睡眠是与清醒交替出现的一种生理状态，并伴随着意识丧失，且参与到学习、记忆以及突触可塑性等高阶认知过程中。因此研究不同觉醒状态下以及不同意识水平下大脑功能的差异对理解睡眠的神经机制具有重要的意义。本文以睡眠剥夺和全麻药物右美托咪定为操控手段，基于静息态功能连接、任务态激活模式以及结构像预测行为学等多模态影像学研究了睡眠剥夺影响大脑的机制，并从系统层面上研究了右美托咪定诱导无意识状态下脑干和丘脑的功能连接模式的变化。此外，本文还评估了非专家检测睡眠过程中的纺锤波这一特征波形的有效性。具体的讲，本文的研究成果主要有以下几个方面：

首先，利用局部独立成分分析（masked Independent Component Analysis, mICA）方法研究24小时睡眠剥夺对海马亚区静息态功能连接的影响。海马在学习，记忆以及认知功能中起到至关重要的作用，但是海马不是一个单一的结构，海马不同亚区的功能连接受睡眠剥夺的影响还不清楚。本文使用mICA将左右海马划分为10个亚区，比较睡眠剥夺前后海马各亚区全脑功能连接的变化。研究发现睡眠剥夺后右侧前海马与左额下回、初级躯体感觉皮层以及视觉相关脑区之间的功能连接由负变正。此外，左侧后海马与左侧罗兰多岛盖部、双侧脑岛、左侧中央后回、听觉皮层、右侧距状回、双侧舌回以及双侧梭状回之间的功能连接在睡眠剥夺后亦显著增加。这些结果表明不同海马亚区的功能连接受睡眠剥夺的影响不同。

其次，本文利用停止信号任务（Stop Signal Task, SST）和任务态功能磁共振技术，研究睡眠剥夺损伤抑制控制能力的脑机制。本文首先比较了睡眠剥夺前后抑制控制任务中的行为表现和脑激活模式，发现睡眠剥夺显著降低了被试的抑制控制能力，且降低了抑制控制网络（额下回，辅助运动区，丘脑底核以及脑岛）和视觉相关皮层的激活强度；然后分析睡眠剥夺对脑区激活强度与抑制控制能力相关性的影响，发现双侧额下回、左侧丘脑底核以及左侧舌回的激活与抑制控制能力的相关性在睡眠剥夺之后发生显著变化。

上述研究发现睡眠剥夺会降低人们的抑制控制能力。而损伤的抑制控制能力会严重影响人们的日常生活。睡眠剥夺后抑制控制能力的降低会造成很严重的后果，能够准确地预测睡眠剥夺对抑制控制的影响可以提前规避危险。因此，本文使用基于体素的形态学分析方法（voxel based morphometry, VBM）以及机器学习的方法评估局部脑区的灰质体积对睡眠剥夺后抑制控制能力变化量（△SSRT）的预测效果。本文首先找到与△SSRT相关的脑区，然后提取出这些脑区的灰质体积，最后使用线性回归模型以及四折交叉验证评估预测效果，发现左侧额中回、右侧额下回岛盖部、左侧额下回三角部、右侧罗兰多区岛盖部、左侧辅助运动区、左侧海马、右侧舌回、右侧中央后回以及左侧颞中回这九个脑区可以准确的预测△SSRT。

接下来，本文以右美托咪定麻醉药物为操控手段，基于EEG/fMRI融合的技术手段研究右美托咪定诱导意识丧失状态下对觉醒系统脑干以及丘脑亚区功能连接的影响。在整个实验过程中被试同时接受同步EEG/fMRI扫描以及右美托咪定药物的注射。为了确定被试的意识状态水平，每隔1分钟给被试一个听觉刺激（被试名字），并要求被试按键，若被试按键，说明被试仍处于清醒状态；若被试不按键，说明被试进入无意识状态。行为学上，发现所有被试在基线状态均对声音刺激有响应，但每个人进入无意识状态所需的时间有很大的个体差异。影像学上，本文使用mICA将脑干分割成9个亚区，并比较清醒状态和无意识状态之间脑干各亚区与全脑体素的功能连接差异，发现进入无意识状态后觉醒中心脑区（中脑导水管周围灰质和背外侧被盖区）所属的脑干亚区与默认网络、额顶网络、辅助运动区、中央后回以及颞叶皮层之间的功能连接显著降低，但脑干去甲肾上腺素中心所属的脑干亚区与默认网络、额顶网络、海马、丘脑、颞叶以及小脑的功能连接在进入无意识状态后显著增加。此外，本文对丘脑进行类似的分析，发现右侧丘脑背内侧核与运动网络、右侧丘脑枕内侧部与视觉相关脑区、左侧丘脑内侧腹侧核与右侧距状回和右侧舌回、左侧丘脑中央内侧核以及左侧丘脑背内侧核所属的丘脑亚区与左侧中扣带和左侧辅助运动区、左侧丘脑枕与全脑大多脑区之间的功能连接在无意识状态下显著降低，而右侧丘脑枕外侧部与右侧楔前叶的正向功能连接在无意识状态下显著增加。

