Transpara (2.1.0)

K241831

ScreenPoint Medical B.V. · cleared 2024-11-25 · product code QDQ · Radiology

Premarket evidence — what FDA accepted

Device typesamd

source quote (p.6)

“Transpara is a software only application designed to be used by physicians to improve interpretation of full-field digital mammography (FFMD) and digital breast tomosynthesis (DBT).”

AlgorithmDeep learning algorithms are applied to images for recognition of suspicious calcifications and soft tissue lesions (including densities, masses, architectural distortions, and asymmetries). Algorithms are trained with a large database of biopsy-proven examples of breast cancer, benign abnormalities, and examples of normal tissue.

source quote (p.6)

“Deep learning algorithms are applied to images for recognition of suspicious calcifications and soft tissue lesions (including densities, masses, architectural distortions, and asymmetries). Algorithms are trained with a large database of biopsy-proven examples of breast cancer, benign abnormalities, and examples of normal tissue.”

Adaptive (vs locked)FDA source did not state this

PCCPNo

Cybersecurity addressedYes

source quote (p.9)

“Cybersecurity in Medical Devices: Quality System Considerations and Content of Premarket Submissions”

Validation studies (3)

Bench

sample size not stated

Retrospective clinical

n=10,207 cases

endpoints: Exam based sensitivity for cancer detection; False positive rates; Area under the ROC Curve

standards: IEC 62366-1, ISO 20417, ISO 14971, IEC 62304, IEC 82304-1

Retrospective clinical

n=5,724 cases

endpoints: Exam based sensitivity for cancer detection; False positive rates

standards: IEC 62366-1, ISO 20417, ISO 14971, IEC 62304, IEC 82304-1

Reported performance (10 observations)

sensitivity0.957CI 93.7 - 97.6

source quote (p.12)

“FFDM with TA 95.7% (93.7 - 97.6)”

specificity0.7

source quote (p.12)

“Sensitivity for Sensitive Mode (70% specificity)”

aurocas written: “auc”0.958CI 0.946 - 0.969

source quote (p.12)

“FFDM with TA ... 0.958 (0.946 - 0.969)”

sensitivityas written: “Sensitivity (80% specificity) - FFDM with TA”0.954CI 93.4 - 97.4

source quote (p.12)

“FFDM with TA ... 95.4% (93.4 - 97.4)”

specificityas written: “Specificity (80% specificity) - FFDM with TA”0.8

source quote (p.12)

“Sensitivity for Specific Mode (80% specificity)”

sensitivityas written: “Sensitivity (97% specificity) - FFDM with TA”0.827CI 79.1 - 86.4

source quote (p.12)

“FFDM with TA ... 82.7% (79.1 - 86.4)”

specificityas written: “Specificity (97% specificity) - FFDM with TA”0.97

source quote (p.12)

“Sensitivity for Elevated Risk (97% specificity)”

sensitivityas written: “Sensitivity (70% specificity) - DBT with TA”0.946CI 91.2 - 98.0

source quote (p.12)

“DBT with TA 94.6% (91.2 - 98.0)”

specificityas written: “Specificity (70% specificity) - DBT with TA”0.7

source quote (p.12)

“Sensitivity for Sensitive Mode (70% specificity)”

aurocas written: “AUC - DBT with TA”0.941CI 0.921 - 0.958

source quote (p.12)

“DBT with TA ... 0.941 (0.921 - 0.958)”

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).