- 學位論文
A Fuzzy System for Early Dementia Evaluation
A Fuzzy System for Early Dementia Evaluation
摘要
Alzheimer’s disease (AD), also known as geriatric dementia, is the most common and serious form of dementia that affects individuals of old age. Gradual loss of memory and general cognitive functions is observed in AD subjects. The so called well-developed countries are categorized as aged societies where AD cases are increasing due to the higher life expectations. However, the traditional medicine is still unable to figure out the real causes of it. This is why there is a strong need for systems that are able to diagnose AD at early stage to slow down the mental decay of the subject. The thesis proposes a novel system to achieve this need that focuses on the brain cortical thickness and the hippocampus, which are specific brain components in charge of short-term and long-term memory. The main goal is to study the disease by giving a preliminary evaluation of the brain status of the patient by applying fuzzy systems. Inference value between 0 and 1 of contracting AD is the output of the system. The higher the value the more likely the subject may contract it in the future. The systems have been tested by running healthy (Control), schizophrenia (SCZ), age-associated memory impairment (AAMI) and mild cognitive impairment (MCI) patients and they result in a reliable system with well-defined boundaries among the different pathology cases.
並列摘要
Alzheimer’s disease (AD), also known as geriatric dementia, is the most common and serious form of dementia that affects individuals of old age. Gradual loss of memory and general cognitive functions is observed in AD subjects. The so called well-developed countries are categorized as aged societies where AD cases are increasing due to the higher life expectations. However, the traditional medicine is still unable to figure out the real causes of it. This is why there is a strong need for systems that are able to diagnose AD at early stage to slow down the mental decay of the subject. The thesis proposes a novel system to achieve this need that focuses on the brain cortical thickness and the hippocampus, which are specific brain components in charge of short-term and long-term memory. The main goal is to study the disease by giving a preliminary evaluation of the brain status of the patient by applying fuzzy systems. Inference value between 0 and 1 of contracting AD is the output of the system. The higher the value the more likely the subject may contract it in the future. The systems have been tested by running healthy (Control), schizophrenia (SCZ), age-associated memory impairment (AAMI) and mild cognitive impairment (MCI) patients and they result in a reliable system with well-defined boundaries among the different pathology cases.