Explainable AI Technique in Lung Cancer Detection Using Convolutional Neural Networks

TL;DR

This study tackles automated lung cancer detection from chest CT while emphasizing interpretability. It compares a custom CNN with fine-tuned backbones DenseNet121, ResNet152, and VGG19, incorporating SHAP explanations to visualize evidence for predictions on the multi-class IQ-OTH/NCCD dataset. DenseNet121 achieves the strongest balance of precision, recall, and F1, while ResNet152 offers the highest accuracy, and SHAP maps provide radiologically plausible attributions. The work demonstrates a practical, explainable framework suitable for screening in settings with limited radiology resources, and outlines concrete paths for broader validation and model enhancement. The combination of model diversity, cost-sensitive learning, and post-hoc explanations advances trustworthy AI-assisted screening in clinical workflows.

Abstract

Early detection of lung cancer is critical to improving survival outcomes. We present a deep learning framework for automated lung cancer screening from chest computed tomography (CT) images with integrated explainability. Using the IQ-OTH/NCCD dataset (1,197 scans across Normal, Benign, and Malignant classes), we evaluate a custom convolutional neural network (CNN) and three fine-tuned transfer learning backbones: DenseNet121, ResNet152, and VGG19. Models are trained with cost-sensitive learning to mitigate class imbalance and evaluated via accuracy, precision, recall, F1-score, and ROC-AUC. While ResNet152 achieved the highest accuracy (97.3%), DenseNet121 provided the best overall balance in precision, recall, and F1 (up to 92%, 90%, 91%, respectively). We further apply Shapley Additive Explanations (SHAP) to visualize evidence contributing to predictions, improving clinical transparency. Results indicate that CNN-based approaches augmented with explainability can provide fast, accurate, and interpretable support for lung cancer screening, particularly in resource-limited settings.

Figures (9)

• Figure 1: Representative CT samples for (left to right): Benign, Normal, Malignant.
• Figure 2: Class distribution in IQ-OTH/NCCD dataset.
• Figure 3: Learning curves for Custom CNN.
• Figure 4: Learning curves for DenseNet121.
• Figure 5: Learning curves for ResNet152.