DI-Artículos

URI permanente para esta colecciónhttps://hdl.handle.net/10953/218

Examinar

Mostrando 1 - 20 de 115

Fuzzy intelligent system for patients with preeclampsia in wearable devices
(Wiley, 2017-10-12) Espinilla-Estévez, Macarena; Medina-Quero, Javier; García-Fernández, Ángel Luis; Campaña, Sixto; Londoño, Jorge
Preeclampsia affects from 5% to 14% of all pregnant women and is responsible for about 14% of maternal deaths per year in the world. This paper is focused on the use of a decision analysis tool for the early detection of preeclampsia in women at risk. This tool applies a fuzzy linguistic approach implemented in a wearable device. In order to develop this tool, a real dataset containing data of pregnant women with high risk of preeclampsia from a health center has been analyzed, and a fuzzy linguistic methodology with two main phases is used. Firstly, linguistic transformation is applied to the dataset to increase the interpretability and flexibility in the analysis of preeclampsia. Secondly, knowledge extraction is done by means of inferring rules using decision trees to classify the dataset. The obtained linguistic rules provide understandable monitoring of preeclampsia based on wearable applications and devices. Furthermore, this paper not only introduces the proposed methodology, but also presents a wearable application prototype which applies the rules inferred from the fuzzy decision tree to detect preeclampsia in women at risk. The proposed methodology and the developed wearable application can be easily adapted to other contexts such as diabetes or hypertension.
Generating implicit object fragment datasets for machine learning
(ELSEVIER, 2024-10-15) Fuertes Garcia, José Manuel
One of the primary challenges inherent in utilizing deep learning models is the scarcity and accessibility hurdles associated with acquiring datasets of sufficient size to facilitate effective training of these networks. This is particularly significant in object detection, shape completion, and fracture assembly. Instead of scanning a large number of real-world fragments, it is possible to generate massive datasets with synthetic pieces. However, realistic fragmentation is computationally intensive in the preparation (e.g., pre-factured models) and generation. Otherwise, simpler algorithms such as Voronoi diagrams provide faster processing speeds at the expense of compromising realism. In this context, it is required to balance computational efficiency and realism. This paper introduces a GPU-based framework for the massive generation of voxelized fragments derived from high-resolution 3D models, specifically prepared for their utilization as training sets for machine learning models. This rapid pipeline enables controlling how many pieces are produced, their dispersion and the appearance of subtle effects such as erosion. We have tested our pipeline with an archaeological dataset, producing more than 1M fragmented pieces from 1,052 Iberian vessels (Github). Although this work primarily intends to provide pieces as implicit data represented by voxels, triangle meshes and point clouds can also be inferred from the initial implicit representation. To underscore the unparalleled benefits of CPU and GPU acceleration in generating vast datasets, we compared against a realistic fragment generator that highlights the potential of our approach, both in terms of applicability and processing time. We also demonstrate the synergies between our pipeline and realistic simulators, which frequently cannot select the number and size of resulting pieces. To this end, a deep learning model was trained over realistic fragments and our dataset, showing similar results.
Change Detection in Point Clouds Using 3D Fractal Dimension
(MDPI, 2024-03-16) Fuertes Garcia, José Manuel
The management of large point clouds obtained by LiDAR sensors is an important topic in recent years due to the widespread use of this technology in a wide variety of applications and the increasing volume of data captured. One of the main applications of LIDAR systems is the study of the temporal evolution of the real environment. In open environments, it is important to know the evolution of erosive processes or landscape transformation. In the context of civil engineering and urban environments, it is useful for monitoring urban dynamics and growth, and changes during the construction of buildings or infrastructure facilities. The main problem with change detection (CD) methods is erroneous detection due to precision errors or the use of different capture devices at different times. This work presents a method to compare large point clouds, based on the study of the local fractal dimension of point clouds at multiple scales. Our method is robust in the presence of environmental and sensor factors that produce abnormal results with other methods. Furthermore, it is more stable than others in cases where there is no significant displacement of points but there is a local alteration of the structure of the point cloud. Furthermore, the precision can be adapted to the complexity and density of the point cloud. Finally, our solution is faster than other CD methods such as distance-based methods and can run at O(1) under some conditions, which is important when working with large datasets. All these improvements make the proposed method more suitable than the others to solve complex problems with LiDAR data, such as storage, time series data management, visualization, etc.
