Scholarlink

Identifying monkeypox is critical as infections occur outside endemic African locations. This paper explores the potential of deep learning models for classifying monkeypox and wart using skin lesion images. Experiments were performed on a publicly available dataset that underwent standardization, augmentation, and normalization. Four deep neural models (ResNet50 V2, InceptionV3, EfficientNetB0, and DenseNet121) and one hybrid model were assessed for performance. The models demonstrated effective diagnosis of these diseases from skin images, with ResNet50 V2, InceptionV3, and the hybrid model showing particularly high accuracies (95.58%, 94.11%, and 97.05%), respectively. Future research will increase the dataset and work with medical experts to further enhance the classification procedure.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Classification of Monkeypox and Wart Skin Lesions Images Using Pre-Trained Deep Learning Models

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Md. Abdul Ahad Rifat,
  • Md. Shaharear Kabir Rabby,
  • Niaz Ahmed,
  • Sajidul Islam Saief,
  • Ahmed Wasif Reza

摘要

Identifying monkeypox is critical as infections occur outside endemic African locations. This paper explores the potential of deep learning models for classifying monkeypox and wart using skin lesion images. Experiments were performed on a publicly available dataset that underwent standardization, augmentation, and normalization. Four deep neural models (ResNet50 V2, InceptionV3, EfficientNetB0, and DenseNet121) and one hybrid model were assessed for performance. The models demonstrated effective diagnosis of these diseases from skin images, with ResNet50 V2, InceptionV3, and the hybrid model showing particularly high accuracies (95.58%, 94.11%, and 97.05%), respectively. Future research will increase the dataset and work with medical experts to further enhance the classification procedure.