Transpara helps Swedish breast screening find more cancers

In the MASAI trial of 105,934 women, Transpara-supported screening found 6.4 cancers per 1,000 against 5.0 with standard double reading and cut screen reading by 44.2%.

42
Benefit status
Reported
Evidence
Independently studied
Observed
Sources
5 publications
Download data CiteSuggest a correction

Benefits

Benefit

A detection rate of 6.4 per 1,000 against 5.0 per 1,000

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

Transpara

Awaiting review

Model

This is an explicit unresolved assignment: the sources on record do not settle this question.

Not reviewedUnresolved

Reported configuration

1.7.0

Awaiting review

Provider

Unresolved

Not reviewedUnresolved

Role

Unresolved

Not reviewedUnresolved

Configuration

Unresolved

Not reviewedUnresolved

Inputs and outputs

Unresolved

Not reviewedUnresolved

Autonomy

Unresolved

Not reviewedUnresolved

Workflow

Transpara 1.7.0 scored every screening examination on a scale from 1 to 10. Examinations scored 10 went to double reading and all others to single reading. Readers saw the AI risk scores and detection marks while reading. The trial ran inside an organized population screening program with experienced breast radiologists.

Awaiting review

Human oversight

Unresolved

Not reviewedUnresolved

Relation to AI

The trial randomly assigned women to AI-supported screening or to standard double reading without AI, so the difference in outcomes reflects the AI-supported workflow. The design does not separate the effect of triage from the effect of the detection marks.

Awaiting review

AI contribution

The trial randomly assigned women to AI-supported screening or to standard double reading without AI, so the difference in outcomes reflects the AI-supported workflow. The design does not separate the effect of triage from the effect of the detection marks.

Awaiting review

Measurement

Studies

Study 1: randomized. Study 2: randomized. [1] [2]

Fully reviewed

Review date: 29 October 2026.

Sources: [1] [2]

Study design

Study 1: randomized. Study 2: randomized. [1] [2]

Fully reviewed

Review date: 29 October 2026.

Sources: [1] [2]

Measurement

Study 1: Interval cancers from registry follow-up; sensitivity and specificity. Study 2: Screen-detected cancers, recalls and screen readings. [1] [2]

Fully reviewed

Review date: 29 October 2026.

Sources: [1] [2]

Setting and sample

Study 1: 105,934 women screened at four sites in southwest Sweden, 12 April 2021 to 7 December 2022. Study 2: 105,915 women analyzed. [1] [2]

Fully reviewed

Review date: 29 October 2026.

Sources: [1] [2]

Measures

Study 1: Interval cancer rate per 1,000 participants. Study 2: Cancer detection rate per 1,000 and screen-reading workload. [1] [2]

Fully reviewed

Review date: 29 October 2026.

Sources: [1] [2]

Comparison

Study 1: Standard double reading without AI. Study 2: Standard double reading without AI. [1] [2]

Fully reviewed

Review date: 29 October 2026.

Sources: [1] [2]

Results

Study 1: 1.55 vs 1.76 per 1,000; proportion ratio 0.88 (95% CI 0.65 to 1.18), non-inferior; sensitivity 80.5% vs 73.8%. Study 2: 6.4 vs 5.0 per 1,000 (ratio 1.29); 44.2% fewer screen readings. [1] [2]

Fully reviewed

Review date: 29 October 2026.

Sources: [1] [2]

Additional value from AI

Study 1: Sensitivity 6.7 percentage points higher with AI. Study 2: 1.4 more cancers per 1,000 women. [1] [2]

Fully reviewed

Review date: 29 October 2026.

Sources: [1] [2]

Limitations

Study 1: Interval cancer difference not statistically significant; one AI product; mortality not measured. Study 2: Some extra cancers may be overdiagnosed in situ disease. [1] [2]

Fully reviewed

Review date: 29 October 2026.

Sources: [1] [2]

Denominators

Study 1: AI triage plus detection support for all examinations in the intervention group. Study 2: AI triage plus detection support. [1] [2]

Fully reviewed

Review date: 29 October 2026.

Sources: [1] [2]

Materials for replication

Study 1: Trial registered as NCT04838756. Study 2: Trial registered as NCT04838756. [1] [2]

Fully reviewed

Review date: 29 October 2026.

Sources: [1] [2]

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: ScreenPoint Medical. Operator: Unilabs Mammography Unit at Skåne University Hospital with Lund University (MASAI trial, Swedish national screening program).

Awaiting review

Jurisdictions

Benefit: Malmö, Lund, Landskrona and Trelleborg, Skåne, Sweden. Developer: Netherlands. Operator: Sweden.

