Scholarlink

<p>Alzheimer’s disease continues to present a significant challenge in global healthcare, necessitating advanced diagnostic and treatment methodologies. This study proposes a deep learning framework for addressing Alzheimer’s disease diagnosis and treatment, employing state-of-the-art convolutional neural network (CNN) architectures and sampling techniques. The motivation behind this research stems from the urgent need for a cure and the potential of deep learning algorithms, combined with artificial intelligence, to revolutionize medical diagnostics. Furthermore, the study addresses the issue of imbalanced datasets in Alzheimer’s disease diagnosis by investigating various techniques for dataset balancing, aiming to enhance the reliability of diagnostic models. During experimentation, the accuracy of the 4-stage CNN, VGG-16, and ResNet-60 with different sampling techniques, including ADASYN, SMOTE, and RandomOverSampler, was analyzed for accurate diagnosis of Alzheimer’s disease. The ResNet-50 with RandomOverSample has achieved the highest test accuracy of 99.94%.</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

Performance analysis of deep learning models for imbalanced Alzheimer’s disease dataset using sampling techniques

  • 期刊论文

  • 发表日期：

    2026年4月15日

查看全文
  • Anupama Phakatkar

摘要

Alzheimer’s disease continues to present a significant challenge in global healthcare, necessitating advanced diagnostic and treatment methodologies. This study proposes a deep learning framework for addressing Alzheimer’s disease diagnosis and treatment, employing state-of-the-art convolutional neural network (CNN) architectures and sampling techniques. The motivation behind this research stems from the urgent need for a cure and the potential of deep learning algorithms, combined with artificial intelligence, to revolutionize medical diagnostics. Furthermore, the study addresses the issue of imbalanced datasets in Alzheimer’s disease diagnosis by investigating various techniques for dataset balancing, aiming to enhance the reliability of diagnostic models. During experimentation, the accuracy of the 4-stage CNN, VGG-16, and ResNet-60 with different sampling techniques, including ADASYN, SMOTE, and RandomOverSampler, was analyzed for accurate diagnosis of Alzheimer’s disease. The ResNet-50 with RandomOverSample has achieved the highest test accuracy of 99.94%.