TY  - JOUR
AU  - Aluri, Pravallika
AU  - Parella, Venu Kumari
AU  - Thati, Jagadeesh
PY  - 2026
DA  - 2026/03/01
TI  - NES-Net: Neuro-Synergy Based Alzheimer’s Detection Using MRI Images
JO  - Frontiers in Biomedical Signal Processing
T2  - Frontiers in Biomedical Signal Processing
JF  - Frontiers in Biomedical Signal Processing
VL  - 1
IS  - 1
SP  - 71
EP  - 78
DO  - 10.62762/FBSP.2025.463290
UR  - https://www.icck.org/article/abs/FBSP.2025.463290
KW  - Alzheimer’s disease
KW  - MRI
KW  - deep learning
KW  - neuro-synergy network
KW  - CNN
KW  - transformer
KW  - attention mechanism
KW  - medical image analysis
AB  - Alzheimer disease (AD) is a progressive neurodegenerative disorder that impairs memory and cognitive function in older adults, placing a substantial burden on global healthcare systems. Early and accurate diagnosis is crucial for timely intervention, yet manual interpretation of magnetic resonance imaging (MRI) scans is often subjective, time-consuming, and prone to inter-observer variability and bias. To overcome these challenges, we propose NES-Net (Neuro-Synergy Network), a novel deep learning model for automated Alzheimer's detection from MRI data. NES-Net employs a multi-branch hybrid architecture that simultaneously captures structural, spatial, and semantic features of brain images. By integrating convolutional neural network (CNN) modules for local pattern extraction with transformer-based attention mechanisms for global contextual understanding, the model effectively models both fine-grained details and long-range dependencies. These complementary representations are fused through a dedicated synergy fusion layer, inspired by inter-regional connectivity in human brain networks. Evaluated on benchmark public datasets including ADNI and OASIS, NES-Net outperforms traditional CNNs and state-of-the-art transformer-based models, achieving an overall accuracy of 94.83%, recall (sensitivity) of 93.92%, and AUC of 0.963. These results demonstrate that NES-Net, through its neuro-synergistic learning paradigm, offers a powerful, interpretable, and clinically valuable tool for early AD detection and decision support.
SN  - request pending
PB  - Institute of Central Computation and Knowledge
LA  - English
ER  - 

@article{Aluri2026NESNet,
  author = {Pravallika Aluri and Venu Kumari Parella and Jagadeesh Thati},
  title = {NES-Net: Neuro-Synergy Based Alzheimer’s Detection Using MRI Images},
  journal = {Frontiers in Biomedical Signal Processing},
  year = {2026},
  volume = {1},
  number = {1},
  pages = {71-78},
  doi = {10.62762/FBSP.2025.463290},
  url = {https://www.icck.org/article/abs/FBSP.2025.463290},
  abstract = {Alzheimer disease (AD) is a progressive neurodegenerative disorder that impairs memory and cognitive function in older adults, placing a substantial burden on global healthcare systems. Early and accurate diagnosis is crucial for timely intervention, yet manual interpretation of magnetic resonance imaging (MRI) scans is often subjective, time-consuming, and prone to inter-observer variability and bias. To overcome these challenges, we propose NES-Net (Neuro-Synergy Network), a novel deep learning model for automated Alzheimer's detection from MRI data. NES-Net employs a multi-branch hybrid architecture that simultaneously captures structural, spatial, and semantic features of brain images. By integrating convolutional neural network (CNN) modules for local pattern extraction with transformer-based attention mechanisms for global contextual understanding, the model effectively models both fine-grained details and long-range dependencies. These complementary representations are fused through a dedicated synergy fusion layer, inspired by inter-regional connectivity in human brain networks. Evaluated on benchmark public datasets including ADNI and OASIS, NES-Net outperforms traditional CNNs and state-of-the-art transformer-based models, achieving an overall accuracy of 94.83\%, recall (sensitivity) of 93.92\%, and AUC of 0.963. These results demonstrate that NES-Net, through its neuro-synergistic learning paradigm, offers a powerful, interpretable, and clinically valuable tool for early AD detection and decision support.},
  keywords = {Alzheimer’s disease, MRI, deep learning, neuro-synergy network, CNN, transformer, attention mechanism, medical image analysis},
  issn = {request pending},
  publisher = {Institute of Central Computation and Knowledge}
}