A new study validates the Sybil AI model for predicting lung cancer risk using low-dose CT in a predominantly Black cohort at an urban safety-net hospital.
Key Details
- 1Study presented at the 2025 IASLC World Conference on Lung Cancer.
- 2Validation conducted at UI Health on 2,092 baseline low-dose CT scans from 2014-2024.
- 3Cohort primarily Non-Hispanic Black (62%), with 13% Hispanic and 4% Asian.
- 4Sybil demonstrated strong Area Under the Curve (AUC): 0.94 (1 yr), 0.90 (2 yr), 0.86 (3 yr), 0.85 (4 yr), 0.80 (5 yr), 0.79 (6 yr).
- 5Results remained strong after restricting to Black participants and excluding recent diagnoses.
- 6The Sybil Implementation Consortium will advance to prospective clinical trials for clinical workflow integration.
Why It Matters
Demonstrating reliable AI performance in racially and socioeconomically diverse settings is crucial for equitable lung cancer screening. This validation may facilitate broader adoption and help address longstanding disparities in lung cancer outcomes.
Source
EurekAlert
Related News
•EurekAlert
Penn Medicine's Percival AI Detects Hidden Health Risks in CT Scans
Penn Medicine's Percival AI leverages CT scans to uncover unseen health risks and disease patterns, expanding the utility of radiological imaging.
•EurekAlert
Rethinking Multimodal Cardiovascular AI: When Models Should Defer Predictions
A commentary in Medicine Discovery argues cardiovascular AI models must learn to defer predictions when key modalities are missing, as missingness and asynchrony are clinical realities.
•EurekAlert
New Study Urges 'Green AI' Practices in Medical Imaging and Healthcare
A new review highlights the urgent need for sustainable AI practices in healthcare, with emphasis on environmental impacts from applications like medical imaging.