Related Experiment Video
Updated: Jul 11, 2025
10:26
Author Spotlight: A 3D Digital Model for the Diagnosis and Treatment of Pulmonary Nodules
Published on: May 19, 2023
1.9K
Prediction of subsolid pulmonary nodule growth rate using radiomics
Zong Jing Ma1, Zhuang Xuan Ma1, Ying Li Sun1
1Department of Radiology, Huadong Hospital, Fudan University, Shanghai, 200040, China.
BMC Medical Imaging
|November 7, 2023
Summary
A new model combining radiomics and radiographic features accurately predicts subsolid pulmonary nodule (SSN) growth rates. This advancement aids in optimizing clinical decisions and improving long-term management for patients with SSNs.
Area of Science:
- Radiology
- Medical Imaging Analysis
- Oncology
Background:
- Accurate assessment of pulmonary nodule growth rate is crucial for managing subsolid pulmonary nodules (SSNs).
- Clinical follow-up of SSNs requires reliable methods to predict their growth trajectory.
Purpose of the Study:
- To develop and validate a predictive model for the growth rate of subsolid pulmonary nodules.
- To compare the performance of a combined radiomics-radiographic model against traditional methods.
Main Methods:
- Retrospective analysis of 273 growing SSNs from 857 CT scans.
- Development of predictive models using radiomics and radiographic features (nodule density, spiculation, vascularity).
- Evaluation of model performance using Area Under the Curve (AUC) on training and validation sets.
Main Results:
- A nomogram integrating radiomics and radiographic features achieved AUCs of 0.928 (training) and 0.905 (validation).
- The combined model significantly outperformed models using only radiographic features (AUCs ~0.67) or radiomics alone (AUCs ~0.85).
Conclusions:
- A nomogram combining radiomics and radiographic features provides accurate prediction of SSN growth rates.
- This model can enhance clinical treatment decisions and long-term management strategies for SSN patients.
