US 10324021

I'll search for the specific patent number and check for any litigation or CAFC docket activity.

Let me run a few more targeted searches on litigation/docket activity and the assignee.

Summary — US 10,324,021 B2

Important caveat up front: I do not have direct query access to USPTO Patent Center/PAIR or to the CAFC docket system. I ran targeted web searches for the literal number 10324021 and for assignee/litigation activity. Those searches returned no CAFC 2026 appeal, IPR, or district-court docket involving this patent. I cannot state affirmatively that no such proceeding exists — only that my searches surfaced none. Everything below is grounded in the patent text itself.

---

Bibliographic data

Note: the title and specification consistently use the spelling "magnetophorisis" (not the standard "magnetophoresis"). I preserve the identifier literally.

---

Abstract (as issued)

A magnetophorisis measuring system comprising a microscope device, a magnetic field generator, an image capturing unit, and a processing unit. The microscope magnifies a sample liquid having objects that each contain magnetic particles. The magnetic field generator applies an external magnetic field so the objects are moved. The image capturing unit captures a dynamic image of the fluid sample in the microscope's view field. The processing unit receives the dynamic image, automatically detects and locks moving objects, determines a motion status for each object, and quantifies the magnetic particles according to that motion status.

---

Plain-language overview of the independent claims

Claim 1 — velocity-based motion-status determination
A magnetophorisis measuring system with four elements:

1. Microscope — magnifies a sample liquid containing objects, each carrying magnetic particles.
2. Magnetic field generator — placed at one or both sides of the sample; drives the objects with an external field, and reverses the field direction on a control signal so objects reciprocate in the sample repeatedly during the measurement.
3. Image acquiring unit — coupled to the microscope, produces a video image of the field of view.
4. Processing unit — receives the video, automatically locks the moved object(s), and analyzes motion status.

The claim's characterizing limitation: the processing unit splits the video into frames, identifies the moved object from at least two frames, and determines motion status from two velocities. The first velocity uses the time taken to travel from a first boundary line to a second boundary line and the known distance between them; the second velocity uses the time from the second to a third boundary line and that distance. Both time periods are computed as frame number × (1/frame rate). Essentially: define virtual lines in the field of view and use frame counts to time how long the object takes to cross each zone.

Claim 16 — acceleration-based motion-status determination
The same four-element system (microscope, reversible magnetic field generator driving reciprocating motion, image acquiring unit, processing unit). The difference is the analysis: motion status is derived from two accelerations, which in turn come from three velocities measured across four boundary lines (B1→B2, B2→B3, B3→B4), each timed by frame number and frame rate. This lets the system distinguish constant-acceleration from variable-acceleration motion rather than simply detecting constant velocity.

(Claim 16's final time-period clause reads "…from the third boundary line to the boundary line…" — the word "fourth" appears to be missing. I am reporting the text literally rather than correcting it.)

---

Dependent claim themes

• Claim 2: motion status = constant velocity, constant acceleration, or variable acceleration.
• Claim 3: the processing unit runs at least one algorithm on the motion status to quantify the magnetic particles inside each moved object — this is the equation (3) step: N = 36ηR_cell·V / (b·M_s·d³·dB/dx), equating Stokes viscous drag (6πηRv) to magnetic force (m_bead·dB/dx).
• Claim 4: field magnitude is varied by control signal to widen the measurable population and size range of objects.
• Claim 5: sample on a 1-D horizontal, 2-D horizontal, or vertical flow channel.
• Claim 6: object = bacterial strain, cell, protein, antibody, antigen, drug, or chemical molecule.
• Claim 7: display unit.
• Claims 8–9: generator is an electromagnet, permanent magnet, or superconducting magnet; the electromagnet has a connection bar, two supporting arms, an accommodating space, and coils wound on each arm.
• Claims 10–12: image processing — frame differencing for object identification, gray-scale conversion for noise reduction, boundary processing with overlapped contour marks.
• Claims 13–15: light source module; for vertical motion, multiple laser light sources at different depths; velocity from known inter-laser spacing and the time difference between projection events.

---

Reported experimental result (specification)

Under 843 Gauss, magnetophorisis velocities across 100 objects distributed 25–29 µm/s (average 33 µm/s); equation (3) yielded ~28,825 particles per object, corresponding to ~0.68 ppm iron, close to the 0.659 ppm measured by ICP-MS. The patent frames this as the commercial advantage: comparable accuracy to ICP-MS without destroying the sample, and without the manual tracking required by Huang et al. (PLoS ONE, vol. 9, e96550, 2014).

---

Prior art of record

Cited references include US 5,974,901 (Cleveland Clinic, determining particle characteristics from velocity in a constant force field), US 2011/0236882 A1 (Academia Sinica, cell mass spectrometry particle endocytosis), JP 2002-022704 A (Osaka Univ., magnetophoretic concentration detection), WO 2012/060163 A1 (cell analyzer), JP 2009-536348 A (Philips, high-speed magnetic biosensor), and JP WO2016/063912 (Univ. of Tokyo, particle detection). Non-patent citations include the Huang PLoS ONE paper and Jin et al., "Quantitative intracellular magnetic nanoparticle uptake measured by live cell magnetophoresis" (FASEB J.).

---

Uncertainty flags

1. No litigation found. My searches produced no district-court or CAFC activity for 10,324,021, and no PTAB proceeding. This is a negative search result from limited tools, not a clearance opinion.
2. Claim 16 typographical gap ("third boundary line to the boundary line") is reproduced as-is per your literal-ID rule.
3. The Google Patents "adjusted expiration 2037-08-10" and "Active" status are database assumptions, not legal conclusions. Verify current status and any certificate-of-correction contents directly in USPTO Patent Center for application 15/587,081.

