Modern anthropomorphic upper limb bioprostheses are typically controlled by electromyographic (EMG) biosignals using a pattern recognition scheme. Unfortunately, there are many factors originating from the human source of objects to be classified and from the human-prosthesis interface that make it difficult to obtain an acceptable classification quality. One of these factors is the high susceptibility of biosignals to contamination, which can considerably reduce the quality of classification of a recognition system. In the paper, the authors propose a new recognition system intended for sEMG based control of the hand prosthesis with detection of contaminated biosignals in order to mitigate the adverse effect of contaminations. The system consists of two ensembles: the set of one-class classifiers to assess the degree of contamination of individual channels and the ensemble of K-nearest neighbours (KNN) classifier to recognise the patient’s intent. For all recognition systems, an original, coherent fuzzy model was developed, which allows the use of a uniform soft (fuzzy) decision scheme throughout the recognition process. In the proposed decision scheme the degree of contamination is used to the dynamic linguistic hedge operation of fuzzy similarity relation by the dilation procedure which next is used in the original fuzzy model of KNN classifiers. The experimental evaluation was conducted using real biosignals from a public repository. The goal was to provide an experimental comparative analysis of the parameters and procedures of the developed method on which the quality of the recognition system depends. The proposed fuzzy recognition system was also compared with similar systems described in the literature. The results show that the proposed method improves robustness to noise for KNN-based classifiers. It also outperforms classifiers based on other paradigmas when the contamination level of sEMG signal is high.

Modern upper limb bioprostheses are typically controlled by sEMG signals using a pattern recognition scheme in the control process. Unfortunately, the sEMG signal is very susceptible to contamination that deteriorates the quality of the control system and reduces the usefulness of the prosthesis in the patient's everyday life. In the paper, the authors propose a new recognition system intended for sEMG-based control of the hand prosthesis with detection of contaminated sEMG signals. The originality of the proposed solution lies in the co-operation of two recognition systems working in a cascade structure: (1) an ensemble of one-class classifiers used to recognise contaminated signals and (2) a naive Bayes classifier (NBC) which recognises the patient's intentions using the information about contaminations produced by the ensemble. Although in the proposed approach, the NBC model is changed dynamically, due to the multiplicative form of the classification functions, training can be performed in a one-shot procedure. Experimental studies were conducted using real sEMG signals. The results obtained confirm the hypothesis that the use of the one-class classifier ensemble and the dynamic NBC model leads to improved classification quality.

The paper presents an original method for controlling a surface-electromyography-driven (sEMG) prosthesis. A context-dependent recognition system is proposed in which the same class of sEMG signals may have a different interpretation, depending on the context. This allowed the repertoire of performed movements to be increased. The proposed structure of the context-dependent recognition system includes unambiguously defined decision sequences covering the overall action of the prosthesis, i.e. the so-called boxes. Because the boxes are mutually isolated environments, each box has its own interpretation of the recognition result, as well as a separate local-recognition-task-focused classifier. Due to the freedom to assign contextual meanings to classes of biosignals, the construction procedure of the classifier can be optimised in terms of the local classification quality in a given box or the classification quality of the entire system. In the paper, two optimisation problems are formulated, differing in the adopted constraints on optimisation variables, with the methods of solving the problems based on an exhaustive search and an evolutionary algorithm, being developed. Experimental studies were conducted using signals from 1 able-bodied person with simulation of amputation and 10 volunteers with transradial amputations. The study compared the classical recognition system and the context-dependent system for various classifier models. An unusual testing strategy was adopted in the research, taking into account the specificity of the considered recognition task, with two original quality measures resulting from this scheme then being applied. The results obtained confirm the hypothesis that the application of the context-dependent classifier led to an improvement in classification quality.

Myopotential pattern recognition to decode the intent of the user is the most advanced approach to controlling a powered bioprosthesis. Unfortunately, many factors make this a difficult problem and achieving acceptable recognition quality in real-word conditions is a serious challenge. The aim of the paper is to develop a recognition system that will mitigate factors related to multimodality and multichannel recording of biosignals and their high susceptibility to contamination. The proposed method involves the use of two co-operating multiclassifier systems. The first system is composed of one-class classifiers related to individual electromyographic (EMG) and mechanomyographic (MMG) biosignal recording channels, and its task is to recognise contaminated channels. The role of the second system is to recognise the class of movement resulting from the patient's intention. The ensemble system consists of base classifiers using the representation (extracted features) of biosignals from different channels. The system uses a dynamic selection mechanism, eliminating those base classifiers that are associated with biosignal channels that are recognised by the one-class ensemble system as being contaminated. Experimental studies were conducted using signals from an able-bodied person with simulation of amputation. The results obtained allow us to reject the null hypothesis that the application of the dual ensemble foes not lead to improved classification quality.

