Cortical Mechanisms of Tongue Sensorimotor Functions in Humans: A Review of the Magnetoencephalography Approach

Towards best practice of interpreting deep learning models for EEG-based brain computer interfaces

Introduction

As deep learning has achieved state-of-the-art performance for many tasks of EEG-based BCI, many efforts have been made in recent years trying to understand what have been learned by the models. This is commonly done by generating a heatmap indicating to which extent each pixel of the input contributes to the final classification for a trained model. Despite the wide use, it is not yet understood to which extent the obtained interpretation results can be trusted and how accurate they can reflect the model decisions.

Methods

We conduct studies to quantitatively evaluate seven different deep interpretation techniques across different models and datasets for EEG-based BCI.

Results

The results reveal the importance of selecting a proper interpretation technique as the initial step. In addition, we also find that the quality of the interpretation results is inconsistent for individual samples despite when a method with an overall good performance is used. Many factors, including model structure and dataset types, could potentially affect the quality of the interpretation results.

Discussion

Based on the observations, we propose a set of procedures that allow the interpretation results to be presented in an understandable and trusted way. We illustrate the usefulness of our method for EEG-based BCI with instances selected from different scenarios.

Electrophysiological Measures of Swallowing Functions: A Systematic Review

The purpose of this systematic review was to examine the application of event-related potentials (ERPs) to investigate neural processes of swallowing functions in adults with and without dysphagia. Computerized literature searches were performed from three search engines. Studies were screened using Covidence (Cochrane tool) and followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement standards (PRISMA-2009). A total of 759 studies were initially retrieved, of which 12 studies met inclusion criteria. Electrophysiological measures assessing swallowing functions were identified in two major ERP categories: (1) sensory potentials and (2) pre-motor potentials. Approximately 80% of eligible studies demonstrated strong methodological quality, although most employed a case series or case-control study design. Pharyngeal sensory-evoked potentials (PSEPs) were used to assess pharyngeal afferent cortical processing. The temporal sequence of the PSEP waveforms varied based on the sensory stimuli. PSEPs were delayed with localized scalp maps in patients with dysphagia as compared to healthy controls. The pre-motor ERPs assessed the cortical substrates involved in motor planning for swallowing, with the following major neural substrates identified: pre-motor cortex, supplementary motor area, and primary sensorimotor cortex. The pre-motor ERPs differed in amplitude for the swallow task (saliva versus liquid swallow), and the neural networks differed for cued versus non-cued task of swallowing suggesting differences in cognitive processes. This systematic review describes the application of electrophysiological measures to assess swallowing function and the promising application for furthering understanding of the neural substrates of swallowing. Standardization of protocols for use of electrophysiological measures to examine swallowing would allow for aggregation of study data to inform clinical practice for dysphagia rehabilitation.

Functional cortical localization of tongue movements using corticokinematic coherence with a deep learning-assisted motion capture system

Corticokinematic coherence (CKC) between magnetoencephalographic and movement signals using an accelerometer is useful for the functional localization of the primary sensorimotor cortex (SM1). However, it is difficult to determine the tongue CKC because an accelerometer yields excessive magnetic artifacts. Here, we introduce a novel approach for measuring the tongue CKC using a deep learning-assisted motion capture system with videography, and compare it with an accelerometer in a control task measuring finger movement. Twelve healthy volunteers performed rhythmical side-to-side tongue movements in the whole-head magnetoencephalographic system, which were simultaneously recorded using a video camera and examined using a deep learning-assisted motion capture system. In the control task, right finger CKC measurements were simultaneously evaluated via motion capture and an accelerometer. The right finger CKC with motion capture was significant at the movement frequency peaks or its harmonics over the contralateral hemisphere; the motion-captured CKC was 84.9% similar to that with the accelerometer. The tongue CKC was significant at the movement frequency peaks or its harmonics over both hemispheres. The CKC sources of the tongue were considerably lateral and inferior to those of the finger. Thus, the CKC with deep learning-assisted motion capture can evaluate the functional localization of the tongue SM1.

Event-Related Desynchronization and Corticomuscular Coherence Observed During Volitional Swallow by Electroencephalography Recordings in Humans

Swallowing in humans involves many cortical areas although it is partly mediated by a series of brainstem reflexes. Cortical motor commands are sent to muscles during swallow. Previous works using magnetoencephalography showed event-related desynchronization (ERD) during swallow and corticomuscular coherence (CMC) during tongue movements in the bilateral sensorimotor and motor-related areas. However, there have been few analogous works that use electroencephalography (EEG). We investigated the ERD and CMC in the bilateral sensorimotor, premotor, and inferior prefrontal areas during volitional swallow by EEG recordings in 18 healthy human subjects. As a result, we found a significant ERD in the beta frequency band and CMC in the theta, alpha, and beta frequency bands during swallow in those cortical areas. These results suggest that EEG can detect the desynchronized activity and oscillatory interaction between the cortex and pharyngeal muscles in the bilateral sensorimotor, premotor, and inferior prefrontal areas during volitional swallow in humans.