上述研究基于核磁数据研究了右美托咪定诱导无意识状态下大脑的变化。在EEG方面，前人发现右美托咪定诱导的无意识状态能产生类似于正常睡眠状态下的脑电波形——纺锤波。该波形不仅是学习和智力的生理指标，还对一些疾病具有一定的辅助诊断价值。准确地检测出该波形是很有必要的。因此，本文基于众包的方法评估非专家检测纺锤波的有效性。本文招募5个专家和168个非专家使用MATLAB界面手动标记N2和N3阶段中的纺锤波。在标记的过程中区分每个纺锤波是确定性纺锤波还是不确定性纺锤波。首先本文介绍一种根据专家和非专家自身的标记结果获取组一致性阈值的方法，对专家组一致性或非专家组一致性取该阈值得到专家组标准和非专家组标准。然后，根据匹配算法将非专家组标准与专家组标准进行比较，从而评估非专家组标准的检测效果，发现当仅考虑确定性纺锤波时，非专家组标准的检测效果最好。本文还实现了四种常用的纺锤波检测自动算法，并分别与专家组标准进行比较，发现仅考虑确定性纺锤波的非专家组标准的检测效果优于这四种自动算法。此外，还发现至少需要6个非专家标记N2阶段，9个非专家标记N3才可以保持非专家的检测效果。

综上所述，不同的觉醒水平以及意识状态下大脑的功能连接模式均发生了显著的变化。睡眠剥夺不仅改变了海马亚区的功能连接模式，还降低了人们的抑制控制能力以及抑制控制网络的激活强度。抑制控制网络脑区、躯体感觉皮层以及海马脑区的灰质体积可以准确的预测睡眠剥夺前后抑制控制的变化。在另一种觉醒水平由右美托咪定诱导的意识丧失状态下，觉醒系统脑干和丘脑亚区的功能连接模式亦发生了显著变化。此外，本文提供了一个详细的以众包的方式由非专家检测纺锤波的检测过程。

Sleep alternates with wakefulness, accompanied by a loss of consciousness, and are involved in higher-order cognitive processes such as learning, memory and synaptic plasticity. Therefore, it is of great significance to study the differences in brain function between different arousal level and different levels of consciousness in order to understand the neural mechanism of sleep. This study investigated the effect of sleep deprivation on brain based on the resting-state functional connectivity, task-state activation patterns and prediction of behavior via structural data, and explored the changes of functional connectivity in brainstem and thalamus during dexmedetomidine-induced unconsciousness. Further, this present study estimated the validation of sleep spindle detection based on non-experts. The main findings in present study are as follows:

Firstly, this study used a masked independent component analysis to partition the hippocampus into ten small regions and investigated the changes in the functional connectivity with the whole brain after 24 hours of sleep deprivation in 40 normal young subjects. Increased functional connectivity was found between the right anterior hippocampus and left inferior frontal gyrus, bilateral postcentral gyrus (PoCG), bilateral precuneus and vision-related regions after sleep deprivation. Similar results were also identified between the left posterior hippocampus and the pars opercularis of left rolandic area, bilateral insula, left PoCG, left superior temporal gyrus, bilateral lingual gyrus and fusiform gyrus. These results reflect differential effects of sleep deprivation on the functional connectivity in specific hippocampal regions and provide new insights into the impact of sleep deprivation on the resting-state functional organization in the human brain.

Secondly, this study used fMRI to examine the effects of 24 h of sleep deprivation on cerebral activation during a stop-signal task in 20 normal young subjects. Behaviorally, subjects showed significantly delayed stop-signal reaction time (SSRT) following sleep deprivation. In addition, reduced cerebral activation was found in the stopping network (including the inferior frontal gyrus [IFG], supplementary motor area, subthalamic nucleus [STN] and insula) and vision-related regions (occipital cortex, lingual gyrus and fusiform gyrus) after sleep deprivation. These findings support the hypothesis that task-related activation in prefrontal cortex is particularly vulnerable to sleep deprivation. Furthermore, we observed significant interaction effects of state (sleep deprivation or rested wakefulness) with activation in bilateral IFG, left STN and left lingual gyrus on SSRT. In conclusion, sleep deprivation is associated with the deficits in inhibition-related neural activation and the altered correlation between SSRT and cerebral activation, especially in the bilateral IFG, left STN and left lingual gyrus.

