AITIS Model Validation: Clinical Data for Sarcopenia Prediction in Stage IV-V CKD Patients

Yang Li published a dataset on 2026-04-28 from a cross-sectional study of 236 patients with stage IV-V chronic kidney disease. The data includes patient demographics, 20 functional measures, and sarcopenia diagnoses based on AWGS 2019 criteria. It was used to validate the performance of the Artificial Intelligence to Identify Sarcopenia (AITIS) model.

Use Cases

• Validating test-free AI models for sarcopenia screening based on functional measures and demographics.
• Analyzing the relationship between specific functional limitations (e.g., jogging 1 km, climbing stairs) and sarcopenia in CKD patients.
• Investigating model performance across patient subgroups, such as those with different CKD stages or physical activity levels.

Strengths

• Dataset includes 236 patient records with a clear sarcopenia prevalence of 26.3% (62 patients).
• Model performance metrics are provided, including an AUC of 0.792 and accuracy of 0.758.
• Data includes 20 functional measures and key predictors like age, sex, height, and weight.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Data may reflect temporal and source bias inherent to a single observational study.

Provenance

Source

Yang Li via figshare.

Collection Method

Observational cross-sectional study enrolling 236 patients with stage IV-V CKD.

Freshness

Last updated 2026-04-28 04:29:23; freshness should be verified.