Scholarlink

<p>Skin cancer is a potentially life-threatening disease resulting from DNA damage, where early detection is critical to improving survival rates. This study introduces an advanced hybrid deep learning framework designed to enhance the accuracy of skin cancer classification, distinguishing between benign and malignant lesions. Our approach begins with data pre-processing to improve input quality, followed by training two high-performing pre-trained deep learning models, InceptionV3 and DenseNet121. We then apply a weighted sum rule for fusion of the model predictions, leading to high accuracy and generalizability across datasets. On the primary dataset, our hybrid model achieved 92.27% accuracy, 90.80% precision, 92.33% sensitivity, 92.22% specificity, and a 91.57% F1 score. Notably, the model also demonstrated robustness on an external dataset, achieving 93.45% accuracy, 91.87% precision, 93.04% sensitivity, 93.18% specificity, and a 92.30% F1 score, outperforming existing models. This study offers a reliable, highly generalizable solution for automated skin cancer diagnosis and represents a valuable contribution to early detection methods, with the potential to improve patient outcomes significantly.</p>

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

An Integrated Deep Learning Model for Skin Cancer Detection Using Hybrid Feature Fusion Technique

  • 期刊论文

  • 发表日期：

    2025年1月9日

查看全文
  • Maksuda Akter,
  • Rabea Khatun,
  • Md. Alamin Talukder,
  • Md. Manowarul Islam,
  • Md. Ashraf Uddin,
  • Md. Khabir Uddin Ahamed,
  • Ansam Khraisat

摘要

Skin cancer is a potentially life-threatening disease resulting from DNA damage, where early detection is critical to improving survival rates. This study introduces an advanced hybrid deep learning framework designed to enhance the accuracy of skin cancer classification, distinguishing between benign and malignant lesions. Our approach begins with data pre-processing to improve input quality, followed by training two high-performing pre-trained deep learning models, InceptionV3 and DenseNet121. We then apply a weighted sum rule for fusion of the model predictions, leading to high accuracy and generalizability across datasets. On the primary dataset, our hybrid model achieved 92.27% accuracy, 90.80% precision, 92.33% sensitivity, 92.22% specificity, and a 91.57% F1 score. Notably, the model also demonstrated robustness on an external dataset, achieving 93.45% accuracy, 91.87% precision, 93.04% sensitivity, 93.18% specificity, and a 92.30% F1 score, outperforming existing models. This study offers a reliable, highly generalizable solution for automated skin cancer diagnosis and represents a valuable contribution to early detection methods, with the potential to improve patient outcomes significantly.