US 12245952

I'll verify current bibliographic and litigation status with targeted searches, then summarize.

Verification note before the summary

I searched for the literal identifiers US 12,245,952 / US12245952B2 and US 18/754,101. Results confirm the patent exists as described below. Two cautions:

• No CAFC docket match found. My searches returned no Federal Circuit appeal or other 2026 appellate docket involving US 12,245,952. One search hit referenced PTAB proceeding IPR2025-00052/54/55/56, but that email concerns other patents (the '069/828/407/061 patents, in an Ellodi matter) and must not be conflated with this patent. I cannot query PACER/CAFC CM-ECF directly, so treat "no litigation found" as absence of evidence, not proof of absence.
• Everything below is grounded in the authoritative full text you supplied, corroborated by Google Patents and an HT Syndication issuance notice (Alexandria, March 12, 2025).

US 12,245,952 B2

Relation to prior grant: This patent shares a specification with US 11,439,514 B2 (issued 2022-09-13) and US 12,251,320 B2 — same title and inventive entity. Note the family also includes US 11,432,943 B2, a different Carlsmed application (16/352,699, provisional 62/643,046) that happens to share the same title but claims an interbody implant with deployable teeth. Do not merge those.

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."

Plain-language overview of the independent claims

Claim 1 — Computer-implemented method of designing a patient-specific implant. Software displays a model/image of at least part of the patient's spine; the user updates it to show a corrected spinal anatomy (i.e., changes the spatial relationship between at least two vertebrae). The system displays predicted post-operative spinal metrics — pelvic tilt, sagittal alignment, and/or lumbar lordosis. Based on that corrected anatomy, a multi-planar patient-specific longitudinal element is designed, containing one or more three-dimensional curved segments sized/shaped to hold the spine in the corrected position. The element is then manufactured.

Claim 11 — Personalized spinal fixation system. The apparatus/system-side counterpart: a surgical planning software tool configured to (i) display the spinal image/model, (ii) update it to a corrected anatomy, (iii) display the predicted post-op metrics (pelvic tilt, sagittal alignment, lumbar lordosis), (iv) design the multi-planar patient-specific longitudinal element, and (v) generate fabrication instructions for making it. Claim 13 adds the actual manufacturing system.

Claim 17 — Patient-specific implant (product claim). A physical implant that matches a virtual longitudinal element generated by the surgical planning software. It is tied to a patient-specific surgical plan for moving the spine into a corrected anatomy; it is sized/shaped to span at least three vertebral segments and stabilize the corrected spine; and it has three-dimensional curved segments designed based in part on the predicted post-op spinal metrics (pelvic tilt, sagittal alignment, and/or lumbar lordosis).

Dependent-claim themes: fewer/more material properties along the element — differing stiffness (claims 4, 15, 18), differing density and/or porosity (claims 5, 19), varying thickness (claim 6); multi-planar rod (8, 16, 21) vs. multi-planar plate (9, 22); plurality of bone-anchor-receiving nodes (10); plurality of bone anchors (20); extension across three or more vertebral levels (7); and the closed-loop guideline check of claims 2 and 12 (compare predicted metrics to predetermined guideline values for pelvic tilt / sagittal alignment / lumbar lordosis; revise the virtual anatomy if not conforming; manufacture only if conforming).

Analytical observations worth flagging

1. Claim scope departs sharply from the abstract and specification. The abstract and the detailed description are dominated by a non-planar "plate" with nodes, holes, bone anchors (sacral/iliac screws), and a locking element. The independent claims, by contrast, are directed to a "multi-planar patient-specific longitudinal element" defined functionally by surgical-planning metrics, and the claims recite no plate, no node, no locking element, and no additive manufacturing. That mismatch between the disclosure's focus and the claim set is the single most consequential feature of this document.
2. The 2018 priority date is doing heavy lifting. The application itself was filed 2024-06-25, but the claims recite subject matter (multi-planar rods/plates, pelvic tilt / SVA / lumbar lordosis predictive guidelines) whose written-description support traces to the 2018 provisional and the 2019 parent. Any validity analysis turns on whether the 2018 provisional supports these claims under §112. The specification's §132-corresponding passage (the three predictive guidelines, citing Pimenta, Journal of Spine, 2015) appears verbatim in the shared disclosure.
3. Term: The listed 2039-04-12 anticipated expiration runs from the 2019 parent's filing (plus any PTA), not from the 2024 continuation filing.
4. Uncertainty: I could not independently confirm current maintenance-fee status, any terminal disclaimer, or the absence of IPR/PGR or district-court activity specifically naming 12,245,952. Those would need a live PatentCenter, PTAB, and PACER/CAFC docket check.

I'll search for litigation involving US patent 12245952 across patent litigation databases.

Initial searches didn't surface case-specific results. Let me search patent litigation databases more specifically.

I found a relevant Carlsmed enforcement action. Let me determine whether the '952 patent is among the asserted patents and check other sources.

Search Results: Litigation Involving U.S. Patent No. 12,245,952

Summary finding: I identified no litigation that specifically names U.S. Patent No. 12,245,952 ("Systems and methods for orthopedic implant fixation," Carlsmed, Inc.).

