Authors and Affiliations

Uzair Khan¹, Joecelyn Kirani Tan², Karran Bhagat³, Aaditya Tiwari⁴, Soumya Arun⁵, Abel Tesfai⁶, Alexandra Naranjo⁶, Maryam Hasanova⁴, Akash Maniam⁷, Giuseppe Banna⁸, Stergios Boussios⁹, Karan Jatwani¹⁰, Yüksel Ürün¹¹, Swarupa Mitra¹², Jeremy Teoh¹³, Benjamin Lamb¹⁴, Balraj Dhesi¹⁵, Atif Khan¹⁶, Sola Adeleke¹⁷, Aruni Ghose⁴

1. Early Cancer Institute, University of Cambridge, Cambridge, UK

2. School of Medical Sciences, University of Manchester, Manchester, UK

3. Royal Surrey Cancer Centre, Royal Surrey NHS Foundation Trust, Guildford, UK

4. Cancer AI Validation Lab, OncoFlow, London, UK

5. Centre for Cancer Screening, Prevention and Early Diagnosis, Wolfson Institute of Population Health, London, UK

6. Data and Evidence, Prostate Cancer UK, London, UK

7. Department of Oncology, Isle of Wight NHS Trust, Newport, UK

8. Department of Oncology, Portsmouth Hospitals University NHS Trust, Portsmouth, UK

9. Department of Medical Oncology, Ioannina University Hospital and University of Ioannina, Ioannina, Greece

10. Urological Oncology Program, George Washington Cancer Centre, Washington DC, USA

11. Ankara University Cancer Research Institute, Ankara, Turkey

12. Fortis Cancer Institute, Fortis Memorial Research Institute, Gurugram, India

13. S.H. Ho Urology Centre, The Chinese University of Hong Kong, Hong Kong, China

14. Department of Urology, Barts Health NHS Foundation Trust, London, UK

15. Department of Radiology, University Hospitals Birmingham NHS Foundation Trust, Birmingham, UK

16. Department of Radiology, Leeds Teaching Hospitals NHS Trust, Leeds, UK

17. School of Biomedical Engineering and Imaging Sciences, King’s College London, London, UK

Background

Multiparametric MRI (mpMRI), established by the PROMIS trial, is the pre-biopsy reference standard for detecting clinically significant prostate cancer (csPCa, ISUP Grade ≥2). The PRIME trial challenges this standard, supporting biparametric MRI (bpMRI) as a faster, lower-cost alternative with comparable diagnostic performance. Artificial intelligence (AI) may further improve detection and efficiency. In low- and lower-middle-income countries (L/LMICs), these efficiencies could reduce per-scan costs and expand MRI access, addressing diagnostic inequities. This review systematically maps AI models developed for csPCa detection using bpMRI and mpMRI, appraised via standard methodological frameworks, and models the direct imaging budget impact and theoretical system-capacity gains of switching to a bpMRI-based workflow.

Methods

Following PRISMA guidelines [PROSPERO: CRD420251037432], MEDLINE, PMC, EMBASE, SCOPUS, and COCHRANE were searched, yielding 6,389 records; 202 underwent full-text review and six studies met inclusion criteria. AI models were assessed using TRIPOD checklists. Deterministic health-system budget-impact and capacity models were developed using GLOBOCAN 2022 prostate cancer incidence data and MRI availability from IAEA. Conservative and realistic scenarios assumed 10% vs 25% MRI referral rates, $100 vs $200 per-scan savings, and one scan per patient, estimating total scans, direct annual savings, and theoretical capacity uplift.

Results

Six studies evaluated AI-assisted csPCa detection using bpMRI (PMID: 40259798, 36222324, 38876123) and mpMRI (PMID: 40016318, 37345961, 33671533). Pooled performance metrics and comparative standard metrics have been presented in Table 1. Conservative and realistic models estimates are depicted in Table 2, highlighting the additional scans per machine, corresponding to 7.35M extra scans/year at $300 per scan (~$2.2B value).

Conclusion

Biparametric MRI, with or without AI, matches mpMRI for csPCa detection and offers substantial efficiencies. Modelling suggests meaningful cost savings and large theoretical scanner capacity gains, which could be redirected to reduce wait times and expand access in L/LMICs, supporting equitable, high-throughput prostate cancer diagnostics.

References

1. Ahmed HU, El-Shater Bosaily A, Brown LC, Gabe R, Kaplan R, Parmar MK, Collaco-Moraes Y, Ward K, Hindley RG, Freeman A, Kirkham AP, Oldroyd R, Parker C, Emberton M; PROMIS study group. Diagnostic accuracy of multi-parametric MRI and TRUS biopsy in prostate cancer (PROMIS): a paired validating confirmatory study. Lancet. 2017 Feb 25;389(10071):815-822

2. Twilt JJ, Saha A, Bosma JS, Padhani AR, Bonekamp D, Giannarini G, van den Bergh R, Kasivisvanathan V, Obuchowski N, Yakar D, Elschot M, Veltman J, Fütterer J, Huisman H, de Rooij M; PI-CAI Consortium. AI-Assisted vs Unassisted Identification of Prostate Cancer in Magnetic Resonance Images. JAMA Netw Open. 2025 Jun 2;8(6):e2515672.

Table 1: Comparative Performance metrics of the AI-Assisted mpMRI and bpMRI study for csPCa Detection, with reference to PROMIS (PMID: 28110982) and PI-CAI (PMID: 40512493) Benchmarks.

Table 2: Budget Impact and Capacity Uplift Estimates using Conservative and Realistic Health-System Perspective Modelling