I'll search litigation-focused sources for the specific patent number.

Let me try litigation-specific databases and other query formulations.

Let me run a few final checks on PTAB and any court activity.

Litigation search result for US Patent 10,324,021

No litigation is known involving US Patent 10,324,021.

I searched for the specific number and found no district court case, no CAFC appeal, no ITC Section 337 investigation, and no PTAB proceeding (IPR/PGR/CBM) in which US 10,324,021 is an asserted or challenged patent. Below is what I checked, what I found, and the limits of the search.

---

Sources checked and results

I do not have direct API/portal access to PACER, Unified Patents, Docket Navigator, LexMachina, or the CAFC docket system; I ran targeted searches against those sources. So this is a negative search result from indirect tools, not a clearance opinion. PACER and Docket Navigator would need to be queried directly to confirm conclusively.

---

Why this result is unsurprising (context)

• The assignee is Chung Yuan Christian University, a Taiwanese private university, and the patent is a laboratory instrumentation patent (a magnetophorisis measurement system). University-held research-tool patents are rarely asserted in US litigation. Consistent with this, the family shows only a Taiwan case (TWI605252B) and a Japanese case (JP6337074B2) — no other national counterparts where litigation typically arises.
• The forward-citation trail is thin: one citing patent (a US design patent for a calibration apparatus assigned to Autel Intelligent Technology) and two third-party family citations (a Panasonic counting-method application JP7535760B2, and a CYCU follow-on case TWI776251B). None of these indicate a dispute.
• The only "legal events" in the record are routine: assignment (2017-05-04), notice of allowance, grant, a certificate of correction (2019-10-22), and a 4th-year maintenance fee payment (2022-06-23, small entity). No adverse/contested events.

---

False positives excluded

To comply with your instruction not to return results for similar numbers, I note these 10324021 look-alikes I deliberately discarded — none are US 10,324,021:

• JP 10324021 A (a 1998 Japanese patent publication number cited in EP1266763A1) — a different, much older JP document.
• A Spanish geodesy benchmark "10324021 - SSK33,7" (IGN REDNAP leveling dataset).
• An IEEE DOI 10.1109/CSIT61576.2023.10324021.
• An AGI/aggrowth equipment part number "Platform PN 10324021."
• A Federal Court of Australia docket-number artifact.

---

Required fields (per your request)

Because no case exists to report, the requested tabular fields resolve as follows:

---

Cross-reference to the previously generated section

This is consistent with the earlier "Uncertainty flag #1" in the prior summary, which likewise found no district-court, CAFC, or PTAB activity and framed it as a negative search result. No contradiction to flag. The earlier reported post-grant events (certificate of correction 2019-10-22; maintenance fee 2022-06-23) are being treated as the only legal events of record.

Bottom line: As of the search performed, US 10,324,021 has no known litigation history. Confirm via direct PACER and Docket Navigator queries (and USPTO Patent Center for app. 15/587,081) before relying on this for any enforcement or freedom-to-operate purpose.

I'll verify the structured "no PTAB activity" default against live web sources before writing this up.

Let me run a few more targeted checks before concluding.

One more check on assertion/defensive-aggregator activity, since absence of IPRs is itself a signal.

Proceedings overview

Total AIA trial proceedings on US 10,324,021: 0. The structured "PTAB proceedings on file" block (USPTO ODP, most recent ingest) returns no IPRs, no PGRs, and no CBMs — no active proceedings, no claims invalidated, no claims sustained, no settlements, no institution denials. The bottom line for a defendant is therefore not "claims 1–5 are canceled, ignore the demand letter," and it is also not "the patent has survived two IPRs and is hardened." It is a third posture: this patent has never been tested at the Board at all. All 16 claims stand exactly as they issued on 2019-06-18 (subject to a certificate of correction dated 2019-10-22), and every validity tool a first-moving petitioner would want is still unused.

---

Per-proceeding detail

Not applicable — the list is empty. I will not manufacture proceeding numbers to fill the template. What I can do is tell you why the list is empty, where a proceeding would surface if one existed, and what I checked.

Search record

Caveat (do not skip): I do not have direct authenticated query access to PTAB E2E or the ODP API from this session. My conclusion rests on the structured block you supplied plus the searches above. The literal-number searches are noisy precisely because 10324021 is not a distinctive string — so an absence of hits is weaker evidence than it would be for a distinctive patent number. Confirm independently at PTAB E2E, the PTAB Decisions page, and CourtListener before relying on this in a filing or an opinion letter.

Why the absence is expected here (structural, not accidental)

• PGR was available and has expired. This application was filed 2017-05-04 with a TW priority of 2016-11-16 — AIA first-inventor-to-file. That made PGR available for nine months from grant: 2019-06-18 to 2020-03-18. That window closed more than six years ago. No PGR will ever be filed on this patent.
• CBM was never a realistic vehicle. This is a measurement instrument (microscope + electromagnet + camera + processor), not a "covered business method" tied to financial products or services. It would have failed the technological-invention screen, and the CBM program has since sunset for patents on applications filed on or after 2020-09-16 in any event.
• IPR remains fully open. IPR is available throughout the patent's life. The only bars that matter are: no petitioner who filed a prior DJ action (§ 315(a)(1)), and no petitioner served with an infringement complaint more than one year earlier (§ 315(b)). Nothing in my searches suggests either trigger has occurred for anyone. And critically, no one has ever been to the Board on this patent, so § 315(e) estoppel has not attached to any party on any ground.

