Scholarlink

<p>The proposed system aims to diagnose early stages of Alzheimer’s Disease (AD) utilizing an automated segmentation tool and supervised machine learning algorithms. T1-weighted, three-dimensional Magnetic Resonance Images (MRI) from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) 1 and ADNI 3 databases are utilized, encompassing subjects classified as Cognitively Normal (CN), Mild Cognitive Impairment (MCI), Early MCI (EMCI), Late MCI (LMCI), and AD. The research comprises two stages. In the initial stage, the VolBrain automated MRI brain volumetry system is employed to derive volumes of brain structures from the MRI scans of distinct subject groups. Ten volumetric features, indicative of AD progression, are selected for subsequent classification. In the second stage, classification of various AD stages is performed using a Fuzzy Inference System (FIS) and Adaptive Neuro Fuzzy Inference System (ANFIS). Two classifier types are introduced in the study: a three-class classifier for AD, MCI, and CN subjects, and a four-class classifier for AD, LMCI, EMCI, and CN subjects. The performance of these classifiers is validated through key metrics, including Accuracy, Sensitivity, Specificity, Precision, Recall, and F-score. The results indicate that ANFIS employing subtractive clustering optimized with a Hybrid algorithm (ANFIS-3) outperforms other configurations, including FIS, ANFIS with its own FIS (ANFIS-1), and ANFIS with Subtracting Clustering optimized with Backpropagation algorithm (ANFIS-2), for both the three- class and four-class classifiers. The proposed system achieves a classification performance of 84.79% accuracy, sensitivity, and F-Score for early stages of AD (EMCI and LMCI) compared to AD and Healthy Control subjects, utilizing the ANFIS classifier on an uneven-sized dataset. The Three-class classifier outperforms the Four-class classifier, and ANFIS-3 demonstrates the highest F-Score (84.79%) among the classifiers, showcasing its superior performance over FIS, ANFIS-1, and ANFIS–2.</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

Diagnosing early stages of Alzheimer’s diseases based on volumetric features from MRI using soft computing algorithms

  • 期刊论文

  • 发表日期：

    2025年5月22日

查看全文
  • Ramya M,
  • Kavitha AR,
  • Kavitha S

摘要

The proposed system aims to diagnose early stages of Alzheimer’s Disease (AD) utilizing an automated segmentation tool and supervised machine learning algorithms. T1-weighted, three-dimensional Magnetic Resonance Images (MRI) from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) 1 and ADNI 3 databases are utilized, encompassing subjects classified as Cognitively Normal (CN), Mild Cognitive Impairment (MCI), Early MCI (EMCI), Late MCI (LMCI), and AD. The research comprises two stages. In the initial stage, the VolBrain automated MRI brain volumetry system is employed to derive volumes of brain structures from the MRI scans of distinct subject groups. Ten volumetric features, indicative of AD progression, are selected for subsequent classification. In the second stage, classification of various AD stages is performed using a Fuzzy Inference System (FIS) and Adaptive Neuro Fuzzy Inference System (ANFIS). Two classifier types are introduced in the study: a three-class classifier for AD, MCI, and CN subjects, and a four-class classifier for AD, LMCI, EMCI, and CN subjects. The performance of these classifiers is validated through key metrics, including Accuracy, Sensitivity, Specificity, Precision, Recall, and F-score. The results indicate that ANFIS employing subtractive clustering optimized with a Hybrid algorithm (ANFIS-3) outperforms other configurations, including FIS, ANFIS with its own FIS (ANFIS-1), and ANFIS with Subtracting Clustering optimized with Backpropagation algorithm (ANFIS-2), for both the three- class and four-class classifiers. The proposed system achieves a classification performance of 84.79% accuracy, sensitivity, and F-Score for early stages of AD (EMCI and LMCI) compared to AD and Healthy Control subjects, utilizing the ANFIS classifier on an uneven-sized dataset. The Three-class classifier outperforms the Four-class classifier, and ANFIS-3 demonstrates the highest F-Score (84.79%) among the classifiers, showcasing its superior performance over FIS, ANFIS-1, and ANFIS–2.