Volume 8 • Issue 1 • PP: 05–12 • 2026
XAI-DermNet: A Dual-Modality Deep Learning and Explainable AI Fusion Framework for Transparent Skin Lesion Diagnosis from Dermoscopic Images
View PDF Abstract
The advancement of trustworthy diagnostic tools in dermatological automation is hindered by the limited transparency of current deep learning systems, which function as opaque models and impede clinical acceptance. This research presents a novel intelligent framework for skin lesion analysis that unites deep learning methodologies with explainable artificial intelligence (XAI) principles to address this interpretability deficit. The proposed approach utilizes a transfer-learned ResNet50 architecture for robust image classification, coupled with Local Interpretable Model-agnostic Explanations (LIME) to furnish clear, visual justifications for the model’s outputs. Performance assessment on the HAM10000 benchmark yielded a classification accuracy of 94.3%, with a validation accuracy of 91.8%. Concurrently, the LIME framework effectively identified and visualized diagnostically critical features in the lesion images, thereby elucidating the model’s reasoning process for medical practitioners. These findings confirm that augmenting high-performance deep learning with post-hoc explanatory techniques yields a credible and understandable diagnostic instrument, thereby promoting clinician trust and facilitating data-informed medical judgments. Subsequent developments will prioritize scalable cloud implementation, interoperability with healthcare information systems, extension to underrepresented lesion categories, and rigorous evaluation in diverse clinical environments.
Keywords
References
[1] N. Codella, V. Rotemberg, P. Tschandl, M. E. Celebi, S. Dusza, M. A. Marchetti, and A. Halpern, “Skin lesion analysis toward melanoma detection: A challenge at the International Skin Imaging Collaboration (ISIC),” IEEE Journal of Biomedical and Health Informatics, vol. 23, no. 2, pp. 519–526, 2019.
[2] A. Esteva, B. Kuprel, R. A. Novoa, J. Ko, S. M. Swetter, H. M. Blau, and S. Thrun, “Dermatologist-level classification of skin cancer with deep neural networks,” Nature, vol. 542, no. 7639, pp. 115–118, 2017.
[3] P. Tschandl, C. Rosendahl, and H. Kittler, “The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions,” Scientific Data, vol. 5, no. 1, p. 180161, 2018.
[4] H. L. Gururaj, N. Manju, A. Nagarjun, V. M. Aradhya, and F. Flammini, “DeepSkin: A deep learning approach for skin cancer classification,” IEEE Access, vol. 11, pp. 50 205–50 214, 2023.
[5] Y. C. Lee, S. H. Jung, and H. H.Won, “WonDerM: Skin lesion classification with fine-tuned neural networks,” arXiv preprint arXiv:1808.03426, 2018.
[6] A. Mahbod, G. Schaefer, C. Wang, R. Ecker, and I. Ellinge, “Skin lesion classification using hybrid deep neural networks,” in ICASSP 2019 – IEEE International Conference on Acoustics, Speech and Signal Processing, 2019, pp. 1229–1233.
[7] J. Saeed and S. Zeebaree, “Skin lesion classification based on deep convolutional neural network architectures,” Journal of Applied Science and Technology Trends, vol. 2, no. 1, pp. 41–51, 2021.
[8] A. S. Shete, A. S. Rane, P. S. Gaikwad, and M. H. Patil, “Detection of skin cancer using CNN algorithm,” International Journal, vol. 6, no. 5, pp. 1–4, 2021.
[9] M. Ahmed, S. Sadek, and A. Mohammed, “Advancements in ResNet-based architectures for multi-class skin lesion classification,” Journal of Medical Imaging and Health Informatics, vol. 14, no. 2, pp. 210–225, 2024.
[10] F. Altini, D. Cascio, and G. Taormina, “A comparison of deep learning performance against healthcare professionals in detecting diseases from medical imaging,” IEEE Journal of Biomedical and Health Informatics, vol. 28, no. 1, pp. 112–124, 2024.
[11] D. Brown and S. Green, “Fairness and bias in AI-driven dermatology: A review of current datasets,” Science Advances, vol. 10, no. 4, p. eadj2024, 2024.
