Machine Learning-Based Quantification of Vesicoureteral Reflux with Enhancing Accuracy and Efficiency

TL;DR

This study applies machine learning to analyze VCUG images for VUR diagnosis, achieving high accuracy and objectivity, and identifying renal calyceal deformation as a key indicator of severe reflux.

Contribution

Introduces a machine learning approach for VUR assessment that improves consistency and objectivity over traditional subjective grading methods.

Findings

All models achieved accurate classification with no false positives or negatives.

Renal calyceal deformation is identified as a key predictor of high-grade VUR.

High AUC values confirm sensitivity to subtle image patterns.

Abstract

Vesicoureteral reflux (VUR) is traditionally assessed using subjective grading systems, which introduces variability in diagnosis. This study investigates the use of machine learning to improve diagnostic consistency by analyzing voiding cystourethrogram (VCUG) images. A total of 113 VCUG images were reviewed, with expert grading of VUR severity. Nine image-based features were selected to train six predictive models: Logistic Regression, Decision Tree, Gradient Boosting, Neural Network, and Stochastic Gradient Descent. The models were evaluated using leave-one-out cross-validation. Analysis identified deformation patterns in the renal calyces as key indicators of high-grade VUR. All models achieved accurate classifications with no false positives or negatives. High sensitivity to subtle image patterns characteristic of different VUR grades was confirmed by substantial Area Under the Curve (AUC) values. The results suggest that machine learning can offer an objective and standardized alternative to current subjective VUR assessments. These findings highlight renal calyceal deformation as a strong predictor of severe cases. Future research should aim to expand the dataset, refine imaging features, and improve model generalizability for broader clinical use.