US 10368795

I'll search for the specific patent number and any related litigation records.

Let me run additional targeted searches for USPTO records and any litigation.

Let me do a final check for any PTAB or appellate proceedings specifically involving this patent.

Search scope and caveats

I searched for the literal identifiers 10368795, US10368795B2, 10,368,795, 14/686,824, and combinations with "Canfield," "acne imaging," "IPR," "PTAB," and "Federal Circuit / CAFC 2026."

Important uncertainty note: I found no CAFC docket and no PTAB (IPR/PGR) proceeding naming U.S. Patent No. 10,368,795. I cannot access PACER/CM-ECF dockets directly, so I cannot rule out a pending 2026 appeal or district-court case that is not indexed in the search results I retrieved. What I did find in 2026 Federal Circuit activity is unrelated to this patent (e.g., AudienceScience v. Google, Katana v. Micron, Enviro Tech v. Safe Foods, Moskowitz v. Globus, Smith Interface v. Apple). The Canfield-related litigation that does exist concerns different patents — see below.

---

US 10,368,795 B2 — Bibliographic summary

Abstract (verbatim from the patent)

"In an arrangement for detecting acne, particularly in its earliest stages before any humanly visible or palpable manifestations thereof have emerged, a green fluorescence image of skin illuminated with blue light is captured and processed to detect the presence of microcomedones. After correcting the green fluorescence image for the non-uniform distribution of excitation light on the skin, fluorescent spots are detected therein. The subset of fluorescent spots having an intensity lower than an intensity threshold and an area smaller than an area threshold are selected as representing candidate microcomedones. The subset can be further refined by eliminating fluorescent spots having irregular shapes, are associated with inflammation, or exhibit comedone features. Additional filtering can be applied by selecting those spots whose locations correspond to pore locations. A dermatoscope with a blue illumination source and a green-filtered viewer allows live imaging of skin spots indicative of acne, including microcomedones not otherwise visible."

---

Plain-language overview of the independent claims

Claim 1 — Method of detecting microcomedones

A three-part method:

1. Illuminate the skin with blue light.
2. Capture an image of that illuminated skin through a filter that passes a green-yellow band of wavelengths. (Note the slight mismatch with the specification, which describes a green filter centered at 540–560 nm.)
3. Process the captured image to find skin areas containing a porphyrin whose fluorescence emission is primarily in an orange-red band (the specification identifies the target as coproporphyrin III, associated with P. acnes). The processing specifically requires:
   • detecting fluorescent spots in the image; and
   • selecting a subset of those spots meeting three quantitative criteria simultaneously: (a) intensity below an intensity threshold, (b) regular shape, and (c) area smaller than a first area threshold but larger than a second (lower) area threshold, where the lower bound is at least 150 microns in diameter or 12 pixels in area. The selected subset is stated to include microcomedones.
4. Generate a further image of the skin indicating the selected spots as possible microcomedones.

A notable feature is the bounded area range (both a floor and a ceiling), which distinguishes the claim from a simple "smaller than X" filter.

Claim 14 — Apparatus for detecting microcomedones

An apparatus with four components:

• an image capture device with a viewing opening;
• a light source arranged about the viewing opening, emitting blue light when activated;
• a filter in the viewing path that passes a green-yellow band; and
• processing circuitry configured to perform the same core detection logic as claim 1 — detecting fluorescent spots representing areas containing the orange-red-emitting porphyrin, selecting the subset meeting the intensity / regular-shape / bounded-area criteria (lower bound ≥150 µm diameter or 12 px), and generating a further image indicating the spots as possible microcomedones.

Dependent claims (summary)

• Orthogonal corrections: correcting the captured image before spot detection (claim 2), including normalizing against a chromophore distribution image (hemoglobin and/or melanin) or an excitation image (claims 9–11).
• Additional filtering: excluding spots associated with inflammation or having comedone features (claim 3, and 22 for the apparatus); retaining only spots whose locations correspond to detected pores (claim 4), with an alternative area threshold based on mean/median pore size (claim 20).
• Numeric ranges: blue light centered at 400–410 nm (claim 5); filter centered at 525–575 nm (claim 6); intensity threshold of 60–80% of full scale based on mean/median pixel intensity (claim 7); first area threshold of 70–90% of the median or 120–140% of the mean spot size (claim 8).
• Comparison/monitoring: comparing the captured image and/or the further image with a stored image (claims 12–13).
• Porphyrin identity: the porphyrin includes coproporphyrin III (claim 18).
• Light source implementation: a blue LED or a blue filter (claim 15); and blue center wavelength 400–410 nm / filter 525–575 nm for the apparatus (claims 16–17).

One rendering note: In the claim text as retrieved, claim 18 reads "The method of claim 1" and claim 20 reads "The method of claim 1," immediately after apparatus claims 14–17 and 19, and claim 18 sits between apparatus claims 17 and 19. This appears to be a formatting/dependency artifact in the source markup rather than a substantive point; I flag it because I cannot verify the printed claim dependencies from the retrieved text alone.

---

Related litigation — careful distinction

The litigation activity around Canfield that appears in search results involves other patent numbers, not the '795 patent:

1. Canfield Scientific, Inc. v. Melanoscan LLC / Dr. Rhett Drugge, D.N.J. No. 2:16-cv-04636 — concerns U.S. 7,359,748, not the '795 patent. Canfield filed IPR2017-02125 against the '748 patent; the PTAB found no challenged claim unpatentable. (See the D.N.J. opinions and the PTAB Final Written Decision available via CourtListener/Justia.)
2. QuantifiCare S.A. v. Canfield Scientific, Inc. — the IPRs (IPR2021-01511, -01518, -01519), the CAFC appeal No. 23-1917 (decided March 25, 2025), and the parallel UPC/Düsseldorf dispute concern U.S. 10,070,119; 10,165,253; 10,681,334 and EP 3 156 843 — again, not the '795 patent.

I found no reference to 10,368,795 in any IPR, district-court, or Federal Circuit record. Given my inability to query PACER directly, treat the "no CAFC 2026 docket" conclusion as a search-result-based finding, not an authoritative docket check.

---

Sources

• Full patent text (authoritative): https://patents.google.com/patent/US10368795/en
• Pre-grant publication: https://patents.justia.com/patent/20150374277
• Canfield virtual patent marking (VISIA-CR ↔ U.S. 10,368,795): https://canfieldresearch.com/patents/
• QuantifiCare v. Canfield CAFC No. 23-1917 (unrelated patents): https://www.courtlistener.com/opinion/[10363459](/patent/10363459)/quantificare-sa-v-canfield-scientific-inc/
• Canfield v. Melanoscan / Drugge, D.N.J. 2:16-cv-04636 (unrelated patent): https://cases.justia.com/federal/district-courts/new-jersey/njdce/2:2016cv04636/[335928/139](https://assignmentcenter.uspto.gov/search/patent/reelFrameDetail?reelFrame=335928-0139)/0.pdf

I'll search for litigation involving US patent 10368795 across patent litigation databases and court sources.

Let me search more targeted litigation databases and the Unified Patents portal.

Litigation Search Results for U.S. Patent No. 10,368,795

Bottom line: No litigation (assertion) specifically involving U.S. Patent No. 10,368,795 was found.

