控制功能性近红外光谱中的与下颌相关的运动伪影。
Controlling jaw-related motion artifacts in functional near-infrared spectroscopy.
机构信息
Stephenson School of Biomedical Engineering, University of Oklahoma, Norman, OK 73019, USA.
Stephenson School of Biomedical Engineering, University of Oklahoma, Norman, OK 73019, USA; Institute for Biomedical Engineering, Science and Technology, University of Oklahoma, Norman, OK 73019, USA.
出版信息
J Neurosci Methods. 2023 Mar 15;388:109810. doi: 10.1016/j.jneumeth.2023.109810. Epub 2023 Feb 2.
BACKGROUND
Functional near-infrared spectroscopy (fNIRS) as a non-invasive optical neuroimaging technique has demonstrated great potential in monitoring cerebral activity. Due to its portability and compatibility with medical implants, fNIRS has seen increasing applications in studying the hearing, language and cognitive functions. However, fNIRS is susceptible to artifacts related to jaw movements, such as teeth clenching, swallowing and speaking, which affect recordings over the temporal, parietal and frontal/prefrontal cortices.
NEW METHOD
We investigated two new approaches to control the jaw-related motion artifacts, an individually customized bite bar apparatus and a denoising algorithm namely PCA-GLM based on multi-channel fNIRS recordings from long-separation and short-separation montage. We first recorded data while subjects performed a clenching task, then an auditory task and a resting-state task with and without the bite bar.
RESULTS
Our results have shown that jaw clenching can introduce spurious, task-evoked-like responses in fNIRS signals. A bite bar customized for each participant effectively suppressed the movement-related activities in fNIRS, at both task and resting-state conditions. Moreover, the bite bar and the PCA-GLM denoising method are shown to improve auditory responses, by significantly reducing the within-subject standard deviation, increasing the task-related contrast-to-noise ratio, and yielding stronger activations to the auditory stimuli.
COMPARISON WITH EXISTING METHOD(S): The current study has demonstrated a novel method to control the jaw-related motion artifacts in fNIRS signals.
CONCLUSIONS
Our method will benefit the study of the hearing, language and cognitive functions in normal healthy subjects and patients.
背景
功能近红外光谱（fNIRS）作为一种非侵入性的光学神经成像技术，在监测大脑活动方面显示出巨大的潜力。由于其便携性和与医学植入物的兼容性，fNIRS 在研究听力、语言和认知功能方面的应用越来越广泛。然而，fNIRS 容易受到与咀嚼运动相关的伪影的影响，例如咬牙、吞咽和说话，这些伪影会影响颞叶、顶叶和额/前额叶皮层的记录。
新方法
我们研究了两种控制与咀嚼相关的运动伪影的新方法，一种是为每个参与者定制的咬棒装置，另一种是基于长分离和短分离布局的多通道 fNIRS 记录的基于 PCA-GLM 的去噪算法。我们首先在受试者执行咬牙任务时记录数据，然后在有和没有咬棒的情况下执行听觉任务和静息状态任务。
结果
我们的结果表明，咬牙会在 fNIRS 信号中引入虚假的、与任务相关的响应。为每个参与者定制的咬棒可以有效地抑制 fNIRS 中与运动相关的活动，无论是在任务状态还是静息状态。此外，咬棒和 PCA-GLM 去噪方法被证明可以通过显著降低个体内标准差、提高与任务相关的对比噪声比以及产生更强的听觉刺激反应来改善听觉反应。
与现有方法的比较
本研究证明了一种控制 fNIRS 信号中与咀嚼相关的运动伪影的新方法。
结论
我们的方法将有益于正常健康受试者和患者的听力、语言和认知功能的研究。
Zotero 插件