陳志宏臺灣大學:電機工程學研究所吳昌衛Wu, Changwei W.Changwei W.Wu2010-07-012018-07-062010-07-012018-07-062009U0001-2101200915383700http://ntur.lib.ntu.edu.tw//handle/246246/188018自1992年以來，以血液含氧程度為對比的功能性磁振造影技術(BOLD-fMRI)已逐漸成為探討腦功能的主流技術與標定方法。因其非侵入性、非放射性的優點，搭配嚴謹的實驗設計，此項造影技術在神經科學領域中被廣為使用。不過由於BOLD-fMRI本身不具有量化的能力再加上日趨複雜的實驗需求，促使學界近年來相繼提出多種新式的功能性磁振造影技術，如腦血容積(VASO)以及平靜狀態(Resting-state)功能性造影，以期能彌補傳統方法之不足。但這些新技術目前仍處於發展階段，其理論基礎與功能信雜比皆未臻完善，如能針對其生理模型與技術層面再加以發展改良，這些新式的功能性磁振造影技術將能更進一步引領腦功能研究邁向嶄新的神經科學領域。論文的主旨在於針對腦血容積以及平靜狀態功能性磁振造影(VASO-fMRI and Resting-state fMRI)做進一步的生理探索及技術改良，期望能提供穩定可靠的技術以進行臨床病理與認知心理的腦功能相關研究。本論文共具有四項主要目標，分列如下：一目標：評估水分子滲透現象對於VASO-fMRI估計腦血容積的影響ASO是近年來被提出用於估計腦血容積這項重要生理參數的功能性磁振造影技術。這項技術是建構於忽略水分子在血管壁與神經細胞間通透現象，然而此一假設並不符合實際生理狀況。因此本論文的第一目標是以電腦模型估計水分子通透現象對於VASO-fMRI的生理意義與影響。結果顯示腦血容積的動態變化在神經活動的瞬間將會受到不同水分子滲透模型的影響，進而造成其估計值的時間延遲(第二章)。二目標：增進VASO-fMRI的功能信雜比要干擾腦血容積估計的因素在於VASO-fMRI這項技術的功能信雜比。為了提昇其功能信雜比，我們在VASO-fMRI脈衝序列上進行造影時間的調校，將原本降低血管信號轉為最小化血管周遭的組織信號，並進行兩種實驗設計的比較，其功能信雜比皆可達到40%左右的增益(第三章)。三目標：以Resting-state fMRI信號探討腦部功能性聯結產生之起源年來研究指出在沒有外在刺激的情境之下，腦區間的功能性聯結仍可從平靜狀態的功能性造影技術中擷取而出。然而，這項功能性聯結的成因眾說紛紜，可能是神經的自發性活動，亦或是血流的自發性振動。在這項研究中，我們以不同的功能性造影技術探討能否追溯此功能性聯結至腦中氧氣代謝的生理層面。肯定的結果暗示此功能性聯結很有可能起因於神經細胞的自發性脈動(第四章)。四目標：探討Resting-state fMRI於腦部各功能聯結區域間的頻率分布能性聯結是基於觀察Resting-state fMRI信號的低頻成分，通常是取0.1 Hz以下的信號做為分析依據。然而其頻率成份尚未被準確地評估。因此，我們以多重濾波器來分析不同腦區間的Resting-state fMRI信號並觀察其頻率響應，結果發現最佳的功能性聯結出現於0.01-0.06 Hz的區間內，而不同腦區之間也各自具有獨特的頻率分布曲線(第五章)。論文的研究解開了腦血容積與平靜狀態功能性磁振造影的部份生理爭議，並增進其功能信雜比。在未來發展中，本論文也整合上述技術進行腦血容積的絕對定量與靜坐狀態下腦部功能性聯結的研究(第六章)。這些初步成果顯示這兩種功能性造影技術未來於臨床病理與認知心理上具有高度的應用潛力。總結而言，本論文在神經科學領域提供了功能性磁振造影的生理知識，而其技術改進未來將對腦功能研究有所裨益。Since it was introduced in 1992, functional magnetic resonance imaging (fMRI) based on the blood oxygenation level dependent (BOLD) mechanism has become a popular modality for non-invasive mapping of brain functions incorporating sophisticated experimental design. Within one decade, a few novel fMRI techniques, such as vascular-space occupancy (VASO)-based fMRI and resting-state fMRI, were subsequently invented and proven useful for quantifiable features and for reducing the demand of external stimulation, which are what typical fMRI procedure fails to offer. However, the physiological bases of these newly-developed techniques remain to be refined to improve contrast-to-noise ratio. Therefore, resolving physiological and technical issues would undoubtedly benefit functional brain research. he general hypothesis of this dissertation was that VASO-fMRI and resting-state fMRI techniques could be optimized for both clinical and cognitive research. pecific aim 1: To evaluate the influence of water permeability on CBV estimation using VASO-fMRI ASO-fMRI has been recently developed as a promising technique for estimating CBV, which is an important physiological indicator. However, this technique was based on a non-practical water exchange model. In this specific aim, the effects of the permeability model at two extreme cases were evaluated for CBV quantification. Results showed that estimated CBV has slight difference between two permeability models, but such disparity would be enhanced during the transient state of brain activity, leading to a temporal delay on dynamic CBV estimations (Chapter 2)pecific aim 2: To improve the contrast-to-noise ratio (CNR) of VASO-fMRI he key factor affecting the CBV estimation is the CNR of VASO-fMRI technique. In order to improve the CNR, the VASO-fMRI pulse sequence was modified by amending the imaging timing. The resulting CNR was increased by 40% in both block-design and event-related experiments with visual stimuli (Chapter 3). pecific aim 3: To trace the origin of functional connectivity based on resting-state fMRI to metabolic levelsing resting-state fMRI signals, the functional connectivity of brain networks can be retrieved without external stimulations, but the source of such phenomenon remains obscure. To explore its physiological source, the functional connectivity maps of the cerebral metabolic rate of oxygenation (CMRO2) were followed independent of the hemodynamic changes (Chapter 4), implying that the functional connectivity observed by resting-state fMRI signal is originated from spontaneous neuronal oscillations.pecific aim 4: To explore the frequency distribution of the functional connectivity in multiple brain networks using resting-state