Clinical applications of machine learning for urolithiasis and benign prostatic hyperplasia: A systematic review
David Bouhadana1, Xing Han Lu2, Wei Xi Luo1, Anis Assad3, Claudia Deyirmendjian4, Abbas Guennoun3, David-Dan Nguyen1, Bilal Chughtai5, Dean Elterman6, Kevin Zorn3, Naeem Bhojani3.
1Faculty of Medicine, McGill University, Montreal, QC, Canada; 2School of Computer Science, McGill University, Montreal, QC, Canada; 3Division of Urology, Centre Hospitalier de l'Université de Montréal , Montreal, QC, Canada; 4Faculty of Medicine, Université de Montréal, Montreal, QC, Canada; 5Department of Urology, Weill Cornell Medical College/New York Presbyterian, New York, NY, United States; 6Division of Urology, Deptartment of Surgery, University of Toronto, Toronto, ON, Canada
Introduction: Currently, studies have looked at the use of machine learning (ML) in aiding with the diagnosis, outcome prediction, and management of urological conditions.1-3 Previous systematic reviews related to ML in urology failed to thoroughly review the literature related to endourology.4-6 Therefore, we aim to conduct a more focused systematic review examining the use of ML algorithms for the management of benign prostatic hyperplasia (BPH) or urolithiasis.
Methods: Searches of MEDLINE, Embase, and the Cochrane CENTRAL databases were conducted from inception through July 12, 2021. Keywords included those related to ML, endourology, urolithiasis, and BPH. Original research articles were included without any language restrictions. Two reviewers screened the citations that were eligible for title, abstract, and full-text screening, with conflicts resolved by consensus. Two reviewers extracted information from the studies, with discrepancies resolved by consensus. The data collected was then qualitatively synthesized.
Results: After identifying 459 unique citations, 63 articles were retained for data extraction. Most articles consisted of tabular (n=23) and computer vision (n=31) tasks. The two most common problem types were classification (n=40) and regression (n=12). In general, most studies used neural networks as their ML algorithm (n=36). Of the 63 studies retrieved, 58 were related to urolithiasis and five to BPH. The urolithiasis studies were designed for outcome prediction (n=21), disease classification (n=15), predicting disease occurrence (n=11), diagnostics (n=8), and therapeutics (n=3). The BPH studies helped with outcome prediction (n=2), diagnostics (n=2), and treatment of BPH (n=1).
Conclusions: The majority of the retrieved studies successfully helped with outcome prediction, disease classification, disease prediction, diagnostics, and therapeutics for both urolithiasis and BPH. While ML shows great promise in improving patient care, it is important to adhere to the recently developed STREAM-URO guidelines to ensure the development of high-quality ML studies.
[1] Chen J, Remulla D, Nguyen JH, et al. Current status of artificial intelligence applications in urology and their potential to influence clinical practice. BJU International 2019;124:567-577.
[2] Bertolo R, Hung A, Porpiglia F, et al. Systematic review of augmented reality in urological interventions: the evidences of an impact on surgical outcomes are yet to come. World journal of urology 2020;38:2167-2176.
[3] Lucas M, Liem EI, Savci-Heijink CD, et al. Toward automated in vivo bladder tumor stratification using confocal laser endomicroscopy. Journal of endourology 2019;33:930-937.
[4] Checcucci E, De Cillis S, Granato S, et al. Applications of neural networks in urology: a systematic review. Current Opinion in Urology 2020;30:788-807.
[5] Shah M, Naik N, Somani BK, et al. Artificial intelligence (AI) in urology-Current use and future directions: An iTRUE study. Turkish Journal of Urology 2020;46:S27-S39.
[6] Hameed B, Shah M, Naik N, et al. The Ascent of Artificial Intelligence in Endourology: a Systematic Review Over the Last 2 Decades. Current Urology Reports 2021;22:1-18.
