Scholarlink

Alzheimer’s disease (AD) is a neurological condition that worsens over time and causes irreversible cognitive impairment, significantly burdening global healthcare systems. Effective management and therapy depend on an early and precise diagnosis. This study presents a robust deep learning-based classification framework leveraging MobileNet and Xception architectures to identify AD stages from MRI images. Pretrained models are adjusted with the use of data augmentation and optimization methods for grouping images into four categories: Non-Demented, Very Mild Demented, Mild Demented, and Moderate Demented. The Xception model achieved 95% accuracy, while MobileNet reached 84.59%, demonstrating promising potential for resource-constrained environments. This work advances diagnostic accuracy in AD detection, facilitating timely medical interventions.

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 Robust Classification Framework for Alzheimer’s Detection Using MobileNet and Xception Networks

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • V. Arpitha,
  • S. Amutha,
  • Anupama Girish,
  • Amith Pradhaan

摘要

Alzheimer’s disease (AD) is a neurological condition that worsens over time and causes irreversible cognitive impairment, significantly burdening global healthcare systems. Effective management and therapy depend on an early and precise diagnosis. This study presents a robust deep learning-based classification framework leveraging MobileNet and Xception architectures to identify AD stages from MRI images. Pretrained models are adjusted with the use of data augmentation and optimization methods for grouping images into four categories: Non-Demented, Very Mild Demented, Mild Demented, and Moderate Demented. The Xception model achieved 95% accuracy, while MobileNet reached 84.59%, demonstrating promising potential for resource-constrained environments. This work advances diagnostic accuracy in AD detection, facilitating timely medical interventions.