Investigation Of The Feasibility Of Performing Measurements Of Optical Neuronal Signals In Human Brain With Near-infrared Spectroscopy

Brain imaging methods provide us a better view of the living human brain, and we can understand more about how the brain functions. Brain activity is associated with changes in the optical properties of brain tissue. Near-infrared spectroscopy (NIRS) has typically been used to monitor slow hemodynamic responses in the brain by measuring the absorption changes of the light penetrating brain tissue. However, different from the neuronal electrical signal, appearing tens of milliseconds after stimulation, the slow hemodynamic response is consequent to neuronal activity and occurs within seconds after stimulation. Neuronal activity can only be indirectly inferred from the slow hemodynamic response via neurovascular coupling. Changes in light scattering after penetrating brain tissue generate another type of optical signal that reflects brain activity. Previous in vitro and animal studies have shown that neuronal activity directly causes variations in optical scattering, referred to as the optical neuronal signal (ONS). The detection of the optical neuronal signal has attracted much attention, as this measurement can provide functional images of neuronal activity directly in the human cortex, making the optical imaging a powerful tool to study the complex interdependency between neuronal excitation and the ensuing hemodynamic changes, as NIRS provides measurements of these two signals simultaneously. However, the optical neuronal signal is two orders of magnitude and time smaller than the slow hemodynamic signal and therefore much more difficult to detect. Consequently, during the last decade, whether optical neuronal signals can be reliably obtained in vivo using a non-invasive approach remains controversial.The optical neuronal signal is weak and localized; thus, a NIRS system with high detection sensitivity, an optical neuronal signal sensitive experimental protocol, a reasonable probe area, and the effective removal of the background interference, are all important factors for non-invasive detection of the optical neuronal signal. This thesis focused on these important factors and aimed to investigate the feasibility of performing measurements of optical neuronal signals in vivo and noninvasively. The main research contents are as follows:(1) Summarize the typical procedures and methods that are generally applied to process the NIRS data, on the basis of which, the data processing methods used in this study are presented and described in detail. A modified heart filter based on mathematical morphology is proposed to overcome the shortage of the algorithm that currently used to extract the optical neuronal signal. This modified heart filter can remove the background interference effectively even when there are unknown artifacts in the optical data. Finally, a software named "fNIRSA" is developed to analyze the NIRS data.(2) The performance of the home-made continuous-wave (CW) near-infrared imaging system is evaluated and simulation experiments are conducted with the system. Our CW NIRS system was developed based on source modulation, digital lock-in with large sample points and simultaneous sampling data acquisition. Performance analysis demonstrates very high detection sensitivity (on the order of0.1pW) and high temporal resolution (-20ms,48channels). Simulation experiments indicate that intensity changes on the order of0.01%can be resolved by the instrument (average over approximately500stimuli) when using the data analysis methods proposed in this study. Thus, the optical neuronal signal can be detected in vivo and noninvasively theoretically.(3) By carefully designing the experimental protocol and optical probe, in vivo experiments over human motor and visual cortex are conducted by using our CW NIRS system and the data analysis methods proposed in this study. Reliable optical neuronal signals that obtained within the motor and visual cortex are62.5%and77%of the measurements respectively. EEG was simultaneously acquired with the optical signal in the experiment over visual cortex, and there are no significant differences between the latencies of the N75component of the visual evoked potential (VEP) and optical neuronal signals at either wavelength.This study investigate the feasibility of performing measurements of optical neuronal signals in vivo and noninvasively gradually through simulation and in vivo experiments. The results indicate that the optical neuronal signal can be detected in vivo and noninvasively by improving the instrumentation, the stimulation protocols, and the signal processing.