The myoelectric signal reflects the electrical activity of skeletal muscles and contains information about the structure and function of the muscles which make different parts of the body move. Advances in engineering have extended electromyography beyond the traditional diagnostic applications to also include applications in diverse areas such as rehabilitation, movement analysis and myoelectric control of prosthesis. This paper aims to study and develop a system that uses myoelectric signals, acquired by surface electrodes, to characterize certain movements of the human arm. To recognize certain hand-arm segment movements, was developed an algorithm for pattern recognition technique based on neuro-fuzzy, representing the core of this research. This algorithm has as input the preprocessed myoelectric signal, to disclosed specific characteristics of the signal, and as output the performed movement. The average accuracy obtained was 86% to 7 distinct movements in tests of long duration (about three hours).

Surface Electromyography (sEMG) signal processing has a disruptive technology potential to enable a natural human interface with artificial limbs and assistive devices. However, this biosignal real-time control interface still presents several restrictions such as control limitations due to a lack of reliable signal prediction and standards for signal processing among research groups. Our paper aims to present and validate our sEMG database through the signal classification performed by the reliable forms of our Extreme Learning Machines (ELM) classifiers, used to maintain a more consistent signal classification. To perform the signal processing, we explore the use of a stochastic filter based on the Antonyan Vardan Transform (AVT) in combination with two variations of our Reliable classifiers (denoted R-ELM and R-Regularized ELM (RELM), respectively), to derive a reliability metric from the system, which autonomously selects the most reliable samples for the signal classification. To validate and compare our database and classifiers with related papers, we performed the classification of the whole of Databases 1, 2, and 6 (DB1, DB2, and DB6) of the NINAProdatabase. Our database presented consistent results, while the reliable forms of ELM classifiers matched or outperformed related papers, reaching average accuracies higher than $99\%$ for the IEEdatabase, while average accuracies of $75.1\%$, $79.77\%$, and $69.83\%$ were achieved for NINAPro DB1, DB2, and DB6, respectively.

This report describes a new crank arm-based force platform designed to evaluate the three-dimensional force applied to the pedals by cyclists in real conditions. The force platform was designed to be fitted on a conventional competition bicycle crankset while data is transmitted wirelessly through a Bluetooth^TM module and also stored on a SD card. A 3D solid model is created in the SolidWorks (Dassault Systèmes SOLIDWORKS Corp.) to analyze the static and dynamic characteristics of the crank arm by using the finite elements technique. Each crankset arm is used as a load cell based on strain gauges configured as three Wheatstone bridges. The signals are conditioned on a printed circuit board attached directly to the structure. The load cell showed a maximum nonlinearity error between 0.36% and 0.61% and a maximum uncertainty of 2.3% referred to the sensitivity of each channel. A roller trainer equipped with an optical encoder was also developed, allowing the measurement of the wheel’s instantaneous velocity.

Recent years have witnessed a rapid development of brain-computer interface (BCI) technology. An independent BCI is a communication system for controlling a device by human intension, e.g., a computer, a wheelchair or a neuroprosthes is, not depending on the brain’s normal output pathways of peripheral nerves and muscles, but on detectable signals that represent responsive or intentional brain activities. This paper presents a comparative study of the usage of the linear discriminant analysis (LDA) and the naive Bayes (NB) classifiers on describing both right- and left-hand movement through electroencephalographic signal (EEG) acquisition. For the analysis, we considered the following input features: the energy of the segments of a band pass-filtered signal with the frequency band in sensorimotor rhythms and the components of the spectral energy obtained through the Welch method. We also used the common spatial pattern (CSP) filter, so as to increase the discriminatory activity among movement classes. By using the database generated by this experiment, we obtained hit rates up to 70%. The results are compatible with previous studies.

Pattern recognition algorithms have introduced increasingly sophisticated solutions. However, many datasets are far from perfect; for example, they may include inconsistencies and have missing data, which may interfere with the classification process. Thus, the use of paraconsistent logic can provide a compelling quantitative analysis approach in classification algorithms because it deals directly with inaccurate, inconsistent and incomplete data. Paraconsistent logic is considered a nonclassical logic, which enables the processing of contradictory signals in its theoretical structure without invalidating the conclusions. In this context, the proposed approach aggregates the power of hybrid classifiers, the low noise susceptibility of the random forest approach and the robustness of paraconsistent logic to provide an intelligent treatment of contradictions and uncertainties in datasets. The proposed method is called paraconsistent random forest. The computational results demonstrated that paraconsistent random forest could classify several databases with satisfactory accuracy in comparison with state-of-the-art methods, namely, LDA, KNN, and SVM. Regarding imperfect datasets, the proposed approach significantly outperforms most of these methods in terms of prediction accuracy.

It is challenging to obtain good results for hand movements classification. Previous studies expended efforts on filters for sEMG data, feature extraction and classifier algorithms to achieve the best results. This paper proposes the insertion of a step in the classification process that selects which features to use in training aiming to increase accuracy and performance. Feature selection was previously used in other classification tasks but is new in wrist/fingers movements classification. Obtained results were positives as the performance gain is huge (39 to 53 features out of 144 are used for classification) and accuracy reach promising values (above 90% for some subjects).