Enhanced Recognition of Amputated Wrist and Hand Movements by Deep Learning Method Using Multimodal Fusion of Electromyography and Electroencephalography

Motion classification can be performed using biometric signals recorded by electroencephalography (EEG) or electromyography (EMG) with noninvasive surface electrodes for the control of prosthetic arms. However, current single-modal EEG and EMG based motion classification techniques are limited owing to the complexity and noise of EEG signals, and the electrode placement bias, and low-resolution of EMG signals. We herein propose a novel system of two-dimensional (2D) input image feature multimodal fusion based on an EEG/EMG-signal transfer learning (TL) paradigm for detection of hand movements in transforearm amputees. A feature extraction method in the frequency domain of the EEG and EMG signals was adopted to establish a 2D image. The input images were used for training on a model based on the convolutional neural network algorithm and TL, which requires 2D images as input data. For the purpose of data acquisition, five transforearm amputees and nine healthy controls were recruited. Compared with the conventional single-modal EEG signal trained models, the proposed multimodal fusion method significantly improved classification accuracy in both the control and patient groups. When the two signals were combined and used in the pretrained model for EEG TL, the classification accuracy increased by 4.18–4.35% in the control group, and by 2.51–3.00% in the patient group.

Gaps and roadmap of novel neuromodulation targets for treatment of gait in Parkinson’s disease

Gait issues in Parkinson’s disease (PD) are common and can be highly disabling. Although levodopa and deep brain stimulation (DBS) of the subthalamic nucleus and the globus pallidus internus have been established therapies for addressing the motor symptoms of PD, their effects on gait are less predictable and not well sustained with disease progression. Given the high prevalence of gait impairment in PD and the limitations in currently approved therapies, there has been considerable interest in alternative neuromodulation targets and techniques. These have included DBS of pedunculopontine nucleus and substantia nigra pars reticulata, spinal cord stimulation, non-invasive modulation of cortical regions and, more recently, vagus nerve stimulation. However, successes and failures have also emerged with these approaches. Current gaps and controversies are related to patient selection, optimal electrode placement within the target, placebo effects and the optimal programming parameters. Additionally, recent advances in pathophysiology of oscillation dynamics have driven new models of closed-loop DBS systems that may or may not be applicable to gait issues. Our aim is to describe approaches, especially neuromodulation procedures, and emerging challenges to address PD gait issues beyond subthalamic nucleus and the globus pallidus internus stimulation.

Median Nerve Stimulation Facilitates the Identification of Somatotopy of the Subthalamic Nucleus in Parkinson’s Disease Patients under Inhalational Anesthesia

Deep brain stimulation (DBS) improves Parkinson’s disease (PD) symptoms by suppressing neuropathological oscillations. These oscillations are also modulated by inhalational anesthetics used during DBS surgery in some patients, influencing electrode placement accuracy. We sought to evaluate a method that could avoid these effects. We recorded subthalamic nucleus (STN) neuronal firings in 11 PD patients undergoing DBS under inhalational anesthesia. Microelectrode recording (MER) during DBS was collected under median nerve stimulation (MNS) delivered at 5, 20, and 90 Hz frequencies and without MNS. We analyzed the spike firing rate and neuronal activity with power spectral density (PSD), and assessed correlations between the neuronal oscillation parameters and clinical motor outcomes. No patient experienced adverse effects during or after DBS surgery. PSD analysis revealed that peripheral 20 Hz MNS produced significant differences in the dorsal and ventral subthalamic nucleus (STN) between the beta band oscillation (16.9 ± 7.0% versus 13.5 ± 4.8%, respectively) and gamma band oscillation (56.0 ± 13.7% versus 66.3 ± 9.4%, respectively) (p < 0.05). Moreover, 20-Hz MNS entrained neural oscillation over the dorsal STN, which correlated positively with motor disabilities. MNS allowed localization of the sensorimotor STN and identified neural characteristics under inhalational anesthesia. This paradigm may help identify an alternative method to facilitate STN identification and DBS surgery under inhalational anesthesia.

