Medical images hold a wealth of information, much of it invisible to the naked eye. Dr. Rebecca Thornhill (Medicine/OHRI), a medical imaging physicist, collaborates with radiologists and cardiologists to deliver safe, high-quality cardiac MRI while pushing the boundaries of what these images can reveal. Her research focuses on radiomics—quantitative features extracted from images—and how these features can feed AI models to detect disease or tissue states. As models grow increasingly powerful, her team is developing explainable AI tools to make complex decisions interpretable for clinicians and patients alike, ensuring that cutting-edge technology can be trusted and effectively integrated into clinical care.

An unconventional path to imaging physics

Dr. Thornhill’s route into imaging physics was far from typical. With an undergraduate degree in biomedical sciences, she never imagined her love of mathematics could be applied to medicine—until her honours project at Guelph, analyzing heart-rate variability in horses. That first encounter with Fourier theory and digital signal processing sparked a fascination that led her to graduate studies in medical biophysics and cardiac MRI. Over nearly three decades, her focus shifted from producing beautiful images of the beating heart to extracting meaningful diagnostic information, inspiring work in radiomics and explainable AI—a space where she draws on insights from psychology, management studies, and human factors. She says she remains perpetually humbled by how much there is still to learn and how much collaboration shapes discovery. 

Mentorship that shaped her journey

Dr. Thornhill credits many mentors for shaping her career and approach. Her 4th-year supervisor, Dr. Peter Physick-Sheard, pushed her from “stamp collecting” to real critical thinking, while Dr. Barb Lehman brought set theory to life through stories behind the theorems. Graduate advisors Dr. Frank Prato and Dr. Maria Drangova provided honest, lasting guidance long after her PhD, helping her navigate complex research challenges with clarity and integrity. Outside science, her childhood drama teacher, Yo Mustafa, taught her to project confidently and not take herself too seriously—a skill she finds invaluable in teaching, conferences, and collaborative research projects.

Advice for the next generation

Her guidance to emerging researchers is practical and human: write freely and often, even if the draft feels messy. AI can refine writing, she notes, but it cannot replace the initial connections your mind makes on the page. Foster interpersonal respect across disciplines, intentionally include all members in labs and workshops, and bring your full humanity to your work. Most importantly, choose to help when in doubt—you’ll rarely regret it.

Goats, giggles, and cosmic jokes

For those who know her, Dr. Thornhill’s love of animals is unmistakable—especially goats. She delights in sharing “goat news,” whether it’s their role in reducing California wildfire risk or their hilariously mischievous antics. Their impudent charm, she says, feels like they’re in on a cosmic joke, a reminder that life is often ridiculous. “Find your people—or animals—and make time to laugh about it together,” she advises. 

In emergency medicine, uncertainty can be dangerous. Dr. Thiruganasambandamoorthy’s (Emergency Med/OHRI) research focuses on reducing that uncertainty for patients with cardiovascular emergencies—particularly syncope, presyncope, and chest pain—by improving risk assessment and decision-making. Central to this work is medical artificial intelligence, using data-driven tools to support faster, safer, and more consistent clinical decisions while complementing the expertise of frontline clinicians.

His program aims to reduce morbidity and mortality through robust risk stratification in the ED and beyond. By combining clinical expertise, large-scale datasets, and emerging AI methods, his work replaces variability in care with evidence-informed precision.

From clinical intuition to algorithm-informed care

A cornerstone of his work is the Canadian Syncope Risk Score (CSRS), developed with input from patients, emergency physicians, cardiologists, paramedics, and health administrators. While grounded in traditional clinical predictors, the CSRS laid the foundation for more advanced, data-driven risk tools.

Building on this, Dr. Thiruganasambandamoorthy is exploring machine learning and natural language processing to refine risk prediction for syncope and chest pain. By analyzing patterns in electronic health records, clinical notes, and patient outcomes, AI models can detect subtle signals that might be missed under time pressure. These approaches are designed to support decision-making, not replace it, offering clinicians a layer of insight grounded in real-world data. 