The study aimed to determine whether four weeks of motor imagery training (MIT) of goal-directed reaching (reaching to grasp task) would affect the cortical activity during motor imagery of reaching (MIR) and grasping (MIG) in the same way. We examined cortical activity regarding event-related potentials (ERPs) in healthy young participants. Our study also evaluated the subjective vividness of the imagery. Furthermore, we aimed to determine the relationship between the subjective assessment of motor imagery (MI) ability to reach and grasp and the cortical activity during those tasks before and after training to understand the underlying neuroplasticity mechanisms. Twenty-seven volunteers participated in MIT of goal-directed reaching and two measurement sessions before and after MIT. During the sessions 128-channel electroencephalography (EEG) was recorded during MIR and MIG. Also, participants assessed the vividness of the MI tasks using a visual analog scale (VAS). The vividness of imagination improved significantly ( P < .05) after MIT. A repeated measures ANOVA showed that the task (MIR/MIG) and the location of electrodes had a significant effect on the ERP's amplitude ( P < .05). The interaction between the task, location, and session (before/after MIT) also had a significant effect on the ERP's amplitude ( P < .05). Finally, the location of electrodes and the interaction between location and session had a significant effect on the ERP's latency ( P < .05). We found that MIT influenced the EEG signal associated with reaching differently than grasping. The effect was more pronounced for MIR than for MIG. Correlation analysis showed that changes in the assessed parameters due to MIT reduced the relationship between the subjective evaluation of imagining and the EEG signal. This finding means that the subjective evaluation of imagining cannot be a simple, functional insight into the bioelectrical activity of the cerebral cortex expressed by the ERPs in mental training. The changes we noted in ERPs after MIT may benefit the use of non-invasive EEG in the brain-computer interface (BCI) context. Trial registration: NCT04048083.

Many dynamic ensemble selection (DES) methods are known in the literature. A previously-developed by the authors, method consists in building a randomized classifier which is treated as a model of the base classifier. The model is equivalent to the base classifier in a certain probabilistic sense. Next, the probability of correct classification of randomized classifier is taken as the competence of the evaluated classifier. In this paper, a novel randomized model of base classifier is developed. In the proposed method, the random operation of the model results from a random selection of the learning set from the family of learning sets of a fixed size. The paper presents the mathematical foundations of this approach and shows how, for a practical application when learning and validation sets are given, one can determine the measure of competence and build a MC system with the DES scheme. The DES scheme with the proposed model of competence was experimentally evaluated on the collection of 67 benchmark datasets and compared in terms of eight quality criteria with two ensemble classifiers which use the previously-proposed concepts of randomized model. The proposed approach achieved the lowest ranks for almost all investigated quality criteria.

In this paper, the issue of tailoring the soft confusion matrix (SCM) based classifier to deal with stream learning task is addressed. The main goal of the work is to develop a wrapping-classifier that allows incremental learning to classifiers that are unable to learn incrementally. The goal is achieved by making two improvements in the previously developed SCM classifier. The first one is aimed at reducing the computational cost of the SCM classifier. To do so, the definition of the fuzzy neighbourhood of an object is changed. The second one is aimed at effective dealing with the concept drift. This is done by employing the ADWIN-driven concept drift detector that is not only used to detect the drift but also to control the size of the neighbourhood. The obtained experimental results show that the proposed approach significantly outperforms the reference methods.

In ensemble regression three steps can be distinguished: first, the set of base models is created, next the ensemble of best models is selected and finally predictions of individual models are integrated to receive common prediction for a given test sample. In this paper a novel method for selection of base models in dynamic mode is developed. According to the proposed concept, models are selected in pairs of complementary models, so as to minimize the mutual prediction error in the predefined area around the test object. In order to define the area in which a pair of complementary models is sought, the concept of orthants in the space of independent variables is used. The proposed method was experimentally tested and compared against 6 literature ensemble regression methods using 14 benchmark data bases and 3 homogeneous ensembles of base models: multilayer perceptrons, 5-nearest neighbor models and linear regression models. The experimental results demonstrate that the proposed method leads to the statistically significant smaller values of the relative root mean squared error of the prediction.