Human femur fracture by mechanical compression: Towards the repeatability of bone fracture acquisition
(ELSEVIER, 2023-09) Pérez-Cano, Francisco Daniel; Jiménez-Pérez, Juan Roberto; Molina-Viedma, Ángel Jesús; López-Alba, Elías; Luque-Luque, Adrián; Delgado-Martínez, Alberto Damián; Díaz-Garrido, Francisco Alberto; Jiménez-Delgado, Juan José
The increase in life expectancy combined with greater bone fragility over the years is causing a rise in the bone fracture cases. Femur fractures are the most important due to their high mortality rate. This multidisciplinary work is carried out in this context and focuses on the experimental reproduction of human femur fractures by compression. We describe a sequence of steps supervised by orthopaedic surgeons for the correct arrangement of specimens on the system set up to perform the experiment. The device applies force by compression until the human bone is fractured. All tests performed have been monitored and evaluated from different knowledge perspectives. The results obtained have demonstrated the repeatability of the fracture type in a controlled environment as well as identifying the main features involved in this process. In addition, the fractured bones have been digitized to analyze the fracture zone to recreate and evaluate future simulations.
Subdivision Strategies for Bone Models: A Comprehensive Analysis of Geometric and Visual Quality
(IEEE, 2024-06-28) Parra-Cabrera, Gema; Pérez-Cano, Francisco Daniel; Jiménez-Delgado, Juan José
Bone fracture modeling is a major challenge in medical image analysis and simulation, requiring accurate strategies to faithfully represent complex fracture patterns. This study conducts a comprehensive analysis of three subdivision strategies: approximation, triangulation, and a hybrid approach. The approximation method preserves mesh topology but exhibits visual inconsistencies with non-horizontal fractures. Triangulation accurately represents fractures but alters mesh topology. The hybrid approach balances geometric accuracy and visual fidelity by dynamically adjusting an approximation threshold. This minimizes deviations from the original fracture pattern and maintains visual quality. Using quality metrics, we evaluate these strategies for geometric accuracy, visual fidelity, and mesh topology. Our results indicate that the hybrid approach effectively balances accuracy and visual quality, making it a promising solution for bone fracture modeling. Expert validation and quantitative metrics underscore the importance of tailored approaches for different fracture patterns. This study significantly advances computational models for clinical and research applications, offering enhanced tools for improving the accuracy and realism of bone fracture simulations, ultimately benefiting surgical planning, prosthetic design, and medical training.
An Approach to Microscopic Cortical Bone Fracture Simulation: Enhancing Clinical Replication
(Springer Nature, 2024-04-24) Pérez Cano, Francisco Daniel; Parra-Cabrera, Gema; Jiménez-Delgado, Juan José
The acquisition of bone models to perform simulations is a complex and expensive process. The hierarchical structure of bones is very complex, so that studies are mainly focused on the larger scales of bones. The objective of this work is to perform a fracture simulation at the microscale level. For this purpose, the first part of the process focuses on segmenting a bone model and selecting an area of it to generate a representation of the microstructures that make up the bone tissue from a microscopic point of view. The second part is dedicated to carry out a fracture simulation in the microscopic bone model. The developed algorithm follows a statistical approach and solves the main problems of the traditional approach (FEM) to perform a bone fracture simulation. The method returns the path that a fracture follows and demonstrates how bone structures affect fracture growth. The parameters used are configurable and can be adapted for specific cases. In addition, users can reproduce as many clinical cases as desired within seconds without have to manually segment images obtained from a microscope. The data obtained may be exported to obtain synthetic images that could be used to generate datasets for machine learning tasks or other purposes.
