Scholarlink

<p>Alzheimer’s Disease (AD) is a major neurodegenerative disorder that requires early detection to initiate preventive care. Magnetic Resonance Imaging (MRI) provides a non-invasive method to assess brain abnormalities associated with AD. However, manual interpretation of MRI scans is time-consuming and requires expert knowledge. This study presents a deep learning-based model, Medical Image Analysis for Alzheimer’s Disease Detection using Optimal Deep Learning (MIAADD-ODL), designed to automate and optimize AD detection from MRI scans. The model incorporates advanced preprocessing, an improved ShuffleNet architecture for feature extraction, and a Stacked Deep Belief Network (SDBN) for classification. Hyperparameters are tuned using the Enhanced Squirrel Search Algorithm (ESSA). The model is trained and evaluated on publicly available OASIS and Kaggle datasets, comprising 2D MRI brain images categorized into four dementia stages. A Restricted Boltzmann Machine (RBM) is a two-layer generative stochastic neural network consisting of a visible layer and a hidden layer, trained to reconstruct input data by minimizing reconstruction error. It serves as the building block for Deep Belief Networks (DBNs). In the SDBN model used here, multiple RBMs are stacked, where the hidden layer of one RBM becomes the visible layer of the next. This hierarchical pretraining enables the network to capture complex data representations before undergoing supervised fine-tuning for classification. In our study, the RBM initialization facilitates stable weight updates in MRI feature learning, enhancing the accuracy of AD stage classification. Results demonstrate high diagnostic accuracy with 99.82% accuracy, 97.52% precision, and 97.46% sensitivity. The integration of Explainable AI techniques such as Grad-CAM and SHAP aids in model interpretability and builds trust in medical environments.</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

Medical image analysis for Alzheimer’s disease detection using optimal deep learning model

  • 期刊论文

  • 发表日期：

    2026年1月5日

查看全文
  • K Renugadevi,
  • T Jayasankar

摘要

Alzheimer’s Disease (AD) is a major neurodegenerative disorder that requires early detection to initiate preventive care. Magnetic Resonance Imaging (MRI) provides a non-invasive method to assess brain abnormalities associated with AD. However, manual interpretation of MRI scans is time-consuming and requires expert knowledge. This study presents a deep learning-based model, Medical Image Analysis for Alzheimer’s Disease Detection using Optimal Deep Learning (MIAADD-ODL), designed to automate and optimize AD detection from MRI scans. The model incorporates advanced preprocessing, an improved ShuffleNet architecture for feature extraction, and a Stacked Deep Belief Network (SDBN) for classification. Hyperparameters are tuned using the Enhanced Squirrel Search Algorithm (ESSA). The model is trained and evaluated on publicly available OASIS and Kaggle datasets, comprising 2D MRI brain images categorized into four dementia stages. A Restricted Boltzmann Machine (RBM) is a two-layer generative stochastic neural network consisting of a visible layer and a hidden layer, trained to reconstruct input data by minimizing reconstruction error. It serves as the building block for Deep Belief Networks (DBNs). In the SDBN model used here, multiple RBMs are stacked, where the hidden layer of one RBM becomes the visible layer of the next. This hierarchical pretraining enables the network to capture complex data representations before undergoing supervised fine-tuning for classification. In our study, the RBM initialization facilitates stable weight updates in MRI feature learning, enhancing the accuracy of AD stage classification. Results demonstrate high diagnostic accuracy with 99.82% accuracy, 97.52% precision, and 97.46% sensitivity. The integration of Explainable AI techniques such as Grad-CAM and SHAP aids in model interpretability and builds trust in medical environments.