Innovative Projects Realized

eQuiet – Étude et développement d’un bouchon d’oreille intelligent pour chevaux

Ce projet vise à étudier et développer une preuve de concept d’un bouchon d’oreille pour chevaux. Les bouchons d’oreille sont utilisés pour réduire le niveau sonore des chevaux afin de réduire les distractions et d’améliorer leur concentration. Toutefois, le niveau d’atténuation des bouchons disponible sur le marché est limité. De plus, on dénote un intérêt grandissant pour le monitorage des signes vitaux des chevaux. Ceci permet d’assurer la santé des chevaux dans plusieurs domaines tel que le dressage, les courses et les calèches. L’objectif du projet est donc d’étudier et de concevoir une preuve de concept pour un bouchon intelligent capable de monitorer les signes vitaux d’un cheval. Ce premier prototype devra fonctionner sans fil pendant quelques heures, pouvoir capter les signes vitaux du cheval et transmettre les données sans fil.

Using Machine Learning to Classify and Access an Audio Database

Voices.com, the largest online marketplace of voice talent, have identified Machine Learning as an enabler for future growth. In particular, incorporating Natural Language Processing (NLP) into structured queries and automatic classification of sample recordings. The first phase of this research will use machine learning to identify and train an NLP learnable parser. The second phase will be to automatically classify sound samples, which has been historically difficult due to low levels of accuracy. The approach will start with attributes for which accurate algorithms exist and new attributes will be added when usable levels of accuracy are achieved. Note that self-reported attributes are also not very accurate, so it is possible that automatic classification would beat user-reported attributes. The classification problem cannot be completely solved within the first year, but these classifications could speed up human processing or be used in the absence of human classification.

Mechanisms underlying stability and accuracy of electrocardiogram textile electrodes and breathing strain sensors

Electrocardiogram (ECG) electrodes are used for cardiovascular health monitoring with applications in health and wellness. When integrated in textiles, these electrodes can detect ECG over longer periods of time and provide more comfort for applications in sports and health monitoring. In collaboration with Vitali Wear, a leading company in the area of smart bras for monitoring of stress and quality of meditation, this project focuses on investigation of the underlying mechanisms related to stability issues of ECG textile electrodes for application in smart bras for detection of heart rate (HR) and heart rate variability (HRV) and textile strain sensors for breathing rate. The UBC team will investigate the response of knitted electrodes within textile to frequent mechanical flexing, washing and temperature cycling. The interplay between the type of yarn, conductive materials as well as protection films on the quality and stability of signals will be investigated.

News Driven Risk

This project centers on the effect of firm-level news on equity volatility. While there has been a strong emphasis on macroeconomic announcements, we all know that the news about the failure of a single bank marked the beginning of the sub-prime crisis. Thus, I study the effect of news on a firm-specific basis. The first part of the project focuses on decomposing firm news’ overall impact into the beta, market, and idiosyncratic volatility channels. Preliminary results suggest that there exist fifty eights news categories related to price jumps. I examine how market processes firm-specific news during different states of the economy and determine the types of firms and news that move the market. TO BE CONT’D

Effect of prepartum dietary starch content on gut function, metabolism, and production of dairy cows in early lactation

Previous studies in dairy cow nutritional management during the calving transition mainly focused on how to improve milk production after calving, yet gut function and health during this critical period is largely unknown. Glucagon-like peptide-2 (GLP-2), a 33-amino acid peptide secreted by enteroendocrine L-cells of the gastrointestinal tract (GIT), has been reported to increase small intestinal mucosal growth and blood flow and reduce gut permeability and mucosal injury. Although most studies investigated the biological effects of GLP-2 used non-ruminants as animal model, recent research in calves demonstrated that GLP-2 has the similar physiological effects as it does in non-ruminant. However, the effect of GLP-2 on gut function of dairy cows and how this related to production remains unknown. Therefore, this study will evaluate the effects of prepartum diets on gut peptide, metabolism, and production of dairy cows in early lactation. TO BE CONT’D

Les effets de la participation à un programme d’éducation musicale sur le développement de la relation parent-enfant et sur le développement du jeune enfant

Depuis quelques années, la communauté scientifique s’intéresse aux effets de la pratique de la musique sur le développement de l’enfant. À ce jour, nous savons que la pratique de la musique soutien le développement cognitif et langagier de l’enfant. Elle encourage aussi l’adoption de comportements prosociaux, tels quel l’entraide et la coopération. Néanmoins, à notre connaissance, peu d’information ne semble disponible quant à l’impact de la pratique d’activités musicales en dyades parents-enfants. Ainsi, ce projet cherchera à mesurer les effets de la participation à un programme d’éducation musicale sur la relation parent-enfant et sur le développement du jeune enfant (24 à 42 mois). En parallèle, le partenariat proposé permettra de favoriser le développement des compétences professionnelles des éducateurs et éducatrices à la petite enfance.

