen Medical action. Any legal-status characterization (e.g., "Active," expiration date) is a USPTO/aggregator assumption, not a legal conclusion.

I'll search litigation-specific sources for this exact patent number.

Let me search the litigation databases directly for this patent number.

Let me search for the specific case details and any other litigation naming this patent.

Litigation search: US 12,251,320 ("Systems and methods for orthopedic implant fixation")

Bottom line: I found no litigation that names US 12,251,320 by its specific number. No PACER docket, no Unified Patents Portal litigation entry, no Federal Circuit appeal, and no PTAB proceeding referencing the '320 patent surfaced in my searches. What follows is what I can and cannot establish, with the uncertainty made explicit.

1. Direct number searches — negative

Searches run against the exact identifiers (never a near-neighbor number):

Per your strict rule, I did not substitute results for similar numbers (e.g., US 12,251,320 vs. the family members 11,439,514 / 12,245,952 / 12,491,085). Those family members are distinct patents and are not evidence of litigation involving the '320.

2. The one identified lawsuit — but the '320 is not confirmed as asserted

Carlsmed, Inc. v. DeGen Medical, Inc.

Sources: GlobeNewswire via Stockhouse; ODT Magazine; FinancialContent/GlobeNewswire; MT Newswires via Bitget; Medical Buyer.

Critically: none of these sources identifies which eight patents are in suit. Because the '320's claims are computer-implemented design-method claims on virtual surgical planning, while the press release describes "implants, methods and systems," it is plausible the '320 is among the eight — but that is inference, not fact. I cannot confirm or deny that US 12,251,320 is asserted.

The complaint itself (the only authoritative source for the asserted numbers) was not retrievable — PACER/CM-ECF for D.S.C. is paywalled and no free docket mirror had indexed the case at the time of search.

3. No other known litigation

• Federal Circuit: no appeal or decision naming the '320 found.
• PTAB: no IPR, PGR, or CBM petition naming the '320 found.
• Other district courts: no other action found. The press coverage characterizes the DeGen suit as "the first legal action of this kind disclosed by Carlsmed" (Wiseek summary of GlobeNewswire), which undercuts the existence of earlier suits by the patentee.

4. ⚠️ Date discrepancy — must be flagged

Your prompt states the current date is April 26, 2026. The DeGen suit is dated September 15, 2026 — after your stated date — and my session date is September 17, 2026. This matches the discrepancy flagged in the earlier summary section of this analysis. I have reported the litigation as found rather than suppressing it, but you must reconcile which date governs. If April 26, 2026 is truly the operative date, then the DeGen filing did not yet exist on your timeline, and as of that date no litigation involving US 12,251,320 would be known.

5. Confidence statement

• High confidence: No litigation naming US 12,251,320 by number is publicly discoverable; the Carlsmed v. DeGen Medical action exists as described.
• Low / undetermined: whether the '320 is one of the eight asserted patents; the case number; the ultimate outcome.
• Explicit non-findings: I could not search PACER directly, and the Unified Patents Portal caselist was not observed to contain a '320 entry — absence of a result is not affirmative proof that no docket exists.
• Any litigation status is a point-in-time observation, not a legal conclusion.

I'll search for any PTAB proceedings involving this patent or its family.

PTAB proceedings on file — US 12,251,320 ("Systems and methods for orthopedic implant fixation")

Source of the proceedings list: the canonical USPTO Open Data Portal (ODP) block supplied in this prompt — "The USPTO ODP API returns no AIA trial proceedings for this patent as of the most recent ingest." I then ran independent web searches to catch anything ODP has not yet indexed ("12251320" PTAB inter partes review petition; Carlsmed IPR petition PTAB inter partes review; "Carlsmed" PTAB IPR petition instituted; PTAB petition Carlsmed "11,439,514" OR "12,245,952" OR "12,491,085"; DeGen Medical IPR petition Carlsmed patent PTAB 2026; Unified Patents Carlsmed spine patent challenge). None returned a PTAB proceeding of any kind. The only hits were the 2026-09-15 Carlsmed v. DeGen Medical complaint press coverage.

Proceedings overview

Total AIA trial proceedings on US 12,251,320: 0. Breakdown: 0 active, 0 with claims invalidated, 0 with claims sustained, 0 settled, 0 institution denials — there is no IPR, PGR, or CBM to categorize. The bottom-line defensive posture is therefore neither "hardened" nor "dead": the patent is completely untested at the PTAB. Its claims carry no PTAB-driven narrowing and no § 315(e)(2) estoppel, which means a defendant retains the full universe of § 102/§ 103 art — but also means there is no free kill; asserting invalidity will require building the record from scratch.

Before drawing conclusions from the zero, three facts matter:

1. The patent is recent. Granted 2025-03-18. As of 2026-09-17 it is ~18 months old.
2. Assertion is brand-new. Carlsmed's complaint against DeGen Medical was filed 2026-09-15 in D.S.C. (Florence Division), asserting eight Carlsmed patents; the press release does not identify which eight, so I cannot confirm '320 is among them. (GlobeNewswire via Barchart)
3. Pre-assertion IPRs are rare. Absent a competitive threat, third parties usually have no reason to spend $300k+ challenging a patent in this space. A zero count in the first ~18 months of a patent's life is unremarkable; it is not yet a meaningful "well-asserted patents eventually attract IPRs" signal.

