Artificial Intelligence for Prostate Cancer Detection and Screening on MRI in Global Populations: The SCARLET-1 Study.
Authors
Affiliations (33)
Affiliations (33)
- Diagnostic Image Analysis Group, Radboud University Medical Center, Nijmegen, the Netherlands.
- Minimally Invasive Image-Guided Intervention Center, Radboud University Medical Center, Nijmegen, the Netherlands.
- Department of Health and Information Technology, Ziekenhuisgroep Twente, Almelo, the Netherlands.
- Department of Radiology, Netherlands Cancer Institute, Amsterdam, the Netherlands.
- Division of Surgery and Interventional Science, University College London, London, United Kingdom.
- Centre for Urology Imaging, Prostate, AI and Surgical Studies Research Group, University College London, London, United Kingdom.
- Department of Radiology, Duke University School of Medicine, Durham, NC, United States.
- Department of Radiology and Biomedical Imaging, University of California, San Francisco, CA, United States.
- Department of Medical Imaging, The Third Affiliated Hospital, Southern Medical University, Guangzhou, China.
- Department of Diagnostic Radiology, Diagnostic Imaging Center, 108 Military Central Hospital, Hanoi, Vi-etnam.
- Department of Radiology, Aga Khan University Hospital, Nairobi, Kenya.
- Department of Radiology and Nuclear Medicine, Erasmus University Medical Center, Rotterdam, the Netherlands.
- Department of Radiology, Isala Hospital, Zwolle, the Netherlands.
- Department of Medicine, Surgery and Dentistry, University of Salerno, Baronissi, Italy.
- Computational Clinical Imaging Group, Centre for the Unknown, Champalimaud Foundation, Lisbon, Portugal.
- Department of Radiology, Royal Marsden Hospital, Sutton, United Kingdom.
- Department of Radiology, Candiolo Cancer Institute, FPO-IRCCS, Candiolo, Turin, Italy.
- Department of Urology, Feinberg School of Medicine, Northwestern University, Chicago, IL, United States.
- Department of Radiology, University Medical Center Groningen, Groningen, the Netherlands.
- Department of Circulation and Medical Imaging, Norwegian University of Science and Technology, Trond-heim, Norway.
- Department of Radiology and Nuclear Medicine, St. Olavs hospital, Trondheim University Hospital, Trond-heim, Norway.
- Department of Radiology, Ziekenhuisgroep Twente, Hengelo, the Netherlands.
- Department of Multi-Modality Medical Imaging, Technical Medical Centre, University of Twente, Enschede, the Netherlands.
- Molecular Imaging Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD, United States.
- Department of Quantitative Health Sciences and Department of Diagnostic Radiology, Cleveland Clinic Foundation, Cleveland OH, United States.
- Paul Strickland Scanner Centre, Mount Vernon Cancer Centre, London, United Kingdom.
- Imperial Prostate, Division of Surgery, Department of Surgery and Cancer, Imperial College London, London, United Kingdom.
- Institute of Molecular Medicine and Surgery, Karolinska Institutet, Stockholm, Sweden.
- Department of Radiology, Capio Saint Göran's Hospital, Stockholm, Sweden.
- Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden.
- Department of Clinical Sciences, Danderyd Hospital, Karolinska Institutet, Stockholm, Sweden.
- Department of Urology, University College London Hospitals NHS Foundation Trust, London, United Kingdom.
- Department of Medical Imaging, Radboud University Medical Center, Nijmegen, the Netherlands.
Abstract
Background Artificial intelligence (AI) systems have demonstrated non-inferiority to radiologists for prostate cancer detection on MRI, but evidence in global diagnostic and population-based screening settings remains limited. Purpose To determine whether an AI system can achieve diagnostic performance comparable to radiology readings for detecting Gleason grade group ≥2 prostate cancer on MRI in external diagnostic and screening datasets. Materials and Methods In the SCARLET-1 retrospective validation study, prostate MRIs from adults with suspected prostate cancer (January 2012-October 2024) were used to evaluate an AI system, with external testing in diagnostic (PRIME) and organized screening settings (STHLM3-MRI and IP1-PROSTAGRAM). The primary outcome was agreement between AI assessments and standard-of-care diagnoses of Gleason grade group ≥2 cancer, with diagnostic interchangeability defined relative to the estimated agreement of an average radiologist. Secondary outcomes included area under the receiver operating characteristic curve (AUC) across datasets and subgroups. Results 22,445 examinations (21,252 male patients; median age, 66 years [IQR 60-71]) were included: 20,471 prostate MRI examinations (19,278 patients; 26 cities, 14 countries) for AI training and internal testing, and 1974 patients (20 external cities, 12 countries) for external testing. External AUCs were 0.86 (95% CI: 0.83, 0.90) in PRIME, 0.89 (95% CI: 0.86, 0.92) in STHLM3-MRI, and 0.84 (95% CI: 0.70, 0.95) in IP1-PROSTAGRAM. The primary outcome was 83.0% (395 of 476; adjusted 95% CI: 78.8%, 87.0%), 87.9% (974 of 1108; adjusted 95% CI: 85.6%, 90.3%), and 87.7% (342 of 390; adjusted 95% CI: 83.6%, 91.5%), respectively, meeting prespecified interchangeability criteria in all datasets. Subgroup performance was consistent across age, risk, cancer location and ethnicity (mean ΔAUC, 0.02), but varied substantially across 25 sites (28% with AUC<0.80; 32% with AUC ≥0.90) and by image quality (AUC, 0.78-0.88). Conclusion AI demonstrated diagnostic performance comparable to that of radiology readings across external diagnostic and screening datasets. © The Author(s) 2026. Published by the Radiological Society of North America under a CC BY 4.0 license.