CT-Based Habitat Imaging Model for Predicting Microvascular Invasion in HCC

Shuangxi Chen's dataset comprises a predictive model developed from a retrospective study of 216 patients with hepatocellular carcinoma. It integrates habitat imaging radiomic features with clinicopathologic factors to noninvasively predict microvascular invasion status. The model achieved an AUC of 0.862 in the training set and 0.814 in the validation set.

Use Cases

• Predicting microvascular invasion risk in hepatocellular carcinoma based on integrated habitat imaging radiomic features.
• Developing noninvasive preoperative risk stratification tools based on CT imaging and clinicopathologic data.
• Comparing the predictive performance of combined models versus single-feature models based on AUC metrics.
• Creating clinical nomograms for visualizing predictive model outputs based on integrated risk scores.

Strengths

• Model performance is validated with specific AUC metrics: 0.862 in training and 0.814 in validation.
• Dataset integrates multiple data types: CT imaging radiomics, habitat analysis, and clinicopathologic factors.
• The study includes 216 patients with pathologically confirmed hepatocellular carcinoma.

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 source bias inherent to a single retrospective study.

Provenance

Source

figshare

Collection Method

Retrospective study collecting preoperative CT scans and postoperative histopathologic data.

Freshness

Last updated 2026-05-22 04:30:29; freshness should be verified.