I searched patent-litigation resources including Unified Patents' litigation portal, CourtListener, PACER-derived dockets (Docket Alarm / Justia / PacerMonitor), Becker's Spine Review, and general news sources. No case was found in which the '952 patent is identified as an asserted patent.

What I did find — and an important caveat

The patent's assignee, Carlsmed, Inc., has an active enforcement campaign, but I could not confirm that the '952 patent is among the patents asserted:

Sources:

• Carlsmed press release: https://investors.carlsmed.com/news-releases/news-release-details/carlsmed-inc-files-patent-infringement-lawsuit-against-degen
• Becker's Spine Review: https://www.beckersspine.com/spinal-tech/carlsmed-sues-degen-medical-over-8-patents/
• PacerMonitor docket: https://cdn.pacermonitor.com/public/case/66840125/Carlsmed,_Inc_v_DeGen_Medical,_Inc
• Justia docket listing: https://dockets.justia.com/browse/state-south_carolina/noscat-10/nos-830

Because the complaint's patent schedule (Exhibit listing of asserted patents) is not reproduced in the accessible public reporting I retrieved, I cannot state that US 12,245,952 is one of the eight asserted patents. It is plausible (the asserted patents are described as relating to Carlsmed's "implants, methods and systems," which is the '952 patent's subject matter), but that would be speculation, not a verified fact. Confirming this would require the actual complaint PDF from PACER (Docket No. 1 in 4:26-cv-04064).

Other notes

• No PTAB proceedings (IPR/PGR) challenging the '952 patent were found.
• No CAFC appeals involving the '952 patent were found.
• The '952 patent is a continuation (filed June 25, 2024) of U.S. App. Ser. No. 17/880,277 (now U.S. Pat. No. 12,251,320) and of Ser. No. 16/383,215 (U.S. Pat. No. 11,439,514), part of a Carlsmed patent family with priority to April 16, 2018.

Flag on dates

You indicated the current date is April 26, 2026, but the Carlsmed v. DeGen filing surfaced in my searches is dated September 15, 2026 (and the system date provided to me is October 1, 2026). If your effective date is truly April 26, 2026, this suit post-dates it and all other findings would be pre-dating the search horizon — worth reconciling.

Bottom line: As of the available public records, no litigation is confirmed to involve U.S. Patent No. 12,245,952 specifically. If you need a definitive answer, the next step is to pull the Docket No. 1 complaint and its patent exhibit list from PACER for 4:26-cv-04064 to see whether the '952 patent appears among the eight asserted patents.

Proceedings overview

Total AIA trial proceedings on US 12245952: 0. The USPTO Open Data Portal returns no AIA trial proceedings (no IPR, PGR, or CBM) associated with this patent as of the most recent ingest, and targeted web searching surfaced no petition, institution decision, FWD, or Federal Circuit appeal naming this patent — so the breakdown is 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denied. The defensive posture this gives a defendant is "hardened by default — but untested": all 22 claims stand, unamended and unchallenged, and nothing about the patent's validity has been adjudicated at the Board. That is not the same as "strong." It means the plaintiff currently holds 22 presumptively valid claims with zero PTAB estoppel on the books, and the first petitioner gets a clean, uncrowded shot at the art.

No proceedings to report. Per the operating instruction, the structured PTAB block is canonical and it is empty; I am not going to manufacture proceeding numbers, panels, or FWDs. Searches for petitions against 12245952, its parents (US 11,439,514; US 12,251,320), and its assignee (Carlsmed, Inc.) returned no Board activity.

One flag that is worth the defendant's attention, though it is not a PTAB proceeding: a related litigation has now been filed, which is the customary trigger for AIA petitions in this space.

Non-PTAB context — Carlsmed, Inc. v. DeGen Medical, Inc. (D.S.C.)

• Forum / docket: U.S. District Court for the District of South Carolina, Florence Division, Case No. 4:26-cv-04064 (Judge Joseph Dawson, III) — PacerMonitor case page
• Filed: 2026-09-15 (press release dated 2026-09-15, Carlsmed IR)
• Scope: Carlsmed states the complaint asserts eight Carlsmed patents "relating to its implants, methods and systems," seeking injunctive relief and damages. I could not confirm from public sources whether US 12245952 is among the eight asserted patents — the press release and secondary coverage (Becker's Spine Review) do not list patent numbers. Treat that as an open item to verify against the complaint itself.
• Defendant's position: DeGen CEO Craig Black publicly disputed the allegations and stated DeGen "intend[s] to vigorously defend" (Becker's, 2026-09-16).
• PTAB relevance: This is the § 315(b) clock-starter. Once DeGen is served, it has one year to petition for IPR/PGR on any asserted patent, and — critically for the estoppel analysis below — no petitioner has yet triggered § 315(e)(2) estoppel against this patent.

