US 9664587

I'll search for information on this specific patent number.

I found solid confirmation of the patent. Let me check for any CAFC 2026 docket activity or litigation.

No litigation hits so far. Let me run a few more targeted checks to be thorough.

US Patent 9,664,587 B2 — Summary

Important caveat up front: I could not access USPTO PatentCenter or the CAFC docket system directly from this session; the findings below come from the full patent text supplied (Google Patents) plus corroborating public sources (Justia, FreePatentsOnline, Google Patents family data). I found no evidence of any CAFC 2026 docket, district-court litigation, or PTAB/IPR proceeding involving 9,664,587. That is an absence of found evidence, not authoritative proof that none exists.

Bibliographic data

Abstract

A method and a device for balancing a CT gantry are provided. The method includes: obtaining a fluctuation chain of pulses generated in one revolution of the CT gantry and obtained when the CT gantry collects data; obtaining an eccentric angle of a center of mass according to the fluctuation chain, and calculating a magnitude of imbalance; and adjusting weight at a weight counterbalancing position according to the eccentric angle and the magnitude of imbalance, to locate the center of mass at the rotation center. The device includes a data collection unit for collecting a fluctuation chain, and a processor for obtaining an eccentric angle of a center of mass and calculating a magnitude of imbalance; the processor adjusts weight at the weight counterbalancing position accordingly.

Plain-language overview of the independent claims

Claim 1 — Method (independent). A three-step method for balancing a CT gantry:

1. Obtain a "fluctuation chain" of pulses generated over one full revolution, taken from data the CT gantry itself already collects during rotation.
2. Derive an eccentric angle (the angular position at which the center of mass sits off the rotation axis) from that fluctuation chain, and calculate a magnitude of imbalance.
3. Adjust weight at a weight-counterbalancing position using that angle and magnitude so the center of mass moves onto the rotation center.

The claim also has a negative/structural limitation built into step 1: sampling is performed once each time the gantry rotates by a constant angle, a pulse curve is obtained for each sampling, and those curves are connected successively in the rotational direction to form the fluctuation chain.

Key concept: when an unbalanced gantry spins, it slows slightly where the heavy side passes and speeds up elsewhere, so pulse durations vary. That varying pulse-duration pattern is the "fluctuation chain," and the patent uses it as the imbalance signal — avoiding a separate balance-measurement device.

Claim 10 — Device (independent). A balancing device for a CT gantry that already has (a) a data collection unit and (b) a weight-counterbalancing position. The device comprises a processor in signal connection with the data collection unit, where:

• the data collection unit collects the fluctuation chain of pulses generated in one revolution;
• the processor derives the eccentric angle of the center of mass from the fluctuation chain and calculates the magnitude of imbalance; and
• the processor adjusts the weight at the counterbalancing position to bring the center of mass to the rotation center.

Like claim 1, claim 10 includes the constant-angle sampling / successive-curve-concatenation requirement to form the fluctuation chain.

Dependent claims (grouped)

• No-X-ray data (claim 2): the sampled data is collected without X-ray exposure (reduces radiation and saves energy).
• Fourier/math handling (claims 3, 6, 8): step 2 comprises performing a Fourier transform on the fluctuation chain, retaining only the fundamental frequency component (other harmonics assumed zero) to get a cosine function whose phase angle is the eccentric angle; then calculating imbalance magnitude from that cosine function and angle.
• Specific imbalance equation (claims 4, 7, 9): magnitude of imbalance = m·r, computed as |P→| = 2πJ(tᵢ − tᵢ₊₁) / (gN tᵢ₊₁ tᵢ² sin α), where P→ is the imbalance vector, J rotational inertia, N the number of constant angles per revolution, tᵢ/tᵢ₊₁ durations for the i-th and (i+1)-th constant angles, g gravity, α eccentric angle.
• Steady-state pre-step (claim 5): before step 1, adjust the torque voltage of the drive motor so the gantry rotates in a steady state.
• Real-time velocity detector (claim 11): a detector senses gantry rotation velocity in real time, is in signal connection with the processor, and the processor adjusts motor torque voltage based on that velocity to reach a steady-state rotation.

Provenance / supporting URLs

• Full text: https://patents.google.com/patent/US9664587/en
• Pre-grant publication (Justia): https://patents.justia.com/patent/20150185107
• FreePatentsOnline: https://www.freepatentsonline.com/y2015/0185107.html
• Chinese counterpart CN103767725B: https://patents.google.com/patent/CN103767725B/en
• Assignee listing: https://onscope.com/ipowner/en/owner/ip/[945474](/patent/945474)-neusoft-medical-systems-co-ltd.html

Explicit uncertainties

1. No CAFC 2026 docket or litigation/PTAB record found for 9,664,587 in the searches performed. I cannot certify a negative; confirm via PACER/CM-ECF and the CAFC docket (and USPTO PatentCenter for assignment/status) if a definitive litigation check is required.
2. Legal status ("Active," 2035-07-17 adjusted expiration) is Google Patents' stated assumption, not a legal conclusion; verify current fee/status at USPTO PatentCenter.
3. Two claim sets exist: the pre-grant publication US 2015/0185107 A1 (as reproduced on Justia) shows a 6-claim set with different numbering (step "11)/12)/13)" and no constant-angle-sampling limitation in claim 1), whereas the granted patent has 11 claims with the constant-angle-sampling limitation added to independent claims 1 and 10. The granted-claim text above is the authoritative version.
4. I did not independently verify the assignee's corporate name change beyond the recorded assignment events listed in the patent data.

I'll search for litigation involving US patent 9664587 across patent litigation sources.

Let me search additional litigation-specific sources.

Litigation Search Results — US Patent No. 9664587

Bottom line: I found no known litigation involving US Patent No. 9664587.

What I searched (and what came back)

Patent identity (for the record, per the authoritative patent text)

• Patent: US 9664587 B2 — "Method and device for balancing CT gantry"
• Application: US 14/513,219; filed 2014‑10‑14; granted 2017‑05‑30
• Priority: CN 201310752409.4, filed 2013‑12‑31 (Chinese counterpart: CN103767725B)
• Inventors: Shanshan Lou; Junfang Ma
• Current assignee: Neusoft Medical Systems Co., Ltd. (originally Shenyang Neusoft Medical Systems Co., Ltd.)

None of the Google Patents "Cited By," "Families Citing this family," or "Citations" entries are litigation — they are patent-document citations (e.g., Tsinghua University US10935691B2 / US11327198B2, Shanghai United Imaging US10702221B2), not lawsuits.

⚠️ Important disambiguation caveat

A search hit surfaced an unrelated case in which the shorthand "the '587 Patent" appears — Invertery v. Apple‑type pleadings concerning EGPRS/bit‑modulation technology (asserting, e.g., an '815 patent and a different '587 patent). That is not US 9,664,587. Per your strict instruction not to auto‑correct or conflate identifiers, I am flagging this explicitly: that '587 reference should not be attributed to US 9664587.

