Jun-Uk Chu

A Real-Time EMG Pattern Recognition System Based on Linear-Nonlinear Feature Projection for a Multifunction Myoelectric Hand

This paper proposes a novel real-time electromyogram (EMG) pattern recognition for the control of a multifunction myoelectric hand from four channel EMG signals. To extract a feature vector from the EMG signal, we use a wavelet packet transform that is a generalized version of wavelet transform. For dimensionality reduction and nonlinear mapping of the features, we also propose a linear-nonlinear feature projection composed of principal components analysis (PCA) and a self-organizing feature map (SOFM). The dimensionality reduction by PCA simplifies the structure of the classifier and reduces processing time for the pattern recognition. The nonlinear mapping by SOFM transforms the PCA-reduced features into a new feature space with high class separability. Finally, a multilayer perceptron (MLP) is used as the classifier. Using an analysis of class separability by feature projections, we show that the recognition accuracy depends more on the class separability of the projected features than on the MLPs class separation ability. Consequently, the proposed linear-nonlinear projection method improves class separability and recognition accuracy. We implement a real-time control system for a multifunction virtual hand. Our experimental results show that all processes, including virtual hand control, are completed within 125 ms, and the proposed method is applicable to real-time myoelectric hand control without an operational time delay

Synthesis and physicochemical characterization of reduction-sensitive block copolymer for intracellular delivery of doxorubicin

AbstractAn amphiphilic diblock copolymer bearing the reduction-sensitive linker, composed of poly(ethylene glycol) (PEG) and hydrophobic poly(γ-benzyl L-glutamate) (PBLG), was prepared as the potential carrier of doxorubicin (DOX) via a facile synthetic method in the presence of a shell-sheddable PEG macroinitiator (PEG-SS-NH2). Owing to its amphiphilic nature, the copolymer (PEG-SS-PBLG) formed spherical micelles (137 nm in diameter) in aqueous conditions. The micelles were stable under the physiologic condition (pH 7.4) and were readily cleaved in the presence of glutathione (GSH), a tripeptide reducing the disulfide bond in the cytoplasm of the cell. DOX, chosen as a model anticancer drug, was effectively encapsulated into the hydrophobic core of the micelle with high loading efficiency (>75%). The micelle released DOX completely within 18 h at 10 mM GSH mimicking the intracellular condition, whereas only 34% of the drug was released from the micelle at 2 μM GSH. In vitro cytotoxicity tests revealed that DOX-loaded reduction-sensitive micelles are more toxic to SCC7 cells than reduction-insensitive control micelles. These results suggest that PEG-SS-PBLG is the promising carrier for the intracellular delivery of DOX.

A Supervised Feature Projection for Real-Time Multifunction Myoelectric Hand Control

EMG pattern recognition is essential for the control of a multifunction myoelectric hand. The main goal of this study is to develop an efficient feature projection method for EMG pattern recognition. To this end, we propose a linear supervised feature projection that utilizes linear discriminant analysis (LDA). We first perform wavelet packet transform (WPT) to extract the feature vector from four channel EMG signals. For dimensionality reduction and clustering of the WPT features, the LDA incorporates class information into the learning procedure and finds a linear matrix to maximize the class separability for the projected features. Finally, the multilayer perceptron (MLP) classifies the LDA-reduced features into nine hand motions. To evaluate the performance of LDA for the WPT features, we compare LDA with three other feature projection methods. From a visualization and quantitative comparison, we show that LDA has better performance for the class separability, and the LDA-projected features improve the classification accuracy with a short processing time. We implemented a real-time control system for a multifunction myoelectric hand. In experiment, we show that the proposed method achieves 97.2% recognition accuracy, and that all processes, including the myoelectric hand control, are completed within 97 msec

An Implantable Wireless Neural Interface System for Simultaneous Recording and Stimulation of Peripheral Nerve with a Single Cuff Electrode

Recently, implantable devices have become widely used in neural prostheses because they eliminate endemic drawbacks of conventional percutaneous neural interface systems. However, there are still several issues to be considered: low-efficiency wireless power transmission; wireless data communication over restricted operating distance with high power consumption; and limited functionality, working either as a neural signal recorder or as a stimulator. To overcome these issues, we suggest a novel implantable wireless neural interface system for simultaneous neural signal recording and stimulation using a single cuff electrode. By using widely available commercial off-the-shelf (COTS) components, an easily reconfigurable implantable wireless neural interface system was implemented into one compact module. The implantable device includes a wireless power consortium (WPC)-compliant power transmission circuit, a medical implant communication service (MICS)-band-based radio link and a cuff-electrode path controller for simultaneous neural signal recording and stimulation. During in vivo experiments with rabbit models, the implantable device successfully recorded and stimulated the tibial and peroneal nerves while communicating with the external device. The proposed system can be modified for various implantable medical devices, especially such as closed-loop control based implantable neural prostheses requiring neural signal recording and stimulation at the same time.

Measurement of MMP activity in synovial fluid in cases of osteoarthritis and acute inflammatory conditions of the knee joints using a fluorogenic peptide probe-immobilized diagnostic kit

Purpose: A fluorogenic peptide probe-immobilized diagnostic kit was used to analyze MMP activity in the synovial fluids (SFs) from patients with osteoarthritis (OA) and acute inflammatory conditions of the knee joint. Methods: The MMP diagnostic kit containing a polymer-conjugated MMP probe immobilized on a 96-well plate was utilized for high-throughput screening of MMP activity in SFs from OA patients (n = 33) and patients with acute inflammatory conditions of the knee joint (n = 5). Results: Compared to SF from OA patients, SF from patients with acute inflammatory conditions of the knee joint presented stronger NIR fluorescent signals. In gelatin zymography, most samples from patients with acute inflammatory conditions of the knee joint also displayed 92 kDa (pro-form) MMP-9 and faint 84 kDa (active form) MMP-9, while SF from OA patients did not display detectable MMP-9 activity . Conclusion: The presence of a strong fluorescence signal from the MMP diagnostic kit corresponded well with patients with acute inflammatory conditions of the knee joint. The results suggest that our MMP diagnostic kit can be useful in differentiation between early stages of OA and acute inflammatory conditions of the knee joint.

