Related Experiment Video
Updated: Jul 5, 2026
04:04
Endobronchial Ultrasound-guided Intratumoral Injection of Cisplatin for the Treatment of Isolated Mediastinal Recurrence of Lung Cancer
Published on: February 12, 2017
Utilizing Clinicopathological and Radiomic Features for Risk Stratification of Lung Cancer Recurrence
Wai Lone J Ho1, Nikolai Fetisov2, Lawrence O Hall2
1University of South Florida, Morsani College of Medicine, Tampa, Florida (W.L.J.H.).
Academic Radiology
|May 16, 2025
Summary
Radiomics combined with clinical data accurately predicts recurrence in non-small cell lung cancer (NSCLC) patients after surgery. This approach improves risk stratification, identifying patients with significantly higher recurrence risk.
Area of Science:
- Oncology
- Radiology
- Medical Imaging Analysis
Background:
- Accurate prediction of recurrence risk is crucial for managing patients with surgically resected non-small cell lung cancer (NSCLC).
- Traditional staging systems may not fully capture individual patient risk profiles.
- Novel approaches are needed to enhance prognostic accuracy.
Purpose of the Study:
- To develop and validate a model predicting recurrence risk in NSCLC patients using radiomic features and clinicopathological factors.
- To compare the performance of radiomic, clinical, and combined radiomic-clinical models in predicting recurrence.
Main Methods:
- Analysis of 293 patients with surgically resected stage IA-IIIA NSCLC, stratified into development and test cohorts.
- Extraction of 107 radiomic features from pre-treatment CT scans using pyRadiomics.
- Development of radiomic, clinical, and radiomic-clinical models using logistic regression and evaluation via Area Under the Curve (AUC).
Main Results:
- The radiomic-clinical model achieved the highest predictive performance on the test set (AUC 0.77), outperforming radiomic (0.76), clinical (0.71), and TNM stage (0.70) models.
- Patients stratified into a high-risk group by the radiomic-clinical model had a five-fold higher recurrence risk (p<0.01).
- Key predictors included lymph node metastasis, "GLDM Large Dependence Emphasis" texture, and "Elongation" shape features.
Conclusions:
- Radiomics analysis, when integrated with clinicopathological features, offers an effective strategy for recurrence risk stratification in surgically treated NSCLC.
- This combined approach enhances prognostic accuracy beyond traditional methods.
- The findings support the use of radiomics in personalized treatment planning for NSCLC.