Status / confidence

• Known litigation: none found. No district court case, CAFC appeal, ITC Section 337 investigation, or PTAB proceeding (IPR/PGR) involving US 9664587 surfaced in any search.
• Likely reason (inference, not established fact): the patent is held by a Chinese medical‑imaging OEM (Neusoft) and appears to be practiced internally rather than asserted — consistent with the absence of US enforcement activity.
• Confidence level: moderate. I could not directly query the paywalled/portal databases that would be definitive (Unified Patents Portal litigation caselist, Lex Machina, Docket Navigator, RPX Insight, PACER). My conclusion rests on open‑web, CourtListener, and RPX‑sourced searches only.
• Recommendation: if you need a litigation‑grade clearance, confirm against Unified Patents Portal (portal.unifiedpatents.com/litigation), Lex Machina, Docket Navigator, and a PACER full‑text search for "9,664,587" before relying on this.

I will not fabricate a case number, court, or party for this patent. If no litigation exists, as my searches indicate, that is the accurate answer.

PTAB / AIA Trial Activity — US 9,664,587 B2 ("Method and device for balancing CT gantry")

Owner: Neusoft Medical Systems Co., Ltd. (originally Shenyang Neusoft Medical Systems Co., Ltd.)
Inventors: Shanshan Lou; Junfang Ma
Priority: 2013-12-31 (CN 201310752409.4) · Filed: 2014-10-14 · Granted: 2017-05-30 · Adjusted expiration on file: 2035-07-17
Source of record: US9664587B2 on Google Patents · USPTO ODP structured block (PTAB proceedings)

---

Proceedings overview

Total AIA trial proceedings on file: ZERO. The canonical structured list supplied from the USPTO Open Data Portal returns no IPR, PGR, or CBM proceeding for US 9,664,587, and independent web searching surfaced none — no petition, no institution decision, no Final Written Decision, no appeal. The breakdown is therefore 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denied, and the bottom line for a defendant is: the patent is un-attacked and intact — all 11 claims stand exactly as granted, and there is no PTAB record, no claim-cancellation order, and no estoppel to hand you. Any defense must be built from scratch, most likely in the district court.

---

Proceedings detail

No proceeding numbers exist to report. Per the operating rule, I do not invent them. For completeness, what was checked and found empty:

I cannot affirmatively rule out a very recently filed petition that neither ODP's ingest nor public search has indexed yet; that is the one residual uncertainty. But on the record available, there is no AIA trial history for this patent.

---

Strategic summary

1. Claim status — everything is UNTESTED. Claims 1–11 have never been construed or adjudicated by the Board. No independent claim has been canceled; no dependent claim has been canceled; no claim has been confirmed. The full claim set is:

• Claims 1–9 (method): independent claim 1 (obtaining a fluctuation chain of pulses in one revolution; deriving eccentric angle + imbalance magnitude; adjusting weight at a counterbalancing position), narrowed by the claim-1 requirement of "performing sampling once each time the gantry rotates by a constant angle" and connecting pulse curves in the rotational direction. Dependents add: no-X-ray data collection (2); Fourier transform with fundamental-only retention, phase angle = eccentric angle (3, 6, 8); the specific imbalance-magnitude equation (4, 7, 9); pre-collection torque-voltage stabilization to steady state (5).
• Claims 10–11 (device): processor in signal connection with the gantry's own data-collection unit, plus the constant-angle sampling limitation (10); real-time rotation-velocity detector feeding back to motor torque voltage (11).

The practical consequence: the "constant angle sampling → per-angle pulse curve → successively connected fluctuation chain" element in claims 1 and 10 is the load-bearing limitation. A § 102/§ 103 case that does not squarely address uniform-angular-sampling pulse-duration chains will not reach the claim.

2. Estoppel landscape — wide open. Because no IPR/PGR was ever instituted, 35 U.S.C. § 315(e)(2) estoppel does not attach to anyone, and neither does § 325(e)(1) post-grant estoppel. A defendant today may raise any § 102 or § 103 ground based on patents and printed publications, on any claim, with no prior-IPR bar. Note two timing points: (a) this is a post-AIA patent (filed 2014-10-14, priority 2013-12-31), so IPR grounds are limited to § 102/§ 103 over patents and printed publications — § 112 and public-use/on-sale art are unavailable in an IPR; (b) the PGR window is long closed (grant 2017-05-30 → 9-month deadline 2018-02-28), and CBM is inapplicable to a CT gantry balancing patent since it is not a "covered business method" claim.

The prior art already of record in the file — and therefore fair game to build on or distinguish — includes US 6,748,806 B2 (GE, "Dynamic balancing system for computed tomography gantry"), US 6,890,100 B2 (GE, "CT gantry balance system"), US 7,236,855 B2 (Siemens), CN 102809464 A (dynamic balance measurement method/device and CT apparatus), WO 2012/095773 A2 (Philips), CN 1575767 A, CN 1828250 A, CN 102772218 A, US 2009/0046835 A1 (Rigaku), and CN 203191158 U. Family-art cited on the sibling GE/Philips side — US 6,412,345 B1 and JP 2004-065477 A — is also worth mining, since the core idea (extract imbalance from a rotating gantry's own angular-position/timing data rather than a dedicated balancer) is old in this field.

3. Pattern signals — none. There is no serial petitioner (no company has filed even one, let alone multiple, IPRs against this patent), no Neusoft appellate history tied to Board decisions on the '587 patent, and no defensive aggregator (Unified Patents, RPX, or similar) visible in the chain. Neusoft's US portfolio in this space has been challenged from the other direction — Neusoft patents appear as cited art in third-party PCT search reports (e.g., CN 103489496 A, US 9,138,196 B2) — but the '587 patent itself has never been put in front of the Board. Notably, a competitor patent in the same technical lane, US 10,702,221 B2 (Shanghai United Imaging Healthcare, priority 2015-12-25, "Methods and systems for CT balance measurement and adjustment"), was filed roughly two years after this patent, which suggests the space is active commercially without anyone choosing to litigate the '587 patent at the Board.

---

Recommended next steps

If you are a defendant facing an assertion of US 9,664,587:

1. Do not expect a ready-made invalidity kill shot — there isn't one. No claim is canceled, no FWD exists to link or quote, and no estoppel protects you. Everything must be built in your own forum.
2. Ask a threshold question first: has the patent ever actually been asserted? I found no district court complaint, ITC action, or demand-letter campaign tied to the '587 patent in public sources. If a demand letter cites claims 1–9 or 10–11, the underlying theory will hinge on the constant-angle-sampling/fluctuation-chain limitation — the demand is worth taking seriously, and worth countering with a Rothschild-style diligence check on real damages exposure, because a never-asserted, un-attacked patent with no litigation history is often priced for settlement rather than for trial.
3. If you do litigate invalidity, focus § 102/§ 103 art on: (i) CT/DXA systems that derive rotation-timing or angular-position data from their own encoder/sampling chain (the GE '806 and '100 patents and CN 102809464 A are the natural starting points); (ii) the routine Fourier-fundamental extraction and the imbalance-magnitude equation, which is applied mathematics with a well-known eccentric-mass torque model (M = Jα = mgr sin θ → |P| = mr = Jα/(g·sinα)); and (iii) uniform-angular-sampling diagnostics in adjacent rotordynamics art. If you are sued in district court, consider an IPR filed within one year of service of the complaint (§ 315(b)) — the patent is post-AIA, so § 102/§ 103 patent-and-printed-publication grounds are available, and there is no Fintiv-style prior-proceeding history to weigh against institution.
4. If you want the practical defensive posture in one line: no IPR has ever been filed, no claim has ever been tested, and the patent runs to 2035 — you are in the earliest, most expensive phase of the fight, not the cheapest.

