Innovative Projects Realized

AI for Laparoscopic Cholecystectomy decision support

My project’s goal is to improve the model, which predicts in real-time zones, where it is safe to dissect during laparoscopic surgery. I am going to work on the ways to mitigate tool bias of the current model, as it presence could result into misleading of the surgeon, who performs an operation and, as a result, cause significant injuries. Eventually, this project will greatly benefit all future research around tool bias problem during surgeries, deepen human knowledge about different biases in AI models and prolong the lives of patients.

Facebook Ad Development For Sexy Shroom Targeting American Men over 50

Sexy Shroom is a mushroom extract supplement created by Mycos to increase the sex drive and sexual function of men. Specifically, Sexy Shroom helps older men with erectile dysfunction by alleviating the three factors that induce erectile dysfunction (ED): endocrinological factors (i.e. Low testosterone), neurogenic factors (i.e. psychosomatic factors), and vasculogenic/metabolic factors (e.g. atherosclerosis, diabetes, obesity). This product has been previously developed in-house, while optimization and validation is ongoing. This internship will be utilized to advertise our product to men in America over the age of 50 to drive sales within America.

Automated Analysis of Call Transcripts Using Large Language Models

Analysis of customer service call transcripts can be challenging due to high volumes of unstructured textual data. To overcome this challenge, Co-operators Insurance Group seeks to explores the application of large language models (LLMs) such as GPT3 in analyzing call transcripts to extract structured data to be used in further modelling. Specifically, the project aims to design an end-to-end pipeline to filter call transcripts based on user keywords and call outcomes, extract transcript data using LLMs and format extracted information in a structured dataset. Extracted data from call transcripts could include classifications of the sentiment of the client and/or call representative, classification of client-expressed reason(s) for cancellation, and identification of call representative tactics attempting to retain the client. Co-operators insurance aims to apply this pipeline to the analysis of calls relating to mid-policy cancellations to understand the causes of these cancellations and potential methods for mitigating cancellations using insights related to customer service approach.

CFD modelling of the thermal behavior of aero engine components subjected to fire

In the proposed project, a series of computer-aided simulations will be conducted to determine the thermal behaviour of engine components in the presence of a fire inside the engine. The goal is to predict the heat transfer rate into the components, and the behaviour of the oil found inside those components. Results of this work would benefit the industry partner, Pratt and Whitney Canada, immensely by reducing the amount of time-consuming and expensive real-world fire testing to be performed during the design and testing of engines, and increase the overall safety and reliability of these components.

Role of flood events in driving geomorphic evolution of the Nicola River

The Nicola River as many parts of British Columbia were hit by major floods in November 2021. These floods caused massive bank erosion and channel widening in the Nicola River. Floods are generally thought be important driver in changing the shape of a river as the high energy of floods can move large particles in the stream or erode banks, which can have implications on nearby infrastructure and property. The observed changes in the shape of the Nicola River in response to the 2021 floods are massive. However, to better quantify the changes it is important to compare these changes with changes associated to previous flood events. Particularly, effects of flood events of different origin are of interest, as snowmelt and rainfall driven floods differ in the timing, duration, and magnitude of peak flows and they have different return periods.

Investigating the Effects of High-Intensity Interval Training Based Multimodal Prehabilitation on Inflammatory and Immune Markers in Patients with Non-Metastatic Non-Small Cell Lung Carcinoma – A feasibility study

Lung cancer is a significant health concern worldwide, and a common type called non-small cell lung carcinoma (NSCLC) is responsible for many cases. Surgery is the preferred treatment for early-stage NSCLC, but it can lead to complications after the operation, affecting patient recovery. In addition, inflammation before surgery has been linked to worse outcomes in NSCLC patients. Multimodal prehabilitation programs have emerged as a promising strategy to optimize patient health prior to surgery, potentially improving surgical recovery and reducing complications. Additionally, high-intensity interval training (HIIT) has been effective in reducing post operative complications and inflammatory markers in other conditions but its effects on NSCLC remain unknown. This feasibility study aims to investigate the effects of a supervised 4-week multimodal prehabilitation program, combining HIIT exercise, nutritional support, and psychological support on functional capacity, postoperative complication and inflammatory markers in patients with NSCLC.

