Δευτέρα 9 Νοεμβρίου 2020

Deep Temporal-Spatial Feature Learning for Motor Imagery-Based Brain–Computer Interfaces

alkiviadis.1961 shared this article with you from Inoreader
Motor imagery (MI) decoding is an important part of brain-computer interface (BCI) research, which translates the subject's intentions into commands that external devices can execute. The traditional methods for discriminative feature extraction, such as common spatial pattern (CSP) and filter bank common spatial pattern (FBCSP), have only focused on the energy features of the electroencephalography (EEG) and thus ignored the further exploration of temporal information. However, the temporal information of spatially filtered EEG may be critical to the performance improvement of MI decoding. In this paper, we proposed a deep learning approach termed filter-bank spatial filtering and temporal-spatial convolutional neural network (FBSF-TSCNN) for MI decoding, where the FBSF block transforms the raw EEG signals into an appropriate intermediate EEG presentation, and then the TSCNN block decodes the intermediate EEG signals. Moreover, a novel stage-wise training strategy is propose d to mitigate the difficult optimization problem of the TSCNN block in the case of insufficient training samples. Firstly, the feature extraction layers are trained by optimization of the triplet loss. Then, the classification layers are trained by optimization of the cross-entropy loss. Finally, the entire network (TSCNN) is fine-tuned by the back-propagation (BP) algorithm. Experimental evaluations on the BCI IV 2a and SMR-BCI datasets reveal that the proposed stage-wise training strategy yields significant performance improvement compared with the conventional end-to-end training strategy, and the proposed approach is comparable with the state-of-the-art method.
View on the web

Muscle Synergy Functions for Estimating Unmeasured Muscle Excitations Using a Measured Subset

alkiviadis.1961 shared this article with you from Inoreader
Estimation of muscle excitations from a reduced sensor array could greatly improve current techniques in remote patient monitoring. Such an approach could allow continuous monitoring of clinically relevant biomechanical variables that are ideal for personalizing rehabilitation. In this paper, we introduce the notion of a muscle synergy function which describes the synergistic relationship between a subset of muscles. We develop from first principles an approximation to their behavior using Gaussian process regression and demonstrate the utility of the technique for estimating the excitation time-series of leg muscles during normal walking for nine healthy subjects. Specifically, excitations for six muscles were estimated using surface electromyography (sEMG) data during a finite time interval (called the input window) from four different muscles (called the input muscles) with mean absolute error (MAE) less than 5.0% of the maximum voluntary contraction (MVC) and that accounts for 82-88% of the variance (VAF) in the true excitations. Further, these estimated excitations informed muscle activations with less than 4.0% MAE and 89-93% VAF. We also present a detailed analysis of a number of different modeling choices, including every possible combination of four-, three- and two-muscle input sets, the size and structure of the input window, and the stationarity of the Gaussian process covariance functions. Further, application specific modifications for future use are discussed. The proposed technique lays a foundation to explore the use of reduced wearable sensor arrays and muscle synergy functions for monitoring clinically relevant biomechanics during daily life.
View on the web

Reducing Calibration Efforts in RSVP Tasks With Multi-Source Adversarial Domain Adaptation

alkiviadis.1961 shared this article with you from Inoreader
Rapid Serial Visual Presentation (RSVP)-based Brain-Computer Interface (BCI) is an efficient information detection technology by detecting event-related brain responses evoked by target visual stimuli. However, a time-consuming calibration procedure is needed before a new user can use this system. Thus, it is important to reduce calibration efforts for BCI applications. In this article, we propose a multi-source conditional adversarial domain adaptation with the correlation metric learning (mCADA-C) framework that utilizes data from other subjects to reduce the data requirement from the new subject for training the model. This model utilizes adversarial training to enable a CNN-based feature extraction network to extract common features from different domains. A correlation metric learning (CML) loss is proposed to constrain the correlation of features based on class and domain to maximize the intra-class similarity and minimize inter-class similarity. Also, a multi-source fram ework with a source selection strategy is adopted to integrate the results of multiple domain adaptation. We constructed an RSVP-based dataset that includes 11 subjects each performing three RSVP experiments on three different days. The experimental results demonstrate that our proposed method can achieve 87.72% cross-subject balanced-accuracy under one block calibration. The results indicate our method can realize a higher performance with less calibration efforts.
View on the web