If you are monitoring rather than defending: the absence of any AIA trial against a patent that issued in 2017 and expires in 2035 is itself the signal — this patent has not been used as a litigation lever in a way that would attract an IPR. Watch for a first petition to appear in PTAB E2E / ODP once any real assertion occurs; that filing, not this one, will generate the claim-level record.

---

Caveats: The ODP structured block is the canonical proceedings list and it is empty; I have not fabricated proceeding numbers, panels, or dispositions to fill the template. Web search indexes PTAB petitions and FWDs imperfectly for older, low-volume patents, so treat "no activity" as accurate as of the most recent ingest and public search (queried 2026-09-30) rather than as an absolute. Verbatim source for the patent record: https://patents.google.com/patent/US9664587/en; for the family, CN103767725B.

I'll research the assignment record for US 9664587, cross-referencing the patent's own legal-events data with assignment databases and NPE directories.

Let me dig into the corporate history of the assignees and search for the assignment correspondent of record.

I've hit the tool-call limit, but I have enough from the patent's own legal-events record plus corroborating public sources to reconstruct the chain. Below is the analysis. Note on primary source: I could not open the USPTO Assignment Center UI directly in this session — the reel/frame entries below are those recorded in the granted patent's legal-events data (Google Patents, mirroring USPTO assignment records). The correspondent-of-record field is not exposed in the data I could retrieve, so I flag that explicitly rather than guess.

---

Inventors

Pattern check: No unusual pattern. Neither inventor shows a departure from the original assignee within 12 months of filing — Lou in particular continues to appear as a Neusoft inventor on later filings (through at least 2020, e.g., US 17/030,014 for a photon-counting CT system). This is a normal employee-inventor assignment to an operating company, not a founder/portfolio-exit pattern.

---

Original assignee

• Entity on the issued patent: Shenyang Neusoft Medical Systems Co., Ltd. (China), later renamed Neusoft Medical Systems Co., Ltd. (China) — see assignment below.
• Product embodiment: Yes. Neusoft Medical is a Chinese medical-imaging OEM that designs and manufactures CT scanners, MRI, ultrasound, X-ray, PET and LINAC systems. Gantry balancing is an in-house manufacturing/design function directly implicated by CT scanner production, so the claims are practiced in the assignee's own CT gantry manufacturing, not licensed out.
• Primary line of business: development, manufacturing and sale of medical imaging equipment, plus device-service/training and IVD reagents (four segments per its HK IPO prospectus).
• Corporate trajectory / current status: Operating. Founded 1998; once a subsidiary of Shanghai-listed Neusoft Corporation (600718.SH); spun off in 2016; Neusoft Corporation remains largest shareholder (~30%). The 2020 "change of name" record aligns with its conversion to a PRC joint-stock company and a series of (so-far unsuccessful) Hong Kong / STAR Market IPO attempts in 2020–2022. No bankruptcy, no dissolution, no fire-sale.

---

Assignment timeline

Two records appear in the patent's legal-events data. Because the correspondent field could not be retrieved, each entry flags that gap.

• 2014-10-10 (executed) / recorded 2014-10-14 — Reel 033976/0097

  • Conveyance: Assignment ("ASSIGNMENT OF ASSIGNORS INTEREST")
  • Assignor: Lou, Shanshan; Ma, Junfang (individual inventors)
  • Assignee: Shenyang Neusoft Medical Systems Co., Ltd.
  • Correspondent: Not retrievable from available sources. ⚠️ The patent's attorney of record is Kilpatrick Townsend & Stockton LLP (per the bibliographic summary), but I cannot confirm that Kilpatrick Townsend is the recording correspondent on this reel/frame — the two roles are often (not always) the same firm. Treat as unverified.
  • Context: Employment/standard inventor assignment at filing — the inventors assigned to their employer contemporaneously with the US national-stage filing.

• 2020-03-29 (effective) / recorded 2020-04-14 — Reel 052398/0489

  • Conveyance: Change of Name ("CHANGE OF NAME; ASSIGNOR: SHENYANG NEUSOFT MEDICAL SYSTEMS CO., LTD.")
  • Assignor: Shenyang Neusoft Medical Systems Co., Ltd.
  • Assignee: Neusoft Medical Systems Co., Ltd.
  • Correspondent: Not retrievable. No recurrence can be assessed (single link).
  • Context: Internal corporate reorg / name change only — no change in beneficial ownership; the same enterprise remains owner. Consistent with the company's 2020 conversion to a PRC joint-stock company ahead of its IPO attempts.

No post-2020 records appear (no transfer to an LLC, no security agreement, no license, no release). Per the task framing: there are not "zero" assignments, but the only conveyance ever recorded is the original employee assignment plus a same-owner name change. The chain has never left the operating company.

Correspondent recurrence: Cannot be assessed — the correspondent field is missing from the retrievable data, and there is no second conveyance (only a name change) against which to check recurrence. If you need this field, pull reels 033976/0097 and 052398/0489 directly at the Assignment Center.

---

Timeline diagram

timeline
    title Ownership of US 9664587
    2013 : CN priority filed by inventors
    2014 : US application filed by Shenyang Neusoft
         : Inventors assign to Shenyang Neusoft
    2017 : Patent issued to Shenyang Neusoft
    2020 : Name change to Neusoft Medical Systems
         : Same owner reorganized for IPO

---

NPE / troll-pattern signals

1. Shell-entity transfer — Not present. The only non-inventor record (Reel 052398/0489) runs from Shenyang Neusoft to Neusoft Medical Systems as a Change of Name, not a sale to a "Holdings/Ventures/Licensing" vehicle. No single-purpose LLC appears anywhere.
2. Known asserter in the chain — Not present. Neither assignee matches any public NPE list (Acacia, Marathon, IV, Wi-LAN/Conversant, Pendrell, etc.). Both are the same Chinese medical-imaging OEM.
3. Repeat correspondent across the chain — Unclear / cannot be assessed. Correspondent data unavailable from the sources retrieved; there is no multi-link conveyance chain in any event, so recurrence cannot arise.
4. Cascading transfers — Not present. One reorg link in six years; no <24-month LLC chains, no shared correspondent addresses.
5. Pre-litigation transfer — Not present. No litigation involving US 9,664,587 was found (see the prior litigation summary), so there is no assignment dated within 6 months of a suit.
6. Bankruptcy fire-sale — Not present. Assignee is an operating company with disclosed revenue (¥2.8B in 2021 per its HK prospectus), not in Chapter 7/11.
7. Privateering — Not present. No operating-company→NPE transfer; ownership stays in-house.
8. Defensive aggregator (anti-NPE) — Not present. The chain does not terminate at RPX/AST/LOT/Unified/OIN.

---

Verdict

Insufficient data (for an NPE finding) — the record is minimal and shows no NPE pattern.

