Real-time intelligent pattern recognition algorithm for surface EMG signals

Background  

Electromyography (EMG) is the study of muscle function through the inquiry of electrical signals that the muscles emanate. EMG signals collected from the surface of the skin (Surface Electromyogram: sEMG) can be used in different applications such as recognizing musculoskeletal neural based patterns intercepted for hand prosthesis movements. Current systems designed for controlling the prosthetic hands either have limited functions or can only be used to perform simple movements or use excessive amount of electrodes in order to achieve acceptable results. In an attempt to overcome these problems we have proposed an intelligent system to recognize hand movements and have provided a user assessment routine to evaluate the correctness of executed movements.     

Application of EMG signals for controlling exoskeleton robots.

Exoskeleton robots are mechanical constructions attached to human body parts, containing actuators for influencing human motion. One important application area for exoskeletons is human motion support, for example, for disabled people, including rehabilitation training, and for force enhancement in healthy subjects. This paper surveys two exoskeleton systems developed in our laboratory. The first system is a lower-extremity exoskeleton with one actuated degree of freedom in the knee joint. This system was designed for motion support in disabled people. The second system is an exoskeleton for a human hand with 16 actuated joints, four for each finger. This hand exoskeleton will be used in rehabilitation training after hand surgeries. The application of EMG signals for motion control is presented. An overview of the design and control methods, and first experimental results for the leg exoskeleton are reported.     

Invariant hip moment pattern while walking with a robotic hip exoskeleton

Robotic lower limb exoskeletons hold significant potential for gait assistance and rehabilitation; however, we have a limited understanding of how people adapt to walking with robotic devices. The purpose of this study was to test the hypothesis that people reduce net muscle moments about their joints when robotic assistance is provided. This reduction in muscle moment results in a total joint moment (muscle plus exoskeleton) that is the same as the moment without the robotic assistance despite potential differences in joint angles. To test this hypothesis, eight healthy subjects trained with the robotic hip exoskeleton while walking on a force-measuring treadmill. The exoskeleton provided hip flexion assistance from approximately 33% to 53% of the gait cycle. We calculated the root mean squared difference (RMSD) between the average of data from the last 15 min of the powered condition and the unpowered condition. After completing three 30-min training sessions, the hip exoskeleton provided 27% of the total peak hip flexion moment during gait. Despite this substantial contribution from the exoskeleton, subjects walked with a total hip moment pattern (muscle plus exoskeleton) that was almost identical and more similar to the unpowered condition than the hip angle pattern (hip moment RMSD 0.027, angle RMSD 0.134, p<0.001). The angle and moment RMSD were not different for the knee and ankle joints. These findings support the concept that people adopt walking patterns with similar joint moment patterns despite differences in hip joint angles for a given walking speed.     

EMG stability as a biofeedback control

Factors that may confound comparisons between electromyographic (EMG) biofeedback training and its control conditions include feedback quality and experience of success. We investigated the usefulness of a control procedure designed to overcome these potential sources of confounding. The procedure consisted of training muscle tension stability. We used it as a control for frontal EMG relaxation training in children with asthma. To equate the groups for feedback quality and experience of success, we gave each child in the control condition audio feedback decreasing in pitch when muscle tension was at or near baseline levels, and feedback increasing in pitch when muscle tension was either substantially above or below baseline levels. Children in both groups were instructed to decrease the pitch of the tone. In comparison to children in the relaxation condition, the children in the control condition exhibited stable levels of muscle tension throughout eight training sessions. We concluded that feedback for stable muscle tension may be a useful control procedure for EMG biofeedback training whenever experimental and control procedures differ in either feedback quality of degree to which they permit subjects to experience success.This research was supported by NIH-Grant HL 27402. We are grateful to Paul Schnitter who constructed the EMG stability feedback device.     

EMG stability as a biofeedback control.

Factors that may confound comparisons between electromyographic (EMG) biofeedback training and its control conditions include feedback quality and experience of success. We investigated the usefulness of a control procedure designed to overcome these potential sources of confounding. The procedure consisted of training muscle tension stability. We used it as a control for frontal EMG relaxation training in children with asthma. To equate the groups for feedback quality and experience of success, we gave each child in the control condition audio feedback decreasing in pitch when muscle tension was at or near baseline levels, and feedback increasing in pitch when muscle tension was either substantially above or below baseline levels. Children in both groups were instructed to decrease the pitch of the tone. In comparison to children in the relaxation condition, the children in the control condition exhibited stable levels of muscle tension throughout eight training sessions. We concluded that feedback for stable muscle tension may be a useful control procedure for EMG biofeedback training whenever experimental and control procedures differ in either feedback quality of degree to which they permit subjects to experience success.     