Electromyogram-Based Classification of Hand and Finger Gestures Using Artificial Neural Networks

Electromyogram (EMG) signals have been increasingly used for hand and finger gesture recognition. However, most studies have focused on the wrist and whole-hand gestures and not on individual finger (IF) gestures, which are considered more challenging. In this study, we develop EMG-based hand/finger gesture classifiers based on fixed electrode placement using machine learning methods. Ten healthy subjects performed ten hand/finger gestures, including seven IF gestures. EMG signals were measured from three channels, and six time-domain (TD) features were extracted from each channel. A total of 18 features was used to build personalized classifiers for ten gestures with an artificial neural network (ANN), a support vector machine (SVM), a random forest (RF), and a logistic regression (LR). The ANN, SVM, RF, and LR achieved mean accuracies of 0.940, 0.876, 0.831, and 0.539, respectively. One-way analyses of variance and F-tests showed that the ANN achieved the highest mean accuracy and the lowest inter-subject variance in the accuracy, respectively, suggesting that it was the least affected by individual variability in EMG signals. Using only TD features, we achieved a higher ratio of gestures to channels than other similar studies, suggesting that the proposed method can improve the system usability and reduce the computational burden.

Motor Point Heatmap of the Calf

Background: Contractions of muscles in the calf induced by neuromuscular electrical stimulation (NMES) may prevent venous thromboembolism. However, compliance to NMES-treatment is limited by the use of suboptimal stimulation points which may cause discomfort. Knowledge of where one is most likely to find muscle motor points (MP) could improve NMES comfort and compliance.Aims: To anatomically map the MPs of the calf as well as to calculate the probability of finding a MP in different areas of the calf. Material and Methods: On 30 healthy participants (mean age 37 years) anatomical landmarks on the lower limbs were defined. The location of the four most responsive MPs on respectively the medial and lateral calf muscle bellies were determined in relation to these anatomical landmarks using a MP search pen and a pre-set MP search program with 3 Hz continuous stimulation. The anatomy of the calves was normalized and subdivided into a matrix of 48 (6x8) smaller areas (3x3cm), from upper medial to lower lateral, in order to calculate the probability of finding a MP in one of these areas. The probability of finding a MP was then calculated for each area and presented with a 95% confidence interval.Results: The MP heat map displayed a higher concentration of MPs proximally and centrally on the calf. However, there were wide inter-individual differences in the location of the MPs. The highest probability of finding a MP was in area 4, located proximally and medially, and in area 29, located centrally and around the maximum circumference, both with 50% probability (95% CI: 0.31-0.69). The second highest probability of finding MPs was in areas 9, 10, 16, proximally and medially, all with 47% probability (95% CI: 0.28-0.66). These areas 4, 9, 10, 16 and 29 exhibited significantly higher probability of finding motor points than all areas with a mean probability of 27% and lower (p<0.05) The lateral and distal outskirts exhibited almost zero probability of finding MPs. Conclusions: This MP heat map of the calf could be used to expedite electrode placement and to improve compliance in order to receive consistent and enhanced results of NMES treatments.

Preliminary Analysis of a Wireless and Wearable Electronic-Textile EASI-Based Electrocardiogram

