ECG-based detection of left ventricle hypertrophy

Introduction Left ventricle hypertrophy (LVH) is an important risk factor for cardiovascular morbidity and mortality. It is characterized by a thickening of the walls of the left ventricle. The transthoracic echocardiogram is a very accurate method for LVH detection. However, the electrocardiogram (ECG) offers an alternative method in diagnosing LVH, besides being less expensive and easier to obtain. In this context, this study proposes an ECG based approach for left ventricle hypertrophy (LVH) classification. Methods According to the literature, several indexes have so far been proposed that suggest specific changes in cardiac structure, however, generally speaking there is no consensus about the best criteria. This way, instead of considering only one LVH criterion, a score derived from electrocardiographic traces was employed which explores the complementarity of the best criteria through a fusion strategy. The best criteria are those which discriminate normal and LVH ECGs. Results The experiments were performed in the Monica database with a group of fifty men. Half of the individuals had LVH diagnosed by calculating the left ventricular mass index measured by transthoracic echocardiography. The score fusion proposed achieved a sensitivity of 78.3% and specificity of 91.3%, outperforming all isolated LVH criteria. Discussion Unlike the other methods, our score must be estimated within a computer because of its high complexity. Even with this limitation it is much less expensive than using the echocardiography.

Commanding a robotic wheelchair with a high-frequency steady-state visual evoked potential based brain-computer interface.

This work presents a brain-computer interface (BCI) used to operate a robotic wheelchair. The experiments were performed on 15 subjects (13 of them healthy). The BCI is based on steady-state visual-evoked potentials (SSVEP) and the stimuli flickering are performed at high frequency (37, 38, 39 and 40 Hz). This high frequency stimulation scheme can reduce or even eliminate visual fatigue, allowing the user to achieve a stable performance for long term BCI operation. The BCI system uses power-spectral density analysis associated to three bipolar electroencephalographic channels. As the results show, 2 subjects were reported as SSVEP-BCI illiterates (not able to use the BCI), and, consequently, 13 subjects (12 of them healthy) could navigate the wheelchair in a room with obstacles arranged in four distinct configurations. Volunteers expressed neither discomfort nor fatigue due to flickering stimulation. A transmission rate of up to 72.5 bits/min was obtained, with an average of 44.6 bits/min in four trials. These results show that people could effectively navigate a robotic wheelchair using a SSVEP-based BCI with high frequency flickering stimulation.

Robotic wheelchair commanded by SSVEP, motor imagery and word generation.

This work presents a robotic wheelchair that can be commanded by a Brain Computer Interface (BCI) through Steady-State Visual Evoked Potential (SSVEP), Motor Imagery and Word Generation. When using SSVEP, a statistical test is used to extract the evoked response and a decision tree is used to discriminate the stimulus frequency, allowing volunteers to online operate the BCI, with hit rates varying from 60% to 100%, and guide a robotic wheelchair through an indoor environment. When using motor imagery and word generation, three mental task are used: imagination of left or right hand, and imagination of generation of words starting with the same random letter. Linear Discriminant Analysis is used to recognize the mental tasks, and the feature extraction uses Power Spectral Density. The choice of EEG channel and frequency uses the Kullback-Leibler symmetric divergence and a reclassification model is proposed to stabilize the classifier.

Premature Ventricular beat classification using a dynamic Bayesian Network.

This paper investigates the viability of using the dynamic Bayesian Network framework as a tool to classify heart beats in long term ECG records. A Decision Support System composed by two layers is considered. The first layer performs the segmentation of each heartbeat available in the ECG record, whereas the second layer classifies the heartbeat as Premature Ventricular Contraction (PVC) or Other. The use of both static and dynamic Bayesian Networks is evaluated through using the records available in the MIT-BIH database, and the results show that the Dynamic one performs better, obtaining 95% of sensitivity and 98% of positive predictivity, showing that to consider the temporal relation among events is a good strategy to increase the certainty about present events.

SSVEP-BCI implementation for 37-40 Hz frequency range.

This work presents a Brain-Computer Interface (BCI) based on Steady State Visual Evoked Potentials (SSVEP), using higher stimulus frequencies (>30 Hz). Using a statistical test and a decision tree, the real-time EEG registers of six volunteers are analyzed, with the classification result updated each second. The BCI developed does not need any kind of settings or adjustments, which makes it more general. Offline results are presented, which corresponds to a correct classification rate of up to 99% and a Information Transfer Rate (ITR) of up to 114.2 bits/min.

Detection of premature ventricular beats in ECG records using bayesian networks involving the P-Wave and fusion of results.