fMRIunctional connectivity can be observed using low-frequency (<0.1 Hz) fMRI fluctuations. However, the observing frequency has yet been systematically assessed. In this specific aim, the frequency specificity of resting-state fMRI signal was explored through multiple brain networks. Generally, the maximum connectivity strength falls on 0.01-0.06 Hz, but results differ between distinct brain networks (Chapter 5).n summary, the dissertation unveiled in part the underlying physiology of VASO-fMRI and resting-state fMRI techniques and improved their functional contrast. In future directions, two practical studies of absolute CBV quantification and meditation-based functional connectivity were also presented incorporating the two developed techniques (Chapter 6), expressing their potential for clinical and psychological investigations. In conclusion, the physiological exploration provides a better understanding of the fMRI mechanism in relation to neural activity and the proposed improvements profits further studies of brain functions based upon VASO- and resting-state fMRI techniques.Title page ipproval iiopyright iiiedication ivcknowledgement vhinese Abstracts viiinglish Abstracts xiable of Contents xivist of Tables xxist of Figures xxihapter 1 Introduction.1 Background and Historical Review on Functional Neuroimaging 1.2 Functional MRI Based on Blood Oxygenation Level Dependent (BOLD) Contrast 6.3 Alternative Functional MRI Techniques 9.3.1 Perfusion-based fMRI 10.3.2 CBV-based fMRI 12.3.3 Resting-state fMRI 14.4 Hypothesis and Specific Aims 16hapter 2 Modeling Dynamic CBV Changes during Brain Activation Based on Blood-nulled Functional MRI Signal.1 Introduction 25.2 Theory 28.2.1 Slow Water Exchange without Flow Effects 28.2.2 Fast Water Exchange with Flow Effects 30.3 Methods 34.3.1 Steady State 34.3.2 Transient State 36.4 Results 38.4.1 Steady-state Simulations 38.4.2 Transient-state Simulations 40.5 Discussion 42.5.1 Water Exchange 42.5.2 Physiological Effects 43.5.3 Biophysical Parameters 44.5.4 Baseline Information of CBV & CSF 45.6 Conclusion 47hapter 3 Vascular Space Occupancy Dependent Functional MRI by Tissue Suppression.1 Introduction 54.2 Theory 57.3 Materials and Methods 60.3.1 Determination of Inversion Time 60.3.2 Block-design Experiment 61.3.3 Event-related Experiment 62.3.4 Data Analysis 63.4 Results 66.4.1 Spatial Localization 66.4.2 T1 vs. Signal Change 66.4.3 CNR Comparison 67.5 Discussion 69.5.1 Influence of Gray Matter T1 70.5.2 BOLD and CBF Contamination 71.5.3 CBV Quantification 73.5.4 Reason for CNR Enhancement 75.5.5 Temporal Dynamics 76.6 Conclusion 77hapter 4 Mapping Functional Connectivity Based on Synchronized CMRO2 Fluctuations during the Resting State.1 Introduction 85.2 Materials and Methods 88.2.1 MRI acquisition 88.2.2 Data analysis 90.3 Results 94.3.1 Temporal Dynamics of Functional Connectivity 94.3.2 Spatial Extent of Functional Connectivity 95.3.3 Effects of Biophysical Parameters 97.4 Discussion 98.4.1 Spatial Discrepancies between Networks 98.4.2 Sensitivity Issue 99.4.3 Vascular Contributions 100.4.4 Physiological Bases of Biophysical Parameters 101.5 Conclusion 103hapter 5 Frequency Specificity of Functional Connectivity in Brain Networks.1 Introduction 108.2 Materials and Methods 111.2.1 MRI Acquisition 111.2.2 Data Analysis 112.2.3 ROI Generation 114.3 Results 116.3.1 Between Networks Discrepancy 116.3.2 Within Networks Discrepancy 118.4 Discussion 120.4.1 Spectral Discrepancies in Brain Networks 120.4.2 Neurobiological Basis of Functional Connectivity 121.4.3 Physiological Noise: Respiration and Cardiac Pulse 124.5 Conclusion 125hapter 6 Conclusion, Discussion and Future Directions.1 Conclusion 132.2 Discussion 136.3 Future Directions 139.2.1 Clinical Applications using CBV-based fMRI 139.2.2 Long-term Mental Process using Resting-state fMRI 142.2.3 Functional Connectivity under Meditation 146bbreviations 155eferences 158ita 1737499766 bytesapplication/pdfen-US功能性磁振造影血氧程度依賴對比血液灌流腦血流量腦血容積腦氧代謝率血管空間佔有率自發性擾動功能性聯結functional MRIblood oxygenation level dependentperfusioncerebral blood flowcerebral blood volumecerebral metabolic rate of oxygenationvascular-space occupancyspontaneous fluctuationfunctional connectivitythesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/188018/1/ntu-98-F90921078-1.pdf