Scholarlink

Alzheimer’s Disease is recognized as a neurological disorder that damages the tissues of the brain, causing long-term memory loss, cognitive difficulties, confusion, inconsistent behavior, and ultimately death. In this study, we identified the three broad stages of this neurodegenerative disease: Non-Demented (Normal), very mild (early stage), mild (middle stage), and moderate demented (late stage). The treatment for the AD symptoms can greatly benefitted from the early identification and categorization. DL and ML techniques are applied. The dataset is obtained from the ADNI and Kaggle research data sharing platforms. By using Visual Studio code in Anaconda Navigator, the source code is tested, and Google Colab is used for the model training. The study’s findings showed that, using the test dataset, we could effectively and accurately categorize the stages of AD with a 98.02% accuracy rate. This test’s accuracy score is noticeably greater than that of previous studies. The findings also showed that these methods can be effectively applied in the medical field to aid in early disease diagnosis and identification.

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 Novel CNN Architecture: Comprehensive Approach to Alzheimer’s Disease Detection and Classification

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • G. Indhumathi,
  • M. Palanivelan,
  • G. Saranya,
  • K. Devadharshini,
  • N. Divya Bharathi,
  • A. Divyadharshini

摘要

Alzheimer’s Disease is recognized as a neurological disorder that damages the tissues of the brain, causing long-term memory loss, cognitive difficulties, confusion, inconsistent behavior, and ultimately death. In this study, we identified the three broad stages of this neurodegenerative disease: Non-Demented (Normal), very mild (early stage), mild (middle stage), and moderate demented (late stage). The treatment for the AD symptoms can greatly benefitted from the early identification and categorization. DL and ML techniques are applied. The dataset is obtained from the ADNI and Kaggle research data sharing platforms. By using Visual Studio code in Anaconda Navigator, the source code is tested, and Google Colab is used for the model training. The study’s findings showed that, using the test dataset, we could effectively and accurately categorize the stages of AD with a 98.02% accuracy rate. This test’s accuracy score is noticeably greater than that of previous studies. The findings also showed that these methods can be effectively applied in the medical field to aid in early disease diagnosis and identification.