Enhancing individual glomerular filtration rate assessment : can we trust the equation? Development and validation of machine learning models to assess the trustworthiness of estimated GFR compared to measured GFR
(2025) In BMC Nephrology 26(1).- Abstract
Background: Creatinine-based estimated glomerular filtration rate (eGFR) equations are widely used in clinical practice but exhibit inherent limitations. On the other side, measuring GFR is time consuming and not available in routine clinical practice. We developed and validated machine learning models to assess the trustworthiness (i.e. the ability of equations to estimate measured GFR (mGFR) within 10%, 20% or 30%) of the European Kidney Function Consortium (EKFC) equation at the individual level. Methods: This observational study used data from European and US cohorts, comprising 22,343 participants of all ages with available mGFR results. Four machine learning and two traditional logistic regression models were trained on a cohort... (More)
Background: Creatinine-based estimated glomerular filtration rate (eGFR) equations are widely used in clinical practice but exhibit inherent limitations. On the other side, measuring GFR is time consuming and not available in routine clinical practice. We developed and validated machine learning models to assess the trustworthiness (i.e. the ability of equations to estimate measured GFR (mGFR) within 10%, 20% or 30%) of the European Kidney Function Consortium (EKFC) equation at the individual level. Methods: This observational study used data from European and US cohorts, comprising 22,343 participants of all ages with available mGFR results. Four machine learning and two traditional logistic regression models were trained on a cohort of 9,202 participants to predict the likelihood of the EKFC creatinine-derived eGFR falling within 30% (p30), 20% (p20) or 10% (p10) of the mGFR value. The algorithms were internally and then externally validated on cohorts of respectively 3,034 and 10,107 participants. The predictors included in the models were creatinine, age, sex, height, weight, and EKFC. Results: The random forest model was the most robust model. In the external validation cohort, the model achieved an area under the curve of 0.675 (95%CI 0.660;0.690) and an accuracy of 0.716 (95%CI 0.707;0.725) for the P30 criterion. Sensitivity was 0.756 (95%CI 0.747;0.765) and specificity was 0.485 (95%CI 0.460; 0.511) at the 80% probability level that EKFC falls within 30% of mGFR. At the population level, the PPV of this machine learning model was 89.5%, higher than the EKFC P30 of 85.2%. A free web-application was developed to allow the physician to assess the trustworthiness of EKFC at the individual level. Conclusions: A strategy using machine learning model marginally improves the trustworthiness of GFR estimation at the population level. An additional value of this approach lies in its ability to provide assessments at the individual level.
(Less)
- author
- organization
-
- Space Planning
- Epidemiology and population studies (EPI@Lund) (research group)
- Centre for Economic Demography
- Division of Occupational and Environmental Medicine, Lund University
- LU Profile Area: Proactive Ageing
- Infect@LU
- LU Profile Area: Nature-based future solutions
- eSSENCE: The e-Science Collaboration
- EpiHealth: Epidemiology for Health
- LUCC: Lund University Cancer Centre
- Radiology Diagnostics, Malmö (research group)
- Cystatin C, renal disease, amyloidosis and antibiotics (research group)
- publishing date
- 2025-12
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- Chronic kidney disease, Creatinine, Glomerular filtration rate, Machine learning, Random forest
- in
- BMC Nephrology
- volume
- 26
- issue
- 1
- article number
- 47
- publisher
- BioMed Central (BMC)
- external identifiers
-
- pmid:39885391
- scopus:85217623809
- ISSN
- 1471-2369
- DOI
- 10.1186/s12882-025-03972-0
- language
- English
- LU publication?
- yes
- id
- 2e244f91-5475-486f-ad0b-21cf76cf7db6
- date added to LUP
- 2025-12-19 14:06:57
- date last changed
- 2026-01-16 16:16:20
@article{2e244f91-5475-486f-ad0b-21cf76cf7db6,
abstract = {{<p>Background: Creatinine-based estimated glomerular filtration rate (eGFR) equations are widely used in clinical practice but exhibit inherent limitations. On the other side, measuring GFR is time consuming and not available in routine clinical practice. We developed and validated machine learning models to assess the trustworthiness (i.e. the ability of equations to estimate measured GFR (mGFR) within 10%, 20% or 30%) of the European Kidney Function Consortium (EKFC) equation at the individual level. Methods: This observational study used data from European and US cohorts, comprising 22,343 participants of all ages with available mGFR results. Four machine learning and two traditional logistic regression models were trained on a cohort of 9,202 participants to predict the likelihood of the EKFC creatinine-derived eGFR falling within 30% (p30), 20% (p20) or 10% (p10) of the mGFR value. The algorithms were internally and then externally validated on cohorts of respectively 3,034 and 10,107 participants. The predictors included in the models were creatinine, age, sex, height, weight, and EKFC. Results: The random forest model was the most robust model. In the external validation cohort, the model achieved an area under the curve of 0.675 (95%CI 0.660;0.690) and an accuracy of 0.716 (95%CI 0.707;0.725) for the P30 criterion. Sensitivity was 0.756 (95%CI 0.747;0.765) and specificity was 0.485 (95%CI 0.460; 0.511) at the 80% probability level that EKFC falls within 30% of mGFR. At the population level, the PPV of this machine learning model was 89.5%, higher than the EKFC P30 of 85.2%. A free web-application was developed to allow the physician to assess the trustworthiness of EKFC at the individual level. Conclusions: A strategy using machine learning model marginally improves the trustworthiness of GFR estimation at the population level. An additional value of this approach lies in its ability to provide assessments at the individual level.</p>}},
author = {{Lanot, Antoine and Akesson, Anna and Nakano, Felipe Kenji and Vens, Celine and Björk, Jonas and Nyman, Ulf and Grubb, Anders and Sundin, Per Ola and Eriksen, Björn O. and Melsom, Toralf and Rule, Andrew D. and Berg, Ulla and Littmann, Karin and Åsling-Monemi, Kajsa and Hansson, Magnus and Larsson, Anders and Courbebaisse, Marie and Dubourg, Laurence and Couzi, Lionel and Gaillard, Francois and Garrouste, Cyril and Jacquemont, Lola and Kamar, Nassim and Legendre, Christophe and Rostaing, Lionel and Ebert, Natalie and Schaeffner, Elke and Bökenkamp, Arend and Mariat, Christophe and Pottel, Hans and Delanaye, Pierre}},
issn = {{1471-2369}},
keywords = {{Chronic kidney disease; Creatinine; Glomerular filtration rate; Machine learning; Random forest}},
language = {{eng}},
number = {{1}},
publisher = {{BioMed Central (BMC)}},
series = {{BMC Nephrology}},
title = {{Enhancing individual glomerular filtration rate assessment : can we trust the equation? Development and validation of machine learning models to assess the trustworthiness of estimated GFR compared to measured GFR}},
url = {{http://dx.doi.org/10.1186/s12882-025-03972-0}},
doi = {{10.1186/s12882-025-03972-0}},
volume = {{26}},
year = {{2025}},
}
