Objective:

To investigate the utility of intranodular and perinodular radiomics features derived from non-contrast CT in differentiating pulmonary cryptococcosis (PC) from lung adenocarcinoma (LAC).

Approach:

Key Findings:

• The combined radiomics model achieved the highest AUC of 0.936 in the training set and 0.922 in the testing set.
• Sensitivity, specificity, and accuracy for the combined model were 0.838, 0.898, and 0.859 in the training set, and 0.854, 0.808, and 0.892 in the testing set, respectively.
• The combined model significantly outperformed the clinical model (p = 0.005) but not the intranodular or perinodular models (p > 0.05).

Interpretation:

The study demonstrates that integrating intranodular and perinodular radiomics features enhances diagnostic accuracy in distinguishing PC from LAC, addressing a significant clinical challenge.

Limitations:

• Retrospective design may introduce selection bias.
• Limited to patients with solitary solid nodules or masses, which may not represent all cases of PC or LAC.
• The study was conducted at two centers, which may affect generalizability.

Conclusion:

A combined radiomics model integrating intranodular and perinodular features can effectively improve diagnostic accuracy in differentiating pulmonary cryptococcosis from lung adenocarcinoma.

Attribution Notice

This content is an AI-generated, fully rewritten summary based on a published scholarly article. It does not reproduce the original text and is not a substitute for the original publication. Readers are encouraged to consult the source for full context, data, and methodology.