Rahimi's 
Research Lab

Canada Research Chair in AI and Advanced Digital Primary Health Care

Our Projects

A quick look at what’s happening in the Our Projects.

Cardio-Twin Study

Cardiovascular diseases (CVD) are the leading cause of death worldwide and second leading cause of death in Canada. Indeed, 206 Canadians (60% of them women) die from CVD every day. CVD cost Canada $20.9 billion per year. Most CVDs can be prevented/managed. Accurate point of care predictions and early and personalized recommendations for prevention of CVD can help.

Our study is developing and testing CARDIO-TWIN, a digital tool that creates a “digital twin” of a patient to predict heart disease risk and suggest personalized prevention strategies. Delivered in primary care, CARDIO-TWIN aims to catch heart problems early, guide prevention, and support shared decision-making between patients and doctors.

Fairness Study

This project focused on making machine learning tools in healthcare more fair and trustworthy. Using data from over 1,600 older adults with COVID-19, the team developed methods to ensure predictive models provide equitable insights across different patient groups. The work advances responsible AI in health, promoting accuracy, transparency, and fairness in real-world clinical applications.

EDAI Study

This study developed a practical framework called EDAI to guide the integration of equity, diversity, and inclusion at every stage of creating and using AI systems in health and oral health care. The researchers brought together experts and community representatives through workshops and review processes to identify principles and concrete indicators that help ensure AI tools consider diverse population needs and reduce biases in data, design, and implementation.

The impact of this work lies in helping health technology developers, clinicians, and policymakers build more responsible and trustworthy AI — systems that are not only technically robust, but also fair, inclusive, and sensitive to the real‑world needs of different patient groups.

HOPE Study

Depression is a common but often underdiagnosed condition among older adults, affecting their quality of life, physical health, and independence. Early detection is critical, yet traditional screening can be burdensome and may miss subtle changes in daily behavior.

We developed the HOPE model, a machine learning approach that uses Wi‑Fi motion sensors in people’s homes to detect patterns in activity and sleep linked to depression — all without requiring wearable devices or clinic visits.

This study shows a nonintrusive, scalable way to monitor mental health in older adults, helping identify signs of depression earlier and enabling timely support. By capturing real-world behavior in daily life, it has the potential to improve care, prevent worsening symptoms, and promote independence and well-being for older adults.

AIFM-ed Curriculum Framework for Postgraduate Family Medicine Education on Artificial Intelligence

As healthcare increasingly relies on artificial intelligence (AI), all healthcare providers need the knowledge and skills to use these tools safely and effectively. Without proper training, there is a risk that AI will be underused or misapplied, which could affect patient care.

To address this, we developed AIFMed, a curriculum framework specifically for family medicine residents, equipping them to confidently apply AI in clinical practice. This approach highlights the broader need. Similar AI education should be adapted and implemented across other care specialties so that all providers can harness AI to improve decision-making, patient outcomes, and the overall quality of care.

Learning and
education

Learn how we are preparing the next generation to use AI safely and effectively. Here, you can explore our collection of various resources to become more familiar with the fundamentals we work on by accessing different educational materials, resources, and information about our upcoming events.