Post-processing of Brain Functional MR Imaging

• Main Authors: Chia-Tan Dai, 戴佳坦
• Other Authors: Jan-Ray Liao
• Format: Others
• Language: zh-TW
• Published: 2002
• Online Access: http://ndltd.ncl.edu.tw/handle/16099945567017229748

碩士 === 國立中興大學 === 電機工程學系 === 90 === In recent years, the research and development of Magnetic Resonance Imaging (MRI) has proceeded very rapidly. The development of ultra-fast and high-resolution imaging methods has opened up brand new opportunities for detecting the status of blood flow in the brain. The technology of functional MRI and its related brain diffusion/perfusion imaging has become more and more mature recently. In a whole body system of 1.5T magnet, we use a high-speed and high-resolution dynamic imaging sequence, to acquire the perfusion images of capillaries after intravenous administration of contrast media (Gd+-DTPA, 0.2mmol/kg, 3cc/sec). The entire imaging process took nine slices, including 45 phases for each slice, in order to obtain the temporal change of the signals . Through image processing, the curve of signal level versus time for the Region of Interest (ROI) in perfusion images was then generated. The relationship between the T2* signal changes and the volume of contrast medium (i.e., blood volume)is linear. From the curve of signals versus time, the Relative Cerebral Blood Volume (rCBV), Relative Cerebral Blood Flow (rCBF), and Tissue Mean Transit Time (tMTT) were obtained. Then, different perfusion mappings were generated. In this thesis, we sent the images obtained from the GE-MRI scanner to a personal computer. After creating the imaging database, the data are processed to obtain the rCBV mapping through auto detection and a number of image-processing techniques, such as filtering and masking.