Spontaneous synchronized burst firing of subthalamic nucleus neurons in rat brain slices measured on multi-electrode arrays.

The current study presents an organotypic rat midbrain slice culture that served as a consistent and informative framework, where the STN neurons and their interconnectivity were closely examined with respect to electrophysiological and pharmacological properties. From multi-electrode array recordings, it was found that the majority of STN neurons spontaneously fired in bursts rather than tonically under control conditions, and the neural activity between pairs of burst-firing STN neurons was tightly correlated. This spontaneous synchronized burst firing was also affected by a glutamate receptor antagonist, yet unaffected by a GABA receptor antagonist. Moreover, even when the STN was isolated from all its known external inputs, spontaneous synchronized burst firing was still observed under control conditions and consistently switched to tonic firing following the application of a glutamate receptor antagonist. Therefore, the results indicated the existence of glutamatergic projections to the STN in the slice preparation, and these excitatory synaptic connections appeared to originate from axon collaterals within the STN rather than other basal ganglia nuclei. It could be concluded that the STN neurons and their interconnectivity are essential requirements in the rat brain slice preparation to produce spontaneous synchronized burst firing.

Feedback control of electrode offset voltage during functional electrical stimulation

Control of the electrode offset voltage is an important issue related to the processes of functional electrical stimulation because excess charge accumulation over time damages both the tissue and the electrodes. This paper proposes a new feedback control scheme to regulate the electrode offset voltage to a predetermined reference value. The electrode offset voltage was continuously monitored using a sample-and-hold (S/H) circuit during stimulation and non-stimulation periods. The stimulation current was subsequently adjusted using a proportional-integral (PI) controller to minimise the error between the reference value and the electrode offset voltage. During the stimulation period, the electrode offset voltage was maintained through the S/H circuit, and the PI controller did not affect the amplitude of the stimulation current. In contrast, during the non-stimulation period, the electrode offset voltage was sampled through the S/H circuit and rapidly regulated through the PI controller. The experimental results obtained using a nerve cuff electrode showed that the electrode offset voltage was successfully controlled in terms of the performance specifications, such as the steady- and transient-state responses and the constraint of the controller output. Therefore, the proposed control scheme can potentially be used in various nerve stimulation devices and applications requiring control of the electrode offset voltage.

Development of Myoelectric Hand with Infrared LED-based Tactile Sensor

This paper proposes an IR (infrared) LED (Light Emitting Diode)-based tactile fingertip sensor that can independently measure the normal and tangential force between the hand and an object. The proposed IR LED-based tactile sensor has several advantages over other technologies, including a low price, small size, and good sensitivity. The design of the first prototype is described and some experiments are conducted to show output characteristics of the proposed sensor. Furthemore, the effectiveness of the proposed sensor is demonstrated through anti-slip control in a multifunction myoelectric hand, called the KNU Hand, which includes several novel mechanisms for improved grasping capabilities. The experimental results show that slippage was avoided by simple force control using feedback on the normal and tangential force from the proposed sensor. Thus, grasping force control was achieved without any slippage or damage to the object.

Prediction of Stress Distribution in the Ceramic Femoral Head after Total Hip Replacement

Ceramic femoral heads are now widely used in Total Hip Replacement (THR). Due to their high biocompatibility and low ductility, ceramic femoral heads are considered to be suitable for young and active patients. However, as in testing the mechanical stability of the femoral head, the conventional proof test (standard ISO 7206-10) has its limit of showing axisymmetric stress distribution on the contact surface, while non-uniformed stress distribution is expected after THR. Since non-uniformed stress distribution can result in the increased probability of ceramic femoral head fracture, it is considerable to evaluate the stress distribution in vivo-like conditions. Therefore, this study simulated the ceramic femoral heads under in vivo-like conditions using finite element method. The maximum stress decreased when increasing the size of the femoral head and stress distribution was concentrated on superior contact surface of the taper region.

A Study on Feature Projection Methods for a Real-Time EMG Pattern Recognition

EMG pattern recognition is essential for the control of a multifunction myoelectric hand. The main goal of this study is to develop an efficient feature projection method for EMC pattern recognition. To this end, we propose a linear supervised feature projection that utilizes linear discriminant analysis (LDA). We first perform wavelet packet transform (WPT) to extract the feature vector from four channel EMC signals. For dimensionality reduction and clustering of the WPT features, the LDA incorporates class information into the learning procedure, and finds a linear matrix to maximize the class separability for the projected features. Finally, the multilayer perceptron classifies the LDA-reduced features into nine hand motions. To evaluate the performance of LDA for the WPT features, we compare LDA with three other feature projection methods. From a visualization and quantitative comparison, we show that LDA has better performance for the class separability, and the LDA-projected features improve the classification accuracy with a short processing time. We implemented a real-time pattern recognition system for a multifunction myoelectric hand. In experiment, we show that the proposed method achieves 97.2% recognition accuracy, and that all processes, including the generation of control commands for myoelectric hand, are completed within 97 msec. These results confirm that our method is applicable to real-time EMG pattern recognition far myoelectric hand control.