Publications

Wearable Health Devices for Diagnosis Support: Evolution and Future Tendencies

The use of wearable devices has increased substantially in recent years. This, together with the rise of telemedicine, has led to the use of these types of devices in the healthcare field. In this work, we carried out a detailed study on the use of these devices (regarding the general trends); we analyzed the research works and devices marketed in the last 10 years. This analysis extracted relevant information on the general trend of use, as well as more specific aspects, such as the use of sensors, communication technologies, and diseases. A comparison was made between the commercial and research aspects linked to wearables in the healthcare field, and upcoming trends were analyzed.

An Innovative 12-Lead Resting Electrocardiogram Dataset in Professional Football

This paper aims to provide a comprehensive and innovative 12-lead electrocardiogram (ECG) dataset tailored to understand the unique needs of professional football players. Other ECG datasets are available but collected from common people, normally with diseases confirmed [1–5], while it is well known that ECG characteristics change in athletes and elite players as a result of their intense long-term physical training [6]. This initiative is part of a broader research project employing machine learning (ML) to analyse ECG data in this athlete population and explore them according to the International Criteria for ECG Screening [7]. The dataset is generated through the establishment of a prospective observational cohort consisting of 54 male football players from La Liga, representing a UEFA Pro level team.Named the Pro Football 12-lead Resting Electrocardiogram Database (PF12RED)[8], it comprises 163 10-second ECG recordings, offering a detailed examination of the at-rest heart activity of professional football athletes. Data collection spans five phases over multiple seasons, including the 2018-2019 postseason, the 2019-20 preseason, the 2020-21 preseason, and the 2021-22 preseason. Athletes undergo medical evaluations that include a 10-second resting 12-lead ECG performed with General Electric's USB-CAM 14 module (https://co.services.gehealthcare.com/gehcstorefront/p/900995-002), with data saved using General Electric's CardioSoft V6.73 12SL V21 ECG Software. (https://www.gehealthcare.es/products/cardiosoft-v7)The data collection adheres to ethical principles, with clearance granted by the Autonomous Community of Andalusia Ethics Committee (Spain) under protocol number 1573-N-19 in December 2019. Participants provide informed consent, and data sharing is permitted following anonymization. The study aligns with the Declaration of Helsinki and adheres to the recommendations of the International Committee of Medical Journal Editors (ICMJE).Findings from the dataset, featuring examples based on the International Criteria for ECG Interpretation in athletes [7], reveal prevalent sport-related sinus bradycardia (SB) in 64.81% of cases, incomplete right bundle branch block (iRBBB) in 66.67%, right T-wave inversions (RTWI) in 66.87%, and Mobitz Type I–2º degree AV block (MTI) in 1.85% of cases. Additionally, one player (1.85%) exhibited T-wave inversions in the left leads V4, V5, and V6, requiring further cardiologist evaluation.The generated dataset, featuring ECG signals from professional football players, serves as a valuable resource for research in sports cardiology and cardiac health. Its potential for reuse encompasses:International Comparison: Enabling cross-regional comparisons of cardiac characteristics among elite football players, enriching international studies.ML Model Development: Facilitating the development and refinement of machine learning models for arrhythmia detection, serving as a benchmark dataset.Validation of Diagnostic Methods: Allowing the validation of automatic diagnostic methods, contributing to enhanced accuracy in detecting cardiac conditions.Research in Sports Cardiology: Supporting future investigations into specific cardiac adaptations in elite athletes and their relation to cardiovascular health.Reference for Athlete Protection Policies: Influencing athlete protection policies by providing data on cardiac health and suggesting guidelines for medical assessments.Health Professionals Training: Serving as a tra

Deep Neural Networks for the Recognition and Classification of Heart Murmurs Using Neuromorphic Auditory Sensors