Beta-Range Corticomuscular Coupling Reflects Asymmetries in Hand Movement

alkiviadis.1961 shared this article with you from Inoreader
Hand movement in humans is verified as asymmetries and lateralization, and two hemispheres make some distinct but complementary contributions in the control of hand movement. However, little research has been done on whether the information transfer of the motor system is different between left and right hand movement. Considering the importance of functional corticomuscular coupling (FCMC) between the motor cortex and contralateral muscle in movement assessment, this study aimed to explore the differences between left and right hand by investigating the interaction between muscle and brain activity. Here, we applied the transfer spectral entropy (TSE) algorithm to quantize the connection between electroencephalogram (EEG) over the brain scalp and electromyogram (EMG) from extensor digitorum (ED) and flexor digitorum superficialis (FDS) muscles recorded simultaneously during a gripping task. Eight healthy subjects were enrolled in this study. Results showed that left hand yield ed narrower and lower beta synchronization compared to the right. Further analysis indicated coupling strength in EEG-EMG(FDS) combination was higher at beta band than that in EEG-EMG(ED) combination, and exhibited distinct differences between descending (EEG to EMG direction) and ascending (EMG to EEG direction) direction. This study presents the distinctions of beta-range FCMC between left and right hand, and confirms the importance of beta synchronization in understanding the mechanism of motor stability control. The cortex-muscle FCMC might be used as an evaluation approach to explore the difference between left and right movement system.
View on the web

Evaluation of the Nino® Two-Wheeled Power Mobility Device: A Pilot Study

alkiviadis.1961 shared this article with you from Inoreader
Novel technologies such as the Nino® two-wheeled powered mobility device are promoted as offering an intuitive improved experience compared to conventional wheelchair mobility use. The Nino® has a smaller footprint than a power wheelchair, a zero-degree turning radius, tiller-based steering, and relies on the user leaning forwards and back to move and brake. This study aimed to evaluate manual wheelchair users' ability to use the Nino® to complete a variety of wheelchair skills, and also investigated task demand, user confidence, and user perceptions. Twelve participants with a mean of 22 years of experience using a wheelchair completed the study; most had spinal cord injuries and one had multiple sclerosis. Our findings indicate that Wheelchair Skills Test scores were significantly higher for individuals in their manual wheelchair than in the Nino®. Results from the Wheelchair Use Confidence Scale showed that confidence scores increased significantly after completing Nin o® training, and that participants were significantly more confident using their manual chair than the Nino®. Cognitive workload, as measured by the NASA-Task Load Index, was significantly higher in the Nino® than in participants' manual wheelchairs. Findings from qualitative interviews suggest that the Nino® is unlikely to be suitable as a functional replacement of an individual's manual wheelchair. Most participants felt unsafe during braking. Other perceptions included that the Nino may be a good alternative for use as a recreational outdoor mobility device, a powered mobility option to help prevent upper extremity overuse injuries, have a positive impact on social interactions, but that a high degree of focus was required during use. In addition to needing to address safety, us- bility, and functional concerns, the data suggests a clinical focus on training individuals to use these new devices may be necessary for effective community use.
View on the web

FPGA-Based Real-Time Simulation Platform for Large-Scale STN-GPe Network

alkiviadis.1961 shared this article with you from Inoreader
The real-time simulation of large-scale subthalamic nucleus (STN)-external globus pallidus (GPe) network model is of great significance for the mechanism analysis and performance improvement of deep brain stimulation (DBS) for Parkinson's states. This paper implements the real-time simulation of a large-scale STN-GPe network containing 512 single-compartment Hodgkin-Huxley type neurons on the Altera Stratix IV field programmable gate array (FPGA) hardware platform. At the single neuron level, some resource optimization schemes such as multiplier substitution, fixed-point operation, nonlinear function approximation and function recombination are adopted, which consists the foundation of the large-scale network realization. At the network level, the simulation scale of network is expanded using module reuse method at the cost of simulation time. The correlation coefficient between the neuron firing waveform of the FPGA platform and the MATLAB software simulation waveform is 0.9 756. Under the same physiological time, the simulation speed of FPGA platform is 75 times faster than the Intel Core i7-8700K 3.70 GHz CPU 32GB RAM computer simulation speed. In addition, the established platform is used to analyze the effects of temporal pattern DBS on network firing activities. The proposed large-scale STN-GPe network meets the need of real time simulation, which would be rather helpful in designing closed-loop DBS improvement strategies.
View on the web

