Transpara
Commercial mammography AI with FDA clearance and CE marking, used as a concurrent or replacement reader in breast screening; the MASAI randomised trial in Sweden used it to triage screening reads.
At a glance
What it does
Transpara supports radiologists reading screening mammograms by highlighting suspicious findings and triaging examinations to single or double reading. It is the system evaluated in MASAI, the first randomized controlled trial of AI in mammography screening, conducted within the Swedish national programme with over 105,000 participants.
Notable uses in oncology
In the MASAI trial, AI-supported reading increased cancer detection while reducing screen-reading workload by 44% (safety analysis, The Lancet Oncology 2023). Final results (The Lancet, 2026) reported 12% fewer interval cancers and 27% fewer aggressive cancer subtypes, with false-positive rates similar to standard double reading.
Tasks, data types and cancers
Architecture
Training data
Not disclosed by the manufacturer. Checked on 2026-09-06: the ScreenPoint Medical site (screenpoint-medical.com) names no training dataset, no number of mammograms or examinations, no source institutions and no architecture; it documents FDA clearance, CE marking and clinical validation (including a randomised controlled trial) instead. Published training-corpus details for this device are not available from the vendor.
Evaluation
| Benchmark / dataset | Metric | Value | External validation | Source |
|---|---|---|---|---|
| MASAI randomised trial (Sweden), first results | cancer detection rate / reader workload | AI-supported screening: similar cancer detection with 44% lower screen-reading workload (Lancet Oncol 2023) | yes | Source |
How to run
Not yet documented on this card.
Regulatory status and intended use
Manufacturer states (verified 2026-09-06): "Transpara is CE marked and FDA cleared for 2D and 3D mammography. CE (0344) certified. FDA 510(k) K241831 and K232096." Cleared for full-field digital mammography (FFDM) and digital breast tomosynthesis (DBT); described by the manufacturer as vendor agnostic. Always check the clearance covering the specific software version deployed at your site.
Regulatory status is quoted from the source linked above and can change. Research-use-only models must not be used for clinical decisions.
Limitations and bias
- Closed weights; performance depends on vendor and population.
- Trial evidence is programme-specific (Swedish double-reading setting).
Sources
This page is educational — it is not medical advice and does not replace consultation with an oncologist. Diagnostic and treatment decisions are made solely by specialist physicians.