3D-AttenNet Model Can Predict Clinically Significant Prostate Cancer in PI-RADS Category 3 Patients: a Retrospective Multicenter Study

Overview

Journal Insights Imaging

Publisher Springer

Date 2025 Jan 29

PMID 39881076

Authors

Jie Bao

Litao Zhao

Xiaomeng Qiao

Zhenkai Li

Yanting Ji

Yueting Su

Libiao Ji

Junkang Shen

Jiangang Liu

Jie Tian

Ximing Wang

Hailin Shen

Chunhong Hu

Affiliations

Soon will be listed here.

Abstract

Purposes: The presence of clinically significant prostate cancer (csPCa) is equivocal for patients with prostate imaging reporting and data system (PI-RADS) category 3. We aim to develop deep learning models for re-stratify risks in PI-RADS category 3 patients.

Methods: This retrospective study included a bi-parametric MRI of 1567 consecutive male patients from six centers (Centers 1-6) between Jan 2015 and Dec 2020. Deep learning models with double channel attention modules based on MRI (AttenNet) for predicting PCa and csPCa were constructed separately. Each model was first pretrained using 1144 PI-RADS 1-2 and 4-5 images and then retrained using 238 PI-RADS 3 images from three training centers (centers 1-3), and tested using 185 PI-RADS 3 images from the other three testing centers (centers 4-6).

Results: Our AttenNet models achieved excellent prediction performances in testing cohort of center 4-6 with the area under the receiver operating characteristic curves (AUC) of 0.795 (95% CI: [0.700, 0.891]), 0.963 (95% CI: [0.915, 1]) and 0.922 (95% CI: [0.810, 1]) in predicting PCa, and the corresponding AUCs were 0.827 (95% CI: [0.703, 0.952]) and 0.926 (95% CI: [0.846, 1]) in predicting csPCa in testing cohort of center 4 and center 5. In particular, 71.1% to 92.2% of non-csPCa patients were identified by our model in three testing cohorts, who can spare from invasive biopsy or RP procedure.

Conclusions: Our model offers a noninvasive screening clinical tool to re-stratify risks in PI-RADS 3 patients, thereby reducing unnecessary invasive biopsies and improving the effectiveness of biopsies.

Critical Relevance Statement: The deep learning model with MRI can help to screen out csPCa in PI-RADS category 3.

Key Points: AttenNet models included channel attention and soft attention modules. 71.1-92.2% of non-csPCa patients were identified by the AttenNet model. The AttenNet models can be a screen clinical tool to re-stratify risks in PI-RADS 3 patients.

References

Netzer N, Weisser C, Schelb P, Wang X, Qin X, Gortz M . Fully Automatic Deep Learning in Bi-institutional Prostate Magnetic Resonance Imaging: Effects of Cohort Size and Heterogeneity. Invest Radiol. 2021; 56(12):799-808. DOI: 10.1097/RLI.0000000000000791. View

Ahmed H, El-Shater Bosaily A, Brown L, Gabe R, Kaplan R, Parmar M . Diagnostic accuracy of multi-parametric MRI and TRUS biopsy in prostate cancer (PROMIS): a paired validating confirmatory study. Lancet. 2017; 389(10071):815-822. DOI: 10.1016/S0140-6736(16)32401-1. View

Ahdoot M, Wilbur A, Reese S, Lebastchi A, Mehralivand S, Gomella P . MRI-Targeted, Systematic, and Combined Biopsy for Prostate Cancer Diagnosis. N Engl J Med. 2020; 382(10):917-928. PMC: 7323919. DOI: 10.1056/NEJMoa1910038. View

Lim C, Abreu-Gomez J, Thornhill R, James N, Al Kindi A, Lim A . Utility of machine learning of apparent diffusion coefficient (ADC) and T2-weighted (T2W) radiomic features in PI-RADS version 2.1 category 3 lesions to predict prostate cancer diagnosis. Abdom Radiol (NY). 2021; 46(12):5647-5658. DOI: 10.1007/s00261-021-03235-0. View

Hiremath A, Shiradkar R, Fu P, Mahran A, Rastinehad A, Tewari A . An integrated nomogram combining deep learning, Prostate Imaging-Reporting and Data System (PI-RADS) scoring, and clinical variables for identification of clinically significant prostate cancer on biparametric MRI: a retrospective multicentre study. Lancet Digit Health. 2021; 3(7):e445-e454. PMC: 8261599. DOI: 10.1016/S2589-7500(21)00082-0. View

Li T, Sun L, Li Q, Luo X, Luo M, Xie H . Development and Validation of a Radiomics Nomogram for Predicting Clinically Significant Prostate Cancer in PI-RADS 3 Lesions. Front Oncol. 2022; 11:825429. PMC: 8825569. DOI: 10.3389/fonc.2021.825429. View

