Scholarlink

Neuroimaging has become widely recognized as an essential clinical tool for diagnosing Alzheimer’s disease (AD) and mild cognitive impairment (MCI) in the realm of neuro-pathological disorders. The main objective of neuroimaging is to leverage visual data to aid in the diagnosis of brain-related conditions. A notable example of this is positron emission tomography (PET), which produces three-dimensional images of the brain. This study explores the use of advanced deep learning (DL), an innovative neuroimaging approach, to evaluate its effectiveness in enhancing the accuracy of AD diagnosis. This research proposes novel techniques in neuroimaging in Alzheimer’s disease detection based on generative adversarial models and deep learning techniques. Here, the input is collected as brain neuroimages and processed for noise removal and normalization. Then, this image is segmented using Fuzzy K-clustering transfer graph cut convolutional U-net neural networks (FKCTGCU). Then, this segmented image has been classified using generative adversarial Gaussian Q-neural network with particle whale colony heuristic optimization (GAGQ-PWCHO). The classified output gives neural system with abnormality in which the AD has been detected. The simulation analysis was conducted on various neuroimaging datasets, focusing on detection accuracy, random precision, recall, F1 score, and the kappa coefficient.

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

Neuro Imaging-Based Alzheimer’s Disease Detection Using Generative Adversarial Model with Deep Learning Algorithm

  • 章节

  • 发表日期：

    2025年

查看全文
  • J. Vijayaraj,
  • B. Satheesh Kumar,
  • M. Umapathy,
  • R. Manikandan,
  • S. Magesh

摘要

Neuroimaging has become widely recognized as an essential clinical tool for diagnosing Alzheimer’s disease (AD) and mild cognitive impairment (MCI) in the realm of neuro-pathological disorders. The main objective of neuroimaging is to leverage visual data to aid in the diagnosis of brain-related conditions. A notable example of this is positron emission tomography (PET), which produces three-dimensional images of the brain. This study explores the use of advanced deep learning (DL), an innovative neuroimaging approach, to evaluate its effectiveness in enhancing the accuracy of AD diagnosis. This research proposes novel techniques in neuroimaging in Alzheimer’s disease detection based on generative adversarial models and deep learning techniques. Here, the input is collected as brain neuroimages and processed for noise removal and normalization. Then, this image is segmented using Fuzzy K-clustering transfer graph cut convolutional U-net neural networks (FKCTGCU). Then, this segmented image has been classified using generative adversarial Gaussian Q-neural network with particle whale colony heuristic optimization (GAGQ-PWCHO). The classified output gives neural system with abnormality in which the AD has been detected. The simulation analysis was conducted on various neuroimaging datasets, focusing on detection accuracy, random precision, recall, F1 score, and the kappa coefficient.