Strategic summary

Claim status: all 22 claims UNTESTED. There is no IPR, PGR, or CBM history on US 12245952, so nothing is CANCELED and nothing is SUSTAINED-by-the-Board. The claim set as granted breaks into three independent groupings: a computer-implemented method of designing a patient-specific implant (claims 1–10, with claim 1 reciting corrected spinal anatomy, predicted post-operative spinal metrics of pelvic tilt / sagittal alignment / lumbar lordosis, a multi-planar patient-specific longitudinal element, and a manufacturing step); a personalized spinal fixation system (claims 11–16); and a patient-specific implant (claims 17–22, which claims the physical multi-planar longitudinal element that "matches a virtual multi-planar patient-specific longitudinal element generated by the surgical planning software"). All three independent claims carry the same core limitation family — multi-planar geometry derived from a corrected spinal alignment and predicted post-op metrics. Note the patent's prosecution history context: this is a continuation of 17/880,277 (US 12,251,320) → 16/383,215 (US 11,439,514) → provisional 62/658,389, all with a 2018-04-16 priority date and an anticipated expiration of 2039-04-12. A petitioner attacking claim 1 must contend with the same 2018 priority date across the family, so the art search is a 2018-or-earlier exercise; the family also keeps generating new continuations (US 19/041,891, published as US 2025/0235324 A1), meaning a design-around or invalidity win on this patent alone may not clear the portfolio.

Estoppel landscape: a blank slate — for now. Because no IPR/PGR has ever been filed on this patent, there is no § 315(e)(2) estoppel binding anyone, and no petitioner or privy is precluded from raising any § 102/§ 103 ground, including grounds that would otherwise be "reasonably could have raised" forfeitures. That cuts both ways for a defendant: the full prior-art universe is available, but you get exactly one bite without estoppel exposure, and a first-filing petitioner (e.g., DeGen) may capture the best art before you do. If DeGen files first, a co-defendant who is not a privy is not estopped, but a supplier/manufacturer indemnitor relationship is a common estoppel trap. There is also no In re IPR joinder or consolidated-proceeding history to leverage.

Pattern signals. (1) Carlsmed is a patent owner with zero PTAB defensive history — the company has never been a patent owner in an instituted AIA trial, so there is no track record of how it litigates at the Board (no prior claim amendments, no prior motion-to-amend practice, no known expert stable at the PTAB). (2) No single petitioner has filed multiple IPRs against this patent, because none has filed any. (3) No defensive aggregator (Unified Patents, RPX, LOT, IPR-focused hedge funds) appears in the chain — Unified's portal surfaces Carlsmed patents as literature (e.g., US-12232980-B2 on Unified's patent portal, cited in an obviousness-style listing), not as challenge targets, so there is no third-party IPR machine to piggyback on. (4) Carlsmed has no documented Federal Circuit appeals, because it has no PTAB appeals. The company is instead behaving as an aggressive plaintiff: first asserted patent suit disclosed 2026-09-15, days after a favorable CMS reimbursement decision and a $250M shelf registration (per third-party news aggregators).

Recommended next steps

• If you are DeGen (or a similarly situated defendant): the patent has no invalidated claims to lean on. There is no FWD to cite and no canceled claim to quote in a Rule 12 or § 101 motion. Your invalidity case must be built from scratch — there is no PTAB record to import.
• Watch the § 315(b) one-year clock. Confirm the service date of the 4:26-cv-04064 complaint. The IPR/PGR petition deadline runs one year from service of a complaint alleging infringement of the patent. If US 12245952 is among the eight asserted patents, that deadline is the single most important date on the docket; if it is not asserted, there is no § 315(b) bar and no estoppel consequence to petitioning early.
• Pre-suit / early-filing IPR is unusually attractive here. With no prior petitions and a 2018 priority date, the art space is uncrowded. The claim 1 combination (corrected spinal anatomy + predicted pelvic tilt / sagittal alignment / lumbar lordosis metrics + multi-planar patient-specific longitudinal element + manufacturing) is a strong § 103 target given the dense pre-2018 surgical-planning literature, and the "matches a virtual ... element generated by the surgical planning software" language in claim 17 invites § 112 indefiniteness arguments that the Board may reach in the alternative.
• No active proceedings means no trial-stage milestones to track. There is no institution-decision deadline, no oral hearing, and no PTAB statutory 1-year FWD date, because no trial has been instituted. Any timing you care about comes from the district court case, not the Board.
• Verify before relying. Confirm directly against the USPTO Patent Trial and Appeal Board E2E / PTAB Decisions system and the USPTO patent page that no proceeding issued after the ODP ingest date; a petition could have been filed between ingest and today (2026-10-01). Absence of PTAB activity on a patent this young and this commercially asserted is a signal, not a guarantee — well-asserted patents eventually attract IPRs, and the filing of 4:26-cv-04064 on 2026-09-15 is exactly the kind of event that precedes one.

Caveats on sourcing: the proceeding list is taken from the structured ODP block in the prompt (canonical, empty). I searched for petition records, Board decisions, and CAFC activity on US 12245952, US 11,439,514, US 12,251,320, and Carlsmed as patent owner, and found none. I did not locate the text of the 4:26-cv-04064 complaint, so I cannot confirm which eight patents are asserted. No FWD, panel composition, or settlement term is reported anywhere in this memo because none exists to report.

I'll verify the assignment record with targeted searches before reconstructing the chain.

Assignment & Ownership Analysis — US 12,245,952 B2

"Systems and Methods for Orthopedic Implant Fixation" · App. 18/754,101 · Assignee of record: Carlsmed, Inc.

