Correlation between lesion location and dysphagia characteristics in post-stroke patients.

OBJECTIVE: To assess the correlation between lesion location and swallowing function characteristics in post-stroke dysphagia (PSD) patients. MATERIALS AND METHODS: We enrolled 133 PSD. The patients were divided into supratentorial and infratentorial stroke groups. We compared the measurements in the videofluoroscopic swallowing study (VFSS) with 3ml and 5 ml of diluted and thickened barium liquid data between supratentorial and brainstem stroke groups. We further compared the difference of VFSS measurements between patients with left hemispheric or right hemispheric lesions (further divided into unilateral hemispheric cortical and subcortical subgroups) and brianstem leison stroke group.To explore the lesion location's effect on different bolus volume, the VFSS measurements of 3ml and 5ml in each subgroups were compared respectively. The measurements of VFSS included the oral transit time, soft palate elevation duration, hyoid bone movement duration (HMD), UES opening duration, pharyngeal transit duration (PTD), stage of ansition duration, and laryngeal closure duration (LCD), the upper esophageal sphincter opening (UESO), hyoid bone superior horizontal displacement, and hyoid bone anterior horizontal displacement. General swallowing function was assessed using the Penetration Aspiration Scale (PAS) and Functional Oral Intake Scale (FOIS). We performed the paired t-test, Spearman's correlation, and Kruskal-Wallis test analysis to characterize the parameters among the groups. RESULTS: Fifty-eight patients were assessed in the final analysis. The HMD (p = 0.019), PTD (p = 0.048) and LCD (p = 0.013) were significantly different between the supratentorial and brainstem lesion groups in 5ml volume. The HMD was significantly different (p = 0.045) between the left cortical and brainstem lesion groups. Significant differences in the HMD (p = 0.037) and LCD (p = 0.032) between the left subcortical and brainstem lesion groups were found in 5ml volume bolus. There was no group different when taking the 3ml volume bolus. Regarding the relationship between food bolus volume and swallowing functions, only the UESO demonstrated a significant difference in the subcortical lesion of the right hemisphere (p = 0.0032) compared the 3 ml and 5 ml volume bolus. The PTD demonstrated a moderate correlation with the PAS scores (r = 0.38, p = 0.0044). The HMD (r = 0.32, p = 0.018) and LCD (r = 0.29, p = 0.039) demonstrated weak correlations with the PAS scores. We did not identify any correlation between the VFSS parameters and FOIS scores in each subgroup level. CONCLUSION: The PSD with brainstem lesion shows more sever dysfunction in the pharyngeal phases. The left hemisphere was engaged in both the oral and pharyngeal phases. Lesions in the bilateral cortical, subcortical, and brainstem regions may impair sensory input.

An Approach for EEG Denoising Based on Wasserstein Generative Adversarial Network.

Electroencephalogram (EEG) recordings often contain artifacts that would lower signal quality. Many efforts have been made to eliminate or at least minimize the artifacts, and most of them rely on visual inspection and manual operations, which is time/labor-consuming, subjective, and incompatible to filter massive EEG data in real-time. In this paper, we proposed a deep learning framework named Artifact Removal Wasserstein Generative Adversarial Network (AR-WGAN), where the well-trained model can decompose input EEG, detect and delete artifacts, and then reconstruct denoised signals within a short time. The proposed approach was systematically compared with commonly used denoising methods including Denoised AutoEncoder, Wiener Filter, and Empirical Mode Decomposition, with both public and self-collected datasets. The experimental results proved the promising performance of AR-WGAN on automatic artifact removal for massive data across subjects, with correlation coefficient up to 0.726±0.033, and temporal and spatial relative root-mean-square error as low as 0.176±0.046 and 0.761±0.046, respectively. This work may demonstrate the proposed AR-WGAN as a high-performance end-to-end method for EEG denoising, with many on-line applications in clinical EEG monitoring and brain-computer interfaces.

The analgesic effect of different interactive modes of virtual reality: A prospective functional near-infrared spectroscopy (fNIRS) study.

