IMDEA Networks Institute Digital Repository
Not a member yet
1915 research outputs found
Sort by
Consistent Comparison of Symptom-based Methods for COVID-19 Infection Detection
Get PDFBackground:
During the global pandemic crisis, various detection methods of COVID-19-positive cases based on self-reported information were introduced to provide quick diagnosis tools for effectively planning and managing healthcare resources. These methods typically identify positive cases based on a particular combination of symptoms, and they have been evaluated using different datasets.
Purpose:
This paper presents a comprehensive comparison of various COVID-19 detection methods based on self-reported information using the University of Maryland Global COVID-19 Trends and Impact Survey (UMD-CTIS), a large health surveillance platform, which was launched in partnership with Facebook.
Methods:
Detection methods were implemented to identify COVID-19-positive cases among UMD-CTIS participants reporting at least one symptom and a recent antigen test result (positive or negative) for six countries and two periods. Multiple detection methods were implemented for three different categories: rule-based approaches, logistic regression techniques, and tree-based machine-learning models. These methods were evaluated using different metrics including F1-score, sensitivity, specificity, and precision. An explainability analysis has also been conducted to compare methods.
Results:
Fifteen methods were evaluated for six countries and two periods. We identify the best method for each category: rule-based methods (F1-score: 51.48\% - 71.11\%), logistic regression techniques (F1-score: 39.91\% - 71.13\%), and tree-based machine learning models (F1-score: 45.07\% - 73.72\%). According to the explainability analysis, the relevance of the reported symptoms in COVID-19 detection varies between countries and years. However, there are two variables consistently relevant across approaches: stuffy or runny nose, and aches or muscle pain.
Conclusions:
Regarding the categories of detection methods, evaluating detection methods using homogeneous data across countries and years provides a solid and consistent comparison. An explainability analysis of a tree-based machine-learning model can assist in identifying infected individuals specifically based on their relevant symptoms. This study is limited by the self-report nature of data, which cannot replace clinical diagnosis.Community of MadridSpanish Ministry of Science and InnovationTRUEpu
In-band multi-connectivity with local beamtraining for improving mmWave network resilience
Get PDFMulti-connectivity is considered a key enabler for 5G networks and beyond, aiming to enhance capacity by combining multiple communication links in the same or different bands. Similarly, in cell-free networks all Access Points (APs) jointly serve users in the same band, boosting capacity through enhanced spectral efficiency. Both approaches can be very effective in Millimeter-Wave (mmWave) networks by addressing key issues of reliability and robustness due to the multiple simultaneous links. Furthermore, the use of narrow directional beams in mmWave spatially separates the signals, allowing for in-band multi-connectivity through local beamtraining. Such in-band multi-connectivity would be an alternative design to
traditional cell-free networks that does not rely on phase-coherent processing or centralized methods for interference suppression. The physical layer processing and resource allocation problem then
simplifies to a local beamtraining challenge, making these networks easier and simpler to implement and deploy, as any connection just has to train and maintain the local beam. We validate this approach
by designing a multi-connectivity mmWave network with minimal network synchronization, relying solely on analog beamforming for spatial separation. Our evaluation results demonstrate that inband multi-connectivity with 4 asynchronous and independent links can provide uninterrupted service even in dense, high-traffic scenarios, compared to up to 20% of service loss in a standard singleconnectivity deployment. Distributing the traffic across multiple APs also had throughput gains of up to 30%, showing that multiconnectivity mmWave networks can provide a high-throughput, reliable and stable service for next-generation applications.European UnionTRUEinpres
Case studies of clinical decision-making through prescriptive models based on machine learning
Get PDFBackground
The development of computational methodologies to support clinical decision-making is of vital importance to reduce morbidity and mortality rates. Specifically, prescriptive analytic is a promising area to support decision-making in the monitoring, treatment and prevention of diseases. These aspects remain a challenge for medical professionals and health authorities.
Materials and Methods
In this study, we propose a methodology for the development of prescriptive models to support decision-making in clinical settings. The prescriptive model requires a predictive model to build the prescriptions. The predictive model is developed using fuzzy cognitive maps and the particle swarm optimization algorithm, while the prescriptive model is developed with an extension of fuzzy cognitive maps that combines them with genetic algorithms. We evaluated the proposed approach in three case studies related to monitoring (warfarin dose estimation), treatment (severe dengue) and prevention (geohelminthiasis) of diseases.
