Claudio Loconsole

A New Gaze-BCI-Driven Control of an Upper Limb Exoskeleton for Rehabilitation in Real-World Tasks

This paper proposes a new multimodal architecture for gaze-independent brain-computer interface (BCI)-driven control of a robotic upper limb exoskeleton for stroke rehabilitation to provide active assistance in the execution of reaching tasks in a real setting scenario. At the level of action plan, the patients intention is decoded by means of an active vision system, through the combination of a Kinect-based vision system, which can online robustly identify and track 3-D objects, and an eye-tracking system for objects selection. At the level of action generation, a BCI is used to control the patients intention to move his/her own arm, on the basis of brain activity analyzed during motor imagery. The main kinematic parameters of the reaching movement (i.e., speed, acceleration, and jerk) assisted by the robot are modulated by the output of the BCI classifier so that the robot-assisted movement is performed under a continuous control of patients brain activity. The system was experimentally evaluated in a group of three healthy volunteers and four chronic stroke patients. Experimental results show that all subjects were able to operate the exoskeleton movement by BCI with a classification error rate of 89.4±5.0% in the robot-assisted condition, with no difference of the performance observed in stroke patients compared with healthy subjects. This indicates the high potential of the proposed gaze-BCI-driven robotic assistance for neurorehabilitation of patients with motor impairments after stroke since the earliest phase of recovery.

An EMG-Controlled Robotic Hand Exoskeleton for Bilateral Rehabilitation

This paper presents a novel electromyography (EMG)-driven hand exoskeleton for bilateral rehabilitation of grasping in stroke. The developed hand exoskeleton was designed with two distinctive features: (a) kinematics with intrinsic adaptability to patients hand size, and (b) free-palm and free-fingertip design, preserving the residual sensory perceptual capability of touch during assistance in grasping of real objects. In the envisaged bilateral training strategy, the patients non paretic hand acted as guidance for the paretic hand in grasping tasks. Grasping force exerted by the non paretic hand was estimated in real-time from EMG signals, and then replicated as robotic assistance for the paretic hand by means of the hand-exoskeleton. Estimation of the grasping force through EMG allowed to perform rehabilitation exercises with any, non sensorized, graspable objects. This paper presents the system design, development, and experimental evaluation. Experiments were performed within a group of six healthy subjects and two chronic stroke patients, executing robotic-assisted grasping tasks. Results related to performance in estimation and modulation of the robotic assistance, and to the outcomes of the pilot rehabilitation sessions with stroke patients, positively support validity of the proposed approach for application in stroke rehabilitation.

Fall detection in indoor environment with kinect sensor

Falls are one of the major risks of injury for elderly living alone at home. Computer vision-based systems offer a new, low-cost and promising solution for fall detection. This paper presents a new fall-detection tool, based on a commercial RGB-D camera. The proposed system is capable of accurately detecting several types of falls, performing a real time algorithm in order to determine whether a fall has occurred. The proposed approach is based on evaluating the contraction and the expansion speed of the width, height and depth of the 3D human bounding box, as well as its position in the space. Our solution requires no pre-knowledge of the scene (i.e. the recognition of the floor in the virtual environment) with the only constraint about the knowledge of the RGB-D camera position in the room. Moreover, the proposed approach is able to avoid false positive as: sitting, lying down, retrieve something from the floor. Experimental results qualitatively and quantitatively show the quality of the proposed approach in terms of both robustness and background and speed independence.

Motion detection technology as a tool for cardiopulmonary resuscitation (CPR) quality training: A randomised crossover mannequin pilot study

INTRODUCTION Outcome after cardiac arrest is dependent on the quality of chest compressions (CC). A great number of devices have been developed to provide guidance during CPR. The present study evaluates a new CPR feedback system (Mini-VREM: Mini-Virtual Reality Enhanced Mannequin) designed to improve CC during training. METHODS Mini-VREM system consists of a Kinect(®) (Microsoft, Redmond, WA, USA) motion sensing device and specifically developed software to provide audio-visual feedback. Mini-VREM was connected to a commercially available mannequin (Laerdal Medical, Stavanger, Norway). Eighty trainees (healthcare professionals and lay people) volunteered in this randomised crossover pilot study. All subjects performed a 2 min CC trial, 1h pause and a second 2 min CC trial. The first group (FB/NFB, n=40) performed CC with Mini-VREM feedback (FB) followed by CC without feedback (NFB). The second group (NFB/FB, n=40) performed vice versa. Primary endpoints: adequate compression (compression rate between 100 and 120 min(-1) and compression depth between 50 and 60mm); compressions rate within 100-120 min(-1); compressions depth within 50-60mm. RESULTS When compared to the performance without feedback, with Mini-VREM feedback compressions were more adequate (FB 35.78% vs. NFB 7.27%, p<0.001) and more compressions achieved target rate (FB 72.04% vs. 31.42%, p<0.001) and target depth (FB 47.34% vs. 24.87%, p=0.002). The participants perceived the system to be easy to use with effective feedback. CONCLUSIONS The Mini-VREM system was able to improve significantly the CC performance by healthcare professionals and by lay people in a simulated CA scenario, in terms of compression rate and depth.