Auscultation is one of the most used techniques for detecting cardiovascular diseases, which is one of the main causes of death in the world. Heart murmurs are the most common abnormal finding when a patient visits the physician for auscultation. These heart sounds can either be innocent, which are harmless, or abnormal, which may be a sign of a more serious heart condition. However, the accuracy rate of primary care physicians and expert cardiologists when auscultating is not good enough to avoid most of both type-I (healthy patients are sent for echocardiogram) and type-II (pathological patients are sent home without medication or treatment) errors made. In this paper, the authors present a novel convolutional neural network based tool for classifying between healthy people and pathological patients using a neuromorphic auditory sensor for FPGA that is able to decompose the audio into frequency bands in real time. For this purpose, different networks have been trained with the heart murmur information contained in heart sound recordings obtained from nine different heart sound databases sourced from multiple research groups. These samples are segmented and preprocessed using the neuromorphic auditory sensor to decompose their audio information into frequency bands and, after that, sonogram images with the same size are generated. These images have been used to train and test different convolutional neural network architectures. The best results have been obtained with a modified version of the AlexNet model, achieving 97% accuracy (specificity: 95.12%, sensitivity: 93.20%, PhysioNet/CinC Challenge 2016 score: 0.9416). This tool could aid cardiologists and primary care physicians in the auscultation process, improving the decision making task and reducing type-I and type-II errors.

A Neuro-Inspired Spike-Based PID Motor Controller for Multi-Motor Robots with Low Cost FPGAs

In this paper we present a neuro-inspired spike-based close-loop controller written in VHDL and implemented for FPGAs. This controller has been focused on controlling a DC motor speed, but only using spikes for information representation, processing and DC motor driving. It could be applied to other motors with proper driver adaptation. This controller architecture represents one of the latest layers in a Spiking Neural Network (SNN), which implements a bridge between robotics actuators and spike-based processing layers and sensors. The presented control system fuses actuation and sensors information as spikes streams, processing these spikes in hard real-time, implementing a massively parallel information processing system, through specialized spike-based circuits. This spike-based close-loop controller has been implemented into an AER platform, designed in our labs, that allows direct control of DC motors: the AER-Robot. Experimental results evidence the viability of the implementation of spike-based controllers, and hardware synthesis denotes low hardware requirements that allow replicating this controller in a high number of parallel controllers working together to allow a real-time robot control.

Non-small cell lung cancer diagnosis aid with histopathological images using Explainable Deep Learning techniques

Lung cancer has the highest mortality rate in the world, twice as high as the second highest. On the other hand, pathologists are overworked and this is detrimental to the time spent on each patient, diagnostic turnaround time, and their success rate.In this work, we design, implement, and evaluate a diagnostic aid system for non-small cell lung cancer detection, using Deep Learning techniques.The classifier developed is based on Artificial Intelligence techniques, obtaining an automatic classification result between healthy, adenocarcinoma and squamous cell carcinoma, given an histopathological image from lung tissue. Moreover, a report module based on Explainable Deep Learning techniques is included and gives the pathologist information about the image's areas used to classify the sample and the confidence of belonging to each class.The results show a system accuracy between 97.11 and 99.69%, depending on the number of classes classified, and a value of the area under ROC curve between 99.77 and 99.94%.The classification results obtain a substantial improvement according to previous works. Thanks to the given report, the time spent by the pathologist and the diagnostic turnaround time can be reduced.

Technical viability study for behavioral monitoring of wildlife animals in Doñana: An 802.15.4 coverage study in a Natural Park

The study and monitoring of wildlife and in semi-freedom has always been a subject of great interest. In recent years the technology allows to design low cost systems that facilitate these tasks: microcontrollers, low-power wireless networks, sensors. GPS, satellite and VHF has been used for position tracking and localization of wildlife. Our aim is to design a low-cost system for local monitoring of wildlife (collar or harness) with local memory and remote access. It will use multiples sensors for behavioral and health monitoring and fuse the processed information locally to reduce the stored data, but allowing to be sporadically transmitted through wireless networks. This collar will be based on an embedded low-power microcontroller with 802.15.4 transceiver and a set of sensors to provide data (activity and health) of the animal under monitoring: accelerometers, compass, humidity, temperature, light, microphone, heart rhythm. This collar is called mote in wireless sensors network (WSN) terminology. Only when one mote is closer to an 802.15.4 mote and under request, the collar will dump the information to a host database server through the 802.15.4 network. In this paper we present a viability study of the WSN for Donana Natural Park for different mote transmission powers, frequencies and distances for coverage. ZigBee and XBee mote has been proven.