Access limitation (read first). I could not query the USPTO Assignment Center UI or its JSON endpoints directly — only general web search was available to me. Consequently I can confirm the existence, type, assignor/assignee identities and the recording date of the assignment from Google Patents' legal-events mirror, but I could not retrieve the reel/frame number, the execution date, or the correspondent of record. Those three fields are flagged as unverified below rather than guessed. The chain is also unusually short — one record — so the "stop after this section" rule does not apply.

Inventors

Pattern assessment — no fire-sale precursor. The classic tell the brief asks about (all inventors departing the original assignee within 12 months of filing) is absent. Both inventors co-founded Carlsmed, both remained continuously employed through the 2025 issuance and the July 2025 IPO, and Cordonnier is still CEO as of the most recent earnings call (Feb 2026). Cordonnier signed the company's Amended and Restated Investors' Rights Agreement as CEO at 1800 Aston Avenue, Suite 100, Carlsbad, CA 92008, which is the assignee's address of record.

One notable sub-pattern: the inventive entity here is only the two founders. Sibling Carlsmed patents in the same technical family add outside clinicians — e.g. US 10,902,944 B2 names Justin Esterberg (Mercer Island, WA) and Jeffrey Roh (Seattle, WA) in addition to Casey and Cordonnier. The omission of the clinical inventors from this patent (and from its two parents) is consistent with deliberate claiming on software/design method rather than on surgical technique — not an ownership anomaly, but worth noting because it narrows the inventorship record to company insiders.

Original assignee

Carlsmed, Inc., 1800 Aston Avenue, Suite 100, Carlsbad, CA 92008. Named as assignee on the face of the patent and as the current assignee. No "IP Holdings," no licensing vehicle — the operating entity itself.

• Primary line of business: commercial-stage medical technology — AI-enabled personalized spine surgery. Product platform aprevo (proprietary surgical-planning software + custom 3D-printed, anatomically matched vertebral interbody implants); the corra cervical platform adds personalization to cervical plates; a bilateral aprevo lumbar construct addresses the posterior approach.
• Does a product embody these claims? The claims read on a multi-planar patient-specific longitudinal element (rod or plate) spanning ≥3 vertebral levels and generated from planning-software metrics. Carlsmed's interbody implant is not that element; however the cervical plate line (corra) and the posterior-approach program are the closest commercial analogues, and a 2025 family member titled "Posterior fixation systems for spinal treatments" (US 12,514,644 B1, filed 2025-01-09, granted 2026-01-06) indicates Carlsmed is actively building the posterior-fixation rod/plate line these claims cover. I would characterize product embodiment as plausible and aligning, but not conclusively documented from my sources.
• Current status: Operating. Publicly listed on Nasdaq (ticker CARL) since its IPO priced 2025-07-22 at $15.00/share and began trading 2025-07-23; gross proceeds ~$100.5M. FY2025 revenue $50.5M (+86% YoY), GAAP net loss $29.6M, cash and investments $109.9M at year-end 2025. Added to the Russell 2000 on 2025-09-22. No bankruptcy, no receivership, no acquisition.

Assignment timeline

Only one assignment is reflected in the indexed record for this patent:

• Execution date: not exposed in the indexed record / recorded 2025-03-11 — Reel/frame not retrievable via the interfaces available to me (unverified)
  • Conveyance: Assignment of assignors' interest ("ASSIGNMENT OF ASSIGNORS' INTEREST — SEE DOCUMENT FOR DETAILS")
  • Assignor: Niall Patrick Casey and Michael J. Cordonnier (individually)
  • Assignee: Carlsmed, Inc. (Carlsbad, CA)
  • Correspondent: not retrieved — could not be read from the available source. This is the single most important open field; see signal 3. For context only (and not a substitute for the correspondent field), the attorney of record on a sibling Carlsmed patent, US 11,793,577 B1, is Perkins Coie LLP — a full-service firm, i.e. the ordinary course for an operating-company patent portfolio, not an NPE recording mill.
  • Context: Original inventor-to-company assignment. This is an in-house/initial prosecution record, not a post-issuance transfer. Its recording date (2025-03-11) coincides exactly with the grant date — consistent with a routine batch recording on issuance, and inconsistent with an assertion-driven transfer.

No other records. There are no security agreements, no confirmatory assignments, no mergers/name changes, no license recordations, no releases, and no terminal disclaimers surfaced. Executed-but-unrecorded documents cannot be ruled out.

Family consistency check: all four members of family ID 68237242 — 16/383,215 (US 11,439,514 B2), 17/880,277 (US 12,251,320 B2), 18/754,101 (this patent), and 19/041,891 (US 2025/0235324 A1, pending) — name Carlsmed as applicant/assignee. Ownership never left the operating company across the entire continuation chain.

Timeline diagram

timeline
    title Ownership of US 12245952
    2018 : Provisional 62/658389 filed
         : Carlsmed incorporated
         : Casey and Cordonnier co-found company
    2019 : Parent 16/383215 filed
    2022 : Continuation 17/880277 filed
    2024 : Continuation 18/754101 filed
    2025 : Patent US12245952 issues Mar 11
         : Inventor assignment to Carlsmed recorded
         : Carlsmed IPO on Nasdaq ticker CARL

NPE / troll-pattern signals

Verdict

Insufficient data.

