Deep Learning in Dermatology: Exploring Convnext Model Hierarchies and Ensembles for Enhanced Diagnostic Precision

  • Authors

    • J Dhanalakshmi Department of Data Science and Business Systems, School of Computing, College of Engineering and Technology, SRM Institute of Science ‎& Technology, India
    • A. Prabhu Chakkaravarthy Department of Networking and Communications, School of Computing, College of Engineering and Technology, SRM Institute of Science ‎& Technology, India
    • B. Dhanalakshmi Associate Professor, Department of Computer Science and Engineering, B.S Abdur Rahman Crescent Institute of Science and Technology, ‎Chennai
    • Gokulnath K. Professor, School of Advanced Computing, CGC University, Mohali
    • Sudha Rajesh Associate Professor, Dept. of Computational Intelligence, College of Engineering and Technology, School of Computing, SRM Institute of ‎Science & Technology, Kattankulathur, Chennai
    https://doi.org/10.14419/1kbqz735

    Received date: August 21, 2025

    Accepted date: November 3, 2025

    Published date: November 9, 2025

  • Skin Cancer Diagnosis; ConvNeXt Models; Ensemble Learning; Deep Learning; HAM10000 Dataset; Stratified Training

  • Abstract

    Skin cancer is one of the most widespread and life-threatening malignancies in the world that requires early and proper accurate ‎diagnosis to be efficiently solved. This research article examines the performance of different ConvNeXt models- ConvNeXt-Tiny ‎to ConvNeXt-XLarge architectures, and also their ensemble representations in malignant categories of dermoscopic images. To guarantee the ‎balanced representations of classes, using the HAM10000 dataset, a stratified 80-20 train-validation configuration was used to train and real-‎ize the models. Using the indicators of performance in terms of accuracy, precision, recall, F1-score, and confusion matrices, the analysis ‎has shown that middle-ground models and layers, like ConvNeXt-Small combination of performance and efficiency. Ensemble learning also ‎contributes to diagnostic robustness, with ConvNeXt-Tiny, Small, and Base models combined, achieving the greatest validation accuracy of ‎‎92.42%. Comparative analysis shows that, although adding depth leads to marginal benefits in the model, the depth poses a risk of overfit-‎ting and the cost of computation further‎.

