碩士 / 國立臺北科技大學 / 電機工程研究所 / 105 / Alzheimers disease, also known as brain degeneration, is one kind of slow the progression, along with the neurological disorder. It is easy to diagnose in older adults and the causes have not fully understood or discovered yet. Early and mid- term patients will have a memory loss, personality changes, and language impairment such speech become incomprehensive and incoherent, as well as repeat some words or actions. When the symptoms get worse, the patients will lose the sense of direction, at the end or late stage, patients eventually unable to move or walk, can become bedridden. Although there is no specific treatment for Alzheimers disease, early detection of Alzheimers disease will be able with early correspondence and prevention, delay the rate of sustained deterioration.
Recently, many studies have pointed out that there is a close connection between the brain degeneration and Alzheimers disease, such as the hippocampus and cerebral cortex. In this study, we used the Freesurfer software developed by Harvard University to analyze the brain Magnetic Resonance Imaging (MRI) images of patients with Alzheimers disease, and compare with the control group. The extraction of various information from different parts of their brains, and then calculate the Alzheimers disease by C4.5 algorithm, with Gain Ratio from patients brain and build a decision tree.
In this study, we used the Freeurfer software developed by Harvard University to analyze the brain Magnetic Resonance Imaging(MRI) images of patients with Alzheimers disease, and compare with the control group. The extraction of various information from different parts of their brains, and then calculate the Alzheimers disease by C4.5 algorithm, with Gain Ratio from patients brain and build a decision tree. Finally, the test data are tested for the accuracy, precision, sensitivity and specificity by using the confusion matrix method to evaluate the prediction efficiency of the decision tree. Based on the results, it is able to effectively filter out from patients with Alzheimers disease from the test data. The rate with prediction is more than 70% accuracy, it can provide an effective identification for the risk assessment of Alzheimers disease.
| Identifer | oai:union.ndltd.org:TW/105TIT05442052 |
| Date | January 2017 |
| Creators | YiMin.Hsu, 許逸民 |
| Contributors | 黃有評 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 49 |
Page generated in 0.0147 seconds