[12] L. Chen, J. Huang, and K. Smith, “Improving clinical trust: A comparative study of LIME vs. Grad-CAM in dermatological AI,” Journal of Biomedical Informatics, vol. 145, p. 104456, 2023.
[13] R. Das and M. Sen, “A hybrid deep learning framework for skin lesion classification on imbalanced datasets,” Expert Systems with Applications, vol. 213, p. 118897, 2023.
[14] V. Gupta, P. Sharma, and R. Kumar, “Real-time skin cancer detection on edge devices using optimized ResNet50,” Sensors, vol. 24, no. 1, p. 312, 2024.
[15] A. Ibrahim, M. Hassan, and S. Ali, “Assessing the reliability of LIME for skin lesion explanations in clinical settings,” JMIR Medical Informatics, vol. 11, p. e45892, 2023.
[16] Z. Khalid, A. Rehman, and M. A. Khan, “Future of explainable AI in personalized healthcare: From theory to practice in dermatology,” Artificial Intelligence Review, vol. 58, no. 1, pp. 45–68, 2025.
[17] A. Khan, M. Rahim, and S. Hussain, “Deep learningbased automated skin lesion segmentation and classification: A 2024 perspective,” IEEE Reviews in Biomedical Engineering, vol. 17, pp. 88–102, 2024.
[18] B. Lopez, M. Garcia, and A. Fernandez, “Multimodality data fusion for skin cancer diagnosis: Deep learning meets clinical history,” Medical Image Analysis, vol. 91, p. 103012, 2024.
[19] H. Malik, S. Farooq, and A. Jamil, “A novel ResNet- 50 based approach for basal cell carcinoma detection,” Biomedical Signal Processing and Control, vol. 85, p. 104832, 2023.
[20] P. Mishra, R. Gupta, and A. Pratap, “Ensemble learning with ResNet50 and InceptionV3 for high-accuracy skin cancer screening,” Journal of Digital Imaging, vol. 36, no. 4, pp. 1450–1465, 2023.
[21] T. Nguyen and P. Vo, “Lightweight CNN architectures for early melanoma detection: A 2025 outlook,” Pattern Recognition Letters, vol. 182, pp. 12–20, 2025.
[22] K. Patel, A. Kumar, and J. Singh, “Beyond black-box models: Applying LIME and SHAP for interpretable skin cancer detection,” Diagnostics, vol. 13, no. 4, p. 725, 2023.
[23] M. Roberts, D. Driggs, and M. Thorpe, “Common pitfalls and recommendations for deep learning in skin cancer: A 2023 update,” The Lancet Digital Health, vol. 5, no. 2, pp. e94–e105, 2023.
[24] S. Sharma and R. Kumar, “Explainable AI for dermatology: A systematic review of XAI techniques in skin cancer diagnosis,” Artificial Intelligence in Medicine, vol. 148, p. 102765, 2024.
[25] S. Singh, A. K. Mondal, and A. Kumar, “Fusion of ResNet and Vision Transformers (ViT) for enhanced skin lesion analysis,” Computers in Biology and Medicine, vol. 168, p. 107784, 2024. [26] J. R. Smith and L. White, “The role of transfer learning in rare skin disease identification,” Nature Machine Intelligence, vol. 5, no. 3, pp. 220–231, 2023.
[27] K. Sun, L. Zhang, and H. Wang, “Robust skin lesion classification under noisy labels using deep learning,” Scientific Reports, vol. 13, p. 11452, 2023.
[28] Y. Wang, X. Liu, and Z. Zhang, “Explainable AI (XAI) for medical diagnosis: Challenges and future directions in dermatology,” IEEE Transactions on Medical Imaging, vol. 43, no. 3, pp. 890–905, 2024.
[29] G. Zhang and F. Li, “Self-supervised learning for skin lesion classification with limited labeled data,” IEEE Transactions on Neural Networks and Learning Systems, vol. 35, no. 6, pp. 7800–7812, 2024.
[30] X. Zhao, Y. Wang, and Y. Sun, “Attention-guided ResNet for fine-grained skin lesion classification,” Computerized Medical Imaging and Graphics, vol. 112, p. 102334, 2024.
Cite This Article
Choose your preferred format