An Automatic Diagnostic System for the Urodynamic Study Applying in Lower Urinary Tract Dysfunction

Overview

Journal Int Urol Nephrol

Publisher Springer

Specialty Nephrology

Date 2023 Sep 27

PMID 37755608

Authors

Zehua Ding

Weiyu Zhang

Huanrui Wang

Hanwei Ke

Dongyu Su

Qi Wang

Kaigui Bian

Feng Su

Kexin Xu

Affiliations

Soon will be listed here.

Abstract

Objective: To establish an automatic diagnostic system based on machine learning for preliminarily analysis of urodynamic study applying in lower urinary tract dysfunction (LUTD).

Methods: The eight most common conditions of LUTDs were included in the present study. A total of 527 eligible patients with complete data, from the year of 2015 to 2020, were enrolled in this study. In total, two global parameters (patients' age and sex) and 13 urodynamic parameters were considered to be the input for machine learning algorithms. Three machine learning approaches were applied and evaluated in this study, including Decision Tree (DT), Logistic Regression (LR), and Support Vector Machine (SVM).

Results: By applying machine learning algorithms into the 8 common LUTDs, the DT models achieved the AUC of 0.63-0.98, the LR models achieved the AUC of 0.73-0.99, and the SVM models achieved the AUC of 0.64-1.00. For mutually exclusive diagnoses of underactive detrusor and acontractile detrusor, we developed a classification model that classifies the patients into either of these two diseases or double-negative class. For this classification method, the DT models achieved the AUC of 0.82-0.85 and the SVM models achieved the AUC of 0.86-0.90. Among all these models, the LR and the SVM models showed better performance. The best model of these diagnostic tasks achieved an average AUC of 0.90 (0.90 ± 0.08).

Conclusions: An automatic diagnostic system was developed using three machine learning models in urodynamic studies. This automated machine learning process could lead to promising assistance and enhancements of diagnosis and provide more useful reference for LUTD treatment.

References

Rosier P, Schaefer W, Lose G, Goldman H, Guralnick M, Eustice S . International Continence Society Good Urodynamic Practices and Terms 2016: Urodynamics, uroflowmetry, cystometry, and pressure-flow study. Neurourol Urodyn. 2016; 36(5):1243-1260. DOI: 10.1002/nau.23124. View

Wessels F, Schmitt M, Krieghoff-Henning E, Jutzi T, Worst T, Waldbillig F . Deep learning approach to predict lymph node metastasis directly from primary tumour histology in prostate cancer. BJU Int. 2021; 128(3):352-360. DOI: 10.1111/bju.15386. View

Lebret T, Pignot G, Colombel M, Guy L, Rebillard X, Savareux L . Artificial intelligence to improve cytology performances in bladder carcinoma detection: results of the VisioCyt test. BJU Int. 2021; 129(3):356-363. DOI: 10.1111/bju.15382. View

Goecks J, Jalili V, Heiser L, Gray J . How Machine Learning Will Transform Biomedicine. Cell. 2020; 181(1):92-101. PMC: 7141410. DOI: 10.1016/j.cell.2020.03.022. View

Minchole A, Rodriguez B . Artificial intelligence for the electrocardiogram. Nat Med. 2019; 25(1):22-23. DOI: 10.1038/s41591-018-0306-1. View

Hogan S, Gammie A, Abrams P . Urodynamic features and artefacts. Neurourol Urodyn. 2012; 31(7):1104-17. DOI: 10.1002/nau.22209. View

Birkhauser V, Sartori A, Bodmer N, Conlon T, Thiel J, Kessler T . Metaepidemiological Inventory of Diagnostic Studies on Urodynamics. Eur Urol Focus. 2019; 6(5):880-908. DOI: 10.1016/j.euf.2019.11.017. View

Raz O, Tse V, Chan L . Urodynamic testing: physiological background, setting-up, calibration and artefacts. BJU Int. 2014; 114 Suppl 1:22-8. DOI: 10.1111/bju.12633. View

Bentellis I, Guerin S, Khene Z, Khavari R, Peyronnet B . Artificial intelligence in functional urology: how it may shape the future. Curr Opin Urol. 2021; 31(4):385-390. DOI: 10.1097/MOU.0000000000000888. View

10.

Sonke G, Heskes T, Verbeek A, de la Rosette J, Kiemeney L . Prediction of bladder outlet obstruction in men with lower urinary tract symptoms using artificial neural networks. J Urol. 1999; 163(1):300-5. View

11.

Wadie B, Badawi A, Abdelwahed M, Elemabay S . Application of artificial neural network in prediction of bladder outlet obstruction: a model based on objective, noninvasive parameters. Urology. 2006; 68(6):1211-4. DOI: 10.1016/j.urology.2006.08.1079. View

12.

Niederhauser T, Gafner E, Cantieni T, Gramiger M, Haeberlin A, Obrist D . Detection and quantification of overactive bladder activity in patients: Can we make it better and automatic?. Neurourol Urodyn. 2017; 37(2):823-831. DOI: 10.1002/nau.23357. View

13.

Wang H, Cahill D, Panagides J, Nelson C, Wu H, Estrada C . Pattern recognition algorithm to identify detrusor overactivity on urodynamics. Neurourol Urodyn. 2020; 40(1):428-434. DOI: 10.1002/nau.24578. View

14.

Cullingsworth Z, Kelly B, Deebel N, Colhoun A, Nagle A, Klausner A . Automated quantification of low amplitude rhythmic contractions (LARC) during real-world urodynamics identifies a potential detrusor overactivity subgroup. PLoS One. 2018; 13(8):e0201594. PMC: 6093663. DOI: 10.1371/journal.pone.0201594. View

15.

Rantell A, Dolan L, Bonner L, Knight S, Ramage C, Toozs-Hobson P . Minimum standards for continence care in the UK. Neurourol Urodyn. 2015; 35(3):400-6. DOI: 10.1002/nau.22717. View

16.

Panicker J . Neurogenic Bladder: Epidemiology, Diagnosis, and Management. Semin Neurol. 2020; 40(5):569-579. PMC: 9715349. DOI: 10.1055/s-0040-1713876. View

17.

Stohrer M, Blok B, Castro-Diaz D, Chartier-Kastler E, Del Popolo G, Kramer G . EAU guidelines on neurogenic lower urinary tract dysfunction. Eur Urol. 2009; 56(1):81-8. DOI: 10.1016/j.eururo.2009.04.028. View