Scholarlink

Multiple sclerosis (MS) is a chronic disorder disease of the central nervous system, causing inflammation and demyelination in the brain and spinal cord. This condition can impact movement and the overall visual system, making early detection crucial for effective management. Currently, specific treatments for MS are limited, emphasizing the importance of early intervention. This study proposes the use of EfficientNetB0, MobileNetV2, ResNet50, InceptionV3, and Xception CNN deep learning models for the detection of MS lesions using magnetic resonance imaging (MRI) images. Experimental results showed that EfficientNetB0 model achieved the best results among other models, and it achieved an accuracy of about 98%; on the other hand, MobileNetV2 and ResNet50 obtained promising results with accuracies of about 97%.

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 Empirical Analysis of Five CNN Models for the Detection of Multiple Sclerosis

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Ahmad T. Al-Taani,
  • Abrar A. Al-Qudah

摘要

Multiple sclerosis (MS) is a chronic disorder disease of the central nervous system, causing inflammation and demyelination in the brain and spinal cord. This condition can impact movement and the overall visual system, making early detection crucial for effective management. Currently, specific treatments for MS are limited, emphasizing the importance of early intervention. This study proposes the use of EfficientNetB0, MobileNetV2, ResNet50, InceptionV3, and Xception CNN deep learning models for the detection of MS lesions using magnetic resonance imaging (MRI) images. Experimental results showed that EfficientNetB0 model achieved the best results among other models, and it achieved an accuracy of about 98%; on the other hand, MobileNetV2 and ResNet50 obtained promising results with accuracies of about 97%.