---

Strategic summary

Claim status: 100% UNTESTED, 0% canceled, 0% sustained

For contrast with your template language: a patent "hardened" by IPR has survived an adverse record. This one has no record. That cuts both ways — the claims are presumptively valid and fully enforceable, and the specification's experimental corroboration (≈28,825 particles/object → 0.68 ppm Fe vs. 0.659 ppm by ICP-MS, at 843 Gauss) reads as the kind of concrete, non-speculative disclosure that tends to survive a § 112 attack. But nothing has been adjudicated, so the claim scope is also un-elucidated — no FWD construction, no district-court construction, no prosecution disclaimer beyond the ordinary file wrapper.

Estoppel landscape

Because no petitioner has appeared, the § 315(e)(2) estoppel field is clean for a defendant. A first IPR petitioner can raise any §§ 102/103 ground based on patents and printed publications, without worrying that a prior petitioner or privy already consumed it. Conversely, a defendant who petitions and loses has the full weight of estoppel applied against it in the parallel litigation — that asymmetry is the single most important planning fact here.

The real constraint is not estoppel but § 325(d) discretion under Advanced Bionics. The prosecution record already contains a meaningful art stack, and the Board will ask whether a petition merely re-presents it:

A petition that recombines these seven will be fighting § 325(d). The more promising angle for a challenger is art not of record that teaches the claim's actual point of novelty — the frame-number × (1/frame rate) timing limitation recited in both independent claims. That is a computer-vision/particle-tracking limitation, not a magnetics limitation, and the natural prior art sits in the flow-cytometry and video particle-tracking literature (the patent itself is classified in G01N15/14, G06T7/246, G06T7/254) rather than in the magnetophoresis papers the examiner saw.

Note also the claim 16 typo: the third time-period clause reads "…from the third boundary line to the boundary line…" with "fourth" apparently omitted. Claim 16 is still presumptively valid as issued, but expect that defect to be a topic in any construction dispute and a possible basis for a § 112 indefiniteness argument if the fourth boundary line is not otherwise ascertainable from the claim and specification.

Pattern signals

• No repeat petitioner — trivially, because there is no petitioner at all.
• No PTAB appeal history — the patent owner has never had occasion to defend a claim at the Board on this patent, and therefore no CAFC appeal. (The owner does have an active patent-family practice — TW I605252B, JP 6337074 B2, JP 2018081069 A, TW 201819913 A, and a later TW I776251B on magnetic control for target-molecule adsorption — so the institution is clearly still working in this space.)
• No defensive aggregator in the chain. No Unified Patents, no IPR-on-file, no evidence of a coordinated validity campaign. Well-asserted patents in the biotech/instrument space routinely attract one within a few years of assertion; the absence of any suggests this patent has not been aggressively asserted in the US.

---

Recommended next steps

If you are a defendant or a prospective petitioner:

1. The absence of PTAB activity is your opening, not a reassurance. Verify the null result yourself at PTAB E2E and pull the file wrapper for application 15/587,081 from Patent Center — specifically the 2019-10-22 certificate of correction, whose contents I could not retrieve and which may have amended claim text. Challenge the corrected claims.
2. A first IPR is available to you with no § 315(b) bar and no estoppel exposure. IPR must be filed within one year of service of an infringement complaint — so if you have already been served, the clock is running and this is the controlling deadline. Ask me for a docket-based date check before you do anything else.
3. Do not build the petition on the art of record. US 5,974,901, JP 2002-022704, the Philips and Univ. of Tokyo references, and the Huang/Jin papers are all before the examiner. A $325(d)-driven denial is the most likely failure mode. Focus your search on video-based particle-tracking and frame-counting timing art that anticipates or renders obvious the frame number × (1/frame rate) limitation.
4. Do not expect PGR or CBM as fallbacks. Both are closed to you on this patent.
5. Budget for the § 315(e)(2) downside. If you petition and lose, you are estopped as to every ground raised or reasonably could have been raised — which, for a defendant also litigating in district court, is a broad bar. This is a one-shot weapon.

If you are the patent owner: the clean sheet is an asset. There is no adverse FWD to distinguish, no estoppel constraining you, and no § 325(d) burden on you. The exposure points are the claim 16 typo (consider a certificate-of-correction or reissue route if it matters), and the fact that your claim 1 novelty largely rests on a timing convention (frame count ÷ frame rate) that a challenger may find trivially disclosed in the particle-tracking literature.

Scope note: this memo covers US AIA trials only. I did not examine foreign post-grant challenges to the family members (TW I605252B, JP 6337074 B2), where invalidation actions are procedurally distinct. Say the word if you want those pulled.

I'll verify the assignment record against USPTO sources and check the asserter directories before writing this up.

Assignment & Ownership Analysis — US 10,324,021 B2

Sourcing note (read first): I could not issue authenticated API queries against Assignment Center / the ODP assignment endpoint in this session. The assignment data below comes from the Google Patents legal-events block for US 10,324,021, which mirrors the USPTO assignment record, plus targeted web searches. Where a field (notably the correspondent of record) is not exposed by that source, I say so rather than fill it in. Verify at Assignment Center / Patent Assignment Search using patent number 10324021 or application 15/587,081.

---

Inventors

Unusual-pattern check — nothing anomalous:

• All six inventors are listed as assignors on a single assignment to CYCU, executed 2016-12-21 and recorded 2017-05-04. This is the standard flow for a university-originated file: the inventors assign to the institution before/at the US filing.
• No inventor-departure signal. There is no evidence of any inventor assigning away a personal interest, retaining rights, or departing CYCU. A joint assignment to the institution by the entire inventive entity is the opposite pattern from a pre-fire-sale inventor exit.
• Timing is routine, not distressed. The TW priority filing is 2016-11-16; the assignment was executed ~5 weeks later (2016-12-21) and recorded on the US filing date 2017-05-04. That is the normal "assign before US filing" sequence, not a reactive transfer.
• Specific departmental affiliations (e.g., Ger's CYCU Biomedical Engineering affiliation) are plausible but I did not verify each inventor's department from a primary source in this session — treat the employer column as "CYCU, per the assignment," not as a per-inventor HR record.

