Marwa Suraj is a physician and data scientist from Ghana with a keen interest in public health and preventive medicine. She currently works as a Public Health Lead with the Canadian Red Cross, serving indigenous communities within Manitoba and Nunavut. Her areas of focus include the judicious use of AI in population health management. Outside work, she is an ardent fan of Manchester United FC and enjoys watching football.

What initially sparked your interest in medicine, and what led you into general practice?

There were definitely multiple contributing factors, but I believe it stemmed from my lived experiences growing up. As a child, I recall falling sick quite frequently. It was uncomfortable for my parents and for me. Whenever we’d visit a hospital, I could see the helplessness in other kids, and I thought to myself, I wish I could help stop this unpleasant feeling.

So as cliche as it might sound, I just wanted to help sick people feel better! I thought I’d become a pediatrician for the longest time, but my interests evolved as I got older.

When did you become interested in medical AI?

It was in 2018. I was new to Canada and Toronto, navigating my way around the healthcare system and passively searching for some inspiration for career advancement. Then I chanced upon a free event by AIMed – it took me 2 trains and 3 buses to barely make it on time for the breakfast meeting. Dr Anthony Chang and a diverse panel of clinicians, engineers and data scientists talked about advancements in AI and healthcare. I listened, took notes, interacted with other attendees, and returned home feeling like I had found my eureka moment.

You are a self-taught data scientist. What are the main resources you used to educate yourself?

People underestimate the value of MOOCs (Massive Open Online Courses), but that is where I learnt most of the basics. I committed to these two courses, and I highly recommend them to any self-starter:

  • Coursera Machine Learning by Stanford University (instructed by Andrew Ng)
  • Udemy Machine Learning A-Z by Kirill Eremenko

In addition to these, I read a lot of publications and listen to podcasts specific to medical AI.

What advice would you give someone starting their career in medicine or medical AI?

Many say the practice of medicine is a calling. It gets really confusing because different people pursue a career in medicine for different reasons, but mostly with good intentions. Still, the tricky part is that one’s intention is a hard thing to prove. For me, the practice of medicine is an art. What matters most is the magnitude of impact you strive to obtain for your beneficiaries or patients.

The premise holds for a career in medical AI as well.

You have continued to practice as a physician, as well as working for humanitarian agencies. How do you manage to balance these diverse roles and responsibilities?

I haven’t had the chance to do a lot of clinical work since I moved to Canada in 2018. However, I have worked as a public health specialist on multiple projects for humanitarian agencies. Juggling diverse roles and responsibilities comes with challenges. Sometimes I fail woefully at time management, but on most days the energy that fuels my enthusiasm is passion. It is true what they say: “If you love what you do, you’ll never work a day in your life.”

Tell us about your first AI project. How did it come about, and what were the outcomes?

I did some pro-bono work with an NGO that used AI and data in the service of community health workers (CHWs) who travel around to provide life-saving primary healthcare to hard-to-reach communities in sub-Saharan Africa.

Despite the CHW sector being a data mature space, we identified significant gaps in recruitment and retention workspaces, deployment and usage of digital health tools, and management of pre-existing workflows. The idea was to identify areas where data science, ML and AI could create more efficient workflows for partnering mission-driven organizations deploying CHW to these countries.

Being a black female physician on the team was an added advantage because I had the concept of social determinants of health (SDoH) at the heart of everything I did. Because of my medical background, working on this project felt like I was visualizing the tiny pieces of a puzzle with binoculars. I saw all the moving parts and anticipated the challenges from a distance.

Like many AI models, we identified a problem, found a practical solution, but did not get a chance to materialize it in the real world at a scale large enough to drive sector-wide impact. My lessons learned could be summarized in one sentence: “Communication trumps technology any day of the week”. Without a mutual understanding among all stakeholders involved, your project is at risk of gathering dust on a deserted shelf.

Overall, It was a great experience that opened my mind to the possibility of doing this as a vocation.

What excites you most about the future of AI?

The mere fact that we are yet to see the best AI innovations in healthcare is enough excitement for me.

Who’s been the biggest influence on your career?

It is a difficult question because I have been lucky enough to find a mentor who cheers me on or motivates me directly or indirectly at every phase of my career.

During the early stages of my profession, Dr Richard Dey (a Ghanaian cardiologist) was my most significant influence. He spurred me on to believe in achieving the seemingly impossible. At the moment, my mentor and friend Dr Simon Treissman is a big reason why I continue to explore AI in healthcare confidently. Also, Dr Eric Topol, whom I have never met, is a huge inspiration because of his numerous works on integrating digital medicine, genomics and AI.

What’s the best piece of advice you’ve ever received?

My father is my greatest cheerleader and has passed on countless invaluable nuggets of wisdom to me over time. But this quote from Marie Curie happens to encapsulate all of my values as a physician:

“You cannot hope to build a better world without improving the individuals. To that end, each of us must work for his own improvement and, at the same time, share a general responsibility for all humanity, our particular duty being to aid those to whom we think we can be most useful.”

If you could return to the past, what would you change or do differently?

A few things. I would probably opt to be more fearless of failure. I’d get rid of the fear that nourished my hesitations and led to a habit of procrastination. My current understanding of AI gives me a whole new perspective on issues that stagnate the quality of primary health care delivery at the community level. With a new perspective comes creative problem-solving skills. If I could go back in time with this current knowledge, I’d probably save myself a lot of time and frustration on my long shifts in the ER.