Virtual reality has demonstrated its analgesic effectiveness. However, its optimal interactive mode for pain relief is yet unclear, with rare objective measurements that were performed to explore its neural mechanism. Objective: This study primarily aimed at investigating the analgesic effect of different VR interactive modes via functional near-infrared spectroscopy (fNIRS) and exploring its correlations with the subjectively reported VR experience through a self-rating questionnaire. Methods: Fifteen healthy volunteers (Age: 21.93 ± 0.59 years, 11 female, 4 male) were enrolled in this prospective study. Three rounds of interactive mode, including active mode, motor imagery (MI) mode, and passive mode, were successively facilitated under consistent noxious electrical stimuli (electrical intensity: 23.67 ± 5.69 mA). Repeated-measures of analysis of variance (ANOVA) was performed to examine its pain relief status and cortical activation, with post hoc analysis after Bonferroni correction performed. Spearman's correlation test was conducted to explore the relationship between VR questionnaire (VRQ) items and cortical activation. Results: A larger analgesic effect on the active (-1.4(95%CI, -2.23 to -0.57), p = 0.001) and MI modes (-0.667(95%CI, -1.165 to -0.168), p = 0.012) was observed compared to the passive mode in the self-rating pain score, with no significant difference reported between the two modes (-0.733(95%CI, -1.631 to.165), p = 0.131), associated with diverse activated cortical region of interest (ROI) in charge of motor and cognitive functions, including the left primary motor cortex (LM1), left dorsal-lateral prefrontal cortex (LDLPFC), left primary somatosensory cortex (LS1), left visual cortex at occipital lobe (LOL), and left premotor cortex (LPMC). On the other hand, significant correlations were found between VRQ items and different cortical ROIs (r = -0.629 to 0.722, p < 0.05) as well as its corresponding channels (r = -0.599 to 0.788, p < 0.05). Conclusion: Our findings suggest that VR can be considered as an effective non-invasive approach for pain relief by modulating cortical pain processing. A better analgesic effect can be obtained by exciting and integrating cortical ROIs in charge of motor and cognitive functions. The interactive mode can be easily tailored to be in line with the client's characteristics, in spite of the diverse cortical activation status when an equivalent analgesic effect can be obtained.

The Importance of Extensional Rheology in Bolus Control during Swallowing.

Thickened fluids are commonly used in the medical management of individuals who suffer swallowing difficulty (known as dysphagia). Previous studies have shown that the rheological properties of a liquid affect the flow behavior of the bolus in swallowing, such as pharyngeal transit time. While there is no doubt that shear rheology is a highly important factor for bolus flow, it is suspected that extensional properties of a liquid bolus also plays an important role in swallowing, due to elongation of the bolus as it flows through the oropharynx. Our aim in this work was to observe the effect of extensional viscosity on pharyngeal transit time and elongation of the bolus during swallowing. Eight samples of thickened liquid barium that were shear-controlled, but varied in extensional viscosity and two samples that were extensional-controlled, but varied in shear viscosity were swallowed by eight healthy individuals. Data were collected under lateral view of videofluoroscopy swallow study (VFSS); measures of pharyngeal transit time and the ratio of the length to the width of the bolus on the frame of Upper Esophageal Sphincter (UES) opening were taken from the VFSS recordings. It was observed that the pharyngeal transit time generally increases when the fluids are thickened to higher IDDSI consistency. Additionally, higher extensional viscosity fluids reduced the elongation of the bolus during swallowing, thus potentially reducing the risk of post-swallow residue due to bolus breakage. This study confirmed the relevance of the extensional viscosity of the bolus in swallowing.

Task-concurrent anodal tDCS modulates bilateral plasticity in the human suprahyoid motor cortex.

Transcranial direct current stimulation (tDCS) is a non-invasive method to modulate cortical excitability in humans. Here, we examined the effects of anodal tDCS on suprahyoid motor evoked potentials (MEP) when applied over the hemisphere with stronger and weaker suprahyoid/submental projections, respectively, while study participants performed a swallowing task. Thirty healthy volunteers were invited to two experimental sessions and randomly assigned to one of two different groups. While in the first group stimulation was targeted over the hemisphere with stronger suprahyoid projections, the second group received stimulation over the weaker suprahyoid projections. tDCS was applied either as anodal or sham stimulation in a random cross-over design. Suprahyoid MEPs were assessed immediately before intervention, as well as 5, 30, 60, and 90 min after discontinuation of stimulation from both the stimulated and non-stimulated contralateral hemisphere. We found that anodal tDCS (a-tDCS) had long-lasting effects on suprahyoid MEPs on the stimulated side in both groups (tDCS targeting the stronger projections: F (1,14) = 96.2, p < 0.001; tDCS targeting the weaker projections: F (1,14) = 37.45, p < 0.001). While MEPs did not increase when elicited from the non-targeted hemisphere after stimulation of the stronger projections (F (1,14) = 0.69, p = 0.42), we found increased MEPs elicited from the non-targeted hemisphere after stimulating the weaker projections (at time points 30-90 min) (F (1,14) = 18.26, p = 0.001). We conclude that anodal tDCS has differential effects on suprahyoid MEPs elicited from the targeted and non-targeted hemisphere depending on the site of stimulation. This finding may be important for the application of a-tDCS in patients with dysphagia, for example after stroke.