Search

Neuroprostheses can be an innovative solution to improve quality of life of upper limb amputees. In this framework, the recovery of sensory feedback is a property widely requested by amputee subjects. Neural prostheses are based on neural interfaces that allow delivering direct current stimuli to the nerve fibers. The study of the interaction between the nerve and the electrode is fundamental to investigate activation properties in the nerve. Furthermore, the results could provide useful insight into improve the design of the electrodes and to advance and ameliorate tactile sensations, elicited by these interfaces, obtaining tactile feedback more like natural sensations. This work aims at studying, by means of a FEM Neuron computational model, the axon fibers activation by means of neural stimulation provided through the intraneural electrodes DS-file. Three different types of stimulation waveforms (i.e. biphasic charge balanced stimulus with inter-pulse delay, biphasic charge balanced stimulus without inter-pulse delay, biphasic charge unbalanced stimulus with inter-pulse delay), three different nerve fascicles, i.e. two sensory and one motor fascicle, and ten distances from the electrode in the fascicles, are considered. The efficacy of the stimulation expressed as the percentage of activation of the fibers, and the safety, in terms of current intensity and used waveform, are studied in the previously described different conditions and the results are compared. The obtained results show that: i. stimulating a sensory fascicle with implanted active sites can activate a fascicle close to it, but not all the fascicles belonging to the same nerve. In fact, in the nerve considered in this study, a motor fascicle cannot be activated due to the values of the electrical potential which are too low to activate the fibers; ii. the current intensity necessary to activate fibers increases according to the distance from the source of the stimulus; iii. by using a biphasic charge unbalanced stimulus, the threshold to activate the fibers is lower than using the other tested waveforms. It is an important result because the stimulation is efficient and safer since current intensity is lower than the one used for the other two waveforms.

A Reinforcement Learning Model of Reaching Integrating Kinematic and Dynamic Control in a Simulated Arm Robot.

A case of reconstructive surgery for saddle nose deformity is presented in this paper. A 22 year-old Bengali female who had completed WHO/MB regimen for 27 months underwent reconstructive surgery for saddle nose deformity. Since a saddle nose is one of the symbols of leprosy, it often causes serious psychological and social troubles to patients. This happens more often when the patient is a young unmarried woman. In this case the saddle nose seemed to be very serious. After being discharged from hospital she got married and had a baby. This operation gave the patient great relief to live in the community, because she no longer had serious visible evidence of leprosy on the body. Though the correction of the deformity without any dysfunction does not always have priority over other surgeries at a busy leprosy control project in Bangladesh, it has merit both in patients themselves and in the society around them because it leads to the elimination of the stigma of leprosy.

The early case detection followed by WHO/MDT is the most important in leprosy control program not only for the destruction of linkage of infection but also the prevention of disability in leprosy. PIME Sisters leprosy program has been carrying out active case detection for leprosy since 1986 in Khulna, the third biggest metropolitan city in Bangladesh. The relation between several detection modes and their associated disability grading was analyzed in order to see the contribution of each detection mode to disability grading. The disability grading of new registered patients from 1995 to 1997, 1,115 cases in total was analyzed according to its detection mode. Voluntary reporting patients from "out of control area" had both high disability grading of G = 2 (38.0%) and high MB rate (39.3%). On the other hand voluntary reporting patients in control area had less disability grading of G = 2 (12.2%) and less MB rate (25.3%). The number of patients referred by local doctors was still small and had both high disability grading of G = 2 (27.0%) and high MB rate (51.4%). Children detected under school surveillance did not have any marked disability of G = 2 (0%) and were almost PB patients (88.6%). (Prevalence rate of school surveillance was low.) Patients detected under general surveillance had low disability grading of G = 2 (2.6%). The disability grading (G = 2) of voluntary patients from control area (12.2%) was three times as much as that of patients from general surveillance (2.6%). Lastly patients of family contacts who were aware of the first symptoms had relatively less disability grading (5.9%). The small number of patients referred by local doctors (3.3% of total number) with high disability grading meant that there would be still great need of information programs on leprosy for local medical professions to think about leprosy. The high disability grading of patients from "out of control area" meant that the more effective leprosy control program should be done in "out of control area".