Intrathecal Delivery of BDNF Into the Lumbar Cistern Re-Engages Locomotor Stepping After Spinal Cord Injury

alkiviadis.1961 shared this article with you from Inoreader
Delivery of neurotrophins to the spinal injury site via cellular transplants or viral vectors administration has been shown to promote recovery of locomotion in the absence of locomotor training in adult spinalized animals. These delivery methods involved risks of secondary injury to the cord and do not allow for precise and controlled dosing making them unsuitable for clinical applications. The present study was aimed at evaluating the locomotor recovery efficacy and safety of the neurotrophin BDNF delivered intrathecally to the lumbar locomotor centers using an implantable and programmable infusion mini-pump. Results showed that BDNF treated spinal cats recovered weight-bearing plantar stepping at all velocities tested (0.3-0.8 m/s). Spinal cats treated with saline did not recover stepping ability, especially at higher velocities, and dragged their hind paws on the treadmill. Histological evaluation showed minimal catheter associated trauma and tissue inflammation, underlinin g that intrathecal delivery by an implantable/programmable pump is a safe and effective method for delivery of a controlled BDNF dosage; it poses minimal risks to the cord and is clinically translational.
View on the web

Longitudinal effects of breast feeding on parent-reported child behaviour

alkiviadis.1961 shared this article with you from Inoreader

11.cover-source.jpg

Objective

Shorter breastfeeding duration has been linked to a range of difficulties in children. However, evidence linking shorter breastfeeding duration to child behavioural problems has been inconclusive. Owing to an almost exclusive focus on early childhood in previous research, little is known about breastfeeding effects on behaviour throughout childhood and adolescence. This study examines the longitudinal effect of breast feeding on parent-reported behaviour in children aged 3–14.

Design

Data come fr om the Millennium Cohort Study, a large, prospective, UK birth cohort study.

Participants

11 148 children, their parents and teachers.

Methods

This study maps the effect of breastfeeding duration on parent-reported child behaviour longitudinally, using latent growth curve modelling and on teacher-reported child behaviour using multiple regression analyses. Breastfeeding duration was assessed through parent interviews when the child was 9 months old. Children's behavioural development was measured using parent-reported Strengths and Difficulties Questionnaires (SDQ) at 3, 5, 7, 11 and 14 years and teacher-reported SDQs at 7 and 11 years.

Results

Breast feeding was associated with fewer parent-reported behavioural difficulties at all ages even after adjusting for potential confounders (<2 months: B=–0.22, 95% CI –0.39 to –0.04; 2–4 months: B=–0.53, 95% CI –0.75 to –0.32; 4–6 months: B=–1.07, 95% CI –1.33 to –0.81; >6 mo nths: B=–1.24, 95% CI –1.44 to –1.04; B=adjusted mean difference of raw SDQ scores at age 3, reference: never breast fed).

Conclusion

This study provides further evidence supporting links between breastfeeding duration and children's socioemotional behavioural development. Potential implications include intervention strategies encouraging breast feeding.

View on the web

Foot and ankle injuries

alkiviadis.1961 shared this article with you from Inoreader

E.cover-source.jpg

Introduction

Musculoskeletal foot and ankle injuries are commonly experienced by soldiers during military training. We performed a systematic review to assess epidemiological patterns of foot and ankle injuries occurring during military training.

Methods

A review of the literature was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The search, done on 14 February 2019, resulted in 1603 reports on PubMed, 565 on Embase and 3 on the Cochrane Library. A fter reading the remaining full-text articles, we included 91 studies.

Results

Among a population of 8 092 281 soldiers from 15 countries, 788 469 (9.74%) foot and ankle injuries were recorded. Among the 49 studies that reported on length of training, there were 36 770/295 040 (18.17%) injuries recorded among women and 248 660/1 501 672 (16.56%) injuries recorded among men over a pooled mean (±SD) training period of 4.51±2.34 months. Ankle injuries were roughly 7 times more common than foot injuries, and acute injuries were roughly 24 times more common than non-acute injuries. Our findings indicated that, during a 3-month training period, soldiers have a 3.14% chance of sustaining a foot and ankle injury. The incidence of foot or ankle injury during military parachutist training was 3.1 injuries per thousand jumps.

