Domenico Buongiorno

A novel BCI-SSVEP based approach for control of walking in Virtual Environment using a Convolutional Neural Network

A non-invasive Brain Computer Interface (BCI) based on a Convolutional Neural Network (CNN) is presented as a novel approach for navigation in Virtual Environment (VE). The developed navigation control interface relies on Steady State Visually Evoked Potentials (SSVEP), whose features are discriminated in real time in the electroencephalographic (EEG) data by means of the CNN. The proposed approach has been evaluated through navigation by walking in an immersive and plausible virtual environment (VE), thus enhancing the involvement of the participant and his perception of the VE. Results show that the BCI based on a CNN can be profitably applied for decoding SSVEP features in navigation scenarios, where a reduced number of commands needs to be reliably and rapidly selected. The participant was able to accomplish a waypoint walking task within the VE, by controlling navigation through of the only brain activity.

A neuromusculoskeletal model of the human upper limb for a myoelectric exoskeleton control using a reduced number of muscles

This paper presents a myoelectric control of an arm exoskeleton designed for rehabilitation. A four-muscles-based NeuroMusculoSkeletal (NMS) model was implemented and optimized using genetic algorithms to adapt the model to different subjects. The NMS model is able to predict the shoulder and elbow torques which are used by the control algorithm to ensure a minimal force of interaction. The accuracy of the method is assessed through validation experiments conducted with two healthy subjects performing free movements along the pseudo-sagittal plane. The experiments show promising results for our approach showing its potential for being introduced in a rehabilitation protocol.

A supervised CAD to support telemedicine in hematology

This paper presents the design and the implementation of a Computer Aided Diagnosis (CAD) system for the clinical analysis of Peripheral Blood Smears (PBS also called Blood Film). The proposed system is able to count and classify the five types of leucocytes located in the tail of a PBS for computing the leukocyte formula. Image processing and segmentation techniques were used to extract 33 leucocytes features (morphological, chromatic and texture-based). Only 7 features, selected by using the Information Gain Ranking algorithm of Weka platform, were used to evaluate the classification performance of two different classifiers: Back Propagation Neural Network (BPNN) and Decision Tree (DT). From the comparison between the two proposed approaches we can argue that the BPNN performed better than the DT on the validation set. Finally, the Neural Network classifier was evaluated with a test set composed of 1274 leucocytes obtaining good results in terms of Precision (87.9%) and Sensitivity (97.4%).

Evaluation of a Pose-Shared Synergy-Based Isometric Model for Hand Force Estimation: Towards Myocontrol

In this work the authors investigated whether the muscle synergies concept could improve the isometric hand force estimation. Electromyographic (EMG) activity from 9 arm muscles and hand forces applied at the Light-Exos Exoskeleton end-effector were recorded during isometric contractions in several workspace points lying on the parasagittal plane crossing the shoulder joint. The muscle synergies were extracted in two different ways according to the statements that the muscle primitives are ‘Arm Pose Related’ or ‘Arm Pose Shared’. From the pre-processed EMG signals the authors then estimated the hand forces using three methods. The results showed that the muscle synergy concept improves the isometric force estimation paving the way for a synergy-based myoelectric control.

A linear optimization procedure for an EMG-driven neuromusculoskeletal model parameters adjusting: Validation through a myoelectric exoskeleton control

This paper presents a linear optimization procedure able to adapt a simplified EMG-driven NeuroMusculoSkeletal (NMS) model to the specific subject. The optimization procedure could be used to adjust a NMS model of a generic human articulation in order to predict the joint torque by using ElectroMyoGraphic (EMG) signals. The proposed approach was tested by modeling the human elbow joint with only two muscles. Using the cross-validation method, the adjusted elbow model has been validated in terms of both torque estimation performance and predictive ability. The experiments, conducted with healthy people, have shown both good performance and high robustness. Finally, the model was used to control directly and continuously a exoskeleton rehabilitation device through EMG signals. Data acquired during free movements prove the model ability to detect the human’s intention of movement.

Advanced classification of Alzheimer's disease and healthy subjects based on EEG markers

In this study, we compared several classifiers for the supervised distinction between normal elderly and Alzheimers disease individuals, based on resting state electroencephalographic markers, age, gender and education. Three main preliminary procedures served to perform features dimensionality reduction were used and discussed: a Support Vector Machines Recursive Features Elimination, a Principal Component Analysis and a novel method based on the correlation. In particular, five different classifiers were compared: two different configurations of SVM and three different optimal topologies of Error Back Propagation Multi Layer Perceptron Artificial Neural Networks (EBP MLP ANNs). Best result, in terms of classification (accuracy 86% and sensitivity 92%), was obtained by a Neural Network with 3 hidden layers that used as input: age, gender, education and 20 EEG features selected by the novel method based on the correlation.

Gait Analysis and Parkinson’s Disease: Recent Trends on Main Applications in Healthcare

There is an increasing interest in the use of Gait Analysis (GA) in Parkinson’s Disease (PD), however no one has previously investigated what are the principal trends on the main applications of quantitative GA in studies involving this neurological disorder. We performed a systematically literature search for articles published through 2013 to present using three electronic databases. We retrieved 76 articles that met the inclusion criteria and identified four main research areas which refers to GA for: pathophysiological mechanisms underlying PD; assessment tool for treatment outcomes; automatic recognition of PD symptoms and algorithms for classification between PD patients and healthy subjects.

Real-Time 3D Tracker in Robot-Based Neurorehabilitation

Abstract The chapter describes a computer vision-based robot-assisted system used in neurorehabilitation of post-stroke patients that allows the subjects to reach for and grasp objects in a defined workspace. The proposed computer vision technique is used to model objects that have not been preprocessed in a real setting, track them in real time, and provide their actual pose to the robotic device in order to accomplish grasping tasks. The robotic device is composed of three integrated modules: (i) a 4-DOF arm exoskeleton that supports the patients impaired arm when reaching for the objects; (ii) a 3-DOF actuated wrist exoskeleton for optimizing the hand pose in the grasping task; and (iii) a 2-DOF (flexion/extension) underactuated hand exoskeleton designed to be automatically adjusted for different grasping tasks based on contact forces. The conducted tests have demonstrated the robustness of the proposed approach, and its performance in the neurorehabilitation scenario through reaching and grasping task experiments.