Future Thinker

What drew you to a career in medicine?
 

When I was going through my late great grandfather’s belongings, I discovered an old doctor’s bag with a stethoscope inside. It didn’t belong to him, I discovered, but apparently he always dreamed that one of his family would become a doctor. None of his kids or grandkids went down that route, so then it came to my generation. Out of 20 or so cousins, I was the only one who showed any interest.

As a child, I really liked science – and I had a very curious mind. My dad noticed my interest and converted our garden shed into a laboratory for me. I would spend hours there, perfecting my animal dissections in preparation for my biology exam. 

I applied for medical school in Pakistan, but was only offered a place for the following year. Rather than wait a year, I decided to study in the US, and enrolled for a biology degree course. My research projects were in infectious diseases, and this led me to a PhD, developing a vaccine. After that I went to medical school.

I really enjoyed doing research, but I was keen to find a discipline that combined research with patient care. It was also noted that in every rotation I did, I was drawn to the lab to look at my patients’ samples under the microscope. I liked the “detective work” that was involved. It became clear that pathology was the perfect fit for me.

How did your interest in digital pathology evolve?
 

I was particularly interested in looking at images of patient samples. In the early 2000s, a couple of years into my residency, the first slide scanners became available. My department had one for research, and I was excited by its potential in terms of sharing images with different users. We set up a website and shared some static images on it for training purposes – and it soon became a part of my daily practice.

I took a job in Pittsburgh, which was at the center of informatics at that time. Soon after that, slide scanners began to be validated for clinical use, and I was involved in some of the early trials and product development. Even as long as 20 years ago, cytologists were already using automated systems to analyze PAP smears, so AI-powered digital pathology was a very natural progression.

During my 10 years in Pittsburgh, we managed to set up telepathology networks, sharing slides across the state – and even across countries. A whole slide image scanner was installed in China and we set up a service consulting on cases with pathologists over there.

Recognizing this potential, I started working with the FDA and the CAP to develop the regulatory frameworks for digital sign-out of cases. 

What have been your greatest achievements?
 

I am proud of being part of the movement to make digital pathology available for primary diagnosis in the US. Initially, the FDA considered whole slide imaging to be a high-risk technology. I was part of a group of pathologists who secured a meeting with the FDA to address this. It was only an hour’s meeting, but we managed, with the help of the DPA, to start a conversation about the regulatory issues surrounding the use of whole slide imaging for clinical use and create a roadmap towards approval.

Our team at Omnyx designed a study with some slide scanner manufacturers to test the hypothesis that digital slides were non-inferior to glass slides for diagnosis. We were able to present the results to the FDA and were involved in several discussions with them that eventually led to the first approval of a slide scanner for use in primary diagnosis in 2017. 

After that, we continued to work with the FDA to secure approval for two more slide scanner devices. I was the Principal Investigator on the trial of that third scanner, which involved 20,000 reads. 

When I moved to The Ohio State University, we set up a digital pathology workflow and were the first hospital in the country to sign out cases digitally. That was in 2018. I remember the day well – I signed out the case on this very monitor on my desk, with my colleagues crowded around me, observing. It felt momentous! And it was a big deal for me, because I dreamed it would happen all those years ago, and I was able to help drive it forward and see the dream become reality. It took many years to attain approval. But, since then, many thousands of patients have benefited from the availability of digital pathology for primary diagnosis and consultations.

What do you think the pathology lab of the future will look like?
 

I think we're in the knowledge age of pathology right now. We start with data – that leads to information – and from that comes our knowledge. And, in the future, AI will take us into the wisdom age. We must remember to separate artificial intelligence from real intelligence, real experience, and the ability to learn from our mistakes. We can rely on AI to take over tasks that are very manual or very labor intensive, and to do them well. But we must see the work as a partnership, with each side sticking to what it’s good at. We must continue to use our wisdom. We might be able to liken the future to the highways, where we will find a combination of self-driving and non-automated cars; in the lab, there will be some processes with AI copilots and some without. 

I also predict that the specialties that are highly dependent on imaging will start to merge because of AI. Diagnostic teams will consist of pathologists, radiologists, and oncologists all working together on cases. There will be a collection of tools and skill sets, which will be highly customized, comprising the patient care team. In fact, diagnostic imaging might become a specialty in its own right. 

In terms of tools, we now have systems that can interrogate the genetic activity within a single cell or in the tumor microenvironment, and these will become available far more widely for diagnoses. Generative AI will allow three-dimensional imaging. In fact, I saw a prototype of a machine that generates an image directly from a tissue sample, just like a CT scan. So, the technologies we need already exist; we need to ask how we can safely and ethically incorporate them into the workflow. 

Finally, there will be no digital or computational pathology in the future – it will all simply be “pathology.”

What does a typical day look like for you?
 

My day starts around 6 am, going through my emails and planning my day. Then I might spend an hour on my publications and papers. My clinical work typically starts at 8 am. In addition, around 20 percent of my day is spent on management tasks. Then around 10 percent of my time goes to my research projects, particularly on AI development. I also devote some time to teaching. 

When I return in the evening, I’m careful to take a break and avoid turning on my laptop so I’m fully rested for each day ahead.

The best part of my job is that it’s not a job! I’m excited to go to work – even after 25 years!

What do you do to relax?
 

I often go to the gym on my way home. Then, once home, I might go out for dinner with my family (I like trying different ethnic restaurants) or I might cook a meal at home. Watching movies and playing board games also helps me to relax. And I love to travel.

During my early career, I worked too much. But I learned to compartmentalize my life into work and personal space; now, my work–life balance is better.

What advice would you give people aspiring towards pathology leadership?
 

Engage with the people around you, communicate your vision, and keep dreaming. It’s also important to identify the problems you’re trying to solve, create the right team around you, and work towards specific goals. Problem-solving is not just about throwing money at it – throw your vision at it.

With your goals laid out, establish the steps needed to achieve them, and make sure you have buy-in from your team. Also, accept that it might take time. It took 12 years from the time I dreamed of digital sign-out to the time I achieved it! 

For people following in my footsteps, it will be important to keep looking at technology developments, and applying those to the future vision. The question should be: “How can I transform the pathology practice of the future.”