CT radiomics model predicts high-grade lung adenocarcinoma patterns and survival risk

A CT radiomics model identified high-grade patterns (≥20%) in stage I lung adenocarcinoma with an external test AUC of 0.84. High-risk prediction was an independent prognostic factor for recurrence-free survival (HR=3.33) and overall survival (HR=1.98).

• 1181 patients from three centers were analyzed; a combined model integrating radiomics, clinical variables, and CT semantic features outperformed radiomics-only models.
• For recurrence-free survival, the model's high-risk prediction carried a hazard ratio of 3.33 (P < .001) in the binary classification.
• The study was retrospective and lacked prospective external validation.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.