Projets novateurs réalisés

Nematic Ferroelectric Liquid Crystal Elastomers

The goal of this project is to develop a class of smart materials known as “Electroactive Liquid Crystal Elastomers” which undergo shape change and exhibit locomotion by an applied electric field. Ferroelectric molecules will be specifically used in the design and synthesis of these materials which are expected to increase the dielectric constant of the elastomer and result in spontaneous polarization and shape change under the applied electric field. This goal is achieved by two methods: (1) Addition of ferroelectric liquid crystal molecules to the previously developed liquid crystal elastomers, and (2) Synthesizing a new class of liquid crystal elastomers that all its molecular components are composed of ferroelectric molecules. Moreover, the developed material will be studied in terms of mechanical, electrical, and optical properties.

Comprendre les accidents de la route : une modélisation micro

Ce projet de recherche, dirigé par une chercheuse en sciences économiques dans le cadre du programme Mitacs Globalink, vise à analyser les accidents de la route en Tunisie en utilisant des méthodes avancées d’analyse spatiale pour identifier les zones à haut risque, ou “zones noires”. Avec une base de données de 30 000 accidents couvrant la période de 2008 à 2023, l’étude appliquera des techniques telles que la fonction K de Ripley et l’analyse de densité spatiale pour comprendre la distribution des accidents et les facteurs contributifs, comme l’infrastructure routière et le comportement des conducteurs. Ce travail a pour objectif de produire un article scientifique significatif, contribuant à une meilleure compréhension des dynamiques des accidents routiers et à l’élaboration de politiques de prévention efficaces, tout en visant à réduire le nombre d’accidents et à améliorer la sécurité routière en Tunisie.

Mechanisms and Algorithms for Optimal Use of Inter-Clouds

Cloud computing is one of the pillars of the modern computing infrastructure mainly because it allows the procurement of computing services on a pay-as-you-go basis. However, despite the many benefits offered by cloud computing, it has several significant drawbacks such as data lock-in, lack of universal geographic proximity, risk of service outages, and variable cost structures. To address these problems, researchers from the academia and industry have proposed an interconnection of many independently managed cloud-computing systems such that workloads can be seamlessly migrated among them to suite the customer requirements. Many research problems are yet to be solved to realize the vision of interconnecting cloud-computing systems. The goal of this project is to develop new lightweight architectures for interconnecting cloud-computing systems. In particular, we will focus on the development of a Linux container-based workload-mapping framework that will benchmark the different clouds, select the appropriate clouds, and migrate workloads to maximize client benefits. We want to realize the inter-cloud architecture without requiring the clouds to agree on standardized cloud management protocols.

Correlation between trophoblast cell-surface antigen-2 (TROP-2) expression measured using conjugated Gold nanoparticles and ADC response in breast cancer cells.

According to a WHO report in 2022, one in five people are at risk of developing cancer, but only 39% of countries provide comprehensive health coverage. The most common cancers are lung, breast, and colorectal. Chemotherapy and radiotherapy, while common, present challenges due to their side effects and low specificity. New targeted therapies, such as immunotherapy with monoclonal antibodies, offer promising options. For example, trastuzumab helps the immune system target tumors, while antibody-drug conjugates (ADCs), such as trastuzumab-deruxtecan, direct drugs directly to cancer cells. These treatments require precise identification of tumor proteins, usually by immunohistochemistry (IHC) after biopsy, but this method can reach an error rate of 30%, particularly for the low expression of the Her2 protein in breast cancer. In response to these challenges, researchers at Polytechnique Montréal, under the direction of Professor Meunier, have developed a plasmonic nanoparticle technology. This technology improves the accuracy of protein detection and quantification and enables multiplexing, thus offering a potentially more reliable solution for the diagnosis and treatment of cancers.

Anomaly Detection with Noisy Labels in Graphs

This research project focuses on improving how we detect fraudulent activities, such as suspicious transactions, in large financial transaction networks like Mastercard’s. Fraudulent transactions are rare and difficult to catch because they often look similar to regular transactions, and there are limited accurate data to rely on. To tackle this, we are using Graph Neural Networks (GNNs), which can analyze transaction networks in ways that traditional methods cannot by incorporating the graph structure in the model. This network view helps GNNs detect patterns of fraud that might otherwise go unnoticed. However, a big challenge is that some of the data we use to train these models is noisy, meaning it contains mistakes or inaccurate labels. These noisy labels can affect the model negatively, causing
it to learn incorrect patterns and reducing its ability to find real fraud.

Design, synthesis and charaterization of new p-AQM macromolecules for electronics applications

This project focuses on synthesizing a new class of conjugated polymers incorporating stable, quinoidal units that enhance electronic properties, targeting applications in organic field-effect transistors (OFETs), organic solar cells, and organic electrochemical transistors (OECTs). By designing polymers that feature reduced bond-length alternation, these materials exhibit lower band gaps and high charge carrier mobilities, essential for efficient energy and electronic applications. Using a green synthesis route, we avoid organometallic reagents and catalysts, opting instead for sustainable methods that lock the polymer backbone in a rigid conformation. This structural control enhances orbital delocalization and interchain coupling, boosting charge transport efficiency. The project will benefit both institutions by advancing sustainable, high-performance materials that bridge the gap between environmental responsibility and cutting-edge electronic functionality, supporting developments in bioelectronics, flexible electronics, and eco-friendly device manufacturing.

