Innovative Projects Realized

Neurostimulateur/neuroenregistreur de l’activité cérébrale pour la surveillance de l’épilepsie

L’épilepsie peut entraîner des crises violentes et touche entre 0,5 et 1% de la population dont ?30% souffrent d’épilepsie réfractaire aux médicaments et bénéficierait de la neurostimulation. Des recherches démontrent la capacité de prévoir les crises 24 heure à l’avance, donnant la possibilité au patient de prendre des mesures d’atténuation. Ce projet vise à développer un système électronique effectuant une lecture de l’activité cérébrale avec résolution cellulaire, extrayant les caractéristiques principales de la décharge neuronale et transmettant cette information avec une compression par codage algébrique dans un lien sans fil. Contrairement aux implants épileptiques actuels limités à quelques dizaines d’électrodes sur plusieurs mm2, chaque circuit, d’une taille < 1 mm2, peut enregistrer simultanément l'activité de 1600 canaux à un pas de 20 ?m. En collaboration avec notre partenaire, nous visons réaliser le premier système d'enregistrement et de stimulation synchronisé et distribué comportant jusqu'à 6400 électrodes adaptées à des implantations à long terme.

Novel Ovulation Research-Recruitment Methods for an App (NORMA) Study

One purpose of the proposed research project is to contribute to the My Normative (MyN) app development by increasing the number of users on the app and maintaining long-term app usage. The second purpose is to verify the accuracy of the app’s menstrual cycle prediction in female persons of varying cycles less than or greater than 28-days, with or without ovulation. The project and MyN will initially target female persons in their reproductive years (18-35 years) with the bigger picture of representing all female persons of diverse backgrounds. The goal is to improve app engagement and create a platform that is inclusive, informative, and empowering to those who experience a menstrual cycle.

Anaerobic Bioaugmentation of a PHC Groundwater Plume: Pilot-Scale Experiment

Bacterial cultures are sometimes added to groundwater to increase the rate of degradation of contaminants. Three cultures that are able to completely biodegrade their primary compound to non-toxic end products in the absence of oxygen have been enriched from contaminated soils. The goal of this project is to demonstrate the efficacy of these cultures in a controlled field experiment. High resolution temporal and spatial data will be collected to estimate degradation rates, and support a process level understanding. This novel work will allow SiREM and Geosyntec to further expand its technical expertise, and provide high-value opportunities for field applications of for a suite of contaminants not currently in their portfolio.

Conception et optimisation d’un capteur de température par fluorescence d’une fibre optique dopée

Le projet proposé consiste à concevoir et à optimiser les performances d’un capteur de température basé sur la fluorescence d’une fibre optique dopée par un élément thermosensible. La dépendance du temps de vie de fluorescence de l’élément dopant la fibre optique sera utilisée pour mesurer la température de façon précise et reproductible. Ce composant photonique novateur a le potentiel d’augmenter la plage de détection des capteurs de température actuellement déployés en plus d’en augmenter significativement la robustesse et la fiabilité, et d’en réduire le coût de fabrication. Le projet proposé est en partenariat avec l’entreprise Rugged Monitoring qui est spécialisée dans le développement de capteurs à fibre optiques adaptés aux environnement hostiles.

Development of a 3D environment for ventricular puncture

Ventricular drainage is commonly performed in neurosurgery departments or in theemergency room. It consists of inserting a catheter into the brain, using a needle, until it reaches the frontal horn to drain cerebrospinal fluid for therapeutic or diagnostic purposes. Nowadays, the learning of this gesture is only performed by companionship: there is no effective simulator for training in this type of surgery. Ampere lab robotics team has been designing haptic training simulators in the medical field for two decades. It is currently designing a haptic simulator that renders the various layers crossed by the needle during such an operation.

The outcome of this internship is the design of a 3D environment from real MRI images so that surgeons get a realistic visual feedback when training on this simulator before a future operation on a patient. The patient-specific MRI images will be obtained through a partnership with a local surgeon.

Study of the synergistic effects of fatigue and corrosion in CA6NM steels and welds for hydraulic turbine applications

In this project we propose to evaluate the synergistic effects of fatigue and corrosion to further understand the degradation mechanism of the alloys used by our partner. This will help the partner defining their total cost of ownership & develop mitigating corrosion strategies to ensure long term sustainability of their infrastructure. Specifically, we will i) study two types of soft martensitic stainless CA6NM (13%Cr-4%Ni) cast low carbon steel (as received and welded), ii) quantify their localized corrosion properties, iii) study the synergistic effect that static applied mechanical stress has on the corrosion behavior.

