Abstract
Background
Accurate estimation of the glomerular filtration rate (GFR) is clinically crucial for determining the status of obstruction, developing treatment strategies, and predicting prognosis in obstructive nephropathy (ON). We aimed to develop a deep learning-based system, named UroAngel, for non-invasive and convenient prediction of single-kidney function level.
Methods
We retrospectively collected computed tomography urography (CTU) images and emission computed tomography diagnostic reports of 520 ON patients. A 3D U-Net model was used to segment the renal parenchyma, and a logistic regression multi-classification model was used to predict renal function level. We compared the predictive performance of UroAngel with the Modification of Diet in Renal Disease (MDRD), Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equations, and two expert radiologists in an additional 40 ON patients to validate clinical effectiveness.
Results
UroAngel based on 3D U-Net convolutional neural network could segment the renal cortex accurately, with a Dice similarity coefficient of 0.861. Using the segmented renal cortex to predict renal function stage had high performance with an accuracy of 0.918, outperforming MDRD and CKD-EPI and two radiologists.
Conclusions
We proposed an automated 3D U-Net-based analysis system for direct prediction of single-kidney function stage from CTU images. UroAngel could accurately predict single-kidney function in ON patients, providing a novel, reliable, convenient, and non-invasive method.
Similar content being viewed by others
Data availability
The data of this study are protected by the Ethics Committee and cannot be published. The codes and supplementary materials involved can be obtained by contacting the corresponding authors upon reasonable request.
References
Truong LD, Gaber L, Eknoyan G (2011) Obstructive uropathy. Contrib Nephrol 169:311–326. https://doi.org/10.1159/000314578
Stevens S (2018) Obstructive kidney disease. Nurs Clin N Am 53(4):569–578. https://doi.org/10.1016/j.cnur.2018.07.007
Renard-Penna R, Rocher L, Roy C, André M, Bellin MF, Boulay I, Eiss D, Girouin N, Grenier N, Hélénon O, Lapray JF, Lefèvre A, Matillon X, Ménager JM, Millet I, Ronze S, Sanzalone T, Tourniaire J, Brunelle S, Rouvière O (2020) Imaging protocols for CT urography: results of a consensus conference from the French Society of Genitourinary Imaging. Eur Radiol 30(3):1387–1396. https://doi.org/10.1007/s00330-019-06529-6
Keramida G, James JM, Prescott MC, Peters AM (2015) Pitfalls and limitations of radionuclide renal imaging in adults. Semin Nucl Med 45(5):428–439. https://doi.org/10.1053/j.semnuclmed.2015.02.008
Iyer RS, Schopp JG, Swanson JO, Thapa MM, Phillips GS (2013) Safety essentials: acute reactions to iodinated contrast media. Can Assoc Radiol J 64(3):193–199. https://doi.org/10.1016/j.carj.2011.12.014
Musso CG, Álvarez-Gregori J, Jauregui J, Macías-Núñez JF (2016) Glomerular filtration rate equations: a comprehensive review. Int Urol Nephrol 48(7):1105–1110. https://doi.org/10.1007/s11255-016-1276-1
Levey AS, Coresh J, Greene T, Stevens LA, Zhang YL, Hendriksen S, Kusek JW, Van Lente F (2006) Using standardized serum creatinine values in the modification of diet in renal disease study equation for estimating glomerular filtration rate. Ann Intern Med 145(4):247–254. https://doi.org/10.7326/0003-4819-145-4-200608150-00004
Ma YC, Zuo L, Chen JH, Luo Q, Yu XQ, Li Y, Xu JS, Huang SM, Wang LN, Huang W, Wang M, Xu GB, Wang HY (2006) Modified glomerular filtration rate estimating equation for Chinese patients with chronic kidney disease. J Am Soc Nephrol 17(10):2937–2944. https://doi.org/10.1681/ASN.2006040368
Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AR, Feldman HI, Kusek JW, Eggers P, Van Lente F, Greene T, Coresh J (2009) A new equation to estimate glomerular filtration rate. Ann Intern Med 150(9):604–612. https://doi.org/10.7326/0003-4819-150-9-200905050-00006
Berns JS (2015) Clinical decision making in a patient with stage 5 CKD–Is eGFR good enough? Clin J Am Soc Nephrol 10(11):2065–2072. https://doi.org/10.2215/CJN.00340115
Michels WM, Grootendorst DC, Verduijn M, Elliott EG, Dekker FW, Krediet RT (2010) Performance of the Cockcroft-Gault, MDRD, and new CKD-EPI formulas in relation to GFR, age, and body size. Clin J Am Soc Nephrol 5(6):1003–1009. https://doi.org/10.2215/CJN.06870909
