Estimated blind people in the world will exceed 40 million by 2025. To develop novel algorithms based on fundus image descriptors that allow automatic classification of retinal tissue into healthy and pathological early stages is necessary. In this paper, we focus one most common pathologies current society: diabetic retinopathy. The proposed method avoids necessity lesion segmentation or candidate map generation before stage. Local binary patterns granulometric profiles are locally computed...

State-of-the-art high-end prostheses are electro-mechanically able to provide a great variety of movements. Nevertheless, in order functionally replace human limb, it is essential that each movement properly controlled. This the goal prosthesis control, which has become growing research field last decades, with ultimate reproducing biological limb control. Therefore, exploration and development control crucial improve many aspects an amputee’s life. Nowadays, large divergence between...

In proportional myographic control, one can control either position or velocity of movement. Here, we propose to use adaptive auto-regressive filters, so as gradually adjust between the two. We implemented this in an system with closed-loop feedback, where both user and machine simultaneously attempt follow a cursor on 2-D arena. tested 15 able-bodied three limb-deficient participants using eight-channel myoelectric armband. The human-machine pairs learn perform smoother movements larger...

Electromyography-based wearable biosensors are used for prosthetic control. Machine learning controllers based on classification and regression models. The advantage of the approach is that it permits us to obtain a smoother more natural controller. However, existing training methods regression-based solutions same as protocol in approach, where only finite set movements trained. In this paper, we present novel myoelectric include feedback term allows explore than automatically adjusted...

The detection and identification of internal defects in a material require the use some technology that translates hidden interior damages into observable signals with different signature-defect correspondences. We apply impact-echo techniques for this purpose. materials are classified according to their defective status (homogeneous, one defect or multiple defects) kind (hole crack, passing through not). Every specimen is impacted by hammer, spectrum propagated wave recorded. This input...

Patient empowerment is seen as the capability to understand health information and make decisions based on it. It a competence that can improve self-care, adherence overall health. The COVID-19 pandemic has increased need for also reduced number of visits centers. Nurses have had adapt in order continue offering quality care different environments such digital world, but this entails assessing level their patients’ adapting material educational messages new realities. aim study is, one hand,...

We present two applications in image processing of an agglomerative clustering method based on mixtures non-Gaussian distributions. The joins pair-wise the mixture models estimated for every cluster building a pyramidal or hierarchical structure by using Kullback-Leibler divergence. This process can be related with feedforward abstraction carried out brain. consist grouping images their content similarities and segmentation regions similar areas. capability to distinguish between natural...

Deep learning-based algorithms have led to tremendous progress over the last years, but they face a bottleneck as their optimal development highly relies on access large datasets. To mitigate this limitation, cross-silo federated learning has emerged way train collaborative models among multiple institutions without having share raw data used for model training. However, although artificial intelligence experts expertise develop state-of-the-art and actively code through notebook...

The analysis and characterization of atrial fibrillation requires the prior extraction activity from electrocardiogram, where independent ventricular activities are combined in addition to noise. An component method is proposed additional knowledge about time statistical structure sources incorporated. Finally, a based on maximum likelihood second order blind identification obtained validated with results that improve those traditional ICA algorithms.

This paper presents a novel procedure to classify data from mixtures of independent component analyzers. The includes two stages: learning the parameters (basis vectors and bias terms) clustering ICA following bottom-up agglomerative scheme construct hierarchy for classification. approach estimation source probability density function is non-parametric minimum kullback-Leibler distance used as criterion merging clusters at each level hierarchy. Validation proposed method presented several...

In this paper, we address the challenge of ensuring stability in bipedal walking robots and exoskeletons. We explore feasibility real-time implementation for Predicted Step Viability algorithm (PSV), a complex multi-step optimization criterion planning future steps gait. To overcome high computational cost PSV algorithm, performed an analysis using 11 classification algorithms stacking strategy to predict if step will be stable or not. generated three datasets increasing complexity through...

Signals obtained from impact-echo techniques can be used to detect and classify the defects in damaged materials. The change wave propagation between impact sensors producing particular spectrum elements, which define feature vector. We propose a hierarchical clustering method that models vector as mixture of Gaussians (MoG) for every class then merges different clusters using distance measure symmetric Kullback-Leibler (KL) divergence. Since there is no closed-form solution KL divergence...

The new times, marked by immediacy, globalization, and technological advances, has forced health professionals to develop competencies adapt the challenges. However, necessary skills such as using digital tools are primarily ignored institutions, hospitals, universities, forcing undertake training in these areas independently. This research aims analyse if there is a transfer of what been learned healthcare field their professional practice patients. To perform study, 104 professionals,...

Atrial fibrillation disorders are one of the main arrhythmias elderly. The atrial and ventricular activities decoupled during an episode, very rapid irregular waves replace usual P-wave in a normal sinus rhythm electrocardiogram (ECG). estimation these wavelets is must for clinical analysis. We propose new approach to this problem focused on quasiperiodicity wavelets. activity characterized by interval 3-12 Hz. It enables us establish as separation original sources from instantaneous linear...

Introduction Digital literacy helps patients to be more informed in order make decisions about their health. Patient empowerment the digital realm is a duty for all healthcare professionals, but nurses are most numerous professionals systems worldwide. In addition, they have an important role health education and patient follow-up. Therefore, use of tools, by empower help know health, crucial. Objective This study was conducted identify nurses’ views on benefits as well constraints encounter...

In this paper, we propose and analyze by means of simulations the use surrogate data algorithms for blind detection nonlinearities in multiple-echo ultrasonic signals. We assume a scheme so that no information about input (emitted pulse) can be used. The metrics equations model some nonlinear situations are carefully reviewed. Also, closed form third-order from simplified second-order Volterra kernel derived. Computer show technique is potentially powerful tool signals if adequate chosen....

We present an application of nonnegative matrix factorization (NMF) in astrophysics. It consists the study ice mixtures obtained laboratory. They simulate real astrophysical ices. The goal is to identify molecules that are mixtures. data correspond infrared absorption spectra ices formed by different combinations molecules. and abundances nonnegative, allowing NMF. In addition, statistics supergaussian signals, so a sparseness restriction can be added. review some NMF algorithms imposing...