Background: With cardiovascular disease continuing to be the leading cause of death and the primary reason for hospitalization worldwide, there is an increased burden on healthcare facilities. Electronic-textile (e-textile)-based cardiac monitoring offers a viable option to allow cardiac rehabilitation programs to be conducted outside of the hospital.Objectives: This study aimed to determine whether signals produced by an e-textile ECG monitor with textile electrodes in an EASI configuration are of sufficient quality to be used for cardiac monitoring. Specific objectives were to investigate the effect of the textile electrode characteristics, placement, and condition on signal quality, and finally to compare results to a reference ECG obtained from a current clinical standard the Holter monitor.Methods: ECGs during different body movements (yawning, deep-breathing, coughing, sideways, and up movement) and activities of daily living (sitting, sitting/standing from a chair, and climbing stairs) were collected from a baseline standard of normal healthy adult male using a novel e-textile ECG and a reference Holter monitor. Each movement or activity was recorded for 5 min with 2-min intervals between each recording. Three different textile area electrodes (40, 60, and 70 mm2) and electrode thicknesses (3, 5, and 10 mm) were considered in the experiment. The effect of electrode placement within the EASI configuration was also studied. Different signal quality parameters, including signal to noise ratio, approximate entropy, baseline power signal quality index, and QRS duration and QT intervals, were used to evaluate the accuracy and reliability of the textile-based ECG monitor.Results: The overall signal quality from the 70 mm2 textile electrodes was higher compared to the smaller area electrodes. Results showed that the ECGs from 3 and 5 mm textile electrodes showed good quality. Regarding location, placing the “A” and “I” electrodes on the left and right anterior axillary points, respectively, showed higher signal quality compared to the standard EASI electrode placement. Wet textile electrodes showed better signal quality compared to their dry counterparts. When compared to the traditional Holter monitor, there was no significant difference in signal quality, which indicated textile monitoring was as good as current clinical standards (non-inferior).Conclusion: The e-textile EASI ECG monitor could be a viable option for real-time monitoring of cardiac activities. A clinical trial in a larger sample is recommended to validate the results in a clinical population.

Use of Neuromuscular Electrical Stimulationfor Swallowing Habilitation in High-Risk Infants: A Survey of Clinical Practice

Purpose: The purpose of this clinical focus article was to identify clinical practice patterns of speech-language pathologists (SLPs) who use neuromuscular electrical stimulation (NMES) with the high-risk infants, specifically, the modality treatment parameters and physiological rationale. Method: An online survey was disseminated to query use of NMES by licensed SLPs who routinely address swallowing habilitation in high-risk infants. Results: Eleven of the 40 practicing SLPs who completed the survey indicated they have used NMES on neonates and infants. Most survey respondents reported not using NMES with infants. Of the SLPs who used this modality, over half reported they relied on clinical judgment for determination of NMES dose, frequency of treatments, electrode placement, and discharge determination. While SLPs acknowledged that little empirical evidence is available, those who used this modality indicated that the American Speech-Language-Hearing Association should support this modality in infants. Conclusions: Reliance on clinical judgment, as indicated by SLPs who use NMES on high-risk infants, is not consistent with the evidence-based practice triad, which encourages the use of high-quality peer-reviewed published evidence to inform clinical decision-making. Additional basic and applied research is needed to support use of NMES as a therapeutic modality in infants.

Electrically evoked and spontaneous neural activity in the subthalamic nucleus under general anesthesia

OBJECTIVE Deep brain stimulation (DBS) surgery is commonly performed with the patient awake to facilitate assessments of electrode positioning. However, awake neurosurgery can be a barrier to patients receiving DBS. Electrode implantation can be performed with the patient under general anesthesia (GA) using intraoperative imaging, although such techniques are not widely available. Electrophysiological features can also aid in the identification of target neural regions and provide functional evidence of electrode placement. Here we assess the presence and positional variation under GA of spontaneous beta and high-frequency oscillation (HFO) activity, and evoked resonant neural activity (ERNA), a novel evoked response localized to the subthalamic nucleus. METHODS ERNA, beta, and HFO were intraoperatively recorded from DBS leads comprising four individual electrodes immediately after bilateral awake implantation into the subthalamic nucleus of 21 patients with Parkinson’s disease (42 hemispheres) and after subsequent GA induction deep enough to perform pulse generator implantation. The main anesthetic agent was either propofol (10 patients) or sevoflurane (11 patients). RESULTS GA reduced the amplitude of ERNA, beta, and HFO activity (p < 0.001); however, ERNA amplitudes remained large in comparison to spontaneous local field potentials. Notably, a moderately strong correlation between awake ERNA amplitude and electrode distance to an “ideal” therapeutic target within dorsal STN was preserved under GA (awake: ρ = −0.73, adjusted p value [padj] < 0.001; GA: ρ = −0.69, padj < 0.001). In contrast, correlations were diminished under GA for beta (awake: ρ = −0.45, padj < 0.001; GA: ρ = −0.13, padj = 0.12) and HFO (awake: ρ = −0.69, padj < 0.001; GA: ρ = −0.33, padj < 0.001). The largest ERNA occurred at the same electrode (awake vs GA) for 35/42 hemispheres (83.3%) and corresponded closely to the electrode selected by the clinician for chronic therapy at 12 months (awake ERNA 77.5%, GA ERNA 82.5%). The largest beta amplitude occurred at the same electrode (awake vs GA) for only 17/42 (40.5%) hemispheres and 21/42 (50%) for HFO. The electrode measuring the largest awake beta and HFO amplitudes corresponded to the electrode selected by the clinician for chronic therapy at 12 months in 60% and 70% of hemispheres, respectively. However, this correspondence diminished substantially under GA (beta 20%, HFO 35%). CONCLUSIONS ERNA is a robust electrophysiological signal localized to the dorsal subthalamic nucleus subregion that is largely preserved under GA, indicating it could feasibly guide electrode implantation, either alone or in complementary use with existing methods.

