Generation of automated nephrometry scores through direct prediction of each component

Betty Wang; Rishi Jonnalagadda; Jayant Siva; Rikhil Seshadri; Sahil H. Patel; Angelica Bartholomew; Clara Goebel; Beatriz Lopez Morato; Gabriel Wallerstein-King; Jason Scovell; Rebecca Campbell; Michal Ozery-Flato; Vesna Barros; Maria Gabrani; Michal Rosen-Zvi; Ryan Ward; Steven Campbell; Robert Abouassaly; Vinay Duddalwar; Nicholas Heller; Erick M. Remer; Christopher J. Weight

doi:10.1016/j.urology.2026.02.039

Key Points

The study aims to evaluate the ability of a deep learning model to automate the generation of R.E.N.A.L. nephrometry scores and predict significant clinical outcomes.
Training a ResNet-50 neural network on preoperative CT images.
Using expert-derived segmentation masks for score component prediction.
Applying five-fold cross-validation for validation of results.
Comparing automated scores to human consensus scores from six raters.
Employing logistic regression and receiver operating characteristic analysis for outcome associations.
Automated scores correlate strongly with human consensus scores (Spearman's = 0.77).
Automated scores outperform individual raters (average score = 0.42, p < 0.01).
Higher predictive accuracy observed for partial vs. radical nephrectomy (AUC 0.87 vs. 0.80, p = 0.0012).
Automated scores predict malignancy and higher pathologic stage effectively.
External validation confirms correlation with human scoring and predicts surgical approaches across cohorts.

No abstract yet

We only have this paper's title and publication details.

Read it at the source

Implication

Deep learning automates nephrometry score generation, predicting clinical outcomes in kidney cancer patients, suggesting enhanced assessment efficiency.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Wang et al. (2026) studied this question.

synapsesocial.com/papers/69be34f26e48c4981c6730e3 https://doi.org/10.1016/j.urology.2026.02.039

View Full Paper

Ask AI

Bookmark

Share

Generation of automated nephrometry scores through direct prediction of each component

Key Points

No abstract yet

Implication

Discussion

Cite This Study

Generation of automated nephrometry scores through direct prediction of each component

Key Points

No abstract yet

Implication

Discussion

Cite This Study

Keep reading

Keep reading