Assessing Risk for Alzheimer's Disease by Features of White Matter in Brain Magnetic Resonance Imaging.

• Main Authors: HSU,MIN-HUEI, 許敏惠
• Other Authors: FU,JA-CHIH
• Format: Others
• Language: zh-TW
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/3y4nvq

碩士 === 國立雲林科技大學 === 工業工程與管理系 === 106 === Taiwan is an aging society, elderly accounts for 13.7% of the total population, and 5.4 people of every hundred people the age over sixty-five having dementia. The aging dementia patients has been increasing year by year. Long-term care for dementia may become more difficult in the future. Alzheimer's disease is one of the degenerative diseases of the brain. The current diagnostic method is using Mini-Mental State Examination and many different scales to evaluate patient behavior and cognitive ability, also can use the Computed Tomography and Magnetic Resonance Imaging(MRI) and other medical equipment to do auxiliary diagnosis. The study looked forward to analyzing data more rapidly and assisting physician diagnosis. Extract Features of the Magnetic Resonance Imaging and Diffusion Tensor Imaging by Voxel-Based Morphometry (VBM) and Tract-Based Spatial Statistics (TBSS).Choosing the important features of 189 features by using Random Forest model. Each feature to be two levels by isometric binning, and make them to be the input of Bayesian Network. According to the results, predict whether patient is Alzheimer's disease. If MR image processing by TBSS method, and the features' level of the class one are B. (except the feature L_WM_Parietal_RD). This patient could be Alzheimer's disease. If MR image processing by VBM method, and the ten features' level of the class one are B (except the features R_WM0.2_Sub.lobar._Cp and W_WM0.2_Temporal_L1).This patient could be Alzheimer's disease. According to the probability of each feature and the above rules, provide a basis for diagnosis.