Machine Learning Identifies NPAR Prognostic Value
Researchers at The First Affiliated Hospital of Zhengzhou University retrospectively analyzed 160 adult-type diffuse glioma patients (June 2019–September 2021) and used LASSO, XGBoost, and random forest to build prognostic models. They found preoperative neutrophil percentage–to–albumin ratio (NPAR), platelet-to-mean platelet volume ratio, and age predicted overall and progression-free survival, with C-indexes of 0.731 (training) and 0.763 (validation).
Key Points
- 1Identify NPAR and platelet-to-mean-platelet-volume ratio as independent prognostic factors for overall and progression-free survival in ADG patients
- 2Demonstrate models achieve C-index 0.731 in training and 0.763 in validation, indicating strong discriminative performance
- 3Enable preoperative risk stratification to inform personalized treatment selection and follow-up intensity for clinicians
Scoring Rationale
Moderate novelty and practical model development, limited by single-center retrospective cohort and need for external validation.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
