Publications

Implementing a Distance Estimator for a Wildlife Tracking System Based on 802.15.4

In this work, a novel distance estimation mechanism using received signal strength indication (RSSI) signals with ZigBee modules is designed, implemented and tested in several scenarios. This estimator was used for a research project focused on a wildlife behavioral classification system deployed in Doñana’s National Park. As a supporting feature for that project, this work was implemented for locating animal’s collars acting as wireless nodes in order to find those who went outside of the coverage area of the network or that were accidentally detached from animals. This work describes the system architecture and the implementation of a mobile assistant capable of recovering devices located beyond the coverage of the network. The analytical model needed for distance estimation and the signal filtering are described, as well as the difficulties that the researchers must deal when building robust location estimators. This theoretical model was applied to three different scenarios and tested with two validation experiments.

Affective State Assistant for Helping Users with Cognition Disabilities Using Neural Networks

Non-verbal communication is essential in the communication process. This means that its lack can cause misinterpretations of the message that the sender tries to transmit to the receiver. With the rise of video calls, it seems that this problem has been partially solved. However, people with cognitive disorders such as those with some kind of Autism Spectrum Disorder (ASD) are unable to interpret non-verbal communication neither live nor by video call. This work analyzes the relationship between some physiological measures (EEG, ECG, and GSR) and the affective state of the user. To do that, some public datasets are evaluated and used for a multiple Deep Learning (DL) system. Each physiological signal is pre-processed using a feature extraction process after a frequency study with the Discrete Wavelet Transform (DWT), and those coefficients are used as inputs for a single DL classifier focused on that signal. These multiple classifiers (one for each signal) are evaluated independently and their outputs are combined in order to optimize the results and obtain additional information about the most reliable signals for classifying the affective states into three levels: low, middle, and high. The full system is carefully detailed and tested, obtaining promising results (more than 95% accuracy) that demonstrate its viability.

A lightweight xAI approach to cervical cancer classification

Cervical cancer is caused in the vast majority of cases by the human papilloma virus (HPV) through sexual contact and requires a specific molecular-based analysis to be detected. As an HPV vaccine is available, the incidence of cervical cancer is up to ten times higher in areas without adequate healthcare resources. In recent years, liquid cytology has been used to overcome these shortcomings and perform mass screening. In addition, classifiers based on convolutional neural networks can be developed to help pathologists diagnose the disease. However, these systems always require the final verification of a pathologist to make a final diagnosis. For this reason, explainable AI techniques are required to highlight the most significant data to the healthcare professional, as it can be used to determine the confidence in the results and the areas of the image used for classification (allowing the professional to point out the areas he/she thinks are most important and cross-check them against those detected by the system in order to create incremental learning systems). In this work, a 4-phase optimization process is used to obtain a custom deep-learning classifier for distinguishing between 4 severity classes of cervical cancer with liquid-cytology images. The final classifier obtains an accuracy over 97% for 4 classes and 100% for 2 classes with execution times under 1 s (including the final report generation). Compared to previous works, the proposed classifier obtains better accuracy results with a lower computational cost.

Efficient Memory Organization for DNN Hardware Accelerator Implementation on PSoC

The use of deep learning solutions in different disciplines is increasing and their algorithms are computationally expensive in most cases. For this reason, numerous hardware accelerators have appeared to compute their operations efficiently in parallel, achieving higher performance and lower latency. These algorithms need large amounts of data to feed each of their computing layers, which makes it necessary to efficiently handle the data transfers that feed and collect the information to and from the accelerators. For the implementation of these accelerators, hybrid devices are widely used, which have an embedded computer, where an operating system can be run, and a field-programmable gate array (FPGA), where the accelerator can be deployed. In this work, we present a software API that efficiently organizes the memory, preventing reallocating data from one memory area to another, which improves the native Linux driver with a 85% speed-up and reduces the frame computing time by 28% in a real application.

Analyzing ECG signals in professional football players using machine learning techniques

BackgroundFootball player's health is important, and preventing sudden cardiac arrest may be a critical issue. Professional football players have different ECG signals than the average population, yet there are considerable gaps in study whereas the general population has been extensively studied.Objectives(a) Generate a reference and innovative resting 12-lead ECG database from 54 UEFA PRO level male football players from La Liga. This is a novel approach to cope the ECG and possible arrythmias in athletes. (b) Manage each XML athlete ECG data and develop a free-use program to visualize, denoise and filter the signal with the capacity to automate the labelling of the waves and save the reports. (c) Study the ECG wave shape and generate models through ML to analyse its utility to automate basic diagnosis.MethodsThe dataset collection is based on a prospective observational cohort and includes 10 s, 12-lead ECGs and rhythm and condition labels for each athlete. Physiological sport arrhythmias, T-Wave shape and other findings were studied and labelled. ECG Visualizer was developed and used for 3 machine learning (ML) methods to automate sinus bradycardia arrhythmia diagnosis.ResultsA dataset with 163 ECGs in XML format was collected comprising the Pro Football 12-lead Resting Electrocardiogram Database (PF12RED). "ECG Visualizer" software was developed, and ML was shown to be useful in detecting sinus bradycardia.ConclusionsThe study demonstrates that AI and machine learning can detect simple arrhythmias with accuracy, also it provides a valuable dataset and a free software application.

