AutoContour Model RADAC V3

K230685

Radformation, Inc. · cleared 2023-04-14 · product code QKB · Radiology

Premarket evidence — what FDA accepted

Device typesamd

source quote (p.16)

“AutoContour is a pure software device and is not supplied sterile because the device doesn't come in contact with the patient. AutoContour is a pure software device and does not have a Shelf Life.”

Algorithmmachine learning based contouring. The deep-learning based structure models are trained using imaging datasets consisting of anatomical organs of the head and neck, thorax, abdomen and pelvis for adult male and female patients

source quote (p.7)

“AutoContour Model RADAC V3 device is software that uses DICOM-compliant image data (CT or MR) as input to: (1) automatically contour various structures of interest for radiation therapy treatment planning using machine learning based contouring. The deep-learning based structure models are trained using imaging datasets consisting of anatomical organs of the head and neck, thorax, abdomen and pelvis for adult male and female patients, (2) allow the user to review and modify the resulting contours, and (3) generate DICOM-compliant structure set data the can be imported into a radiation therapy treatment planning system.”

Adaptive (vs locked)FDA source did not state this

PCCPFDA source did not state this

Cybersecurity addressedFDA source did not state this

Validation studies (4)

Retrospective clinical

n=50 images

endpoints: Mean Dice Similarity Coefficient (DSC) was used to validate the accuracy of structure model outputs; qualitative review validation was performed on image data that was acquired uniquely from the data used for training; independent reviewers (not employed by Radformation) were used to evaluate the clinical appropriateness of structure models

standards: NRG/RTOG guidelines

Retrospective clinical

n=46 images

endpoints: DSC values were calculated between ground truth contour data and AutoContour structures; qualitative clinical appropriateness of AutoContour structures generated on these scans was graded by clinical experts

Retrospective clinical

n=92 images

endpoints: mean training DSC

standards: NRG/RTOG guidelines

Retrospective clinical

n=20 images

endpoints: DSC values; qualitative clinical appropriateness

Reported performance (4 observations)

diceas written: “DSC (A_Pulmonary, CT Initial)”0.88CI 0.83

source quote (p.18)

“A_Pulmonary Medium 0.65 169 43 0.88 0.03 0.83”

diceas written: “DSC (A_Pulmonary, CT External)”0.93CI 0.89

source quote (p.22)

“A_Pulmonary Medium 0.65 20 0.93 0.02 0.89 4.6”

diceas written: “DSC (Cerebellum, MR Initial)”0.93CI 0.91

source quote (p.26)

“Cerebellum Medium 0.65 0.93 0.01 0.91”

diceas written: “DSC (Cerebellum, MR External)”0.93CI 0.91

source quote (p.28)

“Cerebellum Medium 0.65 20 0.93 0.01 0.91 4”

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

recalls in product code, 24mo

MAUDE reports in code, 12mo

—

vs code's own 3-yr baseline

drift signals on this device

re_clearance
The FDA AI/ML device list shows a newer 510(k) K260509 (decision 2026-03-19) from Radformation, Inc. for a matching device line ("AutoContour (RADAC V5)") — a new clearance for the same line is a change event.
first seen 2026-07-08 · k_number:K260509
re_clearance
The FDA AI/ML device list shows a newer 510(k) K242729 (decision 2024-12-09) from Radformation, Inc. for a matching device line ("AutoContour (Model RADAC V4)") — a new clearance for the same line is a change event.
first seen 2026-07-08 · k_number:K242729

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).