An Explainable Hybrid Quantum-Classical Deep Learning Framework for Early Skin Cancer Detection Using Compact Feature Representation

Author

Bhargav Sai Vaddi , B. Sai Venkata Teja, Ch. Hari Prasad, D. Sai Teja


Abstract

 Skin cancer is a major global health concern, and early detection is essential for improving treatment outcomes. This project proposes an explainable hybrid quantum-classical framework for skin cancer detection using the HAM10000 dermoscopic image dataset. ResNet-50 is employed as the deep learning backbone to extract meaningful features from skin lesion images. Three experimental configurations are investigated: direct classification using the ResNet-50 classification head, classification using extracted deep features, and classification using PCA-reduced features with 4, 6, and 8 dimensions. The extracted and reduced features are evaluated using classical machine learning algorithms, including Logistic Regression, Support Vector Machine, Random Forest, and XGBoost, alongside a PennyLane-based Variational Quantum Classifier (VQC). This comparison provides insights into the performance of classical and quantum classification approaches under different feature representations. To improve model interpretability, Grad-CAM generates visual heatmaps highlighting image regions contributing to predictions. The complete framework is integrated into a web application developed using React for the frontend and FastAPI for backend services. SQLite is used to maintain prediction and screening history. The application provides benign or malignant classification, malignancy probability scores, model information, and visual explanations, offering an integrated platform for comparative evaluation and explainable skin cancer screening.



Keywords

Skin Cancer Detection, ResNet-50, Principal Component Analysis (PCA), Variational Quantum Classifier (VQC), Grad-CAM



Full Text:

Download Paper PDF


References


  1. Al Mahmud, A., Azam, S., Khan, I. U., Montaha, S., Karim, A., Haque, A., Hasan, M. Z., Brady, M., Biswas, R., & Jonkman, M. (2024). SkinNet-14: A deep learning framework for accurate skin cancer classification using low-resolution dermoscopy images with optimized training time. Neural Computing and Applications, 36, 18935–18959. https://doi.org/10.1007/s00521-024-10225-y
  2. Aravinda, C. V., Joseph, E. R., & Alasmari, S. (2025). A hybrid quantum-classical approach for multi-class skin disease classification using a 4-qubit model. IEEE Access, 13, 114460–114470. https://doi.org/10.1109/ACCESS.2025.3581030
  3. Aravinda, C. V., Joseph, E. R., & Alasmari, S. (2025). Optimized DenseNet121 and quantum PennyLane fusion for explainable skin disease recognition and classification. IEEE Access, 13, 167957–167969. https://doi.org/10.1109/ACCESS.2025.3608217
  4. Farhatullah, X., Chen, X., Zeng, D., Xu, J., Nawaz, R., & Ullah, R. (2024). Classification of skin lesion with features extraction using quantum Chebyshev polynomials and autoencoder from wavelet-transformed images. IEEE Access, 12, 193923–193936. https://doi.org/10.1109/ACCESS.2024.3502513
  5. Frasca, M., Cutica, I., Pravettoni, G., & La Torre, D. (2025). Optimizing melanoma diagnosis: A hybrid deep learning and quantum computing approach for enhanced lesion classification. Intelligence-Based Medicine, 12, Article 100264. https://doi.org/10.1016/j.ibmed.2025.100264
  6. Gallazzi, M., Biavaschi, S., Bulgheroni, A., Gatti, T. M., Corchs, S., & Gallo, I. (2024). A large dataset to enhance skin cancer classification with transformer-based deep neural networks. IEEE Access, 12, 109544–109559. https://doi.org/10.1109/ACCESS.2024.3439365
  7. Gururaj, H. L., Manju, N., Nagarjun, A., Manjunath Aradhya, V. N., & Flammini, F. (2023). DeepSkin: A deep learning approach for skin cancer classification. IEEE Access, 11, 50205–50214. https://doi.org/10.1109/ACCESS.2023.3274848
  8. Houssein, E. H., Abdelkareem, D. A., Hu, G., Hameed, M. A., Ibrahim, I. A., & Younan, M. (2024). An effective multiclass skin cancer classification approach based on deep convolutional neural network. Cluster Computing, 27, 12799–12819. https://doi.org/10.1007/s10586-024-04540-1
  9. Hussein, A. A., Montaser, A. M., & Elsayed, H. A. (2025). Skin cancer image classification using hybrid quantum deep learning model with BiLSTM and MobileNetV2. Quantum Machine Intelligence, 7, Article 66. https://doi.org/10.1007/s42484-025-00288-y
  10. Paraddy, S., & Virupakshappa. (2025). Addressing challenges in skin cancer diagnosis: A convolutional Swin Transformer approach. Journal of Imaging Informatics in Medicine, 38(3), 1755–1775. https://doi.org/10.1007/s10278-024-01290-9
  11. Reka, S. S., Karthikeyan, H. L., Shakil, A. J., Venugopal, P., & Muniraj, M. (2024). Exploring quantum machine learning for enhanced skin lesion classification: A comparative study of implementation methods. IEEE Access, 12, 104568–104584. https://doi.org/10.1109/ACCESS.2024.3434681
  12. Suryanarayana, G., Ansari, M. D., K., L. N. C. P., Biksham, V., Swapna, N., & Gogineni, S. (2026). Quantum classical convolutional neural network using ResNet-18 and feature optimization for multiclass skin lesion classification. Discover Artificial Intelligence, 6, Article 710. https://doi.org/10.1007/s44163-026-01384-z
  13. Tabibi, S. T., Nikravanshalmani, A., & Saboohi, H. (2024). An ensemble classifier based on diverse convolutional neural networks for skin lesions classification. IEEE Access, 13, 195673–195686. https://doi.org/10.1109/ACCESS.2024.3442827
  14. Toprak, A. N., & Aruk, I. (2024). A hybrid convolutional neural network model for the classification of multi-class skin cancer. International Journal of Imaging Systems and Technology, 34(5), Article e23180. https://doi.org/10.1002/ima.23180
  15. Yang, G., Luo, S., & Greer, P. (2023). A novel vision transformer model for skin cancer classification. Neural Processing Letters, 55, 9335–9351. https://doi.org/10.1007/s11063-023-11204-5

 

 

 

Share your valuable work from Social Media Buttons