32 projects
Status
Domain
Organization Type
DIS-COVID: A privacy-friendly, user-centric framework for estimating crowd mobility patterns and ...
Modelling the spread of an epidemic over a given territory requires detailed understanding of how the population moves, interacts and behaves. This usually relies on location information (e.g. GPS from smartphones) or cellular datasets (i.e. Call Detail Records aggregating communications over a cellular network). Analysis of such data provides the foundation to develop strategies to prevent further spreading. However, access to this data is subject to controversy, first of all because it is not respectful of the individuals' privacy. Furthermore, analysing this data does not provide a full ...
Data science aided adaptive strategy for anti-body testing to manage post-outbreak and future outbreaks
Why? Anti-body tests are paramount to understand the extent of herd immunity developed in the community. Thus, these tests can help plan public outreach by local authorities during and after an outbreak. Further, it can aid in developing containment strategies to prevent a subsequent wave of the outbreak. In a scenario where the current pandemic becomes seasonal, anti-body tests would help to strategize responses seasonally, including vaccination campaigns. On a personal level, it can help an individual take appropriate preventive and precautionary measures. What? A data-driven applica...
Detection Covid19 patients rapidly using cough and voice analyses
Emergency service hospital center receives a huge number of calls in epidemic period, among these calls there are critical cases that need rapid intervention, it may be delayed to answer these cases. Therefore, we suggest a solution by giving an indication of the health status of the person calling directly when hearing his voice and cough. This indicator depends on finding a degree of correspondence between the caller's voice, cough signatures and that of certain patients with the COVID-19 virus. This solution can also be used effectively by health professionals for remote consultations in...
Facilitating optimal exit strategies with smart access control and identification (SmartExit)
This project aims at facilitating exit strategies incorporating access to the public space, border crossing or critical areas such as retirement homes, based on the individual COVID-19 immunity and/or infection status. The smart access control will be based on passports, ID cards, smart cards, or even smartphone apps. We will propose a mechanism, produce a prototype implementation, and present a preliminary formal analysis of the proposed solution. While it might be necessary to waive users' privacy to contain the epidemic in the short term, we will look for mechanisms that waive it to the ...
INFODEMIC - Combating the Infodemic to Curb the Pandemic: A Step Towards a Reliable and Knowledg...
Building on SnT/Uni.lu expertise on text analytics and machine learning, we propose the INFODEMIC project to challenge fake news on social media. Our approach considers a holistic view in combating fake news. To that end, we develop three key work packages: (1) Archiving of viral fake news, where a wikipedia-like crowdsourcing approach is investigated for building a large and publicly-searchable repository of fake news; (2) Automating early detection of fake news, where transfer learning-based models are devised to flag potential fake news in a multilingual setting; (3) Integrating a relia...
Information Diffusion in Tweeter during the COVID-19 Pandemic: the Case of the Greater Region
WHO announced the COVID-19 outbreak as a global health emergency, and later as a pandemic. The outbreak of the coronavirus leads to an outbreak of pandemic information as well in major online social media platforms, such as Twitter, Instagram and Facebook. Such pandemic information diffusion over online social media can strongly influence people's behaviour, and thus have an impact on the effectiveness of control and protective measures deployed by the governments. In this project, we aim to perform in-depth analysis of pandemic information dynamics, based on a dataset collected from Twitte...
Let's keep connected
We will develop an interactive chatbot to provide support to confined people not only during the COVID19 crisis, but also after the confinement, in the exit and post-exit phase. The public authorities and some NGOs are offering support services to help people facing specific problems, and the same should be the case for the COVID19 crisis. It is of prime importance for these Luxembourgish services not to be overwhelmed by wrong enquiries, and also to be able to serve the maximum amount of enquiries. A majority of them could be targeted on same frequently asked questions. The chatbot that we...
MC19: Early Detection of Epidemics and Health Status Prediction Using Intelligent Wireless Body Sensors
MyelinS, a Luxembourgish startup (LLC in the incorporation phase) is proposing a novel solution for early detection of epidemics (particularly Covid-19), health status daily analysis, and prediction of critical health situations caused by the virus without any direct medical intervention, avoiding thereby any risk of transmission of the virus to the healthcare workers, as well as to healthy people. Unlike any existing solutions, the proposed detection and prediction system is non-invasive (on mobile phones), accurate, and rapid (detection in less than 1.5 minutes). The proposed solution ...