MRI-based radiomics analysis in differentiating solid non-small-cell from small-cell lung carcinoma: a pilot study.

Area of Science:

• Radiology
• Oncology
• Medical Imaging Analysis

Background:

• Distinguishing between non-small-cell lung carcinoma (NSCLC) and small-cell lung carcinoma (SCLC) is crucial for treatment planning.
• Conventional MRI features have limitations in accurately differentiating these lung cancer subtypes.

Purpose of the Study:

• To evaluate the efficacy of a radiomics signature derived from T2-weighted (T2W) magnetic resonance imaging (MRI) in differentiating solid NSCLC from SCLC.
• To compare the diagnostic performance of radiomics signature with conventional MRI morphological features.

Main Methods:

• A retrospective study included 152 patients with NSCLC (n=125) or SCLC (n=27).
• Radiomics features were extracted from T2W MRI scans using a 3T scanner.
• A radiomics model was constructed using the least absolute shrinkage and selection operator (LASSO) logistic regression, and its performance was assessed using receiver operating characteristic (ROC) curve analysis.

Main Results:

• Five optimal radiomics features were selected to build the radiomics signature.
• The radiomics signature and specific conventional MRI features (pleural indentation, lymph node enlargement) were independent predictors.
• The area under the ROC curve (AUC) for the radiomics model was 0.85, and for the combined nomogram model was 0.90, outperforming conventional MRI features (AUC=0.69).

Conclusions:

• The T2W MRI-based radiomics signature demonstrates significant potential as a non-invasive tool for distinguishing between solid NSCLC and SCLC.
• Integrating radiomics with conventional MRI features further enhances diagnostic accuracy.