PEPPeR, a platform for experimental proteomic pattern recognition

Quantitative proteomics holds considerable promise for elucidation of basic biology and for clinical biomarker discovery. However, it has been difficult to fulfill this promise due to over-reliance on identification-based quantitative methods and problems associated with chromatographic separation reproducibility. Here we describe new algorithms termed "Landmark Matching" and "Peak Matching" that greatly reduce these problems. Landmark Matching performs time base-independent propagation of peptide identities onto accurate mass LC-MS features in a way that leverages historical data derived from disparate data acquisition strategies. Peak Matching builds upon Landmark Matching by recognizing identical molecular species across multiple LC-MS experiments in an identity-independent fashion by clustering. We have bundled these algorithms together with other algorithms, data acquisition strategies, and experimental designs to create a Platform for Experimental Proteomic Pattern Recognition (PEPPeR). These developments enable use of established statistical tools previously limited to microarray analysis for treatment of proteomics data. We demonstrate that the proposed platform can be calibrated across 2.5 orders of magnitude and can perform robust quantification of ratios in both simple and complex mixtures with good precision and error characteristics across multiple sample preparations. We also demonstrate de novo marker discovery based on statistical significance of unidentified accurate mass components that changed between two mixtures. These markers were subsequently identified by accurate mass-driven MS/MS acquisition and demonstrated to be contaminant proteins associated with known proteins whose concentrations were designed to change between the two mixtures. These results have provided a real world validation of the platform for marker discovery.     

Phosphocholine as a pattern recognition ligand for CD36

We have previously shown that CD36 recognizes oxidation products of phospholipids on oxidized LDL (OxLDL) such as 1-palmitoyl-2-(5'-oxovaleroyl)-sn-glycero-3-phosphocholine (POVPC). The current study was designed to examine whether the phosphocholine (PC) headgroup in POVPC constitutes an obligatory binding target for CD36. To examine the contribution of PC in the binding of POVPC to CD36, we used well-defined synthetic oxidized phospholipids (OxPLs) cross-linked to BSA or to a hexapeptide. The OxPL adducts were then tested for their ability to bind to CD36-transfected cells and for their ability to inhibit OxLDL binding to CD36. Both POVPC-BSA and POVPC-peptide adducts were high-affinity ligands for CD36 and potent inhibitors of OxLDL binding. Enzymatic removal of the entire PC moiety of the POVPC-peptide, or of the choline headgroup alone, as well as substitution of the choline headgroup by ethanolamine abrogated the inhibitory activity of POVPC. Interestingly, PC by itself or cross-linked to BSA did not show any intrinsic competition activity. In conclusion, our data demonstrate that the PC headgroup of OxPL alone is sufficient for binding to CD36, but only if presented in the correct conformation as in OxPL of OxLDL or as in POVPC-peptide adducts.     

Molecular aptamer beacons for real-time protein recognition

One of the most pressing problems facing those attempting to understand the regulation of gene expression and translation is the necessity to monitor protein production in a variety of metabolic states. Thus far, there is no easy solution that will either identify or quantitate proteins in real time. Here we introduce a novel protein probe, molecular aptamer beacon (MAB), for real time protein recognition and quantitative analysis. The MAB combines the signal transduction mechanism of molecular beacons and the molecular recognition specificity of aptamers. An MAB based on a thrombin-binding aptamer was prepared as a model to demonstrate the feasibility. Significant fluorescent signal change was observed when MAB was bound to thrombin, which is attributed to a significant conformational change in MAB from a loose random coil to a compact unimolecular quadruplex. The MAB recognizes its target protein with high specificity and high sensitivity (112 picomolar thrombin concentration) in homogeneous solutions. Ratiometric imaging has been conducted with MAB labeled with two fluorophores, which makes it feasible for protein quantitation in living specimen. The unique properties of the MAB will enable the development of a class of protein probes for real time protein tracing in living specimen and for efficient biomedical diagnosis in homogeneous solutions.     

Surface EMG recordings and pain-related disorders: a diagnostic framework

Surface electromyography provides a very valuable set of information when used diagnostically with pain-related disorders. Unfortunately, most researchers and clinicians limit their investigation to only one level of diagnostic information available concerning the neuromuscular system. This article develops and encourages the clinician/researcher to consider three levels of diagnostic information: emotional, organ-related, and postural aspects. The theoretical background and diagnostic procedures used to investigate each is presented. Differential treatment considerations, given differential diagnostic findings, are discussed.     

