US 5168532

I'll search for the specific patent number 5168532 and any CAFC docket references to it.

Let me search more specifically for any CAFC or litigation activity tied to this exact patent number, and verify USPTO records.

Summary — U.S. Patent No. 5,168,532 (US5168532A)

Note on sources: The patent text you supplied (from Google Patents, fetched 2026-09-03) is the most authoritative document I have, and my searches of FreePatentsOnline and Google Patents corroborate it. I could not access the USPTO's internal database or the CAFC docket system directly through search; web searches for this exact patent number in CAFC/2026 litigation context returned no CAFC docket entries, PTAB proceedings, or district-court litigation involving patent 5,168,532. (Be aware that searches for "5168532" will also surface an unrelated Japanese patent, JP 5168532 B2 — an ion-beam uniformity system owned by Axcelis — which is not the U.S. patent at issue.)

Core bibliographic data

• Patent number: US 5,168,532 A (application serial no. 07/547,451)
• Title: Method for improving the dynamic range of an imaging system
• Inventors: Edward J. Seppi; Edward G. Shapiro; John M. Pavkovich
• Assignee (original): Varian Associates, Inc. (Palo Alto, CA) — assignment recorded Oct. 15, 1990
• Assignee (current, per Google Patents chain): Varian Medical Systems, Inc. (via 2003 assignment to Varian Medical Systems Technologies, Inc. and 2008 merger/change-of-name records)
• Filing date: July 2, 1990
• Issue (publication) date: December 1, 1992
• Legal status: Expired — Lifetime (Google Patents shows anticipated expiration July 2, 2010; the patent has not been the subject of any maintenance/renewal activity in recent years)
• Related filings: PCT/US1991/004777; EP 0489154 B1; JP 3197560 B2; AU 633756 B2 (family members). The specification also references a copending application on "Partial Fan-beam Tomographic Apparatus and Data Reconstruction Method" (Pavkovich), and an incorporated Microfiche Appendix A (six sheets, 76 frames).

Abstract (verbatim, condensed)

"Disclosed is a method for improving the dynamic resolution of an imaging system. The method employs a dual sampling or exposure technique which samples light from an image intensifier over a long and a short sample interval. When the resulting measurement for the long sample interval exceeds a threshold level, the short sample interval measurement is used, multiplied by a scaling factor. Below the threshold the long sample interval is used."

Field and technology in plain language

The invention is directed to computed-tomography (CT) imaging built around an image-intensifier-tube (IIT) detector — specifically a CT-simulator/radiotherapy-planning system (e.g., Varian's Ximatron-class simulator). A photodiode linear array reads the visible-light output of the IIT. The core problem: the IIT's output has a dynamic signal range (~100,000:1) far exceeding the dynamic range of a single photodiode readout (~35,000:1 with the manufacturer's preamp). The patent solves this by sampling each photodiode over two integration periods — a short interval and a long interval — and intelligently combining the two readings so the measurement chain achieves roughly a 400,000:1 (≈19-bit) effective range. The specification also details supporting calibration and correction: scaling-factor calibration via a normalizing photodiode, polynomial linearization of detector response, point-spread-function deconvolution, background subtraction, spatial-linearity correction, and a 32-element "imaging extension detector" that widens the scan circle.

Claim structure

Per the retrieved claim set (FreePatentsOnline and Google Patents), the patent has 10 claims, with only claim 1 independent; claims 2–10 depend from claim 1.

Independent claim 1 — plain-language overview

Claim 1 is a method of improving dynamic range using a dual-exposure technique. In substance it requires:

1. Short sample — during a first (short) time period, sampling the visible-light output of an image intensifier with a semiconductor photodetector array (a plurality of photodiodes) to get a "short interval count."
2. Long sample — during a second, longer time period, sampling the same output to get a "long interval count."
3. Selection logic based on a transition range of counts:
   • If the long interval count is above the transition range → select the short interval count, multiplied by a scaling factor, for further processing;
   • If the long interval count is below the transition range → select the long interval count;
   • If the long interval count falls within the transition range → select a weighted combination of the long and short interval counts.
4. Scaling-factor calibration — the scaling factor is derived by characterizing a normalizing photodiode: (i) applying a series of calibrating signals; (ii) measuring the normalizing photodiode's response over both a long and a short sample interval, with the maximum calibrating intensity chosen so the normalizing photodiode does not saturate during the long interval; (iii) comparing long-interval measurements to short-interval measurements (multiplied by the scaling factor) over a predetermined count range; and (iv) adjusting the scaling factor for a best least-squares fit between the scaled short-interval and long-interval measurements.

Dependent claims (brief)

• Claims 2–3: Add linearizing of the selected count, using a plurality of nth-order polynomials valid over different count ranges; coefficients depend on photodetector-array characteristics.
• Claim 4: Polynomials may be valid over overlapping ranges; interpolation between overlapping polynomial solutions when the count falls in an overlap.
• Claims 5–8: Coefficient selection by least-squares curve fitting of calibration data; the operative range spans 0 to 500,000 counts using three fourth-order polynomials, with overlapping ranges.
• Claim 9: The transition range of counts may collapse to a single selected count (a threshold rather than a band).
• Claim 10: The scaling factor is a two-part factor characterized by the equation Lᵢ(1 + αLᵢ) = κSᵢ, where Lᵢ and Sᵢ are the normalizing photodiode's long- and short-interval measurements, and α and κ are optimized by best fit over the selected range.