Exploring Fracture Patterns: Assessing Representation Methods for Bone Fracture Simulation
(MDPI, 2024-03-30) Pérez-Cano, Francisco Daniel; Parra-Cabrera, Gema; Vilchis-Torres, Ivett; Reyes-Lagos, José Javier; Jiménez-Delgado, Juan José
Fracture pattern acquisition and representation in human bones play a crucial role in medical simulation, diagnostics, and treatment planning. This article presents a comprehensive review of methodologies employed in acquiring and representing bone fracture patterns. Several techniques, including segmentation algorithms, curvature analysis, and deep learning-based approaches, are reviewed to determine their effectiveness in accurately identifying fracture zones. Additionally, diverse methods for representing fracture patterns are evaluated. The challenges inherent in detecting accurate fracture zones from medical images, the complexities arising from multifragmentary fractures, and the need to automate fracture reduction processes are elucidated. A detailed analysis of the suitability of each representation method for specific medical applications, such as simulation systems, surgical interventions, and educational purposes, is provided. The study explores insights from a broad spectrum of research articles, encompassing diverse methodologies and perspectives. This review elucidates potential directions for future research and contributes to advancements in comprehending the acquisition and representation of fracture patterns in human bone.
A compact representation of the bone fracture area. Application to fractured bones of clinical cases
(Taylor & Francis, 2021-01-05) Luque-Luque, Adrián; Jiménez-Pérez, Juan Roberto; Pérez-Cano, Francisco Daniel; Jiménez-Delgado, Juan José
The extraction of the main features of a fractured bone area enables a posterior virtual reproduction of the same fracture on other bones. The utilisation of the fracture zone for other applications is almost an unexplored field of research. Recreating a given fracture on other areas or bones can be directly applied to medical training programmes of traumatologists or to automatic bone fracture reduction algorithms. This paper is focused on the process of generating a fracture pattern taking computed tomography scans as a starting point. A set of alternative representations, for different purposes, is presented and discussed. A study of several clinical cases is analysed. Finally, the potential usages of bone fracture patterns are described.
Fracture of geometric bone models. Multiscale simulation issues
(Taylor & Francis, 2021-01-12) Pérez-Cano, Francisco Daniel; Luque-Luque, Adrián; Jiménez-Delgado, Juan José
Fracturing of osseous models is a field that allows to obtain advances in computer-assisted medical simulations, as well as to prototype fragile objects as is the case of bones. This field of study includes the use of physical features of bones. Not all fracturing methods are valid with osseous models, due to the importance of considering the hierarchical structure of the bones. In this work, we focus on the study of bone fractures at the macroscale level and on obtaining fractures with a realistic appearance. Therefore, a tool has been developed that allows the fracture of osseous models, obtained through a 3D scanning process, by the use of patterns and physics. The different approaches for the fracturing of osseous models are also shown, and discussed the need to obtain more information about the fracturing process.
Fracture pattern projection on 3D bone models as support for bone fracture simulations
(ELSEVIER, 2022-09) Parra-Cabrera, Gema; Pérez-Cano, Francisco Daniel; Jiménez-Delgado, Juan José
Background and objective: Obtaining bone models that represent certain types of fractures is limited by the need for such fractures to occur in real life and to be processed from medical images. This work aims to propose a method that starts from the design of specific fracture patterns in order to be projected on 3D geometric bone models, being prepared for their subsequent geometric fracturing. Methods: The process of projecting expert-generated fracture patterns has been approached in such a way that they contain geometrical and topological information for the subsequent fracture of the triangle mesh representing the bone model, giving information about the validity of the fracture pattern due to the design process, the validation performed, and the relationships between the fracture lines. Results: Different 3D models of long bones have been used (femur, humerus, ulna and fibula). Also, different types of fracture patterns have been created. These patterns have been used to obtain their projection on three-dimensional bones. In this study, an expert validation of the fracture patterns projected on the bone models is performed. A forensic validation of the fracture patterns used as starting point for the projection is also performed for cases in which this fracture is produced by impact, for which there is scientific evidence based on forensic analysis. This validation also supports the experts, giving them the necessary feedback to complete or modify their fracture patterns according to criteria analyzed from a forensic point of view. Conclusions: The patterns fit the bone models correctly, despite the irregularities of the bone models, and correspond to the expected projection. In addition, it provides us with a clear line of work, by using the topological information of the fracture pattern and the bone model, which allows us to establish a consistent basis for future guided fractures.