The rubric defines this category as covering chains consisting of "no records, or only the original assignment" — which is exactly what exists here: a single inventor-to-company assignment recorded on 2025-03-11 (assignors Casey and Cordonnier → Carlsmed, Inc., reel/frame unretrieved). I am not choosing this to imply the patent looks like an NPE — it does not. All eight NPE signals are "not present" or "unclear," and the substantive evidence points squarely to ongoing ownership by a product-shipping, Nasdaq-listed operating company (CARL) that raised ~$100.5M at its 2025-07-22 IPO and booked $50.5M of 2025 revenue. A stricter "Operating-company assertion" verdict was declined only because I found no competitor litigation naming this patent, so the second prong of that category is unevidenced.

Caveat on absence of evidence: no IPR/PGR, no district-court or CAFC docket entry specifically naming 12,245,952, and no current maintenance-fee or terminal-disclaimer status could be independently confirmed through the interfaces available to me. Treat those as gaps, not as cleared findings.

Verify at:

• USPTO Assignment Center (search patent 12245952, then open Reel/frame to read the correspondent): https://assignmentcenter.uspto.gov/
• Legacy assignment search index: https://assignment.uspto.gov/patent/index.html
• Google Patents legal events / reassignment entry used above: https://patents.google.com/patent/[US12245952B2](/patent/US12245952B2)/en

I'll verify the patent in USPTO records and pull the prior-art citation list, then analyze the references against the claims.

Let me pull the complete prior-art citation list and check key references.

Let me verify the details of the most claim-relevant references.

Prior-Art Analysis — U.S. Patent No. 12,245,952 ("Systems and Methods for Orthopedic Implant Fixation")

Scope and verification caveats (read first)

What I could and could not verify. I performed targeted searches of Google Patents, USPTO-derived secondary sources (patents-review.com, uspto.report, HT Syndication), and general web sources. I confirmed the patent's existence and bibliographic identity, but I could not directly query the USPTO Patent Full-Text Search (PatFT/Patent Public Search) API, nor could I retrieve the complete machine-readable "References Cited" list. The authoritative full text you supplied includes the USPTO "Citations (254)" table only in truncated form — the visible entries run from US 4,704,686 A through US 2014/0100886 A1, after which the text cuts off. I therefore cannot enumerate all 254 cited references, and I explicitly flag those references for which I could not confirm citation-list membership.

Identifiers were interpreted literally. I searched the exact string 12245952 / US12245952B2 / US 18/754,101. One unrelated hit (the "US Patent 9197470" wiki page) lists an application number 12245952 and describes an OFDM/DQPSK communications patent — that is a different, unrelated record and I have not confused it with the '952 patent. As instructed, I did not fold in similarly-numbered records.

Anticipation vs. obviousness. A §102 anticipation finding requires that a single reference disclose every element of a claim, arranged as in the claim. The independent claims here (1, 11, 17) require a specific combination: (i) computer-implemented display/update of a corrected spinal anatomy, (ii) display of predicted post-operative spinal metrics specifically including pelvic tilt, sagittal alignment, or lumbar lordosis, (iii) a multi-planar patient-specific longitudinal element with three-dimensional curved segments, and (iv) manufacture/design output. As shown below, no single cited reference appears to disclose that full combination. My claim mapping therefore distinguishes honest §102 candidates from the much larger set of §103 (obviousness) references.

Contradiction flag (per instructions): The previously generated sections note a date discrepancy — an earlier summary referenced April 26, 2026, while the litigation section surfaced a September 15, 2026 filing and an October 1, 2026 system date. This task states the current date is April 26, 2026. I proceed on April 26, 2026, and reiterate that the Carlsmed v. DeGen suit (4:26-cv-04064) cannot be confirmed to assert the '952 patent; that remains an open item and is not resolved by anything below.

1. Patent under analysis

Effective prior-art cutoff is therefore April 16, 2018 (subject to the §112 written-description question for the 2018 provisional, already flagged in the earlier summary).

2. Most claim-relevant cited references

The table below is ordered by probative weight against claims 1/11/17 and their dependents. Dates are as given in the supplied citation table. "Type" distinguishes true §102 candidates from §103 references.

Non-patent literature cited on the face of the specification:

3. Highly probative art that I could not confirm is on the '952 citation list

Because the supplied citation table truncates at US 2014/0100886 A1 and my searches reached a step limit before I could open the full Google Patents citation pane, I flag the following separately. I cannot state these are among the 254 cited references — only that they are the most technically on-point art I located and that they bear directly on the independent claims.

4. Bottom line

1. No single reference in the visible citation record anticipates independent claim 1, 11, or 17 in its entirety. The claims' distinguishing combination — (a) computer-implemented update to a corrected spinal anatomy, (b) display of predicted post-operative metrics specifically PT/ sagittal alignment/LL, (c) a multi-planar patient-specific longitudinal element with 3D curved segments, and (d) manufacture — is not met by any one listed reference. The cited art is overwhelmingly §103 material (patient-specific design and manufacture: Schroeder, Furrer, Biomet, 3Di, Therics, Case Western, Conformis; additive/porous manufacture: Zimmer, Hollister, Conformis).

