Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 20
Encontrada(s) 1 página(s)

Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Applying machine learning for healthcare: A case study on cervical pain assessment with motion capture

  • de la Torre, Juan
  • Marin, Javier
  • Ilarri, Sergio
  • Marin, José J.
Given the exponential availability of data in health centers and the massive sensorization that is expected, there is an increasing need to manage and analyze these data in an effective way. For this purpose, data mining (DM) and machine learning (ML) techniques would be helpful. However, due to the specific characteristics of the field of healthcare, a suitable DM and ML methodology adapted to these particularities is required. The applied methodology must structure the different stages needed for data-driven healthcare, from the acquisition of raw data to decision-making by clinicians, considering the specific requirements of this field. In this paper, we focus on a case study of cervical assessment, where the goal is to predict the potential presence of cervical pain in patients affected with whiplash diseases, which is important for example in insurance-related investigations. By analyzing in detail this case study in a real scenario, we show how taking care of those particularities enables the generation of reliable predictive models in the field of healthcare. Using a database of 302 samples, we have generated several predictive models, including logistic regression, support vector machines, k-nearest neighbors, gradient boosting, decision trees, random forest, and neural network algorithms. The results show that it is possible to reliably predict the presence of cervical pain (accuracy, precision, and recall above 90%). We expect that the procedure proposed to apply ML techniques in the field of healthcare will help technologists, researchers, and clinicians to create more objective systems that provide support to objectify the diagnosis, improve test treatment efficacy, and save resources.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Semantic traffic sensor data: The TRAFAIR experience

  • Desimoni, F.
  • Ilarri, S.
  • Po, L.
  • Rollo, F.
  • Trillo-Lado, R.
Modern cities face pressing problems with transportation systems including, but not limited to, traffic congestion, safety, health, and pollution. To tackle them, public administrations have implemented roadside infrastructures such as cameras and sensors to collect data about environmental and traffic conditions. In the case of traffic sensor data not only the real-time data are essential, but also historical values need to be preserved and published. When real-time and historical data of smart cities become available, everyone can join an evidence-based debate on the city''s future evolution. The TRAFAIR (Understanding Traffic Flows to Improve Air Quality) project seeks to understand how traffic affects urban air quality. The project develops a platform to provide real-time and predicted values on air quality in several cities in Europe, encompassing tasks such as the deployment of low-cost air quality sensors, data collection and integration, modeling and prediction, the publication of open data, and the development of applications for end-users and public administrations. This paper explicitly focuses on the modeling and semantic annotation of traffic data. We present the tools and techniques used in the project and validate our strategies for data modeling and its semantic enrichment over two cities: Modena (Italy) and Zaragoza (Spain). An experimental evaluation shows that our approach to publish Linked Data is effective.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Otros (Other)

Special Issue on Smart Data and Semantics in a Sensor World

  • Ilarri, S.
  • Po, L.R.
  • Trillo-Lado, R.
Introduction Since its first inception in 2001, the application of the Semantic Web [1, 2] has carried out an extensive use of ontologies [3–5], reasoning, and semantics in diverse fields, such as Information Integration, Software Engineering, Bioinformatics, eGovernment, eHealth, and social networks. This widespread use of ontologies has led to an incredible advance in the development of techniques to manipulate, share, reuse, and integrate information across heterogeneous data sources. In recent years, the growth of the IoT (Internet of Things) required to face the challenges of “Big Data” [6–10]. The cost of sensors is decreasing, while their use is expanding. Moreover, the use of multiple personal smart devices is an emerging trend and all of them can embed sensors to monitor the surrounding environment. Therefore, the number of available sensors is exploding. On the one hand, the flows of sensor data are massive and continuous, and the data could be obtained in real time or with a delay of just a few seconds. Then, the volume of sensor data is increasing continuously every day. On the other hand, the variety of data being generated is also increasing, due to plenty of different devices and different measures to record. There are many kinds of structured and unstructured sensor data in diverse formats. Moreover, data veracity, which is the degree of accuracy or truthfulness of a data set, is an important aspect to consider. In the context of sensor data, it represents the trustworthiness of the data source and the processing of data. The need for more accurate and reliable data was always declared, but often overlooked for the sake of larger and cheaper...
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Contribución de congreso (ConferenceOutput)