Caveats / uncertainty

• The official USPTO full-text database and the CAFC's 2026 docket system were not directly queryable in my environment; my "no CAFC activity" finding is based on open-web searches and should be treated as not confirmed rather than a definitive clearance. Given the patent expired in 2010, a 2026 CAFC appeal involving it would be highly unusual.
• Minor discrepancy worth noting: claim 1's preamble uses the phrase "improving the dynamic resolution," while the title, abstract, and specification consistently use "dynamic range" — I've reproduced the claim language literally above.

If you want, I can attempt further targeted searches (e.g., USPTO Patent Public Search / PatentCenter screenshots or assignment database) to verify maintenance history and confirm the absence of CAFC activity.

I'll research litigation involving US Patent 5,168,532 using multiple search approaches.

The initial search only surfaced the patent itself and an unrelated Japanese patent (JP 5168532 B2). Let me search litigation-specific databases.

Let me check dedicated patent litigation databases and search more specifically.

Based on my research across general web search, news archives, and litigation-oriented queries (including Unified Patents-style case-list searches and targeted docket/complaint searches), I found no litigation specifically involving U.S. Patent No. 5,168,532.

Findings

No known cases identified. I searched for:

• "5168532" patent litigation
• "US5168532" OR "US 5,168,532" lawsuit Varian
• "5,168,532" Varian patent infringement case
• site:unifiedpatents.com "5168532"
• "5168532" docket OR complaint OR "civil action" patent
• "patent 5168532" OR "U.S. Patent No. 5,168,532"

None of these queries returned any complaint, docket, judgment, or case that asserts U.S. Patent 5,168,532.

Important caveats on what the searches did surface (and why they are not responsive)

1. JP 5168532 B2 — This is an unrelated Japanese patent (a beam-uniformity measurement system assigned to Axcelis Technologies). It shares only the numeric string "5168532" and is not U.S. Patent 5,168,532. Per your instruction to interpret identifiers literally, this should not be confused with the U.S. patent.
2. Varian Medical Systems litigation involving other patents — News results cover the University of Pittsburgh v. Varian case (U.S. Patent No. 5,727,554; W.D. Pa.; CAFC No. 2012-1575), the Varian–Elekta global disputes (E.D. Mich.; ITC; D. Del.; N.D. Cal.), and Varian/Applied Materials disputes (referencing U.S. Patent No. 5,186,718). None of these involve patent number 5,168,532.

Assessment

U.S. Patent 5,168,532 ("Method for improving the dynamic range of an imaging system," assigned to Varian Associates, Inc., later Varian Medical Systems, Inc.) issued December 1, 1992. Per Google Patents, it shows "Expired – Lifetime" status with an anticipated expiration date of July 2, 2010 (i.e., it expired on its 20-year term, roughly 20 years after the July 2, 1990 filing date). This long-expired status is consistent with the absence of any reported modern litigation.

Conclusion: No litigation involving U.S. Patent 5,168,532 is known based on the searches performed. I cannot identify any plaintiff(s), defendant(s), jurisdiction, case number, filing date, or outcome because no case was found. If you want, I can attempt deeper PACER/CAFC/docket-database queries (e.g., DocketNavigator, RPX, LexisNexis Justia dockets), but note that free web coverage of a patent this old and long-expired is sparse, and the absence of search results should not be treated as an absolute negative without a paid docket-database confirmation.

Proceedings overview

Total AIA trial proceedings on file: 0 — no active proceedings, no claims invalidated or sustained through IPR/PGR/CBM, no settlements, and no institution denials. The USPTO Open Data Portal block in this prompt reports no AIA trial proceedings for US 5168532, and independent web searching surfaced nothing indexed under this patent number at the PTAB (hits for other "532" patents — e.g., bone-graft and medical-device litigations — are different patents and are not this one). The controlling fact for a defendant is not PTAB history but lifecycle: US 5168532 expired on 2010-07-02 (status: "Expired - Lifetime"), roughly sixteen years before today's date, so no AIA trial was ever filed and none can meaningfully revive an assertion position now.

No proceedings to report

There are no IPR / PGR / CBM dockets (no proceeding numbers exist) for US 5168532. I will not invent any. Instead, the material facts:

• Patent status: Expired - Lifetime (Google Patents / USPTO legal-status feed lists anticipated expiration 2010-07-02, i.e., 20 years from the 1990-07-02 filing date; the 17-years-from-grant term would have ended 2009-12-01, so the longer 20-year-from-filing term controls).
• Assignee history: Varian Associates, Inc. → Varian Medical Systems, Inc. — a large operating company, not an NPE; this is the classic reason no IPR was ever filed (competitors settle or cross-license rather than pay PTAB fees against a functioning medical-device business with a soon-to-expire patent).
• Why no IPR is expected even now: AIA review is a practical tool against in-force patents. An IPR petition filed today against a patent that expired in 2010 would be of no defensive value because there is no prospective relief to block and any pre-expiration damages claim is long since time-barred (see below).