L-ficolin is a pattern recognition molecule specific for acetyl groups

L-ficolin and H-ficolin are molecules of the innate immune system. Upon recognition of a suitable target they activate the complement system. The ligand recognition structure of ficolins is contained within a fibrinogen-like domain. We examined the selectivity of the ficolins through inhibiting the binding to bacteria or to beads coupled with N-acetylglucosamine. The binding of L-ficolin to Streptococcus pneumoniae 11F and the beads was inhibited by N-acetylated sugars and not by non-acetylated sugars. However, it was also inhibited by other acetylated compounds. Based on this selectivity L-ficolin is not easily defined as a lectin. The binding of H-ficolin to Aerococcus viridans was not inhibited by any of the sugars or other compounds examined. Based on the selectivity of L-ficolin we developed a new purification procedure involving affinity chromatography on N-acetylcysteine-derivatized Sepharose. The column was loaded in the presence of EDTA and high salt, and L-ficolin was eluted by decreasing the salt concentration. Further purification was achieved by ion exchange chromatography.     

Pattern recognition of number gestures based on a wireless surface EMG system

Using surface electromyography (sEMG) signal for efficient recognition of hand gestures has attracted increasing attention during the last decade, with most previous work being focused on recognition of upper arm and gross hand movements and some work on the classification of individual finger movements such as finger typing tasks. However, relatively few investigations can be found in the literature for automatic classification of multiple finger movements such as finger number gestures. This paper focuses on the recognition of number gestures based on a 4-channel wireless sEMG system. We investigate the effects of three popular feature types (i.e. Hudgins’ time–domain features (TD), autocorrelation and cross-correlation coefficients (ACCC) and spectral power magnitudes (SPM)) and four popular classification algorithms (i.e. k-nearest neighbor (k-NN), linear discriminant analysis (LDA), quadratic discriminant analysis (QDA) and support vector machine (SVM)) in offline recognition. Motivated by the good performance of SVM, we further propose combining the three features and employing a new classification method, multiple kernel learning SVM (MKL-SVM). Real sEMG results from six subjects show that all combinations, except k-NN or LDA using ACCC features, can achieve above 91% average recognition accuracy, and the highest accuracy is 97.93% achieved by the proposed MKL-SVM method using the three feature combination (3F). Referring to the offline recognition results, we also implement a real-time recognition system. Our results show that all six subjects can achieve a real-time recognition accuracy higher than 90%. The number gestures are therefore promising for practical applications such as human–computer interaction (HCI).     

Ensemble classifier for protein fold pattern recognition

MOTIVATION: Prediction of protein folding patterns is one level deeper than that of protein structural classes, and hence is much more complicated and difficult. To deal with such a challenging problem, the ensemble classifier was introduced. It was formed by a set of basic classifiers, with each trained in different parameter systems, such as predicted secondary structure, hydrophobicity, van der Waals volume, polarity, polarizability, as well as different dimensions of pseudo-amino acid composition, which were extracted from a training dataset. The operation engine for the constituent individual classifiers was OET-KNN (optimized evidence-theoretic k-nearest neighbors) rule. Their outcomes were combined through a weighted voting to give a final determination for classifying a query protein. The recognition was to find the true fold among the 27 possible patterns. RESULTS: The overall success rate thus obtained was 62% for a testing dataset where most of the proteins have <25% sequence identity with the proteins used in training the classifier. Such a rate is 6-21% higher than the corresponding rates obtained by various existing NN (neural networks) and SVM (support vector machines) approaches, implying that the ensemble classifier is very promising and might become a useful vehicle in protein science, as well as proteomics and bioinformatics. AVAILABILITY: The ensemble classifier, called PFP-Pred, is available as a web-server at http://202.120.37.186/bioinf/fold/PFP-Pred.htm for public usage.     

Routine libraries for pattern recognition in quasispecies

The results obtained through biological research usually need to be analyzed using computational tools, since manual analysis becomes unfeasible due to the complexity and size of these results. For instance, the study of quasispecies frequently demands the analysis of several, very lengthy sequences of nucleotides and amino acids. Therefore, bioinformatics tools for the study of quasispecies are constantly being developed due to different problems found by biologists. In the present study, we address the development of a software tool for the evaluation of population diversity in quasispecies. Special attention is paid to the localization of genome regions prone to changes, as well as of possible hot spots.     

A role for human Sp alpha as a pattern recognition receptor

Human Sp alpha is a soluble protein belonging to group B of the scavenger receptor cysteine-rich (SRCR) superfamily for which little functional information is available. It is expressed by macrophages present in lymphoid tissues (spleen, lymph node, thymus, and bone marrow), and it binds to myelomonocytic and lymphoid cells, which suggests that it may play an important role in the regulation of the innate and adaptive immune systems. In the present study we show that recombinant human Sp alpha (rSp alpha) binds to the surface of several gram-positive and gram-negative bacterial strains. Competition studies indicated that such binding is mediated by the recognition of lipoteichoic acid (LTA) and lipopolysaccharide (LPS), respectively, through nonoverlapping sites on the Sp alpha molecule. The most conserved part of LPS (2-keto-3-deoxyoctulosonic acid and lipid A) was shown to be involved in the recognition by Sp alpha. Bacterial binding studies using the SRCR domain 1 of Sp alpha showed that this domain retains both the LPS and LTA binding activities, indicating that both bacterial interacting sites are retained in a single SRCR domain. Furthermore, rSp alpha induced aggregation of gram-positive and gram-negative bacteria strains. On the other hand, rSp alpha inhibited tumor necrosis factor-alpha secretion by human monocytes stimulated with LPS or LTA. Binding of Sp alpha to conserved components of bacterial surfaces and modulation of the monocyte response indicate that this molecule is an active constituent of the innate immune response of the host.     

