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

Area of Science:

• Radiology
• Oncology
• Medical Imaging

Background:

• Visceral pleural invasion (VPI) is a critical prognostic factor in non-small cell lung cancer (NSCLC).
• Accurate preoperative prediction of VPI is essential for treatment planning and patient management.
• CT radiomics and subjective features offer potential for non-invasive VPI assessment.

Purpose of the Study:

• To evaluate the combined diagnostic value of contrast-enhanced CT radiomic features, CT subjective features, and clinical information.
• To develop and validate a predictive model for preoperative visceral pleural invasion (VPI) in non-small cell lung cancer (NSCLC).

Main Methods:

• A study involving 326 NSCLC patients, with data split into training and validation sets.
• Construction of a radiomics signature, a clinical model, and an integrated nomogram combining radiomics and clinical features.
• Performance evaluation using Area Under the ROC Curve (AUC) and calibration curves.

Main Results:

• Ten radiomics and four CT subjective features were significantly associated with VPI.
• The integrated nomogram achieved the highest AUC values (0.937 in training, 0.852 in validation) compared to individual models.
• Good consistency was observed between predicted and actual VPI status via calibration curves.

Conclusions:

• The combined approach using CT radiomics, subjective features, and clinical data demonstrates high accuracy in predicting VPI in NSCLC.
• This integrated model can significantly aid in preoperative clinical decision-making for NSCLC patients.
• The findings support the utility of advanced CT analysis for staging and treatment guidance in NSCLC.