The Two-Stage Alzheimer’s Disease Automatic Diagnosis Algorithm Based on ST-MBV3 Model
Abstract Alzheimer’s disease (AD), commonly referred to as senile dementia, is a progressive and degenerative brain disorder that significantly impacts cognitive function and memory. As the population ages, the need for early and accurate diagnostic tools has become increasingly critical to mitigate...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2025-06-01
|
| Series: | International Journal of Computational Intelligence Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s44196-025-00883-8 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Abstract Alzheimer’s disease (AD), commonly referred to as senile dementia, is a progressive and degenerative brain disorder that significantly impacts cognitive function and memory. As the population ages, the need for early and accurate diagnostic tools has become increasingly critical to mitigate the impact of the disease. Recent developments in medical imaging technologies, particularly magnetic resonance imaging (MRI), alongside the application of artificial intelligence (AI), have provided new avenues for improving the early detection and diagnosis of AD. This study focuses on leveraging these advancements to enhance the accuracy of AD diagnosis through a novel two-stage algorithm. The proposed model combines an improved 3D DenseNet segmentation model with an enhanced ST-MBV3 classification model. The first stage of the proposed algorithm involves processing the collected images. The resulting brain MRI images are then segmented using an enhanced 3D DenseNet model, thereby constructing a comprehensive dataset for AD classification. In the second stage, the segmented images are then classified using the ST-MBV3 model, a deep learning architecture designed for high-accuracy classification. Experimental results demonstrate impressive classification accuracies, including 98.6% for distinguishing between AD and normal control (NC), 95.97% for mild cognitive impairment (MCI) versus NC, 94.57% for AD versus MCI, and 93.12% for AD/MCI/NC classification. The proposed approach offers promising results for automated and accurate AD diagnosis. This method could play a crucial role in advancing diagnostic procedures and improving patient outcomes. |
|---|---|
| ISSN: | 1875-6883 |