Severe Acute Pancreatitis Prediction Model with Clinical and CT Features

Wei Wei's retrospective study includes data from 1,128 patients admitted for acute pancreatitis between November 2019 and December 2025. The dataset was used to develop and validate a nomogram model for predicting severe acute pancreatitis (SAP), achieving an AUC of 0.88. The model incorporates clinical characteristics, laboratory results, and non-contrast CT signs.

Use Cases

• Train a binary classification model for severe acute pancreatitis based on clinical and laboratory features mentioned in the description.
• Validate a nomogram-based prediction tool using the described internal and external validation patient sets.
• Analyze the association between specific risk factors (e.g., D-dimer, creatinine, VFR) and pancreatitis severity.
• Conduct subgroup analysis for patients with hyperlipidemic or non-hyperlipidemic acute pancreatitis.

Strengths

• Dataset includes 1,128 patient records split into training, internal validation, and external validation sets.
• Model performance is quantitatively reported with an AUC of 0.88 and a 95% confidence interval (0.84–0.92).
• The study identifies 11 independent predictors of SAP, including age, diabetes, D-dimer, creatinine, and CT signs.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count for the underlying data tables is unknown, which may limit suitability assessment.
• Data may reflect temporal and institutional bias inherent to the retrospective, hospital-based study design.

Provenance

Source

figshare

Collection Method

Retrospective study of hospital admissions.

Time Range

November 2019 to December 2025

Freshness

Last updated 2026-05-07 05:44:36; freshness should be verified.