Breaking Boundaries: AI-Driven Documentation at KidsAbility

Currently, KidsAbility therapists spend a substantial portion of their time on indirect client activities, primarily focused on creating extensive reports and notes that detail each child’s progress. For every minute spent with a client, clinicians spend up to two minutes writing these important documents. This process, while crucial, is time-consuming and diverts valuable resources from direct care.

Through this project, we will implement a Large Language Model (LLM) that has been trained to analyze therapists’ notes and observations and autonomously turn them into detailed assessment reports and progress notes for the clinician’s review. This AI aid is designed to efficiently streamline documentation tasks while maintaining the high-quality reports expected by families, clinicians, and other stakeholders.

By introducing this AI into their workflows, KidsAbility aims to significantly reduce the time therapists spend on indirect care, allowing them to reallocate their time to treating more clients. This project stands as a testament to KidsAbility’s commitment to using technology to serve more clients and reduce wait times.

A middleware solution for remote Monitoring, data collection and communication

The proposed research project is intended to develop a software solution (middleware) that will work in conjunction with a remote communication device (i.e. ability to communicate via satellite and/or cellular networks) to greatly reduce the cost and complexity of monitoring in-field sensors including wireless sensors and critical industrial equipment from a centrally located office environment. The initial target market for this technology is the oil and gas sector within North America with target applications of environmental and critical infrastructure monitoring

Validation/enhancement of IGA as a battery charge optimizer, and automatic general switch for semi trucks

K3 Green Tech Company Inc. aims to make the life of truck drivers and truck fleets easier and safer, bringing to the market a solution that makes the management of truck batteries more smart, cheaper, and reliable. K3 has developed a product named IGA SOLAR which equalizes the consumption of the battery using solar panels leading to a smart consumption. Partnering with Lambton College, this project will validate, improve, add new features and identify more suitable materials for the K3’s product, ensuring the product is in line with the K3’s objectives.

Evaluating Participant Satisfaction and Results Sharing in Clinical Trials

Randomized clinical trials (RCTs) are considered the gold standard for evaluating bio-medical interventions.
Their results can have important impacts on clinical practice and the development of scientific knowledge.
Often recruitment of trial participants is one of the most challenging aspects of successfully conducting a
trial. If participants are appreciated, and involved in the communication of the clinical trial results, this may
have a positive impact of future trial recruitment. Whilst clinical trials researchers may intend to share
study results with their participants, this intention is often passive with no clear plan for knowledge
dissemination. A 2018 Canadian Clinical Research Participation Survey conducted by Clinical Trials BC
(CTBC) revealed that many participants felt disconnected and underappreciated throughout their clinical
trial experience. Following the 2018 survey, to improve participant experiences, CTBC formed a Provincial
Working Group, and a Best Practices Toolkit was established to better communicate study results with
participants. Although the toolkit is currently available on the CTBC website, there is still no standardized
process in which to distribute it to the main stakeholders (i.e., clinical trial participants). The proposed
evaluation plan aims to improve the experiences of both researchers and participants in clinical trials by
producing useful and accessible scientific knowledge.

SARS-CoV-2 Vaccine Strategies

As the SARS-CoV-2 virus mutates over time and variants emerge, it is crucial to stay ahead to keep the public safe. This research aims at capturing new variants that emerge and test them against the current therapies and vaccines in place as well as develop the next generation of vaccines and therapeutics to keep the public safe.

Nonlinear phasic sensing for biomimetic electronic skin: advancing sensory in prosthetic limbs and robotics

This research project aims to enhance the feeling of touch in artificial limbs and robots through a cutting-edge solution called
biomimetic electronic skin. Existing touch technologies are limited, and this study is set out to make them feel more lifelike for
amputees and more capable for robotic systems. We are developing an efficient electronic system and using machine learning
to interpret touch signals for various gestures. The results of this work could significantly improve prosthetic technology, enable
better interaction between humans and machines, and increase the quality of life for individuals with limb amputations. This
will lead to clinical trials where users’ satisfaction and performance with these advanced features will be evaluated. The insights
gained will contribute to the design of future robotic systems, making them more efficient and self-sufficient in interacting with
the world.