Clinical Trials
Getting the right patients into the right research.
Clinical research has a participation problem. The patients most likely to benefit from trials are often the least likely to reach them because the barriers are linguistic, cultural, navigational, and systemic. We have been working on that problem since the beginning.
The Problem
Clinical trials consistently underrepresent the populations who carry the greatest disease burden. Not because those patients are unwilling, but because the path to participation was never designed for them.
Language barriers. Lack of culturally appropriate outreach. Complex eligibility criteria written for researchers, not patients. Geographic distance. Health literacy gaps. A system that asks already-burdened patients to do significant navigation work just to find out if they qualify.
The result: research that does not reflect the populations it is meant to serve. Treatments validated on narrow demographics. Evidence that does not generalize.
What We Have Built
We developed AI-driven patient-to-clinical-trial matching tools designed specifically to reach underrepresented populations. Our approach combines machine learning with an understanding of the communication and access barriers that standard matching tools overlook.
Our published research demonstrates the application of convolutional neural networks to improve the accuracy and equity of patient-trial matching, making it possible to connect patients who would otherwise be missed to research opportunities they qualify for.
Published Work: Improving Patient-Clinical Trial Matching Using Convolution Neural Networks Yousif Salman (Wilfrid Laurier University); Emad Mohammed (Wilfrid Laurier University); Cassandra Hui (Heal Mary)
Why It Matters
When MobiHealthNews covered our work, the framing was straightforward: better data and decentralized trial participation may be what fixes research's diversity problem. The Vancouver Tech Journal described our approach as building the infrastructure to change who gets access to clinical research.
We agree. The tool is only part of it. The other part is building it with the populations you are trying to reach - understanding what makes participation possible, not just technically accessible.
Who We Work With
We work with research organizations, clinical trial sponsors, and care networks interested in meaningfully expanding participation from underrepresented communities. If you are running trials and want to reach patients who are currently being missed, we would like to understand your problem.
→ Start a Conversation