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) that allows pre-operative planning of primary total hip arthroplasty in real time using the Zimmer Biomet Taperloc G7 system.”

AlgorithmThe software uses a series of algorithms to create 3D models and fit femoral stems and acetabular cups. It includes an AI-based automatic image segmentation algorithm trained on CT scans and statistical shape models of the femur and pelvis.

source quote (p.4)

“Using a series of algorithms, the software creates a 3D model and relevant measurements derived from the patient's pre-dimensioned CT scan. Formus Hip generates a 3D model without any user input. Additional algorithms fit the femoral stem and acetabular cup based on the patient anatomy. The software allows the user to adjust the plan interactively to achieve the desired clinical targets. 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)FDA source did not state this

PCCPFDA source did not state this

Cybersecurity addressedFDA source did not state this

Validation studies (2)

Standalone

n=60 patients

endpoints: Difference between the automatic (Formus Hip generated) and manual 3D models were quantified using the Sorensen-Dice coefficient (Dice); Mean Absolute Distance (MAD); Hausdorff Distance (HD)

Standalone

n=133 patients

endpoints: Comparison of recommended implant sizes by Formus Hip with ground truth determined by orthopaedic surgeons; Performance goal: at least 80% of cup and stem sizes recommended by Formus Hip were within ±2 sizes of the ground truth

Reported performance (4 observations)

diceas written: “Average Dice score (Hemipelvis) for 3D models from image segmentation”0.95

source quote (p.7)

“The average Dice score must be equal or greater than 0.9 Hemipelvis: 0.95”

diceas written: “Average Dice score (Femur) for 3D models from image segmentation”0.97

source quote (p.7)

“The average Dice score must be equal or greater than 0.9 Femur: 0.97”

diceas written: “Average Dice score (Hemipelvis) for 3D models generated from statistical shape modelling”0.95

source quote (p.7)

“The average Dice score must be equal or greater than 0.9 Hemipelvis: 0.95”

diceas written: “Average Dice score (Femur) for 3D models generated from statistical shape modelling”0.97

source quote (p.7)

“The average Dice score must be equal or greater than 0.9 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

recalls in product code, 24mo

MAUDE reports in code, 12mo

—

vs code's own 3-yr baseline

drift signals on this device

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).