Strategic summary

• Claims: CANCELED vs. SUSTAINED vs. UNTESTED. No claim of US 5168532 has been canceled or sustained in an AIA trial — every claim (the sole independent claim plus its dependents, as originally issued) is simply UNTESTED at the PTAB. Because the patent has expired, that untested status is effectively permanent: the claims are dead as a practical enforcement matter regardless of validity.

• Estoppel landscape (§ 315(e)(2)). There are no petitioners, no estoppel, and no privy bars. A defendant needs no IPR-based invalidity defense because the patent cannot be enforced prospectively. Under 35 U.S.C. § 286, damages are capped at the six years preceding the filing of a complaint; the last day any infringing act could have occurred was 2010-07-02, and any complaint filed after roughly mid-2016 captures zero recoverable damages. A suit filed today (2026) based on pre-2010 conduct would be barred on its face, independent of validity.

• Pattern signals. None from PTAB activity — no repeat petitioners, no defensive aggregator (e.g., Unified Patents) in the chain, no Patent Owner appeals. The meaningful signal is the ownership/status pattern: an expired Varian Medical Systems patent that was never litigated at the Board. If a demand letter cites US 5168532 today, the more likely scenario is a stale-assertion play by a purchaser of legacy Varian assets — the patent's expiry, not its validity, is the dispositive defense.

Recommended next steps

1. Lead with expiration, not IPR. There is no FWD to quote because no proceeding exists. The operative citations are the USPTO assignment/legal-status record (expired 2010-07-02) and 35 U.S.C. § 286 (six-year damages bar). If you receive a demand letter citing US 5168532, respond that the patent expired over a decade before the earliest date for which damages could be recovered, making any infringement theory sanction-bait under Fed. R. Civ. P. 11.

2. Verify chain of title before assuming the worst. If the asserting entity is not Varian Medical Systems, Inc., confirm the assignment chain via USPTO Assignment (records show Varian Associates, Inc. → Varian Medical Systems Technologies, Inc. (2003-09-25) → Varian Medical Systems, Inc. (2008-10-13 merger)). An entity without a recorded assignment may lack standing, a second independent knockout.

3. No PTAB milestones to calendar. Because there are no pending proceedings, there are no institution-decision deadlines, oral-hearing dates, or FWD due dates. The absence of PTAB activity on a patent this old and this long-expired is unsurprising and carries no adverse inference — well-asserted in-force patents attract IPRs; expired ones attract motions to dismiss.

Sources: USPTO Open Data Portal PTAB ingest (no proceedings, per the structured block in this prompt); Google Patents legal-status record for US5168532A (status "Expired - Lifetime," anticipated expiration 2010-07-02, assignment history to Varian Medical Systems, Inc.); web search of PTAB/CAFC databases returned no proceeding or appeal citing US 5168532 (all "532"-numbered hits referenced unrelated patents, e.g., Surgalign v. LifeNet IPR2019-00570 and Nox Medical v. Natus, neither involving this imaging patent).

Based on the available record (the Google Patents legal-event feed embedded in the authoritative patent text, plus corporate-history research), I can reconstruct the ownership chain. Important limitation up front: the USPTO Assignment Center pages (reel/frame numbers, correspondent-of-record names) were not retrievable with the search tool available in this session. I have not fabricated reel/frame or correspondent entries; where those fields would go, I say so explicitly. The chain-of-title parties and dates below come from the Google Patents legal-events feed, which mirrors USPTO assignment recordation data.

Inventors

Pattern check: All three assigned to their employer — the standard operating-company pattern. No evidence any inventor left Varian within 12 months of filing; Pavkovich's contemporaneous even-filed application assigned to Varian affirmatively indicates he remained with the company. No fire-sale precursor pattern.

Original assignee

• Varian Associates, Inc., Palo Alto, CA, a Delaware corporation — the entity named on the issued patent (also the named "applicant" on the front page).
• Line of business: Defense/microwave electronics and, relevantly, medical radiation equipment. This patent's specification describes the Ximatron CR Radiotherapy Simulator System (manufactured by Varian, the assignee) and the CT-simulator detector chain (IIT + photodiode linear array + dual-interval sampling) that embodies the claims. Product embodiment: yes — the CT option on the Ximatron CR simulator used the image-intensifier/photodiode-array detector with long/short sampling described and claimed here.
• Current status: Varian Associates, Inc. formally changed its name to Varian Medical Systems, Inc. (USPTO name change recorded 2003-09-26). Varian Medical Systems, Inc. was acquired by Siemens Healthineers (announced Aug. 2020; completed Apr. 15, 2021) and continues to operate as a Siemens Healthineers company. Never in bankruptcy.

Assignment timeline

The patent does have recorded post-issuance assignments (five recorded events appear in the Google Patents legal-event feed). Reel/frame numbers and correspondent-of-record names could not be independently pulled from the USPTO Assignment Center within the available tooling; verify at https://assignmentcenter.uspto.gov (patent no. 5168532). No assignments were found that leave the Varian corporate family.

