Quantitative Prediction of Visceral Pleural Invasion in Non-Small Cell Lung Cancer Using Preoperative...

To investigate the value of combining contrast-enhanced CT radiomic features, CT subjective features, and clinical information for preoperative prediction of visceral pleural invasion (VPI) in non-small cell lung cancer (NSCLC).

A descriptive study. Place and Duration of the Study: Department of Radiology, The Second Affiliated Hospital of South China University of Technology, Guangzhou, China, from March 2022 to March 2024.

This study included 326 NSCLC patients who underwent surgery at two hospitals, which were used as the training and validation sets. A radiomics signature was constructed by extracting and selecting tumour radiomic features. Clinical characteristics were selected through univariate and multivariate logistic regression to construct a clinical model. A nomogram was developed by integrating the radiomics signature with clinical features. The models' performance in predicting VPI was evaluated using the area under the ROC curve and calibration curve.

Ten radiomics features and four CT subjective features were identified as significantly associated with VPI. In the training and validation sets, the AUC values for the clinical model, radiomics signature, and nomogram were 0.854, 0.898, 0.937, and 0.706, 0.837, 0.852, respectively. The nomogram exhibited the highest AUC in both sets, and the calibration curve showed good consistency.

The integration of CT radiomics features, CT subjective features, and clinical information can effectively predict VPI in NSCLC patients, thereby facilitating clinical decision-making.

Non-small cell lung cancer, Visceral pleural invasion, CT radiomics.