Scholarlink

<p>Alzheimer’s disease (AD) is a progressive and irreversible brain disorder characterized by degeneration of brain cells. Dementia impacts over 50&#xa0;million people worldwide, with approximately 10&#xa0;million new cases emerging each year. Dementia cannot presently be cured, but it can be managed if it is recognised in its early stages. More precise classification of certain brain illnesses results from analysis of MRI tissue structures. The proposed research is primarily focused on generating a deep learning model for identifying and categorising AD phases. The 1.5T weighted MRI sensory system of the ADNI data is utilized to create this model. Data from all three classes, namely AD, Mild Cognitive Impairment (MCI), and Cognitive Normal (CN) have been utilized. In total, 8379 samples were used in this study. The gathered MRI images are preprocessed utilising the CLAHE (Contrast Limited Adaptive Histogram Equalization) and TV (Total Variation) denoising methods for noise reduction and picture enhancement. The dataset is examined using five different transfer learning models and is put through transfer learning. The better results are achieved using the transfer learning models (Xception-85.08%, VGG19-83.76%, NasNetMobile-84.07%, InceptionV3- 83.76%, VGG16-82.75%). The anticipated CNN architecture is then trained on the dataset and tested. It has been noted that the suggested network produced better outcomes and had a testing accuracy of roughly 91.17%. The proposed methodology’s experimental findings demonstrate that it offers greater accuracy than current schemes.</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

Deep Learning Applications for Improving Early Detection and Staging Accuracy in Alzheimer’s Disease

  • 期刊论文

  • 发表日期：

    2025年9月4日

查看全文
  • R Raksha,
  • Jyoti Metan,
  • Raghavendra M. Ichangi,
  • P Suresh,
  • Shazia Sulthana,
  • M N Anusha,
  • K N Bharath

摘要

Alzheimer’s disease (AD) is a progressive and irreversible brain disorder characterized by degeneration of brain cells. Dementia impacts over 50 million people worldwide, with approximately 10 million new cases emerging each year. Dementia cannot presently be cured, but it can be managed if it is recognised in its early stages. More precise classification of certain brain illnesses results from analysis of MRI tissue structures. The proposed research is primarily focused on generating a deep learning model for identifying and categorising AD phases. The 1.5T weighted MRI sensory system of the ADNI data is utilized to create this model. Data from all three classes, namely AD, Mild Cognitive Impairment (MCI), and Cognitive Normal (CN) have been utilized. In total, 8379 samples were used in this study. The gathered MRI images are preprocessed utilising the CLAHE (Contrast Limited Adaptive Histogram Equalization) and TV (Total Variation) denoising methods for noise reduction and picture enhancement. The dataset is examined using five different transfer learning models and is put through transfer learning. The better results are achieved using the transfer learning models (Xception-85.08%, VGG19-83.76%, NasNetMobile-84.07%, InceptionV3- 83.76%, VGG16-82.75%). The anticipated CNN architecture is then trained on the dataset and tested. It has been noted that the suggested network produced better outcomes and had a testing accuracy of roughly 91.17%. The proposed methodology’s experimental findings demonstrate that it offers greater accuracy than current schemes.