Proceedings — none

No proceeding numbers to report. I will not manufacture them. If you need to double-confirm independently, the authoritative checks are:

• USPTO PTAB E2E / PTAB Decisions: https://developer.uspto.gov/ptab-api or https://www.uspto.gov/patents/patent-trial-and-appeal-board
• Unified Patents' free PTAB tracker (keyword "Carlsmed")
• CourtListener RECAP for any D. Del./D.S.C. docket that might name '320

Discrepancy flag (carried forward): the previously generated section flagged that the prompt initially stated a current date of 2026-04-26 while the session date is 2026-09-17, and that the DeGen filing (2026-09-15) post-dates that earlier prompt date. My analysis here runs off 2026-09-17. All statutory deadline math below is keyed to that date; if a different date governs, the § 315(b) and PGR figures shift accordingly.

Strategic summary

Claim status: every one of the 23 claims is UNTESTED. Nothing is canceled, nothing is confirmed. Claim 1 (the virtual-planning/predictive-guideline method) and claim 9 (the same pipeline plus "a spinal implant and a bone anchor") are the independent claims identified in the prior section; claims 2–8 depend from claim 1 and recite the spinopelvic thresholds (pelvic tilt < 20°, SVA ≤ 5 cm, |PI − LL| < 10°). There is no FWD to quote and no claim-level disposition to report — do not let anyone tell you otherwise. Whether '320 has further independent claims beyond claim 9 remains unverified from my sources, because the claim text available to me is truncated mid-claim-9.

Estoppel landscape: empty. With zero IPRs, § 315(e)(2) estoppel has never attached to anyone. A defendant today faces no estoppel bar and can raise any § 102/§ 103 ground in district court and in a later IPR. Conversely, there is no prior PTAB record (no institution decision, no FWD, no claim construction) to leverage as persuasive authority. Two adjacent constraints are worth knowing:

• § 325(d) discretion remains a live risk. Because the examiner presumably considered the Pimenta reference ("Current Surgical Strategies to Restore Proper Sagittal Alignment," J. Spine 2015, 4(4)) — it is incorporated by reference in the spec and supplies the three numerical guidelines — an IPR built on that reference invites discretionary denial as "substantially the same art previously presented to the Office."
• Trial-vehicle limits. IPR under § 311(b) reaches only §§ 102/103 on patents and printed publications. That is a real problem here, because the claims are computer-implemented method claims whose most obvious attack surface — patent-ineligible abstract idea / mental process under § 101 — cannot be raised in an IPR at all. Note also that any priority/§ 112 challenge to the 2018-04-16 priority claim (the '320 is a continuation of Ser. No. 16/383,215, now US 11,439,514, which claims benefit of provisional 62/658,389) is likewise unavailable in IPR, and the PGR window has closed (see timing below), so those theories now live only in district court.

Pattern signals: none. No repeat petitioner, no Federal Circuit appeal, no defensive aggregator (Unified Patents, RPX, etc.) on this patent. Carlsmed is the original assignee and sole owner of record (assignment recorded 2025-03-11). The company went public on Nasdaq in July 2025 and appointed a Chief Legal Officer in September 2025 — an enforcement posture consistent with the September 2026 complaint, but it has so far produced litigation, not PTAB activity.

Timing math for a defendant served with the DeGen complaint (filed 2026-09-15):

Recommended next steps

1. Proceed on the premise that no PTAB proceeding exists. If you are advising DeGen or another target, do not budget for reading an FWD — there isn't one. Budget for filing the first IPR instead.
2. Confirm '320 is actually in the complaint before spending on it. The 2026-09-15 press release says "eight of Carlsmed's patents" without numbers. Pull the D.S.C. Florence Division docket and the complaint (CourtListener RECAP) and identify the asserted patents. Because the '320 independent claims are method claims, they read on a surgeon's or planning service's process, not on a shipped implant — evaluate whether Carlsmed's infringement theory against DeGen even reaches them, versus the apparatus family members (US 11,439,514; US 12,245,952; US 12,491,085; US 2025/0235324 A1).
3. Treat the family, not the patent, as the target. Invalidating '320 alone would not clear the field given the sibling continuations. A coordinated IPR set (with § 315(b) dates all running from the same service date) is the rational structure.
4. Build the IPR on art outside the Pimenta reference to avoid a § 325(d) discretionary denial, and prioritize art that predates 2018-04-16 (bone-anchor/plate and patient-specific-implant-screw-placement art from the 2000s–2010s), where these claims' substantive novelty is thin.
5. Preserve the § 101 attack for district court. Read the claims on the pre-emption/abstract-idea axis (a generic computer performing measurement-and-comparison of spinopelvic parameters against numeric thresholds) and keep that in your invalidity contentions; it is foreclosed at the PTAB.
6. Diarize the deadlines. IPR target filing on or before ~2027-09-15 (from actual service date), and monitor for any first IPR by a third party, which opens § 315(c) joinder as an alternative route if your own window is at risk.
7. Re-run the PTAB check quarterly. This is a moving target: the September 2026 complaint is the kind of event that generates an IPR within 6–12 months. A zero today does not mean a zero in Q3 2027.