Investigate machine learning algorithms to detect anomalies in computing infrastructures in real-time

The industry partner, Metafor is developing a new class of IT system management solution. As part of this project, Metafor is building a product feature that monitors computer and network activities and looks for signs of anomalies. This is an important problem as anomalies are usually associated with abnormal user or system behaviors that can potentially result in problems such as system breakdown. As the properties of anomalies and normal behaviours are stochastic and dynamic by nature, efficient and intelligent signal processing and machine learning algorithms are required to detect these anomalies. In this project, the intern will do a comprehensive survey on the state-of-the art of real-time anomaly detection; investigate a set of system indicators or features as well as machine learning algorithms that can potentially be useful in detecting anomalies. Finally, the intern will implement suitable algorithms to predict the presence of anomalies in the system in real time.

New Wearable Headbands for Sleep Assessment

The standard approach to quantifying sleep physiology, polysomnography, requires an overnight stay in the laboratory, is poorly tolerated by many people (especially the elderly or those with medical co-morbidities) and is both labour intensive and costly. In response to this limitation, wearable devices have gained support as tools for self-administered measurement of some aspects of sleep physiology in individuals’ homes. A new generation of wearables devices has become available that are simple to use and cost effective. The wearable consumer EEG headbands DREEM and MUSE provide useful information related to sleep, but data comparing these to polysomnography are needed. The purpose of this study is to compare the DREEM 2 headband and MUSE-S headband against standard polysomnography in older adults. In-lab validation will take place at 8 Canadian academic sleep clinics or sleep research labs, with participants undergoing diagnostic polysomnography per usual care or in the context of existing research projects. In addition to the standard polysomnographic apparatus, they will wear one of DREEM 2 Headband or MUSE Headband. We will quantify several sleep parameters from the headband and determine concordance with polysomnography.

Process development with PAT solutions to manufacture the oil-based raw material

The goal of this study will be to development a process to manufacture the oil-based raw material with PAT implementation.

Bezel-Free LCD Display system design and prototyping

Large area digital displays are often made using tiled liquid crystal display (LCD) units to form a video wall. However the seams or bezel gaps that result are often not acceptable. This project aims to demonstrate and prototype a solution to seamlessly tiling liquid crystal displays (LCDs) to form a large formal digital display.
Tilted light emitting diode (LED) arrays are to be modeled to provide the necessary design guidance to achieve high performance backlit displays. These displays will be optimized for large out-of-home installations for retail and other commercial applications in corporate, retail and transporation sectors among others.
The resulting LED backlit LCD panel technology using a novel collimated lighting design with Fresnel lenses will be tested with working hardware to validate the approach. Optimization of the design will enable maximized performance and minimized cost of such displays. The atrium at the McMaster Innovation Park in Hamilton ON will be used as a test location for this new display technology.

What Matters

Nurses and care workers have long struggled to provide high quality care for people with dementia. Struggling to connect is one of the main causes of responsive behaviours and staff injuries within hospitals and long term care homes. Staff must be empowered to deliver individualized care with a focus on what matters to the older person. Our objective is to develop a mobile app called “WhatMatters” to equip the staff with the tools needed to deliver individualized care to people with dementia. We will follow an iterative process and work closely with stakeholders, inviting them into the design process by arranging workshops and encouraging them to test the different iterations.
The anticipated outcome is to reduce staff injuries associated with behavioural events and promote cost savings within the healthcare system. The partner organization will benefit by learning the process of collaboration with multiple academic and healthcare institutions.

Stiffness Properties Improvement of Wind-Tunnel Structural Models

Wind-tunnel testing allows to verify predicted pressure distributions in various configurations on a specific model. For this verification to be successful, both the model deformation and the air flow needs to be predicted accurately. Modern wings tend to be thinner, thus more flexible and deform more under the loads applied in the wind-tunnel. Current methodologies exist to assess these structural properties outside of the wind-tunnel through static mechanical testing. The process is tedious and prone to errors. The goal of this project is to leverage Digital Image Correlation, a non contact measurement method, to accurately measure wing deformations resulting from a known load. This information can be used to re-construct a stiffness simulation model that can reproduce the full wing behavior.