HeartFlow Analysis

K250902

HeartFlow, Inc. · cleared 2025-07-18 · product code PJA · Cardiovascular

Premarket evidence — what FDA accepted

Device typesamd
source quote (p.6)
“The Heartflow Analysis is an AI-based medical device software developed for the clinical quantitative and qualitative analysis of CT DICOM data.”
Algorithmdeep learning (AI and machine learning)
source quote (p.6)
“The core technology remains unchanged from the primary predicate and continues to be trained using deep learning (AI and machine learning) since 2015, to incorporate learnings from the volumes of CT data and studies.”
Adaptive (vs locked)No
source quote (p.6)
“All algorithms are then frozen and validated prior to product release.”
PCCPYes
source quote (p.1)
“FDA's substantial equivalence determination also included the review and clearance of your Predetermined Change Control Plan (PCCP).”
Cybersecurity addressedYes
source quote (p.12)
“Heartflow has instituted cybersecurity controls to maintain the safety and security of the Heartflow Analysis. Study transmission is accomplished using Heartflow Connect or a third-party DICOM transfer device. This enables secure and reliable transmission of CT data. Heartflow assesses real and perceived cybersecurity vulnerabilities and uses specific software testing tools to ensure that the device remains safe and effective. Software de-bugging occurs on a nearly monthly basis to ensure that issues are resolved quickly, and issues are triaged in accordance with the level of risk associated with them.”

Validation studies (1)

Retrospective clinical

n=100 patients · 67 site(s)

endpoints: sensitivity; precision; mean volume error difference

Reported performance (2 observations)

sensitivityas written: “plaque localization sensitivity superiority (compared to predicate)”0.151
source quote (p.12)
“Heartflow Analysis (subject) plaque algorithm's plaque localization sensitivity superiority compared to Heartflow Analysis (predicate) plaque algorithm at point-wise level sensitivity was 0.151 (Goal > 0, p<0.0001).”
diceas written: “point-wise level DICE similarity coefficient”0.8
source quote (p.12)
“Heartflow Analysis (subject) plaque algorithm's plaque localization performance measured with point-wise level DICE similarity coefficient was 0.8 (Goal > 0.7, p<0.0001)”

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

0
recalls in product code, 24mo
14
MAUDE reports in code, 12mo
+35%
vs code's own 3-yr baseline
0
drift signals on this device

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 Cardiovascular 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/K250902