Li Y, Zhang Y, Zhang E, Chen Y, Wang Q, Liu K, Yu HJ, Yuan H, Lang N, Su MY (2021) Differential diagnosis of benign and malignant vertebral fracture on CT using deep learning. Eur Radiol 31(12):9612–9619. https://doi.org/10.1007/s00330-021-08014-5
Cha KH, Hadjiiski L, Samala RK, Chan HP, Caoili EM, Cohan RH (2016) Urinary bladder segmentation in CT urography using deep-learning convolutional neural network and level sets. Med Phys 43(4):1882–1896
Teramoto A, Fujita H, Yamamuro O, Tamaki T (2016) Automated detection of pulmonary nodules in PET/CT images: ensemble false-positive reduction using a convolutional neural network technique. Med Phys 43(6Part1):2821–2827
Kim T, Lee KH, Ham S, Park B, Lee S, Hong D, Kim GB, Kyung YS, Kim C, Kim N (2020) Active learning for accuracy enhancement of semantic segmentation with CNN-corrected label curations: evaluation on kidney segmentation in abdominal CT. Sci Rep UK 10(1):1–7
Çiçek Ö, Abdulkadir A, Lienkamp SS, Brox T, Ronneberger O (2016) 3D U-Net: learning dense volumetric segmentation from sparse annotation. Springer, Berlin, pp 424–432
Levin A, Stevens PE, Bilous RW, Coresh J, De Francisco ALM, De Jong PE, Griffith KE, Hemmelgarn BR, Iseki K, Lamb EJ, Levey AS, Riella MC, Shlipak MG, Wang H, White CT, Winearls CG (2013) Kidney disease: Improving global outcomes (KDIGO) CKD work group. KDIGO 2012 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int Suppl 3(1):1–150. https://doi.org/10.1038/kisup.2012.73
Park J, Bae S, Seo S, Park S, Bang JI, Han JH, Lee WW, Lee JS (2019) Measurement of glomerular filtration rate using quantitative SPECT/CT and deep-learning-based kidney segmentation. Sci Rep 9(1):4223. https://doi.org/10.1038/s41598-019-40710-7
Wang H, Bowe B, Cui Z, Yang H, Joshua SS, Xie Y, Al-Aly Z (2022) A deep learning approach for the estimation of glomerular filtration rate. IEEE Trans Nanobiosci. https://doi.org/10.1109/TNB.2022.3147957
Hu J, Xu X, Zhang K, Li Y, Zheng J, Chen W, Wang X (2020) Comparison of estimated glomerular filtration rates in Chinese patients with chronic kidney disease among serum creatinine-, cystatin-C- and creatinine-cystatin-C-based equations: a retrospective cross-sectional study. Clin Chim Acta 505:34–42. https://doi.org/10.1016/j.cca.2020.01.033
Wang X, Luo Y, Wang Y, Wang C, Zhao X, Wang D, Liu L, Liu G, Wang Y (2014) Comparison of associations of outcomes after stroke with estimated GFR using Chinese modifications of the MDRD study and CKD-EPI creatinine equations: results from the China National Stroke Registry. Am J Kidney Dis 63(1):59–67. https://doi.org/10.1053/j.ajkd.2013.08.008
Stevens LA, Coresh J, Greene T, Levey AS (2006) Assessing kidney function–measured and estimated glomerular filtration rate. N Engl J Med 354(23):2473–2483. https://doi.org/10.1056/NEJMra054415
Acknowledgements
The authors acknowledge the Grant from the Project of Hubei Province Key Research and Development Project of China (Grant no. 2020BCB051), Hubei Province Central Guiding Local Science and Technology Development Project (Grant no. 2022BGE232), and the National Medical Education Development Center Medical Simulation Education Research Project of China (Grant no. 2021MNYB11) for supporting the work.
Author information
Authors and Affiliations
Contributions
Xiuheng Liu reports being director and the sponsor of this study. Qingyuan Zheng, Rui Yang, and Xinmiao Ni performed the research, analyzed the data, and wrote the paper. Jiejun Wu, Panpan Jiao, Song Yang, Lei Wang, and Zhiyuan Chen contributed essential tools.
Corresponding authors
Ethics declarations
Conflict of interest
The authors have no potential conflicts of interest to report.
Ethical approval
This retrospective study was approved by the Ethics Committee of Renmin Hospital of Wuhan University (No. WDRY2021-K065), and the requirement for written informed consent was waived. All patients had signed informed consent forms regarding the use of iodinated contrast agents before undergoing CTU examinations.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Zheng, Q., Ni, X., Yang, R. et al. UroAngel: a single-kidney function prediction system based on computed tomography urography using deep learning. World J Urol 42, 238 (2024). https://doi.org/10.1007/s00345-024-04921-6
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s00345-024-04921-6