Evaluation of the possibility of using neural networks for automatic diagnostics of obstructive urination

封面

如何引用文章

全文:

开放存取 开放存取
受限制的访问 ##reader.subscriptionAccessGranted##
受限制的访问 订阅或者付费存取

详细

Introduction. Obstructive type of urination requires accurate and timely diagnosis to prevent complications and improve the quality of life of patients. Traditional diagnostic methods such as uroflowmetry, although they remain the standard, have their limitations. In this context, videography of urine stream followed by image analysis is a more cost-effective and promising approach that allows for a more detailed picture of urination, accessible not only to urologists, but also to patients.

Aim. To establish the possibility of recognizing and classifying graphs of the urination using neural network and machine learning technologies.

Materials and methods. This retrospective study involved 152 male patients aged 19 to 87 years who underwent examination and treatment at the MC Medassist clinic from June 2024 to January 2025. There were 43 patients (28%) with obstructive type of urination, 39 patients with benign prostatic hyperplasia, 4 patients with urethral stricture, and 109 patients (72%) with normal urination. The diagnostics algorithm included a general urinalysis, kidney and bladder ultrasound and/or MRI of the prostate, as well as uroflowmetry. The neural network architecture was designed based on the Keras framework of the Python programming language.

Results. Three studies with obtained data were carried out, which differed in the architecture of the neural network and the methods of preparing the initial data. The average area under the ROC curve for a network with random image feed averaged 0.5 for both the training and test samples. For a network with linear feed of the entire data set, it was 1 for the training and test samples. A neural network with three inputs differing in two-threshold binarization ranges showed a result of 0.9 for the training and 0.7 for the test sample.

Discussion. An important aspect of this study is the possibility of using neural networks to process large amounts of video data. Automating the analysis of images of urine stream allows not only to reduce the time required for diagnosis, but also to identify hidden patterns that may be overlooked during visual assessment by a specialist. The study of methods for analyzing video recordings of urination based on artificial neural network technologies and machine learning algorithms can become the basis for creating new diagnostic tools that will increase the speed of diagnosis, accelerate drug research, and monitor patients with chronic diseases.

Conclusion. Despite the current limitations, this study confirms that the use of neural networks and machine learning in urology has significant potential and can become the basis for the development of new diagnostic tools that can improve the efficiency of medical care, and thereby improve the quality of life of patients.

全文:

受限制的访问

作者简介

Alexandr Panferov

“Medassist” Medical Center; Southwest State University

Email: panferov-uro@yandex.ru
ORCID iD: 0000-0001-8258-3454

Ph.D., Head of the Urology Center of the “Medassist” Medical Center; Associate Professor of the Department of Biomedical Engineering, Faculty of Fundamental and Applied Informatics, Southwestern

俄罗斯联邦, Kursk; Kursk

Nariman Gadzhiev

St. Petersburg State University

Email: nariman.gadjiev@gmail.com
ORCID iD: 0000-0002-6255-0193

Ph.D., MD, Professor of the Department of Urology at the St. Petersburg State University Medical Institute, Deputy Director for Medical care (Urology) at the N.I. Pirogov Clinic of High Medical Technologies

 

俄罗斯联邦, St. Petersburg

Vitaly Yastrebov

“Medassist” Medical Center

编辑信件的主要联系方式.
Email: yastrebov.vetaly@yandex.ru
ORCID iD: 0000-0003-1388-4194

urologist of the Center of Urology

俄罗斯联邦, Kursk

Sergey Filist

Southwest State University

Email: sfilist@gmail.com
ORCID iD: 0000-0003-1358-671X

Ph.D. in technical sciences, Professor of the Department of Biomedical Engineering, Faculty of Fundamental and Applied Informatics

俄罗斯联邦, Kursk

Kirill Puchenkov

Southwest State University

Email: k.puchenkov@mail.ru
ORCID iD: 0009-0001-2677-5571

Ph.D. student of the Department of Biomedical Engineering, Faculty of Fundamental and Applied Informatics