Performance and healthcare analysis in elite sports teams using artificial intelligence: a scoping review

In competitive sports, teams are increasingly relying on advanced systems for improved performance and results. This study reviews the literature on the role of artificial intelligence (AI) in managing these complexities and encouraging a system thinking shift. It found various AI applications, including performance enhancement, healthcare, technical and tactical support, talent identification, game prediction, business growth, and AI testing innovations. The main goal of the study was to assess research supporting performance and healthcare.

Aprendizaje Basado en Proyectos: Implementación de Interfaces Gráficas para microcontroladores ARM

Este articulo presenta la metodologia empleada para dotar a nuestros alumnos, alumnos de la asignatura Diseno con Microcontroladores en el contexto del Master de Ingenieria de Computadores y Redes de la Universidad de Sevilla, de la capacidad de disenar y desarrollar sistemas empotrados integrados con interfaz grafica. En la actualidad se ha extendido el uso de microcontroladores ARM de 32 bits, existiendo diversas familias y arquitecturas, ademas se han popularizado las pantallas basadas en la tecnologia OLED. Debido a esta popularidad, resulta muy interesante la combinacion de ambos dispositivos para la formacion de la integracion de sistemas empotrados con interfaz grafica, ya que asi se puede proporcionar una interactuacion entre este tipo de sistemas y los usuarios. Este modulo didactico ha sido bien acogido por el alumnado, ya que el manejo de los displays OLED les resulta muy atractivo y se plantean las tecnicas de programacion necesarias para la implementacion de interfaces graficas mas complejas.

Explainable Deep Learning System for Custom Report Generation in Breast Cancer Histology

Abstract Breast cancer is the most lethal type of cancer among women, one of the causes can be due to the lack of professionals to evaluate the results of medical images in time (Sharafaddini et al. Multimed Tools Appl, 1–112 2024). This problem is even greater in developing countries. In recent years, diagnostic tools based on artificial intelligence techniques have been developed to improve diagnosis time and results. In this work, we present a system that analyzes histopathological images obtained from breast tissue biopsies to design a classification system that distinguishes between benign and malignant tissue. To demonstrate that the proposed work is robust, multiple alternatives and combinations are studied to obtain the best cases. Finally, we compare the proposed approach with previous works. Furthermore, the developed system integrates explainable artificial intelligence techniques to produce a report to the physician, including a heat map with the areas the system has determined to be essential for classification.

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, while it is well known that ECG characteristics change in athletes and elite players as a result of their intense long-term physical training. 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 interpretation in athletes. 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), it comprises 163 10-s 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-s 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). The generated dataset serves as a valuable resource for research in sports cardiology and cardiac health. Its potential for reuse encompasses:1.International Comparison: Enabling cross-regional comparisons of cardiac characteristics among elite football players, enriching international studies.2.ML Model Development: Facilitating the development and refinement of machine learning models for arrhythmia detection, serving as a benchmark dataset.3.Validation of Diagnostic Methods: Allowing the validation of automatic diagnostic methods, contributing to enhanced accuracy in detecting cardiac conditions.4.Research in Sports Cardiology: Supporting future investigations into specific cardiac adaptations in elite athletes and their relation to cardiovascular health.5.Reference for Athlete Protection Policies: Influencing athlete protection policies by providing data on cardiac health and suggesting guidelines for medical assessments.6.Health Professionals Training: Serving as a training resource for health professionals interested in interpreting ECGs in sports contexts.7.Tool and Application Development: Facilitating the development of tools and applications related to the visualization, simulation and analysis of ECG signals in athletes.

