Unidade de Investigação - CIIC - Computer Science and Communication Research Centre
URI permanente desta comunidade:
Navegar
Percorrer Unidade de Investigação - CIIC - Computer Science and Communication Research Centre por Domínios Científicos e Tecnológicos (FOS) "Ciências Médicas::Ciências da Saúde"
A mostrar 1 - 8 de 8
Resultados por página
Opções de ordenação
- Body Area Networks in Healthcare: A Brief State of the ArtPublication . Roda-Sanchez, Luis; Olivares, Teresa; Fernández-Caballero, Antonio; Vera, Daniel; Costa, Nuno; Pereira, António Manuel de JesusA body area network (BAN) comprises a set of devices that sense their surroundings, activate and communicate with each other when an event is detected in its environment. Although BAN technology was developed more than 20 years ago, in recent years, its popularity has greatly increased. The reason is the availability of smaller and more powerful devices, more efficient communication protocols and improved duration of portable batteries. BANs are applied in many fields, healthcare being one of the most important through gathering information about patients and their surroundings. A continuous stream of information may help physicians with making well-informed decisions about a patient's treatment. Based on recent literature, the authors review BAN architectures, network topologies, energy sources, sensor types, applications, as well as their main challenges. In addition, the paper focuses on the principal requirements of safety, security, and sustainability. In addition, future research and improvements are discussed. © 2019 by the authors
- A distributed multiagent system architecture for body area networks applied to healthcare monitoringPublication . Felisberto, Filipe; Laza, Rosalía; Fdez-Riverola, Florentino; Pereira, AntónioIn the last years the area of health monitoring has grown significantly, attracting the attention of both academia and commercial sectors. At the same time, the availability of new biomedical sensors and suitable network protocols has led to the appearance of a new generation of wireless sensor networks, the so-called wireless body area networks. Nowadays, these networks are routinely used for continuous monitoring of vital parameters, movement, and the surrounding environment of people, but the large volume of data generated in different locations represents a major obstacle for the appropriate design, development, and deployment of more elaborated intelligent systems. In this context, we present an open and distributed architecture based on a multiagent system for recognizing human movements, identifying human postures, and detecting harmful activities. The proposed system evolved from a single node for fall detection to a multisensor hardware solution capable of identifying unhampered falls and analyzing the users’ movement. The experiments carried out contemplate two different scenarios and demonstrate the accuracy of our proposal as a real distributed movement monitoring and accident detection system. Moreover, we also characterize its performance, enabling future analyses and comparisons with similar approaches.
- Explainable prototype-based image classification using adaptive feature extractors in medical imagesPublication . Vasconcellos, Nicolas; Tavora, Luis M. N.; Miragaia, Rolando; Grilo, Carlos; Thomaz, LucasPrototype-based classifiers are a category of Explainable Artificial Intelligence methods that use representative samples from the data, called prototypes, to classify new inputs based on a similarity criterion. However, these methods often rely on pre-trained Convolutional Neural Networks as feature extractors, which may not be adapted for the specific type of data being used, thus not suited for identifying the most representative prototypes. In this paper, we propose a method named Explainable Prototype-based Image Classification, a cluster-oriented training strategy that enhances the performance and explainability of prototype-based classifiers. Our method uses a novel loss function, called Cluster Density Error, to fine-tune the feature extractor and preserve the most representative feature vectors in the latent space. We also use Principal Component Analysis-based approach to reduce the dimensionality and complexity of the feature vectors. We conduct experiments on four medical image datasets and compare the results with those from different prototype-based classifiers and state-of-the-art non-explainable learning methods. The proposed method demonstrated superior explainable capabilities and comparable classification performance to the compared methods. Specifically, the proposed method achieved up to 95.01% accuracy and 0.992 AUC using only 43 prototypes. This translated to an improvement in accuracy and AUC score of 21.54% and 9.06%, respectively, and a substantial reduction in the number of prototypes by 98,38%
- How Health Literacy impacts Polytechnic of Leiria Students?Publication . Teixeira Ascenso, Rita Margarida; Luis, Luis; Dias, Sara; Gonçalves, DulceIn 2021, aHealth Literacy(HL) evaluation among university students revealed notable limitations in HL. To assess the general HL of populations comprehensively, the European HLSurvey Questionnaire (HLS-EU-Q) was developed, encompassing 12 subdomains to provide a broad perspective on public health. In 2014, the questionnaire was adapted for use in Portugal, resulting in the HLS-EU-PT version, validated through a 16-question survey (HLS-EU-PT-Q16).Global HL andthreedomains’ indexes and levelswere determined, namely Healthcare (HC), Disease prevention (DP), and Health Promotion (HP). The HLSEU-Q16-PT assessment demonstrated satisfactory internal consistency, with 0.8834Cronbach's alpha coefficient.In this study, an online survey distributedbetween 2020-2021among Polytechnic of Leiria academia allowed data collection from various stakeholders, including 251 students, 109 professors, 15 researchers, and 55 other staff. From the430 responses,75 questions were analysed. The saved data wasthefocus of this work, regarding a thesis of the first edition of the master’s in data science to analysethe 251 surveyed studentsand their HL. The results revealed that thesestudents have lower HL index, and, in this case study,health areadegreeor school impactsHL.
