Brain development and aging are two important issues of brain science. Aging is a major risk factor for most common neurodegenerative disease. Early-life development is characterized by dramatic changes, with overlapping and unique patterns of change in aging. Neurological examinations are crucial in the study of brain science, while multimodal imaging combination would apply complementary information for the researches. Nowadays, multimodal image analysis becomes a research hotspot of brain science. In the thesis, we analyzed biomarkers from different modalities to provide complementary information for brain maturation and brain aging. The main contributions of this thesis include following issues: 1. To identify the optimal time window for capturing perfusion information from early 11C-PIB imaging frames (perfusion PIB, 11C-pPIB) and to compare the performance of dual-tracer PET [18F-FDG and amyloid PIB (11C-aPIB)] and "dual biomarker" 11C-PIB PET [11C-pPIB and 11C-aPIB] for classification of AD, MCI and CN subjects. Forty subjects underwent 18F-FDG and 11C-PIB PET studies. Pearson correlation between the 18F-FDG image and sum of early 11C-PIB frames was maximised to identify the optimal time window for 11C-pPIB. The classification power of imaing parameters was evaluated with a leave-one-out validation. The results indicated that 11C-pPIB could serve as a useful biomarker of perfusion for measuring neural activity and improve the diagnostic power of PET for AD in conjunction with 11C-aPIB. 18F-FDG and 11C-PIB dual-tracer PET examination could better detect MCI. 2. Semi-quantitative method and quantitative method are two types of method for PET data analysis. In this thesis, we investigated the correlation between two biomarkers from early frames of 11C-PIB for perfusion (11C-pPIB and 11C-PIB R1) and two biomarkers from late frames of 11C-PIB for amyloid (11C-aPIB and DVR). Specifically, parametric image $R_1$ were generated by simplified reference tissue model (SRTM) analysis of dynamic 11C-PIB data solved by basis function method (BFM). Perfusion image 11C-pPIB was computed as a summation of early 11C-PIB frames. Amyloid image DVR was estimated by Logan plot method and 11C-aPIB was regarded as a summation of late 11C-PIB frames. The results indicated the high correlation between parameter imagings from semi-quantitative method and quantitative method, and provided supportive evidence that 11C-PIB R1 and 11C-pPIB images could be estimated with t...
修改评论