A new bounded jerk on-line trajectory planning for mimicking human movements in robot-aided neurorehabilitation

In this paper we propose a new on-line control strategy that can generate motion primitives mimicking human movement for robot assistance in stroke neurorehabilitation. The proposed strategy, with respect to other methods, allows the generation of bounded jerk trajectories characterized by inter-joint synchronization, e.g. joint variables complete the same percentage of their trajectories at each instant of time. The algorithm can on-line automatically identify, localize and track target objects to be reached, and adapt the level of assistance to be provided to the patient, so that the robot assistance is provided to let the patient operate in a real world setting, where he/she can reach and grasp common everyday life objects, To evaluate the performance of the proposed algorithm, its implementation was derived to control the movement of an upper limb robotic exoskeleton, the L-Exos, and an experimental evaluation was conducted in a group of healthy subjects to assess the plausibility of generated trajectories in terms of similarity with human motion.

An EMG-based approach for on-line predicted torque control in robotic-assisted rehabilitation

This paper proposes a sEMG-based method for on-line torque prediction and control of robot joints. More in detail, the Mean Absolute Value (MAV) features extracted from the sEMG signals acquired from five muscles of the shoulder and of the elbow are used as input to two trained time delayed neural networks (TDNNs) to estimate the joint torque of an active exoskeleton robot for movements executed in the sagittal plane. The sEMG-driven TDNNs, trained with a dataset composed by shoulder and elbow joint torque values registered in isometric conditions, allow to on-line control the exoskeleton joints for slow movements of the upper limbs. Finally, the method was tested and validated through experiments conducted on a healthy subject.

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.

Relive: a serious game to learn how to save lives.

A recent review has provided evidence in support of new nd alternative methods for CPR training.1 Among these, are he “serious games”, which are applications developed using omputer game technologies more often associated with enterainment, but characterized by a serious purpose. Indeed, during he last decade, many serious games have been developed and sed successfully in the field of health, including training of oth technical and non-technical skills relevant to the surgical rea.2 The Italian Resuscitation Council (IRC) has implemented a erious game for the Viva! Campaign 20133 called Viva! Game http://www.viva2013.it/viva-game). Viva! Game is a serious game irected to kids and young adults. It served as a tool to create wareness on cardiac arrest and cardiopulmonary resuscitation CPR) in a soft and enjoyable way. The game has different scenaros, i.e. school, home, stadium, through which the player needs to nteract. More specifically, during the development of the story, he player finds himself in the need to perform a high quality hest compression to save another character from cardiac arrest. iva! CPR (http://www.viva2013.it/vivacpr) is an application for eal time feedback on chest compression quality created for smarthones directed to general population to increase awareness and nowledge about chest compression only manoeuvres. The numer of downloads of Viva! Game and Viva! CPR during the Viva! ampaign 2013 was around 10,000 (Table 1). For the Viva! Camaign 2014, the Italian Resuscitation Council developed a new and ore ambitious project called “Relive” game. Relive is a serious ame focusing on CPR with the main purpose of increasing kids nd young adults’ awareness on CPR and prompting them to attend PR classes and be prepared to intervene in case of cardiac arrest. elive is a first person 3D adventure taking place on planet Mars, n a near future. The game is divided into two different playing odes: a tournament mode and a story mode. The tournament ode is a ready-to-play simulated emergency scene, taken from elected game scenes, where the player faces different rescue sit-

An emg-based robotic hand exoskeleton for bilateral training of grasp

This work presents the development and the preliminary experimental assessment of a novel EMG-driven robotic hand exoskeleton for bilateral active training of grasp motion in stroke. The system allows to control the grasping force required to lift a real object with an impaired hand, through the active guidance provided by a hand active exoskeleton, whose force is modulated by the EMG readings acquired on the opposite unimpaired arm. To estimate the grasping force, the system makes use of surface EMG recordings during grasping, developed on the opposite unimpaired arm, and of a neural network to classify the information. The design, integration and experimental characterization of the system during the grasp of two cylindrical objects is presented. The experimental results show that an optimal force tracking of the interaction force with the object can be achieved.

Hand and arm ownership illusion through virtual reality physical interaction and vibrotactile stimulations

Body awareness has important implications for the use of virtual reality (VR) and its effectiveness. This involves the senses of agency and body ownership, studied in the past by producing the Rubber Hand Illusion (RHI). Recent studies reported the RHI on virtual environments (VE) by giving the participant synchronous 3D visual stimulation and passive tactile stimulation manually on the hidden real hand placed on a static position. In this paper we present a novel study of the RHI within highly dynamic VE sessions with synchronous pure virtual vibrotactile stimulation of the fingers. The hand/arm participants movements are realistically reproduced on the VE and tactile stimulations are self-inflicted by the participant through actively touching the virtual objects. The results revealed that the RHI is possible in active, dynamic and fully multisensored VE sessions.