I searched patent-litigation and court sources — the Unified Patents litigation portal (portal.unifiedpatents.com/litigation/caselist), Justia Dockets, CourtListener/RECAP, Docket Alarm, RPX Insight, Plainsite (Canfield's company litigation profile), JUVE Patent, EPLAW, and general web sources targeting the exact string "10368795" and "10,368,795" in connection with litigation. I did not auto-correct the number and specifically discarded the many false hits for unrelated identifiers that share the same digit string (e.g., a Czech insolvency debtor "IČO 10368795," a Brazilian corporate registration "10368795 HB TRANSPORTES," and a UK company "DMH PLASTERING LTD" company no. 10368795). None of those are the patent.

What I can confirm about the patent itself

• Patent: U.S. 10,368,795 B2, "Acne imaging methods and apparatus"
• Inventor: Sachin V. Patwardhan
• Assignee: Canfield Scientific, Inc. (Parsippany, NJ)
• Priority date: 2014-06-30; Filing date: 2015-04-15; Granted: 2019-08-06
• Family: US20150374277A1; PCT WO2016003561A1; EP3160327B1 (titled "Acne imaging apparatus")
• Maintenance fee ("4th yr, small entity") paid 2023-04-03; legal status listed as Active. (Per Google Patents: https://patents.google.com/patent/[US10368795](/patent/US10368795)/en)

No district-court, Federal Circuit, ITC, PTAB, or foreign case record naming this patent as an asserted or challenged patent appeared in any of the sources searched.

Related Canfield Scientific litigation (found, but on different patents — not the '795 patent)

For completeness, these are adjacent cases involving the same patent owner; none asserts or challenges U.S. 10,368,795:

1. The Procter & Gamble Co. et al. v. Enhanced Image Technologies, LLC et al.

   • Court: U.S. District Court, W.D.N.C. (Charlotte)
   • Case No.: 3:2012cv00796
   • Filed: Nov. 30, 2012
   • Plaintiff included Canfield Scientific, Inc. (with P&G); defendants Enhanced Image Technologies, LLC and PSI Co., Ltd.
   • Outcome: Consent judgment entered July 30, 2013. Patents asserted were skin-imaging patents (the '003 patent family), not the '795 patent.
   • Source: https://dockets.justia.com/docket/north-carolina/ncwdce/3:2012cv00796/69597

2. Canfield Scientific, Inc. v. Dr. Rhett Drugge; Melanoscan LLC (and related IPR)

   • Court: U.S. District Court, D.N.J.
   • Case No.: 2:16-cv-04636 (JMV-JBC)
   • Filed: Aug. 1, 2016
   • Subject: Declaratory judgment of non-infringement of U.S. Pat. No. 7,359,748 ('748, Melanoscan's skin-cancer imaging patent); related PTAB proceeding Canfield Scientific, Inc. v. Melanoscan LLC, IPR2017-02125 (instituted; Board found '748 claims not obvious, later addressed on appeal in Canfield Sci., Inc. v. Melanoscan). Again, the patent at issue is the '748 patent, not the '795 patent.
   • Sources: https://cases.justia.com/federal/district-courts/new-jersey/njdce/2:2016cv04636/[335928/139](https://assignmentcenter.uspto.gov/search/patent/reelFrameDetail?reelFrame=335928-0139)/0.pdf; https://www.plainsite.org/courts/uspto-patent-trial-and-appeal-board/canfield-scientific-inc-v-melanoscan-llc/

3. QuantifiCare S.A. v. Canfield Scientific, Inc. (and affiliates) — multinational, still active as of 2026

   • U.S. appeal: Fed. Cir. Case No. 23-1917 and 23-1918 (from PTAB IPRs IPR2021-01511, -01518, etc. on QuantifiCare's U.S. Pat. Nos. 10,070,119, 10,165,253, and 10,334,xxx '334). Board found claims unpatentable; Federal Circuit affirmed (closed Mar. 25, 2025).
   • Germany: Düsseldorf Regional Court, 4a O 78/20 (judgment Apr. 8, 2025 — injunction re: Vectra H2; appeal to be heard Sept. 2026); German nullity BPatG 2 Ni 73/20(EP) and BGH X ZR 88/22.
   • UPC: UPC_CFI_559/2024 and UPC_CFI_106/2025 (Düsseldorf Local Division, decision Apr. 23, 2026 — patent EP 3 156 843 valid and infringed; pan-territorial injunction for BE/FR/IT/NL).
   • Patents at issue: EP 3 156 843 and its U.S. counterparts — not the '795 patent.
   • Sources: https://www.juve-patent.com/cases/canfield-faces-injunction-in-five-european-countries-in-quantificare-dispute/; https://www.eplaw.org/blog/detail/upc-quantificare-v-canfield-carve-out/; https://www.courtlistener.com/opinion/[10363459](/patent/10363459)/quantificare-sa-v-canfield-scientific-inc/

Caveat

I can only report what the searched public sources show. Litigation databases are not exhaustive (e.g., very recent suits may not yet be indexed, and some dockets require PACER access). If a case asserting the '795 patent has been filed but is not yet reflected in these public trackers, I would not have found it. Based on the sources available to me today, there is no known litigation involving U.S. Patent No. 10,368,795.

If you'd like, I can run a more targeted PACER/Docket Alarm query on the patent number or on Canfield Scientific as a party to confirm the negative result against live dockets.

I'll verify the structured "no proceedings on file" result with targeted searches before writing up.

Let me run two more confirmatory searches — one on the patent owner's PTAB footprint and one on any post-2023 challenges.

One more check on litigation assertion of this specific patent, which informs the defensive posture.

Proceedings overview

Total AIA trial proceedings on US 10,368,795: zero. The structured "PTAB proceedings on file" block (USPTO Open Data Portal) returns no IPR, PGR, or CBM proceeding naming this patent, and my web searches surfaced none either — so the breakdown is 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution-denied. Bottom line for a defendant: nothing has been canceled and nothing has been tested. All 23 claims — including both independents, claim 1 (method) and claim 14 (apparatus) — carry their full statutory presumption of validity, and there is no § 315(e)(2) estoppel on the books running against anyone.

Because there are no proceedings to walk through individually, the analysis below explains the near-miss dockets you must not confuse with this patent, why the absence exists, and what it means strategically.

---

No proceedings on file — and the near-miss dockets you should not confuse with this patent

No per-proceeding entries are possible here; inventing proceeding numbers would be the one unforgivable error in this exercise. What I can do is separate out the real PTAB/CAFC activity in the Canfield orbit, none of which touches the '795 patent:

Note the direction of travel: in every one of these matters Canfield is the petitioner/defendant attacking someone else's patent — it is not a patent owner that has been hauled before the Board. That is a meaningful pattern signal, discussed below.

---

Strategic summary

Claim status. Because no AIA trial has ever been instituted, the claim ledger is trivially simple: claims 1–23 are all UNTESTED, none CANCELED, none SUSTAINED-by-the-Board. That cuts both ways. On the one hand, a defendant asserting invalidity starts from scratch — there is no FWD reasoning to adopt, no claim-construction ruling to leverage, no expert testimony to recycle. On the other hand, nothing about this patent has been "hardened" by surviving an adversarial validity attack either; the patent has never had to defend itself in an AIA forum, so its claims have no § 315(e)(2)-generated patina of confirmed validity, and the file history is the only intrinsic record a challenger must work around.