Justification: The chain consists of exactly two records — the original inventor→employer Assignment (Reel 033976/0097, executed 2014-10-10, recorded 2014-10-14) and an intra-company Change of Name (Reel 052398/0489, effective 2020-03-29, recorded 2020-04-14). Ownership never left the operating company (Shenyang Neusoft → Neusoft Medical Systems, same enterprise), no litigating vehicle or licensing LLC ever appears, and no litigation or assertion activity was found. The closest enumerated label that fits the evidence is a non-asserting operating company; because none of the enumerated NPE/defensive-aggregator verdicts cleanly applies to a bare reorg-only chain, Insufficient data is the honest call for NPE purposes — with the affirmative finding that no NPE/troll signals are present.

Verification link: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search "9664587" or the reel/frame entries 033976/0097 and 052398/0489 to retrieve the correspondent-of-record fields I could not access). Mirrored index: https://assignment.uspto.gov/patent/index.html

Explicit caveats / contradictions to flag:

• Correspondent field unavailable — I did not fabricate it. The patent's attorney of record (Kilpatrick Townsend & Stockton LLP) is not the same as the recording correspondent; do not conflate them without the reel/frame pull.
• The reel/frame numbers 033976/0097 and 052398/0489 are taken from the patent's legal-events record; confirm against the Assignment Center in case of indexing drift.
• This is consistent with — and reinforces — the earlier litigation summary's finding of no known litigation and its disambiguation caveat (the unrelated "'587 Patent" references in certain pleadings are not US 9,664,587).

I'll search for the USPTO record and then examine each cited prior-art reference.

Prior-Art Analysis — US Patent 9,664,587 B2 (“Method and device for balancing CT gantry”)

Scope note / method. I was unable to reach USPTO PatentCenter directly in this session; the patent’s citation list below is taken from the authoritative full patent text supplied (Google Patents, https://patents.google.com/patent/[US9664587](/patent/US9664587)/en), and each reference was researched on the open web. Per your standing instruction I have not normalized or corrected any identifier; all numbers are reproduced literally as they appear. Where my search returned only partial content (mainly the Chinese/Japanese references), I say so explicitly rather than filling gaps with inference presented as fact. Per the instruction to build on prior sections, the bibliographic and litigation findings already established are not repeated here — this section is limited to the prior-art citations and their §102 potential.

---

1. USPTO record for 9664587 — the patent being analyzed

I confirmed the USPTO-mirrored text at Justia (https://patents.justia.com/patent/9664587) and FreePatentsOnline (https://www.freepatentsonline.com/y2015/0185107.html). I did not retrieve the actual USPTO PatentCenter page; the claim text used here is the granted text from the supplied authoritative document.

---

2. Cited references — the substantive prior art

The patent’s citation list (the 10-item “Patent Citations” table, which is the fuller of the two lists shown) plus the 3-item “Family Cites Families” table yields 13 references. These are the only references the applicant/examiner placed of record. Each is treated below.

A. US 6,748,806 B2 — Dynamic balancing system for computed tomography gantry

• Assignee / inventor: GE Medical Systems Global Technology Company, LLC (inventor Halsmer, Matthew A.)
• Filing date: 2002-02-27; granted: 2004-06-15 (app. US 10/084,696; pub. US 2003/0159508 A1)
• Description: A CT gantry carries at least one electronically positionable weight attached to the gantry and movable along the gantry axis via a received position signal to correct imbalance caused by component variation. The weight moves along two axes (parallel to the gantry axis and radial to it), each with an independent position signal.
• §102 potential: Against claim 1 — no. This reference corrects imbalance by moving a weight; it does not teach obtaining a fluctuation chain of pulses from the gantry’s own data collection or deriving an eccentric angle from that chain. No anticipation.
• Against claim 10 — partially relevant but not anticipatory. It discloses a gantry with a weight-correcting element, but the claim-10 processor/method of deriving eccentric angle and imbalance magnitude from the fluctuation chain is absent. Best characterized as §103 background, not §102 art.
• Source: https://patents.google.com/patent/US20030159508A1/en

B. CN 1,575,767 A — CT gantry balance system and method (GE)

• Assignee: GE Medical Systems Global Technology Company; pub. date: 2005-02-09; priority 2003-07-18
• Description: Chinese counterpart of the US 6,890,100 B2 family (see C). Balances the rotating base member using two preselected balance locations and stacks of spacers/shims/weights, with strain-gauge sensors on the stationary base member and trial/test runs to determine the mass and Z-location to be added.
• §102 potential: None against claims 1/10. Like US 6,748,806, it is a weight-adjustment approach driven by separate sensors (strain gauges) and trial runs — the opposite of the patent’s “no additional device / use the gantry’s own collected data” thrust. It cannot anticipate either independent claim. Cited most plausibly as context for the balancing hardware.
• Source: https://patents.google.com/patent/CN1575767A/en

C. US 6,890,100 B2 — CT gantry balance system

• Assignee: GE Medical Systems Global Technology, Inc.; filed: 2003-07-18; granted: 2005-05-10 (pub. US 2005/0013403 A1)
• Description: Two balance locations on the rotating base member hold stacks of spacers, shims, and weights. Strain-gauge sensors on the stationary base member feed a solver; trial and test runs determine the mass and Z-location needed to statically and dynamically balance the gantry. A twice-per-revolution signal checks for a broken strain gauge. The solver outputs an exact “recipe” of spacers/shims/weights.
• §102 potential: None against claims 1/10. This is the closest cited art on the balancing mechanics and solver side, and it explicitly relies on dedicated sensors (strain gauges) — precisely the “additional device” the patent says it avoids. It does not disclose the fluctuation-chain/constant-angle-sampling limitation at all. §103 candidate only, and even then it does not suggest using gantry-collected data for imbalance.
• Sources: https://patents.justia.com/patent/[6890100](/patent/6890100) ; https://www.freepatentsonline.com/6890100.html
• ⚠️ Note: US 6,890,100 B2 appears in the 10-item citation list but is absent from the 9-item list shown elsewhere on the same page. I flag this inconsistency in the source rather than resolve it.

D. US 7,236,855 B2 — Method for compensating an out-of-balance condition of a rotating body

• Assignee / inventors: Siemens Aktiengesellschaft (Danz, Günter; Müller, Hans-Jürgen); filed: 2004-01-28; granted: 2007-06-26 (pub. US 2005/0192709 A1)
• Description: Measures vibrations transferred to the stationary unit by an out-of-balance gantry, determines the out-of-balance vector(s), and adjusts the axis of the measurement device relative to the bearing axis to compensate — no compensation weights required. Sensors measure vibration (radial and axial) and a further sensor determines rotation angle. Adjustment elements (piezoelectric/magnetostrictive) in two parallel planes can compensate radial vectors; compensation can be dynamic during rotation.
• §102 potential: Against claim 1 — no. Although it computes an out-of-balance vector, it does so from dedicated vibration sensors, not from a fluctuation chain of the gantry’s own collected pulses, and it corrects by axis adjustment, not weight adjustment at a counterbalancing position. Both the “obtain fluctuation chain from gantry-collected data” and the “adjust weight at weight counterbalancing position” limitations are missing. No anticipation.
• Against claim 10 — no. The processor is present, but again the data source and the correction mechanism differ.
• Source: https://patents.google.com/patent/US7236855

