Community-Based Early-Stage Chronic Kidney Disease Screening using Explainable Machine Learning for Low-Resource Settings

TL;DR

This study develops an explainable machine learning framework for early-stage CKD screening in low-resource settings, outperforming existing tools and validated across multiple South Asian datasets.

Contribution

It introduces a tailored, interpretable ML model for community-based early CKD detection in low-resource South Asian populations, with superior accuracy and generalizability.

Findings

Achieved 90.40% balanced accuracy with RFECV-selected features.

Minimal feature models outperformed larger feature sets.

External validation showed 78%-98% sensitivity across datasets.

Abstract

Early detection of chronic kidney disease (CKD) is essential for preventing progression to end-stage renal disease. However, existing screening tools - primarily developed using populations from high-income countries - often underperform in Bangladesh and South Asia, where risk profiles differ. Most of these tools rely on simple additive scoring functions and are based on data from patients with advanced-stage CKD. Consequently, they fail to capture complex interactions among risk factors and are limited in predicting early-stage CKD. Our objective was to develop and evaluate an explainable machine learning (ML) framework for community-based early-stage CKD screening for low-resource settings, tailored to the Bangladeshi and South Asian population context. A community-based CKD dataset from Bangladesh was used to develop predictive models. Variables were organized into clinically meaningful feature groups, and ten complementary feature selection methods were applied to identify robust predictor subsets. Twelve ML classifiers were evaluated using nested cross-validation. Model performance was benchmarked against established CKD screening tools and externally validated on three independent datasets from India, the UAE, and Bangladesh. SHAP was used to interpret model predictions. An ML model trained on an RFECV-selected feature subset achieved a balanced accuracy of 90.40%, whereas minimal non-pathology-test features demonstrated excellent predictive capability with a balanced accuracy of 89.23%, often outperforming larger or full feature sets. Compared with existing screening tools, the proposed models achieved substantially higher accuracy and sensitivity while requiring fewer and more accessible inputs. External validation confirmed strong generalizability with 78% to 98% sensitivity.