Estoppel landscape. There is no § 315(e)(2) estoppel attaching to any party on this patent, because no petitioner has ever been through a trial. That means every prior-art ground is theoretically available to a defendant today. Conversely, there is no adverse-estoppel trap: nobody has previously lost on this patent and thereby narrowed the universe of arguments. If you file first, you alone bear the estoppel risk — § 315(e)(2) will bar you (and your privies) from raising at trial any ground you raised or reasonably could have raised in the IPR, so ground selection at the petition stage is the whole ballgame.

Availability of AIA review tools today (2026-09-27). The patent issued 2019-08-06, so the nine-month PGR window closed on or about 2020-05-06 — PGR is unavailable. CBM was sunset for petitions filed after 2020-09-16 and would never have applied anyway: this is a dermatological imaging method/apparatus, not a "financial product or service," and it would have qualified as a technological invention. IPR is the only AIA vehicle left, and it is fully available. Watch the § 315(b) clock: one year from service of a complaint alleging infringement, and § 315(a)(1) bars a petitioner who previously filed a civil action challenging validity.

Pattern signals. No defensive aggregator (Unified Patents, RPX, etc.) appears anywhere in the chain. No petitioner has filed multiple IPRs against this patent (there are none at all). Canfield has litigated hard when its freedom to operate was at stake — it went to the Board three times against QuantifiCare and once against Melanoscan, and it is currently briefing validity and infringement before the UPC Düsseldorf Local Division on EP 3 156 843 (oral hearing held 2026-03-12; decision announced 2026-04-23). The absence of any IPR against the '795 patent across roughly seven years of enforceability, in a company that clearly knows the PTAB, is itself the leading data point: this patent does not appear to have been broadly asserted against well-resourced defendants who would have fired back.

---

Recommended next steps

If you are a defendant and a demand letter or complaint cites US 10,368,795:

1. Assume the patent is fully live and fully enforceable. It is Active, the 4th-year maintenance fee was paid 2023-04-03 (small entity), and the adjusted expiration is 2035-11-09 — roughly nine more years of exposure. Do not expect to find a "survived two IPRs, weakened" narrative; the opposite is true.
2. Do not cite an FWD — there isn't one. If any counterparty tells you this patent was invalidated or narrowed at the PTAB, that is false. The only FWDs in the neighborhood are on the QuantifiCare patents ('119/'253/'334) and the Melanoscan '748 patent.
3. Build the invalidity case from the cited art of record, but plan around § 325(d). The strongest candidates are the examiner-cited references on the face of the patent: US 5,760,407 (Elizabeth Arden / Unilever — "Device for the identification of acne, microcomedones, and bacteria on human skin," 1998-06-02) and WO 98/24360 A1 (Lui, "Fluorescence scope system for dermatologic diagnosis," 1998-06-11). Those are the art most squarely aimed at the core concept. But precisely because they are of record, a petition resting solely on them invites discretionary denial under § 325(d) as art "previously presented to the Office" (see Advanced Bionics v. Med-El). Your petition should lead with art the examiner never saw, or articulate the Becton Dickinson factors for why the Office materially erred.
4. Exploit the claim-drafting gap you already know about. Claim 1 and claim 14 both recite detecting a porphyrin "whose fluorescence emission is primarily in an orange-red band," while the capture filter is claimed as passing a green-yellow band (claims 1, 14) — and the specification describes a green filter centered at 540–560 nm and coproporphyrin III emission at ~615 nm (orange-red). This internal tension between the claimed detection band and the claimed capture band is a written-description/enablement and claim-construction target worth developing early, ideally before you commit to an IPR ground, because § 112 issues are not IPR-eligible grounds (§ 311(b) limits IPR to §§ 102 and 103). Keep § 112 arguments for the district court or an ITC/EDR response.
5. Timing. If you have been served with a complaint, your § 315(b) one-year bar runs from service — diarize it immediately. If a decision to file is contested internally, remember that a stay of litigation pending IPR is far more likely when the petition is filed and instituted early.
6. Verify before relying. I could not query PTAB E2E, PATSTAT, or PACER directly; the zero-count conclusion rests on the USPTO ODP structured block plus web searches. Before filing, confirm on the Board's E2E system and confirm that this patent has not been asserted in any district-court action (I found no such action). A proceeding filed within the last few weeks may not yet be indexed in the sources I could reach.

One flag on the earlier sections. The prior summary's search-scope note is consistent with what I found and I have no contradiction to report — except to underline it: the "no PTAB activity on file" default, and the caution that litigation conclusions are search-result-based rather than authoritative docket checks, both hold. Separately, if you encounter a PTAB petition referencing US 6,571,003 B1 (a P&G "skin imaging and analysis" patent that appears among the '795's cited references), that is a challenge to the reference, not to this patent — do not let it migrate into your '795 file as if it were a proceeding against the '795.

Sources consulted: PTAB proceeding counts per the USPTO Open Data Portal structured block in the prompt; patent bibliographic and claim text at https://patents.google.com/patent/US10368795/en; QuantifiCare v. Canfield oral hearing record and three-case consolidation (APJs McNamara, Hamann, Range) at the IPR2021-01511/01518/01519 docket; CAFC No. 23-1917 disposition at https://www.courtlistener.com/opinion/[10363459](/patent/10363459)/quantificare-sa-v-canfield-scientific-inc/; Canfield v. Drugge / Melanoscan D.N.J. 2:16-cv-04636 and its account of IPR2017-02125 at https://cases.justia.com/federal/district-courts/new-jersey/njdce/2:2016cv04636/[335928/139](https://assignmentcenter.uspto.gov/search/patent/reelFrameDetail?reelFrame=335928-0139)/0.pdf; UPC Düsseldorf Local Division orders in UPC_CFI_559/2024 and UPC_CFI_106/2025 at https://www.unifiedpatentcourt.org/; Canfield virtual patent marking (VISIA-CR ↔ U.S. 10,368,795) at https://canfieldresearch.com/patents/.

I'll search for the USPTO assignment records for this patent.

Let me dig deeper into the assignment records and ownership history.

Let me check for any post-issuance assignments, security interests, or ownership changes.

Ownership analysis — US 10,368,795 B2 ("Acne imaging methods and apparatus")

Verification caveat up front: I was unable to open the USPTO Assignment Center / Patent Assignment Search results pages directly in this session (the search index surfaces the summary of the record, not the searchable portal itself). The reel/frame below is corroborated from the authoritative patent record (Google Patents legal events, which reproduces the recorded assignment abstract), and I cross-checked the ownership picture against Google Patents, PlainSite's Canfield assignor table, patent-review indices, and Canfield's own virtual patent marking. I could not independently verify a second or later recorded assignment, and I could not retrieve the "correspondent" field on the assignment cover sheet. Those gaps are flagged where they matter rather than filled by inference.

---

Inventors

Pattern notes:

• Single-inventor patent. No co-inventor, so no intra-team divergence to track.
• No inventor-departure signal. Patwardhan did not leave after filing — the PlainSite Canfield assignor table shows him executing assignments to Canfield on a continuous stream of later filings (e.g., 14/813,082 in 2015; 16/183,659 in 2018; 16/374,608 and 16/402,149/16/402,155 in 2019; 16/896,849 in 2021). He also appears as a named author on Canfield's RBX white paper and its acne multi-spectral fluorescence abstract.
• Conclusion: the classic "all inventors gone within 12 months → fire-sale" precursor is not present. This is a career inventor at the assignee, which is the opposite pattern.