Sleep deprivation for one night can impair the preformance of inhibition control. Poor response inhibition may profoundly interfere with the requirements of everyday life. Therefore, a decrease in response inhibition ability after sleep deprivation could have deleterious outcomes, and being able to predict the effect of sleep deprivation on inhibitory control could help avoid danger in the future. In this study, structural MRI data were used to predict the change in response inhibition performance (△SSRT) measured by a stop-signal task after 24 hours of sleep deprivation in 52 normal young subjects. For each subject, T1-weighted MRI data were acquired and the gray matter (GM) volumes were calculated using voxel-based morphometry analysis. First, the regions in which GM volumes correlated with △SSRT were explored. Then, features were extracted from these regions and the prediction process was performed using a linear regression model with four-fold cross-validation. We found that the GM volumes of the left middle frontal gyrus, pars opercularis of right inferior frontal gyrus, pars triangularis of left inferior frontal gyrus, pars opercularis of right rolandic area, left supplementary motor area, left hippocampus, right lingual gyrus, right postcentral gyrus and left middle temporal gyrus could predict the △SSRT with a low mean square error and a high Pearson’s correlation coefficient between the predicted and actual values. In conclusion, our results demonstrated that a linear combination of structural MRI data could accurately predict the change in response inhibition performance after SD.

Forthly, this study explored the effect of dexmedetomidine on the functional connectivity in brainstem and thalamus subregions based on the simultaneous EEG/fMRI data. All subjects underwent the simultaneous EEG/fMRI scaning and dexmedetomidine infusion. In order to evaluate the arousal level, the auditory stimuli consisted of the subject’s name were presented every 1 min throughout the second visit and subjects were instructed to press the left button with their right index finger. Behaviorally, all subjects have response to the auditory stimuli during baseline. Significant individual difference of the unconscious latency have been observed. In addition, this study partitioned brainstem into nine small regions using mICA and compared the brainstem subregions-to-whole-brain functional connectivity between wakefulness and dexmedetomidine-induced unconsciousness states. One subregion of brainstem which included the major arousal centers: periaquaductal gray (PAG) and laterodorsal tegmental area (LDT) exhibited significantly reduced functional connectivity with default mode network, frontoparietal network, supplementary motor area, postcentral gyrus and temporal cortex during dexmedetomidine-induced unconsciousness state. The increased functional connectivity between the brainstem noradrenergic centers and default mode network, frontoparietal network, hippocampus, thalamus, temporal cortex and cerebellum during dexmedetomidine-induced unconsciousness state. Furthermore, similar analyses were performed on thalamus. The reduced functional connectivity between the right mediodorsal thalamic nucleus and motor network, between right medial pulvinar and visual related regions and between left medioventral nucleus and right calcarine sulcus and right lingual gyrus during unconsciousness state. One subregion of thalamus which included the left central median nucleus and left mediodorsal nucleus also exhibited reduced functional connectivity with left middle cingulate cortex and left supplementary motor area during unconsciousness state. The left pulvinar showed reduced functional connectivity with various brain regions. However, the increased functional connectivity was observed between the right lateral pulvinar and right precuneus during unconsciousness state.

The above study investigated the changes of brain functional connectivity on fMRI data during dexmedetomidine-induced unconsciousness. For EEG, previous study have found that dexmedetomidine produces sleep spindles and sleep spindles were similar during dexmedetomidine sedation and normal sleep. Sleep spindles may be considered both as a physiological index of intellectual abilities and a marker of the capacity for learning. They may also have important supplementary diagnostic value. Therefore, it is imperative to accurately detect spindles. The present study described a method of using non-experts for manual detection of sleep spindles.We recruited five experts and 168 non-experts to manually identify spindles in N2 and N3 sleep data using a MATLAB interface. Scorers classified each spindle into definite and indefinite spindle (with weights of 1 and 0.5, respectively). First, a method of optimizing the thresholds of the expert/non-expert group consensus according to the results of experts and non-experts themselves is described. Using this method, we established expert and non-expert group standards from expert and non-expert scorers, respectively, and evaluated the performance of the non-expert group standards by compared with the expert group standard (termed EGS). The results indicated that the highest performance was the nonexpert group standard when definite spindles were only considered (termed nEGS-1). Second, four automatic spindle detection methods were compared with the EGS. We found that the performance of nEGS-1 versus EGS was higher than that of the four automated methods. Further, we found that six and nine non-experts were needed to manually identify spindles in stages N2 and N3, respectively, while maintaining acceptable performance of nEGS-1 versus EGS. In conclusion, this study establishes a detailed process for detection of sleep spindles by non-experts in a crowdsourcing scheme.

In summary, the functional connectivity pattern have significant chagnes in different arousla levels and different conscious staes. Sleep deprivation altered the functional connectivity in hippocampal subreigons, and caused significant declines in inhibition control ability and deficits in inhibition-related neural activation. The gray matter volumes of stopping network, somatosensory cortex and hippcampus could accurately predict the change in response inhibition performance after sleep deprivation. During the other arousal level dexmedetomidine-induced unconsciousness, the functional connectivity in brainstem and thalamus subregions have significant changes. Finally, this study established a detailed process for detection of sleep spindles by non-experts in crowdsourcing scheme.