Texture Analysis of Subsolid Nodules Detected on Computed Tomography for Differentiating Minimally Invasive From...

Area of Science:

• Radiology and Medical Imaging
• Oncology
• Pulmonary Medicine

Background:

• Lung cancer is a leading cause of cancer mortality globally.
• Subsolid nodules (SSNs) on CT scans are early indicators of lung cancer.
• Distinguishing between minimally invasive adenocarcinoma (MIA) and invasive adenocarcinoma (IAC) is crucial for treatment planning.

Purpose of the Study:

• To evaluate the utility of computed tomography (CT) texture features in differentiating MIA from IAC in SSNs.
• To assess the diagnostic performance of CT texture analysis for lung adenocarcinoma subtypes.
• To explore noninvasive methods for preoperative risk stratification of SSNs.

Main Methods:

• Retrospective analysis of CT scans from patients with lung adenocarcinoma SSNs.
• Extraction of morphologic characteristics (size, shape, density, lobulation) and CT texture features (histogram parameters, GLCM metrics) using MaZda software.
• Categorization of patients into MIA and IAC groups based on preoperative pathology.

Main Results:

• Significant differences in morphologic features were observed between IAC and MIA groups (e.g., size, shape, lobulation).
• Key CT texture features, including mean attenuation and entropy, effectively distinguished IAC from MIA.
• A predictive nomogram integrating these features demonstrated high diagnostic accuracy (AUC 0.831 in training, 0.893 in validation).

Conclusions:

• Quantitative CT texture analysis, particularly using entropy and mean attenuation, is a valuable tool for differentiating MIA and IAC in SSNs.
• This noninvasive approach improves preoperative risk stratification.
• The findings support personalized surgical planning for lung adenocarcinoma.