---

Original assignee

Canfield Scientific, Incorporated — the entity named on the face of the issued patent and the record assignee.

• Business: designs and sells clinical/clinical-research skin-imaging hardware and analysis software for dermatology, aesthetics, pharmaceutics, and clinical trials. Flagship lines: VISIA / VISIA-CR, VECTRA, NEXA, VEOS, VAESTRO, and the Mirror software platform. It is a genuine instrument maker, not a holding entity.
• Did they ship a product embodying the claims? Yes. Canfield's virtual patent marking page lists U.S. 10,368,795 against four separate product families: VISIA-CR, VEOS, VAESTRO, and the "Image Processing (Analysis & Simulation)" category, alongside EP 3,160,327 and German utility registration DE 60 2015 080 889.4. The patent's own spec names the VISIA-CR system as the exemplary embodiment.
• Status: operating, private, independently held. Founded 1988 (~50–270 employees depending on source); no funding rounds; company description confirms no acquisition of/merger into any parent. Address of record has moved from 253 Passaic Avenue, Fairfield, NJ 07004 (1999–2015 era) to 4 Wood Hollow Road, Parsippany, NJ 07054 (current, per EP register "Titulaire" and the patent's assignment record).
• Not in bankruptcy; no Chapter 7/11 event found.

---

Assignment timeline

One recorded assignment exists in the U.S. chain — the original inventor→company conveyance. There are no recorded post-issuance transfers, security interests, mergers, or change-of-name records for this patent.

• 2015-04-17 (executed) / recorded 2015-04-17 — Reel 035437 / Frame 0921
  • Conveyance: Assignment of assignor's interest
  • Assignor: Sachin V. Patwardhan
  • Assignee: Canfield Scientific, Incorporated
  • Correspondent: ⚠️ Not verifiable from the sources retrieved. The Google Patents legal-event abstract reproduces the reel/frame but not the cover-sheet correspondent field. For context only (and not a substitute for the recorded correspondent), the prosecution attorney of record on this application was Brosemer, Kolefas & Assoc. LLC (per patents-review examiner/agent metadata). I will not assert that this firm is the recording correspondent, since I could not read the PTO-1595 cover sheet. No recurrence finding can be made.
  • Context: Internal acquisition — the standard inventor-to-employer assignment executed two days after the non-provisional was filed, contemporaneous with the application. Ordinary corporate IP capture, not a third-party transfer.
  • Turnaround: executed 2 days after the 2015-04-15 filing date — recordation discipline is tight (well inside the 3-month §261 window).

Note on a look-alike record to exclude: PlainSite shows a different assignment, Mirror Software Corporation → Canfield Scientific, Inc. (paperwork 1999-07-13, recorded Reel 10105/269, 1999-07-19). That is a 1999 acquisition of different IP by Canfield and is not part of the '795 chain. Likewise, Canfield's litigation/PTAB activity (below) involves other patents. Neither belongs in this timeline.

If Assignment Center at the time of your own check returns only the 035437/0921 entry, that is the expected and complete result — a single first-party assignment, which independently indicates the original assignee still owns the patent.

---

Timeline diagram

timeline
    title Ownership of US 10368795
    2014 : Provisional filed by inventor
    2015 : Non-provisional filed 15 April
         : Assigned to Canfield Scientific
         : Application published 31 December
    2019 : Patent US 10368795 issued
    2023 : Fourth year maintenance fee paid
    2035 : Adjusted patent expiration

---

NPE / troll-pattern signals

Signal 3 is the only one I cannot resolve, and it resolves to "unclear" for lack of a single-link recurrence, not to a positive finding. Per the operating rules, a name-based hunch about Canfield's outside patent counsel would not be evidence, and I decline to make one.

---

Verdict

Operating-company assertion.

Canfield Scientific, Incorporated both ships products embodying the claims (the '795 number is marked on VISIA-CR, VEOS, VAESTRO, and the image-processing software line) and asserts/enforces its skin-imaging portfolio against a direct competitor (QuantifiCare — IPR2021-01511/-01518/-01519; CAFC No. 23-1917). Ownership never left the original operating assignee: the sole recorded transfer is the inventor→company assignment at Reel 035437/0921, executed 2015-04-17, with no subsequent links, no security interests, and no NPE, aggregator, or shell entity anywhere in the chain.

Two precision caveats that keep this honest: (a) I found no infringement suit asserting the '795 patent itself — the enforcement activity is at the portfolio/competitor level, so the "suing actual competitors" prong is met by company conduct, not by a '795-specific complaint; and (b) the cover-sheet correspondent on Reel 035437/0921 was not verifiable in this session, so the repeat-correspondent signal could not be scored.

Verify at:

• USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search "10368795" and "14/686,824"; expect the single hit at Reel 035437/0921)
• Legacy assignment search — https://assignment.uspto.gov/patent/index.html
• Google Patents legal events (reproduced in the authoritative text) — https://patents.google.com/patent/US10368795/en
• Canfield virtual marking — https://canfieldresearch.com/patents/

Note on the prior section: the earlier "Related litigation" section correctly distinguished the Canfield/QuantifiCare and Canfield/Melanoscan matters from this patent. That distinction holds and is the basis for the caveat above — this patent has not been shown to be asserted, even though its owner is a demonstrated enforcer.

I'll research the cited prior art for US 10,368,795, focusing on accurate bibliographic details for each reference.

Let me gather details on additional cited references.

I've reached my research limit. Here is my prior-art analysis for US 10,368,795 B2 ("Acne imaging methods and apparatus"), built on the previously generated sections (which I do not repeat), focusing on the 17 references cited on the face of the patent.

---

Methodology and a necessary caveat

I analyzed the 17 patent documents listed in the "Patent Citations" / "Citations" tables of US 10,368,795 (as reproduced in the authoritative full text and confirmed via Google Patents, Justia, FreePatentsOnline, and the EPO register). Two important analytic points up front:

1. Anticipation vs. obviousness. A § 102 anticipation requires every element of a claim to be disclosed in a single reference. As noted in my prior analysis, the independent claims 1 and 14 require a specific combination: blue excitation, capture through a green-yellow pass filter, detection of porphyrin fluorescence (orange-red), and selection of spots on a conjunctive intensity + shape + bounded-area test (lower bound ≥150 µm/12 px). No single cited reference discloses that full combination. Accordingly, most of these references are § 103 obviousness art, not clean § 102 art. I flag below where a reference is genuinely close to anticipating a dependent-claim element versus merely being analogous art.
2. Labeling. The patent's table marks references with an asterisk (*) = "cited by examiner"; unmarked = cited by applicant. I preserve those designations because they indicate which references the applicant brought forward (many of the Canfield family members, i.e., self-citations, are applicant-cited).

