Innovative Projects Realized

Investigating wildlife-road interactions in the Chignecto Isthmus Region

Roads threaten wildlife throughout the world when animals experience increased collisions with vehicles and decreased access to important habitat and resources. This research will investigate where animals are crossing roads in the Chignecto Isthmus of Nova Scotia, a region highly impacted by human development. The results will provide evidence for hotspots of negative wildlife-road interactions, with the goal of recommending effective changes to road infrastructure for the benefit of both animals and humans. The results will also help the Nature Conservancy Canada to better understand where wildlife are moving through the isthmus as they disperse between Nova Scotia and New Brunswick.

Open-domain Contextual Conversation Generation

The objective of the project is to design a system that is able to generate context-wise reasonable and meaningful responses to open-domain conversation queries. In open-domain conversation generation, the retrieval-based methods and neural network generative models are two main approaches; there are also some recent research about improving the context consistency of conversation generation. In this project, we try to use context resolution model to complete the queries to include more information from context, and use ranking models to rank the candidates from the combination of retrieval-based generation and generative model, based on their relevance to both queries and context. We will also try to use generative adversarial network and reinforcement learning in generative models to make responses with higher qualities.

Interoperative Performance Measurement of Surgeons using Deep Learning

Surgery is undoubtedly one of the most important events in a person’s life. It is thus imperative that a feedback system is in place to ensure that proper care is provided to patients during surgery. Currently, such systems involve experienced surgeons watching hours of surgery to determine how well the surgery was performed based on pre-defined criteria. This project aims to assign the surgeon’s performance rating based on data from their previous procedures. In doing so, surgeons will be assigned a technical competence rating based on their performance. A consistent rating will help differentiate surgeons that are very skilled at their craft and ones that require more training. It will also provide surgeons with additional incentive to hone their skills thereby increasing positive outcomes for their patients.

Characterizing use of the Vancouver public bike share system through 2018

Public bikeshares intend to provide an active, accessible, environmentally friendly and compact transportation alternative—particularly useful for travelling short distances and the “last mile” of a trips to and within a busy, dense urban core. In 2016 Vancouver launched a public bikeshare. Vancouver’s climate, culture, and bike route system provides a strong foundation for success, but there are concerns given the all-ages helmet legislation, long debated as a barrier to uptake of cycling and a threat to public bike share success worldwide. A major expansion of the program is planned for spring 2018, with growth from 150 stations and 1500 bikes to 200 stations and 2000 bikes. This partnership project will examine the impact of Vancouver’s bikeshare on travel and health outcomes, in the context of the expanding program. This work will help the partner organization better understand the impacts and opportunities of their program.

System-Level Performance Analysis of Shared Relaying Architecture withRealistic Channel Modeling

Future wireless networks call for seamless coverage and higher data rates at low

infrastructure cost. Cooperative relaying strategies are among the most promising of future

technological solutions toward achieving this goal. The proposed research project intends to

evaluate the system-level performance of a specific relaying architecture called the shared

relaying network. To quantify the benefit of this network, intelligent wireless resource

management schemes will be developed using efficient mathematical optimization tools.

Based on these schemes, simulation platform will be built for systematic performance

evaluation. In order to obtain accurate and reliable assessment results, realistic wireless

heterogeneous channel modeling software from SIRADEL will be used. The results are

expected to provide insights on the advantages of the shared relaying architecture as

compared to other solutions. These insights will be helpful in extending the capabilities of the

Mitacs Template Version –? July 2011 12

SIRADEL’s software product for future advanced wireless solutions, which is of great interest

to SIRADEL’s customers.

Potential roads for recovery of Atlantic salmon in Nova Scotian rivers

Atlantic salmon and its associated fisheries have a rich and complicated history in Nova Scotia. Commercial and recreational overfishing, as well as habitat damage and environmental pollution have all contributed to the species’ decline. For a century, work on rebuilding populations has met with varied successes. This history, and relative successes of different measures, will be reviewed and synthesized in this project, with a specific focus on potential recovery options for the Margaree in Cape Breton. Primary and secondary literature, combined with field work and key informant interviews will be used to make management recommendations for salmon recovery. Wild Salmon Unlimited, the partner in this proposal, will use the recommendations to chart a path forward for Atlantic salmon enhancement and recovery in the short and long term.

