Evidence

ECR: Poster Presentation
Monajemi et al. 2026

Thyroid nodule characterisation on ultrasound is time-intensive and prone to substantial inter-radiologist variability and errors due to the multi-step requirements of ACR TI-RADS (TR) assessment and structured reporting. We evaluate an AI-enhanced thyroid ultrasound assistant that fully automates nodule detection, characterisation, and report generation following TR requirements, and assess its impact on efficiency, agreement, and diagnostic accuracy.   

ECR: Poster Presentation
Wakelin et al. 2026

Despite lung cancer being a leading cause of mortality in Europe, eligibility requirements for screening are complex and region specific, limiting adoption rates. Artificial intelligence (AI) that identifies lung nodules in routine CT scans could help detect lung nodules that would otherwise be missed, but it is unclear if the same AI can be used for both screening and routine CTs given the different patient populations scanned. Here we characterize the performance of an AI algorithm to detect lung nodules within routine chest CT scans and compare to screening low dose CT (LDCT).  

ECR: Poster Presentation
Moeskops et al. 2026

White matter hyperintensities (WMHs) are clinically actionable imaging markers of several neurological conditions, including cognitive impairment and multiple sclerosis (MS). Manual segmentation of WMHs on Fluid-Attenuated Inversion Recovery (FLAIR) MRI is time-consuming and impractical for routine clinical practice, limiting consistent use in risk stratification and longitudinal assessment. This study evaluates an AI algorithm for automated quantification and localization by identifying, segmenting, and labelling anatomical location of WMHs.

ECR: Poster Presentation
Hasegawa et al. 2026

Breast arterial calcifications (BAC) visible on mammography have been associated with cardiovascular and peripheral vascular disease.  However, BAC reporting is inconsistent and subjective due to impractical manual quantification. This study assesses the standalone detection and localization performance of an AI-based BAC algorithm on full-field digital mammography (FFDM) and digital breast tomosynthesis (DBT).

ECR: Poster Presentation
Everitt  et al. 2026

While prostate MRI can improve prostate cancer (PCa) detection and enable more accurate MRI-guided fusion-biopsy, these gains require radiologist expertise. Shortages of expert radiologists and inter-reader variability can lead to inconsistent diagnoses and suboptimal biopsy targeting. AI could improve radiologists’ interpretations. We evaluated an integrated AI web-based platform providing fully automated lesion segmentation and lesion-level risk classification, assessing effects on PCa detection, localisation and inter-reader consistency.

ECR: Poster Presentation
Kim et al. 2026

Lung nodule detection and tracking on low-dose chest CT is time intensive and subject to substantial interobserver variability, impacting early-stage lung cancer detection. This study aimed to determine whether AI-assisted lung nodule detection improves radiologists’ diagnostic performance, consistency, and efficiency across nodule sizes.

RSNA: Podium Presentation

Screening mammography volumes continue to increase, while radiologist availability remains limited. This creates workflow challenges that can lead to fatigue and burnout. Through a radiologist-industry collaboration, an Efficient Clickable Reporting Tool (ECRT) was co-developed to address reporting inefficiencies identified by practicing radiologists. The study evaluates the real-world impact of this collaboration on mammography read times following large-scale clinical deployment. Post-ECRT deployment, the proportion of exams requiring less than 1.5 minutes to review and interpret increased from 39.9% to 49.2%.

RSNA: Podium Presentation

Women with dense breasts have a higher risk of developing breast cancer and a higher risk of cancer going undetected given cancer is more challenging to detect in dense breast tissue. This study assesses the clinical outcomes of a large-scale deployment of a multistage AI-driven workflow, including stratified analyses by breast density. Results show significantly improved clinical outcomes (CDR and PPV1) for women with dense and non-dense breasts.

RSNA: Podium Presentation
McCabe et al. 2025

Interpretive performance in screening mammograms (SMGs) varies widely. An AI-driven workflow was deployed at scale, and performance before and after implementation was compared between radiologists with and without fellowship training in breast or women’s imaging. The Multistage AI-Driven workflow significantly improved CDR and PPV1 for General radiologists to levels comparable with Fellowship-trained radiologists.

RSNA: Poster Presentation
Louis et al. 2025

This study assesses the cancer subtypes detected in a large-scale deployment of a multistage AI-driven workflow compared to the standard of care (SOC). Employing the multistage AI-driven workflow significantly improved clinical outcomes in terms of CDR in a cohort of over 2400 cancers, with the majority of cancers detected being clinically relevant and no corresponding increase in the proportion of DCIS, showing benefits for screening mammography.

Nature Health
Louis et al. Nov 2025

Artificial intelligence shows promising results for improving early breast cancer detection and overall screening outcomes. Here the AI-Supported Safeguard Review Evaluation (ASSURE) study evaluates an AI workflow on digital breast tomosynthesis exams from women across four states to optimize early cancer detection. Implementation of the AI workflow improved screening effectiveness with equitable benefits.

DeepHealth
Kaminer. Nov 2024

The workflows of two centers were analyzed before and after the installation of DeepHealth’s prostate solution software. Time was reduced in the scan evaluation process (radiologists’ read-time) by 14.1% over two centers with 8 out of 9 radiologists decreasing their evaluation time. With the software’s automatic targeting for fusion biopsies, time was also reduced in the overall workflow for biopsy-recommended cases by 232 hours (6 weeks), or 37%.

BTOG: Poster Presentation
Palmer et al. 2025

Clinical outcomes from the SWAG TLHC programme were reviewed to better understand potential harms, including over-investigation, missed diagnoses and cancers developing between screening rounds. Of the scans reviewed, 4098/4198 (97.6%) were negative (confirmed true negatives) and 80/4198 (1.9%) were abnormal (confirmed true positives). One person was diagnosed with lung cancer 18 months after a reportedly normal TLC scan, giving a false negative rate of 0.02%.

RSNA: Poster Presentation
Kim et al. 2024

An AI-driven safeguard review process was implemented prospectively, and its custom-built AI algorithm was used to flag the most suspicious screening DBT exams that had not been recalled by the initial interpreting radiologist. An expert breast imaging specialist performed a second, safeguard review of the 2,296 flagged exams. This resulted in the detection of 41 additional cancers, mostly invasive, 22.0% of which were deemed aggressive.

Ipsos
Mouland et al. Nov 2024

The TLHC programme (2019–2024) tested lung health checks in real-world settings to diagnose lung cancers earlier, following promising pilot results. Across more than 25 sites, 1.22 million invitations were issued, with 44% uptake. Of 163,000 CT scans, 2,748 cancers were diagnosed (1.7% conversion), around three-quarters at stage 1–2. Impact analysis showed an additional 781 early-stage cancers diagnosed in pilot areas, replicating small-scale outcomes and demonstrating real-world effectiveness while identifying demographic and operational challenges.