This article proposes to use the Bayesian network (BN) framework to support medical decision in the problem of heart beat classification in long-term electrocardiogram (ECG) records. The motivation to use the BN approach is to take into account the uncertainty present in the clinical reasoning. The case study is the classification of Premature Ventricular Beats (PVC). Specifically speaking, it is discussed the use of the P-Wave as a network node, to check its capability to improve the performance of the PVC classification. In spite of concluding that the P wave is not definitive for the classification, such results have motivated the main proposal of this work: a fusion of the results obtained by training the implemented BN with two distinct datasets, which has indeed improved the system performance.

Incremental SSVEP analysis for BCI implementation.

This work presents an incremental analysis of EEG records containing Steady-State Visual Evoked Potential (SSVEP). This analysis consists of two steps: feature extraction, performed using a statistic test, and classification, performed by a decision tree. The result is a system with high classification rate (a test with six volunteers resulted in an average classification rate of 91.2%), high Information Transfer Rate (ITR) (a test with the same six volunteers resulted in an average value of 100.2 bits/min) and processing time, for each incremental analysis, of approximately 120 ms. These are very good features for an efficient Brain-Computer Interface (BCI) implementation.

The use of bayesian networks for heart beat classification.

This work proposes to use a static Bayesian network as a tool to support medical decision in the on-line detection of Premature Ventricular Contraction beats (PVC) in electrocardiogram (ECG) records, which is a well known cardiac arrhythmia available for study in standard ECG databases. The main motivation to use Bayesian networks is their capability to deal with the uncertainty embedded in the problem (the medical reasoning itself frequently embeds some uncertainty). Indeed, the probabilistic inference is quite suitable to model this kind of problem, for considering its random character; as a consequence, random variables are used to propagate the uncertainty embedded in the problem. Some topologies of static Bayesian networks are implemented and tested in this work, in order to find out the one more suitable to the problem addressed. The results of such tests are discussed in details along the text, and the conclusions are highlighted.

Incremental HMM training applied to ECG signal analysis.

This work discusses the implementation of incremental hidden Markov model (HMM) training methods for electrocardiogram (ECG) analysis. The HMMs are used to model the ECG signal as a sequence of connected elementary waveforms. Moreover, an adaptation process is implemented to adapt the HMMs to the ECG signal of a particular individual. The adaptation training strategy is based on incremental versions of the expectation-maximization, segmental k-means and Bayesian approaches. Performance of the training methods was assessed through experiments considering the QT and ST-T databases. The results obtained show that the incremental training improves beat segmentation and ischemia detection performance with the advantage of low computational effort.

Human-machine interfaces based on EMG and EEG applied to robotic systems.

BACKGROUND: Two different Human-Machine Interfaces (HMIs) were developed, both based on electro-biological signals. One is based on the EMG signal and the other is based on the EEG signal. Two major features of such interfaces are their relatively simple data acquisition and processing systems, which need just a few hardware and software resources, so that they are, computationally and financially speaking, low cost solutions. Both interfaces were applied to robotic systems, and their performances are analyzed here. The EMG-based HMI was tested in a mobile robot, while the EEG-based HMI was tested in a mobile robot and a robotic manipulator as well. RESULTS: Experiments using the EMG-based HMI were carried out by eight individuals, who were asked to accomplish ten eye blinks with each eye, in order to test the eye blink detection algorithm. An average rightness rate of about 95% reached by individuals with the ability to blink both eyes allowed to conclude that the system could be used to command devices. Experiments with EEG consisted of inviting 25 people (some of them had suffered cases of meningitis and epilepsy) to test the system. All of them managed to deal with the HMI in only one training session. Most of them learnt how to use such HMI in less than 15 minutes. The minimum and maximum training times observed were 3 and 50 minutes, respectively. CONCLUSION: Such works are the initial parts of a system to help people with neuromotor diseases, including those with severe dysfunctions. The next steps are to convert a commercial wheelchair in an autonomous mobile vehicle; to implement the HMI onboard the autonomous wheelchair thus obtained to assist people with motor diseases, and to explore the potentiality of EEG signals, making the EEG-based HMI more robust and faster, aiming at using it to help individuals with severe motor dysfunctions.

A new mobile robot control approach via fusion of control signals.

This paper proposes an alternative approach to address the problem of coordinating behaviors in mobile robot navigation: fusion of control signals. Such approach is based on a set of two decentralized information filters, which accomplish the data fusion involved. Besides these two fusion engines, control architectures designed according to this approach also embed a set of different controllers that generate reference signals for the robot linear and angular speeds. Such signals are delivered to the two decentralized information filters, which estimate suitable overall reference signals for the robot linear and angular speeds, respectively. Thus, the background for designing such control architectures is provided by the nonlinear systems theory, which makes this approach different from any other yet proposed. This background also allows checking control architectures designed according to the proposed approach for stability. Such analysis is carried out in the paper, and shows that the robot always reaches its final destination, in spite of either obstacles along its path or the environment layout. As an example, a control architecture is designed to guide a mobile robot in an experiment, whose results allows checking the good performance of the control architecture and validating the design approach proposed as well.