Development of electrospun PMMA/PEG/Sio2 polymer nano-composites for Carbon Capture

The proposed project fosters collaboration between Sabanci University and the University of Waterloo, offering significant benefits to both institutions. Sabanci University will gain access to advanced imaging and gas analysis techniques available at Waterloo, enabling precise characterization and optimization of the CO2 capture system. This enhances the quality and impact of ongoing research at Sabanci. Meanwhile, the University of Waterloo will benefit from participating in cutting-edge research that aligns with global climate action goals, advancing its expertise in sustainable CO2 capture technologies. The partnership strengthens both institutions’ research profiles, promotes knowledge exchange, and paves the way for future collaborations in innovative environmental solutions.

Jumine : Anomalies en flottation minière : amélioration et optimisation du système de détection existant

Le projet vise à améliorer un système de détection des anomalies dans des cellules de flottation, un procédé employé dans l’industrie minière. Actuellement, un détecteur d’anomalie développé par Jumine dans la cellule de flottation identifie des problèmes tels que la vitesse inappropriée de la mousse, son affaissement ou encore sa coalescence trop importante, mais ne fournit pas d’explications ni de solutions.
Le projet vise à expliquer et résoudre ces anomalies en analysant conjointement a) les données des capteurs dans la cellule de flottation et b) les images en continu de la mousse à la surface des cellules de flottation. Le stagiaire sera chargé de développer un modèle utilisant l’apprentissage automatique (AA) non supervisé pour détecter les différents types d’anomalies (labels faibles). Ces données faiblement étiquetées seront ensuite validées par un humain qui examinera les images correspondantes. Ce nouvel ensemble de données étiquetées permettra d’entraîner un classificateur capable d’identifier chaque classe d’intérêt.
Les méthodes d’explicabilité en AA fourniront des explications détaillées sur les anomalies détectées, permettant aux opérateurs de prendre des mesures correctives plus informées, améliorant ainsi l’efficacité du processus de concentration des minéraux.

MOREAU_MÉRAND_Barcelone

This project will examine how the positions of central governments influence the diplomatic strategies of regional entities, such as Catalonia and Scotland, during sovereignty referendums. It focuses on protodiplomacy, which refers to the diplomatic efforts of subnational regions seeking international recognition for their independence movements. The main goal is to conduct interviews with key people involved in the independence movements, such as local politicians and diplomats, to understand the strategies these regions use to gain support and recognition from other countries. The project will also involve organizing workshops to share knowledge between academics and local experts, helping to build a network of researchers focused on this topic. The expected benefits of this research include a better understanding of how regions influence international opinions and gain support during referendums, as well as contributing new insights into the theory of paradiplomacy and the challenges regions face when they seek to be recognized internationally.

Métallooligomères à base de porphyrines et de diketopyrrolopyrroles pour des applications de cellules solaires

Ce projet vise à fabriquer deux nouveaux polymères qui offrent des propriétés supérieures en vue de construire des cellules solaires plastiques. Ces propriétés consistent d’un meilleur empilement et d’une meilleure géométrie planaire pour améliorer le transport des charges, et de composantes judicieusement choisies pour améliorer la vitesse du photo-transfert du premier électron afin de minimiser les pertes d’énergie lumineuse en chaleur. Ces composantes seront des porphyrines (comme l’on retrouve dans hémoglobine) et du dicétopyrrolopyrrole (un pigment), des ponts thiophène et thiazole. Finalement après caractérisation, des cellules solaires seront fabriquées et testées.

Development of a sub-unit vaccine to protect swine against the infections caused by Streptococcus suis

Streptococcus suis is one of the most important causes of bacterial disease in post-weaned piglets. Affected animals suffer from meningitis (affected brain), endocarditis (affected heart), arthritis (affected joints) and septicemia (general infection). There are 35 serotypes of S. suis (same bacteria but different antigens). Some of these serotypes (such as serotype 2) can also affect humans (transmission from pigs), causing serious disease. Usually, veterinarians use antibiotics to treat and even protect animals. However, the large use of antibiotic induces the increase of antibiotic-resistance and there are now many S. suis bacterial strains that are resistant. Unfortunately, there are no effective vaccines in the market to protect against this infection. We have identified a protein called “Sao” that protects pigs against a bacterial strain of S. suis serotype 2. It is not known if it protects against other strains and other serotypes that cause disease in pigs. Production of Sao protein should also be improved to obtain acceptable yields to have enough material to be tested in a large vaccine trial. In addition, this protein should be mixed with a very good adjuvant that increases or aids the immune response of animals and used in pigs.

Détection précoce des épisodes d’hypoglycémie par intelligence artificielle : une approche personnalisée intégrant des mécanismes d’attention

Le diabète de type 1 expose les patients à des épisodes d’hypoglycémie, des baisses dangereuses du taux de sucre dans le sang, qui peuvent entraîner des symptômes graves tels que des vertiges, des évanouissements, voire des convulsions ou un coma. Une détection précoce de ces épisodes est essentielle pour permettre aux patients de réagir rapidement et ainsi éviter des complications potentiellement graves.

Ce projet propose de développer un système de détection intelligent basé sur l’intelligence artificielle (IA). En utilisant les données issues des dispositifs de mesure continue du glucose (CGM) et des algorithmes d’apprentissage profond, nous cherchons à créer une solution capable de prédire les épisodes d’hypoglycémie avant qu’ils ne deviennent critiques. Cette approche personnalisée permettra de fournir des alertes précises et adaptées à chaque patient, réduisant ainsi les faux positifs et améliorant la gestion quotidienne de la glycémie.

Cette technologie a pour objectif d’améliorer la qualité de vie des patients atteints de diabète en leur offrant des outils fiables et proactifs pour gérer leur santé tout en contribuant à l’avancement de la recherche dans le domaine de la santé numérique.