What question did this study set out to answer?

The aim is to create an explainable AI framework that enhances the diagnostic accuracy and interpretability of lung cancer detection.

April 19, 2026Open Access

Towards Transparent AI for Lung Cancer Diagnosis: A Dual-Pipeline Explainable Framework Using Clinical and CT Imaging Data

View Full Paper

Ask AI

Bookmark

Authors

Develops a dual-pipeline framework for lung cancer diagnosis using clinical and CT imaging data, indicating improved interpretability of AI models.

The aim is to create an explainable AI framework that enhances the diagnostic accuracy and interpretability of lung cancer detection.
Develop a dual-pipeline framework using structured clinical data and CT imaging data.
Compare various machine learning models on clinical data for accuracy.
Evaluate deep learning models on CT imaging data using the IQ-OTH/NCCD dataset.
Use SHAP and LIME for analyzing clinical feature importance and Grad-CAM for CT image insights.
Conduct stability and agreement analyses to validate interpretability.
LightGBM achieved 98.39% accuracy and 0.99 ROC-AUC on clinical data.
MobileNetV2 performed best on imaging data with 97% accuracy.
XAI methods provided complementary insights, enhancing AI system transparency.

elmasry et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark

Develops a dual-pipeline framework for lung cancer diagnosis using clinical and CT imaging data, indicating improved interpretability of AI models.