---

Original assignee

Chung Yuan Christian University (CYCU) — the entity named on the issued patent and the sole assignee of record.

No SEC filings exist for this assignor — CYCU is a private Taiwanese university and files no 10-K/8-K. SEC cross-referencing is therefore not applicable here (there is no public-company assignor or assignee in the chain to check).

---

Assignment timeline

Total recorded assignments: 1 (the original inventor→university assignment). No post-issuance transfer of any kind is recorded.

---

• 2016-12-21 (executed) / recorded 2017-05-04 — Reel 042245 / Frame 0400
  • Conveyance: ASSIGNMENT OF ASSIGNORS' INTEREST (original assignment; the "Assignment" conveyance type)
  • Assignor: GER, TZONG-RONG; CHEN, WEI-YU; WANG, TING-RUEI; and others (the full inventive entity — six inventors per the face of the patent)
  • Assignee: CHUNG YUAN CHRISTIAN UNIVERSITY, Taiwan
  • Correspondent: Not exposed by the sources I could reach. The Google Patents legal-events block reports the reel/frame, conveyance text, assignors, assignee, and effective date, but not the recording correspondent. The attorney/firm of record for this recording is on the cover sheet of reel 042245/0400, which I could not retrieve this session. Recurrence flag: not assessable — with a single recorded link and no correspondent data, I cannot determine whether the same recording attorney appears on other CYCU family filings. Do not treat the absence of a correspondent as a finding of no correspondent.
  • Context: Initial acquisition by the employer institution — the routine inventor-to-university title transfer for a foreign-origin filing, executed five weeks after the TW priority filing and recorded on the US filing date. Not a fire-sale, not a reorg, not a securitization, not a transfer to an asserter.

No later assignments. The remaining legal events — notice of allowance (2019-02-21), grant (2019-05-29), certificate of correction (2019-10-22), and 4th-year maintenance fee (2022-06-23, small entity) — are prosecution and maintenance events, not title transfers. The certificate of correction is not a conveyance; it corrects text, not ownership.

Per your instruction: the Assignment Center record for this patent contains only the original assignment. That is itself the finding — it means the original assignee, CYCU, is still the owner of record.

---

Timeline diagram

timeline
    title Ownership of US 10324021
    2016 : Inventors execute assignment to CYCU
    2017 : US application filed 04 May
         : Assignment recorded 04 May
    2019 : Patent issued 18 Jun
         : Certificate of correction 22 Oct
    2022 : Fourth year maintenance fee paid

---

NPE / troll-pattern signals

Signal score: 0 present, 1 unclear (correspondent recurrence, for lack of data), 7 absent.

---

Verdict

Insufficient data — only the original assignment is on record

Justification: The complete recorded chain for US 10,324,021 consists of a single assignment — reel 042245/0400, executed 2016-12-21 and recorded 2017-05-04 — conveying the six inventors' interest to Chung Yuan Christian University. Under your rubric, a patent with only the original assignment falls squarely into Insufficient data, because there is no later link to characterize and no second correspondent to test for recurrence. Substantively, however, this is an unusually clean ownership record: the patent was never sold, never securitized, never transferred to a licensing-only entity, and never placed with a defensive aggregator; it is still held by the operating academic institution that generated it, with maintenance fees paid through the 4th year (2022-06-23). No NPE pattern is present on the evidence — but the absence of a finding here reflects the thinness of the record (one link, no correspondent data), not a clean bill of health verified against a full Assignment Center abstract.

Two caveats before you rely on this:

1. The correspondent of record for reel 042245/0400 is unknown to me. That is the single most valuable datum your brief asks for, and it is the one I could not extract. Pull the reel cover sheet directly — a US associate firm handling CYCU's inbound filings would appear there, and you can then check that firm against the CYCU family (TW I605252B, JP 6337074 B2, and the later TW I776251B) to test signal #3 properly.
2. The 2019-10-22 certificate of correction is recorded as a post-grant event. I retrieved no evidence that it touches ownership — certificates of correction amend text, not title — but its contents were not available to me, and the earlier prosecution section flagged it as worth pulling.

Search page for verification: https://assignmentcenter.uspto.gov/ · https://assignment.uspto.gov/patent/index.html — enter patent 10324021 or application 15/587,081.

I'll research the cited prior art references for US 10,324,021. Let me run targeted searches.

Let me research the remaining cited references and verify the USPTO record.

Let me research the remaining references and verify the USPTO record.

Prior-Art Analysis — US 10,324,021 B2

1. USPTO record check for patent number 10324021

Result: exact match found — US 10,324,021 B2. No look-alike was substituted.

Caveat: I could not issue a live query against USPTO Patent Center/PAIR directly; the record above is the authoritative full text you supplied (which mirrors the USPTO bibliographic record). All dates and identifiers are quoted literally, including the patent's own spelling "magnetophorisis."

Because the priority date is 2016-11-16, every reference below (all predating 2016) qualifies as prior art under at least 35 U.S.C. § 102(a)(1)/(a)(2)/(b).