---

Tier 1 — Most material references

1. US 5,760,407 A — Device for the identification of acne, microcomedones, and bacteria on human skin

• Assignee/inventor: Elizabeth Arden Co., Division of Conopco, Inc.; inventors Aronson et al. (Unilever family; counterpart EP 0 783 867 B1; ZA 967500).
• Filing/priority: filed Dec. 21, 1995 (continuation of Ser. No. 08/576,526); granted June 2, 1998.
• Description: A device and method for detecting fluorescing follicles — expressly including microcomedones ("clinically non-evident acne lesions") — and for distinguishing bacteria-populated from non-bacterial follicles. It irradiates skin with UV/UVA light (Wood's-lamp-type; longpass >350 nm, or broadband UVA 320–400 nm via bandpass filter) and detects both yellow/green (500–580 nm) and orange/red (580–680 nm) fluorescence through detection filters that cut off radiation below ~450–480 nm. It expressly explains that "the bacteria produce porphyrins which upon excitation, fluoresce orange-red" and connects bacterial presence to P. acnes.
• § 102 relevance: This is the single most on-point prior-art reference on the diagnostic principle of the patent — microcomedone detection via bacterial porphyrin fluorescence. It potentially anticipates the conceptual core recited in the preamble/body of claims 1 and 14 (illuminating skin, filtering to isolate the porphyrin emission band, and associating fluorescent spots with microcomedones). However, it does not disclose: (a) blue (400–410 nm) excitation specifically (claim 5/16); (b) capture through a green-yellow (525–575 nm) emission filter (claim 6/17); or (c) the digital image-processing selection logic of claim 1 (intensity threshold, regular shape, bounded area). So it is best characterized as anticipating/narrowing the field of claims 1 & 14 rather than a literal § 102 hit, and as very strong art against any claim attempting to monopolize generic "detect microcomedones by porphyrin fluorescence" language. It is also directly relevant to the dermatoscope aspect of the disclosure (compare claim 14's viewing opening + filter architecture).

2. WO 98/24360 A1 — Fluorescence scope system for dermatologic diagnosis (Lui et al.)

• U.S. counterpart: US 6,021,344 (Lui, Zeng, MacAulay, Palcic, McLean; assignee Derma Technologies, Inc.), granted Feb. 1, 2000; filed Dec. 3, 1997; priority CA 2,192,036, Dec. 4, 1996.
• WO publication date: June 11, 1998.
• Description: A fluorescence-imaging scope for skin-disease diagnosis: an excitation source (explicitly including a 442 nm He-Cd blue laser, a mercury-arc 405 nm blue line, and a Wood's lamp), a light guide, and viewing goggles carrying bandpass filters ("green light and red light"; green defined as ~480–560 nm) that block the excitation light and pass the longer-wavelength fluorescence to the user. Supports both full-color and "enhanced two-color" fluorescence images.
• § 102 relevance: This reference is highly material to the apparatus claim 14 and to claim 6/17 (green-band filter): it discloses a blue excitation source plus a filtered green-band viewing apparatus (spectacles/goggles) for skin fluorescence — structurally analogous to the FIG. 4 dermatoscope (illumination source + green filter in the optical path). It does not disclose microcomedone-specific processing or the spot-selection logic. Note the spectral mismatch: Lui's green band (480–560 nm) versus the '795 claim's green-yellow band (525–575 nm), and Lui targets skin disease/malignancy diagnosis, not acne. Potentially anticipates the apparatus environment of claim 14 but not its processing circuitry limitations.

3. US 2009/0137908 A1 — Multi-spectral tissue imaging (Patwardhan) — applicant-cited

• Parent grant: US 8,849,380 B2 (granted Sept. 30, 2014; filed Feb. 26, 2008; assignee Canfield Scientific Inc.); PCT counterpart WO 2009/070370.
• Publication date: May 28, 2009.
• Description: Multi-spectral imaging of tissue to map the distribution of fluorophores and chromophores. It expressly tabulates coproporphyrin excitation/emission (405 nm / 620 nm, "P. acnes") and discusses normalizing a fluorescence image against a chromophore absorption distribution map (hemoglobin, melanin) and an excitation image. It also lists applications such as classifying open/closed comedones, papules, pustules and nodules and predicting acne lesion sites.
• § 102 relevance: This is a self-cited Canfield/Patwardhan reference. It is directly material to claim 2 and claims 9–11 (correcting the captured image by normalizing against a chromophore distribution image for hemoglobin and/or melanin, and/or an excitation image). Because it is by the same inventor and same assignee, it is largely § 102(a)/§ 102(e)-type art against the correction sub-features and a § 103 reference for the overall framework — it arguably anticipates claims 9, 10 and 11 on their own (normalizing against chromophore/excitation images), even though it does not anticipate independent claims 1 or 14 as a whole. The specification of the '795 patent itself cross-references US 8,849,280 for these correction techniques.

4. US 6,571,003 B1 — Skin imaging and analysis systems and methods (Hillebrand et al.)

• Assignee: The Procter & Gamble Company.
• Filed June 14, 1999; granted May 27, 2003.
• Description: Acquisition and electronic analysis of digital skin images to locate defect areas (including acne/red spots, follicular pores, and others), generate a further image visually identifying the defects by altering pixels / overlaying, compute a numerical severity, compare to a population, and track over time.
• § 102 relevance: Material to the "generating a further image … indicating the subset of spots" step of claims 1 and 14, and to claims 12–13 (comparing captured/further images with a stored image for treatment monitoring). It does not disclose fluorescence, porphyrins, blue/green filtering, or microcomedones. It is a § 103 reference for the image-annotation/monitoring features; it does not anticipate any claim as a whole. (Note: the '003 patent was itself the subject of Procter & Gamble v. QuantifiCare, where several claims were held ineligible under § 101 — a background point not bearing on anticipation.)

---

Tier 2 — Material to specific sub-features

5. WO 99/66830 A1 — Endoscopes and methods relating to direct viewing of a target tissue

• Applicant: Biomax Technologies Inc.; published Dec. 29, 1999 (U.S. counterpart US 6,110,106).
• Description: Direct-viewing endoscope/optical systems for viewing target tissue (uses filtering at the viewing end; generally relates to tissue fluorescence/reflectance viewing hardware).
• § 102 relevance: Peripheral. Relevant only as general background for filtered optical viewing apparatus (claim 14 environment). No acne, porphyrin, or microcomedone disclosure.

6. US 2002/0090123 A1 — Methods for enabling evaluation of typological characteristics of external body portion, and related devices (Roland Bazin)

• Filed Dec. 21, 2000; published July 11, 2002.
• Description: Imaging/evaluation of external body (skin) typological characteristics.
• § 102 relevance: Peripheral background on skin-attribute imaging and analysis; no fluorescence or microcomedone teaching.

7. US 2004/0174525 A1 — Dermoscopy epiluminescence device employing cross and parallel polarization (Mullani)

• Filed March 7, 2003; published Sept. 9, 2004.
• Description: Dermatoscope using cross- and parallel-polarization for viewing skin.
• § 102 relevance: Material background for the polarized-image processing used in the FIG. 3 inflammation-filtering embodiment (cross-polarized image; RBX). It supports the obviousness of using polarization in a skin-imaging apparatus but does not anticipate claims 1–23.

8. US 2013/0006118 A1 — Radiative fibers (Merck Patent GmbH)

• Filed March 11, 2010; published Jan. 3, 2013.
• Description: Radiative/emissive fiber technology (e.g., light-emitting textile fibers).
• § 102 relevance: Peripheral — likely cited for illumination-source technology (cf. claim 15's LED/emissive light source). No diagnostic or acne content.