Extending the Reach of the Point-To-Set Principle

  • Lutz, Jack H
  • Lutz, Neil
  • Mayordomo Cámara, Elvira
The point-to-set principle of J. Lutz and N. Lutz (2018) has recently enabled the theory of computing to be used to answer open questions about fractal geometry in Euclidean spaces Rn. These are classical questions, meaning that their statements do not involve computation or related aspects of logic. In this paper we extend the reach of the point-to-set principle from Euclidean spaces to arbitrary separable metric spaces X. We first extend two fractal dimensions – computability-theoretic versions of classical Hausdorff and packing dimensions that assign dimensions dim(x) and Dim(x) to individual points x ∈ X – to arbitrary separable metric spaces and to arbitrary gauge families. Our first two main results then extend the point-to-set principle to arbitrary separable metric spaces and to a large class of gauge families. We demonstrate the power of our extended point-to-set principle by using it to prove new theorems about classical fractal dimensions in hyperspaces. (For a concrete computational example, the stages E0, E1, E2, . . . used to construct a self-similar fractal E in the plane are elements of the hyperspace of the plane, and they converge to E in the hyperspace.) Our third main result, proven via our extended point-to-set principle, states that, under a wide variety of gauge families, the classical packing dimension agrees with the classical upper Minkowski dimension on all hyperspaces of compact sets. We use this theorem to give, for all sets E that are analytic, i.e., Σ1 1, a tight bound on the packing dimension of the hyperspace of E in terms of the packing dimension of E itself.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Una generalización del teorema de proyección de Marstrand

  • Mayordomo Cámara, Elvira
Marstrand cuantificó la dimensión de Hausdorff de (casi todas) las proyecciones ortogonales de un conjunto analítico. Recientemente este resultado ha sido generalizado por Lutz y Stull para algunos casos no analíticos, utilizando en su demostración la dimensión efectiva y el principio de punto a conjunto. La dimensión efectiva tiene sus raíces en la computabilidad y la teoría de la información y en este artículo introducimos sus principales ideas a través de la demostración de Lutz y Stull.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Towards MANET-based recommender systems for open facilities

  • Dunkel, Jürgen
  • Hermoso, Ramón
Nowadays, most recommender systems are based on a centralized architecture, which can cause crucial issues in terms of trust, privacy, dependability, and costs. In this paper, we propose a decentralized and distributed MANET-based (Mobile Ad-hoc NETwork) recommender system for open facilities. The system is based on mobile devices that collect sensor data about users locations to derive implicit ratings that are used for collaborative filtering recommendations. The mechanisms of deriving ratings and propagating them in a MANET network are discussed in detail. Finally, extensive experiments demonstrate the suitability of the approach in terms of different performance metrics. © 2021, The Author(s).
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

A phylogenetic study of covid-19 data from Aragon and Catalonia over a year: learning bioinformatics during a world pandemic

  • Peña, F
  • García Díaz, A.
  • Gómez González, A.
  • Martín Marcos, A.
  • Romeo, A.
  • Sánchez Aquilué, J. M.
  • Vallés, A.
  • Mayordomo, E.
Presentamos aquí un análisis de variabilidad durante un año del virus SARS-CoV-2 en el área de Aragón, comparándolo con la vecina Cataluña y centrándonos en la variante del Reino Unido. La investigación se realizó dentro de un curso de Bioinformática. Se extrajeron tanto una evaluación detallada sobre el valor de esta metodología como conclusiones novedosas de la investigación.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

VoxelMorph based Normalization in the Prediction of Stable VS Progressive MCI Conversion with Convolutional Neural Networks

  • Sierra Tome, Daniel
  • Júlvez, Ubaldo Ramón
  • Hernández, Mónica
  • Mayordomo, Elvira
El presente trabajo tiene como objetivo estudiar el efecto de la normalización en el sistema de clasificación de Spasov2019 [2] para el problema de deterioro cognitivo leve estable vs progresivo (sMCI vs pMCI), cuando esta normalización proviene de un algoritmo de deep-learning. Se ha elegido VoxelMorph [8], por ser uno de los algoritmos más utilizados como benchmark desde la introducción del deep-learning en el registro deformable de imágenes médicas. Se realizará una comparativa entre los métodos de registro difeomorfo tradicionales y VoxelMorph para este problema.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Algorithmic Dimensions, the Point-To-Set Principles, and the Complexity of Oracles

Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

I am all EARS: Using open data and knowledge graph embeddings for music recommendations

  • Bertram, Niels
  • Dunkel, Jürgen
  • Hermoso, Ramón
Music streaming platforms offer music listeners an overwhelming choice of music. Therefore, users of streaming platforms need the support of music recommendation systems to find music that suits their personal taste. Currently, a new class of recommender systems based on knowledge graph embeddings promises to improve the quality of recommendations, in particular to provide diverse and novel recommendations. This paper investigates how knowledge graph embeddings can improve music recommendations. First, it is shown how a collaborative knowledge graph can be derived from open music data sources. Based on this knowledge graph, the music recommender system EARS (knowledge graph Embedding-based Artist Recommender System) is presented in detail, with particular emphasis on recommendation diversity and explainability. Finally, a comprehensive evaluation with real-world data is conducted, comparing of different embeddings and investigating the influence of different types of knowledge.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Explainable artificial intelligence toward usable and trustworthy computer-aided diagnosis of multiple sclerosis from Optical Coherence Tomography

  • Hernandez, Monica
  • Ramon-Julvez, Ubaldo
  • Vilades, Elisa
  • Cordon, Beatriz
  • Mayordomo, Elvira
  • Garcia-Martin, Elena
Background: Several studies indicate that the anterior visual pathway provides information about the dynamics of axonal degeneration in Multiple Sclerosis (MS). Current research in the field is focused on the quest for the most discriminative features among patients and controls and the development of machine learning models that yield computer-aided solutions widely usable in clinical practice. However, most studies are conducted with small samples and the models are used as black boxes. Clinicians should not trust machine learning decisions unless they come with comprehensive and easily understandable explanations. Materials and methods: A total of 216 eyes from 111 healthy controls and 100 eyes from 59 patients with relapsing-remitting MS were enrolled. The feature set was obtained from the thickness of the ganglion cell layer (GCL) and the retinal nerve fiber layer (RNFL). Measurements were acquired by the novel Posterior Pole protocol from Spectralis Optical Coherence Tomography (OCT) device. We compared two black-box methods (gradient boosting and random forests) with a glass-box method (explainable boosting machine). Explainability was studied using SHAP for the black-box methods and the scores of the glass-box method. Results: The best-performing models were obtained for the GCL layer. Explainability pointed out to the temporal location of the GCL layer that is usually broken or thinning in MS and the relationship between low thickness values and high probability of MS, which is coherent with clinical knowledge.Conclusions: The insights on how to use explainability shown in this work represent a first important step toward a trustworthy computer-aided solution for the diagnosis of MS with OCT.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Extending the reach of the point-to-set principle

  • Lutz, Jack H.
  • Lutz, Neil
  • Mayordomo, Elvira
The point-to-set principle of J. Lutz and N. Lutz (2018) has recently enabled the theory of computing to be used to answer open questions about fractal geometry in Euclidean spaces [Rn]. These are classical questions, meaning that their statements do not involve computation or related aspects of logic. In this paper we extend the reach of the point-to-set principle from Euclidean spaces to arbitrary separable metric spaces X. We first extend two algorithmic dimensions—computability-theoretic versions of classical Hausdorff and packing dimensions that assign dimensions [dim(x)] and [Dim(x)] to individual points [x E X] —to arbitrary separable metric spaces and to arbitrary gauge families. Our first two main results then extend the point-to-set principle to arbitrary separable metric spaces and to a large class of gauge families. We demonstrate the power of our extended point-to-set principle by using it to prove new theorems about classical fractal dimensions in hyperspaces.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Explainable AI toward understanding the performance of the top three TADPOLE Challenge methods in the forecast of Alzheimer’s disease diagnosis

