AI-Assisted Clinico-Quantitative Imaging Nomogram for Pulmonary Nodule Malignancy Risk

A prediction model dataset for preoperative malignancy risk assessment in pulmonary nodules ≤ 3 cm, developed by Yingding Ruan and last updated in April 2026. The dataset is derived from a retrospective study of 951 consecutive patients who underwent surgical resection, containing 210 benign and 741 malignant cases. It combines AI-extracted quantitative CT features with clinical variables and inflammatory markers.

Use Cases

• Developing machine learning models for malignancy risk prediction based on AI-extracted quantitative imaging features.
• Validating clinical prediction nomograms for solid and part-solid pulmonary nodules.
• Conducting decision curve analysis to evaluate the clinical utility of a risk assessment tool.
• Performing internal validation of multivariable logistic regression models using bootstrap resampling.

Strengths

• Dataset is based on a substantial cohort of 951 patient cases.
• Model demonstrated strong discrimination with an AUC of 0.836 and excellent calibration with a mean absolute error of 0.015.
• Risk stratification categorized patients into three distinct risk strata with observed malignancy rates of 43.3%, 86.7%, and 95.0%.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The study cohort was restricted to surgically resected nodules, indicating a malignancy-enriched preoperative surgical setting, which limits generalizability to screening populations.

Provenance

Source

Yingding Ruan via figshare.

Collection Method

Retrospective study analyzing data from patients who underwent surgical resection for pulmonary nodules.

Time Range

The temporal coverage of the underlying patient data is not specified.

Freshness

Last updated 2026-04-13 05:31:08; freshness should be verified.

Geography

The geographic coverage of the patient cohort is not specified.