9. US 2007/0202504 A1 (Sony) / US 8,472,682 B2 (DVP Technologies) / US 8,290,257 B2 (P&G) / US 2008/0180950 A1 (KETI) / US 2009/0054744 A1 (Kitamura) / US 2005/0195316 A1 (Ethicon) / US 2006/0092315 A1 (J&J) / US 2009/0299268 A1 (MGH)

These eight are lower-relevance background. Briefly:

• US 8,472,682 B2 (Medical image processing, DVP Technologies; filed Sept. 12, 2005; granted June 25, 2013) — generic medical image-processing/segmentation; supports the obviousness of digital spot detection/segmentation (relevant to the "detecting fluorescent spots" step), not acne-specific.
• US 8,290,257 B2 (Method and apparatus for simulation of facial skin aging and de-aging, P&G; filed March 2, 2007; granted Oct. 16, 2012) — expressly cross-referenced in the '795 spec as describing mask/AOI delineation techniques; material to the area-of-interest and image-simulation features, not to anticipation of any claim as a whole.
• US 2008/0180950 A1 (Korea Electro Technology Research Institute; filed Jan. 26, 2007; published July 31, 2008) — Apparatus for photodynamic diagnosis of skin diseases with improved uniformity of illumination — background on uniform skin illumination for fluorescence diagnosis (cf. the '795 step 220 correction for non-uniform excitation light).
• US 2009/0054744 A1 (Kitamura; filed April 28, 2005; published Feb. 26, 2009) — Skin state analyzing method/apparatus — background on skin-state analysis from images (supports obviousness of quantitative skin-metric image analysis).
• US 2005/0195316 A1 (Ethicon; filed Nov. 8, 2001; published Sept. 8, 2005) and US 2006/0092315 A1 (J&J Consumer; filed Oct. 29, 2004; published May 4, 2006) — skin-imaging camera/probe hardware; generic apparatus background for claim 14.
• US 2009/0299268 A1 (The General Hospital Corp.; published Dec. 3, 2009) — Topical ALA-photodynamic therapy for acne vulgaris — acne/porphyrin biology background (photodynamic therapy, not imaging/detection).
• US 2007/0202504 A1 (Sony; priority April 16, 2014; published July 20, 2017) — generic information-processing; peripheral. Note this is a post-priority (2014-04-16) publication yet appears in the citation table; if applicable only as of its later publication date it cannot be § 102(b) art and would be marginal § 102(a)/(e) at best.

10. WO 2014/047712 A1 — Imaging device of facial topography with multiple light source flash photography and method of blending same

• Applicant: Windsor Clinical Research Inc. (Tan, Nhan, Ling); published April 3, 2014 (filed Sept. 26, 2012; CA 2012/000886).
• Description: Multi-light-source flash photography of the face with image blending to enhance topographic/dermatological detail (expressly mentioning acne); combines reflectance/polarized/fluorescence digital photography.
• § 102 relevance: Published just before the '795 priority date (June 30, 2014) and expressly discusses acne and fluorescence digital photography for facial topography; material to the general facial-imaging/flash-capture architecture (claim 1 "capturing an image") and possibly the further-image generation. It does not disclose porphyrin/microcomedone detection or the selection logic; a § 103 reference.

---

Overall § 102 conclusion

• No single cited reference anticipates independent claim 1 or independent claim 14 in their entirety. The conjunction of (i) blue excitation, (ii) green-yellow capture band, (iii) orange-red porphyrin detection, and (iv) the intensity + regular-shape + bounded-area selection test is not disclosed by any one of the 17 references.
• Best § 102 candidates for sub-claim elements:
  • US 5,760,407 — closest art on the microcomedone/porphyrin-fluorescence detection principle; the strongest reference for the preamble concept of claims 1 and 14 (and for the general "small, regularly-shaped bright spots" observation), but overlapping only partially with claim scope.
  • US 2009/0137908 A1 / US 8,849,380 B2 (Patwardhan/Canfield) — arguably anticipates the image-correction limitations of claims 9–11 (chromophore/excitation-image normalization) on their own.
  • WO 98/24360 / US 6,021,344 (Lui) — closest art on the blue-excitation + green-filtered-viewer apparatus (claim 14 environment; claim 6).
• Most of the list functions as § 103 art, typically combinable (e.g., US 5,760,407's porphyrin/microcomedone fluorescence + US 6,571,003's digital defect-location-and-annotate workflow + US 8,472,682's image segmentation/spot analysis) to challenge the obviousness of the claimed processing pipeline — but that is an obviousness theory, not anticipation.

Verification caveat: The bibliographic details above are drawn from the patent's own citation tables cross-checked against Google Patents, Justia, FreePatentsOnline, uspto.report, and the EPO/Espacenet family records identified in the searches. I was unable to independently re-pull the full text of every one of the 17 references before reaching my search limit; where a reference's teaching is described above at a high level (Tier 2 items 5–9), that description is from its title/abstract and general classification rather than a full element-by-element comparison. Treat the § 102 "potentially anticipates" tags as issue-spotting leads to be confirmed by reading each reference's actual disclosure against the claim charts, not as a legal conclusion.

I'll analyze the obviousness challenge using the prior art cited on the patent page. Let me first ground myself on the key references.

Let me gather the remaining cited references that bear on the detection/quantification steps.

Obviousness Analysis — U.S. Patent No. 10,368,795 B2 under 35 U.S.C. § 103

Building on the earlier bibliographic, claim-construction, and litigation sections. This section does not repeat that material; it takes the claim breakdown and the 17 references on the face of the patent (plus the family citations and non-patent items listed on the same page) as the starting point.

---

0. Scope, framework, and a threshold caveat on the art set

Governing law. The earliest priority (provisional 62/018,866, 2014‑06‑30) and the non‑provisional filing (2015‑04‑15) both postdate the AIA's March 16, 2013 effective date. Accordingly, AIA §§ 102/103 govern, not pre‑AIA § 103(a). This matters for two reasons:

1. It opens § 102(a)(2) "secret prior art" (US patent documents with an earlier effective filing date), which is how a reference like US 2017/0202504 A1 (Sony; priority 2014‑04‑16, published 2017‑07‑20) appears in the "Citations" list even though it published after the '795 priority date. Its subject matter is unrelated to acne imaging, so it is of no substantive use here.
2. It triggers the § 102(b)(2)(C) common-ownership exception for Canfield's own later-published documents. This is a live issue for US 2014/0243685 A1 (the Canfield dermatoscope application, published 2014‑08‑28, effective filing 2013‑02‑28) which the '795 specification expressly incorporates — it is within the family's common ownership, so it likely cannot be used as § 103 prior art, only as evidence of the state of the art. By contrast, the more important self‑owned references — US 2009/0137908 A1 (published 2009‑05‑28) and US 8,498,460 (granted 2013‑07‑30) — are § 102(a)(1) printed publications predating the critical date by more than a year, so common ownership does not save them.

Caveat on the "Citations (17)" list. Google Patents' list mixes examiner citations, applicant IDS submissions, and third‑party citations. It is not a holding that each item is § 102 prior art. I flagged the Sony item above. I would also note that the specification text cites U.S. Pat. No. 8,849,280 for the "light distribution and tissue absorption" correction technique, but that number does not appear in the page's 17‑item citation list; what does appear is the closely related US 2009/0137908 A1 (Patwardhan, "Multi‑spectral tissue imaging," assigned to Canfield; a member of the family that produced US 8,849,380). I cannot verify from the retrieved text whether "8,849,280" is a typographical variant, a different patent, or omitted from Google's citation list. I flag this rather than auto‑correct it, and I treat the substance (Patwardhan multi‑spectral fluorescence/absorption correction) as the relevant disclosure.