Confidence: High that no AIA trial proceeding on US 12,251,320 is listed in ODP or surfaced in targeted web search as of 2026-09-17. This is an absence of evidence conclusion — I cannot prove a negative, and a very recently filed, unindexed petition could exist. Low/unverified on whether '320 is among the eight asserted Carlsmed patents, and on the content of claims 10–23.

I'll research the assignment record for US 12,251,320 before writing the section.

I could not retrieve the Assignment Center record directly through my tools — targeted queries for the reel/frame and correspondent returned nothing, and my only usable assignment data is the Google Patents legal-events entry plus the shell of the USPTO record it indexes. I state below exactly what is evidenced and what is not. No reel/frame numbers or correspondent names are available to me for this patent, so I have not invented any.

Inventors

• Both listed addresses are Carlsbad, CA, matching Carlsmed's principal executive offices at 1800 Aston Ave, Suite 100, Carlsbad, California 92008.
• Departure pattern — not present. Neither inventor left the assignee. Cordonnier signed the post-IPO Form 8-K on July 24, 2025 as CEO and President, and was reported pre-IPO as holding 7.9% of shares. There is no "inventors departed within 12 months" precursor here; the inventors and the assignee are the same two people plus their company.
• One evidentiary limitation flagged: Casey's precise corporate title at the time of filing is not verified from the material I retrieved. His co-founder/employee status with Carlsmed is consistent but I cannot cite a source for the exact role. Treat that cell as likely, unconfirmed.

Original assignee

Carlsmed, Inc. — a Delaware corporation, principal executive offices at 1800 Aston Ave, Suite 100, Carlsbad, California 92008; IRS EIN 83-1081863; Commission File No. 001-XXXXX (CIK 0001794546).

• Line of business: commercial-stage medical technology — the aprevo Technology Platform, comprising AI-enabled pre-operative planning software (myaprevo), patient-specific interbody implants, and single-use instruments. Founded 2018.
• Product embodying the claims — yes, but read the mismatch. The '320 abstract and specification SUMMARY describe fixation hardware (non-planar additively-manufactured plate, nodes, bone anchor, locking element). The granted independent claims 1 and 9, however, are computer-implemented design methods reciting virtual anatomical correction gated on spinopelvic predictive guidelines. The commercial embodiment of the claims is therefore the aprevo/myaprevo surgical-planning software, not the plate. The S-1 lists the "aprevo® technology platform, including myaprevo® application" as the product mapped to the family that includes the '514 parent. This mismatch was flagged in the prior section and stands.
• Current status: operating and public. IPO priced July 22, 2025; shares began trading on Nasdaq Global Select Market under "CARL" on July 23–24, 2025 at $14.00–$16.00/share, raising up to ~$107.2M. Largest pre-IPO holders: B Capital Group (35.1%) and US Venture Partners affiliates (28.3%).
• Portfolio self-report: the FY2025 Form 10-K states: "We solely own all of the patents and patent applications in our portfolio. As of December 31, 2025, our patent portfolio contained 45 total issued patents and approximately 120 pending patent applications…" This is the single most probative statement in this analysis — the assignee affirmatively disclaims any outstanding co-ownership, outbound license, or recorded security interest in the portfolio as of year-end 2025.

Assignment timeline

The only assignment event I can evidence for US 12,251,320 is:

• Execution date: not established / recorded 2025-03-11 — Reel/Frame NOT retrieved ⚠️
  • Conveyance: Assignment (USPTO/Google Patents event text: "ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)")
  • Assignor: CORDONNIER, MICHAEL J.; CASEY, NIALL PATRICK
  • Assignee: CARLSMED, INC.
  • Correspondent: not retrievable from my available sources — I will not guess a name or firm.
  • Context: Inbound founder/inventor-to-company assignment, almost certainly a confirmatory (nunc pro tunc) recordation tied to IPO diligence — inferred from timing, not evidenced. Recording occurred 7 days before the March 18, 2025 grant and roughly four months before the S-1/A. ⚠️ This execution-date inference is mine, not a record fact. The underlying agreement may well date to 2018 or 2019; the recordation is what is dated 2025-03-11.

No other assignments, security agreements, mergers, changes of name, licenses, or releases were surfaced for this patent. I found no evidence of:

• any outbound transfer from Carlsmed to any third party,
• any security interest recorded against the '320 (note: venture-debt lenders commonly record security agreements against patent portfolios; none surfaced here, and the 10-K's "solely own" language is inconsistent with a recorded assignment-style security interest),
• any assignment recorded against the parent US 11,439,514 beyond this same founder assignment.

⚠️ Two honest gaps:

1. I could not open assignmentcenter.uspto.gov or assignment.uspto.gov record pages through my tools; my queries returned empty result sets. The 2025-03-11 event and its assignor/assignee names come from the Google Patents legal-events block, which parses USPTO Assignment data but does not expose reel/frame or correspondent.
2. Because the record set is thin, I cannot rule out additional recordings that exist in Assignment Center but were not indexed in the sources I could reach. Anyone relying on this chain-of-title analysis for standing purposes should pull the record directly at assignmentcenter.uspto.gov and assignment.uspto.gov by patent number 12251320.