Results
The performance of the developed prescriptive models demonstrated the ability to estimate warfarin doses in coagulated patients, prescribe treatment for severe dengue and generate actions aimed at the prevention of geohelminthiasis. Additionally, the predictive models can predict coagulation indices, severe dengue mortality and soil-transmitted helminth infections.
Conclusions
The developed models performed well to prescribe actions aimed to monitor, treat and prevent diseases. This type of strategy allows supporting decision-making in clinical settings. However, validations in health institutions are required for their implementation.TRUEpu
Hacia una red 5G/6G con inteligencia nativa: oportunidades e integración en los nuevos desarrollos de redes móviles
No full textThe integration of intelligence in future mobile networks presents both a challenge and an opportunity that we must address now. The impact of native intelligence in networks is yet to be determined. However, this impact will only be significant if we are able to design from scratch network architectures and protocols that seamlessly integrate the intelligence in the network.Cátedra THIN5GFALSEpu
Offloading Augmented Reality Tasks with Smart Energy Source-Aware Algorithms at the Edge
Get PDFThe development of novel use cases in beyond-5G and 6G networks will rely, among other aspects, on the availability of computing resources at the edge, therefore enabling the realization of applications that are both computationally demanding and latency constrained, such as Mobile Augmented Reality (MAR). Indeed, due to
end devices’ intrinsic constraints on computation capabilities and battery, newer MAR applications require offloading their most demanding tasks. However, the constrained nature of edge resources implies that these tasks should be carefully allocated at the edge network in order to guarantee satisfactory Quality of Experience to end-users. In this context, we analyze the edge operator’s resource allocation to support the energy-aware offloading of MAR tasks at the edge of the cellular network with the goal of not only maximizing service acceptance (i.e., revenue), but also optimizing the operator’s business utility, which depends on its carbon footprint
and the profit of operating the service. We leverage Deep Reinforcement Learning to propose an efficient model to operate the edge resource allocation that can adapt to different utilities.TRUEinpres
LiFi for Low-Power and Long-Range RF Backscatter
Get PDFLight bulbs have been recently explored to design Light Fidelity (LiFi) communication to battery-free tags, thus complementing Radiofrequency (RF) backscatter in the uplink. In this paper, we show that LiFi and RF backscatter are complementary and have unexplored interactions. We introduce PassiveLiFi, a battery-free system that uses LiFi to transmit RF backscatter at a meagre power budget. We address several challenges on the system design in the LiFi transmitter, the tag and the RF receiver. We design the first LiFi transmitter that implements a chirp spread spectrum (CSS) using the visible light spectrum. We use a small bank of solar cells for both communication and harvesting, and reconfigure them based
on the amount of harvested energy and desired data rate. We further alleviate the low responsiveness of solar cells with a new low-power receiver design in the tag. We design and implement a novel technique for embedding multiple symbols in the RF backscatter based on delayed chirps. Experimental results with an RF carrier of 17 dBm show that we can generate RF backscatter with a range of 92.1 meters/µW consumed in the tag, which is almost double with respect to prior work.European UnionMinisterio de Asuntos Económicos y Transformación DigitalMinisterio de Ciencia e InnovaciónTRUEpu
Characterizing and Modeling Session-Level Mobile Traffic Demands from Large-Scale Measurements
Get PDFWe analyze 4G and 5G transport-layer sessions generated by a widerange of mobile services at over 282, 000 base stations (BSs) of anoperational mobile network, and carry out a statistical characteriza-tion of their demand rates, associated traffic volume and temporalduration. Our study unveils previously unobserved session-levelbehaviors that are specific to individual mobile applications andpersistent across space, time and radio access technology. Basedon the gained insights, we model the arrival process of sessions atheterogeneously loaded BSs, the distribution of the session-levelload and its relationship with the session duration, using simpleyet effective mathematical approaches. Our models are fine-tunedto a variety of services, and complement existing tools that mimicpacket-level statistics or aggregated spatiotemporal traffic demandsat mobile network BSs. They thus offer an original angle to mobiletraffic data generation, and support a more credible performanceevaluation of solutions for network planning and management. Weassess the utility of the models in practical application use cases,demonstrating how they enable a more trustworthy evaluation ofsolutions for the orchestration of sliced and virtualized networks.TRUEinpres
Prototyping Visible Light Communication for the Internet of Things Using OpenVLC
Get PDFVisible Light Communication (VLC) has emerged in the last few years as a promising technology not only for high-speed communication but also for serving a new generation of Internet of Things (IoT) devices that may leverage the pervasive lighting infrastructures. Integrating VLC in lighting environments for IoT requires the design of networked and intelligent luminaries and new IoT devices, encompassing the development of innovative technologies and new algorithms. A common experimental platform is necessary to lower the entrance barriers of VLC and speed up the research development. In this paper, we provide guidelines for prototyping VLC for IoT applications, assisted by the open-source platform OpenVLC. We also introduce the new development on OpenVLC, which guarantees support for more powerful LEDs and much longer distance (extending the communication distance from 6 m to 19 m), dimming adaption, among other features. Its low-cost, open-source, and open-hardware designs allow researchers in the community to swiftly adapt it to suit their research purposes.Ministerio de Asuntos Económicos y Transformación DigitalEuropean UnionTRUEpu