In this paper, the issue of tailoring the soft confusion matrix classifier to deal with imbalanced data is addressed. This is done by changing the definition of the soft neighbourhood of the classified object. The first approach is to change the neighbourhood to be more local by changing the Gaussian potential function approach to the nearest neighbour rule. The second one is to weight the instances that are included in the neighbourhood. The instances are weighted inversely proportional to the a priori class probability. The experimental results show that for one of the investigated base classifiers, the usage of the KNN neighbourhood significantly improves the classification results. What is more, the application of the weighting schema also offers a significant improvement.

In this paper, an issue of building the RRC model using probability distributions other than beta distribution is addressed. More precisely, in this paper, we propose to build the RRR model using the truncated normal distribution. Heuristic procedures for expected value and the variance of the truncated-normal distribution are also proposed. The proposed approach is tested using SCM-based model for testing the consequences of applying the truncated normal distribution in the RRC model. The experimental evaluation is performed using four different base classifiers and seven quality measures. The results showed that the proposed approach is comparable to the RRC model built using beta distribution. What is more, for some base classifiers, the truncated-normal-based SCM algorithm turned out to be better at discovering objects coming from minority classes.

The paper presents a novel concept of ensemble classifier applied to the classification of electromyographic (EMG) and mechanomyographic (MMG) signals in the system of bioprosthetic hand control. In the developed multiclassifier (MC) system, first the base classifiers are grouped in terms of diversity using the novel criterion based on the type of uncertainty in the classification process. In the next step, the ensemble is pruned by selecting the best classifiers from each cluster. One-feature classifiers have been adopted as the base classifiers, i.e. pruning of classifier ensemble denotes also a feature selection procedure. The classification quality of proposed recognition method was experimentally tested and compared with seven literature ensemble systems with pruning classifiers and feature selection procedure based on Kolmogorov criterion. Real EMG and MMG biosignals for the classification of 11 types of grasping movements were used in experiments.

In this work we addressed the issue of applying a stochastic classifier and a local, fuzzy confusion matrix under the framework of multi-label classification. We proposed a novel solution to the problem of correcting label pairwise ensembles. The main step of the correction procedure is to compute classifier-specific competence and cross-competence measures, which estimates error pattern of the underlying classifier. At the fusion phase we employed two weighting approaches based on information theory. The classifier weights promote base classifiers which are the most susceptible to the correction based on the fuzzy confusion matrix. During the experimental study, the proposed approach was compared against two reference methods. The comparison was made in terms of six different quality criteria. The conducted experiments reveals that the proposed approach eliminates one of main drawbacks of the original FCM-based approach i.e. the original approach is vulnerable to the imbalanced class/label distribution. What is more, the obtained results shows that the introduced method achieves satisfying classification quality under all considered quality criteria. Additionally, the impact of fluctuations of data set characteristics is reduced.

Classifier ensemble pruning is a technique to increase ensemble accuracy and reduce its size. In the paper we propose a novel method of ensemble pruning which is based on the local class and inter-class competences of base classifiers for the test object. In the first step, using the inter-class competences, we identify classifiers that tend to commit the same type of errors. The identified classifiers create a common cluster. In the second step, classifiers in each cluster are ordered according to the class competence and then the top-k ones are selected from each cluster as the final ensemble. Experimental investigations of the proposed pruning method for two homogeneous pools of base classifiers and for three values of parameter k were conducted. In experimental studies we used 16 real data sets related to the various medical decision-making problems. Obtained results clearly show that the use of the ensemble pruning procedure leads to the better classification accuracy.

In this paper, the problem of multilabel classification using the classifier chain scheme is addressed. We deal with the problem of building a diverse ensemble of the classifier-chain-based ensemble. For this purpose, we propose a permutation-based criterion of chain diversity. The final ensemble is build using a multi-objective genetic algorithm, which is used to optimise classification quality and chain diversity simultaneously. The proposed methods were evaluated using 29 benchmark datasets. The comparison was performed using four different multi-label evaluation measures. The experimental study reveals that the proposed approach provides a better classification quality than response-based diversity criteria.

