Nomogram Differentiates Benign and Malignant Small Pulmonary Nodules in Two-Center Study

A predictive nomogram that integrates computed tomography imaging features and clinical indicators has demonstrated strong accuracy in differentiating benign from malignant solid pulmonary nodules measuring two centimeters or smaller, according to findings from a two-center retrospective study published in BMC Medical Imaging. The multi-institutional investigation was conducted to aid clinical judgment and enhance patient management by combining routinely available clinical data with radiological findings into an accessible diagnostic prediction model.

The study evaluated a total of 1,337 participants who underwent surgery across two medical centers. To establish an effective diagnostic tool, the researchers comprehensively documented clinical parameters and computed tomography imaging findings for each patient, focusing specifically on individuals presenting with solid pulmonary nodules measuring up to two centimeters. The central objective was to develop and validate a nomogram incorporating both computed tomography-derived characteristics and clinical indicators to assist clinicians in distinguishing malignant nodules from benign lesions.

To formulate and thoroughly evaluate the model, the investigators partitioned the 1,337 surgical participants into three distinct cohorts representing training, internal testing, and external validation groups. The training set included patients with solid pulmonary nodules measuring two centimeters or smaller who underwent surgery from January 2019 to January 2023 at Yunnan Cancer Hospital. An internal testing set was formed using surgical cases from February 2023 to June 2025 at Yunnan Cancer Hospital. To evaluate the model independently, an external validation set comprised patients treated from January 2020 to January 2024 at the First Affiliated Hospital of Kunming Medical University.

Logistic regression analyses were then carried out to identify variables significantly associated with nodule malignancy. Multivariate logistic regression identified six independent predictors of malignancy in solid pulmonary nodules measuring two centimeters or smaller, comprising five risk factors and one protective factor. The identified risk factors were lobulation, spiculation, carcinoembryonic antigen levels, maximum diameter, and mean computed tomography value, whereas nodule calcification was identified as a protective factor.

Statistical evaluation demonstrated strong discriminatory capacity across all cohorts examined in the study. The formulated nomogram achieved an area under the curve value of 0.869 in the training cohort, an area under the curve value of 0.823 in the testing cohort, and an area under the curve value of 0.902 in the external validation cohort. Furthermore, calibration curves and decision curve analyses confirmed good agreement between predicted probabilities and observed surgical outcomes, demonstrating substantial clinical applicability for the predictive model.

The authors reported that incorporating the mean computed tomography value and calcification provided critical diagnostic value beyond morphological characteristics alone, such as spiculation and lobulation. By combining these routinely available computed tomography features with clinical markers, the nomogram supported more individualized risk stratification for small solid pulmonary nodules. The authors concluded that the model exhibits excellent accuracy in predicting malignancy for solid nodules measuring two centimeters or smaller, while noting its scope remains grounded in the retrospective analysis of surgical patients across the two participating medical centers.

Subject of Research: Nomogram prediction of malignancy in solid pulmonary nodules measuring 2 cm or smaller

Article Title: A nomogram based on clinical and CT features to differentiate benign from malignant solid pulmonary nodules ≤ 2 cm: a two-center study

Article References: Li, C., Yang, Y., Zhu, X., Chang, J., Zhou, C., Zhao, J., Wang, C., Tian, L., Jiang, J., Yang, Z., Zhao, G., Huang, Y., Yang, L., & Ye, L. (2026). A nomogram based on clinical and CT features to differentiate benign from malignant solid pulmonary nodules ≤ 2 cm: a two-center study. BMC Medical Imaging. https://doi.org/10.1186/s12880-026-03028-1

Image Credits: AI Generated

DOI: 10.1186/s12880-026-03028-1

Keywords: pulmonary nodules, computed tomography, nomogram, carcinoembryonic antigen, cancer diagnosis, Yunnan Cancer Hospital, risk stratification, medical imaging, based, clinical, features, differentiate