VRheab: a fully immersive motor rehabilitation system based on recurrent neural network

In this paper, a fully immersive serious game system that combines two Natural User Interfaces (NUIs) and a Head Mounted Display (HMD) to provide an interactive Virtual Environment (VE) for patient rehabilitation is proposed. Patients’ data are acquired in real-time by the NUIs, while by the HMD the VE is shown to them, thus allowing the interaction. A Long Short-Term Memory Recurrent Neural Network (LSTM-RNN), previously trained by healthy subjects (i.e., baseline), processes patients’ movements in real-time during the rehabilitation exercises to provide the degree of their performance. By comparing the functionalities of the proposed system with the ongoing state-of-the-art, it is worth noting that the reported fully immersive serious game system provides a concrete contribute to the current literature in terms of completeness and versatility. The results obtained by three rehabilitation exercises, chosen as reference case studies, performed on real patients affected by Parkinson’s disease have shown the effectiveness of the presented approach. Finally, the analysis of the feedbacks received by the therapists and patients who have used the system have highlighted remarkable results in terms of motivation, acceptance, and usability.     

Carpal tunnel syndrome: which electrodiagnostic path should we follow? A prospective study of 100 consecutive patients

We prospectively studied 100 consecutive patients (200 hands) sent to our Neurophysiopathology Service for a suspected Carpal Tunnel Syndrome (C.T.S.). We performed 6 different motor and sensitive electrodiagnostic tests on both hands of each patients and on 25 healthy subjects. We calculated the sensibility and the specificity of each test by comparing each parameter with all the other ones, and then, again, using the clinical evaluation as a "gold standard". The routinely performed tests proved to have a high sensibility in detecting C.T.S., nevertheless, sometimes they do not give a definite response. In these cases we suggest to carry out the neurophysiological examination by performing some of the newly proposed tests. We end our study by drawing a "flow-chart", which considers reliability, sensibility, difficulty in performing and/or evaluating, and discomfort for patients, of each test; by this we would give a guide and a practical aid in determining the electrophysiological path for C.T.S. diagnosis.     

An adaptive high-order neural tree for pattern recognition

A new neural tree model, called adaptive high-order neural tree (AHNT), is proposed for classifying large sets of multidimensional patterns. The AHNT is built by recursively dividing the training set into subsets and by assigning each subset to a different child node. Each node is composed of a high-order perceptron (HOP) whose order is automatically tuned taking into account the complexity of the pattern set reaching that node. First-order nodes divide the input space with hyperplanes, while HOPs divide the input space arbitrarily, but at the expense of increased complexity. Experimental results demonstrate that the AHNT generalizes better than trees with homogeneous nodes, produces small trees and avoids the use of complex comparative statistical tests and/or a priori selection of large parameter sets.     

The thermal transformation of Man Made Vitreous Fibers (MMVF) and safe recycling as secondary raw materials (SRM)

This work describes the high temperature reaction sequence of commercial Man Made Vitreous Fibers (MMVF) Cerafiber, Superwool, Rock wool and Glass wool which may be used as substitute for asbestos in some industrial applications. Knowledge of the reaction path and transformation sequence is very important to assess whether carcinogenic crystalline phases are formed during devitrification, which may occur when used as insulators. In addition, knowledge about the nature of the phases formed at high temperature is mandatory to assess if thermally transformed MMVF can be safely recycled as secondary raw material (SRM). In this scenario, this study provides useful information for the optimization of the industrial annealing process aimed to attain a safe, recyclable product. The results of this work show that one of the high-temperature products of Cerafiber and Superwool is cristobalite which is classified as a carcinogenic. It was possible to define the temperature interval at which Cerafiber and Superwool fibers can be safely used as thermal insulators (e.g. insulators in tunnel and/or roller kilns, etc.). As cristobalite is formed in both synthetic fiber products at temperatures higher than 1200 degrees C, their use should be limited to devices operating at lower temperatures. Rock and Glass wool melt upon thermal treatment. As far as the industrial process of inertization is concerned, a maximum firing temperature of 1100 and 600 degrees C is required to melt Rock wool and Glass wool, respectively, with the high-temperature products that can be safely recycled as SRM. Recycling of these products in stoneware tile mixtures were subsequently attempted. The addition of 1-2 wt.% of the melts of Rock and Glass wool gave promising results in terms of viscous sintering reactions and resistance to staining with the only weak characteristic being the color properties of the fired bodies which tend to worsen.     

