AI-assisted cognitive targeted biopsy: a new approach to prostate cancer diagnosis
Urology reports (St. - Petersburg) · 10.07.2026 · 16 (2) · 191-199
DOI: 10.17816/uroved703712 ↗Kurz gefasst
Fachartikel in der Zeitschrift Urology reports (St. - Petersburg), online erschienen am 10.07.2026. Verfasst von Marlen E. Topuzov und 5 weiteren Autor:innen. Die Arbeit nennt 6 Literaturangaben.
Abstract (Originalsprache)
BACKGROUND: Prostate cancer ranks second in prevalence among male malignancies and is diagnosed via biopsy. However, the standard 12-core systematic biopsy often demonstrates low accuracy. Advancing targeted navigation techniques is a key factor in improving the detection of clinically significant forms of the disease. AIM: To develop a new strategy for cognitive targeted biopsy using artificial intelligence (AI-COG-TB) and to conduct a comparative analysis of the diagnostic performance of various prostate cancer detection methods. METHODS: This prospective multicenter study included 233 patients with prostate-specific antigen 2.5 ng/mL and Prostate Imaging-Reporting and Data System ≥ 3. Five approaches were compared: transrectal ultrasound-guided systematic biopsy (TRUS-TB, n = 50), cognitive biopsy (COG-TB, n = 40), MRI-transrectal ultrasound fusion-guided biopsy (FUS-TB, n = 49), in-bore magnetic resonance imaging-guided biopsy (IB-TB, n = 49), and cognitive biopsy with artificial intelligence-assisted segmentation (AI-COG-TB, n = 45). Results were evaluated based on the cancer detection rate and risk stratification according to the International Society of Urological Pathology (ISUP) groups. RESULTS: The overall prostate cancer detection rate was 98% for in-bore magnetic resonance imaging-guided biopsy (IB-TB), 91.8% for fusion-guided biopsy (FUS-TB), and 88.9% for cognitive biopsy with artificial intelligence-assisted segmentation (AI-COG-TB), which significantly exceeded the rates for systematic biopsy (TRUS-TB: 78%) and cognitive biopsy (COG-TB: 75%). The in-bore magnetic resonance imaging-guided biopsy (IB-TB) demonstrated the highest efficacy in detecting clinically significant prostate cancer: 75.5% of cases, with the lowest proportion of indolent forms (24.5%). In a multivariate binary logistic regression model, the strongest predictors of cancer detection compared to the standard were in-bore magnetic resonance imaging-guided biopsy (IB-TB) (odds ratio [OR] 26.01; p = 0.003), fusion-guided biopsy (FUS-TB) (OR 5.16; p = 0.015), and cognitive biopsy with artificial intelligence-assisted segmentation (AI-COG-TB) (OR 3.56; p = 0.046). Cognitive biopsy without AI support (COG-TB) showed no statistically significant advantages over the systematic method (p = 0.771). CONCLUSION: The AI-COG-TB technique serves as an effective and economically accessible alternative to fusion biopsy (FUS-TB), significantly improving diagnostic accuracy compared to cognitive biopsy (COG-TB). However, in-bore magnetic resonance imaging-guided biopsy (IB-TB) remains the method of choice for identifying aggressive cancer forms and minimizing the overdiagnosis of indolent tumors.
Zitieren
APA
Topuzov, M. E., Vodopyan, S. S., Kuplevatskaya, D. I., Topuzov, T. M., Atamas, M. M., & Abuzarova, S. F. (2026). AI-assisted cognitive targeted biopsy: a new approach to prostate cancer diagnosis. Urology reports (St. - Petersburg), 16(2), 191-199. https://doi.org/10.17816/uroved703712
Vancouver
Topuzov ME, Vodopyan SS, Kuplevatskaya DI, Topuzov TM, Atamas MM, Abuzarova SF. AI-assisted cognitive targeted biopsy: a new approach to prostate cancer diagnosis. Urology reports (St. - Petersburg). 2026;16(2):191-199. doi:10.17816/uroved703712
BibTeX
@article{Topuzov2026,
author = {Topuzov, Marlen E. and Vodopyan, Sergey S. and Kuplevatskaya, Daria I. and Topuzov, Timur M. and Atamas, Mikhail M. and Abuzarova, Sabina F.},
title = {AI-assisted cognitive targeted biopsy: a new approach to prostate cancer diagnosis},
journal = {Urology reports (St. - Petersburg)},
year = {2026},
volume = {16},
number = {2},
pages = {191--199},
doi = {10.17816/uroved703712}
}
EUVIMED