Virtual nephrobiopsy as a knowledge base from functional-morphological patterns

BACKGROUND. Diagnosis of kidney diseases remains a serious challenge, as in many cases it is based on the examination of a kidney biopsy specimen, since the morphological form of the disease determines the treatment strategy, including the use of aggressive pharmacological agents. Non-invasive determination of the pathogenetic group of nephropathy reduces the risk of drug exposure in cases of treatment without morphological diagnosis. 

THE AIM. Development of a mathematical apparatus for constructing virtual morphological characteristics of the kidneys based on aggregate data on their functional status. 

PATIENTS AND METHODS. Artificial intelligence technologies with machine learning based on data from 571 patients with morphological diagnoses. 

RESULTS. The proposed set of medical informatics methods demonstrates potential for virtual pathomorphological modeling and improving healthcare safety by providing clinical decision support. 

CONCLUSION. The proposed system can be used as a clinical decision support tool, enabling patient stratification into three pathogenetic groups (glomerular, tubulointerstitial, and vascular) that differ in principles of clinical management and drug therapy. This highlights the potential of medical informatics and machine learning methods for improving the clinical interpretability of nephrobiopsy data.

1. Shilov EM, Yesayan AM, Petrova NYu, Artemov DV, Batyushin MM, Bevzenko AYu, Belskikh AN, Veselkova NV, Ivliev SV, Statsenko ME, Stolyar AG, Shilova MM. Chronic kidney disease should become a priority area of Russian healthcare. Statement of the Presidium of the Specialized Commission on Nephrology of the Ministry of Health of Russia. 2025. doi: 10.36485/1561-62742025-29-4-8-21. EDN: PAQGCC

2. Feng C, Liu F. Artificial intelligence in renal pathology: Current status and future. Clin Kidney J 2022;15:100–112

3. Genkin AA. New information technology for medical data analysis. OMIS software package. Polytechnic, St. Petersburg, 1999, 191 p

4. Ryabov SI, Emanuel VL, Plotkin VYa, Klemina IK. The role of lesion of tubules and interstitium in the development of renal insufficiency. Arkh Patol 1982;44(12):55–58

5. Cascarano GD, Debitonto FS, Lemma R, Brunetti A, Buongiorno D, De Feudis I et al. A neural network for glomerulus classification based on histological images of kidney biopsy. BMC Med Imaging 2021;21:12. doi:10.1186/s12880-021-00562-2

6. Luo Y, Liang J, Hu X, Tang Z, Zhang J, Han L et al. Deep learning algorithms for the prediction of posttransplant renal function in deceased-donor kidney recipients: a preliminary study based on pretransplant biopsy. Front Med (Lausanne) 2021;8:664. doi:10.3389/fmed.2021.00664

7. Fayzullin A, Ivanova E, Grinin V, Ermilov D, Solovyeva S, Balyasin M et al. Towards accurate and efficient diagnoses in nephropathology: an AI-based approach for assessing kidney transplant rejection. Comput Biol Med 2022;145:105427. doi:10.1016/j. compbiomed.2022.105427

8. Sawhney R, Malik A, Sharma S, Narayan V. A comparative assessment of artificial intelligence models used for early prediction and evaluation of chronic kidney disease. Healthc Technol Lett 2020;7(6):158–165. doi:10.1049/htl.2020.0084.

9. Poonia RC, Gupta MK, Abunadi I, Albraikan AA, Al-Wesabi FN, Hamza MA, et al. Intelligent diagnostic prediction and classification models for detection of kidney disease. J King Saud Univ Sci 2021;33(6):101429. doi:10.1016/j.jksus.2021.101429

10. Schena FP, Anelli VW, Trotta J, Di Noia T, Manno C, Tripepi G et al. Development and testing of an artificial intelligence tool for predicting end-stage kidney disease in patients with immunoglobulin A nephropathy. Kidney Int Rep 2021;6:123–135. doi:10.1016/j.ekir.2020.10.034

11. Rasmussen DGK, Boesby L, Nielsen SH, Tepel M, Birot S, Karsdal MA, Kamper AL, Genovese F. Collagen turnover profiles in chronic kidney disease. Sci Rep 2019 Nov 5;9(1):16062. doi: 10.1038/s41598-019-51905-3. PMID: 31690732; PMCID: PMC6831687

12. Razumov VV. Chronic renal failure as a recapitulation of patterns of phylogenetic development of renal functions. Modern problems of science and education 2008;8. URL: https:// science-education.ru/ru/article/view?id=3619 (date accessed: 01/08/2026)

13. Yamanouchi M, Hoshino J, Ubara Y et al. Clinicopathological predictors for progression of chronic kidney disease in nephrosclerosis: a biopsy-based cohort study. Nephrol Dial Transplant 2020;35(1):136–144. doi:10.1093/ndt/gfz024

14. Yu Z, Ning X, Qin Y et al. Development and validation of a machine learning-based prognostic model for IgA nephropathy with CKD stage 3 or 4. Kidney Dis 2024;10(6):436–449. doi:10.1159/000536992

15. Noda R, Ichikawa D, Shibagaki Y. Machine learning-based diagnostic prediction of IgA nephropathy. Sci Rep 2024;14:12426. doi:10.1038/s41598-024-12426-y

16. Chagas P, Souza L, Araújo I et al. Classification of glomerular hypercellularity using convolutional features and support vector machine. Comput Biol Med 2021;134:104457. doi:10.1016/j. compbiomed.2021.104457

17. Barros GO, Muller da Silva JNA, Proença HMS et al. Enhancing podocyte degenerative changes identification with pathologist collaboration. IEEE Access 2023;11:119845–119858. doi:10.1109/ACCESS.2023.3315678

18. KDIGO Clinical Practice Guideline for the Management of Glomerular Diseases. Kidney International 2021;100(Suppl 4S):S1–S276

19. KDIGO 2024 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease. Kidney International. 2024;105(Suppl 4S):S117–S314

20. KDIGO Clinical Practice Guideline for Acute Kidney Injury. Kidney International Supplements 2012;2(1):1–138

21. Moledina DG, Perazella MA. Treatment of Drug-Induced Acute Tubulointerstitial Nephritis: The Search for Better Evidence. Clin J Am Soc Nephrol 2018;13(12):1785–1787

22. KDIGO 2021 Clinical Practice Guideline for the Management of Blood Pressure in Chronic Kidney Disease. Kidney International 2021;99(3S):S1–S87