COMPARATIVE ANALYSIS OF DEEP LEARNING ALGORITHMS FOR AUTOMATED SKIN DISEASE DIAGNOSIS FROM DERMOSCOPIC IMAGES
DOI:
https://doi.org/10.5281/zenodo.20643398Abstract
A comparative analysis of CNN, ResNet-50, and EfficientNet-B4 architectures for automated skin
disease classification from dermoscopic images was conducted. The study was performed using the HAM10000
dataset comprising 10,015 images across seven nosological classes. Model performance was evaluated using
accuracy, sensitivity, specificity, F1-score, AUC-ROC, inference time, and memory footprint. EfficientNet-B4
demonstrated the best performance, achieving 93.71 % accuracy and an AUC-ROC of 0.978. The obtained
results indicate the potential of this architecture for integration into clinical decision-support systems
Keywords
deep learning; skin disease diagnosis; convolutional neural network (CNN); ResNet; EfficientNet; dermoscopy; HAM10000; medical image analysis; transfer learning; AUC-ROCReferences
1. Siegel R.L., Miller K.D., Jemal A. Cancer Statistics, 2023. CA: A Cancer Journal for Clinicians. 2023. Vol. 73, No. 1.
2. P. 17–48.
3. Chandler P., Rinner C., Apalla Z., et al. Human-Computer Collaboration for Rapid Land Cover Mapping. Nature
4. Medicine. 2020. Vol. 26, No. 6. P. 869–874.
5. Esteva A., Kuprel B., Novoa R.A., et al. Dermatologist-Level Classification of Skin Cancer with Deep Neural Networks.
6. Nature. 2017. Vol. 542, No. 7639. P. 115–118.
7. Codella N.C., Gutman D., Celebi M.E., et al. Skin Lesion Analysis Toward Melanoma Detection: ISIC 2017 Challenge.
8. Proceedings of the IEEE International Symposium on Biomedical Imaging (ISBI). 2018. P. 168–172.
9. He K., Zhang X., Ren S., Sun J. Deep Residual Learning for Image Recognition. Proceedings of the IEEE Conference
10. on Computer Vision and Pattern Recognition (CVPR). 2016. P. 770–778.
11. Han S.S., Park I., Bang W., et al. Deep Neural Network with Monte Carlo Dropout for Uncertainty Estimation in Skin
12. Lesion Diagnosis. Journal of the American Academy of Dermatology. 2020. Vol. 82, No. 4. P. 929–940.
13. Tan M., Le Q.V. EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. Proceedings of the
14. International Conference on Machine Learning (ICML). 2019. P. 6105–6114.
15. Mahbod A., Schaefer G., Wang C., et al. Transfer Learning Using a Multi-Scale and Multi-Network Ensemble for Skin
16. Lesion Segmentation. Pattern Recognition Letters. 2020. Vol. 131. P. 449–456.
17. Li S., Yu L., Fu C.W., et al. Self-Loop Uncertainty: A Novel Graph Convolutional Network for Semi-Supervised
18. Classification. IEEE Transactions on Neural Networks and Learning Systems. 2020. Vol. 32, No. 2. P. 523–534.
19. Dosovitskiy A., Beyer L., Kolesnikov A., et al. An Image Is Worth 16×16 Words: Transformers for Image Recognition at
20. Scale. International Conference on Learning Representations (ICLR). 2021.
21. Goyal M., Knackstedt T., Yan S., Hassanpour S. Artificial Intelligence-Based Image Classification Methods for Diagnosis
22. of Skin Cancer. Computers in Biology and Medicine. 2020. Vol. 127. Article No. 104065
Downloads
Published
How to Cite
Issue
Section