---

2. Framing: what has to be anticipated

Under § 102, anticipation requires every element of a claim in a single reference. The two independent claims are dominated by three unusual limitations:

• (A) Magnetic field generator that changes field direction on a control signal so the objects reciprocate during the measurement (both claims).
• (B) The processing unit automatically locks the moved object and identifies it from ≥2 frames (both claims).
• (C) Velocity/acceleration computed across virtual boundary lines with time = frame number × (1 / frame rate) — two velocities across three lines (claim 1), or three velocities/four lines and two accelerations (claim 16).

Bottom line up front: None of the 11 cited references discloses all of (A)+(B)+(C). No cited reference fully anticipates claim 1 or claim 16. Their real force is aggregate § 103 material, and several of them are § 102-relevant to the broader dependent claims. That assessment is spelled out per reference below. (Confidence note: I worked from the reference abstracts/claims and the patent's own characterizations; I did not obtain full claim charts for every reference, so the § 102 calls below are ranked by confidence.)

---

3. The 11 patent citations

Tier 1 — most relevant

① US 5,974,901 A

• Full citation: Method for determining particle characteristics, The Cleveland Clinic Foundation. Filed 1998-02-06; granted 1999-11-02. (Companion system patent: US 6,412,359 B1.)
• Description: Generates a region of space with a substantially constant force field (claim 17 lists magnetic, electric, electromagnetic, flow); determines the velocity of a particle by identifying and locating it and its coordinates in at least two temporally defined digital images; and derives particle physical characteristics (size, shape, magnetic susceptibility, magnetic label density, dielectric constant, etc.) from the velocity plus the known force-field magnitude and direction. Particles expressly include cells, organelles, platelets, and inorganic/organic/biological/polymeric optically-visible particles.
• § 102 potential: This is the closest structural reference for the "image-velocity → particle characteristic" spine. It could plausibly be urged against claim 3 (algorithmic quantification from motion status) and against generic imaging/velocity features. However, it does not disclose boundary-line zone timing, frame-number×frame-rate computation, automated object locking, or reversible/reciprocating field — so it does not anticipate claims 1 or 16. Treat as high-value § 103 art and the lead § 102 reference for the concept, not a clean anticipation. Confidence: high on content, high that it is not a § 102 hit on 1/16.

② JP 2002-022704 A (Osaka University)

• Full citation: Magnetophoretic concentration detection method and apparatus. Priority/filing 2000-07-04; published 2002-01-23.
• Description: Magnetophoretic concentration detection — measuring particle concentration via magnetophoresis, i.e., the same physical phenomenon (particle motion driven by a magnetic field gradient) that the '021 patent exploits.
• § 102 potential: Same technical field (magnetophoresis-based measurement). Relevance is against the preamble/purpose of claims 1/16 and against claim 3 (quantifying via magnetophoretic behavior), but it does not disclose the claimed imaging, boundary-line, or reciprocation features. Not an anticipation of 1/16. Confidence: medium (I did not retrieve the full JP disclosure).

③ US 2011/0236882 A1 (Academia Sinica) — a cited reference AND discussed in the '021 background

• Full citation: Quantitative measurement of nano/micro particle endocytosis with cell mass spectrometry. Priority 2010-03-29; published 2011-09-29. (Granted as US 9,459,247 B2; TW counterpart TWI540319B.)
• Description: Cell mass spectrometry (CMS/CMCMS) to determine the number of nanoparticles taken up per individual cell by measuring m/z shift → absolute cell mass difference.
• § 102 potential: It is prior art for the problem ("quantify nanoparticles engulfed by a cell") and the statement that CMS "determines the number of NPs taken up by each individual cell whereas ICP only gets a mean" — useful to show the quantification goal was known. But it is a mass-spectrometry method, categorically different from image-based magnetophoresis. Does not anticipate any claim of '021. Relevant only as background/§ 103 context. Confidence: high.

④ Huang, Chen-Yu et al., PLoS ONE (Non-patent citation, discussed in '021 background)

• Full citation: C.-Y. Huang et al., "Compare analysis for the nanotoxicity effects of different amounts of endocytic iron oxide nanoparticles at single cell level," PLoS ONE, vol. 9, issue 5, e96550, May 2014.
• Description: Magnetic-force-attracted cells moving at constant speed recorded by CCD camera; manual tracking of displacement via an ImageJ plugin; velocity → quantity of magnetic nanoparticles.
• § 102 potential: This is arguably the most technically on-point single reference — it performs the same "image cells, measure their magnetophoretic velocity, back out magnetic-nanoparticle content" method. It supports § 102/§ 103 against the broad concept behind claim 3. Critical distinction: Huang uses manual marking, whereas '021 claims automatic locking + boundary-line/frame-number timing — precisely the limitation the patent says distinguishes it. So Huang does not anticipate claims 1/16 (and may be argued as teaching away from nothing / merely being improved upon). Confidence: high on content.

⑤ Jin, Ying et al., The FASEB Journal (Non-patent citation)

• Full citation: Y. Jin et al., "Quantitative intracellular magnetic nanoparticle uptake measured by live cell magnetophoresis," The FASEB Journal (Research Communication).
• Description: Live-cell magnetophoresis to quantitatively measure intracellular magnetic-nanoparticle uptake — i.e., using magnetophoretic velocity to infer nanoparticle load, the same measurement principle as the '021 patent.
• § 102 potential: Very close in substance to the patent's core method; strong § 103 art and potential § 102 material against the quantification concept (claim 3). No disclosure of the claimed boundary-line/frame-number processing or reciprocation. Not a clean anticipation of 1/16. Confidence: medium (working from the title/abstract as cited; full text not retrieved this session).

Tier 2 — field-adjacent (§ 103 value, no anticipation)

⑥ JP 2009-536348 A (Koninklijke Philips Electronics N.V.)

