Crescentic Glomerulonephritis Patient Data with Risk Scores, 224 Patients

224 patient records from the First Affiliated Hospital of Sun Yat-sen University were analyzed to develop risk scores for survival and kidney failure. K-means clustering identified high-risk and low-risk groups with 10-year survival rates of 63.1% and 89.6%, respectively. The study, published on figshare, includes a renal risk score based on serum creatinine, normal glomeruli percentage, and tubular atrophy.

Use Cases

• Validate unsupervised clustering methods for patient stratification based on clinical variables mentioned in the description
• Develop predictive models for end-stage renal disease based on the described cGN-KS score components
• Train decision tree models to identify high-risk patients using age, serum C3, and proteinuria data
• Compare survival outcomes between traditional cGN classifications and data-driven risk groups

Strengths

• Includes 224 patient records with a median follow-up of 87.10 months
• Risk stratification validated with a hazard ratio of 3.28 for mortality in the high-risk group
• Clinical outcomes are defined: all-cause mortality and end-stage renal disease

Limitations

• Column-level documentation is absent; field semantics must be inferred after download
• Data is from a single hospital in China, which may limit generalizability
• Row count is unknown, which may limit suitability assessment

Provenance

Source

First Affiliated Hospital of Sun Yat-sen University

Collection Method

Retrospective analysis of patient records

Time Range

Median follow-up of 87.10 months (interquartile range: 32.35, 153.97 months)

Freshness

Last updated 2026-04-18 07:55:30; freshness should be verified

Geography

China (likely Guangdong province)