Complex fracture reduction by exact identification of the fracture zone
(ELSEVIER, 2021-08) Luque-Luque, Adrián; Pérez-Cano, Francisco Daniel; Jiménez-Delgado, Juan José
Planning of a fracture reduction is important in order to reduce the surgery time, with the consequent improvement of the recovery process. There are no fully automatic methods that solve an adequate fracture reduction without the intervention of a specialist. Usually there are parameters that must be supervised or adjusted by the specialist, in order to obtain a satisfactory reduction. Furthermore, most of the studies in the literature focus on a certain type of bone and area on it. This paper presents an approach that tries to reduce to some extent the intervention of the specialist, so that it can be closer to an automatic approach. The proposed method can be applied to a wide variety of bones and areas, based on the identification of the complete fracture zone and the use of an ICP algorithm modified to work with the distance between fragments. The cases in which it has been tested are clinical cases of real fractures obtained from CT scan. This method allows working with a wide range of fractures, as well as complex fractures or deformed fragments. Unfortunately, all possible cases and situations could not be obtained and proved, but the method can be successfully applied to cases that meet a set of characteristics. The proposed technique has been validated by experts, both visually and empirically, using a framework based on virtual reality (VR). This VR framework has allowed comparing the reduction performed by the method with a reduction made virtually by specialists. This technique has also been compared with other existing techniques, obtaining a significant improvement over these.
Generation and Validation of Osseous Fracture Patterns by Forensic Analysis
(IEEE, 2020-11-19) Jiménez-Delgado, Juan José; Parra-Cabrera, Gema; Pérez-Cano, Francisco Daniel; Luque-Luque, Adrián
This article presents a method for the generation of bone fracture patterns and their automatic validation through the use of forensic analysis. A tool has been designed that allows the generation of a fracture pattern interactively and guided by the system, based on the study of real cases of fractures. This tool assists the specialist in obtaining fracture patterns according to certain rules taken from the statistical analysis of real cases. Additionally, a parametric fracture pattern generator has been developed. This autonomous generator is able to obtain fracture patterns according to forensic case studies. Once a fracture pattern has been generated, by using one of these two methods the system also provides the validation of this pattern based on a forensic analysis, indicating the feasibility of the fracture pattern being valid and explaining the causes of its validity or non-validity. In addition, these tools provide an analysis not only of the probability of a pattern being correct, but also whether it is capable of detecting some limit patterns that could be valid if experts indicate this circumstance. The system is not closed to new cases, it being possible to include new forensic analysis. Both the interactive tool and the automatic generator, have been validated by experts. The automatic generator tool has been checked for feasibility with forensic statistical analysis. Finally, a usability study was carried out to assess the intuitive use of the interactive tool.
Towards a 2D cortical osseous tissue representation and generation at micro scale. A computational model for bone simulations
(ELSEVIER, 2020-12) Pérez-Cano, Francisco Daniel; Luque-Luque, Adrián; Jiménez-Delgado, Juan-José
Background and objective: the acquisition of microscopic images of human bones is a complex and expensive process. Moreover, the objective of obtaining a large data bank with microscopic images in order to carry out massive studies or to train automatic generation algorithms is not an option. Consequently, most of the current work focuses on the analysis of small regions captured by a microscope. The aim is the development of a tool to represent bone tissue at microscopic levels which is suitable for performing physical simulations, as well as for the diagnosis of various diseases. This work includes the whole process from the digitization of a human bone to the generation of bone tissue in a determined area of the bone selected through a cutting plane. Methods: based on the anatomy of the bone structure, the parameters that allow the representation of the bone tissue at mesoscale level have been analyzed. Although the models are randomly generated, they are based on statistical parameters. The model generator is based on the analysis of images of bone tissue and its parameters, performing a representation of each of its relevant structures in a way that fulfils these parameters. Results: the tool is useful for the virtual generation of bone tissue that satisfies the main characteristics of the cortical bone. The models obtained have been favorably evaluated in two stages. In the first stage, a scientific group has examined a set of images, in which images of the models generated were mixed with images obtained through traditional methods. Then, the physical characteristics of the generated tissue have been compared with the morphology of the bone tissue. Conclusions: the model generator allows us to perform precise simulations in order to obtain realistic images with physical characteristics in accordance with reality. It is necessary to emphasize that even though the most relevant structures are included, the proposed model generator can be expanded to include new parameters or elements, so that it can be adapted to new needs. It could even break down randomness and parameterize it completely in order to allow the recreation of the tissue conditions of other studies.
