Table 2_GLOBE: an explainable machine learning platform for preoperative prediction of thr

427 patients with glioma were included in a retrospective cohort to develop machine learning models for predicting postoperative venous thromboembolism (VTE) and postoperative neurological deterioration (PND). The GLOBE platform provides an online tool for individualized risk assessment, with models achieving AUCs of 0.815 for VTE and 0.741 for PND prediction. This dataset was authored by Feiling Xiang and last updated on June 1, 2026.

Use Cases

• Predicting postoperative venous thromboembolism risk based on patient age, D-dimer, fibrin degradation products, and hemoglobin levels.
• Assessing risk of postoperative neurological deterioration based on factors like recurrent glioma, Caprini score, neutrophil percentage, and hypertension history.
• Benchmarking new predictive models against the reported Top-10 random forest and logistic regression models.
• Developing interpretable clinical tools using SHAP analysis insights from the described feature importance.

Strengths

• Model performance is quantitatively validated, with a VTE prediction AUC of 0.815 and a PND prediction AUC of 0.741.
• The dataset underpins a deployed, accessible online prediction platform (GLOBE) for real-time risk assessment.
• Includes a substantial cohort of 427 patients with glioma for model development and validation.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Data is from a single retrospective cohort, which may limit generalizability.

Provenance

Source

figshare

Collection Method

Retrospective cohort study of 427 patients with glioma.

Freshness

Last updated 2026-06-01 05:25:30; freshness should be verified.