Kamran Muhammad A, Mannan Malik M Naeem, Jeong Myung Yung
Department of Cogno-Mechatronics Engineering, Pusan National University Busan, South Korea.
Front Hum Neurosci. 2016 Jun 9;10:261. doi: 10.3389/fnhum.2016.00261. eCollection 2016.
Functional near-infrared spectroscopy (fNIRS) is a non-invasive neuroimaging modality that measures the concentration changes of oxy-hemoglobin (HbO) and de-oxy hemoglobin (HbR) at the same time. It is an emerging cortical imaging modality with a good temporal resolution that is acceptable for brain-computer interface applications. Researchers have developed several methods in last two decades to extract the neuronal activation related waveform from the observed fNIRS time series. But still there is no standard method for analysis of fNIRS data. This article presents a brief review of existing methodologies to model and analyze the activation signal. The purpose of this review article is to give a general overview of variety of existing methodologies to extract useful information from measured fNIRS data including pre-processing steps, effects of differential path length factor (DPF), variations and attributes of hemodynamic response function (HRF), extraction of evoked response, removal of physiological noises, instrumentation, and environmental noises and resting/activation state functional connectivity. Finally, the challenges in the analysis of fNIRS signal are summarized.
功能近红外光谱技术（fNIRS）是一种非侵入性神经成像方式，可同时测量氧合血红蛋白（HbO）和脱氧血红蛋白（HbR）的浓度变化。它是一种新兴的皮层成像方式，具有良好的时间分辨率，适用于脑机接口应用。在过去二十年中，研究人员已开发出多种方法，用于从观察到的fNIRS时间序列中提取与神经元激活相关的波形。但目前仍没有分析fNIRS数据的标准方法。本文简要综述了用于对激活信号进行建模和分析的现有方法。这篇综述文章的目的是对各种现有方法进行总体概述，以便从测量的fNIRS数据中提取有用信息，包括预处理步骤、微分路径长度因子（DPF）的影响、血流动力学响应函数（HRF）的变化和属性、诱发反应的提取、生理噪声的去除、仪器设备和环境噪声以及静息/激活状态功能连接性。最后，总结了fNIRS信号分析中面临的挑战。