Timeline diagram

timeline
    title Ownership of US 12251320
    2018 : Provisional 62658389 filed
         : Named inventors Casey and Cordonnier
    2019 : Parent application 16383215 filed
    2022 : Continuation 17880277 filed
         : Parent issues as US 11439514
    2025 : Founder assignment recorded to Carlsmed
         : US 12251320 issues March 18
         : Carlsmed IPO begins trading on Nasdaq
    2026 : Carlsmed sues DeGen Medical

NPE / troll-pattern signals

Verdict

Operating-company assertion.

The chain of title is a single link — an inventor-to-Carlsmed assignment recorded 2025-03-11 naming assignors CORDONNIER, MICHAEL J. and CASEY, NIALL PATRICK — and the assignee retained the patent rather than transferring it. Carlsmed, Inc. is a Nasdaq-listed, 510(k)-cleared device maker (ticker CARL) whose FY2025 10-K represents that it "solely own[s] all of the patents and patent applications in our portfolio," and it is the plaintiff in Carlsmed, Inc. v. DeGen Medical, Inc. (D.S.C., announced September 15, 2026) — i.e., an operating company suing a competitor. No NPE signal beyond the underlying founder assignment is present, and that one is inbound, not outbound. The absence of any recorded post-issuance assignment is itself the finding: the original assignee still owns the patent.

Verify the record at: https://assignmentcenter.uspto.gov/ and https://assignment.uspto.gov/patent/index.html — search patent number 12251320.

Confidence: High on the presence and substance of the 2025-03-11 inventor→Carlsmed recording, and high on the operating-company characterization. Low on reel/frame, execution date, and correspondent of record — those fields are unsourced and I have deliberately left them blank rather than estimated. Whether US 12,251,320 is among the eight patents asserted in Carlsmed v. DeGen Medical remains unverified, as noted in the prior section.

I'll search for the patent's front-page citations and prior-art references.

Let me pull the front-page patent citations and related prior-art references.

Let me try PubChem and FreePatentsOnline, which list front-page citations for this patent.

Prior-Art Analysis — US 12,251,320 B2 ("Systems and methods for orthopedic implant fixation")

Bottom line up front: I was unable to retrieve the certified "References Cited" (front-page citation) list for US 12,251,320. The patent text you supplied as authoritative contains no citation section (Google Patents' rendered text for this document stops at the Classifications/Definitions and omits "Patent Citations" / "Non-Patent Citations"), and my searches for that specific section returned nothing for the '320 itself. Rather than fabricate a citation list, I have separated what is verified from what is inferential, and I flag the gap explicitly. I did not substitute near-neighbor numbers — note two live traps: US 12,251,321 is a Globus Medical "Expandable intervertebral implant" (unrelated), and US 12,251,313 is a Carlsmed patent ("Systems and methods for orthopedic implants") that is a different patent from the '320.

Part I — References that are cited within the '320's own authoritative text (high confidence)

The only references I can attribute to the '320 with certainty are those appearing in its CROSS-REFERENCE and DETAILED DESCRIPTION sections. These are the references on which any §102 analysis must start, because two of them are affirmatively incorporated by reference into the patent's disclosure.

Point of substance (not just form): the specification expressly credits the three predictive-correction guidelines to Pimenta and incorporates it "in its entirety." By the applicant's own words, the pelvic tilt < 20°, SVA ≤ 5 cm, and |PI − LL| < 10° thresholds are known prior art. That is a self-inflicted §102/§103 concession that goes directly to dependent claims 2, 4, and 7.

Also relevant as admissions (not "cited references"): FIGS. 1–5 are expressly labelled "prior art" — polyaxial pedicle screw + rod constructs (FIGS. 1–2) and posterior plate + screw + nut + washer constructs (FIGS. 3–5). The '320 specification characterizes these as the existing state of the art. These figures are admissions against the apparatus subject matter in the SUMMARY/abstract (which is not what issued).

Part II — Front-page patent citations (proxy, flagged)

Because the '320's own front page is missing from the authoritative text, the closest verifiable citation list I could obtain is for the parent, US 11,439,514 B2, via PubChem's patent-citation listing. I am labelling this a proxy, not the '320's list. Continuation applications share a specification and are normally searched by the same art unit, so overlap is likely high — but it is not proof, and the '514's issued claims (fixation apparatus) differ in focus from the '320's issued claims (virtual surgical-planning methods), so the examiner's citation emphasis would differ.

