Predicting recurrence within 5 years in Early-Stage lung adenocarcinoma with micropapillary and solid patterns

2026-06-10

PRE

Zhongjie Wang

Jie Chen

Yuanyuan Xu

Tengzhe Lin

Chao Chen

Yanming Shen

Jin Huang

Shaojun Xu *

Shuchen Chen *

DOI：10.1016/j.ijmedinf.2026.106542

Original

Original request for help

Save

Abstract

En 中文

• This study developed and validated a Neural Network machine learning model that effectively predicts 5-year recurrence risk in early-stage lung adenocarcinoma patients with high-risk micropapillary or solid pathological patterns, achieving an AUC of 0.775 in the validation set. • Consolidation-to-tumor ratio (CTR) and surgical procedure were identified as the two most important predictive factors, with CTR ≥ 0.5 being the strongest predictor of recurrence. • The model provides clinically actionable insights, suggesting that patients with multiple high-risk factors may benefit from more aggressive treatment strategies despite having early-stage disease.

Predicting recurrence within 5 years in Early-Stage lung adenocarcinoma with micropapillary and solid patterns

Abstract

Journal

Organization

Cited Papers

Citing Papers