A pilot study of AI-assisted reading of prostate MRI in Organized Prostate Cancer Testing
DOI:
https://doi.org/10.2340/1651-226X.2024.40475Keywords:
Prostate cancer, screening, artificial intelligence, MRI prostateAbstract
Objectives: To evaluate the feasibility of AI-assisted reading of prostate magnetic resonance imaging (MRI) in Organized Prostate cancer Testing (OPT).
Methods: Retrospective cohort study including 57 men with elevated prostate-specific antigen (PSA) levels ≥3 µg/L that performed bi-parametric MRI in OPT. The results of a CE-marked deep learning (DL) algorithm for prostate MRI lesion detection were compared with assessments performed by on-site radiologists and reference radiologists. Per patient PI-RADS (Prostate Imaging-Reporting and Data System)/Likert scores were cross-tabulated and compared with biopsy outcomes, if performed. Positive MRI was defined as PI-RADS/Likert ≥4. Reader variability was assessed with weighted kappa scores.
Results: The number of positive MRIs was 13 (23%), 8 (14%), and 29 (51%) for the local radiologists, expert consensus, and DL, respectively. Kappa scores were moderate for local radiologists versus expert consensus 0.55 (95% confidence interval [CI]: 0.37–0.74), slight for local radiologists versus DL 0.12 (95% CI: −0.07 to 0.32), and slight for expert consensus versus DL 0.17 (95% CI: −0.01 to 0.35). Out of 10 cases with biopsy proven prostate cancer with Gleason ≥3+4 the DL scored 7 as Likert ≥4.
Interpretation: The Dl-algorithm showed low agreement with both local and expert radiologists. Training and validation of DL-algorithms in specific screening cohorts is essential before introduction in organized testing.
Downloads
References
Hugosson J, Månsson M, Wallström J, Axcrona U, Carlsson SV, Egevad L, et al. Prostate cancer screening with PSA and MRI followed by targeted biopsy only. N Engl J Med. 2022;387:2126–37.
https://doi.org/10.1056/NEJMoa2209454
Winkel DJ, Tong A, Lou B, Kamen A, Comaniciu D, Disselhorst JA, et al. A novel deep learn-ing based computer-aided diagnosis system improves the accuracy and effi-ciency of radiologists in reading biparametric magnetic resonance images of the prostate: results of a multireader, multicase study. Invest Radiol. 2021;56:605–13.
https://doi.org/10.1097/RLI.0000000000000780
Turkbey B, Haider MA. Deep learning-based artificial intelligence applications in prostate MRI: brief summary. Br J Radiol. 2022;95:20210563.
https://doi.org/10.1259/bjr.20210563
Válek, V. Council recommendation of 9 December 2022 on strengthening prevention through early detection: a new EU approach on cancer screening. Off J Eur Un-ion. C, 2022, 473..
Alterbeck M, Järbur E, Thimansson E, Wallström J, Bengtsson J, Björk-Eriksson T, et al. Designing and implementing a population-based organised prostate cancer testing programme. Eur Urol Focus. 2022;8(6):1568-74.
https://doi.org/10.1016/j.euf.2022.06.008
Alterbeck M, Thimansson E, Bengtsson J, Baubeta E, Zackrisson S, Bolejko A, et al. A pilot study of an organised population-based testing programme for prostate cancer. Eur Radiol. 2023;33(4):2519-28.
https://doi.org/10.1111/bju.16143
Turkbey B, Rosenkrantz AB, Haider MA, Padhani AR, Villeirs G, Macura KJ, et al. Prostate imaging reporting and data system Version 2.1: 2019 update of prostate imaging reporting and data system Version 2. Eur Urol. 2019;76:340–51.
https://doi.org/10.1016/j.eururo.2019.02.033
Winkel DJ. A fully automated, end-to-end prostate MRI workflow solution incorporating dot, ultrashort biparametric imaging and deeplearning-based detection, classi-fication and reporting. Magnetom Flash (76) 1/202. 2020. [Cited date: 15/03/2024] Available from: https://www.magnetomworld.siemens-healthineers.com/publications/magnetom-flash Siemens Healthineers
Yang D, Xu D, Zhou SK, Georgescu B, Chen M, Grbic S, et al. Automatic liver segmentation using an adversarial image-to-image network. In: International conference on medical image computing and computer-
assisted intervention. Descoteaux M, Maier-Hein L, Franz A, Jannin P, Collins DL, Duchesne S, Eds. Springer International Publishing: Cham, Switzerland; 2017. p. 507–515.
Centres RaCoRC. Swedish National Clinical Cancer Care guidelines for prostate cancer. [Cited date: 15/03/2024] Available in swedish from: https://kunskapsbanken.cancercentrum.se/diagnoser/prostatacancer/vardprogram
Winkel DJ, Wetterauer C, Matthias MO, Lou B, Shi B, Kamen A, et al. Autonomous detec-tion and classification of PI-RADS lesions in an MRI screening population incor-porating multicenter-labeled deep learning and biparametric imaging: proof of concept. Diagnostics (Basel). 2020;10: 951-65.
https://doi.org/10.3390/diagnostics10110951
Sanford T, Harmon SA, Turkbey EB, Kesani D, Tuncer S, Madariaga M, et al. Deep-learning-based artificial intelligence for PI-RADS classification to assist multiparametric prostate MRI interpretation: a development study. J Magn Reson Imaging. 2020;52:1499–507.
https://doi.org/10.1002/jmri.27204
Schelb P, Kohl S, Radtke JP, Wiesenfarth M, Kickingereder P, Bickelhaupt S, et al. Classifi-cation of cancer at prostate MRI: deep learning versus clinical PI-RADS assess-ment. Radiology. 2019;293:607–17.
https://doi.org/10.1148/radiol.2019190938
Rosenkrantz AB, Ginocchio LA, Cornfeld D, Froemming AT, Gupta RT, Turkbey B, et al. In-terobserver reproducibility of the PI-RADS Version 2 Lexicon: a multicenter study of six experienced prostate radiologists. Radiology. 2016;280:93–804.
https://doi.org/10.1148/radiol.2016152542
Muller BG, Shih JH, Sankineni S, Marko J, Rais-Bahrami S, George AK, et al. Prostate can-cer: interobserver agreement and accuracy with the revised prostate imaging re-porting and data system at multiparametric MR imaging. Radiology. 2015;277:741–50.
https://doi.org/10.1148/radiol.2015142818
de Rooij M, Israël B, Tummers M, Ahmed HU, Barrett T, Giganti F, et al. ESUR/ESUI consen-sus statements on multi-parametric MRI for the detection of clinically significant prostate cancer: quality requirements for image acquisition, interpretation and radiologists’ training. Eur Radiol. 2020;30:5404–16.
https://doi.org/10.1007/s00330-020-06929-z
Nordström T, Discacciati A, Bergman M, Clements M, Aly M, Annerstedt M, et al. Prostate cancer screening using a combination of risk-prediction, MRI, and targeted prostate biopsies (STHLM3-MRI): a prospective, population-based, randomised, open-label, non-inferiority trial. Lancet Oncol. 2021;22:1240–9.
Downloads
Additional Files
Published
How to Cite
Issue
Section
Categories
License
Copyright (c) 2023 Erik Thimansson, Sophia Zackrisson, Fredrik Jäderling, Max Alterbeck , Thomas Jiborn , Anders Bjartell , Jonas Wallström
This work is licensed under a Creative Commons Attribution 4.0 International License.