Electroconvulsive therapy, electric field, neuroplasticity, and clinical outcomes

Electroconvulsive therapy (ECT) remains the gold-standard treatment for patients with depressive episodes, but the underlying mechanisms for antidepressant response and procedure-induced cognitive side effects have yet to be elucidated. Such mechanisms may be complex and involve certain ECT parameters and brain regions. Regarding parameters, the electrode placement (right unilateral or bitemporal) determines the geometric shape of the electric field (E-field), and amplitude determines the E-field magnitude in select brain regions (e.g., hippocampus). Here, we aim to determine the relationships between hippocampal E-field strength, hippocampal neuroplasticity, and antidepressant and cognitive outcomes. We used hippocampal E-fields and volumes generated from a randomized clinical trial that compared right unilateral electrode placement with different pulse amplitudes (600, 700, and 800 mA). Hippocampal E-field strength was variable but increased with each amplitude arm. We demonstrated a linear relationship between right hippocampal E-field and right hippocampal neuroplasticity. Right hippocampal neuroplasticity mediated right hippocampal E-field and antidepressant outcomes. In contrast, right hippocampal E-field was directly related to cognitive outcomes as measured by phonemic fluency. We used receiver operating characteristic curves to determine that the maximal right hippocampal E-field associated with cognitive safety was 112.5 V/m. Right hippocampal E-field strength was related to the whole-brain ratio of E-field strength per unit of stimulation current, but this whole-brain ratio was unrelated to antidepressant or cognitive outcomes. We discuss the implications of optimal hippocampal E-field dosing to maximize antidepressant outcomes and cognitive safety with individualized amplitudes.

Pacemaker associated reduction of left ventricle systolic function

In recent years, data have been accumulated on the negative effect of right ventricular (RV) stimulation, leading to left ventricular (LV) asynchrony, proarrhythmias and progressive heart failure (HF). On the other hand, biventricular pacing has been shown to affect ventricular asynchrony, reduce HF manifestations, and improve prognosis in patients with LV dysfunction and wide QRS complex. The induced asynchrony from apical right ventricular pacing is unequivocally associated with changes in myocardial perfusion, LV dysfunction, and poorer prognosis for patients over time. This has led researchers for decades to look for an alternative position for electrode placement. The incidence of pacemaker-induced cardiomyopathy (PICM) ranges from 5.9 to 39% in patients with RV pacing, depending on the given definition and the limit for the degree of pacing. Upgrading to biventricular pacing has been shown to reverse the cardiomyopathy. Recently, there has been evidence of a positive effect of His bundle pacing (HBP) in the treatment of PICM even in patients with no improvement after biventricular pacing. The question about the pathogenetic mechanisms of PICM is currently unanswered. The connection between electrical asynchrony and the negative effect on cardiac pump function is clear. There is also evidence of an established relationship between asynchrony and coronary blood flow. The predisposing individual characteristics of the patient in which these negative effects are manifested are not clear. This is an issue that requires further studies.