Diabetic Foot Ulcer Pre-Screening Model with Clinical Indicators, 2012-2025

Weidi Wang's dataset supports a study to develop and validate a simplified pre-screening model for diabetic foot ulcer (DFU) identification. It includes retrospective clinical data from 1,854 hospitalized patients (2012-2020) and 678 outpatients (2021-2025). The model identifies four predictors: age, injury history, alcohol consumption, and the Alb/HbA1c ratio.

Use Cases

• Developing a binary classification model for DFU risk based on the four identified clinical predictors.
• Validating a pre-screening tool's performance using metrics like AUC, calibration, and Brier score as described.
• Analyzing the interaction between age and the Alb/HbA1c ratio for synergistic risk effects.
• Creating clinical nomograms or risk heatmaps for patient stratification as mentioned in the methodology.

Strengths

• Model performance is validated on a distinct outpatient cohort of 678 patients.
• The model achieved an AUC of 0.817 (95% CI 0.782–0.852) in the validation set.
• Dataset includes a training set of 1,854 patient records for model development.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count for the underlying patient data tables is unknown, which may limit suitability assessment.
• Data may reflect geographic and institutional bias inherent to the single hospital system source.

Provenance

Source

Weidi Wang via figshare.

Collection Method

Retrospective analysis of hospitalized and outpatient clinical records.

Time Range

January 2012 – December 2025

Freshness

Last updated 2026-05-29 04:47:36; freshness should be verified.

Geography

Patients from a geographically distinct branch of a hospital system (location unspecified).