Physikalisch-Technische Bundesanstalt, 10587 Berlin, Germany.
Adv Exp Med Biol. 2010;662:143-8. doi: 10.1007/978-1-4419-1241-1_20.
Near-infrared spectroscopy (NIRS) of the human brain is aiming at the non-invasive determination of concentration changes of oxy- and deoxyhemoglobin in the cortex. However, it usually relies on the assumption of spatially homogeneous absorption changes. To overcome this limitation we performed instrumental and methodological developments of time-resolved NIRS with the aim to achieve depth resolution. We present our recently developed time-domain near-infrared brain imager based on picosecond diode lasers and time-correlated single photon counting (TCSPC) which can be used at the bedside. To achieve depth localization of absorption changes we analysed statistical moments (integral, mean time of flight and variance) of measured time-of-flight distributions of diffusely reflected photons. In particular, variance has a selective sensitivity to deep absorptions changes and provides a suitable representation of cerebral signals. The separation of cerebral and extracerebral changes of hemoglobin concentrations is demonstrated for a motor stimulation experiment.
近红外光谱(NIRS)技术旨在无创地检测大脑皮层中氧合血红蛋白和脱氧血红蛋白的浓度变化。然而,它通常依赖于空间均匀吸收变化的假设。为了克服这一限制,我们对基于皮秒二极管激光器和时间相关单光子计数(TCSPC)的时分辨 NIRS 进行了仪器和方法学的开发,以实现深度分辨率。我们介绍了最近开发的基于皮秒二极管激光器和时间相关单光子计数(TCSPC)的可在床边使用的时域近红外脑成像仪。为了实现吸收变化的深度定位,我们分析了漫反射光子的飞行时间分布的测量的统计矩(积分、平均飞行时间和方差)。特别是,方差对深部吸收变化具有选择性灵敏度,并为脑信号提供了合适的表示。通过对运动刺激实验的血红蛋白浓度的脑内和脑外变化的分离,证明了这一点。