E. CN 1,828,250 A — [Identifying methods for stable establishment of systems] (Siemens)

• Assignee: Siemens; pub. date: 2006-09-06
• Description (limited): My searches returned only the title/assignee/date; I did not obtain the specification text, so I cannot characterize its disclosure with confidence. The English title in the citation list (“Identify methods for stable establishment of systems”) suggests a system-stability/identification subject matter, but I will not assert what it teaches.
• §102 potential: Cannot be assessed — insufficient data. I explicitly decline to speculate. If a §102 opinion is needed, the full CN 1828250 A text should be pulled and a machine translation obtained.
• Source: presence confirmed only via the citation list at https://patents.google.com/patent/US9664587/en

F. US 2009/0046835 A1 — X-Ray CT Apparatus (Rigaku Corporation)

• Filed/priority: 2005-12-20; published: 2009-02-19 (JP priority 2004-184326; JP 2006-38836 A)
• Description: Industrial X-ray CT with a guide arm (rotational support member) carrying the X-ray source and detector on opposite sides of the sample; the arm rotates about a horizontal axis. The arm is designed so the center of gravity of all mounted components lies on the rotational axis, and a movable weight (weight-balance adjusting mechanism) automatically offsets the rotational-moment difference to ~zero as the source/detector move. The movable weight’s passage is a straight line parallel to the source/detector motion and not passing through the rotational axis.
• §102 potential: Against claim 1 — no. This is a gravity-center-on-axis design with a movable counterweight driven by geometry of source/detector motion — not by a measured pulse fluctuation chain, and there is no Fourier/eccentric-angle computation. Against claim 10 — no for the same reason.
• Closest conceptual overlap: it shares the goal (keep center of gravity on the rotation axis) and the use of a movable weight. So it is a §103 background reference, not §102 art.
• Source: https://patents.google.com/patent/US20090046835

G. WO 2012/095773 A2 — Computed tomography (CT) data acquisition (Koninklijke Philips Electronics N.V.)

• Priority: 2011-01-10; published: 2012-07-19
• Description: Generates an integration-period (IP) timing signal for a rotating gantry’s data-acquisition periods based on the time duration of a previous revolution, and corrects predicted IP timing for rotating-gantry imbalance. A gantry-revolution information determiner allows synchronizing IP timing with angular position with/without imbalance correction — explicitly “without using an additional rotating gantry position determining apparatus.” FIG. 8 shows an example error curve representing IP error as a function of angular position.
• §102 potential: This is the most substantively relevant cited reference, but still not anticipatory.
  • Similarities: it addresses rotating gantry imbalance, works from timing of the gantry’s own rotation/data-acquisition cycles, and seeks to avoid an additional position-determining apparatus — thematically close to the patent.
  • Decisive differences for §102: (i) its purpose is correcting data-acquisition (integration-period) timing for image reconstruction, not computing an eccentric angle and imbalance magnitude to drive weight counterbalance; (ii) it does not disclose (a) forming a fluctuation chain from pulse curves each obtained at a constant rotation angle and concatenated in the rotational direction, (b) performing a Fourier transform / fundamental-harmonic extraction to get an eccentric angle, or (c) the specific |P→| = 2πJ(tᵢ−tᵢ₊₁)/(gN·tᵢ₊₁·tᵢ²·sin α) relation (claims 4/7/9); (iii) there is no weight adjustment at a counterbalancing position (claim 1 step 3 / claim 10 processor function).
  • Conclusion: a strong §103 reference (e.g., combined with a weight-balancing reference such as US 6,890,100 B2 or US 6,748,806 B2), but it does not by itself anticipate claims 1 or 10.
• Source: https://patents.google.com/patent/WO2012095773A2/en

H. CN 1,027,722,18 A — [Error identification in computer positron emission tomography instrument] (Siemens)

• Assignee: Siemens; pub. date: 2012-11-14
• Description (limited): Title-level information only. My search did not return usable specification text. Subject appears to concern error identification in a PET instrument, not CT gantry balancing.
• §102 potential: Very low on its face and not assessed on the merits. Given the divergent subject matter (PET error identification), I do not see how it could anticipate claims 1 or 10, but I state this as a prima facie observation, not a verified §102 opinion.
• Source: citation list at https://patents.google.com/patent/US9664587/en

I. CN 1,028,094,64 A — [Dynamic balance measurement method and device and CT apparatus with dynamic balance measurement device] (Shanghai Siemens Medical Equipment Co., Ltd.)

• Assignee: 上海西门子医疗器械有限公司 (Shanghai Siemens Medical Equipment Co., Ltd.); pub. date: 2012-12-05
• Description (limited): Title indicates a dynamic-balance measurement method and device and a CT apparatus incorporating it. This is the Chinese-language counterpart/relative of the balancing-measurement line the applicant explicitly distinguished in its Background (the patent’s Background cites Chinese patent 201110145218.2, “DYNAMIC BALANCE MEASUREMENT METHOD AND DEVICE AND CT MACHINE HAVING THE SAME”). I did not retrieve full claim text, so I cannot map its disclosures to specific claim elements.
• §102 potential: Potentially the closest “measurement-of-imbalance” art among the Chinese citations, but not assessable on §102 without the full text. If a rigorous §102 opinion is required, obtain CN 102809464 A (and CN 102809464 A’s family) and map independent claims. On the record available, I would treat it as §103 background at most.
• Source: citation list at https://patents.google.com/patent/US9664587/en

J. CN 2,031,911,58 U — Portable dynamic balance tester

• Assignee: 苏州工业园区聚博精密设备有限公司; pub. date: 2013-09-11
• Description (limited): A utility model for a portable dynamic balance tester. Title-level only; a general-purpose portable balancer, not CT-specific.
• §102 potential: None. A portable balance tester is the archetype of the “additional device other than the CT machine” that the patent’s Background distinguishes and that the patent claim set avoids. It cannot disclose obtaining imbalance from the gantry’s own data-collection fluctuation chain. No anticipation of claims 1 or 10.
• Source: citation list at https://patents.google.com/patent/US9664587/en

K. US 6,412,345 B1 — Balancing of rotational components of CT imaging equipment (Family Cites)

• Assignee: GE Medical Systems Global Technology Company, LLC; filed: 2000-09-29; granted: 2002-07-02
• Description (limited): Family-level citation; title indicates balancing of CT rotational components. I did not retrieve the full specification in this session.
• §102 potential: Not assessed on the merits; on its face it is a balancing-mechanics reference (the same class as US 6,748,806/US 6,890,100) and would be §103 background, not §102 art, for the same reason: no disclosure of a gantry-data fluctuation chain.
• Source: citation list at https://patents.google.com/patent/US9664587/en

L. JP 2004-065477 A — Method for rotation balance adjustment and X-ray CT apparatus (GE)

• Assignee: GE Medical Systems Global Technology Co., LLC; pub. date: 2004-03-04
• Description (limited): Family-level citation; title indicates rotation-balance adjustment in an X-ray CT apparatus. Full text not retrieved.
• §102 potential: Not assessed on the merits. Title indicates a balance-adjustment method; if it computes imbalance from gantry rotation data rather than a separate sensor, it could be more relevant, but I have no verified basis to say so. Treat as unassessed.
• Source: citation list at https://patents.google.com/patent/US9664587/en

