AI Segmentation

K211881

Varian Medical Systems, Inc. · cleared 2021-09-02 · product code MUJ · Radiology

Premarket evidence — what FDA accepted

Device typesamd
source quote (p.5)
Both devices are software-only medical devices.
Algorithmcombined deep learning and classical-based approach
source quote (p.4)
AI Segmentation is a web-based application, running in the cloud, that provides a combined deep learning and classical-based approach for automated segmentation of organs at risk, along with tools for structure visualization.
Adaptive (vs locked)No
source quote (p.5)
Note: These algorithms are static and non-adaptive; they do not alter their behavior over time based on user input.
PCCPFDA source did not state this
Cybersecurity addressedFDA source did not state this

Validation studies (1)

Bench

sample size not stated

endpoints: DICE similarity index; qualitative scoring system was used to measure the acceptability of auto-generated contours, with a target of 80% of expert scores designating the contours as “acceptable with minor or no adjustments"

standards: IEC 62304 Edition 1.1 2015-06 Medical device software - Software life cycle processes, IEC 62366-1 Edition 1.0 2015-02 Application of usability engineering to medical devices, IEC 62083 Edition 2.0 2009-09 Requirements for the safety of radiotherapy treatment planning systems, IEC 82304-1 Edition 1.0 2016-10 Health software - Part 1: General requirements for product safety

Reported performance (1 observation)

diceas written: “DICE similarity indexstated without value
source quote (p.5)
Each AI model was assessed using the DICE similarity index as a comparative measure of the auto-generated contours against ground truth contours for a given structure. Aggregated DICE scores for each AI model were then compared to literature values or against the performance of the prior model when evaluating an update to an existing algorithm.

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

35
recalls in product code, 24mo
17
MAUDE reports in code, 12mo
-50%
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) K230082 (decision 2023-05-04) from GE Medical Systems, LLC for a matching device line ("Auto Segmentation") — a new clearance for the same line is a change event.

    first seen 2026-07-08 · k_number:K230082

  • recall_reason_pattern

    Software/algorithm-related recall in product code MUJ (Philips Medical Systems (Cleveland) Inc, initiated 2025-08-05): "Due to a software issue, there is a potential image error of the Region of Interest for expansion/contraction for HFP (Head First Prone), FFS (Feet First Supine) and FFP (Feet Firs" Recalling firm is another firm in the same product code.

    first seen 2026-07-08 · recall res_event_number:97049

  • recall_reason_pattern

    Software/algorithm-related recall in product code MUJ (Philips Medical Systems (Cleveland) Inc, initiated 2025-07-17): "Due to software issue, Radiation Therapy Planning system may provide incorrect dataset calculations when performing the "Stopping Power Ratio" (SPR) ," Recalling firm is another firm in the same product code.

    first seen 2026-07-08 · recall res_event_number:97309

Recall and MAUDE counts are product-code-level (reports aren't reliably attributable to one device). 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).

RIGOR™ Precedent · public FDA/CMS data · descriptive decision-support, not regulatory or reimbursement advice. Share this page: radar.healthai.com/precedent/device/K211881