TyG-BMI, CTI, and SII for MASLD Identification in Type 2 Diabetes Patients

A retrospective cross-sectional study of 796 hospitalized patients with type 2 diabetes from January 2020 to January 2025, with an external validation cohort of 394 patients. The dataset, authored by Yue Chen and shared under CC-BY-4.0, evaluates the diagnostic performance of three indices for identifying metabolic dysfunction-associated steatotic liver disease.

Use Cases

• Training machine learning models for liver disease prediction based on anthropometric, biochemical, and inflammatory parameters.
• Benchmarking the diagnostic performance of combined biomarker indices like TyG-BMI, CTI, and SII.
• Conducting external validation studies for clinical risk scores using the provided independent cohort.
• Analyzing correlations between metabolic indices and liver disease status in a diabetic population.

Strengths

• Includes data from 796 primary cohort patients and 394 external validation patients.
• Evaluates three distinct biomarker indices and their combined model, validated with ROC curves and machine learning (random forest AUC up to 0.950).
• Study design and key limitations, such as retrospective design and ultrasound-based diagnosis, are explicitly stated in the description.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset is very small (15.1 KB), indicating limited raw data or a summary document.

Provenance

Source

figshare

Collection Method

Retrospective cross-sectional study collecting anthropometric, biochemical, and inflammatory parameters from hospitalized patients.

Time Range

January 2020 to January 2025

Freshness

Last updated 2026-05-08 05:53:53; freshness should be verified.

Geography

null