  • References

    1. Brinker, T. J., Hekler, A., Enk, A. H., Berking, C., Haferkamp, S., Hauschild, A., ... & von Kalle, C. (2019). Deep neural networks are superior to dermatologists in melanoma image classification. European Journal of Cancer, 119, 11-17. https://doi.org/10.1016/j.ejca.2019.05.003
    2. Chen, Y., Xu, Z., & Wang, J. (2023). Lightweight convolutional neural networks for portable medical diagnosis: Recent trends and challenges. IEEE Access, 11, 55642–55659.
    3. Chen, L., Zhao, Y., & Lin, X. (2024). Improving robustness of AI diagnostic systems to adversarial noise through architecture-aware ensemble models. IEEE Access, 12, 54823–54834.
    4. Codella, N. C., Rotemberg, V., Tschandl, P., Celebi, M. E., Dusza, S. W., Gutman, D., ... & Halpern, A. C. (2018). Skin lesion analysis toward melanoma detection: A challenge at the 2017 International Symposium on Biomedical Imaging (ISBI), hosted by the International Skin Imaging Collaboration (ISIC). arXiv preprint arXiv:1710.05006. https://doi.org/10.1109/ISBI.2018.8363547.
    5. Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun, S. (2017). Dermatologist-level classification of skin cancer with deep neural networks. Nature, 542(7639), 115–118. https://doi.org/10.1038/nature21056.
    6. Fan, H., Kirillov, A., Xie, S., & He, K. (2023). Scaling ConvNets for Foundation Models with ConvNeXt-UL. arXiv:2307.08015.
    7. Goyal, M., Oakley, A., Bansal, P., & Yap, M. H. (2020). Skin lesion segmentation in dermoscopic images with ensemble deep learning methods. IEEE Access, 8, 4171–4181. https://doi.org/10.1109/ACCESS.2019.2960504.
    8. Han, S. S., Kim, M. S., Lim, W., Park, G. H., Park, I., & Chang, S. E. (2020). Classification of the clinical images for benign and malignant cutane-ous tumors using a deep learning algorithm. Journal of Investigative Dermatology, 138(7), 1529–1538. https://doi.org/10.1016/j.jid.2017.12.018.
    9. Jha, D., Smedsrud, P. H., Riegler, M. A., Johansen, D., Halvorsen, P., & Lange, T. (2023). Deep learning-based medical image analysis: a review of recent advances. Frontiers in Artificial Intelligence, 6, 1156304.
    10. Karim, M. R., Othman, N., & Islam, S. M. S. (2024). Advances in deep learning models for medical image analysis: A comprehensive review. IEEE Access, 12, 123456-123478.
    11. Kablan, E. B., & Ayas, S. (2024). Skin lesion classification from dermoscopy images using ensemble learning of ConvNeXt models. Signal, Image and Video Processing, 18, 6353–6361. https://doi.org/10.1007/s11760-024-03321-y.
    12. Litjens, G., Kooi, T., Bejnordi, B. E., Setio, A. A. A., Ciompi, F., Ghafoorian, M., ... & Sánchez, C. I. (2017). A survey on deep learning in medical image analysis. Medical Image Analysis, 42, 60-88. https://doi.org/10.1016/j.media.2017.07.005.
    13. Liu, Z., Mao, H., Wu, C.-Y., Feichtenhofer, C., Darrell, T., & Xie, S. (2022). ConvNeXt: A ConvNet for the 2020s. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 11976–11986. https://openaccess.thecvf.com/content/CVPR2022/html/Liu_ConvNeXt_A_ConvNet_for_the_2020s_CVPR_2022_paper.html. https://doi.org/10.1109/CVPR52688.2022.01167.
    14. Liu, Z., Mao, H., Wu, C.-Y., Feichtenhofer, C., Darrell, T., & Xie, S. (2022). ConvNeXt: A ConvNet for the 2020s. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 11976-11986. https://openaccess.thecvf.com/content/CVPR2022/html/Liu_ConvNeXt_A_ConvNet_for_the_2020s_CVPR_2022_paper.html. https://doi.org/10.1109/CVPR52688.2022.01167.
    15. Liu, Z., Mao, H., Wu, C. Y., Feichtenhofer, C., Darrell, T., & Xie, S. (2022). A ConvNet for the 2020s. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 11976-11986. https://doi.org/10.1109/CVPR52688.2022.01167.
    16. Liu, J., Wu, X., Deng, H., & Li, W. (2023). Deep learning-based diagnostic system for skin cancer classification using dermoscopic images. Com-puters in Biology and Medicine, 161, 106012.
    17. Mahbod, A., Schaefer, G., Wang, C., Dorffner, G., & Ecker, R. (2021). Transfer learning using a multi-scale and multi-network ensemble for skin lesion classification. Computers in Biology and Medicine, 104, 103486. https://doi.org/10.1016/j.compbiomed.2018.11.017.