• 1990-07-02 (execution/recordation date as shown on Google feed) — Conveyance: Assignment of Assignor's Interest

  • Assignor: John Pavkovich
  • Assignee: Varian Associates, Inc. (Palo Alto, CA, a DE corp.)
  • Correspondent: not retrievable — reel/frame not retrievable in this session
  • Context: Original inventors' assignment executed on the filing date (record shows only Pavkovich on this first entry; the 10-15 entry below covers all three inventors).

• 1990-10-15 — Conveyance: Assignment of Assignor's Interest

  • Assignor: John M. Pavkovich, Edward J. Seppi, Edward G. Shapiro
  • Assignee: Varian Associates, Inc.
  • Correspondent: not retrievable — reel/frame not retrievable in this session
  • Context: Second inventors' assignment completing the chain of title to the employer for all three named inventors.

• 2003-09-25 — Conveyance: Assignment of Assignor's Interest (see document for details)

  • Assignor: Varian Medical Systems, Inc.
  • Assignee: Varian Medical Systems Technologies, Inc.
  • Correspondent: not retrievable — reel/frame not retrievable in this session
  • Context: Internal Varian-family transfer into an affiliated subsidiary/technology-holding entity — a routine corporate reorganization, not an arm's-length sale.

• 2003-09-26 — Conveyance: Change of Name (see document for details)

  • Assignor: Varian Associates, Inc.
  • Assignee: Varian Medical Systems, Inc. (Google's feed spells it "VARIAN MEDICAL SYTEMS, INC." — likely a data-entry typo for SYSTEMS)
  • Correspondent: not retrievable — reel/frame not retrievable in this session
  • Context: Corporate name change of the original assignee; documents the formal renaming of Varian Associates into Varian Medical Systems.

• 2008-10-13 — Conveyance: Merger (see document for details)

  • Assignor: Varian Medical Systems Technologies, Inc.
  • Assignee: Varian Medical Systems, Inc.
  • Correspondent: not retrievable — reel/frame not retrievable in this session
  • Context: Merger of the affiliate back into Varian Medical Systems, Inc., consolidating title in the parent operating company.

Net effect: every recorded event is either an inventors→employer assignment or an intra-Varian name change / affiliate transfer / merger. The patent never left the operating-company family, and Google Patents' current-assignee field reads Varian Medical Systems Inc. The patent expired on its 20-year term (anticipated expiration 2010-07-02; status "Expired – Lifetime").

Timeline diagram

timeline
    title Ownership of US 5168532
    1990 : Filed by Varian Associates
         : Inventors assign rights to Varian
    1992 : Patent issued
    2003 : Name change to Varian Medical Systems
         : Transfer to Varian Medical Systems Technologies
    2008 : Merger back to Varian Medical Systems
    2010 : Patent expired on 20 year term
    2021 : Varian acquired by Siemens Healthineers

NPE / troll-pattern signals

1. Shell-entity transfer — Not present. All assignees are the operating company or its named corporate affiliates (Varian Associates, Inc.; Varian Medical Systems, Inc.; Varian Medical Systems Technologies, Inc.). No "IP / Licensing / Holdings / Ventures" LLC suffix, no registered-agent address, no single-member Delaware/Texas LLC anywhere in the chain. Caveat: reel/frame numbers were not retrievable to run a registered-agent check, but the recorded assignee names themselves are all operating-company names.

2. Known asserter in the chain — Not present. No Acacia, Marathon, Intellectual Ventures, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Round Rock, or any Unified Patents/RPX-listed high-frequency plaintiff appears as assignor or assignee. The only assignees are the Varian entities above.

3. Repeat correspondent across the chain — Unclear / unverifiable in this session. Correspondent-of-record names could not be retrieved from the USPTO Assignment Center with the tools available, so no recurrence analysis was possible. This is a data-availability gap, not a finding.

4. Cascading transfers — Not present. Only two family-internal transfers occurred (2003 affiliate transfer; 2008 merger-back), separated by five years. No chained-LLC pattern, no shared-registered-agent cascade.

5. Pre-litigation transfer — Not present. The prior litigation summary for this patent found no infringement suits whatsoever, so no assignment can be tied to suit preparation. The 2003 and 2008 transfers coincide with Varian's corporate restructuring (the Varian Associates→Varian Medical Systems rename era), not litigation.

6. Bankruptcy fire-sale — Not present. Varian Associates / Varian Medical Systems never filed Chapter 7 or 11. The 2021 Siemens Healthineers acquisition was a consensual all-stock purchase of a going concern ($16.4B), not a distressed patent sale.

7. Privateering — Not present. No transfer of this patent to any third-party NPE for assertion against competitors; no SEC 8-K/10-K or Patent Progress/EFF coverage suggests Varian used this patent offensively through a proxy.

8. Defensive aggregator (anti-NPE) — Not present. The chain does not end at RPX, AST, LOT, Unified Patents, or OIN. It ends at the practicing operating company (Varian Medical Systems, now a Siemens Healthineers company). Inverse signal: the patent was never weaponized by an NPE and expired in 2010.