• Full citation: High-speed magnetic biosensor (JP national phase of WO 2007/129275 A2, EP 06113763.4). Priority 2006-05-10; published 2009-10-08.
• Description: Applies a magnetic field gradient parallel to the sensor surface to carry magnetic particles laterally; detects/quantifies bound particles. Uses switchable magnetic field gradients (magnets on both sides of the cartridge) — i.e., reversing field direction to move particles back and forth is contemplated in this art.
• § 102 potential: Interesting for the reciprocation feature (A) because it applies lateral field gradients and switches magnetic configuration. But it is a biosensor binding assay, not image-tracked motion-status determination, and lacks the boundary-line/auto-lock features. Does not anticipate 1/16. Confidence: medium.

⑦ WO 2012/060163 A1 (Kanagawa Academy of Science and Technology et al.)

• Full citation: Cell analyzer (US counterpart US 2013/0029407 A1; granted as US 9,023,294 B2 to Terazono et al.). Priority 2010-11-01; published 2012-05-10.
• Description: Microchannel cell concentration/purification with high-speed image acquisition (~10,000 images/sec) of cells flowing through a micro-flow path, real-time image processing and single-cell recognition/analysis, then separation.
• § 102 potential: Antecedent art for image-based, high-frame-rate, automated single-cell tracking/recognition in a microchannel — supports § 103 against the "video image + automatic object identification" elements of claims 1/16 and claim 10 (frame differencing). It does not disclose magnetophoretic quantification or the boundary-line velocity math. Confidence: medium-high.

⑧ US 2013/0054142 A1 (NEC Corporation)

• Full citation: Flow line detection system, flow line detection method, and flow line detection program. Priority 2010-05-31; published 2013-02-28.
• Description: Image-based flow-line (trajectory) detection and tracking using image processing.
• § 102 potential: Relevant background for automated trajectory/motion tracking from images; supports § 103 against the object-locking/tracking features. No magnetophoresis, no boundary-line velocity computation. Confidence: medium.

⑨ JP WO 2016/063912 A1 (National University Corporation, University of Tokyo)

• Full citation: Particle detection method, particle detection apparatus, and particle detection system. Priority 2014-10-24; published 2017-08-03 (JP national-phase entry of WO 2016/063912).
• Description: Particle detection method/apparatus/system (optically detecting particles in a flow).
• § 102 potential: The most temporally recent cited patent, so closest in time to the 2016 priority, but its subject matter is particle detection generally. It does not disclose the magnetophoretic reciprocation + boundary-line velocity determination. Confidence: low-medium (full text not retrieved; assess on record).

Tier 3 — peripherally cited (no anticipation; possibly not even material)

⑩ US 2008/0023641 A1 (Hitachi High-Technologies Corp.)

• Full citation: Focused Ion Beam Apparatus. Priority 2006-07-27; published 2008-01-31.
• Description: A focused-ion-beam instrument. No apparent relationship to magnetophoresis or image-based particle-velocity measurement.
• § 102 potential: None. Appears to be an unrelated citation (possibly a classification/art-unit artifact). Confidence: high.

⑪ US 2008/0186551 A1 (Carl Zeiss Meditec AG)

• Full citation: Scanning Device. Priority 2005-03-26; published 2008-08-07.
• Description: Optical scanning device.
• § 102 potential: At most generic optical-scanning/illumination context; no anticipation of any claim. Confidence: high.

⑫ CN 100388004 C (中南大学 / Central South University)

• Full citation: A method for identifying bacterial magnetotaxis. Priority 2006-09-08; granted 2008-05-14.
• Description: Method for identifying bacterial magnetotaxis (magnetotactic behavior of bacteria in a magnetic field).
• § 102 potential: Relevant to the "object = bacterial strain" dependent claim 6 and to the concept of magnetically driven microbe motion; too narrow to anticipate 1/16. Worth examining if claim 6 is asserted. Confidence: low-medium (Chinese-language record, not fully retrieved).

⑬ US 2006/0252054 A1 (Ping Lin)

• Full citation: Methods and compositions for detecting non-hematopoietic cells from a blood sample. Priority 2001-10-11; published 2006-11-09.
• Description: Detection of non-hematopoietic (e.g., disseminated tumor) cells in blood, with immunomagnetic/immunofluorescent labeling and enrichment.
• § 102 potential: Background only — magnetic labeling of cells. No anticipation. Confidence: high.

---

4. Summary matrix

---

5. Conclusions and confidence

1. No cited reference anticipates independent claim 1 or claim 16 under § 102. The three-element combination (reversible/reciprocating magnetic field + automated object locking + boundary-line/frame-number×frame-rate velocity-or-acceleration determination) appears in none of the references individually. The nearest miss is US 5,974,901 (image-velocity → particle traits) combined with the magnetophoresis NPL (Huang / Jin), but combining references is a § 103 exercise.

2. Most § 102-relevant (closest) references: US 5,974,901 (structure/algorithm) and, on the method substance, Huang et al. (PLoS ONE 2014) and Jin et al. (FASEB J.). These target claim 3 (the quantification algorithm) and the general "velocity → magnetic-particle count" concept.

3. Weak/unrelated citations (US 2008/0023641 and US 2008/0186551) provide no § 102 or meaningful § 103 support and should be discounted.

4. § 102 labels are provisional. I retrieved abstracts, claims, and search snippets — not full claim charts for every reference, and two records (JP 2002-022704; JP WO 2016/063912; CN 100388004) are Japanese/Chinese-language documents I could not fully read this session. Before relying on any § 102 conclusion for enforcement or FTO, pull the full texts and build element-by-element charts.