In this paper, we deal with the task of building a dynamic ensemble of chain classifiers for multi-label classification. To do so, we proposed two concepts of the classifier chain algorithms that are able to change the label order of the chain without rebuilding the entire model. Such models allow anticipating the instance-specific chain order without the significant increase in the computational burden. The proposed chain models are built using the Naive Bayes classifier and nearest neighbour approaches. To take the benefits of the proposed algorithms, we developed a simple heuristic that allows the system to find relatively good label order. The experimental results showed that the proposed models and the heuristic are efficient tools for building dynamic chain classifiers.

In the study the problem of ensemble regression with fuzzy linear regression (FLR) models is considered. For this case a novel method of integration is proposed in which first fuzzy responses of base FLR models are integrated and next the fuzzy response of a common model is defuzzified. Four different operators are defined for integration procedure. The performance of proposed integration methods of FLR base models on the soft level were compared against state-of-the-art integration method on the crisp level using computer generated datasets with linear, 2-order and 3-order models and different variances of Gaussian disturbances. As a criterion of method quality the root mean square error was applied. The results of computer experiments clearly show that in many cases proposed methods significant outperform the reference approach.

In the paper, the problem of multi-label (ML) classification using the label-pairwise (LPW) scheme is addressed. For this approach, the method of correction of binary classifiers which constitute the LPW ensemble is proposed. The correction is based on a probabilistic (randomized) model of a classifier that assesses the local class-specific probabilities of correct classification and misclassification. These probabilities are determined using the original concepts of a randomized reference classifier (RRC) and a local soft confusion matrix. Additionally, two special cases that deal with imbalanced labels and double labeled instances are considered. The proposed methods were evaluated using 29 benchmark datasets. In order to assess the efficiency of the introduced models and the proposed correction scheme, they were compared against original binary classifiers working in the LPW ensemble. The comparison was performed using four different ML evaluation measures: macro and micro-averaged [Formula: see text] loss, zero-one loss and Hamming loss. Moreover, relations between classification quality and the characteristics of ML datasets such as average imbalance ratio or label density were investigated. The experimental study reveals that the correction approaches significantly outperform the reference method in terms of zero-one loss and Hamming loss.

One of the biggest problems of upper limb transplantation is lack of certainty as to whether a patient will be able to control voluntary movements of transplanted hands. Based on findings of the recent research on brain cortex plasticity, a premise can be drawn that mental training supported with visual and sensory feedback can cause structural and functional reorganization of the sensorimotor cortex, which leads to recovery of function associated with the control of movements performed by the upper limbs. In this study, authors based on the above observations propose the computer-aided training (CAT) system, which generating visual and sensory stimuli, should enhance the effectiveness of mental training applied to humans before upper limb transplantation. The basis for the concept of computer-aided training system is a virtual hand whose reaching and grasping movements the trained patient can observe on the VR headset screen (visual feedback) and whose contact with virtual objects the patient can feel as a touch (sensory feedback). The computer training system is composed of three main components: (1) the system generating 3D virtual world in which the patient sees the virtual limb from the perspective as if it were his/her own hand; (2) sensory feedback transforming information about the interaction of the virtual hand with the grasped object into mechanical vibration; (3) the therapist's panel for controlling the training course. Results of the case study demonstrate that mental training supported with visual and sensory stimuli generated by the computer system leads to a beneficial change of the brain activity related to motor control of the reaching in the patient with bilateral upper limb congenital transverse deficiency.

In this paper a new method for calculating the classifier competence in the dynamic mode is developed. In the method, first decision profile of the classified object is calculated using K nearest objects from the validation set. Next, the decision profile is compared with the support vector produced by the classifier. The competence measure reflects the outcome of this comparison and rates the classifier with respect to the similarity of its support vector and decision profile of the test object in a continuous manner. Three different procedures for calculating decision profile and three different measures for comparing decision profile and support vector are proposed, which leads to nine methods of competence calculation. Two multiclassifier systems (MC) with homogeneous and heterogeneous pool of base classifiers and with dynamic ensemble selection scheme (DES) were constructed using the methods developed. The performance of constructed MC systems was compared against seven state-of-the-art MC systems using 15 benchmark data sets taken from the UCI Machine Learning Repository. The experimental investigations clearly show the effectiveness of the combined multiclassifier system in dynamic fashion with the use of the proposed measures of competence regardless of the ensemble type used.