Conclusions

Our findings provide an overview of epidemiological patterns of foot and ankle injuries during military training. These data can be used to compar e incidence rates of foot and ankle injuries due to acute or non-acute mechanisms during training. Cost-effective methods of preventing acute ankle injuries and non-acute foot injuries are needed to address this problem.

View on the web

Combination of 137Cs and 210Pb Radioactive Atmospheric Fallouts to Estimate Soil Erosion for the Same Time Scale

alkiviadis.1961 shared this article with you from Inoreader

ijerph-17-08292-g001-550.jpg

IJERPH, Vol. 17, Pages 8292: Combination of 137Cs and 210Pb Radioactive Atmospheric Fallouts to Estimate Soil Erosion for the Same Time Scale

International Journal of Environmental Research and Public Health doi: 10.3390/ijerph17228292

Authors: Foued Gharbi Torfa Hamad AlSheddi Rebai Ben Ammar Medhat Ahmed El-Naggar

Naturally occurring 210Pb and artificial 137Cs fallouts are widely used as radioactive tracers for the determination of water-induced soil erosion for different time scales equal to 50 and 100 years, respectively. There exist several calibration models useful to convert the variation of the inventory of these radiotracers in cultivated soil compared to its value on non-disturbed soil to a soil erosion rate. The most comprehensive calibration models are based on a mass balance approach. In the present work, a new calibration model is proposed. It consists on the generalization of the mass balance approach to a cultivated soil subject to two successive and continuous periods of cultivation. The proposed model combines 210Pb and 137Cs fallouts for the same time scale by relaxing the constraint on 210Pb fallout from being used for 100 years&rsquo; time scale. The model was applied successfully to hypothetical cases and can be used to measure soil erosion rates for practical cases . It is important to note that the proposed model has two main advantages. First, the complementarity between 210Pb and 137Cs fallouts is for the same time scale and not for different time scales, as usually considered and believed in this field. Second, 210Pb fallout is used for time scales less than 100 years. This makes the model useful to estimate soil erosion rates for two successive periods of cultivation. To the best knowledge of the authors, the combination of 210Pb and 137Cs fallouts for the determination of soil erosion rate variation due to change in cultivation practices for the same time scale has never been developed or applied in the past.

View on the web

Elevated blood urea nitrogen is associated with recurrence of post-operative chronic subdural hematoma

alkiviadis.1961 shared this article with you from Inoreader

Abstract

Background

Chronic subdural hematoma (CSDH) is fundamentally treatable with about a 2–31% recurrence rate. Recently, there has been renewed interest in the association between Blood Urea Nitrogen (BUN) and intracranial lesion. Therefore, this paper attempts to show the relationship between BUN and CSDH recurrence.

Methods

A total of 653 CSDH cases with Burr-hole Irrigation (BHI) were enrolled from December 2014 to April 2019. The analyzed parameters included age, gender, comorbidities, laboratory investigations, medication use and hematoma location. The cases were divided into recurrence and non-recurrence groups while postoperative BUN concentration was further separated into quartiles (Q1 ≤ 4.0 mmol/L, 4.0 < Q2 ≤ 4.9 mmol/L, 4.9 < Q3 ≤ 6.4 mmol/L, Q4 > 6.4 mmol/L). Restricted cubic spline regressions and logistic regression models were performed to estimate the effect of BUN on CSDH recurrence.

Results

CSDH recurrence was observed in 96 (14.7%) cases. Significant distinctions were found between recurrence and non-recurrence groups in postoperative BUN quartiles of cases (P = 0.003). After adjusting for the potential confounders, the odds ratio of recurrence was 3.069 (95%CI =1.488–6.330, p = 0.002) for the highest quartile of BUN compared with the lowest quartile. In multiple-adjusted spline regression, a high BUN level visually showed a significantly high OR value of recurrence risk.

Conclusions

Elevated BUN at post-operation is significantly associated with the recurrence of CSDH, and it is indicated that high levels of serum BUN after evacuation may serve as a risk factor for CSDH recurrence.

View on the web