Dynamic Study Of Neurovascular Coupling By Multi-modal Optical Imaging

Neurovascular coupling plays an important role in clinical diagnosis and neuroscience research.A variety of optical imaging techniques are currently used to investigate cerebral blood flow(CBF)or oxygenation signal as an indirect measure of neural activity,but the relationship between neural activity and the associated hemodynamic changes has yet to be fully understood.In this paper,we applied laser speckle imaging,optical imaging of intrinsic signals,and voltage sensitive dye imaging to measure the dynamic evolutions of CBF,total hemoglobin(HbT),deoxyhemoglobin(HbR)and neural activity under square-wave grating stimulation.We analyzed the coupling between CBF and neural activity,as well as the coupling between HbT/HbR and neural activity.Our results showed that from a temporal perspective,neural activity was significantly faster than the associated hemodynamic responses(p<0.05).The temporal resolution of neural activity is in milliseconds,while the hemodynamic signals can occur over several seconds.From a spatial perspective,we computed the maximal response area of neural activity and the area evolutions of the CBF,HbT and HbR responses.It was shown that the area of CBF and HbT/HbR responses could be considered as a biomarker of neural activity at particular time points.In terms of the response magnitude,CBF was the strongest,followed by HbT,while the HbR response was weaker,and the neural activity is the weakest.Our findings about neurovascular coupling not only provided a new methodology for neuroimaging,but further promoted physiological and pathological studies of brain circulatory regulatory mechanisms.The relationship between neural activity and hemodynamics was investigated mostly in the capillary network,thus large blood vessels in this study were considered as interferences.Therefore,we implemented cerebral vessel segmentation using laser speckle contrast imaging data.First,cerebral blood vessel images were obtained by laser speckle contrast imaging.Then we applied and compared wavelet analysis,the conventional region growing algorithm,and directional vessel segmentation algorithm for vessel segmentation.It was found that conventional region growing algorithm had the best performance among these techniques as area overlap measure(AOM)reached 69.23% between the segmented vessels and the reference image.Therefore,conventional region growing algorithm was applied to facilitate the removal of large cerebral vessels in the study of neurovascular coupling.As stimulation parameters are important factors in the study of neurovascular coupling,we explored the CBF response under different stimulating currents and pulse widths.We observed that when the stimulus current was relatively large,changing the pulse width resulted in no significant differences in the response peak,maximal response area,and average intensity of CBF response.When the stimulus current was relatively small,the changes in stimulation pulse width significantly altered the peak and average intensity of CBF response.Additionally,when the pulse width was relatively small,current changes significantly affected the peak and average intensity of CBF response(p <0.05).Our results suggested that the interaction among stimulation parameters might explain the inconsistent findings in optimal stimulus parameter selection from different literature,and also provided a reference for the selection of stimulus parameters in the study of neurovascular coupling.