| Summary: | 碩士 === 國立臺北科技大學 === 電機工程系研究所 === 101 === Alzheimer''s disease is one of the dementia forms which has the highest prevalence at the age more than 65. It will cause the decrease of both memory and cognitive ability gradually and we still do not know its causes. In addition, it is also hard to find the disease till the patient shows visible symptoms, and for the families, the burden increases as the disease becomes more severe. However, due to the aging of the society, more and more elderly population are likely to develop Alzheimer''s disease. Hence if we could find some symptoms as early as possible, then we could predict it and slow down the rate of deterioration. In this study, we focus on the correlation between the cortical thickness and Alzheimer''s disease. We use the software named Freesurfer which was developed by Harvard to analyze the brain’s MRI (Magnetic Resonance Imaging). Through it, we can acquire the tissue’s segmentations and reconstruct it into 3D model. Successively, we can obtain the thickness data of the cortex. After that, we use the heat kernel smoothing to filter the thickness features and use the Min-Max diagram to compute the topology of homology, finally we use these results to construct a fuzzy inference system. Results show that the correlation of normal subjects is 32% and the correlation of patients is 73% and it proves the feasibility of proposed system.
|
|---|