Verdict

Operating-company assertion — with the explicit caveat that no assertion ever occurred: this patent remained with its practicing operating-company owner for its entire life and expired unlitigated in 2010.

Justification: Every recorded conveyance (1990-07-02 and 1990-10-15 inventor assignments to Varian Associates, Inc.; 2003-09-25 transfer to Varian Medical Systems Technologies, Inc.; 2003-09-26 name change to Varian Medical Systems, Inc.; 2008-10-13 merger back to Varian Medical Systems, Inc.) stayed strictly inside the Varian operating-company family, and the assignee of record shipped the Ximatron CR CT-simulator products embodying the dual-interval sampling claims. There is no shell LLC, no known asserter, no privateering, and no litigation — the absence of any NPE or bankruptcy signal, combined with an expired 2010 term and a Siemens Healthineers operating-company owner, makes this the cleanest available classification. Reel/frame and correspondent fields remain unverified pending a direct USPTO Assignment Center query at https://assignmentcenter.uspto.gov (search patent no. 5168532).

Prior-Art Analysis — U.S. Patent 5,168,532 (US5168532A)

Database-search caveat (read first)

I was asked to search the USPTO database directly. I could not run live queries against USPTO Patent Public Search / PatentCenter from this environment; I used Google Patents, FreePatentsOnline, Justia Patents, and uspto.report as proxies for the USPTO bibliographic and citation data. I also could not retrieve the complete front-page ("References Cited") section of the issued patent — the Google Patents text dump you supplied cuts off mid-specification and does not include the face-of-the-patent citation list, and the FreePatentsOnline mirror returned only a partial list. Accordingly, the citation list below is verified but incomplete, and I flag every item where the exact date or face-page status is uncertain. I have not invented any reference.

A separate search-hygiene note: queries for the numeric string "5168532" repeatedly surface JP 5168532 B2 (Axcelis Technologies, an ion-beam uniformity measurement patent). That is an unrelated Japanese document and is excluded from this analysis per your literal-ID rule.

Verified references and their § 102 analysis

1. U.S. Patent No. 4,149,248 — Pavkovich, "Apparatus and Method for Reconstructing Data"

• Full citation: U.S. Patent No. 4,149,248 to John M. Pavkovich, "Apparatus and Method for Reconstructing Data," issued April 10, 1979, assigned to Varian Associates, Inc.
• Source/status: Referenced affirmatively in the Background of the Invention of US5168532 ("In U.S. Pat. No. 4,149,248, to John M. Pavkovich, entitled Apparatus and Method for Reconstructing Data, and assigned to the same assignee as is the present patent…"). It is also the only U.S. patent expressly identified anywhere in the specification text supplied to me.
• Description: Discloses fan-beam computed-tomography apparatus and a convolution-method data-reduction technique that operates on fan-beam projection data without reordering fan rays into parallel-ray geometry, eliminating reordering-induced errors and computation delay. This is the foundational Varian fan-beam reconstruction patent and is the technological ancestor of the CT-simulator system in which the '532 invention operates.
• Priority status under § 102: Issued April 10, 1979 — more than one year before the '532 filing date of July 2, 1990 — so it is prior art at least under pre-AIA 35 U.S.C. § 102(b) (and as a same-assignee reference it was available to the examiner under § 102(a)/(e) as well). Because the '532 application was filed July 2, 1990, it was examined under pre-AIA § 102; that is the framework used below.
• Claims potentially anticipated: None, on the evidence available. Claim 1 (the sole independent claim) is a dual-exposure dynamic-range method requiring (a) short-interval and (b) long-interval sampling of an image intensifier's visible-light output with a photodiode array; (c)–(e) count-dependent selection among the short count × scaling factor, the long count, and a weighted combination over a transition range; and (i)–(iv) calibration of the scaling factor by least-squares fitting of a normalizing photodiode's long- and short-interval measurements. The '248 patent is directed to reconstruction mathematics, not to detector readout dynamic range; nothing in it discloses dual integration intervals, threshold/transition selection, a scaling factor, or normalizing-photodiode least-squares calibration. It therefore cannot anticipate claim 1, and since claims 2–10 are all dependent from claim 1, it anticipates none of them. Its relevance is contextual (CT fan-beam background), not anticipatory.

2. Herman et al., "Reconstruction Using Divergent-Ray Shadowgraphs" (1977)

• Full citation: G. T. Herman et al., "Reconstruction Using Divergent-Ray Shadowgraphs," in Reconstruction Tomography from Incomplete Projections (University Park Press, 1977), pp. 105–117.
• Status: Listed in the "References Cited — Other References" portion of the patent as mirrored by FreePatentsOnline. I could not verify from my searches whether it appears on the USPTO face page as an examiner-cited "Other Reference," though the FPO mirror formatting suggests it does.
• Description: Treats tomographic reconstruction from divergent-ray (fan/cone-beam) projection data — the mathematical basis underlying the fan-beam CT reconstruction used in the '532 system.
• § 102 status: Dated 1977 → printed publication more than one year before July 2, 1990 → pre-AIA § 102(b) prior art.
• Claims anticipated: None. Divergent-ray reconstruction mathematics does not disclose the long/short dual-sampling, selection, or normalizing-photodiode calibration steps of claim 1, nor anything in dependent claims 2–10.

