Open-source Sybil AI detects lung cancer risk effectively in Asian heavy smokers on low-dose CT.
Key Details
- 1Sybil is an open-source deep learning model that uses low-dose CT (LDCT) to predict lung cancer risk.
- 2Study included 18,057 Asian patients (with at least one follow-up scan) from 2004–2021.
- 3Sybil achieved overall AUC of 0.91 (1-year) and 0.74 (6-year); heavy smoking subgroup AUCs were 0.94 (1-year, visible cancers) and 0.7 (6-year, future cancers).
- 4Performance was weaker (AUC 0.56 for 6-year, future cancers) in never/light-smoking subgroup.
- 5The model may help optimize follow-up intervals in lung cancer screening programs.
- 6External validation is needed before widespread adoption.
Why It Matters
This research demonstrates promising performance of an AI tool for individualized lung cancer risk estimation using routine LDCT in underrepresented Asian populations, potentially supporting more tailored screening protocols. Reliable, open-source risk assessment models like Sybil may improve early detection and resource allocation in lung cancer care.
Source
AuntMinnie
Related News
•AuntMinnie
Study: Computer Vision Models Best LLMs in Chest CT Breast Abnormality Detection
Computer vision models (CVMs) surpass large language models (LLMs) in accurately labeling incidental breast abnormalities on chest CT scans.
•Cardiovascular Business
Radiology Maintains Lead in FDA-Cleared AI Algorithms, Cardiology Follows
Radiology remains the top specialty for FDA-cleared AI, with cardiology as a strong second, particularly in cardiovascular imaging.
•AuntMinnie
Deep Learning Models Rival Radiologists for Pancreatic Cancer Detection on CT
Deep-learning models achieved comparable or superior accuracy to experienced radiologists in detecting pancreatic cancer on CT scans, especially for small tumors.