Extending AI-supported care beyond the ED

Medical AI also informs efforts to bring monitoring outside hospital walls. The REMOSYNC Study evaluates 15-day remote cardiac monitoring for higher-risk syncope patients discharged from the ED, enabling earlier detection of arrhythmias while reducing unnecessary admissions. By combining remote monitoring with predictive analytics, this model provides continuous, personalized data to guide follow-up care. Simultaneously, the Canadian Prehospital Syncope Risk Score equips paramedics to make data-informed decisions before patients reach the hospital, extending the reach of AI-enhanced risk assessment into prehospital care.

Precision, equity, and responsible innovation

His work also addresses equity in diagnosis. The CODE-MI trial examines female-specific thresholds for high-sensitivity cardiac troponin testing, emphasizing that algorithmic tools must be carefully designed to avoid bias. For Dr. Thiruganasambandamoorthy, responsible medical AI means improving precision while ensuring fairness, interpretability, and clinical relevance.

A research path shaped by frontline care

His focus on AI-enabled decision support grew from early work in a rural hospital, where limited resources highlighted the consequences of uncertainty. Those experiences continue to drive his commitment to practical, interpretable tools that can be used at the point of care—enhancing safety, efficiency, and outcomes for patients and clinicians alike.

Persistence, in research and in life

Asked about mentorship, he cites Srinivasa Ramanujan, the mathematician whose originality and perseverance in the face of adversity remain a source of inspiration. His advice to the next generation is equally direct: never give up, professionally or personally.

Life beyond the lab coat: patience rewarded

The most challenging journey of his life, he says, was becoming a father. After decades of trying, he welcomed his child at age 53—a deeply personal reminder that perseverance shapes outcomes both in medicine and beyond.

Dr. El Emam’s (SEPH/CHEO RI) research sits at the intersection of machine learning, privacy-enhancing technologies, and prediction models, with a clear focus on real-world application. Through his work, he develops methods that allow data-driven insights to be generated responsibly—ensuring that powerful analytics can be used without compromising privacy, trust, or social value. His research program, rooted in digital health and applied computing, is designed not just to advance methods, but to make them usable where they matter most: in complex health and societal systems.

At the core of this work is a commitment to practicality. Rather than treating machine learning as an abstract technical exercise, Dr. El Emam focuses on prediction models and privacy-preserving approaches that can be deployed in real settings, informing decision-making while respecting legal, ethical, and social constraints. It is research shaped as much by implementation realities as by algorithms. 

Motivation grounded in impact

Dr. El Emam was drawn to this field by a desire to work on problems with tangible social and economic benefit. Early on, he recognized that data and machine learning—when thoughtfully designed—could help address large-scale challenges, particularly in health. That motivation continues to guide his research choices today: selecting problems not for their theoretical elegance alone, but for their potential to improve systems, outcomes, and efficiency in the real world.

Learning from many mentors

Reflecting on mentorship, Dr. El Emam emphasizes that his path was shaped not by a single figure, but by many individuals at different stages of his career. Each mentor influenced key decisions—whether encouraging him to explore new directions, take calculated risks, or stay the course when challenges arose. Collectively, these relationships helped him develop both technical depth and a pragmatic approach to problem-solving, reinforcing the value of perspective, experience, and honest feedback. 

Advice for the next generation

When asked what he looks for in the next generation, Dr. El Emam is candid. Focus matters. Time is finite, and learning to say no—to distractions, misaligned projects, or diminishing returns—is essential. He also stresses the value of mastering multiple disciplines. Today’s problems rarely fit neatly within a single field; the most effective people are the ones he calls the “unicorns” - those who can combine, for example, computing and law, or clinical insight and computing, or all three.

Stability, he notes, can be underrated. Staying in one place long enough to build trust and networks often enables deeper collaboration, impact, and influence. Equally important is learning to manage stress early—through habits that support health, resilience, and sustained performance.

Above all, he emphasizes teamwork and execution. Complex problems require strong, motivated teams, and progress depends on people who persist in understanding problems and then follow through to get things done. Pragmatic optimism, even when challenges loom, helps teams move forward. And mentors—plural—remain essential. Identifying people willing to invest their time, and actively engaging with their feedback, can shape a career in lasting ways. 

A fact from early days

One detail not easily found on his CV: Dr. El Emam founded his first research spin-off company while still a graduate student. Fun fact or not, it reflects a consistent theme in his career—turning research into something that works beyond the lab.