Non‑patent items — one is NOT prior art. The "Similar Documents" list on the page includes Chantharaphaichi et al., "Automatic acne detection for medical treatment" (2015). Because it is dated 2015, it postdates the 2014‑06‑30 priority date and is not available as prior art under either § 102(a)(1) or § 102(a)(2). Min et al., "Development and evaluation of an automatic acne lesion detection program using digital image processing" (2013) and Malik et al., "Grading of Acne Vulgaris Lesions" (2014) predate the critical date and are available. I could not retrieve the full text of Min et al. or Malik et al. within this session, so I cite them only for what their titles and the page's listing establish — that automated, image‑processing‑based acne lesion detection/counting was known in the art before June 2014.

Framework. Under Graham v. John Deere and KSR Int'l v. Teleflex, I analyze (a) the scope and content of the prior art, (b) the differences between the claims and the art, (c) the level of ordinary skill, and (d) objective indicia. Under KSR, a combination is obvious where a POSITA would have had a reason to combine the references with a reasonable expectation of success, including where the combination is "use of a known technique to improve a similar device in the same way," "application of a known technique to a known device ready for improvement," or "obvious to try" among a finite number of identified, predictable solutions.

Level of ordinary skill. I adopt the description the PTAB and parties accepted in the parallel '003‑patent proceeding (IPR2017‑02113, addressing US 6,571,003): a POSITA is a dermatologist (or skin‑imaging scientist) familiar with computers and image processing. See PTAB Decision, IPR2017‑02113. That is a fair benchmark here, since the '795 patent pairs a dermatological fluorescence technique with conventional digital image processing.

---

1. Claim 1 — element‑by‑element mapping to the prior art

Observation on the internal claim tension. Claim 1 requires capturing through a green‑yellow filter while detecting a porphyrin whose emission is "primarily in an orange‑red band." US 5,760,407 accommodates this literally, because its detector filter passes the entire 480–680 nm span — encompassing both the green‑yellow band and the orange‑red porphyrin emission — so a single broadband capture satisfies both limitations. This is a mapping point in the challenger's favor, and separately a § 112 written‑description/claim‑consistency concern (the specification's own working example uses a green filter centered at 540–560 nm, not an orange‑red filter). I flag it because it cuts both ways (see § 6).

---

2. Ground 1 (claim 1): US 5,760,407 + US 2005/0195316 A1 + US 6,571,003

The combination. US 5,760,407 supplies the dermatological detection principle (blue/UV excitation → porphyrin orange‑red fluorescence → detection of non‑visible microcomedones), the green‑yellow detection band (500–580 nm), a portable device, and CCD‑camera readout. Kollias '316 supplies the digital blue‑fluorescence implementation (blue 380–430 nm, blue‑filtered flash, long‑pass camera filter, digital camera, touch‑screen review). P&G '003 supplies the automated image‑processing and display framework (digital acquisition, sub‑image/AOI determination, electronic defect location, generation of a marked overlayed image, and comparison against stored/population data).

Motivation to combine (articulated):

1. Express need in the art. US 5,760,407 states: "It is desirable to be able to detect follicular impactions and microcomedones in order to provide a timely skin treatment and to prevent the appearance of comedones." The '795 specification's own background repeats this as the unmet need. A POSITA therefore had a concrete, stated problem to solve.
2. Known technique, same problem. US 5,760,407's detection means is "human eyes, a film emulsion camera, and a CCD camera." Substituting a modern digital camera and adding Kollias's blue‑fluorescence capture mode is "use of a known technique to improve a similar device in the same way," yielding a predictable result (KSR).
3. Automation for reproducibility. P&G '003's stated rationale — that manual adviser assessment "results very subjective," varies "visit to visit," and is hard to communicate — supplies the motivation to move from a human reading a dermatoscope to an automated, quantitative, displayed result. The '795 specification touts exactly these benefits (dispensing, storage, comparison over time).
4. Reasonable expectation of success. All three are in the same field (skin imaging/fluorescence), the imaging hardware is described as conventional ("digital SLR camera," "xenon flashes"), and the '795 specification characterizes the processing primitives (Top‑hat filters, thresholding) as "known techniques."

Relationship to the "detecting fluorescent spots / subset selection" step. The claim's three filtering criteria (intensity, shape, size) are each individually known image‑analysis operations; the references furnish the reason to apply them: US 5,760,407 itself distinguishes among follicular impactions, microcomedones, comedones, and bacteria‑populated vs. non‑populated follicles — i.e., a POSITA was already motivated to discriminate classes of fluorescence spots by measurable attributes rather than simply count them.

---

3. Ground 2 (claims 2, 9, 10, 11 — the correction step): + US 2009/0137908 A1 (Patwardhan)

The dependent claims add correcting the fluorescence image against (a) a chromophore distribution image (hemoglobin and/or melanin) and (b) an excitation image, i.e., normalizing for uneven illumination and tissue absorption.

US 2009/0137908 A1 discloses precisely this: it teaches that skin's main visible chromophores are hemoglobin and melanin, that multi‑spectral imaging uses "specific spectral bands for illumination and specific spectral bands for detection" to enhance the SNR of a selected fluorophore over single‑RGB UV photography, and that "the chromophore distribution information derived from the multi‑spectral absorption images can be used to correct the fluorescence measurements." Critically, its Table 1 lists Coproporphyrin: excitation max 405 nm, emission max 620 nm, physiologic parameter "P. acnes." US20090137908A1 / US 8,849,380.

Motivation. US 2009/0137908 articulates the same technical purpose as the '795 correction step — improving fluorophore SNR by removing (i) illumination non‑uniformity and (ii) chromophore absorption. It is not merely analogous art; it is the same field, by the same inventor, addressing the same defect. Under KSR, a POSITA would apply its teaching to the US 5,760,407 / Kollias blue‑fluorescence image with a reasonable expectation that the SNR of the porphyrin signal would improve. Claim 11's "excitation image … captured while illuminating the skin with the blue light" is the natural companion capture and is reflected in multi‑spectral illumination/detection schemes generally.

---

4. Ground 3 (claim 3 / claim 22 — inflammation and comedone‑feature filtering): + US 8,498,460 (RBX) + US 2004/0174525 (Mullani)

Claim 3 requires excluding spots "associated with inflammation or that do not have comedone features."

• US 8,498,460 (Canfield, "Reflectance imaging and analysis for evaluating tissue pigmentation") is the RBX disclosure the '795 specification itself incorporates. It yields RBX‑Red (hemoglobin/erythema) and RBX‑Brown images from a cross‑polarized image. Family citation listed on the '795 page
• US 2004/0174525 A1 (Mullani) discloses "Dermoscopy epiluminescence device employing cross and parallel polarization" — the acquisition modality RBX depends on.

Motivation. The '795 specification itself concedes that "correcting … for hemoglobin absorption is optional," because microcomedones are non‑inflammatory, but explains that for subjects with inflammatory lesions "hemoglobin absorption correction can be used … to improve the detectability of microcomedones that may otherwise be masked." That is an express, self‑identified reason to bring hemoglobin/inflammation maps into the pipeline. US 8,498,460 provides the map (RBX‑Red) and the RBX‑Brown image used to identify dark enlarged pores (blackheads); Mullani provides the polarization. The combination is the predictable application of a known inflammation‑mapping technique to a known fluorescence‑detection pipeline to improve discrimination — a textbook KSR "known technique to improve a similar device" rationale.

