People with severe physical disabilities are not able of using standard robotic wheelchairs, which generally demand some motor skills, and therefore total usage of associate muscles. Robotic wheelchairs commanded by Brain-Computer Interfaces (BCIs) based on Electroencephalography (EEG) have demonstrated to be an alternative for these end-users. In general, existing robotic wheelchairs commanded by BCIs require special platforms adapted to the EEG-BCI, and end-users need to attend a long training process for safely drive these devices. But many times these potential users do not have access to training sessions; due to mobility problems or technology access restrictions. This study proposes an EEG-based BCI with customizable configuration to be used in cloud architectures for remote control of robotic wheelchairs. This research explores two types of Steady State Visual Evoked Potential (SSVEP)-based BCI by applying Canonical Correlation Analysis (CCA) and compressive sensing (CS) as a novelty, adopting in one free calibration, and other including a calibration stage. The free-calibrated SSVEP recognition approach (CS-ncCCA) using compression ratio (CR) at 60% obtained accuracy (ACC) of 85% and information transfer rate (ITR) of 102 bits per minute (bpm), whereas the calibrated BCI (CS-wcCCA) applying also CR at 62% achieved ACC of 85% and ITR of 195 bpm. As a hilghlight, the proposed BCI allows significant reduction of the transmitted file size, and improves the communication latency that may be useful in remote and Cloud robotics applications, such as for users with severe motor disabilities, to train driving safely robotic wheelchairs via IoT.

This work presents a study of chromatic and luminance stimuli in low-, medium-, and high-frequency stimulation to evoke steady-state visual evoked potential (SSVEP) in the behind-the-ears area. Twelve healthy subjects participated in this study. The electroencephalogram (EEG) was measured on occipital (Oz) and left and right temporal (TP9 and TP10) areas. The SSVEP was evaluated in terms of amplitude, signal-to-noise ratio (SNR), and detection accuracy using power spectral density analysis (PSDA), canonical correlation analysis (CCA), and temporally local multivariate synchronization index (TMSI) methods. It was found that stimuli based on suitable color and luminance elicited stronger SSVEP in the behind-the-ears area, and that the response of the SSVEP was related to the flickering frequency and the color of the stimuli. Thus, green-red stimulus elicited the highest SSVEP in medium-frequency range, and green-blue stimulus elicited the highest SSVEP in high-frequency range, reaching detection accuracy rates higher than 80%. These findings will aid in the development of more comfortable, accurate and stable BCIs with electrodes positioned on the behind-the-ears (hairless) areas.

Recently, studies on cycling-based brain–computer interfaces (BCIs) have been standing out due to their potential for lower-limb recovery. In this scenario, the behaviors of the sensory motor rhythms and the brain connectivity present themselves as sources of information that can contribute to interpreting the cortical effect of these technologies. This study aims to analyze how sensory motor rhythms and cortical connectivity behave when volunteers command reactive motor imagery (MI) BCI that provides passive pedaling feedback. We studied 8 healthy subjects who performed pedaling MI to command an electroencephalography (EEG)-based BCI with a motorized pedal to receive passive movements as feedback. The EEG data were analyzed under the following four conditions: resting, MI calibration, MI online, and receiving passive pedaling (on-line phase). Most subjects produced, over the foot area, significant event-related desynchronization (ERD) patterns around Cz when performing MI and receiving passive pedaling. The sharpest decrease was found for the low beta band. The connectivity results revealed an exchange of information between the supplementary motor area (SMA) and parietal regions during MI and passive pedaling. Our findings point to the primary motor cortex activation for most participants and the connectivity between SMA and parietal regions during pedaling MI and passive pedaling.

The COVID-19 pandemic has highlighted the significance of both telemedicine and measuring physiological signals as peripheral Oxygen Saturation (SpO ₂ ) at home. Aiming to increase the confiability of SpO ₂ readings performed by smartphones, this study presents a novel Deep Neural Network (DNN) model that predicts SpO ₂ levels using fingertip videos captured by smartphones. The model was trained and evaluated using the leave-one-out method, and compared against the SpO ₂ values measured by a set of oximeters. The dataset used in this work is a differential of the study, unlike most of similar studies which use a breath-holding protocol that presents only a few samples of low SpO ₂ values, our tests were evaluated using a publicly available dataset in which volunteers’ SpO ₂ levels were measured in a range from almost 100% to approximately 70% by manipulating the percentage of oxygen inhaled by the volunteers, and the SpO ₂ levels were kept on lower values for a considerable period of time. The proposed DNN model was compared against multiple variations of the Ratio-of-Rations (RoR) method and the model proposed by the dataset creators. The results show that the DNN outputs have a strong relationship with the true SpO ₂ measurements, as indicated by a Pearson coefficient (PC) of 0.95 and a Concordance Correlation Coefficient (CCC) of 0.95, and also a small error, measured by a Root Mean Squared Error (RMSE) of 2.92% and a Mean Absolute Error (MAE) of 1.97%. Compared to the RoR method, which achieved a PC of 0.29, CCC of 0.21, RMSE of 8.83% and MAE of 7.03%, the proposed DNN model exhibits superior performance. Moreover, the proposed DNN model outperforms the model proposed by the dataset authors, which achieved a MAE of 5.00%. These findings demonstrate that the proposed DNN model can accurately predict SpO ₂ levels using smartphone-recorded fingertip videos with lower error and higher precision than a widely used classical method.