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

Readers must have the AI scores and marks in front of them when they read. The AI must perform on the local equipment and population as it did in the trial. The program must use double reading that the AI triage can replace for low-risk examinations.

Awaiting review

Replication attempts

3 follow-up entries. Benefit recurred: 3.

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-0642-source-0001–0005

Awaiting review

Citations

Numbered references below.

Awaiting review

Language

Unresolved

Not reviewedUnresolved

Source relationships

Independent

Awaiting review

Original evidence

Unresolved

Not reviewedUnresolved

Access and versions

Unresolved

Not reviewedUnresolved

Supported findings

Unresolved

Not reviewedUnresolved

Evidence strength

Independently studied. Sources: five independent peer-reviewed studies. The cited evidence includes no finding by a regulator or court. [1] [2] [3] [4] [5]

Fully reviewed

Review date: 29 October 2026.

Sources: [1] [2] [3] [4] [5]

Missing evidence

Unresolved

Not reviewedUnresolved

Grades

Assessed benefits

People. Organizations.

Awaiting review

Depth

People: 4. Organizations: 2.

Awaiting review

Reasons

People: Women in the AI group had 338 cancers found at screening against 262 in the control group, and the extra invasive cancers were mainly small and node-negative. Organizations: The screening service needed 61,248 screen readings in the AI group against 109,692 in the control group, a 44.2% reduction.

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 randomized controlled screening trial. [1]

Fully reviewed

Review date: 29 October 2026.

Sources: [1]

Field

Healthcare.

Awaiting review

Purpose

Transpara was built to detect breast cancer on screening mammograms and to score examinations by risk. In MASAI it triaged examinations to single or double reading and marked suspicious findings; the higher detection and the saved reading time were that purpose.

Awaiting review

Dates

Observed: 7 December 2022. Updated: 29 October 2026.

Awaiting review

Stage

The system was used in a pilot. A randomized trial inside the national screening program at four sites; the papers do not report a later routine rollout. [1]

Fully reviewed

Review date: 29 October 2026.

Sources: [1]

Version

SHA-256: 3c6ec576532e…

Awaiting review

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

  1. Peer-reviewed study Independent author

    Gommers J, Hernström V, Josefsson V, et al. Interval cancer, sensitivity, and specificity comparing AI-supported mammography screening with standard double reading without AI in the MASAI study. Lancet. 2026. doi:10.1016/S0140-6736(25)02464-X

  2. Peer-reviewed study Independent author

    Hernström V, Josefsson V, Sartor H, et al. Screening performance and characteristics of breast cancer detected in the Mammography Screening with Artificial Intelligence trial (MASAI). Lancet Digit Health. 2025. doi:10.1016/S2589-7500(24)00267-X

  3. Peer-reviewed study Independent author

    Lång K, Josefsson V, Larsson AM, et al. Artificial intelligence-supported screen reading versus standard double reading in the Mammography Screening with Artificial Intelligence trial (MASAI). Lancet Oncol. 2023. doi:10.1016/S1470-2045(23)00298-X

  4. Peer-reviewed study Independent author

    Ferre R, Benefield T, Kuzmiak CM. Artificial intelligence-supported double reading in European population breast cancer screening: a systematic review and meta-analysis of prospective programs. Clin Imaging. 2026;139:110923. doi:10.1016/j.clinimag.2026.110923

  5. Peer-reviewed study Independent author

    Ciurescu S, Buciu V, Ilaș DG, Pârvănescu R, Șerban D. Ten years of artificial intelligence in screening mammography: a systematic review and meta-analysis of diagnostic accuracy and clinical implementation (literature published 2015-2025). Diagnostics (Basel). 2026;16(18):3045. doi:10.3390/diagnostics16183045

Collections: Cross-border cases.

Cite this case

BibTeX Download BibTeX

Version 3c6ec576532e, archived in the edition of 29 October 2026

Benefit marks

Depth

The number and color show the depth of a benefit.

  1. A convenience, or time or money saved.
  2. A measurable improvement in a result or ability.
  3. A lasting change, an ability restored, or a risk removed.
  4. A serious harm averted or a serious condition resolved in time.
  5. A life or livelihood preserved, or a change documented across a population.

Beneficiary

  • Circle: a person
  • Square: an organization
  • Triangle: society

Evidence

  • Solid. A court or regulator found a benefit, or researchers independently studied it.

  • Hatched. Independent reporting.

  • Outline. Claimed by the parties. Only the people or organizations involved, or sources whose independence has not been assessed, support the benefit.

The fill changes with the evidence. The depth color stays the same.

How benefits are assessed