Marquetand Justus, Noury Nima, Lu Hongyu, Yang Haodi, Sorrentino Chrystina Montuori, Rüttiger Lukas, Knipper Marlies, Braun Christoph, Löwenheim Hubert, von Fraunberg Johannes, Tropitzsch Anke, Siegel Markus, Wolpert Stephan
Department of Neural Dynamics and Magnetoencephalography, Hertie Institute for Clinical Brain Research, University of Tübingen, Otfried-Müller-Str.47, 72076, Tübingen, Germany.
MEG-Center, University of Tübingen, Otfried-Müller-Str.47, 72076, Tübingen, Germany.
Sci Rep. 2025 May 29;15(1):18900. doi: 10.1038/s41598-025-02956-2.
Phonation is important for our daily communication and requires the activation of internal and external laryngeal muscles, which can be recorded by electromyography (EMG) using surface or needle electrodes. Here we present a new noncontact method, laryngeal magnetomyography. As a proof-of-concept, we investigated the feasibility of differentiating various vocalization conditions using laryngeal MMG in two healthy subjects using optically pumped magnetometers (OPM). We recorded magnetic muscle activity of the larynx and neighboring cervical muscles using a 3 × 5 array of OPMs. Subjects vocalized an /a/ in three different conditions: loud high pitch, loud low pitch, and soft high pitch, in 90 s blocks. After removing cardiac artifacts, MMG signals were in the range of 1.5 pT with significant amplitude differences between conditions. In both subjects, Linear Discriminant Analysis (LDA) was able to significantly classify vocalization conditions based on the spatial pattern of MMG activities. In sum, we show that laryngeal MMG allows contactless differentiation of phonations based on myomagnetic signals. Our results set the stage for future studies to explore this method for clinical diagnostics and therapy. Functional, contactless muscle recordings during vocalization enable new applications for miniaturized quantum sensors, e.g. in linguistic studies and speech rehabilitation.
发声对于我们的日常交流至关重要,并且需要激活喉内肌和喉外肌,这些肌肉活动可以通过使用表面电极或针电极的肌电图(EMG)进行记录。在此,我们提出一种新的非接触式方法——喉磁肌图。作为概念验证,我们使用光泵磁力计(OPM)研究了在两名健康受试者中使用喉磁肌图区分各种发声条件的可行性。我们使用一个3×5的OPM阵列记录了喉部和邻近颈部肌肉的磁肌活动。受试者在90秒的时间段内,在三种不同条件下发声/a/:大声高音、大声低音和轻声高音。去除心脏伪迹后,磁肌图信号幅度在1.5 pT范围内,不同条件之间存在显著的幅度差异。在两名受试者中,线性判别分析(LDA)能够根据磁肌图活动的空间模式对发声条件进行显著分类。总之,我们表明喉磁肌图能够基于肌磁信号对发声进行非接触式区分。我们的结果为未来探索该方法用于临床诊断和治疗的研究奠定了基础。发声过程中的功能性非接触式肌肉记录为小型化量子传感器带来了新的应用,例如在语言学研究和言语康复方面。