M. CN 1,020,460,90 B — Air bearing dynamic Z-axis balancing (Royal Philips Electronics / 皇家飞利浦电子股份有限公司)

• Filing date in citation list: 2008-06-03; granted pub. date: 2014-08-27
• Description (limited): A granted Chinese patent on air-bearing dynamic Z-axis balancing for CT. Full text not retrieved in this session.
• §102 potential: Not assessed on the merits. Subject matter (Z-axis/air-bearing balancing) appears directed to an aspect of rotor balancing distinct from the patented fluctuation-chain method, but I decline to opine without the text.
• Source: citation list at https://patents.google.com/patent/US9664587/en

---

3. Consolidated §102 assessment

Bottom line: On the evidence gathered, none of the 13 cited references anticipates (i.e., discloses every limitation of) claim 1 or claim 10 under 35 U.S.C. §102. The references cluster into two families:

1. Balancing-mechanics references (US 6,748,806; US 6,890,100; CN 1,575,767; US 6,412,345; JP 2004-065477; CN 2,031,911,58 U) — they adjust weights or axes, and all rely on dedicated sensors or trial runs, which is the very “additional device” the patent avoids. They lack the fluctuation-chain/constant-angle-sampling limitation entirely.
2. Measurement/compensation references (US 7,236,855; CN 1,028,094,64 A; CN 1,020,460,90 B) — they measure imbalance by vibration sensing or dynamic-balance instrumentation, again external to the gantry’s own data collection.

WO 2012/095773 A2 (Philips) is the only citation that engages imbalance correction using information intrinsic to the gantry’s rotation/data-acquisition cycle, making it the most plausible §103 combination reference, but it is directed to correcting integration-period timing for reconstruction, not to computing an eccentric angle and imbalance magnitude for weight counterbalance, and it lacks the constant-angle sampling / successive-curve concatenation and the specific magnitude equation.

Novelty edge of the patent (what none of the cited art shows): (a) sourcing the imbalance signal from the fluctuation chain of pulses the CT gantry generates when it collects its own data; (b) the constant-angle sampling + successive concatenation in the rotational direction limitation now recited in both independent claims; (c) Fourier transform retaining the fundamental harmonic to extract the eccentric angle (claims 3/6/8); and (d) the |P→| = 2πJ(tᵢ−tᵢ₊₁)/(gN·tᵢ₊₁·tᵢ²·sin α) magnitude equation (claims 4/7/9).

---

4. Explicit uncertainties and verification recommendations

1. USPTO database access: I could not query USPTO PatentCenter directly; the citation list and claim text derive from the authoritative supplied document (Google Patents) and were corroborated at Justia and FreePatentsOnline. For a filing-grade record, pull the USPTO PTAB/PatentCenter “Patent Application Information Retrieval” page and the printed Notice of Allowance / examiner’s cited-art list.
2. Incomplete texts: I obtained substantive disclosure only for US 6,748,806 B2, US 6,890,100 B2, US 7,236,855 B2, US 2009/0046835 A1, CN 1,575,767 A, and WO 2012/095773 A2. For CN 1,828,250 A, CN 1,027,722,18 A, CN 1,028,094,64 A, CN 2,031,911,58 U, US 6,412,345 B1, JP 2004-065477 A, and CN 1,020,460,90 B I have essentially title/date/assignee metadata only. I have marked those §102 cells “unassessed” rather than guess. Full-text (with machine translation where needed) is required before any definitive §102 opinion on them.
3. Do not conflate identifiers: per your rule, US 9,664,587 B2 is distinguished from any other “’587”-type shorthand that may appear in unrelated patents/litigation; and the two citation-list variants on the patent page (9 vs. 10 entries) mean US 6,890,100 B2’s presence should be verified against the USPTO printed list.
4. Legal-effect caveat: §102 status here is a technical-literature assessment, not a legal validity conclusion. Whether a reference is “prior art” also depends on dates, §102(a)/(b) category, and any swear-behind/priority issues — none of which I can resolve from this dataset alone.

I'll research the cited prior-art references to ground the obviousness analysis in actual reference disclosures.

Let me research the remaining key prior-art references, especially those relating to deriving imbalance from operating data (velocity/timing) rather than dedicated sensors.

Excellent — the Philips reference is highly pertinent. Let me pull deeper detail on its imbalance-estimation disclosure and check the remaining cited references.

Obviousness Analysis — US 9,664,587 B2 under 35 U.S.C. § 103

Note on scope and sourcing. This analysis uses the prior-art references listed on the subject patent's own Google Patents page (the "Patent Citations," "Family Cites Families," and "Cited By" tables) plus the two Chinese references identified in the patent's own Background section. Where I could retrieve and read reference text, I quote it. Where I could not, I say so explicitly rather than characterizing the reference. I do not auto-correct any identifier; all numbers appear as published on the page.

Consistent with the earlier-generated sections, the granted claim set (11 claims; independent claims 1 and 10) is treated as authoritative, and the added "constant-angle sampling / successive concatenation" limitation in claims 1 and 10 — which is absent from the pre-grant publication US 2015/0185107 A1 — is treated as a real limitation that must be met by the art.

---

1. Framework and the person having ordinary skill in the art (PHOSITA)

Legal standard applied. Graham v. John Deere Co., 383 U.S. 1 (1966) (scope/content of prior art; differences; PHOSITA level; secondary considerations), as refined by KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007) (predictable combinations; known techniques; design incentives and market forces; "a finite number of identified, predictable solutions"). A claim is obvious where the differences between the claim and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made, i.e., on or about the effective filing date December 31, 2013 (US 9,664,587 claims priority to CN 201310752409.4, filed Dec. 31, 2013).

PHOSITA. A person with a bachelor's degree in mechanical engineering, electrical engineering, biomedical engineering, or medical physics, plus roughly 3–5 years of experience designing, manufacturing, or field-servicing rotating-gantry CT scanners; alternatively a master's degree plus 1–2 years of such experience. This person knows: (i) rotor dynamics / dynamic balancing (mass–eccentricity vectors m·r, static vs. dynamic imbalance, two-plane correction at fixed counterweight locations); (ii) CT gantry architecture (rotating base member, stationary base member, slip ring, axial motor, X-ray tube/detector), and (iii) CT data-acquisition electronics, including that projection data are collected in discrete angular "views"/integration periods indexed to gantry angle, and (iv) basic Fourier/least-squares harmonic extraction of a once-per-revolution vector.

That profile matters: the disputed limitations in this patent are measurement/inference and signal-processing features layered onto a well-developed balancing art, not new physical mechanisms.

---

2. The asserted prior art — what each reference actually discloses

I have read the primary references listed below.

A note on the patent's own characterization of the art. US 9,664,587's Background states the technical problem as designing a balancer "which may perform weight counterbalance based on imbalance information obtained without adding an additional device," and criticizes the prior art because "additional devices other than the CT machine are generally required to obtain imbalance information." That framing is the patent's own admission that the only meaningful difference over the art is the source of the imbalance signal, not the balancing mechanism. Philips WO2012095773A2 already answers that stated problem.

