ECR 2026

RPS 1002 - Breast cancer screening technology, tools and trends

From Screening to Clinical Diagnosis: Can AI Match Expert Eyes in Mammography?

Speaker: Manuel Rafael López De La Torre Carretero, Pamplona/Spain

Date and time: March 5, 14:30 - 16:00 CET

AI sensitivity was 87.9%, with limited specificity (63.9%), with 64.4% accuracy. Radiologists showed 91.3% sensitivity, with significantly higher specificity (86.3%), overall precision (86.5%), and agreement with the reference standard (kappa = 0.20 vs. 0.06 for AI).

RPS 1102 - Revolutionising breast imaging with artificial intelligence

The effect of AI on retrospectively visible interval cancers in mammography screening – results from the randomised controlled MASAI trial

Speaker: Dr. Veronica Hernstrom, Lund/Sweden

Date and time: March 5, 16:30 - 18:00 CET

The use of AI in mammography screening yielded fewer interval cancers with minimal signs (14 [17%] vs 26 [28%]) at screening compared with standard double reading, indicating its ability to aid in detecting subtle malignancies. A further reduction of interval cancer may be achievable since a substantial proportion of the retrospectively visible interval cancers were assigned elevated risk scores and were correctly localised by AI.

Influence of AI-informed Disease Prevalence on Radiologist Performance: Insights from the Mammography Screening with Artificial Intelligence trial (MASAI)

Speaker: Dr. Jessie Gommers, Nijmegen/Netherlands

Date and time: March 5, 16:30 - 18:00 CET

For examinations classified as low risk (AI score 1–7), AI-supported reading led to a reduction in RR (0.50% vs 0.61%, P=.043) and FPR (0.49% vs 0.59%, P=.049 ), without affecting CDR (0.17‰ vs 0.22‰, P=.804), compared to standard reading. For intermediate-risk examinations, AI-supported reading resulted in increased RR (2.29% vs 1.58%, P<.001) and FPR (1.90% vs 1.30%, P=.003), with no change in CDR (3.94‰ vs 2.75‰, P=.224). For high-suspicion examinations, AI-supported reading increased RR (14.41% vs 9.44%, P<.001), FPR (6.69% vs 3.40%, P<.001), and CDR (77.23‰ vs 60.34‰, P=.004).

Artificial Intelligence in Mammography Screening in Norway (AIMS Norway): a randomized controlled trial

(Featuring Transpara® 2.1)

Speaker: Dr. Solveig Hofvind, Oslo/Norway

Date and time: March 5, 16:30 - 18:00 CET

A new randomised, controlled, non-inferiority, parrallel group, single blind trial that has started. Recruitment began in November 2024 in Western Norway and in September 2025 in the Central region. Recruitment in Northern Norway is planned to start by the end of 2025. The trial aims to enroll 140,000 women. As of October 2025, 79% of women attending the screening program in participating regions have consented.

AI risk score for a new versus old version of a CE-marked AI model for breast cancer detection

(Featuring Transpara® 2.1)

Speaker: Dr. Solveig Hofvind, Oslo/Norway

Date and time: March 5, 16:30 - 18:00 CET

Among the screen-detected cancer cases, 7.9% (58/737) had higher AI score in the new versus old version. Among the interval cancers, 11.5% (23/200) had higher score in the new versus old version. An updated version of the AI model resulted in a higher detection of cancers in the high AI risk score group. However, the total number of examinations in the high-risk group did also increase, which resulted in a higher proportion of high AI score, but no cancer detected.

Can artificial intelligence detect additional cancers on screening digital breast tomosynthesis?

(Featuring Transpara® 2.1)

Speaker: Dr. Victor Carl Martin Dahlblom, Tygelsjö/Sweden

Date and time: March 5, 16:30 - 18:00 CET

During the first interval and screening round, 23% (6/22) of the interval cancers and 26% (15/57) of the screening-detected cancers had score 10 at DBT screening, compared to 3.8% (527/13986) among women without cancer. AI analysis of DBT screening images could potentially detect a substantial amount of additional cancers compared to radiologist double reading of DBT.

RPS 1402 - Hot Topic: Personalised imaging

Epidemiological and deep learning breast cancer risk models compared for an increased-risk population

Speaker: Dr. Machteld Keupers, Leuven/Belgium

Date and time: March 6, 12:30 - 13:30 CET                      

Transpara Risk (0.73 (95% CI [0.52; 0.95]) showed superior performance over Mirai (0.57 (95% CI [0.38; 0.77]) (p=0.042). Transpara also outperformed traditional risk models: CanRisk at 0.58 (95% confidence interval (CI) [0.37; 0.79]) and IBIS lifetime risk it was 0.61 (95% CI [0.49; 0.72]). 

RPS 1905 - Building the AI breast imaging service: from deployment to clinical practice

Beyond Screening: AI Performance in Mammographic Breast Cancer Recurrence Detection

(Featuring Transpara® 2.1)

Speaker: M.D. Zahra Agham, NKI/Amsterdam/Netherlands

Date and time: March 7, 12:30 - 13:30 CET

Sensitivity of stand-alone AI for recurrence detection was lower than for radiologists at the same specificity. Increase in risk category was, however, associated with local recurrence.

Comparing the Performance of Top Ranked AI Models from the RSNA 2023 Screening Mammography Breast Cancer Detection AI Challenge to Commercial AI Models

Speaker: Dr. Yan Chen, Nottingham/UK

Date and time: March 7, 12:30 - 13:30 CET

The Top 7 Challenge algorithms achieved AUCs between 0.903 and 0.947, and the commercial AI product achieved an AUC of 0.933. The Top ranked Challenge model AUC was not different to the commercial AI AUC (Delong’s method: P = .18).