Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{206061,
        author = {ADHEENA TOMSON and SONIA SUNNY},
        title = {Comparative Analysis of EfficientNetV2-S, Swin Transformer-T, and ConvNeXt-Tiny for Automated Nail Disease Classification},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {2},
        pages = {204-207},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=206061},
        abstract = {This paper presents a novel framework for classification of nail diseases using Deep Learning models. Nail diseases such as Onychomycosis and Psoriasis are often misdiagnosed, causing serious clinical complications and delays in effective treatment. This work utilizes three Deep Learning models: EfficientNetV2-S, Swin Transformer-T, and ConvNeXt-Tiny. These models were built and compared to classify nail images into three categories: Healthy, Onychomycosis, and Psoriasis. Transfer learning is applied in all three models and fine-tuned using a public dataset of 1,463 nail images. Mixup augmentation, label smoothing, and the AdamW optimizer are used for training. The models are evaluated on a test set of 299 images. Swin Transformer-T achieves the highest classification accuracy and macro recall of 95.20%, while ConvNeXt-Tiny achieves the highest Psoriasis recall of 92.31%.},
        keywords = {This paper presents a novel framework for classification of nail diseases using Deep Learning models. Nail diseases such as Onychomycosis and Psoriasis are often misdiagnosed, causing serious clinical complications and delays in effective treatment. This work utilizes three Deep Learning models: EfficientNetV2-S, Swin Transformer-T, and ConvNeXt-Tiny. These models were built and compared to classify nail images into three categories: Healthy, Onychomycosis, and Psoriasis. Transfer learning is applied in all three models and fine-tuned using a public dataset of 1,463 nail images. Mixup augmentation, label smoothing, and the AdamW optimizer are used for training. The models are evaluated on a test set of 299 images. Swin Transformer-T achieves the highest classification accuracy and macro recall of 95.20%, while ConvNeXt-Tiny achieves the highest Psoriasis recall of 92.31%.},
        month = {July},
        }