---

3. The strongest § 103 combination: WO2012095773A2 (Philips) + US6890100B2 (GE)

I analyze this as the primary combination because both references are in the same field (rotating CT gantry), the combination uses only known techniques arranged in a predictable way, and the result is the claimed subject matter without any asserted unexpected property.

3.1 Element-by-element mapping of claim 1

Conclusion on claim 1: every limitation is disclosed, and the only element not disclosed by Philips itself (physically adding counterweight at a fixed location) is squarely disclosed by GE US6890100B2 and CN102809464A. The combination is a rearrangement of known elements to achieve their known functions.

3.2 Motivation to combine (KSR rationales)

A PHOSITA would have been motivated to combine Philips with GE/CN102809464, on the following independent grounds — any one of which suffices:

1. Same field, same problem, same rotor. All references address imbalance of a rotating CT gantry and its adverse effects on image quality, bearing life, and image artifacts. GE US6890100B2: imbalance "results in an oscillatory motion of the entire gantry mechanism at a frequency of once per revolution. This motion is detrimental to the quality of the images and can generate artifacts." Philips says the same thing more directly: the "dominant disturbance of the rotating gantry velocity is imbalance."

2. Philips supplies exactly the missing input that GE's balancing hardware requires. US6890100B2 requires some once-per-revolution signal; Philips teaches how to derive that once-per-revolution, angle-dependent signal from the CT system's own timing during data acquisition, avoiding a "complex and expensive" high-resolution encoder. Combining the two is the definition of an obvious substitution of one known signal source for another to obtain the same result (KSR).

3. Philips expressly identifies the design incentive and the finite set of solutions. It frames the problem as a "natural rotating gantry position error" attributable to imbalance, and states the solution can be achieved "without using an additional rotating gantry position determining apparatus." This is a stated market/design pressure (cost and complexity reduction) combined with an identified solution — the KSR "finite number of identified, predictable solutions" scenario. US6412345B1 reinforces the pressure by explaining why conventional balancing hardware is ill-suited to CT (large open bore, overhung load, minimum-RPM limits).

4. The subject patent's own admission supplies the motivation. US 9,664,587's Background states the goal is a balancer using "imbalance information obtained without adding an additional device." Philips already accomplishes that; the patent's remaining step (adding weight at the existing counterweight sites) is old (GE/US6890100B2). An invention that merely uses a known technique (Philips' timing-error derivation) to improve a known device (GE's two-site counterweight balancer) in the same way is obvious.

5. Predictable result, no change in principle of operation. Neither reference's operation is disturbed by the combination: GE's weights still act on m·r; Philips' timing data is the same once-per-revolution angle-dependent quantity GE needs. There is no argument that the combination would be inoperative.

3.3 Reasonable expectation of success

Success was predictable: (i) Philips demonstrates that the once-per-revolution imbalance error can be estimated from timing alone, with an explicit error model and least-squares fit; (ii) US6890100B2 demonstrates that a once-per-revolution magnitude-and-phase vector can be converted into a concrete counterweight recipe at fixed sites with a converging solver. Both were proven, working techniques in the same machines. Even a lower-than-perfect estimate is within the claim: claim 1 requires only "obtaining an eccentric angle … and calculating a magnitude of imbalance" and then "adjusting weight." The patent's own specification acknowledges the derivation rests on simplifying assumptions (constant-angle segments, θ = α, gravity-dominated moment) but claims the method nonetheless.

---

4. Dependent claims

Claim 2 — data collected without X-ray exposure

Philips discloses that gantry revolution information may be obtained "before (e.g., pre-scan) and/or during scanning." A pre-scan is by definition an exposure-free revolution. GE US6748806B2's torque-sensor line and US6890100B2's sensor sampling likewise do not require radiation. Reducing patient dose by not emitting X-rays during a calibration/balancing revolution is a conventional and explicitly-recognized benefit (and the patent itself concedes "X-ray exposure has little influence on the imbalance of the gantry"). Obvious.

Claims 3, 6, 8 — Fourier transform; keep fundamental; phase angle = eccentric angle

This is standard harmonic analysis, and the art states it expressly:

• US6890100B2: "the once-per-revolution portion of the signal is extracted as a vector (magnitude and phase angle) … using a least squares method … the method results in the usual Fourier-based equations when the data is equally spaced in time."
• WO2012095773A2: fits the once-per-revolution error to a curve and solves sine/cosine coefficients via least-squares normal equations.

A PHOSITA would unhesitatingly use a Fourier transform (or the equivalent least-squares fit) to isolate the fundamental once-per-revolution component, because the physical excitation is once-per-revolution by definition (a single eccentric mass). Discarding harmonics is the natural and expected step. Obvious.

Claims 4, 7, 9 — the specific magnitude equation |P→| = 2πJ(tᵢ − tᵢ₊₁) / (gN tᵢ₊₁ tᵢ² sin α)

These claims recite a mathematical relationship whose every input is either a known physical constant (J, g, N), a measured quantity (tᵢ, tᵢ₊₁ from the fluctuation chain), or the previously computed α. The underlying physics — that the imbalance moment equals J·(angular acceleration) and equals m·g·r·sin α, so that |P→| = mr = Ja/(g sin α) — is elementary rigid-body dynamics, and the art reflects it:

• US6890100B2 computes imbalance magnitudes from measured forces and moments and then solves for the corrective mass.
• WO2012095773A2 states the imbalance-moment/inertia/velocity relation E(t)I = (Mo/(Jω³))sin(ωτ) and derives angular error from it.
• CN102809464A reduces measured rotation data to "unbalance weight data."

Where a claim recites a formula that merely expresses a known physical relationship among known quantities, and the quantities are obtained by the previously-claimed steps, the claim adds nothing patentable over the art under § 103. (I also note, without deciding, that a claim whose only novel content is a mathematical expression risks § 101 scrutiny under Mayo/Alice; that is outside this task but relevant to overall validity.)

Claim 5 — steady-state pre-step: adjust motor torque voltage before collecting data

• US6748806B2 discloses a torque sensor line from the motor feeding a balance controller, and its balancing procedure precedes/underlies operation — establishing motor-torque monitoring as part of the balancing loop.
• CN102809464A performs a calibration/rating test to determine the workpiece's rotation attributes after start-up and before measurement — the same purpose (stabilize and characterize rotation before measuring).
• Maintaining a rotating body at a steady speed via closed-loop motor voltage control (the patent's "moment" mode) is a ubiquitous, conventional motor-control technique, and the patent concedes the purpose is merely to "guarantee[] the authenticity and effectiveness of the collected data." Routine engineering — obvious.

Claim 10 — device claim (processor + data-collection unit in signal connection)

The device claim is the structural mirror of claim 1, and is met by the same combination:

• Data collection unit configured to collect the fluctuation chain of pulses generated in one revolution → Philips' IP controller/sensors/timer acquiring per-angle time durations during data acquisition.
• Processor in signal connection with the data collection unit → US6890100B2's table gantry processor 110 with analog-to-digital converter 112 and microprocessor 114 loaded with firmware 116, connected to the sensor circuit; US6748806B2's balance controller 44 connected to sensors 64 and motor 26.
• Processor configured to obtain the eccentric angle and calculate the magnitude of imbalance → US6890100B2's extraction of the once-per-revolution vector (magnitude and phase angle) by least-squares/Fourier; Philips' error-curve/sine-cosine solution.
• Processor configured to adjust the weight at the counterbalancing position → US6748806B2's electronically positionable weights moved under controller command; US6890100B2's solved counterweight configuration at two fixed sites.
• "Sampling … once each time the gantry rotates by a constant angle … successively connected to form the fluctuation chain" → same mapping as claim 1's limitation (1a)–(1c).

