Insights and Recommendation Extraction using advances in Large Language Modeling

Businesses allocate a significant amount of financial resources and time to produce regular financial and managerial reports,
which involve analyzing accounting data. These reports play a vital role in evaluating and comprehending the overall
performance of the business, thereby aiding in making important decisions. Additionally, businesses often engage industry
experts or consultants to derive actionable recommendations and strategies based on the findings from the analysis.

Automatic Brain Tumor Segmentation of Ultrasound Images with Generalized Probabilistic U-Net

The proposed research project aims to develop an advanced computer-based method for automatically identifying and segmenting brain tumors in ultrasound images using deep learning techniques. By training a specialized algorithm called the Generalized Probabilistic U-Net, the intern(s) will create a powerful tool that can accurately detect and locate tumors in ultrasound scans.

Design and evaluation of modular exoskeleton for integration with Agilik knee actuator

The proposed project aims to develop an assistive device that helps mitigate crouch gait in people with Cerebral Palsy (CP) by integrating the Bionic Power Agilik system with a tuneable exoskeleton device to explore the optimal knee ankle foot orthosis (KAFO) parameters for ultimate clinical use of the Agilik. The exoskeleton device will enable more people with CP to use the Agilik system by decreasing cost of the device and improving quality of care. We will approach this project in 3 phases: (1) design development, (2) prototyping and fabrication, and (3) characterization and evaluation.

Assessment of computerized system for diagnosis of hearing disorders based on finite-element modeling

Optical coherence tomography (OCT) vibrometry is a noninvasive tool for functional imaging of the middle ear. It provides vibrational responses and also anatomical images of the same ear. Our objective here was to explore the potential of OCT vibration measurements of the middle ear structures, to distinguish among middle-ear disorders.

Designing an Augmented Immersive Virtual Reality Driving Simulator for Advanced Alzheimer’s Disease Patients and Investigating its efficacy on institutionalized Alzheimer’s Residents

In this project a driving simulator in virtual reality will be designed and developed, in which a user can drive a virtual vehicle in a country road with incoming cars and traffic lights and possibly some animals crossing the road. The users will learn the path to reach a destination through the trial and then they are supposed to drive the virtual vehicle in the same pathway and by doing so, strengthen their spatial navigation skills. The game will be played by a physical steering wheel and two pedals for acceleration and brake like a real car.

A Collaborative Innovation Research Program on Patient Decision-Making, Technology, and Outcomes for Robotic-Assisted Knee Arthroplasty Surgery

The ultimate treatment of end stage osteoarthritis is joint replacement surgery. Over the past several decades there have been important innovations in the design and customization of implants, but many patients continue to be burdened with persistent poor outcomes after surgery, particularly with respect to unmet expectations around mobility and their ability to function and participate fully as desired.

Development of a Digital Co-care System for the Management of Parkinson’s Disease

There is no cure for PD. Limitations of the current Canadian health care environment, compounded by the effects of the pandemic, have led to infrequent and imbalanced access to care for PwP, which may contribute to inaccurate assessments, incorrect medication dosages, and undermanaged disease symptoms. To improve the lives of PwP, a chronic disease management and monitoring system is needed that involves the patients, their caregivers, and clinicians.

Sweating the Small Stuff: A Simple, Rapid, and Effective Sensor for the Biomonitoring of Fireground Carcinogens in the Sweat of Firefighters

In the line of duty, firefighters are exposed to a range of toxic chemicals, some of which are known to cause cancer. Despite the immediate dangers that accompany the occupation, cancer is the leading cause of line-of-duty deaths in firefighters. There is growing concern that long-term exposures to these chemicals at the fire scene, and during the cleaning of contaminated uniforms, gear, and stations, are contributing to these increased cancer rates. Several markers of exposure to known cancer-causing chemicals have been detected in the body fluids of firefighters.

Real-time food analysis using deep learning for Diabetes Self -Monitoring Phase 2

Our proposed research is to create an algorithm capable of pre-evaluating diabetes patients’ meals before they consume them with the snap of a picture. We are attempting to accomplish this goal by employing AI, machine learning as well as computer vision for real-time analysis. Our goal is to analyse a user's meal to return an accurate carb count and offer portion size adjustments to reduce their blood sugar fluctuations.

New technologies for low frequency photoacoustic imaging

This project is focussed on developing an optical imaging technology called photoacoustic imaging. Photoacoustic imaging is able to distinguish healthy tissue from cancerous tissue during breast surgery and could one day eliminate the need for repeat surgeries. The technology uses a pulsed laser to selectively build up pressure in some tissues and microphones to listen for sound as the pressure dissipates. After reconstruction, the sound recordings are converted back into a pressure map representative of the tissues. The project funds will support two PhD students.