Related Experiment Video
Updated: Jul 13, 2026
08:14
MicroRNA Based Liquid Biopsy: The Experience of the Plasma miRNA Signature Classifier MSC for Lung Cancer Screening
Published on: October 26, 2017
15.8K
Maximizing Lung Cancer Screening in High-Risk Population Leveraging ML-Developed Risk-Prediction Algorithms: Danish
Margrethe Bang Henriksen1, Ole Hilberg2, Christian Juul3
1Department of Oncology, Vejle University Hospital, Vejle, Denmark; Institute of Regional Health Research, University of Southern Denmark, Odense, Denmark.
Clinical Lung Cancer
|July 1, 2025
Summary
The LungFlag AI model shows promise for lung cancer (LC) screening, outperforming existing methods in identifying high-risk individuals, especially those with chronic obstructive pulmonary disease (COPD). Further real-world studies are needed to confirm its clinical value.
Area of Science:
- Oncology
- Medical Informatics
- Artificial Intelligence
Background:
- Early lung cancer (LC) detection is vital for treatment success but faces screening challenges.
- Artificial intelligence (AI) offers innovative risk stratification using routine clinical data.
- The LungFlag model utilizes AI for lung cancer risk assessment.
Purpose of the Study:
- To validate the LungFlag AI model's performance in Danish high-risk populations.
- To assess LungFlag's potential for lung cancer screening.
- To compare LungFlag against the PLCOm2012 risk prediction model.
Main Methods:
- Retrospective analysis of data from two Southern Denmark populations (2013-2021).
- Included LC fast-track patients and chronic obstructive pulmonary disease (COPD) outpatients.
- Compared LungFlag and PLCOm2012 using laboratory results, comorbidities, BMI, and smoking history; analyzed with SHAP values and age.
Main Results:
- LungFlag demonstrated superior performance over PLCOm2012 in Population A (AUC: 0.63 vs. 0.60).
- LungFlag showed slightly higher sensitivity in Population B, with minor differences.
- Key predictors identified were smoking, age, and COPD; LungFlag identified younger high-risk individuals.
Conclusions:
- LungFlag shows potential as a decision-support tool for lung cancer detection, especially in COPD patients.
- The model may aid in identifying individuals for targeted lung cancer screening.
- Prospective studies are required to confirm LungFlag's real-world effectiveness and clinical utility.
