Benefits
- Benefit
-
Found 6.7 cancers per 1,000 against 5.7 without it
Awaiting review
- Outcome identifiers
-
Unresolved
Not reviewedUnresolved
- Category
-
Improved Task Accuracy
Awaiting review
- Status
-
Reported benefit
Awaiting review
- Beneficiary groups
-
People. Organizations.
Awaiting review
- Beneficiaries
-
Unresolved
Not reviewedUnresolved
- Reach
-
Unresolved
Not reviewedUnresolved
- Distribution and access
-
Unresolved
Not reviewedUnresolved
- Duration
-
Unresolved
Not reviewedUnresolved
- Expected?
-
Yes
Awaiting review
- Costs and harms
-
Unresolved
Not reviewedUnresolved
AI system
- System
-
Vara MG
Awaiting review
- Model
-
This is an explicit unresolved assignment: the sources on record do not settle this question.
Not reviewedUnresolved
- Release
-
Unresolved
Not reviewedUnresolved
- Provider
-
Unresolved
Not reviewedUnresolved
- Role
-
Unresolved
Not reviewedUnresolved
- Configuration
-
Unresolved
Not reviewedUnresolved
- Inputs and outputs
-
Unresolved
Not reviewedUnresolved
- Autonomy
-
Unresolved
Not reviewedUnresolved
- Workflow
-
Radiologists read in an AI-supported viewer that tagged likely normal examinations. A safety-net alert asked radiologists to look again at examinations that the AI judged suspicious. Final recall decisions stayed with radiologists in a double-reading program.
Awaiting review
- Human oversight
-
Unresolved
Not reviewedUnresolved
- Relation to AI
-
The study compares AI-supported double reading with standard double reading in the same program and period. It is not randomized, because radiologists chose when to use the AI, so confounding is possible.
Awaiting review
- AI contribution
-
The study compares AI-supported double reading with standard double reading in the same program and period. It is not randomized, because radiologists chose when to use the AI, so confounding is possible.
Awaiting review
Measurement
- Studies
-
Controlled observational. [3]
Fully reviewed: supporting details for Studies
Review date: 29 October 2026.
Sources: [3]
- Study design
-
Controlled observational. [3]
Fully reviewed: supporting details for Study design
Review date: 29 October 2026.
Sources: [3]
- Measurement
-
Screen-detected cancers and recalls from program records. [3]
Fully reviewed: supporting details for Measurement
Review date: 29 October 2026.
Sources: [3]
- Setting and sample
-
463,094 women aged 50 to 69 at 12 sites, July 2021 to February 2023; 119 radiologists. [3]
Fully reviewed: supporting details for Setting and sample
Review date: 29 October 2026.
Sources: [3]
- Measures
-
Breast cancer detection rate and recall rate per 1,000. [3]
Fully reviewed: supporting details for Measures
Review date: 29 October 2026.
Sources: [3]
- Comparison
-
Double reading without AI. [3]
Fully reviewed: supporting details for Comparison
Review date: 29 October 2026.
Sources: [3]
- Results
-
6.7 vs 5.7 per 1,000 (+17.6%); recall 37.4 vs 38.3 per 1,000. [3]
Fully reviewed: supporting details for Results
Review date: 29 October 2026.
Sources: [3]
- Additional value from AI
-
1.0 more cancer per 1,000 women. [3]
Fully reviewed: supporting details for Additional value from AI
Review date: 29 October 2026.
Sources: [3]
- Limitations
-
No randomization; reading-behavior bias reported; Vara funded and co-wrote the study. [3]
Fully reviewed: supporting details for Limitations
Review date: 29 October 2026.
Sources: [3]
Money
- Kind of value
-
The kind of monetary value is unclear.
Not reviewedUnresolved
- Amount
-
The monetary value is not established.
Not reviewedUnresolved
- Currency and prices
-
The currency and price basis are unclear.
Not reviewedUnresolved
- Period and unit
-
The period and unit of the monetary estimate are unclear.
Not reviewedUnresolved
- Realization
-
Whether the monetary value was realized is unclear.
Not reviewedUnresolved
- Recipients and payers
-
The recipients and payers are unclear.
Not reviewedUnresolved
- Gross or net
-
Whether the monetary figure includes costs is unclear.
Not reviewedUnresolved
- Included costs
-
The costs included in the estimate are unclear.
Not reviewedUnresolved
- Comparison
-
The financial comparison is not established.
Not reviewedUnresolved
- Method
-
The method used to calculate monetary value is unclear.
Not reviewedUnresolved
- Return and payback
-
The return on investment or payback period is not established.
Not reviewedUnresolved
People and places
- Participants
-
Developer: Vara. Operator: Radiologists at 12 screening units of the German mammography screening program.
Awaiting review
- Jurisdictions
-
Benefit: 12 screening sites,
Germany. Developer: Germany. Operator: Germany.Awaiting review
- Ownership and control
-
Unresolved
Not reviewedUnresolved
- Dependencies
-
Unresolved
Not reviewedUnresolved
- Cross-border connections
-
Unresolved
Not reviewedUnresolved
- Geopolitical implications
-
The geopolitical implications are unclear.