---

5. Ground 4 (claims 4 and 20 — pore detection and pore‑size‑based thresholding): + US 6,571,003 + US 2005/0195316 A1

Claim 4 requires detecting pores and retaining only spots whose locations correspond to pores; claim 20 bases the area threshold on mean/median pore size.

• P&G '003 explicitly lists "(e) follicular pores" among the defects its system locates, and teaches determining a sub‑image/AOI before analysis. US6571003
• Kollias '316 teaches that "polarized" and cross‑polarized photography gives "insight into the number and severity of pores on the cheek and forehead areas of the facial skin," and that parallel‑polarization enhances surface features. US20050195316A1

Motivation. The '795 specification states the rationale itself — microcomedones "being confined to pores, appear relatively round," and the post‑shape‑filter spot population "are most likely representative of microcomedones and comedones and thus confined to pores." Because a microcomedone is by definition a plugged follicle (as US 5,760,407 likewise explains: follicular impactions form in the duct lumen), requiring co‑location with a detected pore is the natural, predictable way to reduce false positives. Deriving the area threshold from measured pore size is a conventional "result‑effective variable" optimization (In re Boesch / In re Aller).

---

6. Ground 5 (claim 14 — the apparatus): US 5,760,407 + US 2005/0195316 A1 + US 2006/0092315 A1 (+ US 6,021,344)

Claim 14 requires: an image capture device with a viewing opening; a light source arranged about the viewing opening; a filter passing a green‑yellow band (blue light emitted when activated); and processing circuitry performing the claim‑1 detection logic and generating a further image.

Motivation. US 5,760,407's device is described as portable and usable "by dermatologists as a research tool"; combining it with the self‑contained, camera‑plus‑light‑source‑plus‑controller architecture of Kollias '316 (housing, camera, light source, display, controller) yields the claimed apparatus with predictable results. The "processing circuitry configured to … detect fluorescent spots / select a subset" limitation is the same algorithmic content as claim 1, so Grounds 1–4 apply mutatis mutandis.

---

7. Dependent‑claim ranges — available or obvious

---

8. Where an obviousness challenge is weakest — arguments the patent owner will press

A responsible § 103 opinion must identify the counter‑arguments:

1. Possible teaching away on the green‑yellow channel. US 5,760,407 expressly correlates orange‑red fluorescence with bacteria‑populated (porphyrin‑containing) follicles and yellow/green fluorescence with non‑bacterial follicular impactions. A POSITA told to detect bacterial porphyrin would be led to the orange‑red channel, not the green‑yellow channel. The '795 patent's choice to detect through a green‑yellow filter while claiming an orange‑red‑emitting porphyrin is arguably contrary to the prior‑art teaching, which the patent owner can frame as (a) teaching away and/or (b) an unexpected result. The specification's stated reason — that coproporphyrin III's 615 nm peak is unusable because of protoporphyrin IX overlap and Bayer‑filter gaps, so off‑peak detection is deliberately used — is the patent's strongest non‑obviousness narrative. It parallels the Sauerermann reference's 610 nm cut‑off approach that the art had adopted.
2. The bounded (two‑sided) area window. Claim 1's lower bound (≥150 µm diameter or 12 px area) is the least well‑met by the art. US 5,760,407/Sauermann measure "individual size," but neither expressly discards small spots as noise; nor does either reference set a floor. The patent owner can argue the references teach only "size" as a metric, not a two‑sided size window, and that a challenger must supply the floor from nothing more than general noise‑rejection reasoning. (Counter: the '795 specification itself justifies the floor as removing "noise or other unwanted features (e.g., lint, skin flakes)," which is a recognition that such a floor was an ordinary engineering choice.)
3. Prosecution history / issued claims. The examiner had these very references before him (they appear as citations), and the claims issued; the EPO likewise granted EP 3 160 327 B1 on the same family. While not dispositive under Graham, a challenger must explain why the examiner erred.
4. Objective indicia. The patent is part of the VISIA‑CR marked product (per Canfield's virtual patent marking), and the family has been litigated/defended, which the owner may use to argue commercial success and industry uptake. Any obviousness case should develop a nexus argument and evaluate whether the claimed combination (not just the commercial system) drove the success.
5. The § 112/§ 103 crossover. The green‑yellow‑filter/orange‑red‑emission tension noted in § 1 cuts against the patent owner on definiteness/enablement but helps them on non‑obviousness (it made the combination non‑routine). A challenger should decide which theory to lead with; they are in some tension.

---

9. Conclusion

On the reference set cited on the face of US 10,368,795, the strongest prima facie § 103 case is a three‑reference combination:

• US 5,760,407 (Conopco) — microcomedone detection by porphyrin fluorescence, green‑yellow + orange‑red bands, CCD readout, patient/portable device, express motivation for early detection;
• US 2005/0195316 A1 (Kollias, J&J) — the digital blue‑fluorescence (380–430 nm) imaging implementation with blue‑filtered flash and camera filter; and
• US 6,571,003 (P&G) — the automated acquire → AOI → locate defect → generate marked further image → compare over time pipeline;

optionally reinforced by US 2009/0137908 A1 (Patwardhan) for claims 2 and 9–11 (chromophore/excitation normalization; coproporphyrin 405/620 spectra), US 8,498,460 + US 2004/0174525 for claim 3 (RBX hemoglobin/inflammation and cross‑polarization), US 6,571,003 + US 2005/0195316 A1 for claims 4/20 (pore detection), and US 6,021,344/WO 98/24360 + US 2006/0092315 A1 for claim 14's apparatus form factor.

The principal vulnerability of that case is the bounded area window (claim 1's ≥150 µm/12 px floor) and the green‑yellow‑filter‑for‑orange‑red‑porphyrin mismatch, which the patent owner can develop as teaching‑away/unexpected‑results. The dependent‑claim numeric ranges and the apparatus claim are, by comparison, highly susceptible to a KSR‑style obviousness challenge.

Explicit limitations of this analysis

• I could not retrieve the full texts of several cited references (US 8,472,682, US 2009/0054744 A1 (Kitamura), US 2008/0180950 A1, US 2013/0006118 A1, US 2009/0299268 A1, US 8,290,257, US 2005/0195316's sibling US 2004/0146290) within this session, and I have cited them only by the titles/assignees shown on the '795 page. A full opinion should verify their disclosures.
• I could not verify the examiner's reasons for allowance or any applicant remarks during prosecution of the '795 patent; these are material to a complete Graham analysis and to any § 325(d)/General Plastic discretionary‑denial issue in an IPR.
• As noted, I did not auto‑correct the identifier "U.S. Pat. No. 8,849,280" cited in the specification versus the citation‑list entry "US 2009/0137908 A1"; the discrepancy is flagged, not resolved.
• The obviousness priority‑date analysis assumes the provisional 62/018,866 supports the issued claims; if any claim lacks provisional support, the effective filing date shifts to 2015‑04‑15, which would expand the available art (e.g., potentially capturing the 2015 Chantharaphaichi paper via § 102(a)(2) if it were a US publication, and removing any grace‑period questions). Chantharaphaichi (2015) is not prior art on the assumed 2014‑06‑30 priority date.

Sources: US10368795 (Google Patents) · US5760407 · US5760407 PDF · WO1998024360A1 · US6021344 (Lui) · US6571003 · US20090137908A1 · US8849380 · US20050195316A1 · IPR2017-02113 Decision