Partial list surfaced (numbers as reported, not auto-corrected; dates supplied from general knowledge and not verified in this session):

• US 4,704,686 A (Aldinger) — ~1987 — process for producing an individually tailored bone prosthesis/implant. → §102(a)(1) pre-AIA §102(b) art; goes to "patient-specific implant designed from imaging data."
• US 4,936,862 A — ~1990 — method of making a custom prosthetic device. → same thrust.
• US 5,431,562 A (Andreiko et al.) — ~1995 — method/apparatus for designing and forming a customized appliance from scan data (orthodontic). → computer-aided custom-medical-device design.
• US 6,772,026 B2 — ~2004 — system/method for rapid design, prototyping and manufacture of a medical device.
• US 6,932,842 B1 (Steinberg / "custom implant") — ~2005 — method for producing a patient-specific implant.
• US 6,978,188 B1 — ~2005 — patient-specific surgical planning/implants.
• US 7,174,282 B2 — ~2007 — computer-aided design methodology for implant/scaffold structures.
• US 7,187,780 B2 — ~2007 — data processing and visualization for surgical/diagnostic planning.
• US 7,747,305 B2 — ~2010 — design of custom orthopedic implants from image data.
• US 8,337,507 B2 and US 8,556,983 B2 (Conformis) — 2012 / 2013 — patient-adapted joint arthroplasty implants and surgical tools designed from patient image data.
• WO 2004/110309 A2; US 2005/049590 A1; US 2005/271996 A1; US 2006/009780 A1; US 2002/007294 A1; US 2004/171924 A1; US 2010/191088 A1; US 2012/010710 A1; etc.

Do not over-read this table. These are design/imaging-into-patient-specific-implant references. They are largely cumulative against the abstract/SUMMARY subject matter (non-planar, additively manufactured plate) — which is not what issued. Against the granted method claims 1–9, this cluster is weak: none of them is understood to teach the claim-1/claim-9 predictive-guideline gate (measure pelvic tilt / sagittal alignment / lumbar lordosis → compare to predetermined target values → then design the patient-specific implant).

Part III — §102 mapping against the issued claims

The issued independent claims are computer-implemented design-method claims (claim 1 and claim 9), not apparatus claims. Any §102 attack must therefore target: (i) 3D virtual spinal model, (ii) user-proposed change in the spatial relationship between at least two vertebrae, (iii) adjusted model, (iv) measurement of pelvic tilt / sagittal alignment / lumbar lordosis, (v) comparison to "predictive guidelines" with predetermined values, (vi) design of a patient-specific implant on conformance.

Additional statutory notes:

• Effective filing date drives everything. Claims 1–9 will assert 2018-04-16 (provisional 62/658,389). If the spinopelvic-guideline subject matter is not supported by that provisional but only by the 2019-04-12 parent, the effective date slips to 2019-04-12, which would open a one-year window (Apr 2018–Apr 2019) of §102(a)(1)/(a)(2) art — including potentially published third-party spinal-planning applications. This is the single highest-value thing to verify next, but it requires the provisional's contents, which I could not obtain.
• §102 ineligibility by disclosure: if any of the recited guideline mathematics had been the applicant's own pre-Apr-2018 public disclosure, §102(b)(1)(A) grace-period exception might apply. No evidence of such a disclosure surfaced.
• §103 overlay: even where no single reference anticipates, Pimenta (numeric sagittal targets) in view of any of the Medicrea/planning-software leads is a plausible §103 combination for claims 1 and 9.

Part IV — What I could not do, stated plainly

1. I could not query USPTO's own systems directly. PatentCenter and the USPTO full-text search are script/JS-gated and were not usable from this session. Everything above is from the authoritative patent text you supplied, cross-checked against Google Patents, FreePatentsOnline, and PubChem listing pages.
2. The '320's front-page citation list was not retrievable and is not present in the supplied text. Part II is a proxy from the parent '514 and is labelled as such. Any reference‑by‑reference §102 table purporting to be "the citations of US 12,251,320" would be fabricated; I have not produced one.
3. The NPL citation set beyond Pimenta is unknown. Examiners in this art unit frequently cite the SRS‑Schwab adult‑spinal‑deformity classification literature and Surgimap-type planning publications. I saw no evidence of those on the '320's face and will not assert them.
4. No litigation prior art (IPR/PGR invalidity grounds) exists to mine, since, as established in the earlier sections, no PTAB proceeding naming the '320 was found.

Part V — Cross-reference contradictions I am flagging (per instructions)

1. Date discrepancy (re-flagged). The task header states the current date is April 26, 2026; my session date is September 17, 2026, and the Carlsmed v. DeGen Medical complaint is dated September 15, 2026. Reconcile which date governs.
2. US 11,432,943 title/subject-matter conflict. The earlier "Patent summary" section described US 11,432,943 (Ser. No. 16/352,699) as "interbody implant with deployable tooth — different disclosure." Search results now show a Carlsmed patent US 11,432,943 B2 bearing the title "Systems and methods for orthopedic implant fixation" with a 2018-03-14 priority date. Carlsmed plainly reuses that title across two distinct priority families (Mar 14, 2018 and Apr 16, 2018), so the number is not wrong but the title-based identification is ambiguous. Verify by application number, not title.
3. The earlier summary's claim-scope warning still stands: the claim text supplied to me is truncated mid-claim-9, so I cannot rule out further independent claims among claims 10–23. If an apparatus/system independent claim exists in that range, the Part II custom-implant cluster (Conformis et al.) becomes materially more relevant than the assessment above reflects.