Not reviewedUnresolved
Follow-up
- Conditions for replication
-
Radiologists must accept a share of safety-net alerts and review them. The normal tag must not lead radiologists to miss cancers in tagged examinations. The AI must perform on the local population as it did in Germany.
Awaiting review
- Replication attempts
-
2 follow-up entries. Benefit recurred: 2.
Awaiting review
- Later findings
-
Unresolved
Not reviewedUnresolved
- Availability
-
Unresolved
Not reviewedUnresolved
- Institutional decision
-
A relevant institutional decision has not been established.
Not reviewedUnresolved
- Grounds
-
The grounds for an institutional decision are unclear.
Not reviewedUnresolved
- Action and result
-
The result of an institutional decision is not established.
Not reviewedUnresolved
- Success criterion
-
The criterion for a successful institutional decision is unclear.
Not reviewedUnresolved
Evidence
- Source numbers
-
AIBD-0644-source-0001–0003
Awaiting review
- Citations
-
Numbered references below.
Awaiting review
- Language
-
Unresolved
Not reviewedUnresolved
- Source relationships
-
Independent, Developer
Awaiting review
- Original evidence
-
Unresolved
Not reviewedUnresolved
- Access and versions
-
Unresolved
Not reviewedUnresolved
- Supported findings
-
Unresolved
Not reviewedUnresolved
- Evidence links
-
Unresolved
Not reviewedUnresolved
- Evidence strength
-
Independently studied. Sources: two independent peer-reviewed studies and the makers' peer-reviewed study. The cited evidence includes no finding by a regulator or court. [1] [2] [3]
- Missing evidence
-
Unresolved
Not reviewedUnresolved
Grades
- Assessed benefits
-
People. Organizations.
Awaiting review
- Depth
-
People: 4. Organizations: 2.
Awaiting review
- Reasons
-
People: The safety net led to 204 breast cancer diagnoses among women whose examinations radiologists had first read as normal. Organizations: Detection was 6.7 per 1,000 with AI against 5.7 per 1,000 without, with a recall rate that did not rise.
Awaiting review
- Observed or intended
-
Unresolved
Not reviewedUnresolved
- Rubric
-
Unresolved
Not reviewedUnresolved
- Assessor and date
-
Unresolved
Not reviewedUnresolved
- Alternative assessments
-
Unresolved
Not reviewedUnresolved
- Fill
-
Independently studied
Awaiting review
Entry details
- Kind
-
The case describes an evaluation or a study. A prospective observational implementation study. [1]
Fully reviewed: supporting details for Kind
Review date: 29 October 2026.
Sources: [1]
- Related entries
-
No incident in the AI Incident Database snapshot of 26 October 2026 concerns the same kind of system harming the same kind of stakeholder.
Partly reviewedUnresolved: supporting details for Related entries
Compared with the incidents in the AI Incident Database snapshot of 26 October 2026.
- Field
-
Healthcare.
Awaiting review
- Purpose
-
Vara MG was built to display mammograms and to preclassify screening examinations as normal or suspicious for radiologists. The higher detection rate was that purpose.
Awaiting review
- Dates
-
Observed: February 2023. Updated: 29 October 2026.
Awaiting review
- Stage
-
The system was in routine use. A CE-certified product used in routine screening at 12 sites, with radiologists choosing when to use it. [1]
Fully reviewed: supporting details for Stage
Review date: 29 October 2026.
Sources: [1]
- Version
-
SHA-256: d1e12cfecae8…
Awaiting review: supporting details for Version
SHA-256 of the public case, its assessments, and its displayed answers and notes, excluding its own identifier and the derived coverage summary.
References
- Other identifiers
-
Unresolved
Not reviewedUnresolved
- Category mappings
-
Unresolved
Not reviewedUnresolved
- Citable summary
-
Unresolved
Not reviewedUnresolved
- Revisions
-
1 revision. Latest: 29 October 2026.
Awaiting review
Sources
-
Peer-reviewed study Independent author
in PRAIM, a program-level safety-net mechanism recovered 204 cancers that would otherwise have been missed
-
Peer-reviewed study Independent author
The largest by far is PRAIM, a prospective evaluation across twelve German screening sites, in which 463,094 women were screened and radiologists chose voluntarily whether to use AI support; detection was 6.7 versus 5.7 per 1000, a ratio of 1.18 (1.06-1.31), with recall slightly lower and non-inferior
-
Peer-reviewed study Developer
Eisemann N, Bunk S, Mukama T, et al. Nationwide real-world implementation of AI for cancer detection in population-based mammography screening. Nat Med. 2025;31:917-924. doi:10.1038/s41591-024-03408-6
Radiologists in the AI-supported screening group achieved a breast cancer detection rate of 6.7 per 1,000, which was 17.6% (95% confidence interval: +5.7%, +30.8%) higher than and statistically superior to the rate (5.7 per 1,000) achieved in the control group.