# Predictive Factors and Nomogram for Malignant Pulmonary Nodules (≤ 1 cm)

## TL;DR

This study identifies key factors and creates a predictive model to assess the likelihood of malignancy in small pulmonary nodules.

## Contribution

The study introduces a new nomogram for predicting malignancy in pulmonary nodules ≤ 1 cm, incorporating clinical and imaging features.

## Key findings

- The nomogram achieved an area under the curve of 0.79 with 70% sensitivity and 79% specificity.
- Partial-solid/nonsolid density, larger diameter, and absence of calcification were significant predictors of malignancy.
- The model requires external validation to confirm its reliability and sensitivity.

## Abstract

Models for predicting malignancy in pulmonary nodules ≤ 10 mm are lacking. This study aimed to identify predictive factors and develop a risk model for such nodules.

A retrospective cohort study analyzed 298 patients with pulmonary nodules ≤ 1 cm. Variables including sex, smoking, nodule position, density, enhancement, diameter, and calcification were considered. A nomogram was developed using forward stepwise selection.

The nomogram, incorporating the seven aforementioned variables, achieved an area under the curve of 0.79. Multivariable analysis identified partial‐solid/nonsolid density (vs. solid), larger diameter, and the absence of calcification as significant independent predictors of malignancy. At its optimal threshold, the nomogram showed 70% sensitivity, 79% specificity, and 77% accuracy. Decision curve analysis indicated a net benefit.

Nodule density, diameter, and calcification status are key independent predictors of malignancy in nodules ≤ 1 cm. The developed nomogram, which also includes other clinical and computed tomography features, shows good predictive performance but requires external validation, especially considering its sensitivity.