Towards Transparent Decisions: CNN Ensemble with XAI-Driven Interpretations

Authors
Ahwaz Darweesh Hayder 1
Jwan Najeeb Saeed 2
1 Information Technology Management Department, Technical College of Administration, Duhok Polytechnic University, Duhok 42001, Kurdistan Region, Iraq.
2 IT Department, Technical College of Duhok Duhok Polytechnic University,
10.24271/psr.2026.244505
Abstract
Skin cancer remains a global health threat with increasing incidence rates. Accurate and early classification of skin lesions into benign or malignant types is essential for timely treatment and prevention of severe outcomes. In this paper, we present a comprehensive deep learning-based framework that leverages three benchmark datasets—PH2, ISIC (Benign vs Malignant), and HAM10000—using transfer learning and ensemble techniques. Pre-trained models including VGG16, ResNet50, and EfficientNetB4 were fine-tuned on each dataset, and majority voting was employed to combine predictions. The Gradient-weighted Class Activation Mapping (Grad-CAM) was also used to improve visual explainability. The findings demonstrate a notable increase in classification accuracy, surpassing current techniques and reaching over 98% accuracy on certain datasets. This study highlights the impact of hybrid architectures and explainable AI in advancing the state of skin cancer diagnosis systems.
Keywords
XAI
Skin lesion classification
ensemble learning
majority voting
Crossmark
Subjects
Passer Journal of Basic and Applied Sciences
Volume 8, Special Issue
Proceedings of the 5th International Conference on Advanced science & Engineering ICOASE 2026, University of Zakho and Duhok Polytechnique University, 00th – 00th September 2026
August 2026
Pages 135-148