Summary
New AI model Sybil predicts future lung cancer risk from a single CT scan, offering hope for personalized screening, especially for Asian never-smokers.
AI Technology Shows Promise in Predicting Lung Cancer Risk
Researchers have developed an artificial intelligence model that can predict a person’s future risk of developing lung cancer from just one chest CT scan. The breakthrough technology, presented at the ATS 2025 International Conference, could revolutionize how doctors screen for lung cancer, especially in populations not typically considered high-risk.
The AI model, called Sybil, was created by scientists from MIT and Harvard Medical School using data from the National Lung Screening Trial. What makes Sybil special is its ability to analyze a single low-dose chest CT scan and identify patterns that human eyes might miss, predicting who is likely to develop lung cancer in the future.
This technology could be particularly valuable in Asia, where lung cancer patterns differ from Western countries. Currently, more than 60% of new lung cancer cases worldwide occur in Asia, and surprisingly, many of these patients have never smoked. Traditional screening guidelines, which focus mainly on smokers and former smokers, miss these at-risk individuals.
Dr. Yeon Wook Kim from Seoul National University Bundang Hospital, who led the research, explains that Sybil can identify two important groups: people at low risk who might safely skip future screenings, and those at higher risk who should continue regular monitoring. This personalized approach could make screening more efficient and effective.
The current disconnect between who gets screened and who actually develops lung cancer is a serious problem. Many people in Asia undergo screening outside of guidelines, often paying out of pocket, without clear evidence about who truly needs it. Sybil could help doctors make more informed decisions about screening recommendations.
This AI breakthrough represents a significant step toward more personalized medicine, potentially saving lives by catching lung cancer early in populations previously overlooked by traditional screening criteria.
