Scholarlink

<p>Recently, multiple investigations were completed on the involuntary analysis of Alzheimer’s disease using various models. The primary intention of most of these investigation works has depended on the recognition of AD from neuroimaging data. On the other hand, diagnosing symptoms promptly as much as possible is critical as disease-adapting drugs will be most effective if managed timely in the development of the disease, before the happening of permanent brain damage. Magnetic Resonance Imaging is a non-invasive process, commonly assumed in hospitals to inspect mental abnormalities. It is one of the most employed neuroimaging modalities for detecting brain atrophy. Even though MRI is an influential device that can be beneficial to identify symptoms of AD in the brain, the acquisition process is long, chiefly owing to the requirement for physical examination of workflow blockages. Currently, numerous methodologies have been proposed for the investigation of Alzheimer’s disease employing image processing, machine learning, and deep learning, which demonstrate superior performance compared to traditional physical methods. This study formulates and constructs a fusion of deep learning models for the detection of Alzheimer’s disease utilizing an optimization algorithm. The primary objective of the proposed FDLM-DADOA model is to detect and categorize Alzheimer’s disease through an ensemble of deep learning models with hyperparameter optimization. The image processing phase employs Wiener filtering to remove noise from the image data. The fusion of deep learning models, specifically the EfficientNet B7, MobileNet, and ResNet-50, is utilized for feature extraction. The FDLM-DADOA model employs a bidirectional long short-term memory system for the identification and categorization of Alzheimer’s disease. The hyperparameter selection for the BiLSTM model can ultimately be implemented through the design of the equilibrium optimization algorithm. The experimental assessment of the FDLM-DADOA approach is conducted utilizing a benchmark picture dataset. The testing results indicated the superior performance of the FDLM-DADOA technology relative to contemporary methods.</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

RETRACTED ARTICLE: Fusion of transfer learning models for detection of alzheimer’s disease using bidirectional long short-term memory with equilibrium optimization algorithm

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年12月1日

查看全文
  • K. Renugadevi,
  • T. Jayasankar

摘要

Recently, multiple investigations were completed on the involuntary analysis of Alzheimer’s disease using various models. The primary intention of most of these investigation works has depended on the recognition of AD from neuroimaging data. On the other hand, diagnosing symptoms promptly as much as possible is critical as disease-adapting drugs will be most effective if managed timely in the development of the disease, before the happening of permanent brain damage. Magnetic Resonance Imaging is a non-invasive process, commonly assumed in hospitals to inspect mental abnormalities. It is one of the most employed neuroimaging modalities for detecting brain atrophy. Even though MRI is an influential device that can be beneficial to identify symptoms of AD in the brain, the acquisition process is long, chiefly owing to the requirement for physical examination of workflow blockages. Currently, numerous methodologies have been proposed for the investigation of Alzheimer’s disease employing image processing, machine learning, and deep learning, which demonstrate superior performance compared to traditional physical methods. This study formulates and constructs a fusion of deep learning models for the detection of Alzheimer’s disease utilizing an optimization algorithm. The primary objective of the proposed FDLM-DADOA model is to detect and categorize Alzheimer’s disease through an ensemble of deep learning models with hyperparameter optimization. The image processing phase employs Wiener filtering to remove noise from the image data. The fusion of deep learning models, specifically the EfficientNet B7, MobileNet, and ResNet-50, is utilized for feature extraction. The FDLM-DADOA model employs a bidirectional long short-term memory system for the identification and categorization of Alzheimer’s disease. The hyperparameter selection for the BiLSTM model can ultimately be implemented through the design of the equilibrium optimization algorithm. The experimental assessment of the FDLM-DADOA approach is conducted utilizing a benchmark picture dataset. The testing results indicated the superior performance of the FDLM-DADOA technology relative to contemporary methods.