2. The most dangerous art is metric-driven, patient-specific spinal-rod design — the Medicrea family (US 10,413,365 B2) and the surgical-planning system of US 10,695,099 B2 — coupled with the Pimenta 2015 guidelines for pelvic tilt, SVA, and PI−LL. Whether these fall inside the '952 citation record is unconfirmed; that is the highest-priority follow-up.

3. Claim 17 is the softest independent claim. It is a product claim whose distinguishing feature is that the physical element "matches a virtual … generated by the surgical planning software" with curved segments "designed based in part on" the predicted metrics. Correlating a manufactured rod to a virtual plan is squarely the Medicrea teaching, making a §102 (or at minimum a strong §103) challenge to claim 17 more plausible than to claims 1 or 11.

4. Dependent claims 4/5/6/15/18/19 (variable stiffness, density/porosity, thickness) are the most vulnerable to §103 over Zimmer WO 2012/154534 + Hollister US 7,174,282 + Conformis US 2014/0086780.

5. Claims 2 and 12 (guideline values for pelvic tilt / sagittal alignment / lumbar lordosis) are the most vulnerable to §102/§103 over Pimenta (2015), which the specification itself incorporates.

Unresolved / needs primary-source verification:

• The complete 254-entry citation list (only the first ~100 patent entries were machine-readable in the supplied text).
• Whether US 10,413,365 B2 and US 10,695,099 B2 are cited on the '952 face; and US 10,695,099's assignee and priority date.
• Whether the '952 is among the eight patents in Carlsmed v. DeGen Medical, 4:26-cv-04064 (D.S.C.) — relevant to third-party validity interest, but not confirmed.
• The 2018 provisional's §112 support for the PT/SVA/LL claim language (already flagged in the prior summary as the pivotal validity question).

If you can supply the uncut USPTO "References Cited" listing or the filing-date/assignee data for US 10,413,365 and US 10,695,099, I can convert items A–Q above into a formal element-by-element §102 chart and rank the §103 combinations.

I'll ground this analysis in the prior-art citations listed on the patent page, verifying the key references before building combinations.

Let me verify the Medicrea pre-2018 rod-design art and the spinal-alignment planning literature that underpins the "predicted post-operative metrics" limitation.

Obviousness Analysis — U.S. Patent No. 12,245,952 B2

Systems and Methods for Orthopedic Implant Fixation (Carlsmed, Inc.), 35 U.S.C. § 103

0. Framework, caveats, and a sourcing note

Effective prior-art date: 2018‑04‑16 (provisional 62/658,389). Everything below is measured against that date. The 2024‑06‑25 filing date of application 18/754,101 is legally irrelevant to the art window — the '952 claims ride the 2018 priority chain (16/383,215 → 17/880,277 → 18/754,101), which is the single most important fact for this analysis.

PHOSITA: a person with an M.S. or equivalent in biomedical/mechanical engineering (or a spine surgeon-engineer team), with 3–5 years' experience in spinal fixation hardware and image-based surgical planning/CAD, familiar with sagittal alignment parameters (pelvic incidence [PI], pelvic tilt [PT]/pelvic version, sacral slope, SVA, PI‑LL mismatch) and with additive/subtractive patient-specific implant manufacturing.

Sourcing note — read this before relying on the grounds below. You asked me to use the Prior Art section of the page. That section is the Citations (254) block. The text supplied to me was truncated after roughly the first 70 entries, so I can see US 4,704,686 through US 2014/0086780 A1 but not the remainder of the 254. I therefore split the art into two tiers:

• Tier 1 (verified on the face of the '952 page, in the visible Citations list). I have quoted only these as "cited on the face."
• Tier 2 (identified by targeted search, same field and window). Chief among these is the Medicrea family, which is materially the closest art I found. I cannot confirm whether Medicrea appears in the unshown portion of the '952 citation list; the specification itself cites the Pimenta, J. Spine 2015 alignment paper, and the '514/'320 parents are of the same vintage, so it is likely the Examiner considered this space — but treat that as unverified.

Date flag: the previously generated Litigation section flagged an inconsistency between the stated analysis date (April 26, 2026) and search results dated September 15–16, 2026 (Carlsmed v. DeGen, 4:26‑cv‑04064). Nothing in this § 103 analysis depends on that case; I note it only so the record is consistent.

1. Claim 1, element-mapped

Claim 1 is a five-step computer-implemented method. Because the claim recites no plate, no node, no hole, no locking element, and no additive manufacturing — a mismatch with the plate-centric abstract and specification that the earlier sections already flagged — the prior art surface it must clear is broad.

Immediate observation. The Medicrea references, read together, disclose every conceptual element of claim 1: sagittal-alignment-parameterized virtual correction → rod geometry → manufacture. The '952 claims are distinguishable, if at all, by (i) the 3-D model/display framing versus Medicrea's 2-D x-ray workflow, and (ii) the explicit display of predicted post-operative metrics. Both gaps are closed by Tier-1 art.

2. Ground 1 (primary) — Medicrea + Catholic Healthcare West + Biomet/Conformis

Combination: US 2016/0210374 A1 and/or US 2016/0235479 A1 / US 9,693,831 B2 (Medicrea) in view of US 2012/0150243 A9 (Catholic Healthcare West) and US 2010/0324692 A1 (Biomet) or US 2014/0086780 A1 (Conformis).

