Athelas Home

K200828

Athelas Inc. · cleared 2022-03-22 · product code GKZ · Hematology

Premarket evidence — what FDA accepted

Device typehardware with ml
source quote (p.5)
“Internet connected device for processing results on cloud server.A microfluidic test strip channel creates a stained monolayer of white blood cells. Multiple images are taken of the monolayer and the cells are counted and classified by computer vision based image analysis.”
Algorithmcomputer vision based image analysis
source quote (p.6)
“Multiple images are taken of the monolayer and the cells are counted and classified by computer vision based image analysis.”
Adaptive (vs locked)FDA source did not state this
PCCPNo
Cybersecurity addressedFDA source did not state this

Validation studies (24)

Bench

n=9 cases

Bench

n=720 other · 3 site(s)

Standalone

n=24 patients

Standalone

n=10 patients

Bench

n=10 cases

standards: CLSI EP6-A

Bench

sample size not stated

endpoints: Interference with measurements

Bench

n=120 other

endpoints: Verification of reference intervals

Bench

n=120 other

endpoints: Limit of Blank

Bench

n=5 cases

endpoints: Limit of Detection

Bench

n=4 cases

endpoints: Limit of Quantification

standards: CLSI EP17-A2

Bench

n=9 cases

endpoints: Whole blood stability duration

standards: CLSI EP25-A

Bench

sample size not stated

endpoints: Shelf life stability

standards: CLSI EP25-A

Retrospective clinical

n=312 patients · 3 site(s)

endpoints: slope; r; bias at medical decision levels; overall bias; bland altman analysis; intercept

Retrospective clinical

n=312 cases

endpoints: Flagging accuracy; Positive Agreement (Sensitivity); Negative Agreement (Specificity); Overall Agreement

Retrospective clinical

n=59 patients

endpoints: Comparability of venous and capillary whole blood

standards: CLSI EP5-A3, H26-A2

Retrospective clinical

n=165 cases

endpoints: slope; r; bias at medical decision levels; overall bias; bland altman analysis; intercept

Standalone

sample size not stated

endpoints: Usability; error rates

Standalone

n=32 patients

endpoints: User ability to set up and run test

Bench

sample size not stated

endpoints: Error codes; accuracy

Bench

sample size not stated

endpoints: Robustness against tilted surfaces; accuracy; error codes

Bench

n=12 other

endpoints: Return accurate results after being dropped; functional tests pass

standards: ASTM 1469 - D03

Bench

n=9 cases

endpoints: Performance after exposure to extreme temperatures

Bench

sample size not stated

endpoints: Function after exposure to vibration; error codes

Bench

n=8 other

endpoints: Performance under extreme humidity conditions

Reported performance (2 observations)

sensitivity90.91
source quote (p.18)
“% Positive Agreement (Sensitivity) = 100 * 90/(90 + 9) = 90.91%”
specificity96.71
source quote (p.18)
“% Negative Agreement (Specificity) = 100 * 206/(206 + 7) = 96.71%”

Each value carries its own analysis unit and task — never compare or pool across devices. Source: 510(k) summary PDF.

Predicate network

Postmarket — what happened after clearance

14
recalls in product code, 24mo
61
MAUDE reports in code, 12mo
-55%
vs code's own 3-yr baseline
3
drift signals on this device
  • re_clearance

    The FDA AI/ML device list shows a newer 510(k) K243348 (decision 2026-02-06) from Athelas, Inc. for a matching device line ("Athelas Home") — a new clearance for the same line is a change event.

    first seen 2026-07-08 · k_number:K243348

  • recall_reason_pattern

    Software/algorithm-related recall in product code GKZ (Abbott Laboratories, initiated 2025-09-18): "Software issue for hq analyzer results in system not visibly applying appropriate flagging to results when saturation is present which may result in incorrect results." Recalling firm is another firm in the same product code.

    first seen 2026-07-08 · recall res_event_number:97788

  • recall_reason_pattern

    Software/algorithm-related recall in product code GKZ (Abbott Laboratories, initiated 2025-03-18): "When expired reagents are scanned or manually entered, the system will change the expiration date to current or future date without notifying user that an expired reagent is being " Recalling firm is another firm in the same product code.

    first seen 2026-07-08 · recall res_event_number:96567

Recall and MAUDE counts are product-code-level (reports aren't reliably attributable to one device); a recall is shown as device-attributed only when the recall record itself lists this clearance number. Signals are descriptive observables with sources — never a judgment that the device is unsafe or drifting. Snapshot 2026-07-08.

Reimbursement — how devices like this got paid

Not yet tracked — no payment pathway indexed for this clearance (the reimbursement corpus is a growing seed set).

Applicable FDA guidance — what the submission is measured against

FDA guidance documents and guiding principles applicable to 510(k) AI/ML devices in the Hematology panel. A curated reference index, not legal or regulatory advice — each item states its own status, and a draft is never binding.

Applicability is derived from the device's FDA advisory panel and pathway — cross-cutting guidances apply to every AI/ML device; panel-specific ones are flagged. Titles, dates, and links verified against fda.gov as of July 2026.

Constat Precedent · public FDA/CMS data · descriptive decision-support, not regulatory or reimbursement advice. Share this page: constat.dev/precedent/device/K200828