- 學位論文
功能性磁化率定量磁振影像於鼠腦視覺光刺激之研究
Functional Quantitative Susceptibility Mapping (fQSM) of Rat Brain during Flashing Light Stimulation
摘要
神經科學家已引入數種評估人類腦部活動的工具。其中，功能性磁振造影（fMRI）因其非侵入性和能將腦部功能對應到解剖結構等特性，成為最受研究者仰賴的工具之一。fMRI技術對大腦活化的血流動力學反應敏感，其取得之結果主要由幅度與相位資訊所組成。迄今，幅度資訊已被深入研究且成為血氧濃度相依（BOLD）fMRI的基礎，但相位資訊則相對地未被重視。本研究中，我們應用了新近發展的fMRI技術，即功能性定量磁化率影像（fQSM），基於相位資訊來評估大鼠腦部受光刺激之活化，並描述腦部的磁化率改變和血氧波動等等。 在光刺激實驗中，動物每0.2秒接受一次閃光以活化視覺路徑，其呼吸氣體包括低濃度(30%)或高濃度(100%)的氧氣，以突顯血氧波動與藉此評估靜脈血氧飽和度（SvO2）。透過fMRI，我們從相位資訊推導出fQSM結果，並與從幅度資訊推導出的BOLD結果進行比較。 從實驗結果，我們發現fQSM有多項特點：一、可行性高。fQSM可經由常用的fMRI程序來獲得。二、信號變化高。光刺激時，於兩種吸入氧氣濃度的條件下，fQSM的信號變化（從相位影像來的磁化率變化）皆是BOLD的信號變化（從幅度影像來的強度變化）的四倍。三、專一性高。從現象學上看，與 BOLD-fMRI產生之活化圖相比，fQSM 產生之活化圖其活化區域更多地局限於視覺路徑。四、非侵入性評估SvO2。透過應用fQSM，我們發現當大鼠呼吸低（高）濃度氧的氣體，在有光刺激時，校正後的SvO2約為84%（88%），而在無光刺激時，校正後的SvO2約為83%（87%）。五、我們還意外發現在峰值上，fQSM的反應比BOLD的反應慢。另外，我們發現fQSM的差異雜訊比(CNR)約是BOLD-fMRI的60%。 因此，與BOLD-fMRI相比， fQSM技術所提供之量化的磁化率資訊，其對外界刺激可展現明顯的反應，並突顯腦部區域的局部反應。此外，靜脈血的定量磁化率值還能更進一步直接轉化為SvO2，這是BOLD-fMRI無法提供的重要生理參數。此外，fQSM反應其峰值模式的代表意義，值得在不久的將來深入探討。就我們所知，本研究是第一個應用fQSM於光刺激的囓齒動物模型研究。我們的發現或可成為小動物fQSM應用的基礎框架。我們相信fQSM技術有潛力研究腦部疾病的機制並評估治療後的恢復情況，有朝一日將可能跟BODL fMRI一樣廣受重視。
並列摘要
Neuroscientists have introduced several tools to assess human brain activity. Among them, functional magnetic resonance imaging (fMRI) probably is the most popular one due to its desirable features, such as non-invasiveness and allowing brain anatomy and function to be correlated. fMRI techniques are sensitive to the hemodynamic response to brain activation. The acquired fMRI data mainly comprises magnitude and phase information. So far, magnitude information has been intensively studied and has become the basis of blood oxygenation level-dependent (BOLD) fMRI, while phase information has been relatively neglected. In this study, we applied a new fMRI technique, namely functional quantitative susceptibility mapping (fQSM), based on phase information to assess light-stimulated rat brain activation and characterized its magnetic susceptibility change and oxygenation fluctuation in the brain. In the light stimulation experiment, the animals received flashing light every 0.2 seconds for activating the visual pathway and breathed gas containing low (30%) or high (100%) oxygen concentration for emphasizing the oxygenation fluctuation to evaluate venous oxygen saturation (SvO2). fMRI was performed, and the fQSM signals were compared with the BOLD signals. Based on our results, there are several characteristics in fQSM. (1) High feasibility. fQSM can be acquired by a common fMRI protocol. (2) High signal change. During light stimulation, fQSM signal change (susceptibility change derived from phase image) was four times larger than BOLD signal change (intensity derived from magnitude image) in both inhalation oxygenation conditions. (3) High specificity. Phenomenologically, the activation maps of fQSM were more restricted to the visual pathway, compared with the relatively diffuse distribution in the activation maps of BOLD-fMRI. (4) Non-invasive assessment of SvO2. By applying fQSM, when the rats breathed the gas at low (high) oxygen concentration, we found that the calibrated SvO2 was about 84% (88%) when the task was on and 83% (87%) when the task was off. (5) We unexpectedly found that the peaked response of fQSM to the light task was slower than that of BOLD. In addition, we found that the contrast-to-noise ratio (CNR) of fQSM is approximately 60% of BOLD-fMRI. Hence, comparing with BOLD-fMRI, fQSM technology provides quantified magnetic susceptibility information showing apparent responses to external stimulations and highlights the local responses in the brain areas. Furthermore, the quantitative susceptibility values of venous blood can also be directly transformed into SvO2, an important physiological parameter that BOLD-fMRI cannot provide. Additionally, the significance of the peaked fQSM response pattern requires further research in the near future. To our best knowledge, this study is the first one applying fQSM to a rodent model stimulated with a flashing light. Our findings may be a framework of fQSM in small animals. We believe that fQSM technology holds the potential to reveal the mechanisms of brain disorders and evaluate their recovery after treatments so that fQSM could be as well accepted as BOLD-fMRI.