- Hybrid Honey Bees Mating Optimisation algorithm to assign terminals to concentratorsPublication . Bernardino, Eugénia M.; Bernardino, Anabela M.; Sánchez-Pérez, Juan Manuel; Gómez-Pulido, Juan Antonio; Vega-Rodríguez, Miguel AngelIn this paper we propose a new approach to assign terminals to concentrators using a Hybrid Honey Bees Mating Optimisation algorithm. Honey Bees Mating Optimisation (HBMO) algorithm is a swarm-based optimisation algorithm, which simulates the mating process of real honey bees. We apply a hybridisation of HBMO to solve a combinatorial optimisation problem known as Terminal Assignment Problem (TAP). The purpose is to connect a given set of terminals to a given set of concentrators and minimise the link cost to form a communication network. The feasibility of Hybrid HBMO is demonstrated and compared with the solutions obtained by other algorithms from literature over different TAP instances.
- A Hybrid Population-Based Incremental Learning algorithm for load balancing in RPRPublication . Bernardino, Anabela M.; Bernardino, Eugénia M.; Sánchez-Pérez, Juan Manuel; Gómez-Pulido, Juan Antonio; Vega-Rodríguez, Miguel AngelWhen managed properly, the ring networks are uniquely suited to deliver a large amount of bandwidth in a reliable and inexpensive way. An optimal load balancing is very important, because it increases the system capacity and improves the overall ring performance. An important optimisation problem in this context is the Weighted Ring Arc Loading Problem (WRALP). It consists of the design, in a communication network of a transmission route (direct path) for each request, such that high load on the ring arcs will be avoided. WRALP asks for a routing scheme such that the maximum load on the ring arcs will be minimum. In this paper we study WRALP without demand splitting and we propose a Hybrid Populationbased Incremental Learning (HPBIL) to solve it. We show that HPBIL is able to achieve good solutions, improving the results obtained by previous approaches.
- Microsoft's Your Phone environment from a digital forensic perspectivePublication . Domingues, Patricio; Andrade, Luis Miguel; Frade, MiguelYour Phone is a Microsoft dual mobile/desktop application that links a Windows 10 environment to a smartphone. The Android version provides the smartphone's user with the ability to control the mobile device from Windows 10, allowing to place/receive calls, send/receive text messages such as SMS, MMS and RCS, access up to the last 2000 photos/screenshots of the device and to receive notifications from applications, all through the Windows 10 Your Phone application and, if configured to do so, within Windows 10 notification center. This work analyzes the Your Phone environment, that is, Your Phone Companion for Android and Your Phone for Windows 10. The paper studies the digital forensic artifacts that can be found in a post mortem analysis, focusing on the SQLite3 databases used by both the Android and Windows 10 applications. We also compare the examined version with a previous version of Your Phone, showing that Your Phone newest functionalities bring new valuable artifacts for forensic examiners. The study shows that Your Phone data left on a Windows 10 device can be useful to access a copy of messages, photos, and document interactions, especially when the Android device is inaccessible or even physically unavailable. To ease the task for digital forensic examiners, we have updated our open-source YPA software that collects and analyzes Your Phone data from a Windows 10 system. YPA runs as a module within the digital forensic Autopsy software.
- Using Secure Multi-Party Computation to Create Clinical Trial CohortsPublication . Borges, Rafael; Ferreira, Bruno; Antunes, Carlos Machado; Maximiano, Marisa; Gomes, Ricardo; Távora, Vitor; Dias, Manuel; Bezerra, Ricardo Correia; Domingues, Patrício; Antunes, Carlos MachadoThe increasing volume of digital medical data offers substantial research opportunities, though its complete utilization is hindered by ongoing privacy and security obstacles. This proof-of-concept study explores and confirms the viability of using Secure Multi-Party Computation (SMPC) to ensure protection and integrity of sensitive patient data, allowing the construction of clinical trial cohorts. Our findings reveal that SMPC facilitates collaborative data analysis on distributed, private datasets with negligible computational costs and optimized data partition sizes. The established architecture incorporates patient information via a blockchain-based decentralized healthcare platform and employs the MPyC library in Python for secure computations on Fast Healthcare Interoperability Resources (FHIR)-format data. The outcomes affirm SMPC’s capacity to maintain patient privacy during cohort formation, with minimal overhead. It illustrates the potential of SMPC-based methodologies to expand access to medical research data. A key contribution of this work is eliminating the need for complex cryptographic key management while maintaining patient privacy, illustrating the potential of SMPC-based methodologies to expand access to medical research data by reducing implementation barriers.