Confidence statement

• High confidence: the identity of the '320; that the only citations verifiable from its own authoritative text are 62/658,389, Ser. No. 16/383,215, Ser. No. 16/207,116, and Pimenta 2015; that Pimenta 2015 is the reference most directly tied to the issued claim language (claims 2, 4, 7); and that the parent's front page cannot be treated as the '320's front page.
• Medium confidence: that the Part II list overlaps substantially with the '320's own examiner citations.
• Low confidence / explicitly undetermined: the '320's complete front-page citation list; whether the spinopelvic-guideline limitations enjoy the 2018-04-16 priority date; the contents of claims 10–23; and the substance of the Medicrea/Ryan leads (unverified leads only).
• Any legal-status or priority characterization above is an assumption of the record, not a legal conclusion.

I'll research the prior art landscape and the cited art of record for this patent before constructing the obviousness analysis.

Excellent — the searches surfaced highly on-point prior art. Let me drill into the references-of-record for this family and a few more specific prior-art areas.

Obviousness Analysis — US 12,251,320 B2 under 35 U.S.C. § 103

Prepared: analysis date stated in task = April 26, 2026. (Consistent with the two prior sections, note that my session clock reads 2026-09-17 and litigation surfaced in the earlier sections post-dates the task date. This does not affect the § 103 analysis, which is keyed to a 2018 priority date.)

0. Threshold note on the "Prior Art section" of the page

Per your instruction, I looked for the Prior Art section of the US 12,251,320 page. The fetched page supplies only a prior-art key-word set and a prior-art date:

Prior art keywords: spinal, patient, anatomy, subject, measurements
Prior art date: 2018-04-16

There is no enumerated prior-art reference list (no "References Cited" / "Cited By" table) in the page text supplied to me. Rather than treat that as an absence of art, I grounded the analysis in (a) the references of record surfaced for the parent patent US 11,439,514 and its sibling family members (the same examiner family, and therefore the art the Office actually considered), and (b) references surfaced for the immediate technical field. Where I rely on something I could only partially verify, I say so.

Sources used (URLs):

• Authority text: https://patents.google.com/patent/US12251320/en
• Parent of record: https://patents.google.com/patent/[US11439514B2](/patent/US11439514B2) ; PDF at https://patentimages.storage.googleapis.com/fa/0e/c3/53bacc7545ae5c/US11439514.pdf
• Citations-of-record list for the family: https://pubchem.ncbi.nlm.nih.gov/patent/US-[11439514](/patent/11439514)-B2 and https://patents.google.com:443/patent/US20200046513A1/en
• Medicrea rod-curvature family: https://www.freepatentsonline.com/y2021/0216671.html ; https://companyprofiles.justatic.com/assignee/medicrea-international ; https://patents.justia.com/assignee/medicrea-international?page=2
• Medicrea predictive-modeling: https://patents.google.com/patent/[US11877801B2](/patent/US11877801B2)/en ; https://patents.google.com/patent/EP3612122A2/en
• NuVasive planning family: https://patents.google.com/patent/US20160242857
• Solla et al. PSR paper: https://www.ovid.com/jnls/jspinaldisorders/abstract/10.1097/bsd.0000000000000721~patient-specific-rods-for-surgical-correction-of-sagittal
• Carlsmed sibling disclosures (for context, not prior art): https://patents.justia.com/patent/20250235324 ; https://www.freepatentsonline.com/y2023/0277246.html

1. Framework

1.1 Claims under analysis

I analyze claims 1–9 — the claims whose full text is in the record. As flagged in the Patent Summary section, claims 10–23 are truncated in the supplied text ("…a bone anchor con…"), so I cannot opine on them. Nothing below should be read as covering claims 10–23.

1.2 Effective filing date

Earliest asserted priority: U.S. Provisional 62/658,389, filed April 16, 2018 → Ser. No. 16/383,215 (Apr. 12, 2019) → Ser. No. 17/880,277 (Aug. 3, 2022). For § 103 I use April 16, 2018 as the presumptive critical date, and note the fallback: if any claim lacks § 112 support in the provisional (a live question for a continuation whose claims were rewritten post-parent — see § 5), the effective date could slip to April 12, 2019, which only adds art and strengthens the conclusions below.

1.3 Person of ordinary skill in the art (POSITA)

A POSITA here is a person with a master's degree (or equivalent experience) in biomedical/mechanical engineering, computer science, or an allied field, plus 2–4 years' experience in spinal implant design and image-based surgical planning, or a clinician (orthopedic/spine surgeon or biomechanical engineer) with equivalent familiarity with spinopelvic parameters. Because the claims are to a software-implemented design pipeline using well-characterized clinical metrics, the POSITA is also charged with the contents of the spine-deformity literature (which the patent itself incorporates — see § 3.1).

1.4 Scope of the independent claims (restated compactly, from the authoritative text)

• Claim 1 = (i) 3D virtual model of spinal anatomy → (ii) receive user-proposed changes including a change to the spatial relationship of ≥2 vertebrae → (iii) adjust the virtual anatomy → (iv) measure pelvic tilt / sagittal alignment / lumbar lordosis → (v) compare to "predictive guidelines" having predetermined values → (vi) if it conforms, then design patient-specific implant(s) achieving the corrected geometry.
• Claim 9 = same pipeline, broader at step (ii) (no express "≥2 vertebrae" requirement) and narrower at step (vi) (implants include a spinal implant and a bone anchor anchoring it to bone).

