An experimental evaluation of the incidence of fitness-function/search-algorithm combinations on the classification performance of myoelectric control systems with iPCA tuning.

BACKGROUND: The information of electromyographic signals can be used by Myoelectric Control Systems (MCSs) to actuate prostheses. These devices allow the performing of movements that cannot be carried out by persons with amputated limbs. The state of the art in the development of MCSs is based on the use of individual principal component analysis (iPCA) as a stage of pre-processing of the classifiers. The iPCA pre-processing implies an optimization stage which has not yet been deeply explored. METHODS: The present study considers two factors in the iPCA stage: namely A (the fitness function), and B (the search algorithm). The A factor comprises two levels, namely A1 (the classification error) and A2 (the correlation factor). Otherwise, the B factor has four levels, specifically B1 (the Sequential Forward Selection, SFS), B2 (the Sequential Floating Forward Selection, SFFS), B3 (Artificial Bee Colony, ABC), and B4 (Particle Swarm Optimization, PSO). This work evaluates the incidence of each one of the eight possible combinations between A and B factors over the classification error of the MCS. RESULTS: A two factor ANOVA was performed on the computed classification errors and determined that: (1) the interactive effects over the classification error are not significative (F0.01,3,72 = 4.0659 > fAB = 0.09), (2) the levels of factor A have significative effects on the classification error (F0.02,1,72 = 5.0162 < fA = 6.56), and (3) the levels of factor B over the classification error are not significative (F0.01,3,72 = 4.0659 > fB = 0.08). CONCLUSIONS: Considering the classification performance we found a superiority of using the factor A2 in combination with any of the levels of factor B. With respect to the time performance the analysis suggests that the PSO algorithm is at least 14 percent better than its best competitor. The latter behavior has been observed for a particular configuration set of parameters in the search algorithms. Future works will investigate the effect of these parameters in the classification performance, such as length of the reduced size vector, number of particles and bees used during optimal search, the cognitive parameters in the PSO algorithm as well as the limit of cycles to improve a solution in the ABC algorithm.

Evaluating different combinations of feature selection algorithms and cost functions applied to iPCA tuning in myoelectric control systems.

A myoelectric control system extracts information from electromyographic (EMG) signals and uses it to control different types of prostheses, so that people who suffered traumatisms, paralysis or amputations can use them to execute common movements. Recent research shows that the addition of a tuning stage, using the individual component analysis (iPCA), results in improved classification performance. We propose and evaluate a set of novel configurations for the iPCA tuning, based on a biologically inspired optimization procedure, the artificial bee colony algorithm. This procedure is implemented and tested using two different cost functions, the traditional classification error and the proposed correlation factor, which involves lower computational effort. We compare the tuned system's performance, in terms of correct classifications, to that of a system tuned using two standard algorithms, the sequential forward selection and the sequential floating forward selection. The statistical analyses of the results don't find a significant difference among the classification performances associated with the search algorithms (p < 0.01). On the other hand, they establish a significant difference among the classification performances related to the cost functions (p < 0.02).

Reconfigurable systems for sequence alignment and for general dynamic programming.

Reconfigurable systolic arrays can be adapted to efficiently resolve a wide spectrum of computational problems; parallelism is naturally explored in systolic arrays and reconfigurability allows for redefinition of the interconnections and operations even during run time (dynamically). We present a reconfigurable systolic architecture that can be applied for the efficient treatment of several dynamic programming methods for resolving well-known problems, such as global and local sequence alignment, approximate string matching and longest common subsequence. The dynamicity of the reconfigurability was found to be useful for practical applications in the construction of sequence alignments. A VHDL (VHSIC hardware description language) version of this new architecture was implemented on an APEX FPGA (Field programmable gate array). It would be several magnitudes faster than the software algorithm alternatives.