Shao W, Banh L, Kunder C, Fan R, Soerensen S, Wang J . ProsRegNet: A deep learning framework for registration of MRI and histopathology images of the prostate. Med Image Anal. 2021; 68:101919. PMC: 7856244. DOI: 10.1016/j.media.2020.101919. View

Kuhl C, Bruhn R, Kramer N, Nebelung S, Heidenreich A, Schrading S . Abbreviated Biparametric Prostate MR Imaging in Men with Elevated Prostate-specific Antigen. Radiology. 2017; 285(2):493-505. DOI: 10.1148/radiol.2017170129. View

Mayerhoefer M, Materka A, Langs G, Haggstrom I, Szczypinski P, Gibbs P . Introduction to Radiomics. J Nucl Med. 2020; 61(4):488-495. PMC: 9374044. DOI: 10.2967/jnumed.118.222893. View

10.

Midiri F, Vernuccio F, Purpura P, Alongi P, Bartolotta T . Multiparametric MRI and Radiomics in Prostate Cancer: A Review of the Current Literature. Diagnostics (Basel). 2021; 11(10). PMC: 8534893. DOI: 10.3390/diagnostics11101829. View

11.

Schelb P, Kohl S, Radtke J, Wiesenfarth M, Kickingereder P, Bickelhaupt S . Classification of Cancer at Prostate MRI: Deep Learning versus Clinical PI-RADS Assessment. Radiology. 2019; 293(3):607-617. DOI: 10.1148/radiol.2019190938. View

12.

Naik N, Tokas T, Shetty D, Hameed B, Shastri S, Shah M . Role of Deep Learning in Prostate Cancer Management: Past, Present and Future Based on a Comprehensive Literature Review. J Clin Med. 2022; 11(13). PMC: 9267773. DOI: 10.3390/jcm11133575. View

13.

Stabile A, Giganti F, Rosenkrantz A, Taneja S, Villeirs G, Gill I . Multiparametric MRI for prostate cancer diagnosis: current status and future directions. Nat Rev Urol. 2019; 17(1):41-61. DOI: 10.1038/s41585-019-0212-4. View

14.

Mottet N, Bellmunt J, Bolla M, Briers E, Cumberbatch M, De Santis M . EAU-ESTRO-SIOG Guidelines on Prostate Cancer. Part 1: Screening, Diagnosis, and Local Treatment with Curative Intent. Eur Urol. 2016; 71(4):618-629. DOI: 10.1016/j.eururo.2016.08.003. View

15.

Mohler J, Antonarakis E, Armstrong A, DAmico A, Davis B, Dorff T . Prostate Cancer, Version 2.2019, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw. 2019; 17(5):479-505. DOI: 10.6004/jnccn.2019.0023. View

16.

Israel B, van der Leest M, Sedelaar M, Padhani A, Zamecnik P, Barentsz J . Multiparametric Magnetic Resonance Imaging for the Detection of Clinically Significant Prostate Cancer: What Urologists Need to Know. Part 2: Interpretation. Eur Urol. 2019; 77(4):469-480. DOI: 10.1016/j.eururo.2019.10.024. View

17.

Jin P, Yang L, Qiao X, Hu C, Hu C, Wang X . Utility of Clinical-Radiomic Model to Identify Clinically Significant Prostate Cancer in Biparametric MRI PI-RADS V2.1 Category 3 Lesions. Front Oncol. 2022; 12:840786. PMC: 8913337. DOI: 10.3389/fonc.2022.840786. View

18.

Michaely H, Aringhieri G, Cioni D, Neri E . Current Value of Biparametric Prostate MRI with Machine-Learning or Deep-Learning in the Detection, Grading, and Characterization of Prostate Cancer: A Systematic Review. Diagnostics (Basel). 2022; 12(4). PMC: 9027206. DOI: 10.3390/diagnostics12040799. View

19.

Hou Y, Bao M, Wu C, Zhang J, Zhang Y, Shi H . A radiomics machine learning-based redefining score robustly identifies clinically significant prostate cancer in equivocal PI-RADS score 3 lesions. Abdom Radiol (NY). 2020; 45(12):4223-4234. DOI: 10.1007/s00261-020-02678-1. View

20.

Zhao L, Bao J, Qiao X, Jin P, Ji Y, Li Z . Predicting clinically significant prostate cancer with a deep learning approach: a multicentre retrospective study. Eur J Nucl Med Mol Imaging. 2022; 50(3):727-741. PMC: 9852176. DOI: 10.1007/s00259-022-06036-9. View