"Signal connection … in a wired or wireless manner" (the patent's own definition) is a generic, conventional coupling that imposes no patentable weight. Obvious.

Claim 11 — real-time rotation-velocity detector feeding motor torque voltage

• US6748806B2 monitors the motor (torque sensor line) and adjusts weights based on motor-derived signals; US6890100B2 samples sensor output continuously at a fixed rate; CN102809464A uses a rotatable phase sensor on the gantry and a rating test of rotation attributes.
• A detector that reads rotation velocity in real time and closes a loop on motor torque voltage to hold a steady state is textbook servo control; the patent describes it as no more than an improvement to "the accuracy of subsequent data collection."

---

5. Alternative and secondary combinations

Non-analogous-art and "teaching away" defenses are unavailable. Every reference is directed to balancing rotating bodies, most to CT gantries specifically. The closest thing to a "teaching away" is US6412345B1's observation that traditional external balancing machines are unsuited to CT — but that reference itself solves the CT balancing problem and does not disparage deriving imbalance from the gantry's own rotation data. Philips' criticism is of the complex and expensive high-resolution encoder, which favors the claimed approach rather than discouraging it.

---

6. Secondary considerations / rebuttal of non-obviousness arguments

No record of secondary considerations appears on the patent page (no litigation found per the earlier sections; no unexpected-results data; no industry praise or licensing evidence surfaced). Anticipated arguments and responses:

1. "Philips corrects timing; it doesn't balance." Under KSR, "if a technique has been used to improve one device, and a person of ordinary skill in the art would recognize that it would improve similar devices in the same way, using the technique is obvious." Philips uses the same physical signal (once-per-revolution, angle-dependent timing deviation) and expressly names imbalance as its cause; GE US6890100B2 shows the very next step (convert that vector into counterweight mass and position). Using Philips' measurement to feed GE's balancer is an obvious use of a known technique for its established function.
2. "The patent avoids an 'additional device.'" Philips makes the same promise — synchronization "without using an additional rotating gantry position determining apparatus." The patent's advantage is therefore not unexpected; it is disclosed in the art.
3. "The specific equation is new." The equation is a restatement of |P→| = Ja/(g·sin α) with a = (ωᵢ₊₁ − ωᵢ)/tᵢ and ω = φ/t. Its inputs are all measured or constant; the art already computed imbalance magnitude/phase from rotation data (US6890100B2; CN102809464A; WO2012095773A2's E(t)I = (Mo/(Jω³))sin(ωτ)). Obvious.
4. "The granted claims added the constant-angle-sampling limitation." That amendment narrows claim 1 relative to the pre-grant publication, so the validity question is whether the narrowing feature is itself in the art. It is: Philips partitions each revolution into fixed-angular-range integration periods and determines time durations per angular position; GE US6890100B2 extracts the once-per-revolution vector from uniformly/time-equally-spaced samples. The amendment does not create a patentable distinction; it merely recites the sampling scheme the references already use.

Overall probability assessment (informal): For claims 1, 2, 3, 5, 6, 8, 10, and 11, a § 103 rejection over WO2012095773A2 in view of US6890100B2 (and/or US6748806B2 and CN102809464A) appears strongly supported. For claims 4, 7, and 9, the rejection is supported but somewhat more contestable, because an examiner may need to articulate why the specific algebraic form would have been obvious to the PHOSITA, not merely that the underlying physics was known — though a formula expressing a known relationship among known variables is ordinarily obvious and the art supplies the relationship (Philips' Mo/(Jω³) equation).

---

7. Explicit uncertainties and limits of this analysis

1. Several cited references were not read in full. I retrieved and quoted US6748806B2/US2003/0159508A1, US6890100B2/US2005/0013403A1, WO2012095773A2/ US2013/0287165A1, CN102809464A, and US6412345B1. I did not obtain the full text of US7236855B2, US20090046835A1, JP2004065477A, CN1828250A, CN102772218A, CN203191158U, CN1575767A, or CN102046090B; I have not attributed any specific limitation to those documents, and the analysis would need revision if any of them is more on-point (or if any contains a teaching away).
2. Publication dates vs. grant dates. For prior-art purposes I used the earliest publication of each reference (e.g., CN102046090B's underlying application publication date is uncertain — the B-grant is dated 2014-08-27, after the 2013-12-31 priority date; I therefore did not rely on it as the sole source of any limitation). US9173627B2 (granted 2015) is the US family member of WO2012095773A2; the operative prior-art documents are the WO publication (2012-07-19) and US2013/0287165A1 (2013-10-31), both before the priority date. US10935691B2 and US11327198B2 (Tsinghua) appear in the "Cited By" table with 2017 priorities and are not prior art to US 9,664,587.
3. "Fluctuation chain" construction. I have treated "fluctuation chain of pulses generated in one revolution … obtained when the CT gantry collects data" as a sequence of per-angle rotation-timing values. If the claims were construed more narrowly — e.g., as requiring pulse durations of a specific CT data-acquisition encoder/pulse signal, or as excluding any auxiliary flag/sensor — the Philips mapping would weaken, and the analysis would shift toward US6890100B2's fixed-rate sampling of an existing signal combined with CN102809464A. I flag this because the patent's figures were not fully legible from the retrieved text.
4. This is not a validity opinion. It is a § 103 obviousness analysis of the record on the patent page plus the primary references. A formal clearance/validity opinion would require the full file wrapper (prosecution history and any examiner art cited against the claims, including the reasons the constant-angle-sampling limitation was added), expert declaration evidence, and a full-text review of the unread references.

Key URLs relied upon

• Subject patent: https://patents.google.com/patent/US9664587/en
• Philips WO2012095773A2: https://patents.google.com/patent/WO2012095773A2/en
• Philips US9173627 / US2013/0287165A1: https://patents.google.com/patent/US9173627 ; https://patentimages.storage.googleapis.com/97/f3/9b/b48742d4a14e61/US20130287165A1.pdf
• GE US6748806B2 / US2003/0159508A1: https://patentimages.storage.googleapis.com/c7/d9/9b/67891e9f38e218/US20030159508A1.pdf
• GE US6890100B2 / US2005/0013403A1: https://patents.google.com/patent/US6890100B2/en ; https://uspto.report/patent/grant/[6890100](/patent/6890100)
• Siemens (Shanghai) CN102809464A: https://patents.google.com/patent/CN102809464A/en
• GE US6412345B1: https://patentimages.storage.googleapis.com/d6/4d/7a/821d7aa23a58b8/[US6412345](/patent/US6412345).pdf
• Philips air-bearing z-axis balancing (pub. US2013/0308759): https://patents.justia.com/patent/20130308759