LacdiNAc-glycans constitute a parasite pattern for galectin-3-mediated immune recognition

Although Gal beta 1-4GlcNAc (LacNAc) moieties are the most common constituents of N-linked glycans on vertebrate proteins, GalNAc beta 1-4GlcNAc (LacdiNAc, LDN)-containing glycans are widespread in invertebrates, such as helminths. We postulated that LDN might be a molecular pattern for recognition of helminth parasites by the immune system. Using LDN-based affinity chromatography and mass spectrometry, we have identified galectin-3 as the major LDN-binding protein in macrophages. By contrast, LDN binding was not observed with galectin-1. Surface plasmon resonance (SPR) analysis and a solid phase binding assay demonstrated that galectin-3 binds directly to neoglycoconjugates carrying LDN glycans. In addition, galectin-3 bound to Schistosoma mansoni soluble egg Ags and a mAb against the LDN glycan inhibited this binding, suggesting that LDN glycans within S. mansoni soluble egg Ags contribute to galectin-3 binding. Immunocytochemistry demonstrated high levels of galectin-3 in liver granulomas of S. mansoni-infected hamsters, and a colocalization of galectin-3 and LDN glycans was observed on the parasite eggshells. Finally, we demonstrate that galectin-3 can mediate recognition and phagocytosis of LDN-coated particles by macrophages. These findings provide evidence that LDN-glycans constitute a parasite pattern for galectin-3-mediated immune recognition.     

Surface EMG: The issue of electrode location

This paper contributes to clarifying the conditions under which electrode position for surface EMG detection is critical and leads to estimates of EMG variables that are different from those obtained in other nearby locations. Whereas a number of previous works outline the need to avoid the innervation zone (or the muscle belly), many authors place electrodes in the central part or bulge of the muscle of interest where the innervation zone is likely to be. Computer simulations are presented to explain the effect of the innervation zone on amplitude, frequency and conduction velocity estimates from the signal and the need to avoid placing electrodes near it. Experimental signals recorded from some superficial muscles of the limbs and trunk (abductor pollicis brevis, flexor pollicis brevis, biceps, upper trapezius, vastus medialis, vastus lateralis) were processed providing support for the findings obtained from simulations. The use of multichannel techniques is recommended to estimate the location of the innervation zone and to properly choose the optimal position of the detection point(s) allowing meaningful estimates of EMG variables during movement analysis.     

基于定量分析方法的动作表面肌电信号分析

介绍了非线性数据处理方法递归图法(recurrence plots,RP)及其定量分析方法(recurrence quantifi-cation analysis,RQA),并利用RP和RQA研究了动作表面肌电信号。研究发现,表面肌电信号在不同动作模式下其所对应的RP图在结构上差异明显,通过计算两通道肌电信号的RQA指标递归率,发现不同动作信号的RQA指标递归率值具有不同的聚类分布。该方法为肌电信号的动作模式分类提供了一种新的思路。     

Effects of a passive upper extremity exoskeleton for overhead tasks

Overhead work is a major cause of upper extremity work-related musculoskeletal disorders (WMSD). In this paper, the potential effects of a Passive Upper-Limb Exoskeleton (PULE) were evaluated in the tasks of overhead works. This proposed PULE has a higher degree of freedom and does not impede the user's upper limb movements. Fifteen male volunteers participated in the study by performing the repeated overhead bolt installation tasks. The electromyographic (EMG) values of anterior deltoid (AD), mid deltoid (MD), descending trapezius (TR), and triceps (TB) of the left and right arms of the participants were measured at three different overhead task heights with Intervention (with/without the PULE). Moreover, the rankings of perceived discomfort (RPD) obtained on the neck, shoulders, upper arms, forearms, upper back, waists, and legs were rated for each participant. The preliminary experiment results show that the initial nEMG of right anterior deltoid (AD) decreased by 38.5%, median nEMG values decreased by 45.1%, and total RPD decreased by 52.4%. The use of the PULE could bring the benefits of less upper extremity muscle contraction and lower RPD compared to the non-use, which may potentially reduce or slow down the level of upper extremity WMSD across the overhead work.