Formaldehyde degradation in an anaerobic packed-bed bioreactor

The development of appropriate technologies for the treatment of formaldehyde discharged into the environment is important to minimize its impact. Aerobic systems have been employed, although alternative anaerobic treatments have also been widely studied, mainly due to their low energy consumption and sludge production. However, toxic substances can lead to disturbances in anaerobic reactors. Some research has already been developed on formaldehyde anaerobic biological treatment, but no consensus has yet been reached about its behavior nor has the most efficient system been identified. Aiming at finding supporting evidence for this issue, therefore, this study investigated the degradation and toxicity of formaldehyde in a Horizontal-Flow Anaerobic Immobilized Sludge Reactor. Formaldehyde concentrations of 26.2-1158.6 mg HCHO/L were applied in the reactor, resulting in formaldehyde and chemical oxygen demand removal efficiencies of 99.7% and 92%, respectively. Volatile fatty acids with up to five carbons, found during the degradation of formaldehyde, are believed to indicate that the degradation followed routes unlike those suggested in the literature, which reports the formation of intermediates such as methanol and formic acid. The Monod kinetic model adhered to the experimental data well, with apparent kinetic parameters estimated as r(app)max) = 2.79 x 10(-3) mg HCHO/mg SSVh and K (app)(s) = 242.8 mg HCHO/L.     

Influence of the agitation rate on the treatment of partially soluble wastewater in anaerobic sequencing batch biofilm reactor

This work reports on the influence of the agitation rate on the organic matter degradation in an anaerobic sequencing batch reactor, containing biomass immobilized on 3 cm cubic polyurethane matrices, stirred mechanically and fed with partially soluble soymilk substrate with mean chemical oxygen demand (COD) of 974+/-70 mg l(-1). Hydrodynamic studies informed on the homogenization time under agitagion rates from 500 to 1100 rpm provided by three propeller impellers. It occurred very quickly compared to the total cycle time. The results showed that agitation provided good mixing and improved the overall organic matter consumption rates. A modified first-order kinetic model represented adequately the data in the entire range of agitation rate. The apparent first-order kinetic constant for suspended COD rose approximately 360% when the agitation rate was changed from 500 to 900 rpm, whereas the apparent first-order kinetic constant for soluble COD did not vary significantly.     

A distributed probabilistic system for adaptive regulation of imageprocessing parameters

A distributed optimization framework and its application to the regulation of the behavior of a network of interacting image processing algorithms are presented. The algorithm parameters used to regulate information extraction are explicitly represented as state variables associated with all network nodes. Nodes are also provided with message-passing procedures to represent dependences between parameter settings at adjacent levels. The regulation problem is defined as a joint-probability maximization of a conditional probabilistic measure evaluated over the space of possible configurations of the whole set of state variables (i.e., parameters). The global optimization problem is partitioned and solved in a distributed way, by considering local probabilistic measures for selecting and estimating the parameters related to specific algorithms used within the network. The problem representation allows a spatially varying tuning of parameters, depending on the different informative contents of the subareas of an image. An application of the proposed approach to an image processing problem is described. The processing chain chosen as an example consists of four modules. The first three algorithms correspond to network nodes. The topmost node is devoted to integrating information derived from applying different parameter settings to the algorithms of the chain. The nodes associated with data-transformation processes to be regulated are represented by an optical sensor and two filtering units (for edge-preserving and edge-extracting filterings), and a straight-segment detection module is used as an integration site.