The paper presents a biologically-inspired perception-action scheme for robots interacting with real-world environments, grounded in the broad hypothesis that, like in humans, perception and action are improved in speed and accuracy by expectation mechanisms: perception crucially involves comparison processes between incoming stimuli and expected perceptions (EPs). In the proposed approach, EPs are built on the basis of previous perceptions, current motor commands, proprioception, and internal models of the robot and the environment. In contrast to pure reactive approaches, background knowledge plays here a helpful role, as it reduces the computational burden of perception and motor coordination tasks, in partially structured environments. In the work presented here, an EP mechanism has been applied in the visuo-motor coordination of an anthropomorphic 8 d.o.f. robotic manipulator equipped with a vision system, in order to validate the viability and effectiveness of the initial hypothesis.

The use of robotic technology for assessment has the potential to provide therapists with objective, accurate, repeatable measurements of subject's functions. However, despite the increasing number of clinical studies examining the effect of robotic training on stroke rehabilitation, body functions and structures assessment is typically carried out through traditional human-administered clinical impairment scales.The paper aims at providing a complete set of kinematic and dynamic indices for an objective measure of the effect of robot-aided therapy, and testing their correlation with clinical scales.An intervention pilot study applying robotic therapy was carried out.The clinical study was focused on outpatients and was carried out at Università Campus Bio-Medico of Rome, Italy.Fifteen community-dwelling persons with chronic stroke met inclusion criteria and volunteered to participate.Upper limb robotic therapy was administered to patients. Kinematic and dynamic performance indices were extracted from position and force data recorded with the InMotion2 robot. A linear regression analysis was carried out to study correlation with clinical scales to extract a core set of performance indicators.Robotic outcome measures showed a significant improvement of kinematic motor performance; the improvement of dynamic components was significant only in resistive motion and highly correlated with Motor Power.Preliminary results showed that arm motor functions and strength of the paretic arm can be objectively measured by means of the proposed bunch of robotic measures. Correlation with Motor Power was high, while correlation with Fugl-Meyer was moderate.An improvement of clinical body functions assessment is expected in terms of objective, accurate and repeatable measurements of subject's performance during recovery.

Models proposed within the literature of motor control have polarised around two classes of controllers which differ in terms of controlled variables: the Force-Control Models(FCMs), based on dynamic control, and the Equilibrium-Point Models (EPMs), based on kinematic control. This paper proposes a bioinspired model which aims to exploit the strengths of the two classes of models. The model is tested with a 3D physical simulator of a 2DOF-controlled arm robot engaged in a reaching task which requires the production of curved trajectories to be solved. The model is based on an actor-critic reinforcementlearning algorithm which uses neural maps to represent both percepts and actions encoded as joint-angle desired equilibrium points (EPs), and a noise generator suitable for fine tuning the exploration/exploitation ratio. The tests of the model show how it is capable of exploiting the simplicity and speed of learning of EPMs as well as the flexibility of FCMs in generating curved trajectories. Overall, the model represents a first step towards the generation of models which exploit the strengths of both EPMs and FCMs and has the potential of being used as a new tool for investigating phenomena related to the organisation and learning of motor behaviour in organisms.

This short paper presents the core ideas of the IM-CLeVeR Project. IM-CLeVeR aims at developing a new methodology for designing robot controllers that can: (a) cumulatively learn new skills through autonomous development based on intrinsic motivations, and (b) reuse such skills for accomplishing multiple, complex, and externally-assigned tasks. This goal will be pursued by investigating three fundamental issues: (a) the mechanisms of abstraction of sensorimotor information; (b) the mechanisms underlying intrinsic motivations; (c) hierarchical architectures that permit cumulative learning. The study of these issues will be conducted on the basis of empirical experiments run with monkeys, children, and human adults, with bio-mimetic models aimed at reproducing and interpreting the results of such experiments, and through the design of innovative machine learning systems. The models, architectures, and algorithms so developed will be validated with experiments and demonstrators run with the simulated and real iCub humanoid robot.