نوع مقاله : علمی-پژوهشی
نویسندگان
1 دانشکده مهندسی پزشکی، دانشگاه صنعتی سهند تبریز، تبریز، ایران
2 دانشگاه صنعتی سهند تبریز
3 عضو هیات علمی / دانشکده مهندسی پزشکی، دانشگاه صنعتی سهند
4 گروه علوم مهندسی، دانشکده فناوریهای نوین، دانشگاه محقق اردبیلی، نمین، ایران
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
Alzheimer's disease is a neurodegenerative disorder characterized by a gradual decline in cognitive function, including memory loss and impaired thinking. Common symptoms include progressive memory decline, personality changes, difficulty with reasoning and decision-making, and decreased ability to perform daily tasks. Magnetic Resonance Imaging (MRI) can provide high-resolution images of brain structures, which can aid in identifying signs of brain disorders, including Alzheimer's. This study proposes a method for detecting Alzheimer's disease using MRI images from the Kaggle Alzheimer's disease database, which comprises four stages of the disease: healthy, very mild demented, mild demented, and moderate demented. The proposed method involves segmenting MRI images using Fuzzy C-means clustering and particle swarm optimization, followed by feature extraction using Histogram Oriented Gradients (HOG) and Local Binary Pattern (LBP). To address class imbalance, the Synthetic Minority Over-sampling Technique (SMOTE) is applied to generate a balanced dataset. Feature selection is then performed using the ReliefF algorithm. The evaluation results demonstrate that the proposed method achieves high accuracy (99.87%), sensitivity (99.74%), and specificity (99.91%) in classifying different stages of Alzheimer's disease, showing promise for early detection and diagnosis.
کلیدواژهها [English]