Scholarlink

<p>Alzheimer’s disease (AD), a leading cause of dementia in the elderly, requires accurate and early diagnosis via MRI. This study introduces IGANN-CAD-SBC, an interpretable neural network that classifies AD stages mild, moderate, very mild, and non-demented based on brain changes. The model enhances MRI quality using the Unscented Trainable Kalman filter and applies IGANN for feature extraction and classification. Achieving accuracies up to 99.6% and precision up to 99.8% across all stages, it also delivers high specificity (97.9–99.5%). Compared to models like ADSP-ADSC-FBCMRI and EDAD-BiLSTM-ANN, IGANN-CAD-SBC offers up to 31% higher accuracy, making it a robust and interpretable solution for clinical AD detection.</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

Interpretable generalized additive neural network for classifying Alzheimer’s disease stages via brain changes

  • 期刊论文

  • 发表日期：

    2025年6月24日

查看全文
  • Surendra P. Ramteke,
  • Nilima Surendra Ramteke

摘要

Alzheimer’s disease (AD), a leading cause of dementia in the elderly, requires accurate and early diagnosis via MRI. This study introduces IGANN-CAD-SBC, an interpretable neural network that classifies AD stages mild, moderate, very mild, and non-demented based on brain changes. The model enhances MRI quality using the Unscented Trainable Kalman filter and applies IGANN for feature extraction and classification. Achieving accuracies up to 99.6% and precision up to 99.8% across all stages, it also delivers high specificity (97.9–99.5%). Compared to models like ADSP-ADSC-FBCMRI and EDAD-BiLSTM-ANN, IGANN-CAD-SBC offers up to 31% higher accuracy, making it a robust and interpretable solution for clinical AD detection.