Multimodal Clinical and Imaging Data for Predicting Hepatic Encephalopathy Post-TIPS

Lin-Feng Zhou's dataset supports a study developing a multimodal model to predict overt hepatic encephalopathy (OHE) within one year after a transjugular intrahepatic portosystemic shunt (TIPS) procedure. The data includes manual CT features, radiomics, and clinical data from 338 patients treated between November 2015 and January 2022. The combined model (Model MRC) achieved an area under the ROC curve of 0.902.

Use Cases

• Training predictive models for post-TIPS complications based on integrated clinical and imaging data.
• Benchmarking radiomics feature extraction against manual CT feature analysis for liver disease outcomes.
• Comparing the performance of unimodal versus multimodal models for clinical risk prediction.

Strengths

• Includes data from 338 patients, providing a substantial cohort for model development.
• Model performance is validated with reported metrics, including an AUC of 0.902 for the combined model.
• Data integrates three distinct modalities: manual imaging features, radiomics, and clinical variables.

Limitations

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

Provenance

Source

Lin-Feng Zhou via figshare.

Collection Method

Retrospective study of patients who underwent TIPS.

Time Range

November 2015 to January 2022.

Freshness

Last updated 2026-05-12 04:19:06; freshness should be verified.