Emotion Recognition from Spatio-Temporal Representation of EEG Signals via 3D-CNN with Ensemble Learning Techniques.

The recognition of emotions is one of the most challenging issues in human-computer interaction (HCI). EEG signals are widely adopted as a method for recognizing emotions because of their ease of acquisition, mobility, and convenience. Deep neural networks (DNN) have provided excellent results in emotion recognition studies. Most studies, however, use other methods to extract handcrafted features, such as Pearson correlation coefficient (PCC), Principal Component Analysis, Higuchi Fractal Dimension (HFD), etc., even though DNN is capable of generating meaningful features. Furthermore, most earlier studies largely ignored spatial information between the different channels, focusing mainly on time domain and frequency domain representations. This study utilizes a pre-trained 3D-CNN MobileNet model with transfer learning on the spatio-temporal representation of EEG signals to extract features for emotion recognition. In addition to fully connected layers, hybrid models were explored using other decision layers such as multilayer perceptron (MLP), k-nearest neighbor (KNN), extreme learning machine (ELM), XGBoost (XGB), random forest (RF), and support vector machine (SVM). Additionally, this study investigates the effects of post-processing or filtering output labels. Extensive experiments were conducted on the SJTU Emotion EEG Dataset (SEED) (three classes) and SEED-IV (four classes) datasets, and the results obtained were comparable to the state-of-the-art. Based on the conventional 3D-CNN with ELM classifier, SEED and SEED-IV datasets showed a maximum accuracy of 89.18% and 81.60%, respectively. Post-filtering improved the emotional classification performance in the hybrid 3D-CNN with ELM model for SEED and SEED-IV datasets to 90.85% and 83.71%, respectively. Accordingly, spatial-temporal features extracted from the EEG, along with ensemble classifiers, were found to be the most effective in recognizing emotions compared to state-of-the-art methods.

A self-aligning end-effector robot for individual joint training of the human arm.

AIM: Intense training of arm movements using robotic devices can help reduce impairments in stroke. Recent evidence indicates that independent training of individual joints of the arm with robots can be as effective as coordinated multi-joint arm training. This makes a case for designing and developing robots made for training individual joints, which can be simpler and more compact than the ones for coordinate multi-joint arm training. The design of such a robot is the aim of the work presented in this paper. METHODS: An end-effector robot kinematic design was developed and the optimal robot link lengths were estimated using an optimization procedure. A simple algorithm for automatically detecting human limb parameters is proposed and its performance was evaluated through a simulation study. RESULTS: A six-degrees-of-freedom end-effector robot with three actuated degrees-of-freedom and three non-actuated self-aligning degrees-of-freedom for safe assisted training of the individual joints (shoulder or elbow) of the human arm was conceived. The proposed robot has relaxed constraints on the relative positioning of the human limb with respect to the robot. The optimized link lengths chosen for the robot allow it to cover about 80% of the human limb's workspace, and possess good overall manipulability. The simple estimation procedure was demonstrated to estimate human limb parameters with low bias and variance. DISCUSSION: The proposed robot with three actuated and three non-actuated degrees-of-freedom has a compact structure suitable for both the left and right arms without any change to its structure. The proposed automatic estimation procedure allows the robot to safely apply forces and impose movements to the human limb, without the need for any manual measurements. Such compact robots have the highest potential for clinical translation.

Supported nanocomposite membranes: bridging microtechnology with nanotechnology.

Bridging top-down and bottom-up manufacturing approaches is desirable to exploit the advantages of both approaches. A simple method that combines microfabrication technique with layer-by-layer self-assembly for in situ fabrication of supported nanocomposite membrane structures is presented. To our knowledge, our approach has yielded the largest nanocomposite membrane size. The micro/nanostructures presented in this work can find broad applicability for sensors, MEMS/bioMEMS, cell biology, microfluidics, and lab-on-chip devices.

Microfluidic device for multimodal characterization of pancreatic islets.

A microfluidic device to perfuse pancreatic islets while simultaneously characterizing their functionality through fluorescence imaging of the mitochondrial membrane potential and intracellular calcium ([Ca(2+)](i)) in addition to enzyme linked immunosorbent assay (ELISA) quantification of secreted insulin was developed and characterized. This multimodal characterization of islet function will facilitate rapid assessment of tissue quality immediately following isolation from donor pancreas and allow more informed transplantation decisions to be made which may improve transplantation outcomes. The microfluidic perfusion chamber allows flow rates of up to 1 mL min(-1), without any noticeable perturbation or shear of islets. This multimodal quantification was done on both mouse and human islets. The ability of this simple microfluidic device to detect subtle variations in islet responses in different functional assays performed in short time-periods demonstrates that the microfluidic perfusion chamber device can be used as a new gold standard to perform comprehensive islet analysis and obtain a more meaningful predictive value for islet functionality prior to transplantation into recipients, which is currently difficult to predict using a single functional assay.

Polymer/colloid surface micromachining: micropatterning of hybrid multilayers.

Fabrication of multicomponent patterned films comprising polymer/nanoparticle multilayers using conventional lithography and bottom-up layer-by-layer nanofabrication techniques is described. The work is motivated by the potential to extend polymer surface micromachining capabilities toward construction of integrated systems by connecting discrete domains of active materials containing functional nanoparticles. Modified surfaces illustrate tunability of the physical (thickness, roughness, 3D structures) and chemical (inorganic/organic material combinations) properties of the nanocomposite micropatterns. Intriguing nanoscale phenomena were observed for the structures when the order of material deposition was changed; the final multilayer thickness and surface roughness and mechanical integrity of the patterns were found to be interdependent and related to the roughness of layers deposited earlier in the process.

Microfluidic add-on for standard electrophysiology chambers.

We have developed a microfluidic brain slice device (microBSD) that marries an off-the shelf brain slice perfusion chamber with an array of microfluidic channels set into the bottom surface of the chamber substrate. As this device is created through rapid prototyping, once optimized, it is trivial to replicate and share the devices with other investigators. The device integrates seamlessly into standard physiology and imaging chambers and it is immediately available to the whole slice physiology community. With this technology we can address the flow of neurochemicals and any other soluble factors to precise locations in the brain slice with the temporal profile we choose. Dopamine (DA) was chosen as a model neurotransmitter and we have quantified delivery in brain tissue using cyclic voltammetry (CV) and fluorescence imaging.

Rapid prototyping for neuroscience and neural engineering.

Rapid prototyping (RP) is a useful method for designing and fabricating a wide variety of devices used for neuroscience research. The present study confirms the utility of using fused deposition modeling, a specific form of RP, to produce three devices commonly used for basic science experimentation. The accuracy and precision of the RP method varies according to the type and quality of the printer as well as the thermoplastic substrate. The printer was capable of creating device channels with a minimum diameter of 0.4 or 0.6mm depending on the orientation of fabrication. RP enabled the computer-aided design and fabrication of three custom devices including a cortical recording/stroke induction platform capable of monitoring electrophysiological function during ischemic challenge. In addition to the recording platform, two perfusion chambers and a cranial window device were replicated with sub-millimeter precision. The ability to repeatedly modify the design of each device with minimal effort and low turn-around time is helpful for oft-unpredictable experimental conditions. Results obtained from validation studies using both the cortical recording platform and perfusion chamber did not vary from previous results using traditional hand-fabricated or commercially available devices. Combined with computer-aided design, rapid prototyping is an excellent alternative for developing and fabricating custom devices for neuroscience research.