3. B. E. Oppenheim, "Reconstruction Tomography from Incomplete Projections" (1977)

• Full citation: B. E. Oppenheim, "Reconstruction Tomography from Incomplete Projects," in Reconstruction Tomography from Incomplete Projections (University Park Press, 1977), pp. 155–183. (Title as printed on the patent mirror reads "…from Incomplete Projects"; I reproduce it literally.)
• Description: Addresses reconstruction algorithms when projection data is incomplete or truncated — directly relevant to the '532 system's partial-fan-beam ("asymmetric fan") body-scan mode, in which the image intensifier is offset and a full 360° scan is used so the entire object is not viewed in every projection.
• § 102 status: 1977 printed publication → pre-AIA § 102(b) prior art.
• Claims anticipated: None. It concerns image reconstruction from incomplete data, not detector dynamic-range improvement; no claim of '532 is directed to reconstruction mathematics, and the reference discloses none of claim 1's sampling/selection/calibration elements.

4. Huang et al., "Effect of Out-of-field Objects in Transaxial Reconstruction Tomography" (1977)

• Full citation: Huang et al., "Effect of Out-of-field Objects in Transaxial Reconstruction Tomography," in Reconstruction Tomography from Incomplete Projections (University Park Press, 1977), pp. 185–198.
• Description: Analyzes artifacts/errors introduced in transaxial reconstruction by objects lying outside the scanned field — again an incomplete-projection/partial-fan topic bearing on image quality in the '532 CT-simulator geometry, not on detector readout.
• § 102 status: 1977 printed publication → pre-AIA § 102(b) prior art.
• Claims anticipated: None, for the same reason as items 2 and 3.

5. Hamamatsu, "Characteristics and Use of PCD Linear Image Sensors" (Technical Information SD-03)

• Full citation: "Characteristics and Use of PCD Linear Image Sensors," Hamamatsu Technical Information SD-03 (cover page and p. 9), as listed in the "Other References" on the FreePatentsOnline mirror.
• Description: Manufacturer's technical literature describing Hamamatsu photodiode-array linear image sensors of the PCD (photodiode array) family. This is the closest reference to the hardware of claim 1, because the '532 preferred embodiment uses the Hamamatsu S2301 512-channel linear silicon diode array, whose charge-accumulation, saturation (22 pC), noise, and dynamic-range (≈35,000:1 with the manufacturer's preamp) characteristics the specification discusses at length and which motivated the dual-exposure invention. The SD-03 literature presumably describes the array's charge-integration readout and saturating behavior.
• § 102 status: Undated in my retrieval — flag. Hamamatsu literature of this series circulated in the 1980s, and if publicly distributed before July 2, 1989 it would be § 102(b) art; if only available after the invention date it would not qualify. I cannot confirm the publication date from the search results, and I will not guess. Treat its § 102(b) qualification as unverified.
• Claims anticipated: None, on the available evidence. Even taken at its strongest, a data sheet disclosing a saturating photodiode linear array does not disclose: sampling the image intensifier's output over two different intervals, selecting short-vs-long counts by a transition range, weighted blending, or normalizing-photodiode least-squares scaling-factor calibration (claim 1 elements (a)–(e) and (i)–(iv)). It is best characterized as teaching the problem (limited single-channel dynamic range) that claim 1 solves.

Reference list the specification cites but that are not anticipatory prior art

For completeness, the specification text you supplied also mentions:

• N. Diffrient et al., Humanscale 1/2/3 Manual (MIT Press, 1979) — cited only for anthropometric population-coverage statistics (95% of U.S. males within a 21-cm head scan circle). No technical disclosure; anticipates nothing.
• Copending application "Partial Fan-beam Tomographic Apparatus and Data Reconstruction Method" (Pavkovich), filed even date (July 2, 1990) — the '532 specification cross-references it, but as a copending, same-day, same-assignee application it is not prior art under any § 102 subsection.
• Microfiche Appendix A (6 sheets, 76 frames) — incorporated by reference; part of the '532 disclosure, not prior art.

Bottom-line § 102 assessment

No verified reference anticipates claim 1 — and because claims 2–10 are all directly or indirectly dependent on claim 1, no verified reference anticipates any claim of US 5,168,532. The closest art in substantive terms is:

• Pavkovich '248 for the CT/fan-beam context (background only);
• Hamamatsu SD-03 for the photodiode-array dynamic-range limitation (problem recognition only).

Claim 1's anticipatory burden requires a single reference disclosing every element of the dual-exposure method, including the combination of a weighted-combination transition band and a normalizing-photodiode least-squares scaling-factor calibration. None of the verified references comes close to that combination.

Honest limitation: This conclusion rests on the partial citation list I could verify. The issued patent's full front page almost certainly carries additional U.S. and foreign patent citations and possibly more "Other References" that the truncated Google Patents text and my searches did not surface. Before relying on this as a complete § 102 map, the front-page "References Cited" should be pulled from USPTO Patent Public Search (PN 5168532) or the USPTO PatentCenter image of the printed patent — I was unable to do so from this environment.