  • Hernandez, Mónica
  • Ramón-Julvez, Ubaldo
  • Ferraz, Francisco
The Alzheimer0s Disease Prediction Of Longitudinal Evolution (TADPOLE) Challenge is the most comprehensive challenge to date with regard to the number of subjects, considered features, and challenge participants. The initial objective of TADPOLE was the identification of the most predictive data, features, and methods for the progression of subjects at risk of developing Alzheimer0s. The challenge was successful in recognizing tree-based ensemble methods such as gradient boosting and random forest as the best methods for the prognosis of the clinical status in Alzheimer’s disease (AD). However, the challenge outcome was limited to which combination of data processing and methods exhibits the best accuracy; hence, it is difficult to determine the contribution of the methods to the accuracy. The quantification of feature importance was globally approached by all the challenge participant methods. In addition, TADPOLE provided general answers that focused on improving performance while ignoring important issues such as interpretability. The purpose of this study is to intensively explore the models of the top three TADPOLE Challenge methods in a common framework for fair comparison. In addition, for these models, the most meaningful features for the prognosis of the clinical status of AD are studied and the contribution of each feature to the accuracy of the methods is quantified. We provide plausible explanations as to why the methods achieve such accuracy, and we investigate whether the methods use information coherent with clinical knowledge. Finally, we approach these issues through the analysis of SHapley Additive exPlanations (SHAP) values, a technique that has recently attracted increasing attention in the field of explainable artificial intelligence (XAI). Copyright: © 2022 Hernandez et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Dimension and the structure of complexity classes

  • Lutz, Jack H.
  • Lutz, Neil
  • Mayordomo, Elvira
We prove three results on the dimension structure of complexity classes. The Point-to-Set Principle, which has recently been used to prove several new theorems in fractal geometry, has resource-bounded instances. These instances characterize the resource-bounded dimension of a set X of languages in terms of the relativized resource-bounded dimensions of the individual elements of X, provided that the former resource bound is large enough to parametrize the latter. Thus for example, the dimension of a class X of languages in EXP is characterized in terms of the relativized p-dimensions of the individual elements of X. Every language that is =mP-reducible to a p-selective set has p-dimension 0, and this fact holds relative to arbitrary oracles. Combined with a resource-bounded instance of the Point-to-Set Principle, this implies that if NP has positive dimension in EXP, then no quasipolynomial time selective language is =mP-hard for NP. If the set of all disjoint pairs of NP languages has dimension 1 in the set of all disjoint pairs of EXP languages, then NP has positive dimension in EXP.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Insights into traditional Large Deformation Diffeomorphic Metric Mapping and unsupervised deep-learning for diffeomorphic registration and their evaluation

  • Hernandez, Monica
  • Ramon Julvez, Ubaldo
This paper explores the connections between traditional Large Deformation Diffeomorphic Metric Mapping methods and unsupervised deep-learning approaches for non-rigid registration, particularly emphasizing diffeomorphic registration. The study provides useful insights and establishes connections between the methods, thereby facilitating a profound understanding of the methodological landscape. The methods considered in our study are extensively evaluated in T1w MRI images using traditional NIREP and Learn2Reg OASIS evaluation protocols with a focus on fairness, to establish equitable benchmarks and facilitate informed comparisons. Through a comprehensive analysis of the results, we address key questions, including the intricate relationship between accuracy and transformation quality in performance, the disentanglement of the influence of registration ingredients on performance, and the determination of benchmark methods and baselines. We offer valuable insights into the strengths and limitations of both traditional and deep-learning methods, shedding light on their comparative performance and guiding future advancements in the field.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

EPDiff-JF-Net: Adjoint Jacobi Fields for Diffeomorphic Registration Networks

  • Ramón-Júlvez, Ubaldo
  • Hernández, Mónica
  • Mayordomo, Elvira
Large Deformation Diffeomorphic Metric Mapping (LDDMM) is a well-established diffeomorphic registration method, a critical step for many medical imaging applications. Of special interest are its geodesic shooting variants, which enable statistical shape analysis from transformations usable in computational anatomy studies. This paper introduces a novel deep learning-based unsupervised approach for diffeomorphic image registration called EPDiff-JF-Net. Our method predicts an initial velocity field, performs geodesic shooting to obtain the corresponding path of diffeomorphisms, and utilizes an adjoint Jacobi fields layer to calculate the relevant gradients for training from parallel transport along the geodesic. The model is trained in a fully unsupervised end-to-end manner, with no requirement of ground-truth in the training loss. Experimental results on 3D brain MRI datasets demonstrate the effectiveness of EPDiff-JF-Net, outperforming EPDiff-based LDDMM and deep learning methods while significantly reducing computation time.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

