Projects
Here are some of the Centre for Applied AI’s (formerly known as the Centre for Mobile Innovation) current and recently completed projects and collaborations.
-
Clinic of the future
- Faculty researchers: Prof. Sasiprya Arun, Prof. David Horachek, Dr. Rachel Jiang, Prof. Magdin Stoica, Dr. Ed Sykes
- Industry partner: Cloud DX
Project description:
This research project involves making the Clinic of the Future a reality. Designed for doctors and nurses on the ward, this Augmented Reality (AR) app and supporting patient-care centre uses Microsoft technologies and the HoloLens.
Industry partner Cloud DX is a progressive medical company led by Anthony Kaul, Robert Kaul and Dr. Sonny Kohli. Dr. Kohli, a physician in Internal Medicine/Critical Care at Oakville Hospital, is co-founder and Chief Medical Officer for Cloud DX.
Since August 2017, our Centre for Applied AI (formerly known as the Centre for Mobile Innovation) research team has worked on this project involving the MS HoloLens and a Clinic of the Future UWP app. The app uses Cloud DX’s Vitaliti wearable device to collect a patient’s vitals in real-time and presents the information in the HoloLens as augmented panels and holograms. This work was aired on the Discovery Channel (Daily Planet).
-
RAMADA – Remote monitoring for informal caregivers
- Faculty researchers: Dr. Aeiman Gadafi, Dr. Tarek El Salti
- Industry partner: PointClickCare
Project description:
The Centre for Applied AI (formerly known as the Centre for Mobile Innovation), in collaboration with PointClickCare, has created an unobtrusive app that ensures that informal caregivers (e.g., son, daughter, etc.) can monitor the activities of the recipient (e.g., resident in clinical applications, or an elder). Our system relies only on smartphones to detect and monitor activities, and aims to be secure, confidential and private. Intuitive UI and informative notifications are provided to the informal caregivers.
-
Risk fracture prediction using osteoporosis data
- Faculty researchers: Dr. Volodymyr Voytenko
- Industry partner: Inovex
Project description:
Fracture risk assessments are essential to evaluate and prevent osteoporotic fractures.
Osteoporosis Canada has been collecting data for decades. However, often the crucial Bone Mineral Density (BMD) is missing – even at the request of a physician, only 50% of people complete a BMD test.
This project focuses on creating a Machine Learning (ML) model that predicts the BMD T-score and assesses the fracture risk based on relevant patient data features. Currently, our best model (Polynomial Regression Model) performs at an accuracy of 96.1% (± 1.8 of their Actual BMD T-Scores).
-
Smart towns: optimizing urban planning using IoT
- Faculty researchers: Dr. Haya El Ghalayini
- Industry partner: Inovex and the Town of Oakville
Project description:
The Town of Oakville is interested in improving their capability of urban planning by using leading-edge technologies including IoT, Machine Learning and Computer Vision.
The Town of Oakville has a strong, long-standing relationship with Inovex. Inovex is a leading-edge digital transformation partner of choice for leading enterprises and SMEs. Inovex helps businesses skillfully build technology and custom software solutions to drive innovation and operational efficiency using cloud, web, mobile and IoT solutions.
Our aim in this project is to work collaboratively with Inovex and the Town of Oakville to deploy IoT devices and appropriate infrastructure, and to develop software systems that provide the most complete and accurate information to enable urban planners to make good decisions about creating or improving services for the citizens of the town.
-
5G autonomous vehicle research
- Faculty researchers: Dr. Khaled Mahmud
- Industry partner: Rogers
Project description:
Rogers Communications and Sheridan College are engaged in a two-year partnership with a focus on 5G autonomous vehicle research and development through Sheridan's Centre for Applied AI (formerly known as the Centre for Mobile Innovation). The collaboration involves studying the potential of integrating navigation, diagnostics, and infotainment systems into autonomous vehicles over new 5G wireless technologies and networks, and research into the areas of intelligent transport systems in a multi-user environment.
As part of the joint project, researchers and students are also using machine learning techniques combined with traffic modelling and simulations to generate insights into how Rogers can deliver innovative services to drivers and prepare a roadmap for autonomous vehicles operating on its 5G networks. The research focuses on in-car non-driving systems in autonomous vehicles like navigation, diagnostics and infotainment and explores connectivity with Intelligent Transport Systems using Cellular vehicle-to-everything (C-V2X) – a foundational technology for vehicles to communicate with each other and all devices around them.
The 5G research focuses on the following applications:
- autonomous public transportation systems
- driverless taxis
- autonomous delivery systems
- assisted driving for seniors and people with disabilities
Further reading
News release from Sheridan: Rogers and Sheridan College partner on innovative 5G autonomous vehicle research
-
Adaptive personalization using ML
- Faculty researchers: Dr. Jigisha Patel
- Industry partner: Royal Bank of Canada
Project description:
Adaptive personalization; recommender systems – collaborative filtering; present meaningful suggestions to a mobile-based RBC client.
