Prediction of Alzheimer’s disease with Selected Serum Proteins Using Support Vector Machine / 以支持向量機預測血清中蛋白與阿茲海默症之關係

碩士 / 臺北醫學大學 / 醫學檢驗暨生物技術學系所 / 102 / The problem of elderly care has become increasingly evident in developed countries in recent years due to their aging societies. One of the most important concerns of elderly care is the prevention and treatment of Alzheimer''s disease (AD). AD is a progressive neurodegenerative disease. Early symptoms are not obvious, and the diagnosis relies heavily on the clinical staff’s experience, such as using MMSE scores. There is no definitive diagnosis biomarker to date. Therefore we wanted to find specific proteins in patients’ serum to identify effective diagnostic biomarkers for AD and help track and treat the disease. Studies have shown that the neurodegenerative Huntington’s disease (HD)-associated protein (Hap1) was involved in the intracellular transport of AD-related amyloid precursor protein (APP).We checked the changes of these related proteins in AD patients’ serum and tried to develop a new diagnostic method without subjective bias. In the present study, we detected the concentrations of Abelson helper integration site-1 (AHI-1; Jouberin), dynein-1 light intermediate chain (DYNC1LI)-2 and kinesin light chain (KLC)-2 in the serum of healthy people and patients using enzyme-linked immunosorbent assay (ELISA). We found significant differences between the control healthy subjects and AD patients. Then we used support vector machine (SVM) to establish a model combining AHI-1, DYNC1LI2, KLC2, age, and sex to predict AD risks. Our results showed that the best AD biomarker in serum is DYNC1LI2. A combination of DYNC1LI2 with other markers showed a higher area under curve (AUC) in receiver operating characteristic (ROC) curve.

Identiferoai:union.ndltd.org:TW/102TMC05108015
Date January 2014
CreatorsJian-Yu Chen, 陳建宇
Contributors林詠峯
Source SetsNational Digital Library of Theses and Dissertations in Taiwan
Languagezh-TW
Detected LanguageEnglish
Type學位論文 ; thesis
Format74

Page generated in 0.0015 seconds