EPDIFF-JF-NET: Adjoint Jacobi Fields for Diffeomorphic Registration Networks

  • Ramón Júlvez, Ubaldo
  • Hernández Giménez, Mónica
  • Mayordomo Cámara, Elvira
This paper presents a deep learning unsupervisedapproach for diffeomorphic image registrationcalled EPDiff-JF-Net. We propose a novel paralleltransport layer to compute the gradients necessaryfor training with adjoint Jacobi fields. We test ourmethod on two independent brain MRI datasets andobtain state-of-the-art results.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Traffic and Pollution Modelling for Air Quality Awareness: An Experience in the City of Zaragoza

  • Ilarri, S.
  • Trillo-Lado, R.
  • Marrodán, L.
Air pollution due to the presence of small particles and gases in the atmosphere is a major cause of health problems. In urban areas, where most of the population is concentrated, traffic is a major source of air pollutants (such as nitrogen oxides or NO x and carbon monoxide or CO). Therefore, for smart cities, carrying out an adequate traffic monitoring is a key issue, since it can help citizens to make better decisions and public administrations to define appropriate policies. Thus, citizens could use these data to make appropriate mobility decisions. In the same way, a city council can exploit the collected data for traffic management and for the establishment of suitable traffic policies throughout the city, such as restricting the traffic flow in certain areas. For this purpose, a suitable modelling approach that provides the estimated/predicted values of pollutants at each location is needed. In this paper, an approach followed to model traffic flow and air pollution dispersion in the city of Zaragoza (Spain) is described. Our goal is to estimate the air quality in different areas of the city, to raise awareness and help citizens to make better decisions; for this purpose, traffic data play an important role. In more detail, the proposal presented includes a traffic modelling approach to estimate and predict the amount of traffic at each road segment and hour, by combining historical measurements of real traffic of vehicles and the use of the SUMO traffic simulator on real city roadmaps, along with the application of a trajectory generation strategy that complements the functionalities of SUMO (for example, SUMO’s calibrators). Furthermore, a pollution modelling approach is also provided, to estimate the impact of traffic flows in terms of pollutants in the atmosphere: an R package called Vehicular Emissions INventories (VEIN) is used to estimate the amount of NO x generated by the traffic flows by taking into account the vehicular fleet composition (i.e., the types of vehicles, their size and the type of fuel they use) of the studied area. Finally, considering this estimation of NO x, a service capable of offering maps with the prediction of the dispersion of these atmospheric pollutants in the air has been established, which uses the Graz Lagrangian Model (GRAL) and takes into account the meteorological conditions and morphology of the city. The results obtained in the experimental evaluation of the proposal indicate a good accuracy in the modelling of traffic flows, whereas the comparison of the prediction of air pollutants with real measurements shows a general underestimation, due to some limitations of the input data considered. In any case, the results indicate that this first approach can be used for forecasting the air pollution within the city.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Normality, Relativization, and Randomness

  • Calvert, Wesley
  • Gruner, Emma
  • Mayordomo, Elvira
  • Turetsky, Daniel
  • Villano, Java Darleen
Normal numbers were introduced by Borel and later proven to be a weak notion of algorithmic randomness. We introduce here a natural relativization of normality based on generalized number representation systems. We explore the concepts of supernormal numbers that correspond to semicomputable relativizations, and that of highly normal numbers in terms of computable ones. We prove several properties of these new randomness concepts. Both supernormality and high normality generalize Borel absolute normality. Supernormality is strictly between 2-randomness and effective dimension 1, while high normality corresponds exactly to sequences of computable dimension 1 providing a more natural characterization of this class.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Contribución de congreso (ConferenceOutput)

A Point to Set Principle for Finite-State Dimension

  • Mayordomo, Elvira
Effective dimension has proven very useful in geometric measure theory through the point-to-set principle [9] that characterizes Hausdorff dimension by relativized effective dimension. Finite-state dimension is the least demanding effectivization in this context [3] that among other results can be used to characterize Borel normality [2]. In this paper we prove a characterization of finite-state dimension in terms of information content of a real number at a certain precision. We then use this characterization to give a robust concept of relativized normality and prove a finite-state dimension point-to-set principle. We finish with an open question on the equidistribution properties of relativized normality.