PARADIGM
Research type
Research Study
Full title
A study assessing whether artificial intelligence is non-inferior to radiologists in the diagnosis of clinically significant prostate cancer
IRAS ID
338739
Contact name
Veeru Kasivisvanathan
Contact email
Sponsor organisation
University College London
Clinicaltrials.gov Identifier
Z6364106/2025/08/46, UCL Data Protection Registration Number
Duration of Study in the UK
3 years, 0 months, 1 days
Research summary
Prostate MRI Analysis by Radiologists and AI – Disease Identification and Guided Management (PARADIGM) will be the first prospective, international, multicentre, non-inferiority, within-patient, diagnostic yield study assessing whether a best-in-class AI algorithm is non-inferior to radiologists in the detection of clinically significant prostate cancer. 500 biopsy naïve men with a clinical suspicion of prostate cancer will undergo multiparametric MRI of the prostate. The scan will first be reported by the AI algorithm (report 1), and subsequently, by the radiologist blinded to the initial report (report 2), and a secondary AI algorithm (report 3, if applicable).The radiologist will be unblinded to the AI algorithm(s)' report and produce a merged report. Men with suspicious lesions on either the AI or radiologist report will undergo prostate biopsy. The primary outcome is the proportion of men with clinically significant
prostate cancer detected. Key secondary outcomes will include the proportion of clinically insignificant prostate cancer detected, as well as the diagnostic performance of the AI algorithm, and cost-effectiveness analysis.REC name
Wales REC 3
REC reference
26/WA/0100
Date of REC Opinion
30 Apr 2026
REC opinion
Further Information Favourable Opinion