New research has been presented during the American Thoracic Society 2025 International Conference. This new deep learning model was able to predict lung cancer risk by using low-dose CT screening exam. The DL models was tested on over 21000 scans of patients aging 50 to 80 who had undergone LDCT. The scans took place between 2009 and 2021 and followed through 2024.
In this research, the utilize Sybil, developed with the help of data acquired from National Lung Cancer Screening Trial (NLST). Sybil was able to predict lung cancer based on imaging alone. It performed well at predicting the risk at both one and six year marks. Sybil was proven to be accurate in identifying true low-risk individuals. Sybil was developed in the year 2023.
Sybil was firstly trained by feeding LDCT images which are largely absent of any signs of cancer. This AI tool was able to give hundreds of scans with visible cancerous tumors. In this Korean study, analyzes were performed for individuals with smoking histories ranging from over 20 pack/year to never smokers. The results hold the promise of helping to regularize lung cancer screening in Asia. Asia bears the highest burden of lung cancer and accounts for more than 60% of new cases and related deaths. “In Korea, more than 85% of female lung cancer patients are non-smokers. As a result, increasing attention has been given to evaluating the effectiveness of lung cancer screening, or LCS, in traditionally low-risk populations in Asia”, Kim said.
Sybil could also be used to develop personalized strategies for patients, while further validation will be needed to confirm its potential for clinical use. Researchers are actively working on expanding Sybil’s uses into other personalized health applications as well.
The information shared in this blog is for educational purposes only. You should always consult your medical practitioner for any medical needs.
Subscribe to our newsletter or follow us on Facebook or X today to stay informed and never miss an update!