Department of Biomedical Engineering, Medical School, Tsinghua University , Beijing , China.
Int J Audiol. 2013 Dec;52(12):824-31. doi: 10.3109/14992027.2013.834537. Epub 2013 Sep 23.
The frequency-following response (FFR) is the compound phase-locked brainstem response to periodic components of sound stimuli, and is closely related to pitch perception. Its weak amplitude often prevents its measurement with a high signal-to-noise ratio (SNR). Recording of FFR using multichannel EEG is possible but expensive and it involves the manual screening of raw data.
We describe a new method to extract FFR features by prescreening the raw data using automatic monitoring of sound pressure in the ear canal. Removal of stimulus artifacts, noise reduction, and data selection were systematically studied.
The reliability of our new method was tested by comparing FFRs tracking accuracy and pitch perception in fifteen individuals with normal hearing.
The extracted FFRs tracking accuracy was significantly correlated with behavioral measures of pitch perception, indicating that FFR could be used to represent individual differences in pitch perception ability among a population with similar hearing characteristics.
The designed system could extract FFR signals more accurately with high SNR after signal prescreen and noise reduction.
频率跟随反应（FFR）是对声音刺激周期性成分的复合锁相脑干反应，与音高感知密切相关。其弱幅度往往阻止了其与高信噪比（SNR）进行测量。使用多通道 EEG 记录 FFR 是可行的，但昂贵，并且涉及原始数据的手动筛选。
我们描述了一种新方法，通过使用耳道中声压的自动监测对原始数据进行预筛选来提取 FFR 特征。系统地研究了刺激伪影的去除、降噪和数据选择。
我们通过比较 15 名听力正常个体的 FFR 跟踪准确性和音高感知，测试了我们新方法的可靠性。
提取的 FFR 跟踪准确性与音高感知的行为测量显著相关，表明 FFR 可用于代表具有相似听力特征的人群中个体音高感知能力的差异。
设计的系统在信号预筛选和降噪后，可以以更高的 SNR 更准确地提取 FFR 信号。
