Risk stratification for early-stage NSCLC progression: a federated learning framework with large-small model synergy.

Authors: Huang Z, Feng B, Chen Y, Huang L, Chen Y, Duan X, Chen X, Lin H, Li K, Li Y, Li Q, Ruan X, Chen X, Long W.

DOI: 10.3389/fonc.2025.1719433

Abstract Summary

A new AI-powered risk stratification system called FedCPI significantly outperforms traditional methods in predicting progression of early-stage lung cancer. Tested on 926 patients across multiple centers, this deep learning model accurately identifies high-risk cases requiring aggressive treatment while sparing low-risk patients from unnecessary interventions. Remarkably, the system also proved effective for gastric and endometrial cancers.

Why Brain? 🧠

AI model using federated learning accurately predicts early-stage lung cancer progression risk, outperforming clinical methods, enabling personalized treatment decisions and reducing overtreatment burden.

License: CC BY.

The image is AI-generated for illustrative purposes only. Courtesy of Midjourney.