Medicrea discloses a surgical-planning interface that analyzes a pre-operative image of the vertebral column, simulates a correction to the lumbar segment, and computes a desired curvature of a customized spinal rod from pelvic parameters and lordosis, then transfers the data to produce the rod (US 2016/0210374 A1; US 2016/0235479 A1; US 9,693,831 B2).

Catholic Healthcare West (cited on the face of the '952 patent) supplies the missing 3-D display architecture: importing and segmenting the CT/MRI, converting to a CAD model of the patient's spine, running a virtual surgical procedure in which vertebrae "react realistically to applied forces," suggesting revisions, and rapid-prototyping post-operative models and implants — verbatim the claim 1 display-update-design-manufacture arc (US 2012/0150243 A9).

Biomet (cited on the face) supplies patient-specific implant manufacture from imaging data (US 2010/0324692 A1); Conformis (cited) supplies optimization of implant design/manufacture by solid freeform fabrication (US 2014/0086780 A1).

Motivation (KSR): All three are in the same field and address the same, expressly-identified problem: the '952 specification itself complains that "surgeons approximate the desired shape of the rod based on an estimate," with "no tool or method to determine the optimal degree of correction required," and that bending occurs at the end of surgery when the surgeon is fatigued. Medicrea articulates the identical problem — rod curvature is "done at the surgeon's discretion," and the "trial and error necessary… has the significant drawback of extending the operation time." When two references in the same field articulate the same defect and propose overlapping remedies, KSR makes the combination obvious as a matter of "known elements… according to known methods," with predictable results.

3. Ground 2 — Catholic Healthcare West as primary, Medicrea/Pimenta as secondary

For an Examiner or defendant who prefers a § 102(a)(2)-clean primary reference:

• Primary: US 2012/0150243 A9 — patient-specific spinal planning + virtual surgery + implant fabrication.
• Secondary: Medicrea US 2016/0210374 A1 (or the Pimenta 2015 article the '952 specification itself incorporates) for the predicted post-operative metric limitations. Pimenta teaches the identical guideline set the patent recites in claims 2/12: PT < 20°, C7 SVA ≤ 5 cm, |PI − LL| < 10°.
• Secondary: Spinemedica US 2007/0276501 A1 (cited on the face) for displaying a virtual implant in an "electronic anatomical graphic model of a spinal column," letting the clinician adjust features, and "graphically simulating a post-surgical effect on posture, height and/or wedge angle" (US 2007/0276501 A1). This is a direct, pre-2018 disclosure of "displaying one or more predicted post-operative spinal metrics."

The combination is a textbook KSR "predictable use of prior-art elements according to their established functions."

4. Ground 3 — plate-centric art (for claims 9, 22, and the node/bone-anchor dependents)

If the patentee attempts to read the claims onto the disclosed non-planar plate embodiment (abstract/Figs. 6–10), the Tier-1 art is stronger still:

• US 6,978,188 B1 (Medical Modeling, cited on the face): computer-designed, pre-contoured bone reconstruction plates, using medical image data, three-dimensional surface reconstruction, virtual surgical simulation, and Solid Freeform Fabrication — and expressly noting "the strongest plates… are for reconstruction of defects in the spine, pelvis and mandible" (US 6,978,188 B1). This anticipates/renders obvious a patient-specific, multi-planar, anatomy-conforming plate made additively.
• US 2014/0074438 A1 (Furrer, cited): "Method for producing patient-specific plate" (link).
• WO 2004/110309 A2 (Case Western, cited): computer-aided design of skeletal implants.
• WO 2012/154534 A1 (Zimmer, cited): patient-specific manufacturing of porous metal prostheses — directly on point for claims 5/19 (differing density/porosity) (link).
• US 8,457,930 B2 (Schroeder), US 8,532,806 B1 (Masson), US 8,644,568 B1 (O.N. Diagnostics, automated patient-specific bone-implant biomechanical analysis), US 6,932,842 B1 (3Di, generating patient-specific implants), US 6,772,026 B2 (Therics), US 7,174,282 B2 (Hollister, design methodology for scaffolds) — all cited on the face, all supplying patient-specific design/manufacture with variable internal structure.
• A bone plate with nodes/holes receiving bone anchors secured by a locking nut is not merely known — it is the '952 patent's own admitted prior art (Figs. 3–4, plate 32 / screws 34 / nut 36 / slots 33). Admission-of-prior-art art is the hardest to overcome.

Supplementary (searched, not confirmed on the face): US 2008/0195240 A1 (Martin et al.) expressly recites designing an orthopedic plate by varying "a local and a total thickness," segment length/width, "out-of-plane bending and a torsion" — the limitation of claim 6 (varying thickness) and the "does not conform to a single plane" concept.

5. Claim 11 (system) and claim 17 (product)

Claim 11 requires the software tool to "generate fabrication instructions." Medicrea's interface is configured to "transfer data relating to the desired curvature of the customized spinal rod to a service provider responsible for physically producing a spinal rod" — that is a fabrication instruction. Catholic Healthcare West's rapid-prototyping interface and Biomet's imaging-to-implant data pipeline complete the limitation. Claim 13's "manufacturing system" is met by US 9,693,831 B2 step (m) and by Conformis/Zimmer additive systems.