    18. Nagpal, K., Foote, D., Liu, Y., Chen, P. H. C., Wulczyn, E., Tan, F., ... & Corrado, G. S. (2019). Development and validation of a deep learning algorithm for improving Gleason scoring of prostate cancer. NPJ Digital Medicine, 2(1), 48. https://doi.org/10.1038/s41746-019-0112-2.
    19. Singh, S., Gupta, R., & Banerjee, A. (2023). A comprehensive survey on performance metrics for classification problems in medical AI. Journal of Biomedical Informatics, 137, 104297. https://doi.org/10.1016/j.jbi.2023.104297.
    20. Singh, S., Gupta, P., & Verma, A. (2023). Improving clinical diagnosis through ensemble learning: A review and application in cancer detection. Artificial Intelligence in Medicine, 133, 102470.
    21. Singh, R., Banerjee, A., & Sinha, A. (2023). Addressing data imbalance and noise in medical image classification using hybrid deep learning en-sembles. Computerized Medical Imaging and Graphics, 104, 102228. https://doi.org/10.1016/j.compmedimag.2023.102228.
    22. Tan, M., & Le, Q. V. (2019). EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. Proceedings of the 36th International Conference on Machine Learning (ICML), 6105–6114. http://proceedings.mlr.press/v97/tan19a.html.
    23. Tan, M., & Le, Q. V. (2019). EfficientNet: Rethinking model scaling for convolutional neural networks. International Conference on Machine Learning (ICML), 6105–6114. http://proceedings.mlr.press/v97/tan19a.html.
    24. Tschandl, P., Rinner, C., Apalla, Z., Argenziano, G., Codella, N., Halpern, A., ... & Kittler, H. (2019). Human–computer collaboration for skin can-cer recognition. Nature Medicine, 26, 1229–1234. https://doi.org/10.1038/s41591-020-0942-0.
    25. Wang, H., Zhang, Y., Chen, M., & Yang, J. (2023). Ensemble deep learning for medical image classification: Recent advances and future trends. IEEE Transactions on Neural Networks and Learning Systems, 34(4), 1456-1472.
    26. Wang, H., Zhang, Y., Chen, M., & Yang, J. (2023). Ensemble deep learning for medical image classification: Recent advances and future trends. IEEE Transactions on Neural Networks and Learning Systems, 34(4), 1456–1472.
    27. World Health Organization. (2023). Skin cancers. https://www.who.int/news-room/fact-sheets/detail/ultraviolet-(uv)-radiation
    28. Woo, S., Li, D., Han, S., Kirillov, A., & Feichtenhofer, C. (2023). ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencod-ers. arXiv:2301.00808. https://doi.org/10.1109/CVPR52729.2023.01548.
    29. Yu, F., Chen, X., Wang, L., Xiong, Y., & Sun, J. (2023). ConvNeXt for Dense Prediction Tasks. CVPR Workshop on Efficient Deep Learning for Computer Vision.
    30. Zhang, W., Liu, Y., & Tan, R. (2023). Efficient deep learning models for edge-based medical imaging applications. Computers in Biology and Medicine, 158, 106944. https://doi.org/10.1016/j.compbiomed.2023.106944.
    31. Zhang, Y., Li, X., Wang, S., & Chen, Y. (2023). Enhanced feature extraction techniques in convolutional neural networks for medical image classi-fication. Journal of Biomedical Informatics, 140, 104252.
    32. Zhao, J., Wang, Y., Chen, T., & Li, Z. (2022). Diversity-driven ensemble learning for robust medical image classification. Pattern Recognition Let-ters, 160, 53-61. https://doi.org/10.1016/j.patrec.2022.02.006.
    33. Zhao, J., Xu, K., Wang, F., & Yu, Z. (2023). Hybrid ensembles of deep learning models for robust medical image classification. Artificial Intelli-gence in Medicine, 139, 102484.
    34. Zenebe, A. Y. (2024). SkinAACN: An Efficient Skin Lesion Classification Based on Attention Augmented ConvNeXt with Hybrid Loss Func-tion. Proceedings of the ACM Conference.

  • Downloads

  • How to Cite

    Dhanalakshmi , J. ., Chakkaravarthy , A. P. ., Dhanalakshmi , B., K. , G. ., & Rajesh, S. . (2025). Deep Learning in Dermatology: Exploring Convnext Model Hierarchies and Ensembles for Enhanced Diagnostic Precision. International Journal of Basic and Applied Sciences, 14(7), 254-265. https://doi.org/10.14419/1kbqz735