Predicting chronic kidney disease progression using small pathology datasets and explainable machine learning models

Aug 4, 2026

Upload by Violet

Chronic kidney disease (CKD) affects over 700 million people worldwide and often progresses to kidney failure. This study builds explainable machine learning models that use pathology data to predict CKD trajectory, aiming to improve prognostic performance in early stages despite limited datasets. Models rely on clinical variables including age, gender, most recent eGFR, mean eGFR, and eGFR slope, and use decision trees and random forests for interpretability. Internal validation (n=706) and external validation on a Japanese cohort (n=597) demonstrate strong ROC-AUC results with variable-level explanations.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages12

Reading Time30 min

AccessFree to download

Abstract

Background

Methods

Results

Conclusion

Introduction

This may be the author's version of a work that was submitted/accepted for publication in the following source: Reddy, Sandeep, Singh, Supriya, Choy, Kay Weng, Sharma, Sourav, Dwyer, Karen M., Manapragada, Chaitanya, Miller, Zane, Cheon, Joy, & Nakisa, Bahareh (2024) Predicting chronic kidney disease progression using small pathology datasets and explainable machine learning models. Computer Methods and Programs in Biomedicine Update, 6, Article number: 100160 . This ﬁle was downloaded from: [https://eprints.qut.edu.au/251482/](https://eprints.qut.edu.au/251482/) © Consult author(s) regarding copyright matters This work is covered by copyright. Unless the document is being made available under a Creative Commons Licence, you must assume that re-use is limited to personal use and that permission from the copyright owner must be obtained for all other uses. If the document is available under a Creative Commons License (or other speciﬁed license) then refer to the Licence for details of permitted re-use. It is a condition of access that users recognise and abide by the legal requirements associated with these rights. If you believe that this work infringes copyright please provide details by email [to qut.copyright@qut.edu.au](to qut.copyright@qut.edu.au) License: Creative Commons: Attribution-Noncommercial-No Derivative Works 4.0 Notice: Please note that this document may not be the Version of Record (i. e. published version) of the work. Author manuscript versions (as Submitted for peer review or as Accepted for publication after peer review) can be identiﬁed by an absence of publisher branding and/or typeset appear ance. If there is any doubt, please refer to the published source. [https://doi.org/10.1016/j.cmpbup.2024.100160](https://doi.org/10.1016/j.cmpbup.2024.100160) Computer Methods and Programs in Biomedicine Update 6 (2024) 100160 Contents lists available at ScienceDirect Computer Methods and Programs in Biomedicine Update journal [homepage: www.sciencedirect.com/journal/computer-methods](homepage: www.sciencedirect.com/journal/computer-methods)and-programs-in-biomedicine-update | Predicting chronic kidney disease progression using small pathology datasets and explainable machine learning models
Sandeep Reddy a, * , Supriya Roy b , Kay Weng Choy c , Sourav Sharma b , Karen M Dwyer d , Chaitanya Manapragadaa , Zane Miller e, Joy Cheone , Bahareh Nakisab
a School of Medicine, Deakin University, Geelong, Australia
b School of Information Technology, Deakin University, Geelong, Australia c Northern Health, Melbourne, Australia
d The Royal Melbourne Hospital, Melbourne, Australia e School of Medicine, University of Melbourne, Australia | | | | --- | --- | --- | | A R T I C L E I N F O | | A B S T R A C T | | Keywords:
Chronic kidney disease prediction Explainable machine learning Transfer learning
Shapley additive exPlanations Counterfactual analysis | | Background: Chronic kidney disease (CKD) poses a major global public health burden, with over 700 million affected. Early identification of those in whom the disease is likely to progress enables timely therapeutic interventions to delay advancement to kidney failure.
Methods: This study developed explainable machine learning models leveraging pathology data to accurately predict CKD trajectory, targeting improved prognostic capability even in early stages using limited datasets. Key variables used in this study include age, gender, most recent estimated glomerular filtration rate (eGFR), mean eGFR, and eGFR slope over time prior to the incidence of kidney failure. Supervised classification modelling techniques included decision tree and random forest algorithms selected for interpretability. Internal validation on an Australian tertiary centre cohort (n = 706; 353 with kidney failure and 353 without) achieved exceptional predictive accuracy. To address the inherent class imbalance, centroid-cluster-based under-sampling

FAQs

What is the main goal of the study on CKD progression prediction?

To predict the trajectory of chronic kidney disease progression and identify patients likely to reach kidney failure earlier, using pathology data and explainable machine learning despite limited datasets.

Which data and variables are used to build the predictive models?

The models use pathology-informed inputs and key clinical variables including age, gender, most recent estimated glomerular filtration rate (eGFR), mean eGFR, and eGFR slope over time prior to kidney failure.

How well do the models perform in internal and external validation?

Internal validation on an Australian tertiary centre cohort (n=706) achieved high ROC-AUC values (0.94 for decision tree and 0.98 for random forest). External validation on a Japanese CKD registry cohort (n=597) also showed strong performance (0.88 for decision tree and 0.93 for random forest).

Aug 4, 2026

Upload by Violet