The novelty center of gravity is a gating/validation step, not the modeling, not the metrics, and not the implant per se.

2. Primary prior art

Key structural observation: References A, B, D, E and admitted reference F are all in the same field of endeavor (A61B 34/10 — computer-aided surgical planning; A61F 2/30 — patient-specific orthopaedic implants) and all address the same problem statement recited in the '320 specification (unpredictable intra-operative rod bending; excess fluoroscopy; achieving sagittal alignment). Reference B states the exact targets recited in dependent claims 2, 4, and 7.

3. Element-by-element analysis — Claims 1 and 9

3.1 Claim 1

Preliminary conclusion (claim 1): Every element is disclosed, and the only arguable "gap" is the computing-implementation framing ("receiving, from a user…"). That framing is a software/GUI implementation of a method the spine-planning literature already performed — i.e., a classic § 103 gimme, and possibly a § 102 anticipation by B (if the "dedicated program" is read as generating the virtual model) or by A read against the 3D-model element with the admittedly conventional imaging/segmentation steps.

3.2 Claim 9

Claim 9 = claim 1's pipeline (A, B, C, D, E, F) plus "the one or more patient specific implants include a spinal implant and a bone anchor configured to anchor the spinal implant to a bony structure of the patient."

• The bone-anchor element is the most ubiquitous feature in spinal fixation. Reference A's workflow ends in a rod anchored by pedicle screws or laminar hooks; B's PSR construct is a pedicle-screw construct ("average pedicle screws construct was of 6.4 levels"); D/E likewise.
• The '320 specification itself recites pedicle/iliac/sacral screws as conventional.
• Conclusion: claim 9 rises or falls with claim 1; adding a bone anchor is an obvious design choice (§ 103; KSR, "a court must ask whether the improvement is more than the predictable use of prior art elements according to their established functions").

3.3 The "design objective/condition" caveat

Even if one argued that the art "suggests" rather than "requires" the conform-then-design ordering, the Federal Circuit's repeated position is that reciting a desired outcome or a designer's decision rule does not confer patentability where the prior art supplies the motivation — and here the prior art supplies not just the motivation but the same numbers. The claim's conditional ("in response to … conforming") is functionally the art's convergence criterion.

4. Dependent claims 2–8 — the "quotation-mark" problem

This is where the § 103 case is at its strongest, because the dependent claims recite exactly the numeric thresholds already published in the field and explicitly admitted by the patent:

Important corroboration: These are not obscure numbers — they are the well-known SRS-Schwab / spinopelvic-alignment targets, and the '320 specification itself attributes them to Pimenta 2015 (reference F), expressly incorporated by reference. That incorporation is effectively an admission that the guidelines per se are prior art; the only question is whether applying them inside a virtual-surgery tool that then designs an implant is non-obvious. It is not.

5. Motivation to combine — why a POSITA would have combined these references

The following KSR-compliant rationales apply, and none of the references teaches away:

1. Same field, same problem, same solution architecture. A, B, C, D, E, F are all in computer-assisted spinal deformity planning. A and C identify and solve the identical problem the '320 recites in its Background: unpredictable rod bending, intra-operative fluoroscopy, and inconsistent sagittal correction. A POSITA improving a virtual-planning tool would naturally look to these.

2. "Combining prior art elements according to known methods to yield predictable results." Each step of claim 1 (model → simulate correction → measure spinopelvic metrics → compare to a target → output a patient-specific implant) is a discrete, known element; their serial arrangement is the conventional workflow of the field (A and B both describe it end-to-end). The combination yields nothing more than the predictable sum of its parts.

3. "Use of a known technique (numeric threshold gating) to improve a similar device/process in the same way." Threshold-based validation of a planned correction was already used (A, B, E) to answer the question the surgeon asks: "Have I corrected enough?" Automating that comparison in the planning software is the obvious mechanization of a manual/mental step.

4. "Obvious to try" with a finite, identified, predictable set of solutions. KSR makes clear that where the art identifies a finite number of predictable solutions, the claimed option is obvious. The spinopelvic targets are a finite, enumerated set (PT, SVA, LL–PI, T1/SPI…), and the art discloses the same numeric values. Claims 2, 4, 7 literally recite one of the art's enumerated options.

5. Design incentives and market pressure. The '320's own Background frames the demand: eliminate intra-operative radiation, reduce OR time, reduce surgeon-dependent variability. References A and B explicitly state these same objectives ("A randomized trial on sagittal correction using PSR is ongoing"; "no longer requiring the bending of rods during surgery"). The field was under an admitted, well-documented motivation to move from intra-operative judgement to pre-operative virtual planning.

6. Reasonable expectation of success. A and B report practical success (B: 60 patients; PI-LL < 10° improved from 29/60 to 50/60, OR 5). No unpredictability or teaching-away is present; the combination is a matter of routine software engineering plus a known clinical rule.

