A new study has shown that artificial intelligence (AI) can significantly reduce the workload for radiologists while maintaining accuracy in detecting lung cancer.
Researchers from the University of Liverpool and the Research Institute for Diagnostic Accuracy in the Netherlands found that AI can rule out negative lung cancer screenings, allowing radiologists to focus on high-risk cases.
Published in the European Journal of Cancer, the study suggests that AI integration could make lung cancer screening programs more efficient, cost-effective, and scalable.
The Importance of Early Lung Cancer Detection
Lung cancer is a major health concern, affecting more than 48,000 people in the UK each year. The disease is often diagnosed at a late stage, making it difficult to treat. However, low-dose CT (LDCT) scans have been shown to improve survival rates by detecting lung cancer in high-risk individuals before symptoms appear.
The UK Lung Cancer Screening (UKLS) trial has already demonstrated that LDCT screenings save lives, but the process is time-consuming for radiologists, who must carefully examine thousands of scans—many of which turn out to be negative.
How AI Can Help
In this study, researchers tested an AI tool developed by Coreline Soft, Co Ltd., South Korea, using data from the UKLS trial. The AI was programmed to analyze LDCT scans and identify those without significant lung nodules, which account for the majority of cases.
The results were striking:
- The AI successfully ruled out negative scans, reducing the number of cases that required a radiologist’s review.
- This could cut the workload of radiologists by up to 79%.
- No confirmed lung cancer cases were missed—all cancers were flagged by the AI for further examination.
By filtering out scans that do not require expert evaluation, AI allows radiologists to focus their expertise on complex cases, making the screening process more efficient and effective.
A Step Toward AI in Lung Cancer Screening
Professor John Field, lead author and Professor of Molecular Oncology at the University of Liverpool, highlighted the potential impact of this research:
“Lung cancer screening with low-dose CT is highly beneficial, but it also presents logistical and financial challenges. Our study suggests that AI could help make screening programs more efficient while maintaining diagnostic confidence.”
Co-lead author Professor Matthijs Oudkerk, Chief Scientific Officer of the Institute for Diagnostic Accuracy, called this a milestone for AI validation in real-world lung cancer screening. The study was unique because it used real patient data with confirmed lung cancer diagnoses and over five years of follow-up, making it one of the most rigorous AI studies to date.
Implications for Healthcare
As lung cancer screening programs expand globally, AI tools like this could:
- Improve efficiency by reducing the number of scans radiologists need to review.
- Lower costs by streamlining the screening process.
- Ensure timely diagnoses by allowing experts to focus on cases that truly need attention.
While further validation is needed, this research paves the way for AI integration into routine lung cancer screening, helping to optimize healthcare resources and potentially saving more lives.
The research findings can be found in the European Journal of Cancer.
Copyright © 2025 Knowridge Science Report. All rights reserved.