Incremental Learning For Fundus Image Segmentation

SVITE: A Spike-Based VITE Neuro-Inspired Robot Controller

Multilayer Spiking Neural Network for Audio Samples Classification Using SpiNNaker

PaGER-Sync ADICVIDEO: Player affective gaming experience & responses - synchronized ADICVIDEO dataset

The <b>PaGER-Sync ADICVIDEO dataset</b> is a multimodal, temporally synchronized repository of physiological and facial expression data recorded during controlled, immersive video game sessions designed to simulate realistic home gaming environments. It integrates biosignals from the Empatica E4 wristband -including Electrodermal Activity (EDA), Blood Volume Pulse (BVP), and Skin Temperature (TEMP) - with facial expression features extracted from video recordings using FaceReader software. Additionally, the dataset includes scores from pre-session psychometric questionnaires (Gaming Addiction Scale, Scale of Positive and Negative Experience, Emotion Regulation Questionnaire) and demographic gender data, providing psychological and individual difference context. A summary file detailing the two strongest emotions expressed by each participant with their respective percentages is included. A total of 25 participants played three commercial video games (Tetris, Sonic Racing, and Fall Guys) under controlled conditions, while their physiological responses were continuously recorded and their facial expressions captured on video for subsequent analysis. All data streams were precisely aligned using a common video-based timestamp, enabling frame-level synchronization across modalities, and the data were segmented by game. The dataset supports a wide range of research applications in affective computing, human-computer interaction, and behavioral analysis, and is particularly well-suited for the development and evaluation of multimodal affect detection models, as well as for exploring the interplay between psychological traits and real-time emotional responses.

Evolution and Tendency on the Feature Extraction Process for Diagnostic Aid in Healthcare

Diagnostic support systems based on artificial intelligence are of great interest in the field of healthcare since they offer an early and highly accurate diagnosis. In recent years, the use of machine learning models that are less complex than the most widely used ones, such as deep neural networks, has been of interest. For this purpose, it is necessary to apply previous steps of feature extraction of different types, according to the problem that has to be solved. In this chapter, the authors intend to provide an overview of the current panorama of these systems, differentiating between those that extract features and those that do not. An exhaustive analysis is made of the works carried out in the last five years, checking the artificial intelligence models used, the features extracted, and the countries responsible for each work.

EMG-Controlled Robotic Prosthetic Arm With Neural Network Training

Sound Recognition System Using Spiking and MLP Neural Networks

Design and Development of a Sensorised Sports Waistcoat

Neuronal Specialization for Fine-Grained Distance Estimation Using a Real-Time Bio-Inspired Stereo Vision System

The human binocular system performs very complex operations in real-time tasks thanks to neuronal specialization and several specialized processing layers. For a classic computer vision system, being able to perform the same operation requires high computational costs that, in many cases, causes it to not work in real time: this is the case regarding distance estimation. This work details the functionality of the biological processing system, as well as the neuromorphic engineering research branch—the main purpose of which is to mimic neuronal processing. A distance estimation system based on the calculation of the binocular disparities with specialized neuron populations is developed. This system is characterized by several tests and executed in a real-time environment. The response of the system proves the similarity between it and human binocular processing. Further, the results show that the implemented system can work in a real-time environment, with a distance estimation error of 15% (8% for the characterization tests).

Does Two-Class Training Extract Real Features? A COVID-19 Case Study

Diagnosis aid systems that use image analysis are currently very useful due to the large workload of health professionals involved in making diagnoses. In recent years, Convolutional Neural Networks (CNNs) have been used to help in these tasks. For this reason, multiple studies that analyze the detection precision for several diseases have been developed. However, many of these works distinguish between only two classes: healthy and with a specific disease. Based on this premise, in this work, we try to answer the questions: When training an image classification system with only two classes (healthy and sick), does this system extract the specific features of this disease, or does it only obtain the features that differentiate it from a healthy patient? Trying to answer these questions, we analyze the particular case of COVID-19 detection. Many works that classify this disease using X-ray images have been published; some of them use two classes (with and without COVID-19), while others include more classes (pneumonia, SARS, influenza, etc.). In this work, we carry out several classification studies with two classes, using test images that do not belong to those classes, in order to try to answer the previous questions. The first studies indicate problems in these two-class systems when using a third class as a test, being classified inconsistently. Deeper studies show that deep learning systems trained with two classes do not correctly extract the characteristics of pathologies, but rather differentiate the classes based on the physical characteristics of the images. After the discussion, we conclude that these two-class trained deep learning systems are not valid if there are other diseases that cause similar symptoms.

Simulating Building Blocks for Spikes Signals Processing

Corrigendum to NAVIS: Neuromorphic Auditory VISualizer Tool. Neurocomputing 237C (2017) pp. 418-422.

An Approach to Distance Estimation with Stereo Vision Using Address-Event-Representation