Phonendo: a platform for publishing wearable data on distributed ledger technologies
(Springer, 2023-08-07) Moya, Francisco; Quesada, Francisco J.; Martínez, Luis; Estrella, Fco Javier
Nowadays, Internet of Things (IoT) devices, especially wearable devices, are commonly integrated into modern intelligent healthcare software. These devices enable medical practitioners to monitor pervasively patients’ parameters outside the clinical environment. However, the ease of manipulating wearable devices and their data streams raises concerns regarding patient privacy and data trust. Distributed ledger technologies (DLT) offer solutions to enhance resistance against information manipulation and eliminate single points of failure. By leaveraging DLT, wearable-based solutions can be developed with a wider range of capabilities. This paper carries out an analysis of shortcomings, limitations, potential applications and needs in the medical domain, to introduce Phonendo 1.0, a DLT–IoT-based platform designed to capture data streams from wearable devices and publishing them on a distributed ledger technology infrastructure. The architecture and its difference services are justified based on the identified needs and challenges in the medical domain.
A Consensus-Driven Group Recommender System
(Wiley Periodicals, Inc, 2015) Castro, Jorge; Quesada, Francisco J.; Palomares, Iván; Martínez, Luis
Recommender systems aim at filtering large amounts of information for users, providing them with those pieces of information which better meet their preferences or needs. Such systems have been traditionally used in diverse areas, such as e-commerce or tourism. Within this context, group recommender systems address the problem of generating recommendations for groups of users who might have different interests. Although different aggregation processes have been extensively utilized in real-life applications to generate group recommendations, such processes do not guarantee that the list of products recommended to the group reflect a high agreement level among its members' individual preferences. Given the need for considering the added value of obtaining group recommendations under a high agreement level, this paper presents a novel group recommender system methodology that attempts to reach a high level of consensus among individual recommendations of group members. To do this, and inspired by existing group decision-making approaches in the literature, a consensus reaching process is carried out to bring such individual recommendations closer to each other before delivering the group recommendations.
Managing experts behavior in large-scale consensus reaching processes with uninorm aggregation operators
(Elsevier, 2015-10) Quesada, Francisco J.; Palomares, Iván; Martínez, Luis
In many real-life large scale group decision making problems, it can be necessary and convenient a consensus reaching process, which is an iterative procedure aimed at seeking a high degree of agreement amongst experts’ preferences before making a group decision. Although a wide variety of models and approaches have been proposed and developed to support consensus reaching processes, in large groups there are some important aspects that still require further study, such as the treatment of experts’ behaviors that could hamper reaching the wanted agreement. More specifically, it would be necessary an approach to deal with experts properly, based on the overall behavior they present during the discussion process, as well as reinforcing repeated patterns of cooperative (or uncooperative) behavior adopted by experts. This paper presents an expert weighting methodology for consensus reaching processes in large-scale group decision making, that incorporates the use of uninorm aggregation operators. Such operators, which are characterized by their property of full reinforcement, are used in the proposed methodology to allow the experts’ weighting based on their overall behavior during the consensus process and the behavior evolution across the time. This proposal is integrated in a consensus model for large-scale group decision making problems under uncertainty, and it is put in practice to show an illustrative example of its effectiveness and improvements with respect to other approaches.
Semantic Segmentation of Natural Materials on a Point Cloud Using Spatial and Multispectral Features.
(MDPI, 2020) Jurado Rodríguez, Juan Manuel; Cárdenas Donoso, José Luis; OGAYAR ANGUITA, CARLOS JAVIER; ORTEGA ALVARADO, LIDIA M.; FEITO HIGUERUELA, FRANCISCO RAMON
The characterization of natural spaces by the precise observation of their material properties is highly demanded in remote sensing and computer vision. The production of novel sensors enables the collection of heterogeneous data to get a comprehensive knowledge of the living and non-living entities in the ecosystem. The high resolution of consumer-grade RGB cameras is frequently used for the geometric reconstruction of many types of environments. Nevertheless, the understanding of natural spaces is still challenging. The automatic segmentation of homogeneous materials in nature is a complex task because there are many overlapping structures and an indirect illumination, so the object recognition is difficult. In this paper, we propose a method based on fusing spatial and multispectral characteristics for the unsupervised classification of natural materials in a point cloud. A high-resolution camera and a multispectral sensor are mounted on a custom camera rig in order to simultaneously capture RGB and multispectral images. Our method is tested in a controlled scenario, where different natural objects coexist. Initially, the input RGB images are processed to generate a point cloud by applying the structure-from-motion (SfM) algorithm. Then, the multispectral images are mapped on the three-dimensional model to characterize the geometry with the reflectance captured from four narrow bands (green, red, red-edge and near-infrared). The reflectance, the visible colour and the spatial component are combined to extract key differences among all existing materials. For this purpose, a hierarchical cluster analysis is applied to pool the point cloud and identify the feature pattern for every material. As a result, the tree trunk, the leaves, different species of low plants, the ground and rocks can be clearly recognized in the scene. These results demonstrate the feasibility to perform a semantic segmentation by considering multispectral and spatial features with an unknown number of clusters to be detected on the point cloud. Moreover, our solution is compared to other method based on supervised learning in order to test the improvement of the proposed approach.