Claim 17 claims a physical element that "matches a virtual multi-planar patient-specific longitudinal element generated by the surgical planning software," is tied to a plan for "changes to a spatial relationship between at least two vertebrae," spans "at least three vertebral segments," and has 3-D curved segments based on predicted post-op metrics. A rod manufactured by Medicrea's method is that article: the Medicrea constructs span L5–L1 and T11–C6 (far more than three levels), and the rod is produced to the geometry derived from the corrected alignment. Note that claim 17's "matches a virtual … element" language is also vulnerable under § 112 as a functional/product-by-procedure limitation, and would be construed to cover any rod made by the disclosed process — which the Medicrea method supplies.

6. Dependent claims

Claims 2 and 12 deserve emphasis: the '952 patent incorporates by reference the very Pimenta 2015 paper that supplies the three guideline thresholds recited in those claims, and Medicrea independently recites the PI−LL ≤ 10° rule. An obviousness rejection or invalidity contention on claims 2/12 is close to self-proving from the face of the '952 specification.

7. Motivations to combine — KSR rationales

1. Same field, same problem, same solution (express motivation). Both the '952 specification and Medicrea identify intraoperative rod bending as error-prone, time-consuming, and experience-dependent. Where the references themselves identify the deficiency the claim addresses, motivation is established. KSR, 550 U.S. 398, 420 (2007).
2. Combining known elements to yield predictable results. Image-based patient-specific planning (CHW) + sagittal alignment parameters (Medicrea/Pimenta) + freeform/additive manufacture (Conformis, Zimmer, Biomet) each perform its known function; the combination simply serializes them.
3. Design incentive / market forces. Reduced OR time, reduced fluoroscopy radiation, fewer revision surgeries, and better sagittal-correction outcomes are recognized, quantifiable incentives.
4. "Obvious to try." With a finite, small set of known alignment parameters (PT, SVA, PI−LL, SS) and known rod/plate geometries, the claimed selection is an obvious candidate.
5. Use of a known technique to improve a similar device in the same way. Contouring bone reconstruction plates pre-operatively with SFF (US 6,978,188 B1) applied to spinal rods/plates.
6. No teaching away. Nothing in the cited art disparages patient-specific multi-planar geometry, software-driven correction, or additive manufacture.
7. Author-inventor admissions. "Most device solutions include implants… having standard sizes or shapes" and the Fig. 3–4 plate construct are admitted prior art, narrowing the distance the asserted art must travel.

8. Anticipated patent-owner responses and rebuttals

• "Medicrea is 2-D x-ray-based; the claims require a 3-D corrected model." Rebutted by CHW's segmented 3-D CAD model + virtual surgery and Spinemedica's 3-D display of a post-surgical effect; and Medicrea itself defines "a two- or three-dimensional model of the rod."
• "Claim 1 requires displaying predicted metrics, not merely computing them." Rebutted by Spinemedica's "graphically simulating a post-surgical effect on posture, height and/or wedge angle" and CHW's suggested revisions/post-operative models.
• "The specific combination is not taught." Not the standard post-KSR; the inquiry is whether the combination was obvious, not whether it was explicitly spelled out.
• Objective indicia. Carlsmed may assert commercial success, industry praise, and the issued claim set. To date I found no evidence of unexpected results, copying, licensing, or nexus evidence in the public record; note the earlier sections' finding of no PTAB history and an untested claim set. Any such evidence should be tested for nexus to the claimed multi-planar-element features rather than to the plate/locking-element disclosure.

9. Bottom line and confidence

• Strongest ground: Medicrea (US 2016/0210374 A1; US 2016/0235479 A1; US 9,693,831 B2) + US 2012/0150243 A9 + Biomet/Conformis, with Pimenta 2015 for the metric-based guideline limitations. This combination maps to every element of claims 1, 11, and 17 and to the dependent-claim themes.
• Confidence: moderate-high on the substantive mapping; the § 103 case looks materially stronger than the patentee's presumption of validity suggests, principally because the claims were deliberately drafted away from the disclosed plate/locking-element embodiment (as the earlier sections flagged) and onto a functional, metrics-based longitudinal element for which a dense pre-2018 literature exists.
• Cannot confirm: (i) whether Medicrea appears in the unshown ~184 entries of the '952 citation list (the supplied text truncated after US 2014/0086780 A1); (ii) whether any of these grounds was traversed during prosecution of the 16/383,215 → 17/880,277 → 18/754,101 chain — the file histories are needed to know what the Examiner already considered and what the applicant argued; (iii) full-text claim charts against Medicrea, which require reading the complete Medicrea specifications rather than the abstracts/claims I retrieved.
• Recommended next verification: pull the 254-entry citation list in full, the prosecution histories of US 11,439,514 and US 12,251,320, and the complete specifications of US 2016/0210374 A1 and US 2016/0235479 A1 to build element-by-element charts. Also confirm whether the Carlsmed family generates obviousness-type double patenting exposure among '952, '320, and '514 — a related vulnerability raised by the earlier continuity analysis.