5. Cross-reference / no contradiction: This is consistent with the earlier sections of this analysis (bibliographic data; the "no litigation found" result). One consistency note worth flagging: the prior summary listed US 2011/0236882 A1 (Academia Sinica) among "prior art of record" — correct, and I confirm it was both cited on the face of '021 and discussed in the '021 background. No contradiction to flag.

Obviousness Analysis — US 10,324,021 B2 under 35 U.S.C. § 103

I'll ground this in the prior-art references of record shown on the patent page. Let me pull details on the key references before mapping claim elements.

Obviousness Analysis — US 10,324,021 B2 under 35 U.S.C. § 103

Analysis date: April 26, 2026 (per task). Critical date: priority 2016‑11‑16 (TW 105137453); U.S. filing 2017‑05‑04. All references below predate the critical date.

Scope note / caveat up front: This analysis works from the prior art of record on the patent page — the 11 "Citations," the 2 "Family Cites Families," and the 4 "Non‑Patent Citations." I do not have the USPTO file wrapper for application 15/587,081, so I cannot report the examiner's actual § 103 rejections or the applicant's remarks. What follows is an independent, element‑by‑element obviousness assessment using only the record. I flag the one limitation where the record is genuinely thin rather than papering over it.

---

1. Legal framework

Under § 103 (as construed in Graham v. John Deere and KSR Int'l v. Teleflex), a claim is obvious if the differences between the claimed subject matter and the prior art are such that the subject matter as a whole would have been obvious to a person of ordinary skill ("POSITA") at the critical date. The inquiry requires: (1) the scope and content of the prior art; (2) the differences between the prior art and the claims; (3) the level of ordinary skill; and (4) objective indicia. KSR permits reliance on any articulated rationale — e.g., combining references according to known methods to yield predictable results, a simple substitution of one known element for another, or a "design incentive" arising from the problem to be solved. The POSITA here is a person with a graduate‑level background in biomedical engineering / analytical instrumentation and routine skill in video image processing (a combination of magnetics, optics, and computer vision), consistent with the mixed G01N/G06T classification of the patent.

---

2. What the record actually teaches

---

3. Claim 1 — element‑by‑element mapping

Claim 1 is a system claim. Broken into limitations:

Difference analysis: Every element of claim 1 other than 1c (reciprocating/reversed field) is disclosed or rendered obvious by U.S. 5,974,901 in view of Huang 2014 and the automated‑tracking art (NEC / WO 2012/060163). The "boundary line + frame‑number ÷ frame‑rate" timing of 1g is, at most, an obvious design choice / optimization — a POSITA quantifying speed from a fixed‑frame‑rate video would naturally time crossings between fixed reference lines in the field of view, because that is the simplest way to implement the velocity measurement U.S. 5,974,901 already requires.

---

4. Claim 16 — element‑by‑element mapping

Claim 16 repeats 1a–1e (including the same reciprocating‑field limitation 1c) and differs only in the analysis: it computes two accelerations from three velocities measured across four boundary lines, each velocity timed by frame number ÷ frame rate.

• Three velocities across four reference lines: straightforward extension of the claim‑1 boundary construction — adding one more boundary line/zone is a matter of degree, not kind.
• Acceleration from successive velocities: elementary kinematics (a = Δv/Δt), well within the ordinary skill of the POSITA; neither a new result nor an unpredictability. U.S. 5,974,901 already frames its method around characterizing particle motion in a force field from image‑derived coordinates, and a POSITA distinguishing constant‑velocity from accelerating magnetophoretic motion (a distinction the patent itself acknowledges is needed because objects "closing to the magnetic field generator" behave differently) would compare successive‑interval velocities as a matter of routine.

Note on literal text: Claim 16's final clause recites "…moved from the third boundary line to the boundary line…" — the word "fourth" appears to be missing. Per the literal‑identification rule, this is reproduced as‑is; it does not change the obviousness analysis, since the surrounding text ("distance between the third and fourth boundary lines") supplies the fourth line.

---

5. Dependent claims

---

6. The pivotal limitation 1c — reciprocating magnetic field

Claim 1 (and claim 16) both require the field generator to change the magnetic direction on a control signal so the objects perform a reciprocating motion during the measurement.

• What the record does show: the patent's own background describes causes of inaccuracy ("when the objects move closing to the boundary of flow channel or closing to the magnetic field generator, the accuracy of quantifying the magnetic particles may be affected"). Reversing the field to obtain multiple passes and averaging is a known error‑reduction strategy (repeated measurements + statistical averaging). U.S. 5,974,901 explicitly contemplates applying its force field "for [a] predetermined time period," and general magnetic‑particle‑manipulation art routinely switches/alternates drive direction (e.g., alternating DC drive for magnetophoresis in EP 1 916 032). So a POSITA seeking better statistics on the same magnetophoretic measurement had a motivation to run the objects through the field in both directions.
• What the record does not cleanly show: none of the cited references (U.S. 5,974,901, US 2011/0236882, JP 2002‑022704 as cited, WO 2012/060163, JP 2009‑536348, JP WO2016/063912, US 2013/0054142, or the two family‑cited references) is described as reversing the external field within a single measurement to drive a reciprocating motion.

Assessment: Limitation 1c is the one place a patentee can mount a genuine non‑obviousness argument. But the argument is not strong, because the rationale is a classic "known technique to improve similar devices in the same way" / "predictable use of a known field‑control capability to reduce measurement error." Switching the polarity of an electromagnet is trivial; the reason to do so (repeat/average to cancel the boundary and field‑gradient errors the patent itself identifies) follows directly from the recognized problem. Where there is a design incentive and predictable results, KSR supports obviousness. I would rate 1c as obvious but arguable, not a clean win for either side.