Obviousness Analysis — U.S. Patent No. 5,168,532 ("Method for improving the dynamic range of an imaging system")

Preliminary caveat on the "Prior Art" section

The full patent text supplied to me (the Google Patents fetch) does not include a "Prior Art" or "References Cited" section — the fetched text omits the front-page citation list, and the "Prior Art" keyword listing embedded in the Google page is merely a text-mining artifact of the specification's use of phrases like "prior art" (e.g., "apparatus of the first generation suffered from many shortcomings"). My web searches for the examiner's citation list (e.g., "US5168532" "References Cited", uspto.report patent grant 5168532 references cited) returned no usable front-page reference list before hitting the search-step limit. I therefore cannot identify the exact references the examiner applied during prosecution, and the element-by-element mapping below should be verified against the USPTO file wrapper before it is used in an adversarial setting.

What I can rely on with confidence:

• US 4,149,248 (Pavkovich, "Apparatus and Method for Reconstructing Data," Varian Associates) — expressly discussed in the Background section of the '532 specification (confirmed in the supplied text), issued 1979, well before the July 2, 1990 filing date.
• The specification's own admissions of the state of the art: image-intensifier-tube (IIT) fluoroscopy/radiographic simulators, IIT-to-TV imaging, and self-scanning silicon photodiode linear arrays (the Hamamatsu S2301 device is named in the spec) all existed before the filing date.
• Published family-member/related filings (EP 0489904 A1 "Radiation therapy x-ray simulator"; the copending Pavkovich/Seppi partial-fan-beam application, WO 1992/000566) share specification text with the '532 patent but were filed the same day or later — not § 102 prior art, though useful for claim-scope context.

Where I refer below to art by teaching rather than by patent number, that reflects the fact that the specific citation list was unavailable, not that the teachings were unavailable in 1990.

1. Legal framework and the PHOSITA

§ 103 framework (Graham factors): scope and content of the prior art; differences between the prior art and the claimed invention; the level of ordinary skill; and secondary considerations. Under KSR Int'l Co. v. Teleflex, obviousness may be shown by a "predictable variation," by a combination of known elements "each performing the same function it had been known to perform," and by a showing that a PHOSITA had a "reasonable expectation of success" and a reason to combine — a rigid teaching-suggestion-motivation test is not required.

Person of ordinary skill in the art (PHOSITA) circa July 1990: an engineer or physicist with (i) 3–7 years' experience in x-ray/CT imaging-systems design, including image-intensifier chains and solid-state photodiode-array readout electronics; (ii) working knowledge of charge-integrating preamplifiers, A/D conversion, and detector calibration; and (iii) familiarity with CT data-reduction/reconstruction methods such as those of Pavkovich. This person would routinely read IEEE TNS / Medical Physics, Hamamatsu/Reticon device literature, and the patent literature of Varian, GE, Siemens, Toshiba, and Hitachi.

2. Claim scope, in one paragraph

Independent claim 1 is a method of dual-exposure dynamic-range extension: sample the IIT's visible output on a photodiode array over a short interval (short count) and a long interval (long count); use the scaled short count when the long count is above a transition band, the raw long count when below the band, and a weighted blend inside the band; and calibrate the scaling factor by characterizing a normalizing photodiode — apply a calibration-signal series whose maximum intensity does not saturate the normalizing photodiode during the long sample, compare long-interval vs. scaled short-interval measurements over a predetermined count range, and tune the scaling factor by least-squares curve fit. Dependent claims 2–10 add: linearizing the selected count with multiple nth-order polynomials (claims 2–3), overlapping polynomial ranges with interpolation (claim 4), least-squares coefficient selection (claim 5), three fourth-order polynomials over 0–500,000 counts (claims 6–8), a single-count transition threshold (claim 9), and a two-part scaling factor per the equation Li(1+αLi) = κSi (claim 10).

3. Prior-art building blocks (representative, pre-July 1990)

4. The obviousness theory — primary combination

Combination 1 (core, against claim 1): A + B + C + D + E

Element-by-element mapping:

Why a PHOSITA would combine (motivation):

1. Same field and same players. A, B, C are not merely analogous art — they are the same imaging chain the '532 inventors were extending. Pavkovich (A) is same-assignee background art the spec itself relies on. The problem the '532 patent solves is stated in its own text: the IIT's point-spread response implies "a dynamic signal range of at least 100,000:1," while "the single channel dynamic range of the photodiode linear array 44 has been measured to be only 35,000:1." Extending a sensor's range by using two exposure times was the textbook fix for exactly that saturation/dynamic-range mismatch (D), so the combination is "the combination of familiar elements according to known methods" that KSR treats as obvious.
2. Predictable result. The spec itself quantifies the outcome as arithmetic: a 9:1 interval ratio extends the range by ~3 bits ("a factor of about 9"). That is the predictable product of the chosen duty cycle, not an unexpected synergy.
3. The normalizing-photodiode calibration is conventional. CT systems already needed per-channel normalization (the spec's own x-ray normalization detector 66 does exactly this job for source flux). Using a reference photodiode to derive the short/long scale factor, keeping the reference unsaturated, and least-squares fitting are steps E any calibration engineer would take; claim 1 adds no structural limitation beyond "characterizing a normalizing photodiode," and the spec's Figure-29 procedure describes it as routine calibration data processing.
4. Design necessity of the transition band. Once a person uses two samples, a discontinuity at the switching point is an artifact risk; weighted blending is the obvious smoothing, and claim 9 confirms the "band" may collapse to "a single selected count" — i.e., a threshold, the simplest possible implementation.

