Samsung Medical Center Develops AI Model to Predict Lung Cancer Recurrence

The model, called RADAR CARE, integrates multiple data types, including clinical records, pathology test results, CT scans, and patient-specific medical information.

Samsung Medical Center (SMC) has introduced an AI model designed to predict the recurrence risk of non-small cell lung cancer (NSCLC) up to a year in advance.

The model, called RADAR CARE (Real-time Risk-Adapted Surveillance Comprehensive Strategy AI Model for Early-Stage NSCLC), integrates multiple data types, including clinical records, pathology test results, CT scans, and patient-specific medical information. It assigns patients into three risk categories: low, intermediate, and high.

Developed using data from 14,177 early-stage NSCLC patients who underwent surgery at SMC between 2008 and 2022, the transformer-based deep learning model demonstrated that recurrence risks vary significantly within the same cancer stage.

In the study, high-risk patients had a 10% recurrence rate, compared with 5% in the intermediate group and 1% in the low-risk group. The findings also showed that patients with stage 1 disease and high-risk scores may have higher recurrence rates than those with stage 3 disease and low-risk scores. Overall, the likelihood of recurrence or death was 3.59 times higher in the intermediate-risk group and 9.67 times higher in the high-risk group compared with low-risk patients. Stage-specific analyses indicated elevated risks in high-risk patients across all stages, including 5.83 times higher in stage 1, 1.75 times higher in stage 2, and 1.84 times higher in stage 3.

Dr. Kim Hong-Kwan, professor in the Department of Thoracic and Cardiovascular Surgery at SMC and lead of the RADAR CARE study, noted, "This is because existing staging classifications alone make it difficult to accurately predict patient prognosis." He added, "Within the same cancer stage, recurrence risks can still vary among patients."

Currently, follow-up examinations for NSCLC patients are scheduled every three to six months without risk-based differentiation. Prof Kim highlighted that the AI model is now applied post-surgically at SMC to guide personalized treatment planning for patients with NSCLC.

The development aligns with broader trends in healthcare, where AI is increasingly being used to predict cancer recurrence. Recent efforts include an AI-driven scoring system in Singapore for hepatocellular carcinoma, achieving 82% predictive accuracy, and Lunit's collaboration with AstraZeneca for genomic testing to anticipate NSCLC outcomes.

Stay tuned for more such updates on Digital Health News