Model Combining Clinical, Radiomics, and Deep Learning Predicts 5-Year Growth of Subsolid Pulmonary Nodules

Combined clinical, radiomics, and deep learning model predicted 5-year growth of pulmonary subsolid nodules with AUC 0.81 (95% CI 0.70–0.92) at external validation.

• The study included 574 patients from two hospitals who had subsolid nodules with at least 5 years of follow-up.
• A nomogram was developed to estimate individual nodule growth probability.
• Risk stratification by the model showed significantly different nodule stability rates between high- and low-risk groups.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.