The study evaluated over 21,000 individuals aged 50-80 who underwent self-initiated LDCT screening from 2009 to 2021, tracking their outcomes until 2024.

The increasing disconnect between risk factors and screening practices in Asia raises concerns, especially as the region accounts for a significant proportion of new lung cancer cases and deaths globally.

Dr. Yeon Wook Kim from Seoul National University Bundang Hospital emphasized that Sybil can identify low-risk individuals who may not require further screening, as well as those at higher risk who should continue with screening.

This model could guide personalized strategies for individuals who have undergone LDCT screening but lack further recommendations for follow-up or additional screening.

Researchers have developed a deep learning model named Sybil that can predict lung cancer risk from a single low-dose chest CT (LDCT) scan, a breakthrough presented at the ATS 2025 International Conference in San Francisco.

The model aims to improve personalized lung cancer screening strategies, which is crucial in Asia where lung cancer rates are rising among nonsmokers.

Sybil has the potential to guide decisions on whether individuals should continue or discontinue lung cancer screening based on their personalized risk assessments.

Developed using data from the National Lung Screening Trial (NLST) by researchers from MIT and Harvard Medical School, Sybil seeks to tailor lung cancer screening strategies to individual risk profiles.

Despite increasing lung cancer rates among never-smokers, current international guidelines do not recommend screening for this demographic, particularly concerning given that Asia accounts for over 60% of global lung cancer cases.

Sybil has shown the ability to effectively predict lung cancer diagnoses at one and six years, including among individuals who have never smoked, highlighting its potential for independent risk assessment beyond traditional demographic factors.

Researchers aim to conduct these prospective studies to confirm Sybil's effectiveness in clinical settings and to explore its prediction capabilities for other significant outcomes.

Future prospective studies are planned to validate Sybil's clinical application and enhance its predictive capabilities regarding lung cancer-specific mortality.