Misinformation in third-party voice applications
Get PDFThis paper investigates the potential for spreading misinformation via third-party voice applications in voice assistant ecosystems such as Amazon Alexa and Google Assistant. Our work fills a gap in prior work on privacy issues associated with third-party voice applications, looking at security issues related to outputs from such applications rather than compromises to privacy from user inputs. We define misinformation in the context of third-party voice applications and implement an infrastructure for testing third-party voice applications using automated natural language interaction. Using our infrastructure, we identify — for the first time — several instances of misinformation in third-party voice applications currently available on the Google Assistant and Amazon Alexa platforms. We then discuss the implications of our work for developing measures to pre-empt the threat of misinformation and other types of harmful content in third-party voice assistants becoming more significant in the future.TRUEpu
Advanced Methods to Audit Online Web Services
Get PDFOnline web services have grown dramatically in size and diversity in the last years, becoming essential components of our daily life and allowing us to conduct elementary tasks like working, getting informed, or keeping in contact with relatives and friends.
However, all the changes and evolution experimented on by the online web services had not have been possible without implementing a profitable economic model that sustains it. Despite a suitable percentage of these services being fee-based, they represent a lucrative business that generates billions of dollars, allowing the creation of some of the biggest companies in the world in terms of market capitalization, like Alphabet Inc. or Meta Inc. (Previously known as Facebook Inc.). Being costless and lucrative is possible due to an advertising-based monetization model, which consists of delivering ads to the users in exchange for their services (e.g., Facebook or YouTube). Although online advertising dates back to the middle of the 90s, its popularity has experienced an increase among brands and advertising agencies in the last decade, mainly due to its capacity to reach precise audiences at a low cost.
Converting online web services into advertising walls is a double-edged sword for the users. The capacity offered by online advertising to segment their audiences requires a massive collection of personal data from the users, including their web browsing histories or even more invasive data such as age, gender, or location to infer the online profile of the users. This data collection is possible due to implementing a complex tracking ecosystem by online advertising companies from which multiple stakeholders collect, process, and exchange information. The many privacy cases of abuse inflicted by this industry motivated the implementation of new regulatory efforts to protect consumers’ privacy in the last years. Some notable examples are the General Data Protection Regulation (GDPR)[1] in the European Union or the California Consumer Privacy Act Regulations (CCPA)[2]in California, USA. Further, these privacy regulations typically contain specific provisions and strict requirements for websites that provide sensitive material to end users, including sexual, religious, and health services.
Implementing new regulatory frameworks, alongside the growth of online web services, forces an endless evolution of current techniques to study and audit online web services. Furthermore, there is a need to emphasize the online advertising ecosystem, as it represents the primary economic support of a high percentage of web services. Also, the activities and abuses conducted by this ecosystem drove the implementation of current privacy regulations to control the use and collection of personal data.
This dissertation falls within the topics of Internet measurements, tackling the need for new measurement techniques and methodological approaches to audit and study online web services. Precisely, this dissertation analyzes three aspects of the web. First, we implement a methodology to study sensitive websites, including their potential lack of regulatory compliance. Then, we put into practice our approach by analyzing the pornographic web ecosystem, opening the debate on the need to study and identify web privacy problems from a macroscopic perspective, as the web contains semi-decoupled and highly sensitive subsystems. Second, we look deeply at the suitability and adequacy of domain classification services commonly used by the research community to conduct domain-dependent research studies, including those studying sensitive websites. Finally, we implement a novel methodology to audit the quality and performance of the profiles that Meta (Facebook) and Google create about the users and their ad targeting algorithms. This study also includes an analysis of the transparency tools these two companies offer to the users concerning the process of distributing tailored ads.
In summary, this dissertation brings new methodologies and results to increase our limited knowledge about the web.Telematics EngineeringUniversidad Carlos III de Madrid, Spai