俄罗斯联邦, Kursk

参考

  1. Gusev V.V., Ober’sheva I.A. Artificial intelligence in medicine: opportunities and prospects. Moscow, 2020. Russian (Гусев В.В., Ожерельева И.А. Искусственный интеллект в медицине: возможности и перспективы. М., 2020).
  2. Goodfellow I., Bengio Y., Courville A. Deep learning. MIT Press. 2016, 800 pp, ISBN: 0262035618.
  3. LeCun Y., Bengio Y., Hinton G. Deep learning. Nature. 2015;521(7553):436–44. doi: 10.1038/nature14539
  4. Flores-Mireles A.L., Walker J.N., Caparon M., Hultgren S.J. Urinary tract infections: epidemiology, mechanisms of infection and treatment options. Nat Rev Microbiol. 2015;13(5):269–84. doi: 10.1038/nrmicro3432.
  5. Madersbacher S., Dmochowski R. New technologies and strategies in the management of lower urinary tract dysfunction. European Urology. 2007;52(5):1227–1242. doi: 10.1016/j.eururo.2007.08.015
  6. Wein A.J. et al. Campbell-Walsh Urology. 11th Edition. Elsevier. 2016.
  7. Hashim H., Abrams P. Is the bladder a reliable witness for predicting detrusor overactivity? The Journal of Urology. 2006;175(1):191–194. doi: 10.1016/S0022-5347(05)00558-7
  8. Blaivas J.G., Groutz A. Bladder outlet obstruction nomogram for women with lower urinary tract symptomatology. Neurourology and Urodynamics. 2000;19(5):553–564. doi: 10.1002/1520-6777(2000)19:5<553::AID-NAU1>3.0.CO;2-8
  9. Jeong S.J. et al. Correlation between uroflowmetry parameters and International Prostate Symptom Score in patients with benign prostatic hyperplasia. Korean Journal of Urology. 2012;53(6):410–413. doi: 10.4111/kju.2012.53.6.410
  10. Rosier P.F. et al. Video-urodynamic studies: a comprehensive guide for clinicians. European Urology. 2016;69(3):501–512. doi: 10.1016/j.eururo.2015.12.021
  11. Kuo H.C. Clinical application of videourodynamics in the diagnosis and treatment of lower urinary tract dysfunction. Urological Science. 2003;14(1):1–7.
  12. Watanabe H. et al. Non-invasive diagnostic method for prostatic hypertrophy using ultrasonic imaging. Urologia Internationalis. 1980;35(3):167–175.
  13. Gonzalez R.C., Woods R.E. Digital Image Processing. Pearson. 2018.
  14. Szeliski R. Computer Vision: Algorithms and Applications. Springer, 2010. doi: 10.1007/978-1-84882-935-0
  15. Ronneberger O., Fischer P., Brox T. U-Net: Convolutional networks for biomedical image segmentation. Medical Image Computing and Computer-Assisted Intervention (MICCAI). 2015;234–241. doi: 10.1007/978-3-319-24574-4_28
  16. Litjens G., Kooi T., Bejnordi B.E., Setio A.A.A., Ciompi F., Ghafoorian M., van der Laak JAWM, van Ginneken B., Sánchez C.I. A survey on deep learning in medical image analysis. Med Image Anal. 2017;42:60–88. doi: 10.1016/j.media.2017.07.005
  17. Esteva A. et al. Dermatologist-level classification of skin cancer with deep neural networks. Nature. 2017;542(7639):115–118. doi: 10.1038/nature21056
  18. Shen D. et al. Deep learning in medical image analysis. Annual Review of Biomedical Engineering. 2017;19:221–248. doi: 10.1146/annurev-bioeng-071516-044442
  19. Otsu N. A threshold selection method from gray-level histograms. IEEE Transactions on Systems, Man, and Cybernetics. 1979;9(1):62–66. doi: 10.1109/TSMC.1979.4310076
  20. Chapple C.R. et al. The role of urodynamics in the evaluation of lower urinary tract symptoms. European Urology Focus. 2017;3(1):54–61. doi: 10.1016/j.euf.2016.07.002

补充文件

附件文件
动作
1. JATS XML
2. Fig. 1. Test bench diagram

下载 (1MB)
3. Fig. 2. Model architecture for the first study

下载 (1MB)
4. Fig. 3. ROC curve obtained from the training sample for the model with random image feed

下载 (1MB)
5. Fig. 4. ROC curve obtained from the test sample for the model with random image feed

下载 (1MB)
6. Fig. 5. ROC curve obtained from the training sample for the model with linear data feed

下载 (899KB)
7. Fig. 6. ROC curve obtained from the test sample for the model with linear data feed

下载 (2MB)
8. Fig. 7. Two-threshold binarization of images, where a) is the threshold from 10 to 70, b) is from 71 to 127, c) is from 140 to 255

下载 (1MB)
9. Fig. 8. Architecture of a model with three inputs with different ranges of two-threshold binarization

下载 (2MB)
10. Fig. 9. ROC curve obtained from the training sample for a model with three inputs and linear data feed

下载 (1MB)
11. Fig. 10. ROC curve obtained from the test sample for a model with three inputs and linear data feed

下载 (1MB)
12. Images obtained as a result of video shooting

下载 (202KB)

版权所有 © Bionika Media, 2025