-
Enhancing local ambient temperature accuracy using vehicle data
- Faculty researchers: Dr. Alex Babanski
- Industry partner: Geotab
Project description:
High-resolution gridded temperature datasets are extremely important for many modelling applications such as regional climate prediction models, agriculture and surface transportation systems. Currently, the primary method for gathering air temperature readings is via irregularly spaced weather stations. Most high-resolution datasets use interpolation algorithms to fit temperature curves to the station data or incorporate some models of physical processes into the interpolation methods.
One of the most promising possibilities to improve the temperature datasets is the potential to use vehicles as local weather observation systems. Geotab, a Canada-based company, is a global leader in commercial fleet telematics that collect a wealth of information from Geotab’s GO device connected to vehicle’s computer including the vehicle’s location, speed, acceleration, temperature, barometric pressure, and detailed engine diagnostics. Geotab continuously collects temperature data from more than two million vehicles in real-time.
Due to potential errors in GO device sensor readings, the accuracy of the dataset must be analyzed and improved. In this proposal, Sheridan College’s Centre for Applied AI (formerly known as the Centre for Mobile Innovation), with its experience and expertise in the field of data analytics and cloud computing teams, will create high-accuracy real-time high‐resolution temperature datasets desirable for ecological research, weather predictions and surface transportation systems.
-
An objective method for diagnosis of chronic pain
- Faculty researchers: Dr. Rachel Jiang
- Industry partner: Karmy Pain Clinic
Project description:
Workplace and motor accident fraud is a significant problem in Canada. The sums paid out for workplace and motor claims was $15 billion in 2013 in Canada and is substantially higher in other countries. In terms of fraudulent claims, experts have assessed that this represents at least 20% of the total of all such claims. Beyond the ethical, social and financial impacts, these fraudulent claims lead to higher insurance rates for all Canadians as insurance companies pass on the cost to consumers. The ability to accurately distinguish real from fake pain is crucial in the determination of the appropriate level of compensation for a patient that claims to be suffering from chronic pain.
In this research project we will address the problem of accurately detecting authentic pain by creating an Objective Method for Diagnosis of Chronic Pain system. We will develop and evaluate an automated chronic pain diagnostic system that will output a computer-diagnosed pain score with the ability to accurately distinguish real from fake pain. The proposed Automatic Pain Diagnosis System (APDS) will enhance patient care and improve clinical practice efficiencies compared to the current use of the manual Facial Action Coding System (FACS) procedure, which is laborious and difficult to use in busy clinical settings. The proposed APDS will outperform a human observer in accuracy and efficiency by using the most advanced 3D technologies and Artificial Intelligence (AI) algorithms such as Deep Convolutional Neural Networks (DCNNs). The Sheridan team, in collaboration with Karmy Clinics, will carry out the design, develop, and evaluate the proposed APDS in the clinic setting of Karmy Clinics.
-
Lifecycle management of phones and mobile devices
- Faculty researchers: Dr. Abdul Mustafa
- Industry partner: SOTI
Project description:
SOTI Inc. develops world-leading technology that solves the unique challenges involved in managing, securing, supporting and tracking remote mobile and desktop computing devices.
Today, more than 12,000 customers around the world — in retail, manufacturing, health care, government, logistics and numerous other industries — rely on SOTI software products within the mobile device management industry.
Companies strive to reduce costs by enabling the central management, security and support of remote mobile field-forces. However, despite the rapid growth and recent success SOTI has had, there are many applied research problems yet to be solved. One of the areas SOTI is interested in is lifecycle management of mobile devices. SOTI’s main functionality is in their SOTI One app. The platform is an integrated suite of solutions designed to reduce the cost, complexity and downtime related to business-critical mobility. This helps businesses remove functional silos, eliminate downtime, build apps faster, manage all mobile and IoT devices in one place and deliver actionable insights.
As SOTI’s platform runs mostly on mobile devices, it is important for SOTI to look after the health of mobile devices. The proposed research in lifecycle management of mobile devices will help in two ways:
- monitor the physical health of devices using built-in sensors of a smartphone
- monitor network security and safeguard from malicious threats using on-device computing in the form of fog or edge computing
-
Virtual Immersion Clinical Training (VCIT) program – talent accelerator
- Faculty researchers: Prof. David Horachek
- Industry partners: Toronto Metropolitan University, Medtech Canada, Unity Health Toronto
Project description:
Create an enhanced full VR environment for a typical hospital supporting interactivity in VR with surgical equipment (for the operating room) and procedures, and specialized room housing equipment such as MRI, CT-scan, Ultrasound, etc.
The goal of this one-year project is to enhance the fully functional immersive VR environment and experience including rich realistic interactivity to support several real-life end-to-end surgical procedures.
-
Beyond Silence app
- Faculty researchers: Dr. Sandra Moll (McMaster University), Dr. Ed Sykes and others
- Industry partners: McMaster University, School of Rehabilitation
Project description:
Dedicated app and research for peer-to-peer anonymous support for first responders.