Mapping blood brain barrier permeability- towards a novel biomarker for psychiatric disorders

Research into brain diseases has until recently focused on neurons, but in the last several decades, blood vessels have taken a new spotlight. When blood vessels are damaged, substances can leak from the blood into the brain. Such leak can affect memory and mental health, but surprisingly, its impact on psychiatric disorders has yet to be studied in detail, largely due to lack of technology to map leaky blood vessels in living patients. Recent advances in MRI technology and computer processing power offer new avenues for solving this issue, allowing us to track blood flow and identify affected brain regions and their role in brain disorders.
In this project, we are developing novel computational methods to automatically analyze brain scans and produce detailed mapping of the regions affected by blood leakage. We are also creating algorithms to see whether leakage corresponds to abnormal patterns of brain activity. TO BE CONT’D

Analysis and optimisation of a sensor network for the remote detection of physiological parameters

Measurement of physiological parameters is the cornerstone of medical diagnosis. The rising costs of healthcare and the aging of the population are factors that influence researches in the medical industry, mainly for the development of assisted living and smart home. Several technologies have been proposed in the literature for monitoring people and health care. Recently, radar technology for monitoring the vital functions of a patient has attracted a lot of attention. With the bio-radar, it is also possible to send a rapid emergency alarm in the case of falls of people, sleep apnea syndromes or the sudden death in newborns, and rescue of survivors after an earthquake. The combination of communication standards and the reconfiguration of architectures have led to an evolution of radar architectures and the development of new technological concepts which can alleviate these important health problems. TO BE CONT’D

Community dynamics in restored salt marshes

Salt marshes are important coastal ecosystems because they provide many services to surrounding areas. Due to the high fertility of salt marsh soil, they have a long history of being converted to farmland in Maritime Canada. In recent years, there has been growing interest in restoring salt marshes to protect against coastal erosion, mitigate sea level rise, and provide increased habitat for birds, fish, etc. Ducks Unlimited Canada (DUC) and partners initiated a large salt marsh restoration project in 2009-2010 in Aulac, NB, with the goals of reverting farmland to salt marsh and assessing the effectiveness of restoration methods. TO BE CONT’D

Clinical Decision Support Software for Prediction of Postoperative Atrial Fibrillation Following Bypass Surgery

The problem of accurately predicting the onset of sustained postoperative atrial

fibrillation (AF) in patients undergoing coronary artery bypass grafting (CABG) remains open.

Investigators have reported many clinical indices currently associated with postoperative AF

following CABG. Contemporary machine learning techniques are well-suited to recognizing

underlying trends in ‘training’ data consisting of several labeled examples, and to using the

results to classify new unlabeled data with remarkable sensitivity and specificity. We propose

the development of advanced clinical decision support software capable of automatically

gathering and analyzing relevant clinical data from patients undergoing CABG in order to

provide physicians with objective and non-invasive insights into the likelihood of sustained

postoperative AF so that patient morbidity and mortality, as well as healthcare costs, can be

significantly reduced by targeting appropriate preventative therapies.

Investigation of the strength of bonded joints for lightweight vehicle assembly

In this research project, strength and behaviour of bonded joints is investigated under various loading and geometrical conditions. The interns will prepare specimen in accordance with industry guidelines, test the specimen and record data. Interns will study and investigate the joint behavior under several different scenarios, will model material behavior from the test data. This will enable making reliable prediction of joint performance under different application-specific conditions, using finite element modeling and analysis. The result of this study will enable automakers to use adhesively bonded joints instead of spot welds, thus benefiting from the cost savings.

Abstraction of temporal memory in the human cortex

Episodic memories are made as we go about our daily life: what event has happened, with whom, where and when. The proposed project will target mechanisms of episodic memory retrieval in life-like conditions using advanced neural data analytical tools.
The main notion is that in real-life conditions different episodes/events are not independent, but are connected in a meaningful manner. For example, if a friend sends an SMS about meeting at a train station (event 1) and then sees their friend in front of the station (event 2), these two events are coupled. One could think about many other different types of “event-coupling” (e.g., different stages/missions in a video-game). We aim to formulate formal ways of representing these links, which would then allow us to search for the corresponding neural correlates in the brain. Specifically, it will include triggering episodes in the real-life of the participants using mobile phone technology and studying the interaction between different kinds of temporal-order memory. The research methodology will include advanced functional imaging methods, electroencephalogram (EEG), as well as transcranial magnetic stimulation (TMS).