Predicting Central Lymph Node Metastasis in Papillary Thyroid Carcinoma

Yalin Zhu's study presents a dataset for predicting central lymph node metastasis in clinically node-negative T1–T2 papillary thyroid carcinoma. The dataset integrates pathological, ultrasound features, thyroid function, and systemic inflammatory indicators from 710 patients (971 lesions). The model was developed using a gradient boosting decision tree and validated on an independent temporal cohort.

Use Cases

• Training predictive models for lymph node metastasis based on multimodal clinical features like tumor size and laterality.
• Validating feature importance for thyroid cancer prognosis using SHAP interpretability frameworks.
• Assessing clinical utility of inflammatory biomarkers like SII and PLR for surgical decision-making.

Strengths

• Model achieved an AUC of 0.830 in the test set, indicating strong predictive performance.
• Dataset includes 710 patients (971 lesions), providing a substantial sample for analysis.
• Model was externally validated on an independent temporal cohort of 50 patients (AUC 0.800).

Limitations

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

Provenance

Source

figshare, author Yalin Zhu

Collection Method

Retrospective analysis of patient records.

Freshness

Last updated 2026-04-27 04:12:01; freshness should be verified.