A research team at Samsung Medical Center has developed an artificial intelligence (AI) model, RADAR CARE, that can predict the risk of recurrence in patients with early-stage non-small cell lung cancer (NSCLC) up to one year in advance. Non-small cell lung cancer (NSCLC) accounts for approximately 85% of all lung cancers.
Patients in the early stages primarily undergo surgery; however, even within the same stage, the risk of recurrence varies significantly, limiting opportunities for personalized management. Accordingly, the team led by Professors Hongkwan Kim and Hyeonae Jeong developed a Transformer-based deep learning model using clinical, pathological, and imaging data from 14,177 surgical patients treated between 2008 and 2022. The model achieved an accuracy of 0.823 at the time of surgery and 0.854 when follow-up data were included. The resulting RADAR score categorized patients into low-, intermediate-, and high-risk groups. The one-year recurrence rate was 10% in the high-risk group compared with 1% in the low-risk group. The score predicted prognosis more accurately than stage, with patients in the high-risk group exhibiting up to a 9.67-fold higher risk of recurrence and death.
The research team explained that personalized strategies are feasible, allowing adjustment of treatment intensity and follow-up intervals based on changes in RADAR scores. This study was published in JCO Precision Oncology, a journal of the American Society of Clinical Oncology.
AI Predicts Recurrence in Early-Stage Lung Cancer Patients One Year in Advance
Meet SMC Doctors
Hong Kwan Kim
Professor of Thoracic Surgery, Samsung Medical Center, Sungkyunkwan University
Esophageal Cancer, Lung Cancer, Robotic Surgery for Esophageal Cancer, Minimally Invasive Esophageal Cancer Surgery, Thoracoscopic Surgery
Hyun Ae Jung
Associate Professor, Department of Hematology & Oncology, Samsung Medical Center, Sungkyunkwan University
Lung Cancer, Head and Neck Cancer, Esophageal Cancer, Brain Tumor