Laser Speckle Imaging Of Cerebral Blood Flow: From Vasculature To Function

First introduced in the1980s, laser speckle contrast imaging (LSCI),based on the coherent scattering properties of moving particles, has been apowerful tool for full-field and real-time imaging of blood flow in vivo.Compared with other conventional methods which measure the blood flowsuffer from low spatial or temporal resolution, or injection of exogenoussubstances, or scanning regional blood flow for full-field imaging, LSCI isnon-contact with high spatio-temporal resolution without scanning. Recently,LSCI has gained increased attention, in part due to its rapid adoption forblood flow studies in the brain under physiological and pathologicalconditions.In this dissertation, we studied LSCI from three aspects for itsapplication. First, we obtained information on both deoxy-hemoglobinsaturation (based on optical intrinsic signal) and cerebral blood flow (CBF)changes simultaneously from processed LSCI images to study the specifichemodynamic change in response to hypothermia. Second, LSCI was used toinvestigate the functional brain responses to somatosensory stimulation underdifferent conditions, e.g. hypothermia. Third, we studied the real-time changein microcirculation in a rodent model of focal ischemia and post-ischemicreperfusion. In addition, considering the existing problems in LSCI:1)because of the imaging noise and ambient effect in background, the dynamicrange of contrast data is much limited and thus make it difficult to visualizeand analyze the data;2) disturbances due to the breath, heart beating and othernoises decrease the spatial resolution, we successfully dealt with the issuesrespectively according to the methods developed by our lab, i.e. dynamic range enhancement of contrast data based on monotonic point transformationand spatial resolution enhancement by registration method.The major contributions and conclusions of the thesis are summarized asfollow:1) Analysis of the influence of blood speed and exposure time on CBFmonitoring. Simulated experiments were carried out to obtain the linearrelationship between contrast of time-varying speckle and real speed. Theresults showed that the optimum speed range was0-5mm/s, which suggestthat the LSCI system was suitable for monitoring the dynamics of CBF.We also considered about the best exposure time for different conditions inblood flow measurement.2) Influence of hypothermia on CBF and deoxy-hemoglobin change.Hypothermia can happen in daily life accidentally, e.g., in cardiovascularsurgery, or applied as therapeutics in neurosciences critical care unit(NCCU). Clinical induced hypothermia has been demonstrated withneuroprotective effects in patients with traumatic brain injury, stroke andvarious other disorders. It is meaningful to investigate the influences oftemperature change on the CBF. Eighteen male Sprague Dawley rats wereanesthetized with sodium pentobarbital and randomly assigned to mild andmoderate hypothermia groups (n=9each). To study the relative CBFchange and deoxy-hemoglobin saturation in arteriole, venule and capillarylevel under mild (35℃) and moderate (32℃) hypothermia, laser speckleimaging trials were acquired during baseline (37℃), hypothermia (35℃or32℃) and post-rewarming (37℃) phases. By segmentation methodspackaged in Insight Toolkit (ITK) software, the cortical vessels weresegmented to map the CBF. In the mild group, mean CBF in differentvessels all increased throughout the hypothermia and post-rewarmingphases. Deoxy-hemoglobin saturation in veins and capillaries decreasedwithout arteries. On the contrast, mean CBF was reduced by20%at32℃and could return to~90%of the baseline level during post-rewarming inthe moderate group. And the change of deoxy-hemoglobin saturation was not significant at32℃. We would suggest that moderate hypothermia bemore applicable for clinical purpose in the aspect of CBF recovery levelafter the rewarming. But the physiological foundation of CBF change invascular level under hypothermia still need further study.3) Functional CBF changes under normothermia and hypothermia. LSCI wasused to in the rats with high spatio-temporal resolution. In many studies onfunctional neuroimaging, change in local cerebral blood flow (LCBF)induced by somatosensory stimulation is used as a marker for corticalneuronal activity. Until now, a full description of the relationship betweenthe evoked LCBF and neuronal activity has not been given. In thisexperiment, different activation levels caused by stimulus amplitude(0.5mA,1.5mA and2.5mA) were quantitatively investigated, and were ingood agreement with previous laser Doppler measurements. Electricalstimulation of the hind paw was also carried out with1.5mA pulses (0.3ms pulse width) applied at the frequencies of5Hz for4and8s durations.During the8s stimulation, the evoked LCBF exhibited an initial peakfollowed by a plateau phase. These results suggest that the response ofevoked LCBF reflects the integrated neuronal activity during thestimulation period, and it is modulated by a temporal slow function.Furthermore, we investigated the functional change of CBF, accompaniedwith neuronal activation, during hypothermia. So far thetemperature-induced spatiotemporal responses of neural function have notbeen fully understood. Laser speckle images from Sprague Dawley rats(n=8, male) were acquired under normothermia (37℃) and moderatehypothermia (32℃). For each animal, ten trials of electrical hindpawstimulation (2.5mA,0.3ms,5Hz,4s) were delivered under bothtemperatures. Using registered laser speckle contrast analysis, which wassupposed to eliminate the disturbances from breath and heart beating, andtemporal clustering analysis (TCA), we found a delayed response peak anda prolonged response window under hypothermia. Hypothermia alsodecreased the activation area and the amplitude of the peak CBF. Moreover, capillaries and arterioles are the major contributors in the vascularresponses to functional activation.The results suggest that the combinationof LSCI and TCA is a high resolution functional imaging method toimprove the understanding of the spatiotemporal neurovascular couplingof neuronal activity and hemodynamics under normal and pathematologicbrain function.4) Real-time monitoring of CBF changes in rodent model of distal middlecerebral artery occlusion during ischemia and post-ischemic reperfusion. Itis shown that the CBF perfusion of the ischemic area at the ischemichemisphere decreased to about30%of the baseline during30min ischemiaand recovered to more than80%after reperfusion. The CBF of thenonischemic area at the ischemic hemisphere sustained from70%to100%of the baseline during the whole phases. On the other hand, the CBF of theintact hemisphere increased to～120%of baseline during the ischemicperiod and recovered to the baseline level after reperfusion. The highspatio-temporal resolution of LSCI also made it possible to map thedynamic changes in stroke-induced collateral blood flow after middlecerebral artery occlusion. Investigation of the spatial and temporaldistribution and hemodynamics of both unaffected and severe ischemiccortex can not only help understand the hemodynamic mechanism ofstroke, but also offer newaspect of the evaluation the effectiveness ofclinical therapy.In summary, we utilize laser speckle imaging technique to get thespatiotemporal changes in CBF and deoxy-hemoglobin under lowtemperatures and the re-warming, particularly for the clinical mild (35℃) andmoderate (32℃) hypothermia. Furthermore, we investigated the functionalchange in CBF by laser speckle imaging under moderate hypothermia. Andwe were able to monitor the CBF changes of both ischemic and intacthemisphere after dMCAO during ischemia and post-ischemic reperfusion.LSF has distinct advantages over LDF for visualizing CBF changes becauseof its excellent spatiotemporal resolution. This property combined with relatively simple instrumentation has resulted in rapid adoption of LSCI,particularly in neuroscience.