Gait-Based Parkinson’s Disease Detection Using Recurrent Neural Networks for Wearable Systems

Parkinson’s disease is one of the neurodegenerative conditions that has seen a significant increase in prevalence in recent decades. The lack of specific screening tests and notable disease biomarkers, combined with the strain on healthcare systems, leads to delayed detection of the disease, which worsens its progression. The development of diagnostic support tools can support early detection and facilitate timely intervention. The ability of Deep Learning algorithms to identify complex features from clinical data has proven to be a promising approach in various medical domains as support tools. In this study, we present an investigation of different architectures based on Gated Recurrent Neural Networks to assess their effectiveness in identifying subjects with Parkinson’s disease from gait records. Models with Long-Short term Memory (LSTM) and Gated Recurrent Unit (GRU) layers were evaluated. Performance results reach competitive effectiveness values with the current state-of-the-art accuracy (up to 93.75% (average ± SD: 86 ± 5%)), simplifying computational complexity, which represents an advance in the implementation of executable screening and diagnostic support tools in systems with few computational resources in wearable devices.

Performance study of synthetic AER generation on CPUs for Real-Time Video based on Spikes

Address-Event-Representation (AER) is a neuromorphic interchip communication protocol that allows for real-time virtual massive connectivity between huge number neurons located on different chips. When building multi-chip muti-layered AER systems it is absolutely necessary to have a computer interface that allows (a) to read AER interchip traffic into the computer and visualize it on screen, and (b) convert conventional frame-based video stream in the computer into AER and inject it at some point of the AER structure. This is necessary for test and debugging of complex AER systems. Previous work presented several software methods for converting digital frames into AER format. Those methods were not feasible for real-time conversion those days because the processor performance was insufficient. Nowadays, Multi-core processor architectures and cache hierarchies have evolved and the performance is much better than Pentium 4 Mobile of those years. In this paper we study frame-to-AER methods for real-time video applications (40ms per frame) using modern processor architectures, compilers, and processors oriented for stand-alone applications (mini-PC processors)

Disease screening using Artificial Intelligence

Abstract This work presents a disease detection classifier based on symptoms encoded by their severity. This model is presented as part of the solution to the saturation of the healthcare system, aiding in the initial screening stage. An open-source dataset was used, which underwent pre-processing and served as the data source to train and test various machine learning models, including SVM, Random Forests, KNN, and ANNs. A 3-phase optimization process is developed to obtain the best classifier: first, the dataset is pre-processed; secondly, a grid search is performed with several hypermarameters variations to each classifier; and, finally, the best models obtained are subjected to an additional filtering processes. The best-results model, selected based on the performance and the execution time, is a KNN with 2 neighbors, which achieves an accuracy and F1 score of over 98\%. These results demonstrate the effectiveness and improvement of the evaluated models compared to previous studies, particularly in terms of accuracy. Although the ANN model has a longer execution time compared to KNN, it is retained in the work due to its potential to handle more complex datasets in a real clinical context.

Embedded neural network for real-time animal behavior classification

ECG Pre-processing and Feature Extraction Tool for Intelligent Simulation Systems

Sudden cardiac death events and fatal cardiac problems are a field of vital importance for physicians working with elite athletes. For this reason, it is common to periodically perform cardiac monitoring with professional ECG devices to detect certain risk markers. As these doctors often work with many athletes (as is the case with professional football teams), an artificial intelligence-based system would help mass screening and allow these exams to be carried out more regularly. Because physicians often evaluate the printed reports generated by ECG devices, few manufacturers provide powerful and configurable software tools. Moreover, for teaching purposes, a simulation tool that would allow working with previously collected ECG files would be very useful. In this paper, we present a software tool to be used with General Electric CardioSoft 12SL electrocardiograph. This tool allows importing the XML files generated by this device, perform a manual or automatic signal filtering process and PQRST peak detection, and finally generate a customisable report as a CSV file containing the features obtained after signal analysis. This pre-processed information can be used as input of ECG simulators and in artificial intelligence systems to develop diagnostic support systems.

Dual Machine-Learning System to Aid Glaucoma Diagnosis Using Disc and Cup Feature Extraction

Glaucoma is a degenerative disease that affects vision, causing damage to the optic nerve that ends in vision loss. The classic techniques to detect it have undergone a great change since the intrusion of machine learning techniques into the processing of eye fundus images. Several works focus on training a convolutional neural network (CNN) by brute force, while others use segmentation and feature extraction techniques to detect glaucoma. In this work, a diagnostic aid tool to detect glaucoma using eye fundus images is developed, trained and tested. It consists of two subsystems that are independently trained and tested, combining their results to improve glaucoma detection. The first subsystem applies machine learning and segmentation techniques to detect optic disc and cup independently, combine them and extract their physical and positional features. The second one applies transfer learning techniques to a pre-trained CNN to detect glaucoma through the analysis of the complete eye fundus images. The results of both systems are combined to discriminate positive cases of glaucoma and improve final detection. The results show that this system achieves a higher classification rate than previous works. The system also provides information on the basis for the proposed diagnosis suggestion that can help the ophthalmologist to accept or modify it.

