Formus Hip

K213272

Formus Labs, Ltd · cleared 2023-03-31 · product code QIH · Radiology

Premarket evidence — what FDA accepted

Device typesamd
source quote (p.4)
“Formus Hip is a semi-automated Software as a Medical Device (SaMD)”
AlgorithmAI-based automatic image segmentation algorithm trained on CT scans of male and female subjects with typical and atypical bony anatomy between the ages of 21 and 94. Formus Hip also uses statistical shape models of the femur and pelvis trained on segmented 3D models of male and female subjects with typical and atypical bony anatomy between the ages of 18 and 89.
source quote (p.4)
“Formus Hip uses an AI-based automatic image segmentation algorithm trained on CT scans of male and female subjects with typical and atypical bony anatomy between the ages of 21 and 94. Formus Hip also uses statistical shape models of the femur and pelvis trained on segmented 3D models of male and female subjects with typical and atypical bony anatomy between the ages of 18 and 89.”
Adaptive (vs locked)No
source quote (p.4)
“The training datasets are independent from testing and validation datasets. Training data and internal testing data are tracked in a single record file under version control where they are labelled as either training and testing. Code used for training and testing is read from this record file so that a data point is never mixed between the training and testing datasets. Validation data was sourced from different geographies and stored in locations separate from training and internal testing data to ensure independence.”
PCCPFDA source did not state this
Cybersecurity addressedFDA source did not state this

Validation studies (2)

Retrospective clinical

n=60 patients

endpoints: Dice score; Mean Absolute Distance (MAD); Hausdorff Distance (HD)

Retrospective clinical

n=133 patients

endpoints: Proportion of cup and stem sizes within ±2 sizes of ground truth

Reported performance (4 observations)

diceas written: “Dice score (Hemipelvis) from image segmentation”0.95
source quote (p.7)
“Hemipelvis: 0.95”
diceas written: “Dice score (Femur) from image segmentation”0.97
source quote (p.7)
“Femur: 0.97”
diceas written: “Dice score (Hemipelvis) from statistical shape modelling”0.95
source quote (p.7)
“Hemipelvis: 0.95”
diceas written: “Dice score (Femur) from statistical shape modelling”0.97
source quote (p.7)
“Femur: 0.97”

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
3
MAUDE reports in code, 12mo
—
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 Radiology 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/K213272