---

7. The three strongest § 103 combinations, with motivations

Combination A — Huang 2014 + U.S. 5,974,901 + NEC US 2013/0054142 + Jin 2008 (likely the primary rejection)

• Huang establishes the field, the problem, the magnetophoresis apparatus, video capture, per‑cell velocity→nanoparticle‑count workflow — but is manual, hence slow and impractical at scale (the exact deficiency the '021 spec recites).
• U.S. 5,974,901 supplies the missing automation backbone: velocity from ≥2 temporally defined digital images, background subtraction to isolate the particle, tracking, and derivation of a magnetic particle characteristic from velocity + known force field — i.e., limitations 1d, 1f, 1g and the substance of claims 3/10/11.
• NEC US 2013/0054142 supplies automated multi‑object locking/tracking across video frames, i.e., the "automatically locks at least one moved object" of 1e.
• Jin 2008 supplies the quantification algorithm (mobility/velocity → intracellular magnetic‑particle mass) of claim 3.

Motivation to combine: All four are in the same technical endeavor (measuring/characterizing magnetic‑particle‑laden cells from their motion in a magnetic field, using imaging). The problem Huang itself leaves open is that manual tracking "will become more difficult and complicated … time‑consuming" with many objects — a classic design incentive for automating the very velocity measurement U.S. 5,974,901 already performs automatically, using the tracking methods NEC teaches for multiple objects. Reasonable expectation of success: high; each step is individually known and the combination is the predictable union of their functions.

Combination B — U.S. 5,974,901 + WO 2012/060163 + Jin 2008 + JP 2002‑022704

• U.S. 5,974,901 (image velocity ⇒ magnetic particle characteristic; transparent flow channel; electromagnet/pole geometry), WO 2012/060163 (micro flow channel + ≥200 fps camera + image‑based single‑object recognition — supplies the one‑dimensional flow‑channel and high‑speed video elements and claims 5/13), Jin 2008 (algorithm), and JP 2002‑022704 (magnetophoretic velocity/concentration detection as a characterization method, with controllable driving force → claim 4).
• Motivation: combining a known magnetophoretic characterization method with a known high‑speed imaging flow‑channel analyzer to measure many particles in parallel is the straightforward application of two known techniques to achieve predictable results.

Combination C — Combination A/B plus the reciprocating‑field teaching (eventual § 103 rationale for 1c)

• Because none of the cited references states "reverse the field to reciprocate," the strongest available rationale for 1c is not a single reference but the explicit problem statement in the '021 background itself (accuracy degrades near flow‑channel boundaries and near the field generator) + the general, well‑known capability to switch electromagnet polarity (U.S. 5,974,901's controllable force field; alternating‑drive magnetophoresis in the art). Under KSR, using a known control capability (field reversal) to obtain repeated samples and average away a recognized error is an obvious, predictable improvement. Expect this to be argued rather than cleanly proven.

---

8. Counter‑considerations (why some claims might survive)

1. Limitation 1c is not squarely anticipated in the record. If the patentee can show the record lacks any teaching of reversing the field within a measurement to produce reciprocation, they have a non‑obviousness foothold for claims 1 and 16 — subject to the KSR "known technique / design incentive" rebuttal in § 7C.
2. Claim 9's specific electromagnet frame (connection bar + two supporting arms + accommodating space + coils) and claims 14–15's depth‑staged laser sources are the least well‑supported by the cited references; those dependents are the most defensible if the independent claims fall.
3. Objective indicia (secondary considerations). The patent asserts a concrete technical result — ~28,825 particles/object ⇒ 0.68 ppm iron vs. 0.659 ppm by ICP‑MS — and a practical advantage (non‑destructive, reusable sample, no manual marking). If the patentee can show a nexus between that result and the claimed reciprocating‑field/automated‑locking features, unexpected‑results and long‑felt‑need arguments become available. The best‑supported nexus runs to the automation ("the steps of manual procedure for marking the moved object is not necessary"), directly targeting the Huang deficiency — but automation is exactly what NEC/U.S. 5,974,901 supply, which weakens the nexus.
4. Foreign counterpart grants. The record shows the family was granted in TW (I605252B) and JP (JP 6337074 B2); the page also lists Japanese and Taiwanese prosecution actions (JP official action Oct. 24, 2016; TIPO action Jun. 22, 2017). Foreign allowance is not evidence of U.S. patentability (different art, different standards), but it is weak circumstantial evidence that at least some offices did not find a clean § 103 killer.

---

9. Bottom line

• Claims 2–8, 10–13 and most of claim 9 appear highly likely obvious over U.S. 5,974,901 in view of Huang 2014, Jin 2008, and NEC US 2013/0054142 (and/or WO 2012/060163) — each element is present in a reference in the same field, and the motivations are strong and well‑articulated.
• Claims 1 and 16 rise or fall on the reciprocating‑field limitation (1c) and, secondarily, on the boundary‑line/frame‑rate velocity/acceleration construct. I assess 1c as obvious‑but‑arguable under KSR; it is the single most consequential point in any § 103 contest for this patent.
• Claims 14–15 (depth‑staged laser sources) are the weakest subject matter for an obviousness challenge on the present record.

Verification needed: the authoritative determination depends on the actual USPTO file wrapper for application 15/587,081 — the specific art the examiner applied and the applicant's distinguishing remarks. I do not have that document; the above is an independent analysis of the references of record only, not a legal opinion. Confirm all reference contents and the prosecution history in USPTO Patent Center before relying on this for any invalidity or drafting purpose.