Weakest point of the combination for a challenger: the specific combination of (i) choosing the calibration maximum intensity so the normalizing photodiode does not saturate during the long sample interval and (ii) fitting the scale factor only over a predetermined count range, is somewhat fact-specific. A challenger would need prior art showing a normalization detector deliberately kept unsaturated over the long interval while the imaging array is allowed to saturate — or would argue that keeping a reference detector in its linear region is so elementary that no express teaching is needed. This is a real litigation risk area for the obviousness case, and the examiner's file-wrapper citations (unavailable here) should be checked for whether this was the point of allowance.

5. Dependent claims

• Claims 2–3 (polynomial linearization, one polynomial per count range): F teaches fitting separate polynomial segments to different response ranges of a non-linear detector. Adding this to the selected count of claim 1 is a routine correction layer; a PHOSITA linearizing a 100,000:1 chain would naturally need a range-split polynomial because a single low-order polynomial cannot fit that span. Obvious if the base claim is.
• Claim 4 (overlapping ranges + interpolation): overlap-and-interpolate between adjacent fitted curves is the standard way to avoid seams at polynomial boundaries — the same rationale as the claim-1 transition-band blend. Obvious design choice.
• Claims 5–8 (least-squares coefficient selection; three 4th-order polynomials over 0–500,000 counts): least-squares polynomial fitting to calibration data is E/F black-letter practice; the specific ranges (0–4,000 / 2,000–62,000 / 44,000–500,000) track the device's 16-bit long-count range (~0–62,000) and the ~×9-extended short-count range (to ~500,000) — i.e., the ranges are dictated by the hardware the PHOSITA already chose, not by an inventive insight.
• Claim 9 (single-count transition): collapses the band to a threshold — this narrows nothing conceptually; a threshold trigger is the most obvious form of "above X use short, below X use long."
• Claim 10 (two-part factor Li(1+αLi) = κSi): this is the claim most likely to survive an obviousness attack as drafted, because it encodes a nonlinear (α) plus linear (κ) correction — a second-order linearity term on the long-interval reading. However, a challenger would argue that a two-parameter (slope + curvature) least-squares fit is the ordinary way to model a slightly non-linear photodiode (the spec concedes "the response of the photodetectors in photodiode linear array 44 are slightly non-linear"), and the α term's magnitude ("on the order of 10⁻⁷") is a fitted constant, not a structural feature. This claim's validity likely turns on whether the α-nonlinearity model was specifically taught in the art.

6. Secondary considerations (the other side)

The record shows no litigation and no PTAB challenge to this patent (it expired 2010-07-02, and searches found no district-court or Board activity). Secondary considerations a patent owner would raise if validity were ever tested:

• Long-felt need: diagnostic-class CT simulators with ~100,000:1 detector range were commercially desired (the spec frames the IIT-vs-array mismatch as a problem).
• Unexpected results: achieving an effective 400,000:1 (~19-bit) per-channel range, and the "photon-statistics preserving" property (the spec stresses that using the short count "discards ninety percent" of photons only when counts are already high enough to tolerate it).

A § 103 challenger's rejoinder: the "need" was for a CT simulator, and the solution was assembled from known parts already in the same Varian simulator product line; the 19-bit figure is the arithmetic product of a 16-bit ADC-scale reading plus a 9:1 (≈3-bit) interval ratio — precisely the predictable result of the disclosed 9T/1T timing; and the "photon-statistics" argument is the textbook reason long exposures exist in the first art (D), i.e., it is the known purpose of the known element, not a new effect.

7. Bottom line

• Most vulnerable: claim 1 and its calibration-agnostic dependents (claims 2–9), on a combination of the IIT-CT simulator chain (B), charge-integrating photodiode-array readout (C), known dual-exposure dynamic-range extension (D), and conventional reference-detector least-squares calibration (E), with US 4,149,248 (A) supplying the CT context and motivation. The primary rationale is KSR-style: a finite set of identified, predictable solutions to a known dynamic-range-mismatch problem, all drawn from the same field.
• Most defensible: claim 10's two-part nonlinear+linear scaling model, and claim 1's specific non-saturating-normalizing-photodiode calibration protocol — unless file-wrapper citations surface art squarely teaching those details.
• Key open item: because the examiner's "References Cited" list was not in the supplied text and could not be retrieved in my searches, the concrete § 102/§ 103 references of record remain unidentified. Before relying on this analysis, pull the USPTO file wrapper / front-page citation list for US 5,168,532 and map the claims against the actual cited art (likely candidates in this technology area would include additional Varian/Pavkovich CT patents, image-intensifier CT patents, and CCD/photodiode dual-exposure-range patents of the 1980s).