NeuroFusionNet: a hybrid EEG feature fusion framework for accurate and explainable Alzheimer’s Disease detection
8.5
来源：
Nature
关键字：
ADC
发布时间：
2025-12-15 23:40
摘要：
NeuroFusionNet is a novel hybrid framework for the early detection of Alzheimer's Disease using EEG signals. It combines handcrafted features with deep learning techniques, achieving a remarkable accuracy of 94.27% and a macro-F1 score of 0.94. The model demonstrates robustness across various EEG datasets, ensuring its clinical applicability. Furthermore, it incorporates explainability features such as SHAP and Grad-CAM, enhancing interpretability for clinical use. This framework not only addresses the urgent need for early Alzheimer's detection but also supports real-time deployment in clinical settings.
原文：
查看原文
价值分投票
评分标准
新闻价值分采用0-10分制，综合考虑新闻的真实性、重要性、时效性、影响力等多个维度。
评分越高，表示该新闻的价值越大，越值得关注。
价值维度分析
domain_focus
1.0分+1.0分
business_impact
0.8分+0.8分
scientific_rigor
1.5分+1.5分
timeliness_innovation
1.5分+1.5分
investment_perspective
2.5分+2.5分
market_value_relevance
1.0分+1.0分
team_institution_background
0.5分+0.5分
technical_barrier_competition
1.0分+1.0分
关键证据
NeuroFusionNet achieved 94.27% accuracy and 0.94 macro-F1 score.
The model integrates handcrafted features with CNN-derived features for improved classification.
Cross-validation showed minimal variance (SD <0.3%), confirming robustness.
真实性检查
否
AI评分总结
NeuroFusionNet is a novel hybrid framework for the early detection of Alzheimer's Disease using EEG signals. It combines handcrafted features with deep learning techniques, achieving a remarkable accuracy of 94.27% and a macro-F1 score of 0.94. The model demonstrates robustness across various EEG datasets, ensuring its clinical applicability. Furthermore, it incorporates explainability features such as SHAP and Grad-CAM, enhancing interpretability for clinical use. This framework not only addresses the urgent need for early Alzheimer's detection but also supports real-time deployment in clinical settings.