7. Apparatus/process equivalence. Even treating the '320's claim 1 as a "new use" of a general-purpose computer, it is the same method A/B/C/D already performed — merely with a distinct computational implementation. Nothing in the claims recites an unconventional improvement in computer functionality (no novel data structure, no architectural change, no addressed technical deficiency), so the implementation adds no patentable weight.

Potential counter-argument the patentee might raise, and why it is weak: the patentee may argue that "the art plans a rod," whereas the claim covers any patient-specific implant (including a plate), and that the specific ordering/gating is new. Neither position survives: (i) claim 1 is "one or more patient-specific implants," and B's "from the virtually corrected spine, the rod curvature and length were defined" is the design step generically; (ii) the ordering is the natural workflow and is expressly described in B ("simulated… to reach the following objectives… from the virtually corrected spine, the rod… was defined").

6. Secondary considerations (Graham factors) — preliminary

No file-history or commercial-evidence record was available to me, so this is a placeholder with the analysis flagged:

• Commercial success / industry praise (Carlsmed's Apropos®-type patient-specific implants, the DeGen suit described in the Litigation Summary): to be probative, the patentee must establish nexus to the claimed inventive feature (the conformance gate), not to patient-specific implants generally — and the art shows the gate itself was known.
• Licensing / copying: the 2026 Carlsmed v. DeGen Medical action (from the earlier sections) is not evidence of non-obviousness; litigation is not a Graham "secondary consideration."
• Long-felt need / failure of others: the field's rapid, successful adoption of virtual planning (A since 2014; B reporting success) cuts against a long-felt unsolved need.
• Unexpected results: none identified in the disclosure; the results reported are the expected improvement in sagittal alignment.

7. Conclusion

Claims 1 and 9 (and dependents 2–8) are, on the record I can see, vulnerable to a § 103 challenge. The most efficient combinations are:

• Primary combination (best): Medicrea US 10,318,655 [A] in view of Solla et al. 2019 [B], further in view of the Pimenta 2015 reference the patent itself incorporates [F].
  — [A] supplies the 3D/CAD virtual model, the simulation of correction, the spinopelvic metrics, and the numeric targets (LL = PI ± 10°, SVA < 5 cm), and the manufacture of the resulting implant; [B] supplies the explicit conform-then-design gate with the identical targets (PT < 20°, PI-LL < 10°, SVA < 50 mm) and the virtual-correction-to-implant-definition pipeline; [F] is the admitted source of the guideline framework. Result: claims 1, 2, 3, 4, 5, 7, 8, 9 obvious; claim 6 obvious over the same plus routine level-specific lordosis targets (e.g., [E]).

• Secondary combination: Aurouer et al. 2009 [C] in view of Medicrea US 10,318,655 [A] and US 11,877,801 / EP 3612122 [E], with the NuVasive planner family [D] supplying the general-purpose planning-system implementation. This covers any argument that [C]/[A] are 2D-x-ray-centric by supplying the computerized 3D planning and threshold-rule layer.

• § 102 alternative (worth noting, not relied on): If Solla et al. [B] is shown to have published online before April 16, 2018 (its issue date is March 2019, but Clinical Spine Surgery publishes ahead of print), it is a plausible single-reference anticipation of claim 1 (and of 2, 4, 7), leaving only the "3D virtual model" phrasing as a construction dispute.

8. Confidence statement and explicit gaps (per operating rules)

High confidence — grounded in the authoritative text and retrieved sources:

• The claim text for claims 1–9, the priority chain, and the claim/abstract mismatch.
• That Medicrea US 10,318,655 discloses the spinopelvic metric set, the simulation-of-correction step, LL = PI ± 10°, and SVA < 5 cm, and produces a manufactured, curved, patient-specific rod (URL: https://www.freepatentsonline.com/y2021/0216671.html).
• That the Solla et al. abstract discloses a simulated correction of osteotomies using a dedicated program with objectives PI-LL < 10°, PT < 20°, SVA < 50 mm, and definition of rod curvature/length from the virtually corrected spine (URL above).
• That the '320 specification incorporates the Pimenta 2015 paper as the source of the three predictive guidelines (admitted-into-the-record art).

Low / undetermined — do not treat as established:

• Claims 10–23 — text truncated in the supplied record; I have not analyzed them. (Note: the sibling publication US 2025/0235324 A1 surfaced claims 25–32 in my search, which appear to be a different claim set in a different family member — I have not conflated them with the '320.)
• The exact online-publication date of Solla et al. relative to April 16, 2018 — this determines whether it is § 102(a)(1) art or only § 102(a)(2)/obviousness material. I could not retrieve the "published-ahead-of-print" date.
• The complete "References Cited"/IDS of record for US 12,251,320 specifically — the page supplied to me contains only prior-art keywords, not a reference list, so I cannot state which of A/B/C/D/E the examiner actually considered.
• Any file-history admissions or narrowing amendments in Ser. No. 17/880,277 that might affect the § 103 posture.
• Whether the continuation claims enjoy the April 16, 2018 provisional priority.

Explicit caveat: This is a technical analyst's preliminary obviousness assessment based on public sources, not a legal opinion or a validity determination. A full § 103 analysis would require the complete prosecution file, the full claim set, and the actual prior-art references of record.