Fuzzy monitoring of in-bed postural changes for the prevention of pressure ulcers using inertial sensors attached to clothing
(ACADEMIC PRESS INC ELSEVIER SCIENCE, 2020) Bernal, Edna; Polo, Aurora; Espinilla Estévez, Macarena; Medina, Javier
Postural changes while maintaining a correct body position are the most efficient method of preventing pressure ulcers. However, executing a protocol of postural changes over a long period of time is an arduous task for caregivers. To address this problem, we propose a fuzzy monitoring system for postural changes which recognizes in-bed postures by means of micro inertial sensors attached to patients’ clothes. First, we integrate a data-driven model to classify in-bed postures from the micro inertial sensors which are located in the socks and tshirt of the patient. Second, a knowledge-based fuzzy model computes the priority of postural changes for body zones based on expert-defined protocols. Results show encouraging performance in the classification of in-bed postures and high adaptability of the knowledge-based fuzzy approach.
Assessment of sustainable development objectives in Smart Labs: technology and sustainability at the service of society
(ELSEVIER, 2021-11-06) Verdejo, Ángeles; Espinilla, Macarena; López Ruiz, Jose Luis; Francisco, Jurado
Sustainable development is the working basis of engineering research and cities are becoming increasingly flexible, inclusive and intelligent. In this context, there is a need for environments that emulate real-life spaces in which cutting-edge technologies can be implemented for subsequent deployment in society. Smart Labs or Living Labs are spaces for innovation, research and experimentation that integrate systems, devices and methodologies focused on people and their environments. The technologies studied and developed in such labs can then be deployed in human spaces to provide intelligence, comfort, health and sustainability. Health and wellness, energy and environment, artificial intelligence, big data and digital rights are some of the disciplines being studied. At the same time, the UN 2030 Agenda provides a comprehensive framework to promote human well-being through the Sustainable Development Goals. In this work, an evaluation model of its indicators in smart environments is performed through a mixed review methodology. The objective of this work is the analysis and implementation of the SDGs in Smart Labs through a literature review and a case study of UJAmI, the smart laboratory of the University of Ja´en. The results provide quantitative and qualitative data on the present and future of the smart devices implemented in the UJAmI lab, providing a roadmap for future developments.
Toward an Interpretable Continuous Glucose Monitoring Data Modeling
(IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC, 2024-10) Gaitán-Guerro, Juan Francisco; Lopez Ruiz, Jose Luis; Espinilla Estévez, Macarena; Martínez Cruz, Macarena
The ongoing global health challenge posed by diabetes necessitates a critical understanding of all generated data streamed from sensors. To address this, our study presents a robust fuzzy-logic-based descriptive analysis of glucose sensor data. This analysis is embedded within the context of an innovative architecture designed to support multipatient monitoring, with the goal of assisting healthcare professionals in their daily tasks and providing essential decision-making tools. Our novel approach captures and interprets complex data patterns from glucose sensors, and also introduces the capability of creating high-quality linguistic summaries, to highlight the most relevant phenomena through the use of natural language (NL). These descriptions facilitate clear communication between healthcare professionals and people with diabetes, enhancing a deeper understanding of intricate data patterns and promoting collaboration in diabetes care. A comparative evaluation between our proposal and the one obtained using GPT-4 underscores the sustainability, effectiveness, and efficiency of our methodology, positioning it as a new standard for empowering diabetic patients in terms of care and prevention, contributing to their progress and well-being.

RUJA: Repositorio Institucional de Producción Científica

Examinar

Envíos recientes