On the feature extraction process in machine learning. An experimental study about guided versus non-guided process in falling detection systems

Falls are current events that can lead to severe injuries and even accidental deaths among the population, especially the elderly. Since them usually live alone and their contact with other people has decreased since pandemic, recent years studies have focused on automatic fall detection systems with wearable devices using machine learning algorithms. Overall, and according to other works, these systems can be classified as non-guided, if the machine learning model directly uses raw data without feature extraction, or as guided systems, if a previous step of feature extraction is needed to reduce complexity of the algorithm. However, no recommendations are made in the literature on which system could be more advantageous for detecting fall events. Therefore, in this work, a detailed comparison between both types of systems is carried out, using the same process for different machine learning models in order to obtain an accurate classification of activities of daily living, falling risks, and falls. This process includes the optimization of models' hyperparameters to obtain the best classifiers, followed by an assessment using common evaluation metrics, confusion matrices, ROC curves and execution times. Results show a better classification of models' three classes for the non-guided models. However, the guided models show more stable metrics and lower computational load.

Worker’s physical fatigue classification using neural networks

Physical fatigue is not only an indication of the user's physical condition and/or need for sleep or rest, but can also be a significant symptom of various diseases. This fatigue affects the performance of workers in jobs that involve some continuous physical activity, and is the cause of a large proportion of accidents at work. The physical fatigue is commonly measured by the perceived exertion (RPE). Many previous studies have attempted to continuously monitor workers in order to detect the level of fatigue and prevent these accidents, but most have used invasive sensors that are difficult to place and prevent the worker from performing their tasks correctly. Other works use activity measurement sensors such as accelerometers, but the large amount of information obtained is difficult to analyse in order to extract the characteristics of each fatigue state. In this work, we use a dataset that contains data from inertial sensors of several workers performing various activities during their working day, labelled every 10 min based on their level of fatigue using questionnaires and the Borg fatigue scale. Applying Machine Learning techniques, we design, develop and test a system based on a neural network capable of classifying the variation of fatigue caused by the physical activity collected every 10 min; for this purpose, a feature extraction is performed after the time decomposition done with the Discrete Wavelet Transform (DWT). The results show that the proposed system has an accuracy higher than 92% for all the cases, being viable for its application in the proposed scenario.

Live demonstration — Multilayer spiking neural network for audio samples classification using SpiNNaker

In this demonstration we present a spiking neural network architecture for audio samples classification using SpiNNaker. The network consists of different leaky integrate-and-fire neuron layers. The connections between them are trained using firing rate based algorithms. Tests use sets of pure tones with frequencies that range from 130.813 to 1396.91 Hz. Audio signals coming from the computer are converted to spikes using a Neuromorphic Auditory Sensor and, after that, this information is sent to the SpiNNaker board through a PCB that translates from AER to 2-of-7 protocol. The classification output obtained in the spiking neural network deployed on SpiNNaker is then shown in the computer screen. Different levels of random noise are added to the original audio signals in order to test the robustness of the classification system.

Performance evaluation over HW/SW co-design SoC memory transfers for a CNN accelerator

FPGA devices have demonstrated to be one of the most popular prototyping solutions. N anoelectronics is one of the fields that can prototype its architectures in these devices. Many FPGA vendors have recently included embedded processors in their devices, such as Xilinx with ARM -Cortex A cores, together with programmable logic cells. These devices are known as Programmable System on Chip (PSoC). Their ARM cores (embedded in the processing system or PS) communicates with the programmable logic cells (PL) using ARM-standard interface buses. ARM proposed the Advanced Microcontroller Bus Architecture (AMBA) as an open-standard. Its third generation included the Advanced eXtensible Interface (AXI) to reach higher performances. In this paper we analyse the performance of exhaustive data transfers between PS and PL for a Xilinx Zynq FPGA in a co-design real scenario for Convolutional Neural Networks (CNN) accelerator. This CNN accelerator processes, in dedicated hardware, a stream of visual information from a neuromorphic visual sensor for classification. In the PS side, a Linux operating system is running, which recollects visual events from the neuromorphic sensor into a normalized frame, and then it transfers these frames to the CNN accelerator of multi-layered CNNs; and read results, using an AXI-DMA bus in a per-layer way. As these kind of accelerators try to process information as quick as possible, data bandwidth becomes critical. Maintaining a good balanced data throughput rate requires some considerations, such as data partitioning techniques to balance RX and TX transfers, or different transfer management techniques: polling versus dedicated interrupt-based kernel-level driver. For longer enough packets, the kernel-level driver solution improves global computation timings in a CNN classification example. Kernel-level driver ensures safer solutions and enables OS tasks, scheduling for better computation distribution.