A machine learning based approach for supporting triage of HIV-related documents

The number of biomedical scientific publications available in multiple repositories is huge and rapidly growing. As of April 2014, PubMed, the largest knowledge source for biomedical and life science literature, comprises more than 23 million citations. Querying PubMed with the keyword HIV provides a list of almost three hundred thousand citations. Retrieving data of particular interest for a specific research field in such a large volume of publications is often like looking for a needle in a haystack.

Content centric networks

The advent of smartphones and the associated change in usage patterns indicates that carrier voice revenues are constantly going down and are expected to almost vanish in the near future. Data revenues are picking up but not at a rate which will make up for the shortfall in voice revenues. Also, the increase in data revenue is accompanied by a corresponding increase in traffic. This leads to higher costs in deploying and operating networks as well as a significant growth in network complexity. In this project we will be working on

An efficient implementation of a computational camera for smartphones

A camera is a device that captures light from scenes. Over the last century, the evolution of cameras has been truly remarkable. Through this evolution, the underlying camera has been improved by using a better optical lens. However, the new improved optical lenses, have been remained fixed in terms of size and weight which makes it hard to use in portable devices. In contrast to optical trend, according to Moore’s law, the number of transistor in the chip doubles approximately every two years. This leads to a huge improvement in computational devices.

LuCID Dryer Process Optimization

The project concentrates on MEMS (Micro-Electro-Mechanical Systems) fabrication process optimization. MEMS devices comprise suspended structures, such as cantilevers, gyros, membranes, etc., which are created by removing sacrificial material around them in order to have these structures released to perform their designated function. Currently, these structures are being released with the use of either dry etch or chemical vapor techniques. Wet chemistry release of suspended structures is favorable: minimized costs, parallel process and most importantly reduced impact on the environment.

EMG Driven Human-Humanoid Robot Interaction

The electromyography (EMG) signals are biological signals, which are related to the electrical activity produced by skeletal muscles. To measure and analyze EMG signals, we will use noninvasive EMG surface electrodes. The EMG signals allow knowing the intention of the movement before that the movement happens; this is because of the electromechanical delay of the muscle. The proposed project would exploit this property to generate humanoid robot movements in order to interact properly with a human being.

Développement de coupleurs sur le substrat MnM pour applications en onde millimétriques

Mon stage au sein du laboratoire de microélectronique (LME) du département des systèmes électroniques (PSI) de l’École polytechnique de l’Université de Sao Paulo, (EPUSP) sera d’une durée de quatre mois. Mon mandat sera de développer un dispositif (coupleur) qui sera utilisé dans des systèmes de communication sans fils à courte distance (quelques mètres). Le matériau (substrat) utilisé pour concevoir ce couleur relève d’une technologie développer et breveté par le groupe de recherche du LME en collaboration avec un partenaire français.

Parallel intelligent multiple-run facility for optimal design of power systems

The research aims to develop a new computer algorithm for automating simulations of complex power systems when simulations are used for design. The new algorithm will be able to find multiple optimal solutions, which will give the designer the ability to scrutinize and select the most suited option. Additionally the algorithm will be parallel in nature, which means it can be executed simultaneously and in unison on several computing machines.

Autonomous, Adaptive, and Real-time Cognitive Intrusion Detection System for the Cloud

Currently, cloud and Internet service providers typically use antivirus, firewall, packet inspection and intrusion detection systems (IDSs) to protect against cyber-security threats on the cloud and internet. These protection systems rely on the knowledge of a fixed and known database of threats to detect malicious activity, but they have no ability to detect new, mutated, threats dynamically. Once an undetected cyber-attack has caused damage and has subsequently been identified by technicians or third party sources, only then can these systems be updated by going offline.

Influence des tests d’intrusion sur l’analyse de risque en technologie de l’information

La montée exponentielle des différentes technologies servant au traitement de l’information digitale a marqué les dix dernières années par des vagues d’événements de sécurité aussi importante les unes que les autres. Les systèmes appartenant aux entreprises, gouvernements, organisations sont exposés à de nombreuses menaces et les mesures misent en place ne sont souvent pas suffisantes pour réduire de manière considérable le risque d’attaque. Ces bris d’information conduisent la plupart du temps à de graves pertes monétaires pour les entreprises.

Deformable shape-based registration for MR-compatible robotically assisted prostate interventions

Prostate cancer is the second most cancer related death in men, accounting for over 111 M$ yearly in Canada. Current prostate biopsy or therapy procedures such as transperineal brachytherapy use transrectal ultrasound guidance and a template to guide needles along parallel trajectories. However, pubic arch interference (PAI) with the implant path obstructs part of the prostate from being targeted by the brachytherapy or laser ablation needles.