We recently introduced the concept of a new human-machine interface (the myokinetic control interface) to control hand prostheses. The interface tracks muscle contractions via permanent magnets implanted in the muscles and magnetic field sensors hosted in the prosthetic socket. Previously we showed the feasibility of localizing several magnets in non-realistic workspaces. Here, aided by a 3D CAD model of the forearm, we computed the localization accuracy simulated for three different below-elbow amputation levels, following general guidelines identified in early work. To this aim we first identified the number of magnets that could fit and be tracked in a proximal (T1), middle (T2) and distal (T3) representative amputation, starting from 18, 20 and 23 eligible muscles, respectively. Then we ran a localization algorithm to estimate the poses of the magnets based on the sensor readings. A sensor selection strategy (from an initial grid of 840 sensors) was also implemented to optimize the computational cost of the localization process. Results showed that the localizer was able to accurately track up to 11 (T1), 13 (T2) and 19 (T3) magnetic markers (MMs) with an array of 154, 205 and 260 sensors, respectively. Localization errors lower than 7% the trajectory travelled by the magnets during muscle contraction were always achieved. This work not only answers the question "how many magnets could be implanted in a forearm and successfully tracked with a the myokinetic control approach?", but also provides interesting insights for a wide range of bioengineering applications exploiting magnetic tracking.Reliable control of assistive devices using surface electromyography (sEMG) remains an unsolved task due to the signal's stochastic behavior that prevents robust pattern recognition for real-time control. Non-representative samples lead to inherent class overlaps that generate classification ripples for which the most common alternatives rely on post-processing and sample discard methods that insert additional delays and often do not offer substantial improvements. In this paper, a resilient classification pipeline based on Extreme Learning Machines (ELM) was used to classify 17 different upper-limb movements through sEMG signals from a total of 99 trials derived from three different databases. The method was compared to a baseline ELM and a sample discarding (DISC) method and proved to generate more stable and consistent classifications. The average accuracy boost of ≈ 10% in all databases lead to average weighted accuracy rates higher as 53,4% for amputees and 89,0% for non-amputee volunteers. The results match or outperform related works even without sample discards.Intellectual Developmental Disorder (IDD) is a neurodevelopmental disorder involving impairment of general cognitive abilities. This disorder impacts the conceptual, social, and practical skills adversely. There is a growing interest in exploring the neurological behavior associated with these disorders. Assessment of functional brain connectivity and graph theory measures have emerged as powerful tools to aid these research goals. The current research contributes by comparing brain connectivity patterns of IDD individuals to those typical controls. Considering the intellectual deficits linked to the IDD population, we hypothesized an atypical connectivity pattern in the IDD group. Brain signals were recorded by a dry-electrode Electroencephalography (EEG) system during the rest and music states observed by the subjects. We studied a group of seven IDD subjects and seven healthy controls to understand the connectivity within the human brain during the resting-state vis-à-vis while listening to music. Findings of this research emphasize (1) hyper-connected functional brain networks and increased modularity as potential characteristics of the IDD group, (2) the ability of soothing music to reduce the resting state hyper-connected pattern in the IDD group, and (3) the effect of soothing music in the lower frequency bands of the control group compared to the higher frequency bands of the IDD group.Motor imagery (MI) decoding is an important part of brain-computer interface (BCI) research, which translates the subject's intentions into commands that external devices can execute. The traditional methods for discriminative feature extraction, such as common spatial pattern (CSP) and filter bank common spatial pattern (FBCSP), have only focused on the energy features of the electroencephalography (EEG) and thus ignored the further exploration of temporal information. However, the temporal information of spatially filtered EEG may be critical to the performance improvement of MI decoding. In this paper, we proposed a deep learning approach termed filter-bank spatial filtering and temporal-spatial convolutional neural network (FBSF-TSCNN) for MI decoding, where the FBSF block transforms the raw EEG signals into an appropriate intermediate EEG presentation, and then the TSCNN block decodes the intermediate EEG signals. Moreover, a novel stage-wise training strategy is proposed to mitigate the difficult optimization problem of the TSCNN block in the case of insufficient training samples. Firstly, the feature extraction layers are trained by optimization of the triplet loss. Then, the classification layers are trained by optimization of the cross-entropy loss. Finally, the entire network (TSCNN) is fine-tuned by the back-propagation (BP) algorithm. Experimental evaluations on the BCI IV 2a and SMR-BCI datasets reveal that the proposed stage-wise training strategy yields significant performance improvement compared with the conventional end-to-end training strategy, and the proposed approach is comparable with the state-of-the-art method.We present a real-time monocular 3D reconstruction system on a mobile phone, called Mobile3DRecon. Using an embedded monocular camera, our system provides an online mesh generation capability on back end together with real-time 6DoF pose tracking on front end for users to achieve realistic AR effects and interactions on mobile phones. Unlike most existing state-of-the-art systems which produce only point cloud based 3D models online or surface mesh offline, we propose a novel online incremental mesh generation approach to achieve fast online dense surface mesh reconstruction to satisfy the demand of real-time AR applications. For each keyframe of 6DoF tracking, we perform a robust monocular depth estimation, with a multi-view semi-global matching method followed by a depth refinement post-processing. read more The proposed mesh generation module incrementally fuses each estimated keyframe depth map to an online dense surface mesh, which is useful for achieving realistic AR effects such as occlusions and collisions. We verify our real-time reconstruction results on two mid-range mobile platforms.