Mettre en oeuvre l’économie circulaire dans la filière laitière : la biométhanisation du lactosérum

Cette recherche consistera en un mandat de recherche commerciale effectuée pour la compagnie Dominion & Grimm qui contribuera au développement et à la commercialisation d’une technologie développée pour valoriser un extrant laitier sous la forme d’énergie renouvelable. Ce récent développement technologique présente un fort potentiel de commercialisation et un développement d’affaires intéressant pour l’entreprise partenaire. Cette intervention du stagiaire auprès de l’entreprise partenaire lui permettra de réaliser une série d’entretiens à Saint-Hyacinthe, qui soutiendront la rédaction d’un mémoire portant sur la réflexion et aux recherches sur la mise en oeuvre de l’Économie circulaire à partir de l’angle de la gestion.

Real-Time Operating System for Safety Critical Systems

Mannarino Systems & Software Inc. is currently developing M-RTOS, which is a Real-Time Operating System (RTOS), designed specifically to meet the requirements of the aeronautical industry’s ARINC 653 Avionics Application Software Standard Interface specification, and supporting a wide range of avionic systems. Throughout this phase of the project, the Concordia partner will provide a synthesis document covering: (i) The existing standards and guidance currently governing the aerospace industry, (ii) An evaluation of future threads and trends regarding cyber security in the aerospace industry, and (iii) A catalog of the different threats scenario and mitigation measures commonly identified and discussed in the aerospace industry. TO BE CONT’D

Mobile Analytics Interface for an Ocean Vessel Profiling Device

The goal of this project is to design and implement a mobile “dashboard” interface for an ocean vessel profiling device that captures speed, location, torque, and other data relevant to ship powering measurement and prediction. The interface will provide real-time visualizations that a pilot, engineer, or scientist can use to monitor the impact of ship maneuvers, conditions, etc. on performance. In addition, anomalous measurements or pronounced shifts in measurements will trigger prompts for annotations on the tablet, to aid further analysis and classification of events impacting powering performance.

A reinforcement learning approach to establishing a Q&A symptom checker to evolve the performance of the visual diagnosis system for dermatological diseases

The partner is creating artificial intelligence which can help diagnose over 1,300 skin conditions with dermatologist-level accuracy. They are focused on building the functionality that is to be deployed through their app and web-interface that makes it possible to snap a photo, ask questions, and get an instant diagnosis. The partner is very focused on the project given its critical need as part of the overall system/solution. The main objective of the project is to enhance and finalize Triage’s diagnostic system to the point where it is ready to be launched. Their current technology can perform a visual examination of a patient’s skin, with the existing capability limited by the inability to incorporate other context (e.g. symptom data, patient history) from the user for higher accuracy. TO BE CONT’D

Predation on Chinook Salmon Smolts by Great Blue Herons

Predation on smolts as they out-migrate down natal rivers may significantly contribute to the decline and lack of recovery of Chinook salmon in British Columbia. A prior study of mortality of Chinook smolts released by the Cowichan Hatchery suggests only a small portion of the fish reached the ocean due to predation by raccoons, river otters, mergansers and trout. The goal of this study is to analyze the data collected from Chinook smolts equipped with PIT-tags in 2016 and 2017 to estimate mortality rates (from antennae arrays that recorded movements of marked fish, and scans of predator latrines for PIT-tags), and determine sources of predation (from camera-traps and antennae arrays where animals passed; as well as from tags recovered in scats). These data are needed to confirm the apparent high rates of mortality, and provide better information about who the predators are and where and when predation is occurring.

Numerical Prediction of Mode Splitting on Rotating Disks

This project aims to develop a numerical tool for predicting accurately resonance frequencies of high-heads hydraulic turbines. Both the presence of water and the rotation of the disk-like structure induce shifts in these frequencies compared to a standing disk in air, respectively added mass and mode split. The project is articulated around the following steps: 1) understanding the physical phenomenon causing mode split, 2) developing the numerical tool for natural frequency prediction of a disk, 3) comparing model results with experimental data (communicated by Andritz Hydro) and 4) if time permits, adapt the model to real turbine geometry.
The developed tool will enable the prevention of resonance induced fatigue, which may cause turbines to fail prematurely. TO BE CONT’D