Scholarlink

<p>Skin diseases require immediate investigation and diagnosis to prevent life-threatening situations. Although there is a growing interest in leveraging fog computing and Internet of Things (IoT) applications for diagnosing skin diseases, the availability of accurate deep learning models and frameworks for real-time diagnosis remains limited. This research aims to bridge this gap by proposing a fog-IoT-based solution that combines the transfer learning Visual Geometry Group-19 (VGG-19) model with fog computing and IoT devices for real-time skin disease diagnosis. The proposed architecture utilizes the Dermatology Network (Derm-Net) dataset for training, validating, and testing of the model. Subsequently, the trained model is integrated into edge computing to enable real-time diagnosis of skin diseases. To evaluate and verify the performance of the model, FogBus modules are employed across various fog computation scenarios, demonstrating promising results in terms of arbitration, latency, jitter, and execution time.</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

A deep learning-based approach for a real-time diagnosis of skin diseases

  • 期刊论文

  • 发表日期：

    2025年9月29日

查看全文
  • Ghaleb Al-Gaphari,
  • Nashwan Ahmed Mohammed Morshed,
  • Khadijah AlAidarous

摘要

Skin diseases require immediate investigation and diagnosis to prevent life-threatening situations. Although there is a growing interest in leveraging fog computing and Internet of Things (IoT) applications for diagnosing skin diseases, the availability of accurate deep learning models and frameworks for real-time diagnosis remains limited. This research aims to bridge this gap by proposing a fog-IoT-based solution that combines the transfer learning Visual Geometry Group-19 (VGG-19) model with fog computing and IoT devices for real-time skin disease diagnosis. The proposed architecture utilizes the Dermatology Network (Derm-Net) dataset for training, validating, and testing of the model. Subsequently, the trained model is integrated into edge computing to enable real-time diagnosis of skin diseases. To evaluate and verify the performance of the model, FogBus modules are employed across various fog computation scenarios, demonstrating promising results in terms of arbitration, latency, jitter, and execution time.